Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
Frontiers in Earth Sciences
Series Editors: J.P. Brun, O. Oncken, H. Weissert, W.-C. Dullo
.
Victor A. Melezhik Editor-in-ChiefLee R. KumpAnthony E. FallickHarald StraussEero J. HanskiAnthony R. PraveAivo Lepland Editors
Reading the Archiveof Earth’s OxygenationVolume 3: Global Events and theFennoscandian Arctic Russia -Drilling Early Earth Project
Editor-in-ChiefVictor A. MelezhikGeological Survey of NorwayTrondheim
Centre of Excellence in GeobiologyUniversity of BergenNorway
EditorsLee R. KumpDepartment of GeosciencesPennsylvania State UniversityPennsylvaniaUSA
Anthony E. FallickEnvironmental Research CentreScottish UniversitiesEast KilbrideUnited Kingdom
Harald StraussInstitut f€ur GeologieWestf€alische Wilhelms-Univ. M€unsterM€unsterGermany
Anthony R. PraveDepartment of Earth ScienceUniversity of St AndrewsUnited Kingdom
Eero J. HanskiDepartment of GeosciencesUniversity of OuluOuluFinland
Aivo LeplandGeological Survey of NorwayTrondheimNorway
ISSN 1863-4621ISBN 978-3-642-29669-7 ISBN 978-3-642-29670-3 (eBook)DOI:10.1007/978-3-642-29670-3Springer Heidelberg New York Dordrecht London
Library of Congress Control Number: 2012944339
# Springer-Verlag Berlin Heidelberg 2013This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material isconcerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproductionon microfilms or in any other physical way, and transmission or information storage and retrieval, electronicadaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed.Exempted from this legal reservation are brief excerpts in connection with reviews or scholarly analysis or materialsupplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by thepurchaser of the work. Duplication of this publication or parts thereof is permitted only under the provisions of theCopyright Law of the Publisher’s location, in its current version, and permission for use must always be obtained fromSpringer. Permissions for use may be obtained through RightsLink at the Copyright Clearance Center. Violations areliable to prosecution under the respective Copyright Law.The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does notimply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws andregulations and therefore free for general use.While the advice and information in this book are believed to be true and accurate at the date of publication, neither theauthors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made.The publisher makes no warranty, express or implied, with respect to the material contained herein.
Printed on acid-free paper
Springer is part of Springer Science+Business Media (www.springer.com)
Dedication
The editors respectfully dedi-
cate this three-volume treatise
to Dr. Alexander Predovsky of
the Geological Institute of the
Russian Academy of Sciences
in Apatity. He is one of the
earliest explorers of the Pre-
cambrian geology in Russian
Fennoscandia, and his half cen-
tury of active work on the geo-
chemistry of sedimentary and
igneous rocks provided impor-
tant foundations for the current
understanding of Palaeopro-
terozoic stratigraphy, geochem-
istry of sedimentary and
volcanic processes and ore for-
mation in the region.
.
Contributors to Three-Volume TreatiseReading the Archive of Earth’s Oxygenation,Volume 3: Global Events and the FennoscandianArctic Russia: Drilling Early Earth Project
Wladyslaw Altermann Department of Geology, University of Pretoria, Private Bag X20,
Hatfield, Pretoria 0028, South Africa
Dan Asael Institut Universitaire Europeen de la Mer, UMR 6538, Technopole Brest-Iroise,
Place Nicolas Copernic, 29280 Plouzane, France
Alex T. Brasier Scottish Universities Environmental Research Centre, Rankine Avenue,
East Kilbride, Glasgow G75 0QF, Scotland, UK
Ramananda Chakrabarti Department of Earth and Planetary Sciences, Harvard University,
20 Oxford Street, Cambridge, MA 021 38, USA / Center for Earth Sciences, Indian Institute of
Science, Bangalore 560 012, India
Daniel J. Condon NERC Isotope Geosciences Laboratory (NIGL), Keyworth, Nottingham,
UK
Alenka E. Crne Geological Survey of Norway, Postboks 6315 Sluppen, NO-7491
Trondheim, Norway
Nicolas Dauphas Origins Lab, Department of the Geophysical Sciences and Enrico Fermi
Institute, The University of Chicago, 5734 South Ellis Avenue, Chicago, IL 60637, USA
Yulia E. Deines Institute of Geology, Karelian Research Centre, Russian Academy of
Sciences, Pushkinskaya 11, 185610 Petrozavodsk, Russia
Patrick G. Eriksson Department of Geology, University of Pretoria, Private Bag X20,
Hatfield, Pretoria-Tshwane 0028, South Africa
Anthony E. Fallick Scottish Universities Environmental Research Centre, Rankine Avenue,
East Kilbride, Glasgow G75 0QF, Scotland, UK
Juraj Farkas Department of Geochemistry, Czech Geological Survey, Geologicka 6, 152 00
Prague 5, Czech Republic / Faculty of Environmental Sciences, Czech University of Life
Sciences, Kamycka 129, Prague 6, 165 21 Suchdol, Czech Republic
Michael M. Filippov Institute of Geology, Karelian Research Centre, Russian Academy of
Sciences, Pushkinskaya 11, 185610 Petrozavodsk, Russia
Harald Furnes Department of Earth Science and Centre for Geobiology, Allegaten 41,
Bergen N-5007, Norway
Igor M. Gorokhov Institute of Precambrian Geology and Geochronology, Russian Academy
of Sciences, Makarova 2, 199034 St. Petersburg, Russia
vii
Judith L. Hannah Geological Survey of Norway, 7491 Trondheim, Norway / AIRIE Pro-
gram, Department of Geosciences, Colorado State University, Fort Collins, CO 80523-1482,
USA
Eero J. Hanski Department of Geosciences, University of Oulu, P.O. Box 3000, 90014 Oulu,
Finland
Christian J. Illing Institut f€ur Geologie und Pal€aontologie, Westf€alische Wilhelms-
Universit€at M€unster, Corrensstr. 24, 48149 M€unster, Germany
Stein B. Jacobsen Department of Earth and Planetary Sciences, Harvard University, 20
Oxford Street, Cambridge, MA 021 38, USA
Emmanuelle J. Javaux Department of Geology, University of Liege, 17 allee du 6 Aout
B18, 4000 Liege, Belgium
Lauri Joosu Department of Geology, Institute of Ecology and Earth Sciences, University of
Tartu, Ravila 14a, 50411 Tartu, Estonia
Kalle Kirsim€ae Department of Geology, Tartu University, Ravila 14A, 50411 Tartu, Estonia
Lee R. Kump Department of Geosciences, Pennsylvanian State University, 503 Deike
Building, University Park, PA 16870, USA
Anton B. Kuznetsov Institute of Precambrian Geology and Geochronology, Russian
Academy of Sciences, Makarova 2, 199034 St. Petersburg, Russia
Aivo Lepland Geological Survey of Norway, Postboks 6315 Sluppen, NO-7491 Trondheim,
Norway
Kevin Lepot Departement de Geologie, Palaeobiogeology-Palaeobotany-Palaeopalynology,
Universite de Liege, 4000 Liege, Belgium
Timothy W. Lyons Department of Earth Sciences, University of California, Riverside, CA
92521, USA
Adam P. Martin NERC Isotope Geosciences Laboratory (NIGL), Keyworth, Nottingham,
UK
Nicola McLoughlin Department of Earth Science and Centre for Geobiology, Allegaten 41,
Bergen N-5007, Norway
Pavel V. Medvedev Institute of Geology, Karelian Research Centre, Russian Academy of
Sciences, Pushkinskaya 11, 185910 Petrozavodsk, Russia
Victor A. Melezhik Geological Survey of Norway, Postboks 6315 Sluppen, NO-7491
Trondheim, Norway / Centre for Geobiology, University of Bergen, Allegaten 41, Bergen
N-5007, Norway
Dominic Papineau Department of Earth and Environmental Sciences, Boston College,
Devlin Hall 213, 140 Commonwealth avenue, Chestnut Hill, MA 02467, USA
Anthony R. Prave Department of Earth Science, University of St. Andrews, St Andrews
KY16 9AL, Scotland, UK
Christopher T. Reinhard Department of Earth Sciences, University of California, River-
side, CA 92521, USA
Marlene Reuschel Institut f€ur Geologie und Pal€aontologie, Westf€alische Wilhelms-
Universit€at M€unster, Corrensstrasse 24, 48149 M€unster, Germany
Alexander E. Romashkin Institute of Geology, Karelian Research Centre, Russian Acad-
emy of Sciences, Pushkinskaya 11, 185610 Petrozavodsk, Russia
viii Contributors to Three-olume Treatise
Olivier Rouxel IFREMER, Department of Ressources physiques et Ecosystemes de fond de
Mer, Technopole Brest-Iroise, 29280 Plouzane, France
Dmitry V. Rychanchik Institute of Geology, Karelian Research Centre, Russian Academy
of Sciences, Pushkinskaya 11, 185610 Petrozavodsk, Russia
Paula E. Salminen Department of Geosciences and Geography, University of Helsinki,
P.O. Box 64, (Gustaf H€allstr€omin katu 2a), 00014 Helsinki, Finland
Ronny Schoenberg Department for Geosciences, University of Tuebingen, Wilhelmstrasse
56, 72074 Tuebingen, Germany
Hubert Staudigel Scripps Institution of Oceanography, University of California, 0225 La
Jolla, San Diego, CA 92093-0225, USA
Holly J. Stein Geological Survey of Norway, 7491 Trondheim, Norway / AIRIE Program,
Department of Geosciences, Colorado State University, Fort Collins, CO 80523-1482, USA
Harald Strauss Institut f€ur Geologie und Pal€aontologie, Westf€alische Wilhelms-Universit€at,
Corrensstrasse 24, 48149 M€unster, Germany
Roger E. Summons Department of Earth, Atmospheric and Planetary Sciences,
Massachusetts Institute of Technology, 77 Massachusetts Avenue, E25-633 Cambridge, MA
02139, USA
Francois L.H. Tissot Origins Lab, Department of the Geophysical Sciences and Enrico
Fermi Institute, The University of Chicago, 5734 South Ellis Avenue, Chicago, IL 60637,
USA
Grant M. Young Department of Earth Sciences, University of Western Ontario, London
N6A 5B7, ON, Canada
Mark van Zuilen Institut de Physique du Globe de Paris, Equipe Geobiosphere Actuelle et
Primitive, 1 rue Jussieu, 75238 cedex 5 Paris, France
Contributors to Three-olume Treatise ix
.
Reviewers for Three-Volume Treatise:Reading the Archive of Earth’s Oxygenation
Prof. Ariel Anbar
Arizona State University, Tempe, AZ, USADr. Victor Balagansky
Geological Institute, Apatity, Russia
Prof. Mark Barley
University of Western Australia, Crawley, Australia
Prof. Jochen Brocks
The Australian National University, Canberra, AustraliaProf. Louis Derry
Cornell University, Ithaca, NY, USA
Prof. Anton Eisenhauer
GEOMAR, Kiel, Germany
Prof. Anthony E. Fallick
Scottish Universities Environmental Research Centre,East Kilbride, Scotland, UK
Prof. David Fike
Washington University, St. Louis, USAProf. Karl F€ollmi
University of Lausanne, Lausanne, Switzerland
Prof. Robert Frei
University of Copenhagen, Copenhagen, Denmark
Dr. Dieter Garbe-Sch€onberg
Christian-Albrechts-Universit€at, Kiel, GermanyProf. Eero Hanski
University of Oulu, Oulu, FinlandProf. Raimo Lahtinen
Geological Survey of Finland, Espoo, Finland
Prof. Michail Mints
Geological Institute, Moscow, Russia
Prof. David Mossman
Mount Allison University, Sackville, NB, CanadaProf. Richard Ojakangas
University of Minnesota, Duluth, USA
Prof. John Parnell
University of Aberdeen, Aberdeen, Scotland, UK
Prof. Adina Paytan
University of California, Santa Cruz, USADr. Bernhard Peucker-Ehrenbrink
Woods Hole Oceanographic Institution, Woods Hole, MA, USA
xi
Prof. Igor Puchtel
University of Maryland, College Park, MD, USA
Prof. Robert Raiswell
University of Leeds, Leeds, UKProf. Robert Riding
University of Tennessee, USA/
Cardiff University, Wales, UKDr. Nathaniel Sheldon
University of Michigan, Ann Arbor, MI, USA
Dr. Graham Shields-Zhou
University College, London, UK
Dr. Craig Storey
University of Portsmouth, Portsmouth, UKProf. Kari Strand
Thule Institute, University of Oulu, Oulu, Finland
Prof. Frances Westall
Centre de Biophysique Moleculaire, CNRS, Orleans, France
xii Reviewers for Three-Volume Treatise: Reading the Archive of Earth’s Oxygenation
Acknowledgements
The idea of making an atlas with comprehensive descriptions and illustrations of the
Palaeoproterozoic rocks from the Fennoscandian Shield was initiated in 2009 during a
workshop held in Trondheim, Norway, under the auspices of the International Continental
Scientific Drilling Program (ICDP). Starting from this workshop, a plan was developed and
finalised. Chris Bendall, Senior Editor for Springer, is acknowledged for encouragement and
editorial supervision of the project.
The three-volume set has three major underpinnings. The first is many years of research in
Precambrian geology of the Fennoscandian Shield by many workers, and we acknowledge
particularly the support of the Geological Survey of Norway; the University of Oulu, Finland;
and the Institute of Geology, Petrozavodsk, Russia.
The second is the unique core material obtained during the drilling operations by the
Fennoscandian Arctic Russia – Drilling Early Earth Project (FAR-DEEP). The drilling
operations were largely supported by the ICDP and by additional funding from several other
agencies and institutions. We are grateful for the financial support to the Norwegian Research
Council (NFR), the German Research Council (DFG), the National Science Foundation
(NSF), the NASA Astrobiology Institutes, the Geological Survey of Norway (NGU) and the
Centre of Excellence in Geobiology, the University of Bergen, Norway. The core archive and
associated analytical work were supported by NGU, the Scottish Universities Environmental
Research Centre (SUERC) and by the Pennsylvanian State University.
The third is a multidisciplinary approach to investigate complicated geological processes.
This was provided by the international scientific community and we acknowledge the support
of many universities in Scandinavia, Europe and the USA.
Many individuals helped in the preparation of the drilling operations and offered logistical
support. We are sincerely grateful to Anatoly Borisov (Kola Geological Information and
Laboratory Centre) for providing geological assistance for precise positioning of drillholes
in unexposed, boggy and forested terrains in the Imandra/Varzuga Greenstone Belt. Stanislav
Sokolov (Kola Mining Metallurgical Company) is acknowledged for logistic support and
geological guidance in the Pechenga Greenstone Belt. Logistical organisation of the drilling
operations and core transport across national borders by the State Company Mineral, St.
Petersburg, Russia, is appreciated. The Finnish company, SMOY, performed the drilling.
Many organisations and people have provided rock samples, photographs, SEM images and
permission to use figures. Thanks are due to the Geological Museum of the Department of
Geosciences, University of Oulu; Geological Museum of the Geological Institute, Kola
Science Centre, Apatity; Geological Survey of Finland; and to the following people: Wlady
Altermann, Lawrence Aspler, Alex Brasier, Ronald Conze, Alenka Crne, Kathleen Grey, Jens
Gutzmer, Eero Hanski, Emmanuelle Javaux, Yrj€o K€ahk€onen, Vadim Kamenetsky, Reino
Kesola, Andrew Knoll, Kauko Laajoki, Reijo Lampela, Aivo Lepland, Kevin Lepot, Zhen-
Yu Luo, Vladimir Makarikhin, Tuomo Manninen, Jukka Marmo, Nicola McLoughlin,
Pavel Medvedev, Victor Melezhik, Satu Mertanen, Tapani Mutanen, Lutz Nasdala,
xiii
Richard W. Ojakangas, Domenic Papineau, Petri Peltonen, Vesa Perttunen, Vladimir
Pozhilenko, Anthony Prave, Igor Puchtel, Jorma R€as€anen, Pentti Rastas, Raimo Ristim€aki,
Alexander Romashkin, Dmitry Rychanchik, Ronny Schoenberg, Evgeny Sharkov, Igor
Sokolov, Hubert Staudigel, Kari Strand, Sergey Svetov, Vladimir Voloshin, Frances Westall,
Grant Young, Valery Zlobin and Bouke Zwaan.
Unpublished geochemical data were kindly provided by the Geological Survey of Finland,
Zhorzh Fedotov, Valery Smol’kin and Peter Skuf’in. Permission to use published material was
kindly given by the Royal Society of Edinburgh, Elsevier, John Wiley and Sons and Blackwell
Publishing Ltd.
The book preparation was supported by the NFR (grant 191530/V30 to Victor Melezhik),
Natural Environment Research Council (grant NE/G00398X/1 to Anthony Fallick, Anthony
Prave and Daniel Condon), DFG (grants Str281/29, 32, 35 to Harald Strauss), NASA (grant
NNA09DA76A to Lee Kump), NSF (grant EAR 0704984 to Lee Kump) and the Academy of
Finland (grant 116845 to Eero Hanski).
To be embedded in the family of science always requires sacrifices such as time lost in
family contact. We wish to extend our gratitude to our families for patience, understanding
and constant encouragement.
Finally and most importantly, the editors wish to thank those colleagues and students who
will use and read these books or some parts of them. We hope that this will encourage them to
reach a more complete understanding of those processes that played an important role in the
irreversible modification of Earth’s surface environments and in shaping the face of our
emerging aerobic planet. We would also like to thank those scientists who will use the offered
advantage of rich illustrative material linked to the core collection to undertake new
research projects.
xiv Acknowledgements
Preface to Volume 3
Earth’s present-day environments are the outcome of a 4.5-billion-year period of evolution
reflecting the interaction of global-scale geological and biological processes. Punctuating that
evolution were several extraordinary events and episodes that perturbed the entire Earth
system and led to the creation of new environmental conditions, sometimes even to funda-
mental changes in how planet Earth operated. One of the earliest and arguably the greatest of
these events was a substantial increase (orders of magnitude) in the atmospheric oxygen
abundance, sometimes referred to as the Great Oxidation Event. Given our present knowledge,
this oxygenation of the terrestrial atmosphere and the surface ocean, during the Palaeopro-
terozoic Era between 2.4 and 2.0 billion years ago, irreversibly changed the course of Earth’s
evolution. Understanding why and how it happened and what its consequences were are
among the most challenging problems in Earth sciences.
The three-volume treatise entitled “Reading the Archive of Earth’s Oxygenation” (1) provides
a comprehensive review of the Palaeoproterozoic Eon with an emphasis on the Fennoscandian
Shield geology; (2) serves as an initial report of the preliminary analysis of one of the finest
lithological and geochemical archives of early Palaeoproterozoic Earth history, created
under the auspices of the International Continental Scientific Drilling Programme (ICDP); (3)
synthesises the current state of our understanding of aspects of early Palaeoproterozoic events
coincident with and likely related to Earth’s progressive oxygenation with an emphasis on still-
unresolved problems that are ripe for and to be addressed by future research. Combining this
information in three coherent volumes offers an unprecedented cohesive and comprehensive
elucidation of the Great Oxidation Event and related global upheavals that eventually led to the
emergence of the modern aerobic Earth System.
The format of these books centres on high-quality photo-documentation of Fennoscandian
Arctic Russia – Drilling Early Earth Project (FAR-DEEP) cores and natural exposures of the
Palaeoproterozoic rocks of the Fennoscandian Shield. The photos are linked to geochemical
data sets, summary figures and maps, time-slice reconstructions of basinal and palaeoenvir-
onmental settings that document the response of the Earth system to the Great Oxidation
Event. The emphasis on a thorough, well-illustrated characterisation of rocks reflects the
importance of sedimentary and volcanic structures that form a basis for interpreting ancient
depositional environments, and chemical, physical and biological processes operating on
Earth’s surface. Most of the structural features are sufficiently complex as to challenge the
description by other than a visual representation, and high-quality photographs are themselves
a primary resource for presenting essential information. Although nothing can replace the
wealth of information that a geologist can obtain from examining an outcrop first hand, the
utility of photographs offers the next best source of data for assessing and evaluating
palaeoenvironmental reconstructions. This three-volume treatise will, thus, act as an informa-
tion source and guide to other researchers and help them identify and interpret such features
elsewhere, and will serve as an illustrated guidebook to the Precambrian for geology students.
xv
Finally, the three-volume treatise provides a link to the FAR-DEEP core collection archived
at the Geological Survey of Norway. These drillcores are a unique resource that can be used to
solve the outstanding problems in understanding the causes and consequences of the multiple
processes associated with the progressive oxygenation of terrestrial environments. It is
anticipated that the well-archived core will provide the geological foundation for future research
aimed at testing and generating new ideas about the Palaeoproterozoic Earth. The three-volume
treatise will be of interest to researchers involved directly in studying this hallmark period in
Earth history, as well as professionals and students interested in Earth System evolution in
general.
Volume 3: “Global Events and the Fennoscandian Arctic Russia – Drilling Earth Project”
represents another kind of illustrated journey through the early Palaeoproterozoic, provided by
syntheses, reviews and summaries of the current state of our understanding of a series of global
events that resulted in a fundamental change of the Earth System from an anoxic to an
oxic state. The book discusses traces of life and possible causes for the Huronian-age
glaciations; addresses radical changes in carbon, sulphur and phosphorus cycles during the
Palaeoproterozoic; and provides a comprehensive description and a rich photo-documentation
of the early Palaeoproterozoic supergiant, petrified oil-field. Terrestrial environments are
characterised through a critical review of available data on weathered and calcified surfaces
and travertine deposits. Potential implementation of Ca, Mg, Sr, Fe, Mo, U and Re-Os isotope
systems for deciphering Palaeoproterozoic seawater chemistry and a change in the redox state
of water and sedimentary columns are discussed. The volume considers in detail the definition
of the oxic atmosphere, possible causes for the oxygen rise, and considers the oxidation of
terrestrial environment not as a single event but a slow-motion process lasting over hundreds
of millions of years. Finally, the book provides a roadmap as to how the FAR-DEEP cores
may facilitate future interesting science and provide a new foundation for education in
earth-science community.
Welcome to the illustrative journey through one of the most exciting periods of planet
Earth!
xvi Preface to Volume 3
Contents to Volume 1
Part I Palaeoproterozoic Earth
1.1 Tectonic Evolution and Major Global Earth-Surface
Palaeoenvironmental Events in the Palaeoproterozoic . . . . . . . . . . . . . . . . . . 3
V.A. Melezhik, L.R. Kump, E.J. Hanski, A.E. Fallick, and A.R. Prave
Part II The Fennoscandian Arctic Russia: Drilling Early Earth
Project (FAR-DEEP)
2.1 The International Continental Scientific Drilling Program . . . . . . . . . . . . . . . 25
Victor A. Melezhik
Part III Fennoscandia: The First 500 Million Years of the Palaeoproterozoic
3.1 The Early Palaeoproterozoic of Fennoscandia: Geological
and Tectonic Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
V.A. Melezhik and E.J. Hanski
3.2 Litho- and Chronostratigraphy of the Palaeoproterozoic
Karelian Formations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
E.J. Hanski and V.A. Melezhik
3.3 Palaeotectonic and Palaeogeographic Evolution of Fennoscandia
in the Early Palaeoproterozoic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
V.A. Melezhik and E.J. Hanski
3.4 Evolution of the Palaeoproterozoic (2.50–1.95 Ga) Non-orogenic
Magmatism in the Eastern Part of the Fennoscandian Shield . . . . . . . . . . . . . 179
E.J. Hanski
Part IV Geology of the Drilling Sites
4.1 The Imandra/Varzuga Greenstone Belt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
V.A. Melezhik
4.2 The Pechenga Greenstone Belt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
V.A. Melezhik and E.J. Hanski
4.3 The Onega Basin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387
V.A. Melezhik, P.V. Medvedev, and S.A. Svetov
xvii
.
Contents to Volume 2
Part V FAR-DEEP Core Archive and Database
5.1 FAR-DEEP Core Archive and Database . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 493
A. Lepland, M. Mesli, R. Conze, K. Fabian, A.E. Fallick, and L.R. Kump
Part VI FAR-DEEP Core Descriptions and Rock Atlas
6.1 The Imandra/Varzuga Greenstone Belt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 505
V.A. Melezhik
6.1.1 Seidorechka Sedimentary Formation: FAR-DEEP Hole 1A
and Neighbouring Quarries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 510
V.A. Melezhik, A.R. Prave, A. Lepland, E.J. Hanski, A.E. Romashkin,
D.V. Rychanchik, Zh.-Y. Luo, E.V. Sharkov, and M.M. Bogina
6.1.2 Polisarka Sedimentary Formation: FAR-DEEP Hole 3A . . . . . . . . . . . . . . . 530
V.A. Melezhik, E.J. Hanski, A.R. Prave, A. Lepland, A.E. Romashkin,
D.V. Rychanchik, A.T. Brasier, A.E. Fallick, Zh.-Y. Luo,
E.V. Sharkov, and M.M. Bogina
6.1.3 Umba Sedimentary Formation: FAR-DEEP Hole 4A . . . . . . . . . . . . . . . . . . 551
V.A. Melezhik, E.J. Hanski, A.R. Prave, A. Lepland, A.E. Romashkin,
D.V. Rychanchik, Zh.-Y. Luo, E.V. Sharkov, and M.M. Bogina
6.1.4 Umba Sedimentary Formation: Sukhoj Section . . . . . . . . . . . . . . . . . . . . . . 567
V.A. Melezhik
6.2 The Pechenga Greenstone Belt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 591
V.A. Melezhik
6.2.1 The Neverskrukk Formation: Drillholes 3462, 3463 and Related
Outcrops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 595
V.A. Melezhik, A.E. Fallick, A.R. Prave, and A. Lepland
6.2.2 Kuetsjarvi Sedimentary Formation: FAR-DEEP Hole 5A, Neighbouring
Quarry and Related Outcrops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 617
P.E. Salminen, V.A. Melezhik, E.J. Hanski, A. Lepland, A.E. Romashkin,
D.V. Rychanchik, Zh.-Y. Luo, E.V. Sharkov, and M.M. Bogina
6.2.3 Kuetsjarvi Volcanic Formation: FAR-DEEP Hole 6A
and Related Outcrops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 650
E.J. Hanski, V.A. Melezhik, A. Lepland, A.E. Romashkin, D.V. Rychanchik,
Zh.-Y. Luo, E.V. Sharkov, and M.M. Bogina
xix
6.2.4 Kolosjoki Sedimentary and Kuetsjarvi Volcanic Formations: FAR-DEEP
Hole 7A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 678
E.J. Hanski, V.A. Melezhik, A. Lepland, A.E. Romashkin, D.V. Rychanchik,
Zh.-Y. Luo, E.V. Sharkov, and M.M. Bogina
6.2.5 Kolosjoki Sedimentary Formation: FAR-DEEP Holes 8A and 8B
and Related Outcrops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 693
V.A. Melezhik, A.R. Prave, A. Lepland, E.J. Hanski, A.E. Romashkin,
D.V. Rychanchik, and Zh.-Y. Luo
6.2.6 Kolosjoki Volcanic Formation: FAR-DEEP Hole 9A . . . . . . . . . . . . . . . . . 758
E.J. Hanski, V.A. Melezhik, A. Lepland, A.E. Romashkin, D.V. Rychanchik,
Zh.-Y. Luo, E.V. Sharkov, and M.M. Bogina
6.3 The Onega Basin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 769
V.A. Melezhik
6.3.1 Tulomozero Formation: FAR-DEEP Holes 10A and 10B . . . . . . . . . . . . . . 773
V.A. Melezhik, A.R. Prave, A.T. Brasier, A. Lepland, A.E. Romashkin,
D.V. Rychanchik, E.J. Hanski, A.E. Fallick, and P.V. Medvedev
6.3.2 Tulomozero Formation: FAR-DEEP Hole 11A . . . . . . . . . . . . . . . . . . . . . . 889
V.A. Melezhik, A.R. Prave, A. Lepland, A.E. Romashkin,
D.V. Rychanchik, and E.J. Hanski
6.3.3 Zaonega Formation: FAR-DEEP Holes 12A and 12B, and Neighbouring
Quarries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 946
A.E. Crne, V.A. Melezhik, A.R. Prave, A. Lepland, A.E. Romashkin,
D.V. Rychanchik, E.J. Hanski, and Zh.-Y. Luo
6.3.4 Zaonega Formation: FAR-DEEP Hole 13A . . . . . . . . . . . . . . . . . . . . . . . . . 1008
A.E. Crne, V.A. Melezhik, A.R. Prave, A. Lepland,
A.E. Romashkin, D.V. Rychanchik, E.J. Hanski, and Zh.-Y. Luo
.
xx Contents to Volume 2
Contents to Volume 3
Part VII Earth’s Oxygenation and Associated Global Events:
The FAR-DEEP Perspective
7.1 The End of Mass-Independent Fractionation of Sulphur Isotopes . . . . . . . . . . 1049
M. Reuschel, H. Strauss, and A. Lepland
7.2 Huronian-Age Glaciation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1059
V.A. Melezhik, G.M Young, P.G. Eriksson, W. Altermann, L.R. Kump,
and A. Lepland
7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle:
The Lomagundi-Jatuli Isotopic Event . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1111
V.A. Melezhik, A.E. Fallick, A.P. Martin, D.J. Condon, L.R. Kump,
A.T. Brasier, and P.E. Salminen
7.4 An Apparent Oxidation of the Upper Mantle versus Regional Deep
Oxidation of Terrestrial Surfaces in the Fennoscandian Shield . . . . . . . . . . . . 1151
K.S. Rybacki, L.R. Kump, E.J. Hanski, and V.A. Melezhik
7.5 Abundant Marine Calcium Sulphates: Radical Change of Seawater
Sulphate Reservoir and Sulphur Cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1169
H. Strauss, V.A. Melezhik, M. Reuschel, A.E. Fallick, A. Lepland,
and D.V. Rychanchik
7.6 Enhanced Accumulation of Organic Matter: The Shunga Event . . . . . . . . . . . 1195
H. Strauss, V.A. Melezhik, A. Lepland, A.E. Fallick, E.J. Hanski,
M.M. Filippov, Y.E. Deines, C.J. Illing, A.E. Crne, and A.T. Brasier
7.7 The Earliest Phosphorites: Radical Change in the Phosphorus
Cycle During the Palaeoproterozoic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1275
A. Lepland, V.A. Melezhik, D. Papineau, A.E. Romashkin,
and L. Joosu
7.8 Traces of Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1297
7.8.1 Introductory Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1297
A. Lepland
7.8.2 Palaeoproterozoic Stromatolites from the Lomagundi-Jatuli Interval
of the Fennoscandian Shield . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1298
N. McLoughlin, V.A. Melezhik, A.T. Brasier, and P.V. Medvedev
7.8.3 Palaeoproterozoic Microfossils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1352
E.J. Javaux, K. Lepot, M. van Zuilen, V.A. Melezhik, and P.V. Medvedev
xxi
7.8.4 Seeking Textural Evidence of a Palaeoproterozoic Sub-seafloor
Biosphere in Pillow Lavas of the Pechenga Greenstone Belt . . . . . . . . . . . . 1371
N. McLoughlin, H. Furnes, E.J. Hanski, and H. Staudigel
7.8.5. Biomarkers and Isotopic Tracers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1395
R.E. Summons, C.J. Illing., M. van Zuilen, and H. Strauss
7.9 Terrestrial Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1407
7.9.1 Introductory Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1407
L.R. Kump
7.9.2 Palaeoproterozoic Weathered Surfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . 1409
K. Kirsimae and V.A. Melezhik
7.9.3 Caliche . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1419
A.T. Brasier, V.A. Melezhik, and A.E. Fallick
7.9.4 Earth’s Earliest Travertines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1435
A.T. Brasier, P.E. Salminen, V.A. Melezhik, and A.E. Fallick
7.10 Chemical Characteristics of Sediments and Seawater . . . . . . . . . . . . . . . . . . 1457
7.10.1 Introductory Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1457
L.R. Kump
7.10.2 Sr Isotopes in Sedimentary Carbonates . . . . . . . . . . . . . . . . . . . . . . . . . 1459
A.B. Kuznetsov, I.M. Gorokhov, and V.A. Melezhik
7.10.3 Ca and Mg Isotopes in Sedimentary Carbonates . . . . . . . . . . . . . . . . . . . 1468
J. Farkas, R. Chakrabarti, S.B. Jacobsen, L.R. Kump, and V.A. Melezhik
7.10.4 Iron Speciation and Isotope Perspectives on Palaeoproterozoic
Water Column Chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1483
C.T. Reinhard, T.W. Lyons, O. Rouxel, D. Asael, N. Dauphas,
and L.R. Kump
7.10.5 Cr Isotopes in Near Surface Chemical Sediments . . . . . . . . . . . . . . . . . . 1493
M. van Zuilen and R. Schoenberg
7.10.6 Mo and U Geochemistry and Isotopes . . . . . . . . . . . . . . . . . . . . . . . . . . 1500
F.L.H. Tissot, N. Dauphas, C.T. Reinhard, T.W. Lyons, D. Asael,
and O. Rouxel
7.10.7 Re-Os Isotope Geochemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1506
J.L. Hannah and H.J. Stein
Part VIII The Great Oxidation Event: State of the Art and Major
Unresolved Problems
8.1 The Great Oxidation Event . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1517
L.R. Kump, A.E. Fallick, V.A. Melezhik, H. Strauss, and A. Lepland
Part IX FAR-DEEP Core Archive: Future Opportunities for Geoscience Research
and Education
FAR-DEEP Core Archive: Further Opportunities for Earth Science Research
and Education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1537
V.A. Melezhik and A.R. Prave
xxii Contents to Volume 3
Part VII
Earth’s Oxygenation and Associated Global Events:The FAR-DEEP Perspective
7.1 The End of Mass-Independent Fractionationof Sulphur Isotopes
M. Reuschel, H. Strauss, and A. Lepland
7.1.1 Introduction
The Archaean-Proterozoic transition is marked by a number
of fundamental upheavals in respect to geological, tectonic,
geochemical, biological and climatic aspects. Of these, the
most significant change appears to be a substantial increase
in atmospheric oxygen concentration initiating the irrevers-
ible oxygenation of our planet. It has been proposed that a
major oxygenation event occurred during the early
Palaeoproterozoic some 2.3 Ga ago, widely termed the
“Great Oxidation Event” (“GOE”, Holland 1999, 2006).
Evidence for this generally accepted view (but see Ohmoto
1999; Ohmoto et al. 2006, for a different view) stems from
geological, mineralogical and geochemical data. Of these,
the study of multiple sulphur isotopes, i.e. the analysis of all
four stable isotopes of sulphur (32S, 33S, 34S and 36S) devel-
oped recently into the central approach for reconstructing the
chemical composition of Earth’s early atmosphere, and sec-
ular variations thereof. Specifically, it has been suggested
that mass-independently fractionated sulphur isotopes,
archived in sediments of Archaean and early Palaeopro-
terozoic age, provide a reliable tool for reconstructing past
atmospheric oxygen concentrations (Farquhar et al. 2000;
Pavlov and Kasting 2002; Ueno et al. 2009).
The time interval archived in the FAR-DEEP drill cores
straddles this crucial time of Earth’s initial oxygenation.
This chapter reviews the temporal record of mass-
independently fractionated sulphur as a proxy for atmo-
spheric oxygen. We will proceed by briefly introducing the
relevant systematics of stable isotope geochemistry includ-
ing the analysis of multiple sulphur isotopes. This will lead
into a discussion about the implications for reconstructing
the temporal evolution of atmospheric oxygen abundance.
Finally, the FAR-DEEP rock record that archives the termi-
nation of mass-independently fractionated sulphur isotopes
will be introduced.
7.1.2 Multiple Sulphur Isotope Systematics
The geochemistry of light stable isotopes (i.e. H, C, N, O, S)
has witnessed a long history of applications in earth and life
sciences (e.g. Hoefs 2009). For decades, variations in the
stable isotopic composition of geomaterials have been
utilised for reconstructing environmental conditions and/or
for tracing physical, chemical or biological processes in the
low- and high-temperature realm. Most frequently, the ratio
of a rare over a major stable isotope was considered. Given
their natural abundances, the two stable sulphur isotopes 34S
(natural abundance of 4.21 %) and 32S (natural abundance of
95.01 %) are generally considered with results reported in
the so-called delta notation:
d34S ‰;V� CDT½ � ¼ 34S=32Ssa�34S=32Sstd
� �=34S=32Sstd
h i
� 1;000
(1)
Variations in isotopic composition are based on the fact
that two different isotopes of the same element (here 34S and32S) react differently during a given reaction/process. This is
a consequence of different numbers of neutrons for a given
number of protons, expressed in their differing isotope mass.
The observation that the isotopic composition of a given
element prior to (reactant) and after (product) a given reac-
tion is different is termed isotopic fractionation.
Fractionation of stable sulphur isotopes is associated with
inorganic as well as biologically mediated reactions and
reflects thermodynamic equilibrium and/or kinetic isotope
effects (for a recent review, see e.g. Canfield 2001). In
principle, isotopic fractionation is a consequence of the
relative mass difference between the stable isotopes,
H. Strauss (*)
Institut f€ur Geologie und Pal€aontologie, Westf€alische Wilhelms-
Universit€at M€unster, Corrensstrasse 24, 48149 M€unster, Germany
e-mail: [email protected]
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_1, # Springer-Verlag Berlin Heidelberg 2013
1049
resulting in different behaviour during physical, chemical
and/or biological processes. For sulphur, the mass difference
between 32S and 33S is 1 amu (atomic mass unit), which is
half of the mass difference between 32S and 34S (2 amu).
Consequently, mass dependency dictates the following frac-
tionation relationships between the four different stable sul-
phur isotopes:
d33S ¼ 0:515 � d34S (2)
and
d36S ¼ 1:9� d34S: (3)
Empirical and experimental observations resulted in the
general notion that all reactions/processes occurring on
Earth, whether physical, chemical or biological, result in
mass-dependent isotopic fractionation (MDF). As a conse-
quence, researchers have concentrated for decades on deter-
mining the d34S value of a given sulphur-bearing compound,
accepting that mass-dependency would allow calculation of
the other isotope ratios (if needed).
In strong contrast to these observations in terrestrial
materials, it was discovered early on that extraterrestrial
materials are characterised by a clearly different behaviour
in isotopic fractionation and exhibit so-called mass-
independently fractionated isotopes (e.g. sulphur: Hulston
and Thode 1965; oxygen: Clayton et al. 1977). In addition,
Thiemens and Heidenreich (1983) observed the existence of
similar mass-independent (oxygen) isotope fractionation
during ozone production reflecting photochemically induced
processes in Earth’s present atmosphere.
Following this reasoning,mass-independent fractionation
of sulphur isotopes (MIF-S) means that:
d33S 6¼ 0:515� d34S (4)
and
d36S 6¼ 1:9� d34S: (5)
Renewed interest in the study of multiple sulphur isotopes
and the application to geological and biological questions
required the formulation of a (now commonly excepted) way
for reporting results, notably as D33S (respectively D36S).
This term quantifies the deviation of a measured d33S (d36S)value from the calculated d33S (d36S) value if mass-
dependent fractionation (MDF) would have happened
(Farquhar et al. 2000):
D33S ¼ d33S� 1;000
� 1þ d34S=1;000� �0:515 � 1
� �(6)
and:
D36S ¼ d36S� 1;000 � 1þ d34S=1;000� �1:9 � 1
� �: (7)
In general, studies on modern geo- and biomaterials
revealed that variations in D33S of � 0.3 ‰ are attributed
to mass-dependent sulphur isotope fractionation (given an
external precision for D33S of � 0.008 ‰, Zerkle et al.
2009, 2010). In contrast, deviations from this array (i.e. a
D33S that is larger than � 0.3 ‰) reflect the presence of
mass-independent fractionation processes. Currently, no
threshold value has been agreed for D36S, which allows a
distinction between MDF or MIF-S as known for D33S, but
we note that external precision for D36S is ~0.3 ‰ (e.g. Bao
et al. 2007).
7.1.3 The Multiple Sulphur Isotope Recordof Precambrian Sedimentary Rocks
For decades, our understanding of the Precambrian global
sulphur cycle was based on temporal records of the tradi-
tional d34S value measured in sedimentary sulphates and
sulphides (for reviews, see e.g. Strauss 2002, 2004; Lyons
et al. 2004; Kah et al. 2004; Canfield 2004). These records
suggest a low-sulphate Precambrian ocean (e.g. Lyons and
Gill 2010), a non-linear increase in d34Ssulphate, as a conse-
quence of the growing importance of biological sulphur
cycling in the sedimentary realm (e.g. Strauss 2004; Guo
et al. 2009; Lyons and Gill 2010), and the onset of bacterial
sulphate reduction in the early Palaeoproterozoic (e.g.
Strauss 2002) or in the Neoarchaean (e.g. Grassineau et al.
2001) or possibly as early as in the Palaeoarchaean (e.g.
Shen et al. 2009). These conclusions are based on two simple
observations, notably (1) a sizeable difference in the sulphur
isotopic composition between (reconstructed) seawater
sulphate and sedimentary sulphide, and (2) a substantial
deviation in d34S from the crustal average value. Given the
paucity of preserved sulphate occurrences in the Precam-
brian sedimentary record (see Chap. 7.5), the interpretation
in respect to early sulphur cycling is largely based on the
sedimentary sulphide record. Here, special emphasis is
placed on detecting the activity of sulphur-utilising
microbes, specifically bacterial sulphate reduction. Today,
this form of microbial sulphur cycling is associated with
substantial isotope fractionation. Sulphate reducing bacteria
preferentially metabolise 32SO4 causing a proposed maxi-
mum fractionation of 45 ‰ between seawater sulphate and
hydrogen sulphide (Canfield 2001). Pyrite, which is formed
if reduced iron is available, is depleted in 34S and its burial
leads to an enrichment of 34S in the seawater sulphate pool.
The Precambrian record of sulphide d34S is characterised by
1050 M. Reuschel et al.
near zero d34S values during the Archaean (with few
exceptions, e.g. Grassineau et al. 2001; Philippot et al.
2007; Shen et al. 2009), indicating the apparent absence of
microbial turnover of sulphate. Unquestionable evidence for
bacterial sulphate reduction, based on high-magnitude iso-
tope fractionation, with strongly negative d34S as low as
�34.7 ‰ (Bekker et al. 2004) at seawater sulphate sulphur
isotopic compositions between +10 and +20 ‰ (Schr€oder
et al. 2008; Guo et al. 2009) does not occur in the rock record
before 2.3 Ga. This increase in microbially induced sulphur
isotope fractionation between parent seawater sulphate and
produced hydrogen sulphide (deposited as pyrite) is
suggested to be the result of increasing oceanic sulphate
concentrations, connected to the rise in atmospheric oxygen
concentration and subsequent onset of oxidative weathering
of sulphides.
The discovery of MIF-S in terrestrial sedimentary
sulphides and sulphates of Precambrian age by Farquhar
et al. (2000) strongly stimulated the field of stable sulphur
isotope geochemistry. Multiple sulphur isotope studies over
the past 10 years (e.g. Farquhar et al. 2000, 2007; Mojzsis
et al. 2003; Ono et al. 2003, 2009a, b; Ohmoto et al. 2006;
Guo et al. 2009; Thomazo et al. 2009) revealed a distinct
MIF-S signature in sedimentary sulphides and sulphates of
Archaean and early Palaeoproterozoic age (Fig. 7.1). During
this time the D33S values display a spread between �2.5 and
+11.2 ‰ with clearly discernible temporal variations in the
range of D33S values. While the Eoarchaean and
Palaeoarchaean sediments yielded D33S values between
�1.7 and +6.1 ‰, the subsequent Mesoarchaean time inter-
val between 3.2 and 2.7 Ga displays only minimal anomalies
in D33S for sedimentary sulphides. In strong contrast, high-
magnitude fractionations between �2.5 and +11.2 ‰ char-
acterise the MIF-S record for the Neoarchaean and early
Palaeoproterozoic. Finally, the distinct MIF-S signature
disappears from the sedimentary rock record in the early
Palaeoproterozoic between 2.3 and 2.5 Ga ago (e.g. Guo
et al. 2009). Younger Proterozoic and Phanerozoic rocks
through to the modern world display exclusively mass-
dependently fractionated sulphur isotope values.
MIF-S is thought to originate from photochemical
reactions of sulphur dioxide, induced by short-wave UV
rays in an oxygen-free atmosphere (e.g. Farquhar et al.
2001). Under the modern oxidising atmospheric conditions,
sulphur is only present as aerosol of sulphuric acid with a
homogenous sulphur isotopic composition. However, given
a chemically reducing atmosphere as suggested for the
Archaean and early Palaeoproterozoic (prior to the GOE),
isotopically different sulphur species (such as elemental
sulphur and aerosols of sulphate) would have been
generated, preserved, and subsequently transferred to
Earth’s surface (Kasting et al. 1989; Pavlov and Kasting
2002; Ono et al. 2003). This way, distinct MIF-S signatures
were archived in the ancient sedimentary rock record as
sulphate and sulphide.
Following the early observations by Farquhar et al.
(2000) and subsequent experimental work (Farquhar et al.
2001), different atmospheric sulphur-bearing phases are
thought to carry distinct MIF-S signals. Elemental sulphur
is regarded as the main carrier of a positive D33S signature
whereas sulphate aerosols are thought to be the carrier for a
corresponding negative D33S signal (e.g. Farquhar and Wing
2003; Ono et al. 2003). However, modelling by Ueno et al.
(2009) revealed that sulphate aerosols might also acquire
positive D33S values. In their model, all SO2 that enters the
atmosphere is converted into carbonyl sulphide (OCS). This
would lead to high levels (>1 ppm) of atmospheric OCS
given a volcanic sulphur flux that is three times higher than
today (Bluth et al. 1993; Ueno et al. 2009). Due to the high
UV shielding effect of OCS, the sulphate acquires a negative
D33S signature, a feature consistent with MIF-S data for the
Archaean sulphate occurrences. A decrease in the volcanic
sulphur flux could, hence, explain the positive D33S recorded
in carbonate-associated sulphate (CAS; for a detailed
description about CAS as palaeoproxy see Chap. 7.5) from
the Transvaal Supergroup (South Africa) and Hamersley
Basin (Western Australia) (Guo et al. 2009; Domagal-
Goldman et al. 2008).
Considering the different magnitudes of the MIF-S signal
and the apparent temporal variation of it, the MIF-S record
has been interpreted to reflect differences in the chemical
composition of the atmosphere. Most conspicuous in that
respect are sedimentary sulphides of Mesoarchaean age
between 3.2 and 2.7 Ga that display attenuated D33S values
only minimally above the threshold for non-zero D33S, i.e.
0 � 0.3 ‰. Respective data have been interpreted to reflect
an early oxic atmosphere (Ohmoto et al. 2006). However,
subsequent studies regard even these greatly attenuated D33S
values to reflect MIF-S (Farquhar et al. 2007; Domagal-
Goldman et al. 2008; Thomazo et al. 2009). This interpreta-
tion is based on considering D33S and D36S relationships. On
a plot of D36S versus D33S, the Archaean and earliest
Palaeoproterozoic samples plot along a regression line with
a slope of around �1. This is quite different for Phanerozoic
sulphides, which define a slope varying between �4.4 and
�9.8 in a D36S/D33S diagram, with an average slope of D36S/
D33S ¼ �6.9 (Fig. 7.2, Ono et al. 2006). The fact that
observational data agree with respective multiple sulphur
isotope results from experimental studies performed under
simulated anoxic atmospheric conditions (Farquhar et al.
2001) indicates the significance of D36S/D33S relationships
in identifying MIF-S even when D33S is substantially
minimised (Farquhar et al. 2007; Ono et al. 2006). Conse-
quently, low yet clearly non-zero D33S values cannot be
considered as evidence for an early oxygenation of Earth’s
atmosphere.
1 7.1 The End of Mass-Independent Fractionation of Sulphur Isotopes 1051
Fig. 7.1 (a) Temporal evolution of D33S recorded in sedimentary
sulphides and sulphates (Source of data: Domagal-Goldman et al.
2008; Guo et al. 2009; Ono et al. 2009a,b; Shen et al. 2009; Thomazo
et al. 2009; Johnston et al. 2005, 2006, 2008; Partridge et al. 2008; Bao
et al. 2007; Farquhar et al. 2007; Hou et al. 2007; Kamber and
Whitehouse 2007; Kaufman et al. 2007; Papineau et al. 2005, 2007;
Papineau and Mojzsis 2006; Philippot et al. 2007; Cates and Mojszis
2006; Jamiesson et al. 2006; Ohmoto et al. 2006; Whitehouse et al.
2005; Hu et al. 2003; Mojzsis et al. 2003; Ono et al. 2003; Farquhar
et al. 2002, 2000). (b) Variations in D33S within the Huronian Super-
group after Papineau et al. (2007); Formation names are listed in
Fig. 7.3. (c)Temporal variations in D33S of carbonate-associated
sulphates and sulphides of the Duitschland Formation (Transvaal
Supergroup/South Africa) after Guo et al. (2009)
1052 M. Reuschel et al.
In contrast, the low D33S signal in the Mesoarchaean
samples is viewed to reflect either changing atmospheric
chemistry (i.e. different mechanism for MIF production) or
the occurrence of an organic haze layer in the troposphere
due to the activity of methanotrophs that prevents deep UV
penetration of the atmosphere (Domagal-Goldman et al.
2008; Thomazo et al. 2009). A shield of organic haze, in
addition to preventing the SO2 photochemistry of short-
wave UV rays, may also lead to a global cooling of the
Earth’s surface. Diamictites within the 2.9 Ga South African
Witwatersrand Supergroup and faceted clasts within the
coeval Pongola Supergroup are thought to be of glacial
origin (Young 1988; Crowell 1999), which would yield a
consistent environmental picture if the glaciation was of
global nature. The disappearance of the organic haze and
thus a return to greenhouse climate conditions could have
led to the final rise of the variations in D33S (Ono et al.
2009a, b; Kaufman et al. 2007) until the large magnitude
MIF-S disappears seemingly abruptly after ~2.4 Ga
(Farquhar et al. 2000; Farquhar and Wing 2003; Papineau
et al. 2005).
Neoarchaean sulphides from the Transvaal Supergroup
and the Hamersley Basin display the largest variation in
D33S throughout the Archaean and include the highest posi-
tive anomaly with a D33S value of 11.2 ‰ (Kaufman et al.
2007). Zahnle et al. (2006) argue that a decrease in the
atmospheric methane concentration could have lead to the
decrease in elemental sulphur production and this would
explain subsequent vanishing of the highly positive D33S
anomalies. Within temporal resolution, the largest docu-
mented positive D33S signal (Kaufman et al. 2007) is
followed by a sharp decline and ultimate loss of the MIF-S
signature in the sedimentary rock record. This change from
the mass-independent signal to solely mass-dependent frac-
tionation of sulphur isotopes is associated with the rise in
atmospheric oxygen above 10�5 PAL (present atmospheric
level, Pavlov and Kasting 2002). Few systematic studies
provide a record of this temporal window between 2.50
and 2.35 Ga ago, but most prominent are those of the
Huronian Supergroup in North America (between 2.5 and
2.2 Ga) and the Transvaal Supergroup in South Africa
(between 2.4 and 2.3 Ga).
The Palaeoproterozoic Huronian Supergroup represents a
more than 10-km-thick volcano-sedimentary succession that
is presently best exposed north of Lake Huron, Ontario,
Canada. The age of the succession is constrained between
2.49 and 2.45 Ga (basal Copper Cliff Rhyolite; Krogh et al.
1984) and 2.219 Ga (intrusive Nipissing diabase; Corfu and
Andrews 1986). The Huronian Supergroup comprises three
stratigraphic units containing sedimentary rocks of glacial
origin (from bottom to top: Ramsay Lake Formation, Bruce
Formation, Gowganda Formation), separated by respective
interglacial units (for a detailed description see Chap. 7.2).
Ion microprobe multiple sulphur isotope analyses for
sulphides from the Huronian Supergroup were presented in
Papineau et al. (2005, 2007). The characterization of
authigenic sedimentary or hydrothermal sulphides, and
detrital pyrite, and their mass-independent as well as mass-
dependent isotopic fractionation patterns have been used in
these studies to reconstruct the evolution of atmospheric
oxygen abundance, and to trace the activity of sulphate-
reducing bacteria within the depositional environment.
Most notably, the record of mass-independently fractionated
sulphur isotopes indicates that oxygen levels increased irre-
versibly in the aftermath of the Huronian glaciations
(Fig. 7.3). The sedimentary rocks of the Pecors Formation
that post-date the Ramsay Lake glacials and represent the
first interglacial interval still show small magnitude MIF-S.
In contrast, the strata above the second glacial (Bruce
Formation) level only show large-range mass-dependent
fractionation (Papineau et al. 2007). This indicates that
atmospheric oxygen levels increased enough to prevent fur-
ther mass-independent sulphur isotope fractionation. Increas-
ing atmospheric oxygen levels would have triggered oxidative
continental weathering, enhancing weathering rates and likely
delivering more nutrients to the ocean, which would have
Fig. 7.2 Generalised plot for D36S and D33S relationships. The blue
line shows the regression line observed for Palaeo-, Meso-, and
Neoproterozoic and Phanerozoic sulphur species with an average
slope of �6.9 (MDF ¼ mass-dependent sulphur isotope fractionation;
variations in the slope between �4.4 and �9.9 have been observed;
Ono et al. 2006, 2007; Johnston et al. 2006). Archaean and earliest
Palaeoproterozoic sulphur species are typically plotting on a regression
line with a slope of �1 (MIF-S ¼ mass-independent fractionated sul-
phur isotopes; Farquhar et al. 2001; Ono et al. 2003, and references of
Fig. 7.1), shown by the red line
1 7.1 The End of Mass-Independent Fractionation of Sulphur Isotopes 1053
further stimulated oxygenic photosynthesis. Further, such
weathering would have also activated the oxidative decompo-
sition of continental sulphide minerals resulting in an
enhanced delivery of dissolved sulphate to the ocean, which
likely stimulated bacterial sulphate reduction. In this respect,
the large range in d34S for sulphides of the second and third
glacial and interglacial strata is thought to result from the
enhanced microbial turnover of sulphate under variable sea-
water sulphate concentrations (see Chap. 7.5). This change
from MIF to MDF sulphur isotope pattern is not only visible
in the sedimentary record from North America, but also in
post-glacial strata from Finland and South Africa (Papineau
et al. 2005).
The Transvaal Supergroup in South Africa also captures
the critical time window of the Archaean-Palaeoproterozoic
transition. Here, the Duitschland Formation, which is
sandwiched between glaciogenic deposits, is suggested to
record the increase of the oceanic sulphate reservoir and the
coeval loss of the MIF-S signature archived in sedimentary
sulphur, both connected to the rise in atmospheric oxygen.
Guo et al. (2009) presented a record of paired multiple
sulphur isotope measurements (d34S and D33S) from
sulphides and carbonate-associated sulphate. Multiple sul-
phur isotope data reveal that the Duitschland Formation
records the demise of MIF-S, in parallel with an increase
in the range of mass-dependent sulphur isotope fractionation
(Fig. 7.4). Although a global correlation of the glacial events
of the Huronian glaciation remains speculative, the loss of
MIF-S within the upper Duitschland Formation is consistent
with the data from the Huronian Supergroup described
above. Positive d34S values in the upper Duitschland Forma-
tion also point to an increase in the seawater sulphate con-
centration with progressive bacterial sulphate reduction
linked to an enhanced nutrient delivery during the intergla-
cial period.
At present, strata from the Huronian Supergroup in North
America and from the Transvaal Supergroup in South Africa
are exclusive recorders of the termination of mass-
independent sulphur isotope fractionation. They indicate
that an irreversible increase in atmospheric oxygen occurred
between the first and second glacial event of the Huronian
Glaciation (Papineau et al. 2007; Guo et al. 2009). Following
Pavlov and Kasting (2002), the disappearance of the MIF-S
signature indicates that atmospheric oxygen levels rose from
less than 10�5 PAL to more than 10�2 PAL during the early
Palaeoproterozoic. The sulphur cycling after this first rise in
atmospheric oxygen is associated exclusively with mass-
dependent sulphur isotope fractionation and d34S values
that clearly reflect the activity of sulphate reducing bacteria
(Bekker et al. 2004; Papineau et al. 2005; Guo et al. 2009).
Fig. 7.3 Stratigraphic evolution of D33S across the Huronian glaciations (Modified after Papineau et al. 2007)
1054 M. Reuschel et al.
These observations are accompanied by a seemingly rapid
and substantial rise in carbonate d13C followed by a gentle
decline to normal values prior to the major Jatuli-Lomagundi
positive C-isotope excursion (Bekker et al. 2001;
Frauenstein et al. 2009). Thus, the determination of a causal
relationship, and precise timing between global changes in
carbon and sulphur cycling and the rise in atmospheric
oxygen requires further work.
While the overall isotopic pattern appears to be under-
stood and reasonably constrained, subsequent work should
focus on constructing a more detailed picture. Most obvious
is the causal relation between the observed trend(s) in D33S
(and d34S) and the rise in atmospheric oxygen, and the
temporal resolution thereof. The latter aspect in particular
will be important for reconstructing rates of evolution. As a
first approach, determining the stratigraphic position of the
termination of MIF-S in relation to the (multiple) glacial
horizons will suffice. Ultimately, however, detailed age
constraints throughout the Huronian time interval are crucial
for translating changes in the magnitude in isotopic fraction-
ation to changes in the magnitude of environmental
transformations.
7.1.4 The Termination of Mass-IndependentlyFractionated Sulphur: Implications forthe FAR-DEEP Core Material
The sedimentary succession deposited on the Fennoscandian
Shield covers this critical interval in Earth history with pre-
glacial (Seidorechka Formation), glacial (Polisarka Forma-
tion) and post-glacial (Umba Formation) sedimentary rocks.
These are available in the FAR-DEEP core material and the
strata contain abundant sulphides (Fig. 7.5). Consequently,
such sedimentary rocks represent a prime target for varied
research aiming at identifying the time of disappearance of
mass-independent sulphur isotope fractionation in the Early
Palaeoproterozoic. Moreover, a higher resolution multiple
sulphur isotope record will allow a detailed qualitative
understanding of the style of the termination of MIF-S
(abrupt or gradually) and ultimately a quantitative under-
standing of the related environmental changes. Further, the
MIF-S pattern might assist in resolving existing uncertainties
in chronostratigraphic correlation between the glacial deposits
on the Fennoscandian Shield and those of South Africa and
Fig. 7.4 Stratigraphic evolution of D33S across the Duitschland Formation (Redrawn after Guo et al. 2009)
1 7.1 The End of Mass-Independent Fractionation of Sulphur Isotopes 1055
Fig. 7.5 Outcrop photo (a) and thin section scans obtained by using
optical scanner (b, c) and secondary electron microscope with
backscattered electron detector (d, e, f, g) of sulphide-containing strata
from the Seidorechka Sedimentary Formation. Couplets of limestone-
dolostone and siltstone-shale of the Limestone-Shale member (see
Chap. 6.1.2 for details about Seidorechka Sedimentary Formation)
1056 M. Reuschel et al.
Canada (see Chap. 7.2). The presence of several generations
of sulphides ranging from texturally early sedimentary pyrite
and clearly later metamorphic pyrrhotite in cross-cutting
veinlets, reflecting sulphide mobilization, in the FAR DEEP
cores (Fig. 7.5) provides additionally the opportunity to inves-
tigate the importance of secondary effects on MIF-S and d34Ssignals.
References
Bao H, Rumble D, Lowe DR (2007) The five stable isotope
compositions of Fig Tree barites: implications on sulfur cycle in
3.2 Ga oceans. Geochim Cosmochim Acta 71:4868–4879
Bekker A, Kaufman AJ, Karhu JA, Beukes NJ, Swart QD, Coetzee LL,
Eriksson KA (2001) Chemostratigraphy of the Paleoproterozoic
Duitschland Formation, South Africa: implications for coupled
climate change and carbon cycling. Am J Sci 301:261–285
Bekker A, Holland HD, Wang PL, Rumble D, Stein HJ, Hannah JL,
Coetzee LL, Beukes NJ (2004) Dating the rise of atmospheric
oxygen. Nature 427:117–120
Bluth GJ, Schnetzler CC, Krueger AJ, Walter LS (1993) The contribu-
tion of explosive volcanism to global atmospheric sulphur dioxide
concentrations. Nature 366:327–329
Canfield DE (2001) Biogeochemistry of sulphur isotopes. In: Valley
JW, Cole DR (eds) Stable isotope geochemistry, vol 43, Reviews in
mineralogy and geochemistry. Geological Society of America,
Washington, DC, pp 607–633
Canfield DE (2004) The evolution of the Earth surface sulphur reser-
voir. Am J Sci 304:839–861
Cates N, Mojszis SJ (2006) Chemical and isotopic evidence for wide-
spread Eoarchean metasedimentary enclaves in southern West
Greenland. Geochim Cosmochim Acta 70:4229–4257
Clayton RN, Onuma N, Grossman L, Mayeda TK (1977) Distribution
of pre-solar component in Allende and other carbonaceous
chondrites. Earth Planet Sci Lett 34:209–224
Corfu F, Andrews AJ (1986) U–Pb age for mineralized Nipissing
diabase, Gowganda, Ontario. Can J Earth Sci 23:107–109
Crowell JC (1999) Pre-Mesozoic ice ages; their bearing on understand-
ing the climate system. Mem Geol Soc Am 192:1–106
Domagal-Goldman SD, Kasting JF, Johnston DT, Farquhar J (2008)
Organic haze, glaciations and multiple sulphur isotopes in the Mid-
Archean Era. Earth Planet Sci Lett 269:29–40
Farquhar J, Wing BA (2003) Multiple sulphur isotopes and the evolu-
tion of the atmosphere. Earth Planet Sci Lett 213:1–13
Farquhar J, Bao H, Thiemens M (2000) Atmospheric influence of
Earth’s earliest sulphur cycle. Science 289:756–758
Farquhar J, Savarino J, Airieau S, Thiemens MH (2001) Observation of
wavelength sensitive mass-independent sulphur isotope effects dur-
ing SO2 photolysis: implications for the early atmosphere. J
Geophys Res 106:1–11
Farquhar J, Wing BA, McKeegan D, Harris JW, Cartigny P, Thiemens
MH (2002) Mass-independent sulfur of inclusions in diamond and
sulfur recycling on early Earth. Science 298:2369–2372
Farquhar J, Peters M, Johnston DT, Strauss H, Masterson A, Wiechert
U, Kaufman AJ (2007) Isotopic evidence for Mesoarchaean anoxia
and changing atmospheric sulphur chemistry. Nature 449:706–710
Frauenstein F, Veizer J, Beukes N, Van Niekerk HS, Coetzee LL
(2009) Transvaal supergroup carbonates: implications for Paleopro-
terozoic d18O and d13C records. Precambrian Res 175:149–160
Grassineau NV, Nisbet EG, Bickle MJ, Fowler CMR, Lowry D, Mattey
DP, Abell P, Martin A (2001) Antiquity of the biological sulphur
cycle: evidence from sulphur and carbon isotopes in 2700 million
year old rocks of the Belingwe belt, Zimbabwe. Proc Roy Soc Lond
B268:113–119
Guo Q, Strauss H, Kaufman AJ, Schr€oder S, Gutzmer J, Wing B, Baker
MA, Bekker A, Jin Q, Kim S-T, Farquhar J (2009) Reconstructing
Earth’s surface oxidation across the Archean-Proterozoic transition.
Geology 37:399–402
Hoefs J (2009) Stable isotope geochemistry. Springer, Heidelberg,
p 285
Holland HD (1999) When did the Earth’s atmosphere become oxic? A
reply. Geochem News 100:20–22
Holland HD (2006) The oxygenation of the atmosphere and oceans.
Philos Trans Roy Soc Lond B361:903–915
Hou KJ, Li YH, Wan DF (2007) Constraints on the Archean atmo-
spheric oxygen and sulfur cycle from mass-independent sulfur
records from Anshan-Benxi BIFs, Liaonig Province, China. Sci
China Ser D Earth Sci 50:1471–1478
Hu G, Rumble D, Wang PL (2003) An ultraviolet laser microprobe for
the in situ analysis of multisulfur isotopes and its use in measuring
the Archean sulfur isotope mass-independent anomalies. Geochim
Cosmochim Acta 67:3101–3117
Hulston JR, Thode HG (1965) Variations in the 33S, 34S, 36S contents of
meteorites and their relation to chemical and nuclear effects. J
Geophys Res 70:3475–3484
Jamiesson JW, Wing BA, Hanington MD, Farquhar J (2006)
Evaluating isotopic equilibrium among sulfide mineral pairs in
Archean ore deposits: case study from the Kidd Creek VMS deposit,
Ontario, Canada. Econ Geol 101:1055–1061
Johnston DT, Wing BA, Farquhar J, Kaufman AJ, Strauss H, Lyons
TW, Kah LC, Canfield DE (2005) Active microbial sulphur dispro-
portionation in the Mesoproterozoic. Science 310:1477–1479
Johnston DT, Poulton SW, Farquhar J, Wing BA, Fralick PW, Canfield
DE (2006) Evolution of the oceanic sulfur cycle in the late
Paleoproterozoic. Geochim Cosmochim Acta 70:5723–5739
Johnston DT, Farquhar J, Summons R, Shen Y, Kaufman AJ,
Masterson AL, Canfield DE (2008) Sulfur isotope biogeochemistry
of the Proterozoic McArthur basin. Geochim Cosmochim Acta
72:4278
Kah LC, Lyons TW, Frank TD (2004) Low marine sulphate and
protracted oxygenation of the Proterozoic biosphere. Nature
431:834–838
Kamber BS, Whitehouse MJ (2007) Micro scale sulphur isotope evi-
dence for sulphur cycling in the late Archean shallow ocean.
Geobiology 5:5–17
Kasting JF, Zahnle KJ, Pinto JP, Young AT (1989) Sulfur ultraviolet
radiation, and the early evolution of life. Orig Life 19:95–108
Kaufman AJ, Johnston DT, Farquhar J, Masterson AL, Lyons TW,
Bates S, Anbar AD, Arnold GL, Buick R (2007) Late Archean
biospheric oxygenation and atmospheric evolution. Science
317:1900–1903
�
Fig. 7.5 (Continued) contain abundant sulphides that give rise to the
rusty-weathering patches in the outcrop (a). In the Limestone-Shale
member, sulphides, including pyrite (Py, typically euhedral) and pyr-
rhotite (Po, typically irregular particles), occur most commonly as
disseminations (b, d). Pyrrhotite is also found in veinlets that cross-
cut layering in the Limestone-Shale member (b, e). Pyrrhotite is the
dominant sulphide in the Sandstone-Siltstone member occurring typi-
cally as irregular, up to mm-size particles in the sandstone beds (c, f).
Euhedral pyrite crystals, rimmed by pyrrhotite, occur in the Dolostone
member (g) (Images by Aivo Lepland)
1 7.1 The End of Mass-Independent Fractionation of Sulphur Isotopes 1057
Krogh TE, Davis DW, Corfu F (1984) Precise U–Pb zircon and
baddeleyite ages for the Sudbury area. In: Pye EG, Naldrett AJ,
Gilblin PE (eds) The geology and ore deposits of the Sudbury
structure, Ontario Geological Survey, Sudbury. Geol Surv Spec
Pap 1:431–446
Lyons TW, Gill BC (2010) Ancient sulfur cycling and oxygenation of
the early biosphere. Elements 6:93–99
Lyons TW, Kah LC, Gellatly AM (2004) The Precambrian
sulphur isotope record of evolving atmospheric oxygen. In:
Eriksson PG et al (eds) The Precambrian earth: tempos and events,
developments in Precambrian geology. Elsevier, Amsterdam,
pp 421–440
Mojzsis SJ, Coath CD, Greenwood JP, McKeegan KD, Harrison TM
(2003) Mass-independent isotope effects in Archean (2.5 to 3.8 Ga)
sedimentary sulphides determined by ion microprobe analysis.
Geochim Cosmochim Acta 67:1635–1658
Ohmoto H (1999) When did the Earth’s atmosphere become oxic?
Geochem News 93:12–27
Ohmoto H, Watanabe Y, Ikemi H, Poulson SR, Taylor BE (2006)
Sulphur isotope evidence for an oxic Archaean atmosphere. Nature
442:908–911
Ono S, Eigenbrode JL, Pavlov AA, Kharecha P, Rumble D, Kasting JF,
Freeman KH (2003) New insights into Archean sulphur cycle from
mass-independent sulphur isotope records from the Hamersley
Basin, Australia. Earth Planet Sci Lett 213:15–30
Ono S, Boswell W, Johnston DT, Farquhar J, Rumble D (2006) Mass-
dependent fractionation of quadruple stable sulfur isotope system as
a new tracer for sulfur biogeochemical cycles. Geochim Cosmochim
Acta 70:2238–2252
Ono S, Shanks WC, Rouxel OJ, Rumble D (2007) 33S constraints on the
seawater sulfate contribution in modern seafloor hydrothermal vent
sulfides. Geochim Cosmochim Acta 71:1170–1182
Ono S, Kaufman AJ, Farquhar J, Sumner DY, Beukes NJ (2009a)
Lithofacies control on multiple-sulphur isotope records and
Neoarchean sulphur cycles. Precambrian Res 169:58–67
Ono S, Beukes NJ, Rumble D (2009b) Origin of two distinct multiple-
sulphur isotope compositions of pyrite in the 2.5 Ga Klein Naute
Formation, Griqualand West Basin, South Africa. Precambrian Res
169:48–57
Papineau D, Mojzsis SJ (2006) Mass-independent fractionation of
sulfur isotopes in sulfides from the pre-3770 Ma Isua Supercrustal
Belt, West Greenland. Geobiology 4:227–238
Papineau D, Mojzsis SJ, Coath CD, Karhu JA, McKeegan KD (2005)
Multiple sulphur isotopes of sulfides from sediments in the after-
math of Paleoproterozoic glaciations. Geochim Cosmochim Acta
69:5033–5060
Papineau D, Mojzsis SJ, Schmitt AK (2007) Multiple sulphur isotopes
from Paleoproterozoic Huronian interglacial sediments and the rise
of atmospheric oxygen. Earth Planet Sci Lett 255:188–212
Partridge MA, Golding SD, Baublys KA, Young E (2008) Pyrite
paragenesis and multiple sulfur isotope distribution in late Archean
and early Paleoproterozoic Hamersley Basin sediments. Earth
Planet Sci Lett 272:44–49
Pavlov A, Kasting J (2002) Mass-independent fractionation of sulphur
isotopes in Archean sediments: strong evidence for an anoxic
Archean atmosphere. Astrobiology 2:27–41
Philippot P, van Zuilen M, Lepot K, Thomazo C, Farquhar J, van
Kranendonk M (2007) Early Archean microorganisms preferred
elemental sulfur not sulfate. Science 317:1534–1537
Schr€oder S, Bekker A, Beukes NJ, Strauss H, van Niekerk HS (2008)
Rise in seawater sulphate concentration associated with Paleopro-
terozoic positive carbon excursion: evidence from sulphate
evaporites in the 2.2–2.1 Gyr shallow marine Lucknow Formation,
South Africa. Terra Nova 20:108–117
Shen YN, Farquhar J, Masterson A, Kaufman AJ, Buick R (2009)
Evaluating the role of microbial sulfate reduction in the early
Archean using quadruple isotope systematics. Earth Planet Sci
Lett 279:383–391
Strauss H (2002) The isotopic composition of Precambrian sulphides –
seawater chemistry and biological evolution. Spec Publ Int Assoc
Sediment 33:67–105
Strauss H (2004) 4 Ga of seawater evolution: evidence from the sulphur
isotopic composition of sulphate. In: Amend JP, Edwards KJ, Lyons
TW (eds) Sulphur biogeochemistry-past and present. Geological
Society of America, Boulder, pp 195–205
Thiemens MH, Heidenreich JE (1983) The mass-independent fraction-
ation of oxygen; a novel isotope effect and its possible cosmochem-
ical implications. Science 219:1073–1075
Thomazo C, Ader M, Farquhar J, Philippot P (2009) Methanotrophs
regulated atmospheric sulfur isotope anomalies during the
Mesoarchean (Tumbiana Formation, Western Australia). Earth
Planet Sci Lett 279:65–75
Ueno Y, Johnson MS, Danielache SO, Eskebjerg C, Pandey A, Yoshida
N (2009) Geological sulphur isotopes indicate elevated OCS in the
Archean atmosphere, solving the faint young sun paradox. Proc Natl
Acad Sci 106:14784–14789
Whitehouse MJ, Kamber BS, Fedo CM, Lepland A (2005) Integrated
Pb- and S-isotope investigation of sulphide minerals from the early
Archean of southwest Greenland. Chem Geol 222:112–131
Young GM (1988) Proterozoic plate-tectonics, glaciation and iron
formations. Sediment Geol 58:127–144
Zahnle K, Claire M, Catling D (2006) The loss of mass-independent
fractionation in sulphur due to a Paleoproterozoic collapse of atmo-
spheric methane. Geobiology 4:271–283
Zerkle AL, Farquhar J, Johnston DT, Raymond CP, Canfield DE (2009)
Fractionation of multiple sulfur isotopes during phototrophic oxida-
tion of sulfide and elemental sulfur by a green sulfur bacterium.
Geochim Cosmochim Acta 73:291–306
Zerkle AL, Kamyshny A Jr, Kump LR, Farqhuar J, Oduro H, Arthur
MA (2010) Sulfur cycling in a stratified euxinic lake with moder-
ately high sulfate: constraints from quadruple S isotopes. Geochim
Cosmochim Acta 74:4953–4970
1058 M. Reuschel et al.
7.2 Huronian-Age Glaciation
Victor A. Melezhik, Grant M. Young, Patrick G. Eriksson,Wladyslaw Altermann, Lee R. Kump, and Aivo Lepland
7.2.1 Introduction
Victor A. Melezhik
Glaciations have occurred throughout much of Earth’s his-
tory, episodically and with different durations. Physical and
chemical evidence of Earth’s earliest glacially derived rocks
was reported from the c. 2.9 Ga Mozaan Group in South
Africa (Young et al. 1998) and possible correlative rocks in
the West Rand Group of the Witwatersrand Supergroup. No
new findings of glacial rocks have been made in the
Archaean since, and it remains unresolved whether the
South African Archaean diamictites were locally developed
or could have broader implications.
It seems that early Palaeoproterozoic diamictites and
associated glaciogenic rocks represent the oldest known
glaciation having global significance (Fig. 7.6). This event
has been termed the Huronian Glaciation after the Huronian
Supergroup in Canada where glaciogenic deposits have been
know since Coleman (1907, 1908) reported diamictites and
dropstones in southern Ontario. Three discrete glacial units,
the Ramsey Lake, Bruce and Gowganda formations
(Figs. 7.7 and 7.8), have been studied in detail and
summarised by Young (1970, 1981a, b) and Miall (1983).
Other apparent Palaeoproterozoic glaciogenic units in
Canada, which may be correlative to the Huronian
diamictites, include the Chibougamau Formation in Quebec
(Long 1981) and Padlei Formation of the Hurwitz Group in
the Northwest Territories west of Hudson Bay (Young and
McLennan 1981). In the United States, Palaeoproterozoic
diamictites and beds with dropstones have been reported
from the Reany Creek, Enchantment Lake and Fern Creek
formations at the base of the Marquette Supergroup in north-
ern Michigan (e.g. Gair 1981), and from the Campbell Lake,
Vagner and Headquarters formations in SE Wyoming
(Houston et al. 1981). Lonestones have been found in
Palaeoproterozoic diamictites in the Black Hills of South
Dakota (Kurtz 1981).
In South Africa, several discrete diamictite beds have
been reported from the Transvaal basin. In its eastern
domain, the Transvaal Supergroup contains one
(Duitschland Formation, Bekker et al. 2001) or two
(Duitschland and Boshoek diamictites, Kopp et al. 2005)
glacial diamictites (Fig. 7.21). Athough Martini (1979)
recognised several diamictite units in the Duitschland For-
mation, only the basal diamictite has a basin-wide extent and
contains striated, bullet-shaped clasts of quartzites, base-
ment rocks and banded iron formations (Coetzee 2001);
thus it was interpreted as glacial in origin (Bekker et al.
2001). In the western Transvaal basin, a single glacial unit
is associated with the Makganyene Formation of the
Postmasburg Group (Visser 1971).
In Western Australia, glaciomarine formations, appar-
ently correlative to Huronian glacial deposits, have been
described by Trendall (1976) from the Meteorite Bore Mem-
ber of the Kungarra Formation in the southwestern
Hamersley basin (Fig. 7.21). Martin (1999) provided a
detailed description of glaciomarine lithofacies for these
rocks, which include two diamictite beds.
In Fennoscandia, Huronian-age glaciogenic deposits are
associated with the Sariolian sedimentary formations and
their equivalents (Marmo and Ojakangas 1984) (Fig. 7.9).
In Finland, Palaeoproterozoic diamictites and beds with
dropstones have been reported from the Urkkavaara Forma-
tion in the North Karelia Belt (Marmo and Ojakangas 1984).
A correlative glaciogenic unit has been described in the
Kainuu Belt of central Finland (Strand and Laajoki 1993).
On the Russian side of the Fennoscandian Shield in southern
Karelia (Negrutsa and Negrutsa 1981a, b), a glacial origin
has been suggested for parts of the Sariolian diamictites,
varved schists, and schists with dropstones (Eskola 1919;
V.A. Melezhik (*)
Geological Survey of Norway, Postboks 6315 Sluppen, NO-7491
Trondheim, Norway
Centre for Geobiology, University of Bergen, Allegaten 41, Bergen
N-5007, Norway
e-mail: [email protected]
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_2, # Springer-Verlag Berlin Heidelberg 2013
1059
Fig. 7.6 The Palaeoproterozoic Huronian glacial deposits represent a
world classic type locality. (a) Dropstone in laminated silty mudstone
rhythmites, and (b) varved sedimentary rocks with scattered dropstones
signify the presence of floating icebergs. The Gowganda Formation,
Huronian Supergroup, Ontario (Photographs by Grant Young)
1060 V.A. Melezhik et al.
Fig. 7.7 Cross-section through key stratigraphic horizons containing
the Huronian-age glacial deposits (Modified from Melezhik (2006) and
based on data from Young (1970, 2002), Miall (1983), Halls and Bates
(1990), Barley et al. (1997), Heaman (1997), Vogel et al. (1998),
Martin (1999), Bekker et al. (2001, 2005), Young et al. (2001), Pickard
(2003), Hannah et al. (2004), and Long (2004))
2 7.2 Huronian-Age Glaciation 1061
Fig. 7.8 Glacial and associated rocks from the Huronian Supergroup
in Canada, Ontario. (a) Diamictite from the Ramsey Lake Formation;
hammer for scale – 35 cm. (b) Pecors Formation varved sedimentary
rocks with dropstones on the Ramsay Lake Formation diamictite; pen
length is 14 cm. (c) Dropstone near the base of a diamictite bed
piercing lamination in rhythmically bedded siltsone-sandstone;
Gowganda Formation
1062 V.A. Melezhik et al.
Ojakangas 1985, 1988). Beds with dropstones and varve-like
laminations are known from the Polisarka Sedimentary For-
mation in the Imandra/Varzuga Belt (Melezhik 2006) as well
as from the Seletzky horizon in the Shambozero and Lekhta
belts (cf. Negrutza 1984) (Fig. 7.10).
Fig. 7.8 (continued) (d) Dropstone-piercing lamination in rhythmi-
cally bedded siltsone-sandstone of the Gowganda Formation; hammer
head length is 16 cm. (e, f) The Gowganda Formation diamictite; note
polymict composition and angular nature of scattered outsized
fragments; hammer length is 35 cm (Photographs courtesy of Richard
Ojakangas)
2 7.2 Huronian-Age Glaciation 1063
In the following sections, we provide a more detailed
overview of each of these Palaeoproterozoic glacial
successions. We then consider the possible causes of this,
the first widespread glacial interval in Earth’s history, in the
context of the other significant changes in the Earth system
during the Great Oxidation Event. We conclude with a
consideration of the potential of the FAR-DEEP core to
elucidate the timing, mechanisms, and consequences of
Palaeoproterozoic glaciation.
Fig. 7.9 Geological map and lithological sections across the Sumi-
Sarioli boundary (Modified from Melezhik 2006). (a) Map of
Fennoscandia showing the Archaean craton, younger Svecofennian
multiphase orogen, distribution of 2505–2430 Ma layered gabbro
intrusions and coeval Sumi continental flood-basalts. In the Archaean
craton, all rocks younger than the Sumi formations (i.e., younger than
2430 Ma) are omitted. (b) Lithological profiles across the Sumi-Sariola
boundary
1064 V.A. Melezhik et al.
Fig. 7.10 Sedimentological features of glacial deposits from the Kola
and Karelian cratons. (a) Andesite dropstone piercing lamination in
fine-grained greywacke. (b) Alternating coarse-grained greywacke and
siltstone with dropstone. (c) Finely-laminated sandstone-siltstone with
scattered oversized clasts (rain-out clasts from melting floating ice)
overlain by greywacke. (d) Finely-laminated, varve-like, fine-grained
sandstone-siltstone couplets. (e) Dropstone in finely laminated varved,
glaciomarine clayey siltstone
2 7.2 Huronian-Age Glaciation 1065
Fig. 7.10 (continued) (f) Finely-laminated, varve-like, fine-grained
sandstone-siltstone couplets. (g) Finely-laminated, calcareous
sandstone-siltstone which elsewhere contains lonestones and over-
sized clasts (Negrutza 1984.) (h) Diamictite composed of polymict,
unsorted clasts floating in massive siltstone matrix. (a–e) – the
Polisarka Sedimentary Formation in the Imandra/Varzuga Green-
stone Belt; (f–h) – the Pajozerskaja Formation from the Shambozero
Greenstone Belt (Photographs by Victor Melezhik)
1066 V.A. Melezhik et al.
7.2.2 Palaeoproterozoic Glacial Depositsof North America
Grant M. Young
In North America, world-famous examples of Palaeopro-
terozoic glacial deposits are associated with the Huronian
Supergroup. The supergroup outcrops extensively on the
southern margin of the Canadian Shield in the area north
of Lake Huron (Fig. 7.11), Ontario, Canada, but a much
wider distribution of similar rocks has been inferred from
near-identical stratigraphic successions in widely separated
regions such as SE Wyoming (the Snowy Pass Supergroup)
and the west side of Hudson Bay (Hurwitz Group). The
Huronian Supergroup is a mainly siliciclastic succession
that is estimated to be up to about 12 km in thickness in
the southern part of the outcrop belt. The Huronian Super-
group has an unconformable relationship with Archaean
rocks of the Superior province and deep palaeosols are
locally preserved. In the northern part of the outcrop belt,
the lower Huronian succession is missing and the Gowganda
and Lorrain formations commonly lie directly on Archaean
basement rocks. Chemical sedimentary rocks (limestones,
dolostones and sulphate evaporites) comprise a small pro-
portion of the total thickness. A succession of bimodal
volcanic rocks is present near the base. The time involved
in deposition of the Huronian Supergroup is constrained
between the age of volcanic and related intrusive rocks
dated at 2450–2480 Ma and that of a widespread suite of
mainly mafic intrusive rocks (the Nipissing diabase), which
has yielded an age of 2219 � 4 Ma (see references in Young
et al. 2001 and Melezhik 2006). These dates set a fairly firm
limit on the time when deposition of the Huronian began but
the ~2200 Ma diabase date merely indicates that Huronian
rocks accumulated before that time. It was, however, argued
by Young et al. (2001) that some of the Huronian sedimen-
tary rocks were not completely consolidated at the time
of intrusion of the Nipissing diabase, suggesting that the
~2200 Ma date is close to the time of deposition of the
upper part of the Huronian succession. If this were correct,
then Huronian deposition would have involved a time period
of about 230 Ma. A significant portion of the Huronian
Supergroup comprises three megacycles (Roscoe 1969),
each of which begins with a diamictite-bearing formation,
overlain by fine-grained siliciclastic rocks (in one case with
a significant carbonate content), followed by thick, cross-
bedded sandstones. The stratigraphic position of these cycles
is shown in Fig. 7.12. The glaciogenic nature of the
diamictites is widely (but not universally) accepted, and
some of the evidence for their depositional environment is
presented below.
Ramsay Lake Formation
The Ramsay Lake Formation has a maximum thickness of
about 150 m. It occurs in an east-west-trending region in the
central part of the Huronian outcrop belt (Young 1981a). To
the north it is overstepped by younger Huronian formations
and its southern limits are not known because the oldest
formation exposed in the core of the most southerly major
structure (the McGregor Bay Anticline) is the Mississagi
Formation. The nature of the formation boundaries is vari-
able. In some places, especially near its northern limits, it
lies unconformably on Archaean basement rocks but it has
conformable-to-disconformable relationships with underly-
ing Huronian rocks in more southerly outcrops. The forma-
tion is generally composed of massive, structureless
diamictite (Fig. 7.13a) and poorly sorted orthocon-
glomerates. In Agnew Lake, just west of the Sudbury basin
(Fig. 7.11) there is local development of stratification, and
possible ice-wedge structures were described from this unit
by Young and Long (1976), who also published a partial
stratigraphic section. Near the northern limits of its outcrop
in the Bruce Mines area, finely bedded and laminated
siltstones of the overlying Pecors Formation contain large
isolated clasts that appear to depress or penetrate underlying
layers (Fig. 7.13b). Rounded and angular clasts make up
5–85 % of the rock. They are mostly pebbles and cobbles
but some boulders are present. Grey granitic rocks are com-
mon but chert, quartz, volcanic rocks and rare quartzite
fragments are also present. In some areas the Ramsay Lake
Formation includes fragments of underlying Huronian
formations. Striated stones have not been reported but
some have suggested that elongate clasts have a preferred
orientation (generally in a NW-SE direction).
Because the formation is composed mainly of near-
structureless diamictites, which can form in a number of
different ways (e.g. as glaciogenic deposits and mass flows,
or by some combination of such processes), its depositional
environment is not immediately apparent. The widespread,
blanket-like distribution of the relatively thin diamictite-
dominated unit might favour a glacial interpretation but the
most persuasive evidence of glacial influence derives from the
occurrence of isolated clasts (ice-rafted dropstones?) in
laminated strata of the immediately overlying Pecors Forma-
tion (Fig. 7.13b) and possible ice-wedge structures described
by Young and Long (1976). Local development of layering
(sandstone beds) indicates that subaqueous processes played a
role in deposition of some parts of the formation.G.M. Young (*)
Department of Earth Sciences, University of Western Ontario, London
N6A 5B7, ON, Canada
2 7.2 Huronian-Age Glaciation 1067
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_2, # Springer-Verlag Berlin Heidelberg 2013
1067
Bruce Formation
Distribution of the Bruce Formation is similar to that of the
Ramsay Lake Formation, except that it is known to extend
farther south into the more tightly folded part of the
Huronian (Young 1981a). In most areas, the Bruce Forma-
tion overlies fluvial cross-bedded sandstones of the
Mississagi Formation, whereas the RLF generally overlies
sandstones and mudstones of the McKim Formation. The
Bruce Formation has a maximum thickness of about 300 m
but in some areas it is considerably thinner. It thickens in a
southerly direction but appears to become thinner again at its
known southern limits. Like the Ramsay Lake Formation, it
is composed mainly of structureless pebble- and cobble-
bearing diamictites (Fig. 7.14a), with a poorly sorted matrix
of muddy sandstone, characterised by the presence of dis-
tinctive, commonly rounded, sand-size quartz grains. Clasts
are generally smaller and rarer than those in the Ramsay
Lake Formation, making up about 20–30 % of the rock.
Clast types include grey and pink granites, vein quartz,
metavolcanic and metasedimentay rocks. In some areas the
Bruce Formation includes crudely bedded, coarse pebbly
sandstones, which may contain rare outsize clasts that dis-
rupt bedding and appear to be dropstones (Fig. 7.14b).
Fig. 7.11 Sketch map to show the distribution of Huronian rocks in
the southern part of the Canadian Shield. The Ramsay Lake and Bruce
formations are distributed in an east-west-trending belt that straddles
the Murray Fault zone, whereas the Gowganda Formation is known
throughout the entire outcrop area of the Huronian Supergroup. Meta-
morphic rocks of the younger Grenville province are juxtaposed against
the Huronian rocks in the southeast and the Huronian Supergroup is
unconformably overlain by lower Palaeozoic rocks to the south.
Whitewater Group refers to a succession of sedimentary rocks pre-
served within the elliptical outcrop of the Sudbury Igneous Complex.
The rocks of the Whitewater Group are younger than the Huronian and
may be a remnant of once-extensive foreland basin deposits related to
closure of the Huronian ocean during the Penokean orogeny
1068 V.A. Melezhik et al.
Fig. 7.12 Schematic geological column to show the stratigraphy of the Huronian Supergroup. Note the position of three diamictite-bearing
formations and the large-scale cycles with which they are associated. See text for description and explanation
2 7.2 Huronian-Age Glaciation 1069
Fig. 7.13 Photographs of glacial deposits from the Ramsay Lake
Formation. (a) Diamictite of the Ramsay Lake Formation near Quirke
Lake in the northern part of the Bruce Mines area (Fig. 7.11). Note the
highly varied clast size and angular nature of some. (b) Laminated and
finely bedded mudstones of the Pecors Formation, containing large
dropstones, suggesting the presence of floating glacier ice. This outcrop
is at the northwest end of Quirke Lake in the northern part of the Bruce
Mines area (Fig. 7.11). The hammer shaft rests on diamictite forming
the upper part of the Ramsay Lake Formation. Hammer in (a) and (b) is
about 35 cm long (Photographs by Grant Young)
1070 V.A. Melezhik et al.
Fig. 7.14 Photographs of glacial depositis from the Bruce Formation.
(a) Sandy diamictite of the Bruce Formation in the southern part of the
Espanola area (Fig. 7.11). Note that there are few clasts. Lines running
from top right to bottom left are Pleistocene glacial striations. Coin is
2.8 cm in diameter. (b) Bedded sandstones and granule conglomerates
with dropstones in the Bruce Formation, in the Bruce Mines area.
Hammer head is ~12 cm long (Photographs by Grant Young)
2 7.2 Huronian-Age Glaciation 1071
Where the diamictite surfaces are weathered, they com-
monly show a distinctive rusty-brown colour due to oxida-
tion of pyrite. In some areas the Bruce Formation, together
with associated Huronian Formations, appears to have been
involved in early, large-scale slump folding prior to deposi-
tion of the Gowganda Formation (Young 1983; Young et al.
2001). Some intrusive clastic dykes and sills in the overlying
Espanola Formation were probably injected from the still-
unconsolidated Bruce Formation, possibly as a result of
earth movements that brought about the soft-sediment defor-
mation prior to, or contemporaneous with, deposition of the
Gowganda Formation (Young 1983).
As in the case of the Ramsay Lake Formation, the wide
aerial distribution of a relatively thin diamictite-dominated
unit and the presence of rare dropstones in associated
stratified pebbly sandstones suggest glacial influence on
deposition of the Bruce Formation, although some stratified
intervals represent reworking of glaciogenic debris in a
subaqueous setting.
Gowganda Formation
The Gowganda Formation is one of the earliest recognised
Palaeoproterozoic glacial deposits (Coleman 1908). It has a
much wider distribution than the two earlier-deposited
diamictite-rich Huronian formations. This is mainly because
it oversteps the older units and rests unconformably on
Archaean basement rocks over a huge area to the northeast
in the Cobalt Plain and in a smaller area in the northwest,
near Flack Lake. The Gowganda Formation is much thicker
than the Ramsay Lake and Bruce formations and has a much
more complex internal stratigraphy (Fig. 7.15). In some
northern areas details of the stratigraphic succession within
the Gowganda Formation are known (Lindsey 1969; Miall
1983) but due to the scattered nature of outcrops in the northern
areas and the lack of lateral continuity of many stratigraphic
units, there are no satisfactory detailed regional correlation
charts of the formation in this area. In some areas, however,
there is a large-scale twofold informal subdivision into a lower,
diamictite-dominated unit (Fig. 7.16a) and an upper bedded
mudstone-sandstone unit (e.g. Thomson 1957). In northern
areas, some mudstones are characterized by fine, regular
laminations (with dropstones) that resemble Pleistocene
varves (Fig. 7.16c). Less-regularly stratified siltstones/
mudstones also contain large dropstones (Fig. 7.16d).
In the area near Lake Huron, where there is exceptionally
good exposure and the beds are relatively steeply dipping,
the Gowganda Formation is about 1,600 m thick. In north-
western areas it tends to be thinner (a few hundred metres)
but some areas display highly variable thicknesses, perhaps
related to contemporaneous faulting, and a thickness of
3,000 m was reported by Schenk (1965) from the Cobalt
area in the northeasterly part of the Huronian outcrop belt.
Card et al. (1977) noted thinning of the Gowganda formation
to the southeast, in the vicinity of the Grenville Front. It
should be emphasized that thickness measurements in the
southern, tightly folded portion of the Huronian outcrop belt
are minimal, for there has been considerable flattening of
some stratigraphic units. Regional mapping by Card et al.
(1977) in the southern part of the Huronian fold belt was
complemented by detailed mapping of the Gowganda For-
mation by Lindsey (1969) and Young and Nesbitt (1985).
As with the Ramsay Lake and Bruce Formations, the
notable characteristic of the Gowganda Formation is the
occurrence of abundant diamictites (Fig. 7.16a) and coarse,
granite-boulder-rich orthoconglomerates (Fig. 7.16b). An
important component of the Gowganda Formation is the
occurrence of varve-like rhythmically alternating laminae
of siltstone and mudstone with isolated clasts that were
probably transported by glacier ice (Fig. 7.16c, d). In some
areas there are crudely bedded and massive, pink arkosic
sandstones and orthoconglomerates that display normal and
inverse grading (Fig. 7.16e). The clast content in the
diamictites is highly variable with an average estimated at
about 30–40 %. Granitoid fragments (up to several metres in
diameter) are dominant (up to 80 %) whereas mafic volcanic
and intrusive rocks constitute about 15–20 %. The remainder
is mainly sedimentary rocks, including both siliciclastic and
chemical varieties. In some areas there is evidence of
“cannibalisation”, with incorporation of fragments of
laminated argillite and diamictite that were clearly derived
from the same formation. Preferred orientation of elongate
clasts has been reported by Young (1968) and Lindsey
(1969). Rare faceted and striated clasts are present in
diamictites of the Gowganda Formation. It also includes a
higher proportion of finer-grained, stratified rocks than the
two lower diamictite-rich formations. In northern areas the
Gowganda Formation commonly lies on Archaean rocks but
in the south it overlies arkosic sandstones of the Serpent
Formation. In southern areas the contact is commonly irreg-
ular and shows evidence of penetration of clastic dykes into
the underlying sandstones (Chandler 1973; Bernstein and
Young 1990; Young and Nesbitt 1985). There is much
evidence suggesting that the Serpent Formation was uncon-
solidated or partially consolidated when deposition of the
Gowganda began (Young 1983). The upper boundary of the
Gowganda Formation appears to be gradational, involving
a transition from coarsening-upward deltaic cycles into
fluvial sandstones of the Lorrain Formation (Fig. 7.15). For
purposes of lithological description, the well-exposed sec-
tion near Whitefish Falls in the southern part of the Huronian
outcrop belt is used. In this area the Gowganda Formation
can be divided into several stratigraphic units, which are
represented in the generalized stratigraphic column shown
in Fig. 7.15. A brief description of each of the informal
subdivisions is given below. More details are given in
Young and Nesbitt (1985).
1072 V.A. Melezhik et al.
Fig. 7.15 Simplified stratigraphic section to represent the succession of rock types in the Gowganda Formation near Whitefish Falls in the
southern part of the Espanola Area (Fig. 7.11). See text for description and interpretation
2 7.2 Huronian-Age Glaciation 1073
Fig. 7.16 Photographs of the Gowganda Formation. (a) Diamictites of
the lower part of the Gowganda Formation in the south limb of the
Quirke syncline, Elliot Lake area. Note the dark-coloured, fine matrix
and scattered rounded and angular clasts. Hammer shaft is 35 cm long.
1074 V.A. Melezhik et al.
Lower Gowganda MemberThe Lower Gowganda member starts with a diamictite com-
plex (40–240 m) consisting mainly of grey diamictites with
abundant clasts of grey and pink granitoid rocks but includ-
ing metavolcanic and sedimentary rock fragments. The
diamictites are commonly laminated and bedded but some
are massive. At the base there are some lenses of orthocon-
glomerate that have erosive contacts with the underlying
Serpent Formation. In most places, however, the contact
between the two formations is somewhat diffuse. Clastic
dykes penetrate downwards from the Gowganda into the
Serpent sandstones but there is evidence that the Serpent
Formation was not completely consolidated prior to deposi-
tion of the Gowganda Formation (Young 1983). These
characteristics have been interpreted to indicate a “soft sedi-
ment” contact with penecontemporaneous deformation pos-
sibly related, in part, to emplacement of the diamictites
(Young and Nesbitt 1985). The basal part of the lower
diamictite complex (Fig. 7.15) contains some thin turbidites
that have been locally disrupted by early folding and
faulting. The lower diamictite complex also includes beds
and lenses of orthoconglomerate, sandstone and mudstone.
A thin (few metres) unit of bedded and cross-laminated fine
sandstones near the top of the lower diamictite complex also
indicates subaqueous deposition. Thus, the lower part of
the Gowganda Formation in this area provides evidence
of turbidity currents and mass flow activity but massive
diamictites are also present. Whereas much of the material
was probably of glacial origin, it was mostly subjected to
resedimentation processes in a marine (?) setting.
The middle argillite unit (100–240 m) consists of
laminated, grey and green argillite displaying parallel, wavy
and lenticular bedding. This unit contains no outsize clasts
except for a few small rock fragments and thin diamictite
lenses in the top and basal metre or so. The basal contact is
gradational over a thickness of about 1 m. Apart from these
minor zones near the contacts, the thick argillite displays no
physical evidence of glacial influence. It probably represents
distal glacial outwash materials deposited in a deepening
basin during an interglacial period. Sporadic thin lenses of
sandstone and conglomerate are present at the top of the
argillite unit.
The upper diamictite complex (130–400 m) is stratified,
lithologically heterogeneous and contains many discontinu-
ous stratigraphic units so that it is difficult to portray in a
summary description. The interested reader is referred to
Young and Nesbitt (1985). The lowest unit of the upper
diamictite complex is grey to blue-green diamictite with
clasts up to cobble size. The contact with the underlying
argillites is commonly gradational. The basal diamictite is
characterised by the presence of fine bedding and lamina-
tion, and many clasts show evidence of having been
emplaced vertically, as from melting glacial ice. The upper
diamictite complex represents a second glacial advance/
retreat cycle, following the recession represented by the
middle argillite unit.
In some areas the remainder of the upper diamictite
complex is a stratified diamictite unit up to about 200 m in
thickness. It has a sharp but locally irregular basal contact
and may contain ragged fragments of the underlying sand-
stone. These diamictites commonly display stratification and
include thin sandstones, siltstones and mudstones. When the
diamictites are traced eastwards in the southern part of the
Espanola area (Fig. 7.11), a laminated mudstone or argillite
is introduced. The argillite thickens to the east, dividing the
diamictite into two parts and reaching a thickness of about
50 m. These diamictites are interpreted as glacial-marine
deposits, formed either by rain-out of debris or as
resedimented deposits transported as subaqueous mass
flows. Associated finer siliciclastic deposits commonly dis-
play graded beds and other characteristics of deposition by
turbidity currents although there is also evidence of traction
current activity. The argillite unit is laminated and finely
bedded. It contains rare dropstones. It comprises coarsening
upward cycles in some areas and displays evidence of soft
sediment movements in the form of slumped beds. The
general absence of dropstones in this unit is surprising but
the ice may have been grounded at some distance to the
north, precluding transport of coarse debris into the area.
The coarsening upward sequences and evidence of slope
instability support the idea that the unit was formed by
progradation.
The stratified diamictites are followed by 25–50 m of
grey-to-buff orthoconglomerates with clasts up to boulder-
size. Most of the clasts are granitic and mafic igneous rocks
but there are also some large fragments of contemporane-
ously deposited siltstones and mudstones. The basal contact
is erosive into the underlying diamictite. The coarse basal
conglomerate passes up into an interbedded succession of
conglomerates, sandstones and mudstones that displays
graded bedding (Fig. 7.17a) and slump structures. A detailed
log of this unit is illustrated in Young and Nesbitt (1985,
�
Fig. 7.16 (continued) (b) Granite boulder conglomerate of the
Gowganda Formation in the central part of the Cobalt area. Compass
is about 12 cm long. (c) Laminated (varved?) mudstones in the
Gowganda Formation at Wells Township in the Bruce Mines area.
Note small dropstone (about 3 cm in diameter). (d) Large dropstone
in laminated and finely bedded mudstones and fine sandstones of the
Gowganda Formation near Timagami in the Cobalt area. Hammer is
about 35 cm long. (e) Inverse graded conglomerate with many pink
granite clasts, overlying massive pink arkosic sandstone. These
“resedimented” deposits form part of the Gowganda Formation on the
south limb of the Quirke Syncline, Elliot Lake area. Hammer shaft is
35 cm long (Photographs by Grant Young)
2 7.2 Huronian-Age Glaciation 1075
their Fig. 10). These rocks were clearly deposited as a series
of mass flows and turbidites but the large clasts and mixed
nature of some of the material suggest derivation from
glaciogenic deposits that formed elsewhere (probably in
more ice-proximal areas to the north). The basal orthocon-
glomerate appears to be through-going but many units in the
upper diamictite complex are discontinuous. In some areas
the upper part of this resedimented unit comprises wedge-
shaped bodies that are bounded by faults at their western
limits. Each of these wedge-shaped bodies comprises a basal
diamictite followed by a coarsening-upward sequence from
argillites that contain many rip-up clasts of mudstone, silt-
stone and sandstones. These have been interpreted as sub-
marine fan deposits shed from contemporaneous fault scarps
(Young and Nesbitt 1985). In some areas this largely
resedimented sequence is followed by a thin (8 m) stratified
diamictite and a thin (2–3 m) argillite with rare dropstones,
followed by up to 20 m of sandstone and interbedded
Fig. 7.17 Photographs of the Gowganda Formation in the southern
part of the Espanola area (Fig. 7.11). (a) Normal and inverse grading in
sandstones and pebble conglomerates of the thick sandy unit near the
base of the upper diamictite complex (Fig. 7.15). The lower (fine
grained) portion is normally graded whereas the pebbly, upper partdisplays inverse grading. Pen is ~13 cm long. (b) Laminated, wavy- and
lenticular-bedded siltstones and mudstones in the middle portion of one
of the coarsening upward sequences of the upper deltaic complex
(Fig. 7.15), Coin is 2.4 cm in diameter. (c) Climbing ripples in wavy
bedded siltstones forming the middle part of one of the coarsening
upward cycles of the upper deltaic complex (Fig. 7.15). Coin is ~2.4 cm
in diameter. (d) Ball-and-pillow structures near the top of the upper-
most coarsening upward cycle of the Gowganda Formation close to the
contact with the Lorrain Formation. Coin is ~2.4 cm across
(Photographs by Grant Young)
1076 V.A. Melezhik et al.
siltstones that show much evidence of soft-sediment defor-
mation. Thus the lower half of the upper diamictite complex
(Fig. 7.15) is dominated by a variety of resedimented
deposits of highly variable grain size and complex geometry,
probably attesting to the influence of glacial ice and to the
onset of contemporaneous fault activity in the southern part
of the area.
Upper Gowganda MemberThe upper part of the Gowganda Formation was
differentiated in the northeastern part of the outcrop area as
the Firstbrook Formation (Thomson 1957) and a similar
subdivision, restricting use of the name Gowganda Formation
to the lower, diamictite-rich part was proposed by Lindsey
(1969) for the southern area. Lindsey (1969) suggested the
name La Cloche Formation for the upper portion of the
Gowganda. Neither of these names has received wide accep-
tance in the literature and the upper portion of the succession
is therefore retained as part of the Gowganda Formation.
Because of the dominantly coarsening-upwards character of
much of the Upper Gowganda member, Lindsey (1969) pro-
posed that prograding deltas played an important role in their
deposition. This suggestion was supported by subsequent
detailed work by Junnila and Young (1995) in the southern
area and by Rainbird and Donaldson (1988) in the northeast.
A significant difference between the two areas is that Rain-
bird and Donaldson described a single coarsening upward
sequence, whereas there are several in the south.
In the southern outcrop area Junnila and Young (1995)
documented the presence of four coarsening-upward cycles
in the upper Gowganda Formation (schematically
represented in Fig. 7.15). These have a combined thickness
that ranges from 380 to 780 m, so that these rocks, which are
mainly non-glacial, make up a significant proportion of the
Gowganda Formation in this area. The basal part of each
cycle is made up of fine-grained, finely laminated
mudstones, which generally have a sharp contact with the
underlying coarser grained rocks. The lowest of these argil-
lite units is pervasively slumped and disrupted. In upward
succession the argillites become coarser grained and some
portions exhibit wavy bedding and ripple cross laminations
(Fig. 7.17b, c). Ball-and-pillow and slump structures are
common (Fig. 7.17d). The upper parts of the cycles typically
comprise sandstones and siltstones, some of which are cross-
bedded. Isolated clasts of granitoid and other basement rocks
are present in laminated mudstones of the first cycle
(Fig. 7.15) and in eastern parts of the study area there is a
thick (up to 200 m) diamictite in this cycle. The diamictite
thins and disappears westward. In eastern areas a second,
thinner diamictite is also present higher in the lowest coars-
ening upward cycle. This diamictite is much more wide-
spread but it too becomes thinner to the west and is
represented in some places by scattered dropstones in bed-
ded siltstones (Fig. 7.15). The coarsening-upward cycles
forming the upper part of the Gowganda Formation are
interpreted as the deposits of advancing deltaic lobes. The
basal fine-grained portion represents the prodelta; siltstones
and fine sandstones containing evidence of slumping and
resedimentation processes formed in a delta slope environ-
ment; fine- to coarse-grained, cross-bedded sandstone are
thought to be distributary-mouth sand sheets that were
influenced by shallow marine processes (Junnila and Young
1995). The coarsening-upward cycles were deposited sub-
aqueously; each represents progradation of a braid delta into
a shallow marine basin that was moderately wave-influenced.
The Upper Gowganda member is represented in the northeast
by a single coarsening-upward sequence that is about 500 m
thick (Rainbird and Donaldson 1988). It contains evidence of
diagenetic reddening, a feature that is absent in the south,
although reduction may have accompanied low-grade meta-
morphism in the latter area. As in the south, there is evidence
of a marine tidal influence. The thick, laterally variable deltaic
deposits forming the upper part of the Gowganda Formation
are overlain gradationally by a thick succession of cross-
bedded arkosic sandstones and minor mudstones forming the
base of the Lorrain Formation. The character of the Lorrain
Formation changes upwards as it passes into a thick and
widespread blanket of quartz arenites.
Summary of Huronian Supergroup
The Gowganda Formation is a thick (>1,600 m) succession
of sedimentary rocks containing evidence of glacial influ-
ence in the form of striated rock surfaces (rare) and grooved
and furrowed soft-sediment surfaces, faceted and striated
clasts, varve-like laminated mudstones and abundant
dropstones. It contains numerous diamictites. It has been
suggested (Lindsey 1969) that in northern areas, the forma-
tion was deposited in a continental setting, whereas in the
south it was probably marine. This interpretation was
challenged by Miall (1983) who thought that even in north-
ern areas the formation was deposited under a marine influ-
ence. Miall’s interpretation was influenced by the evidence
of resedimentation processes but such processes are also
active in tectonically influenced, rapidly subsiding lakes. In
the better-known, southern part of the Huronian outcrop belt
the lithology of the Gowganda Formation is extremely het-
erogeneous and it displays some lateral variability. Although
the formation is justifiably world-famous as an example of
Palaeoproterozoic glaciation, the majority of its rocks con-
tain no physical evidence of such conditions. For example,
the thick “middle argillite” bears no dropstones and probably
formed in an ice-free, interglacial period and the upper,
deltaic complex, which comprises almost half the thickness
2 7.2 Huronian-Age Glaciation 1077
of the formation, is also devoid of evidence of glacial influ-
ence, with the exception of local development of diamictites
and dropstones in the lowest cycle.
The tectonic setting of these ancient rocks remains con-
tentious but it has been suggested that the thick Huronian
Supergroup represents the rift-to-drift part of a Wilson Cycle
(Young and Nesbitt 1985). It was suggested by Long (2004)
that the early stages of ocean opening were characterised by
development of a transtensional extensional basin but the
overall setting (rift-to-drift transition) is similar. Noting
remarkable stratigraphic similarities with the Snowy Pass
Supergroup of Wyoming (Young 1975; Houston et al.
1992), Roscoe and Card (1993) proposed that the Wyoming
area may represent a craton that was formerly on the “other
side” of the developing Huronian ocean, then rotated in a
clockwise direction as it was carried about 2,000 km to the
SW (present coordinates). Alternatively, the strong
similarities between these Palaeoproterozoic basins may
reflect their development on the same margin of a large-
scale developing ocean (Young 2004). Similar glaciogenic
rocks are known from Michigan, southern Wyoming, the
west side of Hudson Bay (Hurwitz Group) and from
Chibougamau in northern Quebec, leading Young (1970)
to speculate that there was an extensive Palaeoproterozoic
glaciation in North America and possibly elsewhere (Young
2004).
It is likely that the Ramsay Lake Formation was deposited
during an early stage of the rifting process. The diamictites
of the Bruce Formation were formed at a later stage, in a
widening rift basin, just prior to break-up of the continent.
The relatively restricted nature of the sedimentary basin may
explain the limited distribution of these early glacial
formations. The restricted distribution and typical sandy
nature of the early diamictites may be due to their local
provenance (glaciers developed on or around uplifted rift
shoulders) and incorporation of a high proportion of coarse
matrix material from earlier-deposited sands. Their rela-
tively simple stratigaphy may reflect the fact that they were
largely continental deposits and were not subjected to as
much reworking/resedimentation as those of the marine-
influenced portion of the Gowganda Formation. The
Gowganda Formation is much more extensive, thicker and
more stratigaphically complex than the two lower
glaciogenic formations. It contains abundant and unequivo-
cal evidence for the existence of glaciers during deposition
but more than half of its contained sedimentary rocks bears
no physical evidence of glacial influence. Some, such as the
thick middle mudstone unit, were formed in relatively deep
water during an interglacial period and the thick deltaic
deposits forming the upper half of the formation were mostly
formed in a post-glacial phase. The wide distribution of the
Gowganda Formation and its overstepping relationship with
the older Huronian formations suggest that it formed just
after the rift-drift transition, when there was widespread
crustal subsidence along a newly formed passive margin.
Deposition in such a tectonically active period provides an
explanation for the high percentage of resedimented and
slumped materials.
Other Occurrences of PalaeoproterozoicGlaciogenic Rocks in North America
Lake Superior AreaThe nearest Palaeoproterozoic rocks to those of the Huronian
Supergroup are on the south shore of Lake Superior, in
northern Michigan. Generalised correlations between the
Proterozoic rocks of the Lake Superior area and the Huronian
of the north shore of Lake Huron persisted for more than
a 100 years (see Young 1966) but James (1958) suggested
that this practice be discontinued and proposed that the
term “Animikie Series” be used for the iron-formation-rich
Proterozoic successions of Michigan. When the stratigraphic
successions in the two areas became better-known, Young
(1966) and Young and Church (1966) proposed that the
upper part of the Huronian succession in Ontario (Cobalt
Group) may be correlated to distinctive rock types of the
Chocolay Group in Michigan. These distinctive rock types
include glacial diamictites (Gowganda Formation), followed
by unusual aluminous orthoquartzites (part of the Lorrain
Formation), and carbonate- and evaporate-bearing fine
siliciclastic rocks of the Gordon Lake Formation. In particu-
lar, it was proposed that thin glaciogenic units that lie uncon-
formably on Archaean basement rocks in several places in
northern Michigan (Fern Creek, Enchantment Lake and
Reany Creek formations), as described by Pettijohn (1943),
Gair (1981) and Puffett (1969), may be equivalent to the
much more extensive Gowganda Formation. This lithostra-
tigraphic correlation invoked equivalence of the upper part of
the Huronian succession (Cobalt Group) and the lowest part
(Chocolay Group) of the Palaeoproterozoic succession in
northern Michigan. For many years this suggestion met
with widespread disapproval (e.g. Cannon 1973; Morey
1973; van Schmus 1976; Sims and Peterman 1983), mainly
because available radiometric age determinations suggested
to these workers that the Lake Superior rocks were much
younger. In spite of early suggestions such as those of Young
(1966) and Young and Church (1966) and that of Ojakangas
(1988) and Ojakangas et al. (2001b), correlation between the
Palaeoproterozoic successions of the Lake Huron and Lake
Superior regions was not generally accepted. New
geochronological work on detrital zircons and xenotime
cements from rocks of the Chocolay Group in northern
Michigan (Vallini et al. 2006) provided convincing data in
support of the old correlations. The new age data showed that
the depositional age of the Chocolay Group, including the
1078 V.A. Melezhik et al.
Fig.7
.18
Generalised
columnsillustratingthestratigraphiccontextofPalaeoproterozoicglaciogenicrocksintheHuronianSupergroupandequivalentselsewhereinN.A
merica.Therearethree
glaciogenic
diamictite-bearingform
ationsin
theHuronianbasin
andin
theSnowyPassSupergroupofS.E.Wyoming,whereasitis
likelythat
thesingle
occurrencesofPalaeoproterozoic
diamictite-bearingform
ationsin
N.Michigan
andin
theHurw
itzbasin
arecorrelativeto
theGowgandaForm
ation.Notethelargetimegap
thathas
beendocumentedin
allbuttheSEWyoming
area.T
hesignificance
oftheverylonghiatusbetweendepositionofwhataremostlyconsidered
tobepassivemargindepositsandthose
form
edbysubsequentocean
closureisnotunderstood.S
ee
textfordiscussion
2 7.2 Huronian-Age Glaciation 1079
glaciogenic diamictites at its base, is probably between about
2.3 and 2.2 Ga, thus validating the lithostratigraphic correla-
tion proposed 40 years earlier.
South DakotaTo the west of the Great Lakes region, possible Palaeopro-
terozoic glaciogenic rocks, including diamictites and fine-
grained sedimentary rocks with isolated clasts, have been
described from the Black Hills region of South Dakota by
Kurtz (1981) who consider them to be in age between c.
2560 and 1620 Ma. These rocks are poorly preserved
(metamorphosed and deformed). Recent geochronological
data (Dahl et al. 2006) suggest that there was an episode of
gabbroic magmatism in this area at c. 2480 Ma, comparable
to that described from the base of the Huronian Supergroup
in Ontario. They suggested that these data indicated the
presence of an earliest Proterozoic axial rift zone that
extended from the Sudbury area to South Dakota.
SE WyomingA thick development of Palaeoproterozoic rocks in SE
Wyoming (Fig. 7.18) contains a succession of sedimentary
rocks that closely resembles that of the Huronian Super-
group (Young 1973, Fig. 6; Young 1975, Fig. 1). These
rocks were investigated by Blackwelder (1926) and subse-
quently by Houston and associates at the University of
Wyoming in Laramie (Karlstrom et al. 1984; Houston
et al. 1992). The thick succession of sedimentary rocks is
now known as the Snowy Pass Supergroup. It is thought to
have been deposited on an Atlantic-type passive margin
between c. 2.5 and 2.00 Ga and to have been deformed in
an orogeny at c. 1.8 to 1.7 Ga (Karlstrom et al. 1984). Close
stratigraphic resemblance to the Huronian succession (and to
the Hurwitz Group) on the west side of Hudson Bay was
pointed out by Young (1975). The similarities between the
Snowy Pass and Huronian supergroups are so striking that
they led Roscoe and Card (1993) and Dahl et al. (2006) to
suggest that the Wyoming Craton represents the “other side”
of the Huronian basin that has been subsequently displaced
about 2,000 km to the southwest and rotated, in a clockwise
manner, through about 180 �. Whereas this interpretation is
possible, a simpler and more conservative alternative is that
the Palaeoproterozoic rocks of the Snowy Pass Supergroup
were deposited on the same continental margin as the
Huronian and record the same tectonic and palaeoclimatic
history (Young 2004, Fig. 6).
Hurwitz Group, NunavutThe Hurwitz Group crops out widely in part of Nunavut, on
the west side of Hudson Bay. These rocks were included in
early regional studies by the Geological Survey of Canada but
one of the first attempts to place them in a plate-tectonic
context was by Bell (1970) who considered the Hurwitz
Group rocks to have been deposited on a “metastable craton”
although he hinted at the possibility that they may have
“geosynclinal” attributes. The Hurwitz Group has subse-
quently received considerable attention, notably in papers by
Aspler and associates (see Aspler et al. 2001 and references
therein). The Hurwitz Group is now considered to consist of
two quite separate divisions; the lower part is younger than
2.45 Ga and is cut by sills that have given dates of 2.11 Ga.
There follows a gap that is reckoned to represent about
200 Ma, for the second part of the Hurwitz Group is younger
than 1.91Ma.Diamictites were recognised in the lower part of
the Hurwitz Group (Bell 1970), and Pettijohn (1970)
commented on the possibility that they might be related to
some of the Huronian glaciogenic formations. Bell (1970,
p. 168) was of the opinion that the Hurwitz Group might be
younger than the Huronian. Wanless and Eade (1975)
reiterated that opinion, based on geochronological data
(K-Ar and Rb-Sr dates) and, as a corollary, questioned the
correlation suggested by Young (1973) between parts of the
Hurtwitz Group and the Huronian Supergroup. Subsequent
geochronological studies (summarized by Aspler et al. 2001)
have substantiated the correlation (see Young 1975, p. 1252,
Fig. 1). The provenance of the sedimentary rocks of the
Hurwitz Supergroup is complex because some units were
probably formed in an intracratonic setting and other, more
extensive units such as the orthoquartzites that overlie the
glacial diamictites of the Padlei Formation, are shallow-
water deposits subject to wind-driven water movements.
Aspler et al. (2001) believed that the Hurwitz Group was
mainly deposited in an intracratonic basin. Regional strati-
graphic and structural considerations suggest, however, that
the majority of the siliciclastic rocks were probably derived
from the northwest (Aspler et al. 2001) although some of
the youngest sedimentary rocks of the Hurwitz Group and
the overlying Kiyuk Group (1.90–1.82 Ga) may have been
derived from an orogen that lay to the southeast (Young 1988,
Fig. 3).
Chibougamau Area, QuebecDiamictites and dropstone-bearing laminated mudstones of
probable Palaeoproterozoic age occur in Quebec, c. 400 km
northeast of the nearest outcrops of the Gowganda Forma-
tion near Rouyn. Early investigation of these rocks is
documented in Young (1970) and by Long (1974; 1981),
who illustrated the occurrence of dropstones in laminated
mudstones. The relationship between the Chibouganau For-
mation and rocks of the Mistassini Group, which lie to the
northeast, is uncertain but it is likely that the latter are
younger and possibly equivalent (in part) to higher (post-
Gowganda) formations of the Huronian Supergroup. The
presence of diamictite-filled clastic dykes in Archaean
rocks up to 80 km west of the main outcrops of
Chibougamau Formation led Chown and Gobeil (1990) to
1080 V.A. Melezhik et al.
suggest that the glacial deposits may have formerly been
much more extensive. These clastic dykes reinforce the
suggested correlation between the Chibouganau Formation
and the Gowganda Formations, which unconformably
overlies Archaean rocks south of Rouyn, Quebec, c. 400 km
to the SW.
Summary of North American Occurrencesof Palaeoproterozoic Glaciogenic Rocks
The Huronian Supergroup includes the best examples of
Palaeoproterozoic glaciations in North America (and possi-
bly in the world). There is evidence of three separate glacial
periods, the youngest of which (the Gowganda Formation) is
by far the thickest and most extensive. Young and Nesbitt
(1985) suggested that the two less-widespread Huronian
glaciogenic units (Ramsay Lake and Bruce Formations)
formed in a rift basin prior to continental separation respon-
sible for development of the continental margin on which the
extensive Gowganda Formation was deposited. Early
attempts at lithostratigraphic correlation of these ancient
glacial deposits were frustrated by spurious age data and
erroneous interpretations of these data but as
geochronological techniques have become more sophisti-
cated, new results are showing that many of the early
lithostratigraphical correlations are correct. Correlation of
the Gowganda Formation with thinner glaciogenic deposits
in Northern Michigan is now supported by modern geochro-
nology (Vallini et al. 2006). Deposition of Palaeoproterozoic
sedimentary rocks in the N. Michigan area was initiated at
the time of deposition of the Gowganda Formation so that the
earlier glaciogenic formations and associated rocks are not
preserved there. The subsidence that resulted in widespread
deposition of the Gowganda Formation in the Huronian fold
belt also affected areas on the south side of Lake Superior,
where there was accommodation for glaciogenic deposits
such as those of the Reany Creek and Fern Creek formations
at the base of the Palaeoproterozoic succession.
The most remarkable lithological correlation is that
between the Huronian Supergroup and the Snowy Pass
Supergroup in southeastern Wyoming, where virtually
every formation may be matched between the two widely
separated areas. In order to explain these striking
similarities, it was suggested by Roscoe and Card (1993)
that the Wyoming province was formerly adjacent to the
Huronian basin and has been rotated and translocated to its
present distant position. Likewise Bleeker and Ernst (2006)
proposed that the Hearne province, where the Hurwitz
Group is now preserved, on the west side of Hudson Bay,
may have been in the same location (see Dahl et al. 2006,
Fig. 6). As argued by Young (2004), it is unreasonable to
propose juxtaposition of all the Huronian-age glacial
deposits against the Huronian basin. A more conservative
interpretation (Fig. 7.19) is that the stratigraphic similari-
ties may be due to accumulation of Palaeoproterozoic
successions along the same continental margin that
underwent a similar palaeoclimatic history. The palaeogeo-
graphic situation of the Hurwitz Group is not fully under-
stood. It was suggested by Aspler et al. (2001, and references
therein) that the Hurwitz Group mainly accumulated in an
intracratonic basin. Earlier work by Bell (1970) mentioned
the possibility that the Hurwitz succession represents some
kind of “Wilson cycle” and Young (1988) suggested that the
stratigraphic successions and tectonic history of the Hurwitz
and adjacent basins, such as the Amer basin to the north,
might be accommodated in a model that involves a
southeastward-facing continental margin that probably lay
at some distance to the southeast.
The distribution of known glaciogenic rocks is shown on
a tectonic map of North America (Fig. 7.19). Most of the
known glacial rocks of Palaeoproterozoic age presently
occur along the southeastern margin of a large Archaean
craton. The one exception is the Hurwitz Group, which is
located in the Hearne domain on the west side of Hudson
Bay. The extent of Palaeoproterozoic ice sheet (or sheets) is
not known but its possible extent during deposition of the
Gowganda Formation and correlatives is shown by the
heavy dashed line. The Huronian continental margin may
have extended to the northeast (present co-ordinates) across
Greenland to the Karelian region (Young 2004). Ojakangas
et al. (2001b) proposed the Canadian and Fennoscandian
shields may have been even been connected. It has been
suggested (Williams and Schmidt 1997) that the Huronian
rocks may have accumulated at low latitudes.
2 7.2 Huronian-Age Glaciation 1081
Fig. 7.19 Tectonic map of North America (After Whitmeyer and
Karlstrom 2007) to show the distribution of Palaeoproterozoic
successions containing glaciogenic rocks. Dashed line shows the
possible extent of glacial ice during deposition of the Gowganda
Formation and equivalents. See text for discussion
1082 V.A. Melezhik et al.
7.2.3 Palaeoproterozoic Glacial Depositsof South Africa
Patrick G. Eriksson and Wladyslaw Altermann
At ~2500 Ma, much of the Kaapvaal craton was covered by a
shallow (water depth estimated to be�100 m) epeiric carbon-
ate platform, which is present in all three sub-basins where the
Transvaal Supergroup is preserved: Transvaal itself (northern
Kaapvaal), Kanye in Botswana (NW Kaapvaal) and
Griqualand West (SW of Kaapvaal craton) (Eriksson and
Altermann 1998) (Fig. 7.20). This carbonate platform lasted
from<2642 � 3Ma (Walraven andMartini 1995) until about
<2516 � 4 Ma (Altermann and Nelson 1998) (Fig. 7.21). At
c. 2500 Ma, based on SHRIMP dating of tuffs below and
above the base of the Kuruman and Penge banded iron forma-
tion (BIF) (see Fig. 7.21; Pickard 2003), a major drowning
event affected the entire epeiric basin, and carbonate deposi-
tion was replaced by accumulation of BIF under deep shelf-
type conditions (water depth possibly 100–200 m) (Altermann
and Nelson 1998). BIF deposition continued for at least 30 Ma
and possibly until 2432 � 31 Ma (Trendall et al. 1990);
precise zircon ages from the BIFs range from 2489 � 33 Ma
and 2480 � 6Ma to 2465 � 7Ma (Nelson et al. 1999; Martin
et al. 1998b; Pickard 2003) (Fig. 7.21). From c. 2432 Ma
onwards the Transvaal and Griqualand West basins do not
have a common development (Fig. 7.21).
Griqualand West Basin
The Kuruman BIF in the Griqualand West basin is grada-
tionally (e.g. Beukes 1984) succeeded by the Koegas Sub-
group (Fig. 7.21) composed of alternating clastic and
chemical (BIF and dolostones) sedimentary rocks. This
alternation was interpreted to have been caused by changes
from regressive prograding deltaic to transgressive basinal
systems (Beukes 1983; Eriksson et al. 2006). There is one
unpublished date for the Koegas units: the uppermost forma-
tion has yielded an age of 2415 � 6 Ma (cited by Kirschvink
et al. 2000 as a personal communication regarding unpub-
lished Pb-Pb data). Prior to deposition of the Makganyene
Formation, the Koegas rocks were affected by a major
thrusting event.
The Makganyene Formation lies stratigraphically above
the Koegas rocks with a regional, high-angle, deeply erosive
unconformity (Altermann and H€albich 1990, 1991). This
unconformity has penetrated deeply into the Koegas
Subgroup in the southwestern part of the Griqualand West
basin, and over the larger platform area to the northeast, it
even cuts down into the uppermost formation of the under-
lying Kuruman BIF (e.g. Altermann and Nelson 1998). From
boreholes in this northern part of the platform, it is known to
cut down to the Campbellrand carbonates below the
Kuruman BIF (Altermann, unpublished data). Preserved
thicknesses of the Makganyene Formation are highly vari-
able, from 3 to 70 m (Fig. 7.22a) to a known maximum of
500 m. The formation consists predominantly of massive to
coarsely bedded diamictites (Fig. 7.23a), with bedding being
indicated mostly by parallel orientation of elongated large
clasts. Lenticular, small-to-large-pebble conglomerates,
sandstones and mudrocks, some of the latter being varved
(Visser 1971; Polteau et al. 2006), occur in association with
the diamictites. Diamictite clasts are generally 0.5–30 cm
long, predominantly comprise BIF, with chert, sandstone
and uncommon carbonate compositions as well, and are set
in a fine-grained, ferruginous matrix (Polteau et al. 2006).
Well-defined striations (Figs. 7.23b, c) on large chert clasts,
as well as rafted stones and localised remnants of glacial
pavements underline the inferred glacial origin of the unit
(Visser 1971, 1999; Eyles and Januszczak 2004). Glaciation
is thought to have been limited in scale, and centred over the
Vryburg Rise between the Transvaal and Griqualand West
sub-basins (Fig. 7.20); a mountain glaciation is thus inferred,
with fluvial and marine reworking of tills having occurred
(Visser 1971). A precise depositional age of this important
glacigenic diamictite is not available.
The Makganyene Formation is overlain with a low-angle
unconformity by the Ongeluk lavas (Vajner 1974;
Altermann and H€albich 1991), which were possibly extruded
at a near-equatorial position (palaeomagnetic reconstruction
of Evans et al. 1997). The Ongeluk lavas form part of a large
flood-basalt province comprising the Hekpoort Formation in
the Transvaal basin (Figs. 7.20 and 7.21), and the Tsatsu
Formation in the Kanye basin. The lavas have been dated at
2224 � 21 Ma (Pb-Pb, Cornell et al. 1996), although this
result has been challenged (Moore et al. 2001; Polteau et al.
2006) on the basis of a Pb-Pb age (2394 � 26 Ma; Bau et al.
1999) obtained from the overlying Mooidraai Formation
carbonates, and on the inferred age of 2415 � 6 Ma for the
upper Koegas Subgroup, cited in Kirschvink et al. (2000).
This alternative viewpoint has not been reconciled with
regional relationships, nor with the thrusting of the Koegas
succession prior to deposition of Magkanyene diamictites.
Notably, the unconformity separating the Magkanyene and
Ongeluk formations has yet to be considered in modelling
the Snowball Earth scenario (Evans et al. 1997, and
subsequent publications).
P.G. Eriksson (*)
Department of Geology, University of Pretoria, Private Bag X20,
Hatfield, Pretoria-Tshwane 0028, South Africa
2 7.2 Huronian-Age Glaciation 1083
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_2, # Springer-Verlag Berlin Heidelberg 2013
1083
Transvaal Basin
In the northeast part of the Transvaal basin, the uppermost
members of the Penge BIF are cherty and shale-rich, and in
the past have been confusingly described as a carbonate-rich
succession of the so-called “Tongwane Formation” (viz.
Martini 1977). Although locally the Duitschland Formation
appears to have a conformable relationship to the
“Tongwane Formation”, elsewhere in the basin it rests with
an erosive contact and an angular unconformity either on
the Penge BIF (Fig. 7.21) or on carbonate rocks of the
Chuniesport Group. Moreover, prior to deposition of the
Duitschland rocks, the entire Chuniesport Group was
subjected to relatively gentle folding in the northeast of the
basin. Thus, there is a distinct hiatus between the BIF depo-
sition (minimum age 2465 � 7 Ma and possibly 2432 � 31
Ma, as discussed above) and the onset of Duitschland depo-
sition. However, the Duitschland Formation rocks them-
selves remain undated.
The Duitschland Formation is a difficult unit to quantify
in terms of characteristics (beyond variability) and inferred
genesis, but remains important due to ongoing studies of
carbon isotopic values supportive of glaciation and
Palaeoproterozoic atmospheric compositional changes. The
formation also includes two diamictite beds. The thickness
of the Duitschland Formation is highly variable, ranging
from as little as 15 m (e.g. H€albich et al. 1993) to
c. 1,000 m (e.g. Potgieter 1992; Bekker et al. 2001;
Frauenstein et al. 2009). The thickest development of the
Duitschland Formation occurs where the Penge BIF has
been totally removed and it lies directly upon carbonate
rocks. The maximum preserved thickness of the Penge BIF
Fig. 7.20 Sketch map showing the three Transvaal (Supergroup) sub-basins: Transvaal itself and Griqualand West (separated by the Vryburg
Rise, a palaeohigh), with the Kanye to the north
1084 V.A. Melezhik et al.
Fig. 7.21 Lithostratigraphy for the Chuniespoort-Ghaap Groups, in
the Transvaal and Griqualand West sub-basins, showing inferred
correlations, age data and interpreted regressive-transgressive trends.
The two left-hand columns refer to the Prieska and Ghaap Plateau
2 7.2 Huronian-Age Glaciation 1085
Fig. 7.22 Lithological profiles through three diamictite-bearing units
in the Kaapvaal craton. (a) Typical profile through the Makganyene
Formation, Griqualand West sub-basin (see Fig. 7.20); profile from
Visser (1971), measured on farm Bolham Ku. Q 825, situated about
45 km south of Kuruman. (b) Vertical profile through the Duitschland
Formation on the farm Duitschland, simplified from original in
Frauenstein et al. 2009. Note three subdivisions defined by the latter
authors: (1) thin basal diamictite succeeded by conglomerate-quartzite
couplet; (2) thick black shales, marls and thin carbonate interbeds; (3)an upper interval, with basal conglomerate-quartzite beds, and less
shale and more quartzite and carbonate beds within the interval, as
well as a second, thin diamictite. (c) Profile through the upper c. 50 m of
the Timeball Hill Formation showing 35-m-thick diamictite succeeded
by locally varved mudrocks and a thin chert conglomerate bed (Profile
measured by Pat Eriksson in Magaliesberg village (location in
Fig. 7.20))
Fig. 7.21 (continued) divisions of the Griqualand West sub-basin.
Note that vertical scale refers to time and not thickness. Note also
contact relationships with succeeding units of the Duitschland Forma-
tion, Pretoria and Postmasburg Groups (Modified after Eriksson et al.
2006 (references for age data: 1 – Walraven and Martini 1995; 2 –
Trendall et al. 1990; 3 – Nelson et al. 1999; 4 – Martin et al. 1998b; 5 –
Kirschvink et al. 2000; 6 – Cornell et al. 1996; 7 – Hannah et al. 2004;
8 – Trendall et al. 1995; 9 – Altermann and Nelson 1998; 10 – Sumner
and Bowring 1996; 11 – Barton et al. 1994))
1086 V.A. Melezhik et al.
elsewhere in the Transvaal basin is 640 m (e.g. Eriksson and
Altermann 1998). This would imply that the formation of
Duitschland rocks was related to major weathering and
erosion that occurred during the hiatus separating the
Chuniespoort and Pretoria groups in the Transvaal basin.
The Duitschland Formation is composed largely of
marlstones and mudrocks, with a combination of dolostones
and limestones being next in abundance, followed by minor
thin beds of quartzite, conglomerate and diamictite (e.g.
Frauenstein et al. 2009) (Fig. 7.22b). Although a “basin-
wide distribution” is claimed for the Duitschland Formation
(Bekker et al. 2001, citing Coetzee 2001), the unit is actually
restricted to two relatively limited outcrop areas in the
northeast of the Transvaal basin: (1) one around the town
of Mokopane (cf., old name, Potgietersrus) where most
detailed work has been done on a set of exposures on the
farms Duitschland and De Hoop (e.g. Bekker et al. 2001;
Frauenstein et al. 2009); (2) the other is some 40 km to the
east and southeast, where multiple outcrops of the
Duitschland Formation are a result of folding, leading to
Fig. 7.23 Makganyene diamictite on the Farm Neuwevlei, in
Griqualand West, where it forms lenses with a lateral extent of several
kilometres and is between 1 and 30 m thick (Altermann and H€albich1991). (a) Angular and rounded clasts of highly variable size are
embedded in shaley to sandy matrix. Below and within the diamictite
extensive fluvial sandstone lenses are interbedded and can be traced for
hundreds of meters laterally. (b) Striated clasts in the diamictite: chert
clast in carbonate matrix containing chiefly gritty, much smaller clasts
of chert, carbonate, and BIF. (c) Striated clasts in the diamictite:
striated quartzite in shale matrix (Photographs by Wlady Altermann)
2 7.2 Huronian-Age Glaciation 1087
repeated outcrops within the underlying deformed dolomite-
BIF succession (Fig. 7.24). In this area the Duitschland rocks
overstep the underlying Penge BIF and come to lie on older
carbonate rocks, and are relatively thin (e.g. Potgieter 1992;
H€albich et al. 1993). In the area around Mokopane, the
Duitschland also locally lies directly upon Chuniespoort
Group carbonates due to removal of intervening Penge BIF
prior to Duitschland deposition. In this area that the
Duitschland Formation attains its maximum thickness of c.
1,000 m, and is made up of three units: (1) a basal diamictite
overlain by a conglomerate-quartzite couplet; (2) an inferred
deep-water unit of black shales, laminated marlstones and
thin carbonate interbeds, with quartzites and dolostones in
the upper parts; and (3) an inferred shallow-water facies with
less shale, more quartzites and several carbonate beds
(Fig. 7.22b). The third unit contains a thin diamictite in its
upper levels and, at its base, a distinct conglomerate-
quartzite, interpreted by Bekker et al. (2001) as a “notable”
sequence boundary. The lowermost diamictite has been
interpreted as glacigenic (Bekker et al. 2001; Frauenstein
et al. 2009), based on heterogeneous clast content, consisting
mainly of chert, BIF and basement rocks, a basin-wide
distribution (which remains to be proven, currently known
outcrops being restricted to the NE part of the sub-basin),
and the presence of striations on bullet-shaped quartzite
pebbles and cobbles. Earlier workers did not go as far as a
definite glacigenic interpretation for these diamictites, and
preferred the vague term “tilloid” (Martini 1977; Potgieter
1992) with the latter describing them as “chert breccias”.
Although the Duitschland conglomerates and diamictites
contain locally derived BIF clasts as well as evidence for
sub-Transvaal basement sources, the predominant fragments
in the formation as a whole are claystones and carbonate
rocks. This material appears to have been derived from
outside the depositional site of the Duitschland Formation
in the region northeast of the preserved Transvaal basin, and
was possibly derived from uplifted areas to the south
(Eriksson et al. 2001). In the latter areas, up to 800 m of
Chuniespoort carbonates (as well, apparently, as all
overlying Penge BIF) were removed by erosion prior to
Pretoria Group deposition, specifically at the base of the
Rooihoogte Formation (Eriksson et al. 2001). Chert breccias
(Rooihoogte Formation) are thickest where the carbonate
rocks have been most strongly eroded.
The Rooihoogte Formation rests with a regionally defined
angular unconformity on the Duitschland rocks. Reworked
residual chert breccias of the Chuniespoort Group have been
redeposited in the northwestern part of the Rooihoogte basin
in the form of an alluvial lobe, up to 250 m thick, whereas it
is only 30 m thick in its northeastern part (Eriksson et al.
2001, their Fig. 7.18). For this reason a general equivalence
between the Duitschland and Rooihoogte formations is com-
monly cited (e.g. Bekker et al. 2001; Eriksson et al. 2001;
Frauenstein et al. 2009). Eriksson et al. (2001) postulate that
the Duitschland Formation, predominantly of marly
Fig. 7.24 Geological sketch map (After Bekker et al. 2001) of the NE portion of the Transvaal sub-basin, showing major outcrops of the
Duitschland Formation in the area of Mokopane/Potgietersrus and to the east thereof
1088 V.A. Melezhik et al.
material, might reflect resedimented Chuniespoort Group
carbonate detritus derived from the uplifted southern part
of the Transvaal basin, and transported down a south-to-
north palaeoslope, being deposited within a localized, deep
Duitschland basin in the northeast. This would equate
Duitschland deposition with Rooihoogte sedimentation.
The Rooihoogte Formation varies in thickness from 250 m
to less than 1 m and in many areas forms a karst-fill above
weathered Chuniespoort Group dolomites. It comprises
chert breccias, chert conglomerates, quartzitic sandstones
and mudrocks, with variable stacking patterns of these
lithologies, although the coarser rocks tend to be overlain
by the finer ones in many areas (Eriksson 1988).
The Timeball Hill Formation lies immediately above the
Rooihoogte rocks (Fig. 7.21). Black shales at the base of the
Timeball Hill Formation have been dated at 2316 � 7 Ma
(Re-Os; Hannah et al. 2004). The formation is made up
largely of mudrocks and is about 2 km thick in many areas,
with basal black mudrocks passing up into ‘normal’ shales,
which gradually coarsen up into siltstsones and fine
sandstones, ascribed to a relatively deep, rift-related portion
of an epeiric marine basin (Eriksson and Reczko 1998;
Catuneanu and Eriksson 1999; Eriksson et al. 2008). Medial
regressive fluvial sandstones are succeeded by a repetition of
the lower mudrock interval, with the inferred glacial
diamictites locally preserved within the upper c. 50-m-
thick part of the argillaceous succession (Eriksson and
Reczko 1998). The diamictites are associated with slumped
wackes and conglomerates, contain striated pebbles, and
are associated with varved shales (Visser 1971; Eriksson
et al. 1994).
One of the best diamictite outcrops occurs in
Magaliesberg (Fig. 7.20), where 35 m of diamictite
(Fig. 7.22c) consists of a predominant sandy-silty mudstone
matrix supporting subordinate clasts (5 % of the rock by
volume) ranging in diameter from 1 to 15 cm. Clasts are
predominantly chert, with minor sandstones and mudrocks.
Many clasts are preferentially oriented with long axes
approximately parallel to regional bedding. There rocks
display upward fining and a decrease in degree of roundness.
Together, these characteristics support reworking and a
periglacial setting, as suggested by Visser (1971). Above
the diamictites at Magaliesberg, the upper part of the
Timeball Hill Formation (Fig. 7.22c) comprises about 10 m
of locally varved shales and 3 m of chert pebble conglomer-
ate (glacio-fluvial?).
Lenticular bodies of similar diamictite (of unknown lat-
eral extent – possibly several hundred metres) have been
recorded in boreholes in the upper Timeball Hill Formation
in the southern part of the Transvaal basin (Coetzee et al.
2006). Shale interbeds in the diamictite, and shales immedi-
ately below the diamictites are characterised by soft-
sediment deformation. Faceted and bullet-shaped striated
pebbles have variable composition but are mostly chert.
Diamictite Correlation and Implicationfor Their Depositional Environments
The correlation of diamictites across the Transvaal – Kanye –
Griqualand West sub-basins is not straightforward. Within
the Griqualand West sub-basin, there is only one diamictite
unit, the Makganyene Formation (up to 500 m thick locally,
mostly only up to about 70 m). In the Transvaal sub-basin,
there are two thin beds of diamictite in the Duitschland
Formation and lenticular occurrences of reworked
diamictites in the upper Timeball Hill Formation (up to
100–200 m thick; Eriksson et al. 2001). The two Duitschland
Formation diamictite beds (only the lower is generally
acknowledged as glacigenic) are separated by c. 700 m of
intervening stratigraphy. The lensoidal occurrence of the
Timeball Hill Formation is about 2 km stratigraphically
above the top of the Duitschland Formation (Fig. 7.21).
The question thus remains: do the undated Makganyene
diamictites correlate with those of the undated Duitschland
Formation, or with the <2316 � 7 Ma–>2224 � 21 Ma
Timeball Hill lenses? Both Makganyene and Timeball Hill
diamictites are significantly thicker than those in the
Duitschland, and the former two units lie stratigraphically
close to the 2224 � 21 Ma Hekpoort-Ongeluk Formation
flood basalts, being separated from these overlying basaltic
andesites by a low angle unconformity in each case. The
Duitschland Formation appears to be related to the
Rooihoogte Formation, and the two together to the chrono-
logical hiatus separating Chuniespoort and Pretoria Groups
in the Transvaal sub-basin, a hiatus estimated at c. 80 My
(Eriksson and Reczko 1995; see also Mapeo et al. 2006,
who estimate this hiatus at c. 200 My in the Kanye sub-
basin). An analogous significant angular unconformity,
related to thrusting, separates the Koegas Subgroup at the
top of the Ghaap Group (equivalent to the Chuniespoort
Group in the Transvaal sub-basin) in the Griqualand West
sub-basin from the base of the equivalent there of the Pretoria
Group (the Postmasburg Group; Fig. 7.21), with the
Makganyene Formation being the basal unit of the
Postmasburg Group. Both lensoidal upper Timeball Hill
diamictites and the more widespread diamictites within the
Makganyene Formation are related by Visser (1971) to a
shared centre of montane glaciation lying on the Vryburg
Rise palaeohigh (Fig. 7.20) between the two sub-basins. The
balance of circumstantial evidence, such as it is currently,
thus favours a Makganyene-Timeball Hill correlation, with
the Duitschland Formation being a small and more localised,
earlier occurrence.
2 7.2 Huronian-Age Glaciation 1089
7.2.4 Palaeoproterozoic Glacial Depositsof Australia
Aivo Lepland
Diamictites of the Meteorite Bore Member (Kungarra For-
mation, Turee Creek Group, Mount Bruce Supergroup) that
contain polymict pebbles and boulders (up to 1 m) with
striations, facets and polished surfaces in siltstone matrix
(Trendall 1976, 1979) have been interpreted as debris flows
and turbidites accumulating in a glaciomarine environment
of a foreland basin (Martin 1999). In the type locality area in
the Hardey syncline (Fig. 7.25a), these diamictites reach a
thickness of 270 m and occur c. 1,800 m above the base of
the Kungarra Formation (Martin 1999), which marks the
conformable contact between the Turee Creek Group and
underlying Hamersley Group. The age of diamictites is
bracketed between 2449 � 3 Ma obtained from the
Woongarra Rhyolite in the upper part of the Hammersley
Group (Barley et al. 1997), and 2209 � 15 Ma derived from
the Cheela Springs Basalt within the unconformably
overlying Wyloo Group (Martin et al. 1998a). A penetrative
axial-planar cleavage is present in the matrix of diamictites
in the type locality area (Fig. 7.25b), resulting in alignment
of elongate clasts in the plane of the cleavage (Trendall
1976) without causing major shape or structural changes of
the clasts themselves (Fig. 7.25b). Circumferential flanges of
the matrix sediment have been formed in the stress shadow
zones at the sides of many clasts (Fig. 7.25c). In addition to
the type locality area, diamictites with ice-rafted polymict
lonestones and outsized clasts (Fig. 7.25d) have also been
identified at localities in the Duck Creek syncline
(Fig. 7.25a) and at Yeera Bluff, c. 200 km NNE of the
Duck Creek syncline (Martin 1999). There is no tectonic
fabric developed at these localities, and bending and disrup-
tion of layering in the host sedimentary rocks is preserved
around clasts, characteristic of glacial dropstones (Martin
1999). Diamictite intervals are thin (<0.5 m) in the Duck
Creek Syncline and at Yeera Bluff, and occur at the base of
the Kungarra Formation, immediately above the youngest
formation of the Hamersley Group (Boolgeeda Iron Forma-
tion). Martin (1999) considers the diamictite horizons in the
Hardey and Duck Creek synclines and at Yeera Bluff to be
coeval, related to the same glacial event. He explains the
thickness variation of diamictites and their stratigraphic
relation to thin BIFs that occur above the main part of
the Boolgeeda Iron Formation, and locally above the
diamictites, as due to facies differences. Diamictites in the
Hardey syncline were interpreted to represent laterally
emplaced turbidites and debris flows in an ice-proximal
environment, whereas those in the Duck Creek syncline
and at Yeera Bluff were considered as more distal, vertically
emplaced ice-rafted deposits (Martin 1999).
A. Lepland (*)
Geological Survey of Norway, Postboks 6315 Sluppen, NO-7491
Trondheim, Norway
1090 V.A. Melezhik et al.
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_2, # Springer-Verlag Berlin Heidelberg 2013
1090
Fig. 7.25 (a) Map showing the distribution of late Archaean and early
Palaeoproterozoic (2650–2060 Ma) rocks in the Hamersley Province,
Western Australia (Data from the Atlas of 1:250 000 Geological Series
Map Images, Western Australia; Geological Survey of Western
Australia January 2005 update). (b) Foliated diamictite of the Meteorite
Bore Member in the Hardey syncline; clasts within diamictite are
typically aligned in plane of the tectonic fabric; the head of the hammer
is 16 cm long. (c, d) Rhyolite pebbles with striations, facets and
polished surfaces from diamictite in the Hardey syncline; cleaved
matrix siltstone adheres to sides of pebbles in the stress shadow
zones; the lens cap on (d) is 5.8 cm in diameter. (e, f) Rhyolite
dropstones with bent layering in the host siltstone at the base of
Kungarra Formation in the Duck Creek syncline (Photos (b), (c) and
(e) by Aivo Lepland; (d) and (f) courtesy of Martin van Kranandonk)
2 7.2 Huronian-Age Glaciation 1091
7.2.5 Palaeoproterozoic Glacial Depositsof Fennoscandia
Victor A. Melezhik
The geological evolution of the Fennoscandian Shield dur-
ing pre-Huronian time has been exhaustively discussed in
Chap. 3. A brief description of the type locality in Finland,
and summary data relevant to the onset of the Huronian
glacial conditions in the region is given below. As in other
late Archaean cratons, the Archaean-Palaeoproterozoic tran-
sition was marked on the Fennoscandian Shield by emplace-
ment of many plume-generated 2505–2440 Ma layered
gabbro intrusions (Fig. 4a, b) (Vogel et al. 1998; Amelin
et al. 1995; Puchtel, et al. 1996, 1997; Hanski et al. 2001)
followed by rifting, uplift and erosional episodes. Pre-
Huronian rift-bound basins were filled by chemically mature
quartzites and up to 3,000 m of Sumian volcanic rocks
identified as continental flood basalts (Heaman 1997;
Puchtel, et al. 1996, 1997); these were apparently coeval
and comagmatic with 2400 Ma layered gabbro-norite
complexes (Melezhik and Sturt 1994; Puchtel, et al. 1996,
1997). Based on the present-day distribution of the layered
gabbro intrusions, dykes and coeval volcanic formations,
the areal extent of the Sumian flood basalt province is
c. 700,000 km2 (Melezhik 2006, Fig. 7.9a). However, prior
to crustal shortening associated with the 1940–1860 Ma
Kola orogeny (Daly et al. 2006), the areal extent of the
Sumian flood basalt province was very likely much greater.
Warm climatic conditions during pre-Huronian-glacial
time have been suggested based on geochemistry and min-
eralogy of a post-2505 Ma regolith developed on Archaean
pegmatites in the Pechenga Greenstone Belt (Sturt et al.
1994). A hot and wet climate (or rainwater with an anoma-
lously low pH, perhaps related to high atmospheric pCO2?)
may also be indicated by widespread deposition of mature
Sumian quartzites, which imply a high degree of chemical
weathering. Warm and humid climatic conditions appear to
be corroborated by an apparent sub-equatorial position of the
region at 07–27� (palaeomagnetic data obtained from the
2440 Ma mafic dykes, Mertanen et al. 1999).
The Sumian sedimentary-volcanic successions and
2505–2430 Ma layered gabbro-norites underwent significant
erosion, with removal of up to 2–3 km, followed by a phase of
rifting prior to deposition of Huronian-age equivalent rocks
(Melezhik 2006), locally termed Sariola (e.g. Ojakangas
et al. 2001a). Basal Sariolian rocks are commonly polymict
conglomerates filling juvenile intracratonic rift basins across
the Fennoscandian Shield (e.g. Lahtinen et al. 2008). How-
ever, Ojakangas et al. (2001a) suggested that rather than
deposition in separate rift basins, the rifts have simply pre-
served remnants of a more widespread sheet of glacial
deposits. In the Pechenga Belt (Neverskrukk Formation)
and the Per€apohja Belt (Sompuj€arvi Formation), the basal
conglomerates erode into the c. 2500 Ma General’skaya and
c. 2430 Ma Kemi layered gabbro-norite intrusions, respec-
tively, with some containing reworked clasts of the 2505 Ma
gabbro-norite.
Huronian-age glaciogenic deposits of Fennoscandia are
associated with the Sariolian sedimentary formations and
their equivalents (Marmo and Ojakangas 1984; Strand and
Laajoki 1993; Fig. 7.26). Because the first unequivocal evi-
dence of a glacial origin of Palaeoproterozoic rock was in the
Koli-Kaltimo area in eastern Finland (Salop 1983; Marmo
and Ojakangas 1984; Ojakangas et al. 2001a), the
Urkkavaara Formation is recognised as the stratotype of
the Huronian-age glacial deposit in Fennoscandia.
The Urkkavaara Formation
The Urkkavaara Formation has been previously assigned to
the Sariolian group (Fig. 7.27), although its precise deposi-
tional age remains unknown. Based on long-distance
lithostratigraphic correlation with Sariolian rocks dated else-
where on the Finnish and Russian sides of the
Fennoscandian Shield, the time of deposition is estimated
to be between 2450 and 2300 Ma (Marmo and Ojakangas
1984). A frost-shattered basement appears to be overlain by
a basal till (Fig. 7.28a, b) at the base of the Urkkavaara
Formation, but the contact is rarely exposed. Glacial rocks
of the Urkkavaara Formation per se rest on conglomerates
and arkosic sandstones of the Sarioli group and are overlain
by a deep palaeoweathering crust (Marmo et al. 1986).
Despite amphibolite facies metamorphism and folding and
faulting involving nappe tectonics, primary depositional
features are preserved on both a macro- and microscale.
The formation has a cumulative thickness of 200 m but its
original thickness remains unknown. It has been divided into
seven informal lithological members (Fig. 7.27), among
which one diamictite and two siltstone beds with lonestones
and dropstones have been recognised in the lower part of the
formation. Another diamictite bed, though much thinner,
was documented in the upper part of the succession.
The Lower and the Upper siltstone-argillite members
(Fig. 7.27) are characterised by parallel lamination
expressed by grey, graded siltstone and darker argillite
laminae. Lonestones are commonly felsic plutonic rocks
with rare clasts of argillite and greywacke. Many lonestones
V.A. Melezhik (*)
Geological Survey of Norway, Postboks 6315 Sluppen, NO-7491
Trondheim, Norway
Centre for Geobiology, University of Bergen, Allegaten 41, Bergen
N-5007, Norway
1092 V.A. Melezhik et al.
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_2, # Springer-Verlag Berlin Heidelberg 2013
1092
are dropstones having either pierced or bent lamination
beneath them, whereas laminae above the clasts are either
horizontal or gently arched (Marmo and Ojakangas 1984).
The Upper siltstone-argillite member has a gradational con-
tact with overlying diamictites (Fig. 7.28c). The Lower
graded sandstone member is comprised of alternating sand-
stone and siltstone beds, the latter containing lonestones.
Both contacts of the member are gradational. The Diamictite
member is represented by massive, matrix-supported
greywackes containing poorly sorted plutonic clasts and
rare intraclasts. Rare beds of sandstones have also been
documented. Although the member has gradational contacts
with the Upper siltstone-argillite (Fig. 7.28c) and the Upper
graded sandstone members, it has been observed to pass
laterally into the Upper siltstone-argillite, thus suggesting
overlapping deposition with and erosion of the lower three
members of the formation (Marmo and Ojakangas 1984).
The Upper graded sandstone member is composed mostly
of 10- to 100-cm-thick beds of graded coarse-grained
sandstones. Silty laminae with lonestones are common in
the lower half of the member whereas the upper part contains
conglomerate beds that become more common upwards
and eventually pass gradationally into the overlying unit,
the parallel-bedded conglomerate member. This member is
Fig. 7.26 Sedimentological features of Sarioli conglomerates. (a) A
thin bed of polymict conglomerate lies erosively on 2432 Ma layered
gabbro-norite and is overlain by amygdaloidal basalt at the base of the
Per€apohja Schist Belt. Clasts in the conglomerate are Archaean
granites, amphibolites and gneisses. (b) Drillcore showing mafic
matrix-supported, polymict conglomerate comprising clasts of vein
quartz (white) and gabbro-norite derived from the underlying
2505 Ma layered intrusion at the base of the Pechenga Greenstone
Belt (Photographs by Victor Melezhik)
2 7.2 Huronian-Age Glaciation 1093
Fig. 7.27 Lithological profile of the Urkkavaara Formation and its stratigraphic position in a generalised lithostratigraphic column of central
Finland (Modified by Victor Melezhik after Marmo and Ojakangas (1984) and Marmo et al. (1986))
1094 V.A. Melezhik et al.
Fig. 7.28 Sedimentological features of the Urkkavaara glacial
deposits and their basement. (a) Frost-shattered Archaean basement
below the Urkkavaara Formation; coin for scale is 24 mm. (b) Terres-
trial basal till. The Upper graded sandstone member is composed
mostly of 10- to 100-cm-thick beds of graded, coarse-grained
sandstones. Silty laminae with lonestones are common in the lower
half of the member whereas the upper part contains conglomerate beds
that become more common upwards; compass for scale is 11 cm long.
(c) Upper Siltstone-argillite member with dropstones, grading upward
into the Diamictite member; note the gradational contact; hammer
length is 60 cm in (Photographs courtesy of Jukka Marmo)
2 7.2 Huronian-Age Glaciation 1095
composed of thick beds of conglomerate interbedded with
pebbly sandstones with arkosic sandstone beds and lenses
of diamictites in the base. Upwards, the conglomerate
beds become gradually thicker, accompanied by increasing
clast size and decreasing matrix, resulting in dominance of
clast-supported textures. Clasts are mainly well-rounded
fragments of felsic plutonic rocks and siltstone-argillite.
Beds in this member display both normal and reverse grad-
ing. The upper contact of the member is generally
gradational, although there are some erosional features.
The cross-bedded conglomerate member is the uppermost
unit of the formation. The lower part of the member is
marked by massive, clast-supported, cobble- to boulder-
dominated conglomerate with lenses of diamictite. The
upper part is composed of small-pebble conglomerate and
pebbly arkosic sandstone with horizontal bedding, low-angle
cross-bedding, and large trough cross-bedding sets and
cosets. The member was subjected to erosion, so that its
original thickness remains unknown (Marmo et al. 1986).
Although neither striated nor faceted rock fragments, nor
scoured bedrock surfaces, have been reported from the
Urkkavaara Formation, the presence of abundant dropstones
having either pierced or caused downward bending of subja-
cent laminations in thinly laminated units associated with
diamicte might be considered among the best evidence
supporting a glaciogenic origin in such deformed and
metamorphosed rocks (Hambrey and Harland 1981, p. 14;
Marmo and Ojakangas 1984).
Deposition of the Urkkavaara Formation may have
involved two successive advance-retreat glacial cycles.
Marmo and Ojakangas (1984) suggested that overall
glaciomarine deposition took place within a nearshore marine
environment including grounded glaciers and floating
icebergs. The Lower dropstone-bearing unit and the Lower
graded sandstone unit, were interpreted to have formed in
front of the glacier during the first glacial advance. Glacial
retreat resulted in the accumulation of the Upper dropstone
unit followed by deposition of the diamictite. The Upper
graded sandstone, parallel- and cross-bedded conglomerate
succession was assigned to the second advance-retreat glacial
cycle, followed by isostatic uplift, erosion and deep
weathering. Palaeocurrent directions obtained from overlying
sedimentary rocks in the region suggest that sedimentary
material was transported from the east to the west (Marmo
and Ojakangas 1984). Similar transport directions were
documented in rocks underlying the Urkkavaara Formation.
Based on these observations it was suggested that the ice
probably also moved westward off the Karelian massif.
1096 V.A. Melezhik et al.
7.2.6 Palaeoproterozoic Snowball Earth?
Lee R. Kump, Victor A. Melezhik, WladyslawAltermann, Patrick G. Eriksson, Aivo Lepland,and Grant M Young
An overview of Huronian-era sedimentation from what are
now disparate continents reveals a common history involv-
ing rifting, banded iron formation and associated volcanism
in Australia and South Africa, with erosion, and glaciation.
These observations suggest widespread, perhaps global ice-
house conditions. But was this the first “Snowball Earth,”
with low-latitude continental ice sheets and global sea-ice
coverage (Evans et al. 1997; Kirschvink et al. 2000)?
Prior to the onset of the Huronian ice age, Archaean plate
reconstructions show the assembly of two apparent
supercontinents (Aspler and Chiarenzelli 1998), Kenorland,
comprising the North American, Fennoscandian and
Siberian shields (Williams et al. 1991), and the other, though
less definite, an assembly of the Zimbabwe, Kaapvaal,
Pilbara, Sao Francisco and Indian cratonic blocks. Both
supercontinents experienced protracted break-up driven by
inferred mantle plumes and associated intraplate rifting.
The break-up of Kenorland started at around 2450Ma in a
low palaeolatitude (0–20�; Christie et al. 1975) and was
associated with the formation of a large igneous province
including voluminous continental flood-basalts, giant
radiating dyke swarms and layered gabbro intrusions
(Heaman 1997; Vogel et al. 1998). Break-up was followed
soon after by the onset of ‘icehouse’ conditions. These are
expressed by three separate glacial intervals (Young 1970;
Miall 1983; Young et al. 2001), the lowermost of which is
separated, in southern, basinal areas, from continental flood-
basalts by a c. 2,000-m-thick, rift-bound, siliciclastic succes-
sion (Fig. 7.29). An upper age limit for the glaciogenic rocks
makes them older than 2219 Ma (Young et al. 2001; Long
2004).
The thickness of the Canadian Huronian glacial deposits
suggests an active hydrological regime. Likewise the com-
plex stratigraphy indicates waxing and waning of the ice,
which is similar to glacial activity during the Pleistocene.
Abundant evidence of waterlain sediments and the presence
of sediments formed in interglacial periods also imply a
relatively temperate glaciation. The presence of varve-like
sediments (Figs. 7.13b and 7.16c) suggests that glacial lakes
were subject to annual freeze-thaw cycles – an unlikely
situation under the extreme cold of the snowball Earth
(cf. Bekker and Eriksson 1999; Kirschvink et al. 2000).
The carbonate-rich Espanola Formation has been likened
to cap carbonates (Bekker et al. 2005) that are known from
many Neoproterozoic glaciogenic successions (e.g. Hoffman
and Schrag 2000). However, the absence of cap carbonate
above the other diamictite-bearing formations poses a
challenge to this interpretation, as does the thickness and
stratigraphic complexity of the Espanola Formation (e.g.
Bernstein and Young 1990). An alternative explanation for
the carbonate-rich Espanola Formation is that it represents
evaporites formed during a period of restricted circulation
prior to break-up and formation of a continental margin
(Burke and Dewey 1973; Young and Nesbitt 1985).
Attenuation of the probable southern supercontinent
involved crust- and mantle-driven magmatism followed by
rifting. In South Africa, this was manifested by an igneous
event and formation of extensive 2470–2430 Ma mafic tuffs
(Jones et al. 1975; Hamilton 1977). Almost coeval with this
was deposition of 2480–2465 Ma banded iron formations
(for references see Bekker et al. 2001). As in Kenorland, the
rifting and igneous events were followed by the onset of
‘icehouse’ environments. The South African Duitschland
Formation glacial diamictites have been constrained to
between 2450 and 2320 Ma (Hannah et al. 2004) and rest
with a regional unconformity on jaspilites, banded iron
formations, dolostones and quartzites (Fig. 7.21, Bekker
et al. 2001). The duration of this pre-glacial unconformity
would have been on the order of several millions of years.
The younger c. 2320–2200 Ma Makganyene diamictites are
separated from the Duitschland Formation diamictites both
geographically and by a hiatus (of unknown duration). In
Australia, a 2449 Ma large igneous province was also
emplaced equatorially (0–5�; Evans 2003) and accompanied
by deposition of the largest accumulation of Palaeopro-
terozoic banded iron formations (Barley et al. 1997; Pickard
2003). Here, there is no obvious depositional break between
the pre- and syn-glacial history (Martin 1999) (Figs. 7.21
and 7.22).
Overall, the nature of sedimentation during the Huronian
glaciation indicates an active hydrological regime, inconsis-
tent with the “hard snowball Earth” interpretation of these
deposits, and the challenges are similar to those levelled
against a Neoproterozoic hard Snowball with sea ice cover-
ing the tropical source of atmospheric moisture (Allen and
Etienne 2008). However, the occurrence of evidence for
glacial ice at low latitudes and likely low elevations suggests
glaciation considerably more extensive (and thus likely
more persistent) than that of the Pleistocene glacial intervals.
Perhaps, as has been argued for the Neoproterozoic glacial
diamictites, the Huronian deposits were lain down during the
deglaciation of a snowball Earth, rather than the glacial
zenith during which essentially no sediments would have
been deposited (Hoffman and Schrag 2002). The key to
resolving the present array of conflicting interpretations of
L.R. Kump (*)
Department of Geosciences, Pennsylvanian State University, 503
Deike Building, University Park, PA 16870, USA
2 7.2 Huronian-Age Glaciation 1097
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_2, # Springer-Verlag Berlin Heidelberg 2013
1097
Fig. 7.29 Tentative correlation of Huronian glacial units across dif-
ferent continents. The permanent disappearance of the mass-
independent fractionation of sulphur isotopes is taken as the base line
(shown by grey-lined white bar) for correlation. References for
radiometric dates are given in Fig. 7.7. Sulphur isotope data are from
Papineau et al. (2005, 2007), Guo et al. (2009), Reuschel et al. (2009),
and Chap. 6.2.1. Carbon isotope data are from Veizer et al. (1992),
Bekker et al. (2005), Guo et al. (2009), and Chap. 6.1.2
1098 V.A. Melezhik et al.
the Huronian glaciation is more precise correlation among
the continents and better palaeogeographic control on the
latitude of diamictite deposition.
Improving the Chronology of HuronianGlaciation
There is evidence from both supercontinents of multiple
episodes of glaciation occurring at low latitudes. Correlation
of these episodes, however, is problematic. There is compel-
ling evidence for at least three glaciations in the Huronian of
Canada, two, or less probably, three in South Africa, one or
two closely spaced advance-retreat cycles in Fennoscandia,
and one in Australia. It is unclear whether the lack of evi-
dence for multiple glaciations in Fennoscandia and Australia
is the consequence of postdepositional erosion or is in fact
the result of lack of glaciation.
Significant improvements in the geochronology of the
Huronian interval seem unlikely given the dearth of dateable
volcanic materials in these sequences. A more promising
stratigraphic tool may be the multiple isotopic composition
of pyrites. Archaean rocks exhibit a wide range of non-mass
dependent isotopic values (see Chap. 7.1), as do pyrites in
basal Huronian-age sequences in South Africa (Guo et al.
2009) and Canada (Papineau et al. 2007). The permanent
disappearance of mass-independent fractionation (MIF) of
sulphur occurs between the first and second diamictite in
SouthAfrica andCanada, suggesting that atmospheric oxygen
rose above the threshold for MIF during this time, and more-
over, that one can correlate the lower Duitschland with the
Ramsay Lake, the upper Duitschland with the Bruce, and the
Makganyene-Timeball Hill with the Gowganda diamictites
(Fig. 7.29). The presence of both MIF and mass-dependent
fractionation in the pre Huronian rocks (Reuschel et al. 2009),
and a pronounced mass-dependent fractionation in the
Huronian interval (see Chap. 6.2.1) from Fennoscandia sug-
gest that the glacial deposits likely correlate either with the
Gowganda/Makganyene-Timeball Hill or with the upper
Duitschland/Bruce diamictites (Fig. 7.29). Carbon isotope
data are available from sedimentary carbonates in all
continents (Fig. 7.29), but their relevance to correlation
remains uncertain until the Palaeoproterozoic global carbon
cycle ismore clearly understood, and the d13C reference curve
is better constrained.
What Caused Huronian Glaciation?
Models advanced for an explanation of the onset of the
Huronian global glacial event include (1) drawdown of
atmospheric CO2 as a result of increased weathering caused
by accretionary and collisional tectonics (Young 1991); (2)
lowering of CO2 concentrations due to enhanced
weathering of silicates caused by rifting of supercontinents
in low latitudes (Evans et al. 1997; Evans 2003); (3) elimi-
nation of the CH4 greenhouse by oxidation due to the rise of
O2 (Pavlov and Kasting 2000; Kasting 2004, 2005); (4)
methane greenhouse removal at the onset of oxygenic pho-
tosynthesis (Kopp et al. 2005); and (5) multiple causes
(Melezhik 2006). Evaluating these ideas and the Palaeopro-
terozoic snowball Earth hypothesis remains an important
challenge for the future and requires expanding the
palaeomagnetic and geochronologic database for
Huronian-age rocks worldwide and establishing robust
correlations among the glacial deposits. The FAR-DEEP
core materials present a unique opportunity to begin to
address some of these problems.
2 7.2 Huronian-Age Glaciation 1099
7.2.7 Implications of the FAR-DEEP Core 3A
Victor A. Melezhik
Significant improvements in the geochronology of the
Huronian interval represent one of the most challenging
tasks. Such improvements are unlikely in the Huronian and
Transvaal basins given the dearth of dateable volcanic
materials. However, the Fennoscandian successions represent
a different case. Pre-Huronian, Huronian, and post-Huronian
sequences contain abundant volcanic rocks (Fig. 7.29) ranging
in composition from mafic to andesitic and offer a great
potential for providing better geochronological constraints
on theHuronian interval. The FAR-DEEPHole 3A intersected
various lithologies associated with rocks of undoubted glacial
origin (Fig. 7.30) and provides the most complete succession
outside of, and an apparent time-equivalent to, the Huronian-
age glacial deposits in Canada and South Africa. Carbonate-
shale “varves” with dropstones rest depositionally on
andesites of the Seidorechka Volcanic Formation and are
overlain by volcanic rocks ranging in composition from
mafic komatiites to andesites. Intensive volcanism throughout
the succession suggests that some dateable volcanic ash beds
may be found within the glacial unit as well.
Unlike other Huronian-age successions in Scandinavia,
the drilled section contains a high proportion of carbonate
rocks, mainly shaly and clayey limestones with subordinate
lime-rich varves. This is somewhat reminiscent of the sec-
ond glacial unit (Bruce Formation) in North America where
carbonates (Espanola Formation) rest directly above glacial
diamictites. However, in Scandinavia, only one glacial unit
is known, and in the drilled section limestones are stratigra-
phically below it and contain dropstones. d13C values fluc-
tuate within a narrow range (�2.2 to +0.8 ‰ V-PDB) of
“normal seawater” values. d18O shows a strong depletion in18O with d18O ranging between 8.0 and 12.2 ‰ V-SMOW
(see Chap. 6.1.2). A similar phenomenon, namely, “normal”,
near-“zero” d13C values, in combination with depleted d18Ovalues with no evidence of obvious post-depositional alter-
ation, has been reported from the Espanola Formation
limestones (Veizer et al. 1992). Here, Robertson (1964,
1986), Young (1973) and Bernstein and Young (1990) pro-
posed that the basinal infill accumulated either in a shallow
sea or in a large lake during the retreat of the Bruce glaciers.
Veizer et al. (1992) further concluded that 18O depletion
could have been caused by influx of high latitude/altitude
melt-water. The high Sr content in the Polisarka limestones
is in conflict with a lacustrine depositional system, and thus,
at this stage, we leave the interpretation of the C- and O-
isotopic data open for future detailed research.
An enigmatic feature is also associated with the
Huronian-age Sariola conglomerates at the base of the
Pechenga Belt, in the Neverskrukk Formation. The previ-
ously drilled hole 3462 intersected a 300-m-thick forma-
tion of conglomerate, gritstone and sandstone containing
several thin beds of calcite-cemented pebbly polymict
conglomerates (Figs. 6.39ac, ae and 6.43a from Chap.
6.2.1) located approximately 35–60 m above the
formational base. A similar 2-m-thick horizon of calcite-
cemented conglomerate (Figs. 6.43b, c and 7.30f) was
documented in the surface outcrop located at Brattli,
35 km northwest of the drilling site (Fig. 4.15 in Chap.
4.2). In both cases the calcite occurs in the form of sparite
surrounding and, in some cases, supporting clasts. This
suggests a very early cementation, which was apparently
accomplished in pore space prior to early compaction, and
thus likely associated either with surface or groundwaters.
Earlier isotopic measurements from the surface outcrop at
Brattli yielded �4.0 ‰ and 8.6 ‰ for d13C and d18O,respectively (Melezhik and Fallick 1996).
Understanding the spatial and temporal links of Protero-
zoic carbonates associated with glacial rocks, both those that
are overlying, known as “cap carbonates” (e.g. Shields 2005)
and those occurring below, like the Polisarka limestones
(perhaps termed, “basal carbonates”), represents a challenge
and deserves more attention. Several conflicting models
have been advanced for the formation of Neoproterozoic
(reviewed in Shields 2005), and putative Palaeoproterozoic
(Bekker et al. 2005), “cap carbonates”. However these post-
glacial “greenhouse” deposits are not analogous to the
Polisarka limestones, which likely accumulated under a
cold climate.
Observations made on some recent seasonal river ice in
Siberia point to the fact that a significant volume of
carbonates can be “expelled” from the ice melt waters
(Fig. 7.31). Melt-ice-carbonates have been seen occurring
as infill in coarse clastic river-bed sediments (Fig. 7.31i).
During massive seasonal ice melt such sources provide a
considerable volume of carbonates, and several major
Siberian rivers have been observed discharging “milky”
water rich in suspended, micron-size calcite particles.
Washed into a basin, such carbonate components, in prin-
ciple, may form carbonate-shale varves with seasonal
lamination. The environment illustrated by Fig. 7.31h, i
represents a likely model for the formation of Sariolian
calcite-cemented conglomerates (Fig. 7.30f). We also spec-
ulate that such carbonate material may represent the source
for “cap carbonates” deposited on top of glaciomarine
deposits.
V.A. Melezhik (*)
Geological Survey of Norway, Postboks 6315 Sluppen, NO-7491
Trondheim, Norway
Centre for Geobiology, University of Bergen, Allegaten 41, Bergen
N-5007, Norway
1100 V.A. Melezhik et al.
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_2, # Springer-Verlag Berlin Heidelberg 2013
1100
Fig. 7.30 FAR-DEEP Core 3A representing glacial and associated rocks from the Polisarka Sedimentary Formation, and surface outcrop of the
Neverskrukk Formation; both apparent stratigraphic equivalents of Huronian-age glacial deposits elsewhere on the Fennoscandian Shield
2 7.2 Huronian-Age Glaciation 1101
Fig. 7.30 (continued) (a) Clasts of andesite in diamictite of
glaciomarine origin; large, bright fragment may represent tectonically
modified faceted clasts. (b) Finely laminated limestone-shale couplets
in a varve-like rock deposited in a distal glaciomarine environment. (c)
Glaciomarine diamictite containing tectonically flattened clasts of
andesite, quartzite, granite, limestone and schist emplaced in originally
massive clayey siltstone matrix; apparent fine lamination is due to
extremely flattened incompetent clasts. (d) Parallel-bedded, fine-
grained greywacke beds with shale (pale brown) layers. (e) Beddedlimestone; a high Sr content (760–1030 mg·g�1) suggests aragonite
precursor and may provide a strong buffer for Sr-isotope systematics.
(f) Calcite-cemented polymict conglomerate of the Neverskrukk For-
mation, Pechenga Greenstone Belt; note that calcite cement is not
corrosive, fills available space and in several places supports clasts
(Photographs by Victor Melezhik)
1102 V.A. Melezhik et al.
Fig. 7.31 Seasonal river ice filling part of Sukhoy Kumikulakh river
bed in central Siberia. (a) A general view of the river ice in mid July;
note that the surface and the fringe are veneered with calcite “powder”
(micron-size calcite crystals); hammer length is 45 cm. (b) Close-up
view of the ice surface covered with soft clumps comprised of micron-
size crystals of calcium carbonate. (c) Close-up view of ice fringe
covered with thick, clumpy, soft crust of calcite, which also occurs as
a pocket of clumps in melting ice
2 7.2 Huronian-Age Glaciation 1103
Fig. 7.31 (continued) (d) Vertical profile through the river ice
showing that during the course of the melt carbonate material gradually
accumulated on the ice surface; note that different ice layers show
variable content of expelled calcite, which occurs as dusty particles
reducing transparency of the ice. (e) River ice surface unevenly covered
with brownish calcite crystals, which gradually accumulated in the
form of clumps during the course of ice melt; hammer length is 45.
(f, g) Close-up views of some of the calcite clumps demonstrating that
accumulated calcite occurs in the form of bladed crystals
1104 V.A. Melezhik et al.
References
Allen PA, Etienne JL (2008) Sedimentary challenge to snowball Earth.
Nat Geosci 1:817–825
Altermann W, H€albich IW (1990) Thrusting, folding and stratigraphy
of the Ghaap Group along the southwestern margin of the Kaapvaal
craton. S Afr J Geol 93:553–556
Altermann W, H€albich IW (1991) Structural history of the southwest-
ern corner of the Kaapvaal craton and the adjacent Namaqua realm:
new observations and reappraisal. Precambrian Res 52:133–166
Altermann W, Nelson DR (1998) Sedimentation rates, basin analysis
and regional correlations of three Neoarchaean and Palaeopro-
terozoic sub-basins of the Kaapvaal craton as inferred from precise
U-Pb zircon ages from volcaniclastic sediments. Sediment Geol
120:225–256
Amelin YuV, Heaman LM, Semenov VS (1995) U-Pb geochronology
of layered mafic intrusions in the eastern Baltic Shield: implications
for the timing and duration of Palaeoproterozoic continental rifting.
Precambrian Res 75:31–46
Aspler LB, Chiarenzelli JR (1998) Two Neoarchean supercontinents?
Evidence from the Paleoproterozoic. Sediment Geol 120:75–104
Aspler LB, Wisotzek IE, Chiarenzelli JR, Losonczy MF, Cousens BL,
McNicoll VJ, Davis WJ (2001) Paleoproterozoic intracratonic pro-
cesses from breakup of Kenorland to assembly of Laurentia:
Hurwitz basin, Nunavut, Canada. Sediment Geol 43:287–318
Barley ME, Pickard AL, Sylvester PL (1997) Emplacement of a large
igneous province as a possible cause of banded iron formation 2.45
billion years ago. Nature 385:55–58
Barton ES, Altermann W, Williams IS, Smith CB (1994) U-Pb zircon
age for tuff in the Campbell Group, Griqualand West sequence,
South Africa: implications for early Proterozoic rock accumulation
rates. Geology 22:343–346
Bau M, Romer RL, L€uders V, Beukes NJ (1999) Pb, O, and C isotopes
in silicified Mooidraai dolomite (Transvaal Supergroup, South
Africa): implications for the composition of Paleoproterozoic sea-
water and “dating” the increase of oxygen in the Precambrian
atmosphere. Earth Planet Sci Lett 174:43–57
Bekker A, Eriksson KA (1999) Paleoproterozoic record of biogeo-
chemical events and ice ages. Geol Soc Am Abstr Progr 31:A–487
Bekker A, Kaufman AJ, Karhu JA, Beukes NJ, Swart QD, Coetzee LL,
Eriksson KA (2001) Chemostratigraphy of the Paleoproterozoic
Duitschland Formation, South Africa: implications for coupled
climate change and carbon cycling. Am J Sci 301:261–285
Fig. 7.31 (continued) (h) When river ice melts away, boulders
beneath it get covered while the space between boulders is gradually
filled with the carbonate powder; width of the view is 2 m. (i) Close-up
views of bolders veneered by white calcite left after the ice melted
away; note that free interclast space is also filled with calcite. Coin
diameter in (b), (c), (f), (g) and (i) is 2 cm (Photographs by Victor
Melezhik)
2 7.2 Huronian-Age Glaciation 1105
Bekker A, Kaufman AJ, Karhu JA, Eriksson KA (2005) Evidence for
Paleoproterozoic cap carbonates in North America. Precambrian
Res 137:167–206
Bell RT (1970) The Hurwitz Group – a prototype for deposition on
metastable cratons. In: Baer AJ (ed) Basins and geosynclines of the
Canadian Shield. Geol Surv Can Pap 70–40:159–169
Bernstein L, Young GM (1990) Depositional environments of the early
Proterozoic Espanola Formation, Ontario, Canada. Can J Earth Sci
27:539–551
Beukes NJ (1983) Paleoenvironmental setting of iron formations in the
depositional basin of the Transvaal Supergroup, South Africa. In:
Trendall AF, Morris RC (eds) Iron-Formation: facts and problems.
Elsevier, Amsterdamm, pp 131–210
Beukes NJ (1984) Sedimentology of the Kuruman and Griquatown
iron-formations, Transvaal Supergroup, Griqualand West, South
Africa. Precambrian Res 24:47–84
Blackwelder E (1926) Pre-Cambrian geology of the Medicine Bow
Mountains. Geol Soc Am Bull 37:615–658
Bleeker W, Ernst R (2006) Short-lived mantle generated magmatic
events and their dike swarms: the key unlocking earth’s paleogeo-
graphic reord back to 2.6 Ga. In: Hanski E, Mertanen S, R€am€o T,
Vuollo J (eds) Dyke swarms – time markers of crustal evolution. A.
A. Balkema, Rotterdam, pp 3–26
Burke K, Dewey JF (1973) Plume-generated triple junctions: key
indicators in applying plate tectonics to old rocks. J Geol
81:406–433
Cannon WF (1973) The Penokean orogeny in northern Michigan. In:
Young GM (ed) Huronian stratigraphy and sedimentation. Geol
Assoc Can, Spec Pap 12: 251–271
Card KD, Innes DG, Debicki RL (1977) Stratigraphy, sedimentology
and petrology of the Huronian Supergroup in the Sudbury-Espanola
area. Ont Div Mines Geosci Study 16:98
Catuneanu O, Eriksson PG (1999) The sequence stratigraphic concept
and the Precambrian rock record: an example from the 2.7–2.1 Ga
Transvaal Supergroup, Kaapvaal craton. Precambrian Res
97:215–251
Chandler FW (1973) Clastic dykes at Whitefish Falls and the base of
the Gowganda Fomation. In: Young GM (ed) Huronian stratigraphy
and sedimentation. Geol Assoc Can, Spec Pap 12: 199–209
Chown EH, Gobeil A (1990) Clastic dykes of the Chibugamau Forma-
tion: distribution and origin. Can J Earth Sci 27:1111–1114
Christie KW, Dawidson A, Fahrig WE (1975) The paleomagnetism of
the Kaminak dikes – No evidence of significant Hudsonian plate
motion. Can J Earth Sci 12:2048–2064
Coetzee LL (2001) Genetic stratigraphy of the Paleoproterozoic Pretoria
Group in the western Transvaal, M.Sc. thesis, Rand Afrikaans
University, Johannesburg, p 184
Coetzee LL, Beukes NJ, Gutzmer J, Kakegawa T (2006) Links of
organic carbon cycling and burial to depositional depth gradients
and establishment of a snowball Earth at 2.3 Ga. Evidence from the
Timeball Hill Formation, Transvaal Supergroup, South Africa.
S Afr J Geol 109:109–122
Coleman AP (1907) A lower Huronian ice age. Am J Sci 23:187–192
Coleman AP (1908) The lower Huronian ice age. J Geol 16:149–158
Cornell DH, Sch€utte SS, Eglington BL (1996) The Ongeluk basaltic
andesite formation in Griqualand West, South Africa: submarine
alteration in a 2222 Ma Proterozoic sea. Precambrian Res 79:
101–123
Dahl PS, Hamilton MA, Wooden JL, Foland KA, Frei R, McCombs JA,
Holm DK (2006) 2480 Ma mafic magnetism in the northern Black
Hills, South Dakota: a new link connecting the Wyoming and
Superior cratons. Can J Earth Sci 43:1579–1600
Daly JS, Balagansky VV, Timmerman MJ, Whitehouse MJ (2006) The
Lapland-Kola orogen: Palaeoproterozoic collision and accretion of
the northern lithosphere. In: Gee DG, Stephenson RA (eds)
European lithosphere dynamics, vol 32, Geological Society of
London Memoirs. Geological Society, London, pp 561–578
Eriksson PG, Bose PK, Catuneanu O, Sarkar S, Banerjee S (2008)
Precambrian clastic embayments: examples from South Africa
and India, In: Holmden C, Pratt BR (eds) Dynamics of epeiric
seas: sedimentological, paleontological and geochemical
perspectives. Geol Assoc Can, Spec Pap 48: 119–136
Eriksson PG (1988) Sedimentology of the Rooihoogte Formation,
Transvaal sequence. S Afr J Geol 91(4):477–489
Eriksson PG, Reczko BFF (1995) The sedimentary and tectonic setting
of the Transvaal Supergroup floor rocks to the Bushveld complex.
J Afr Earth Sci 21:487–504
Eriksson PG, Altermann W (1998) An overview of the geology of the
Transvaal Supergroup dolomites (South Africa). Environ Geol 36
(1–2):179–188
Eriksson PG, Reczko BFF (1998) Contourites associated with pelagic
mudrocks and distal delta-fed turbidites in the lower Proterozoic
Timeball Hill Formation epeiric basin (Transvaal Supergroup),
South Africa. Sediment Geol 120:319–335
Eriksson PG, Reczko BFF, Merkle RKW, Schreiber UM, Engelbrecht
JP, Res M, Snyman CP (1994) Early Proterozoic black shales of the
Timeball Hill Formation, South Africa: volcanogenic and palaeoen-
vironmental influences. J Afr Earth Sci 18:325–337
Eriksson PG, Altermann W, Catuneanu O, van der Merwe R, Bumby
AJ (2001) Major influences on the evolution of the 2.67–2.1 Ga
Transvaal basin, Kaapvaal craton. Sediment Geol 141–142:
205–231
Eriksson PG, Altermann W, Hartzer FJ (2006) The Transvaal Super-
group and its precursors. In: Johnson MR, Anhaeusser CR, Thomas
RJ (eds) The geology of south Africa. Geological Society of South
Africa, Johannesburg and Council for Geoscience, Pretoria,
pp 237–260
Eskola PE (1919) Hufvudragen av Onega-Karelens geologi, Helsingin
geol. Yhd. Tiedonantoja 1917 u. 1918, 13–18, and Teknikern, 1919:
37–39
Evans DAD (2003) A fundamental Precambrian-Phanerozoic shift in
earth’s glacial style? Tectonophysics 375:353–385
Evans DA, Beukes NJ, Kirschvink JL (1997) Low latitude glaciation in
the Palaeoproterozoic era. Nature 386:262–266
Eyles N, Januszczak N (2004) ‘Zipper-rift’: a tectonic model for
Neoproterozoic glaciations during the breakup of Rodinia after
750 Ma. Earth Sci Rev 65(1–2):1–73
Frauenstein F, Veizer J, Beukes N, van Niekerk HS, Coetzee LL (2009)
Transvaal Supergroup carbonates: implications for Paleopro-
terozoic d18O and d13C records. Precambrian Res 175:149–160
Gair JE (1981) Lower Proterozoic glacial deposits of northern
Michigan, USA. In: Hambrey MJ, Harland WB (eds) Earth’s pre-
Pleistocene glacial record. Cambridge University Press, Cambridge,
pp 803–806
Guo Q, Strauss H, Kaufman AJ, Schr€oder S, Gutzmer J, Wing B, Baker
MA, Bekker A, Jin Q, Kim S-T, Farquhar J (2009) Reconstructing
earth’s surface oxidation across the Archean-Proterozoic transition.
Geology 37:399–402
H€albich IW, Scheepers R, Lamprecht DF, van Deventer JL, De Kock
NJ (1993) The Transvaal-Griqualand West banded iron for-
mation: geology, genesis, iron exploitation. J Afr Earth Sci 16:
63–120
Halls HC, Bates MP (1990) The evolution of the 2.45 Ga Matachewan
dyke swarm, Canada. In: Parker AJ, Rickwood PC, Tucker DH
(eds) Mafic dykes and emplacement mechanism. Balkema,
Rotterdam, pp 237–250
Hambrey MJ, Harland WB (1981) Earth’s pre-Pleistocene glacial
record. Cambridge University Press, Cambridge, p 1004
Hamilton J (1977) Sr isotope and trace element studies of the Great
Dyke and Bushveld mafic phase and their relation to early Protero-
zoic magma genesis in southern Africa. J Petrol 18:24–52
1106 V.A. Melezhik et al.
Hannah JL, Bekker A, Stein HJ, Markey RJ, Holland HD (2004)
Primitive Os and 2316 Ma age for marine shale: implications for
Paleoproterozoic glacial events and the rise of atmospheric oxygen.
Earth Planet Sci Lett 225:43–52
Hanski E, Huhma H, Vaasjoki M (2001) Geochronology of northern
Finland: a summary and discussion. In: Vaasjoki M (ed) Radiomet-
ric age determination from Finnish Lapland and their bearing on the
timing of Precambrian volcano-sedimentary sequences. Geol Surv
Finl Spec Pap 33:255–279
Heaman LM (1997) Global mafic volcanism at 2.45 Ga: remnants of an
ancient large igneous province? Geology 25:299–302
Hoffman PF, Schrag P (2000) Snowball earth. Sci Am 282:68–75
Hoffman PF, Schrag DP (2002) The snowball earth hypothesis: testing
the limits of global change. Terra Nova 14:129–155
Houston RS, Lanthier LR, Karlstrom KK, Sylvester G (1981) Early
Proterozoic diamictites of southern Wyoming. In: Hambrey MJ,
Harland WB (eds) Earth’s pre-Pleistocene glacial record.
Cambridge University Press, Cambridge, pp 795–799
Houston RS, Kalstrom KE, Graff PJ, Flurkey AJ (1992) New strati-
graphic subdivisions and redefinitions of subdivisions of Late
Archean and Early Proterozoic metasedimentary and metavolcanic
rocks of the Sierra Madre and Medicine Bow Mountains, southern
Wyoming. U.S. Geol Surv Prof Pap 1520: 50
James HL (1958) Stratigraphy of pre-Keweenawan rocks in parts of
northern Michigan. United States Geol Surv Prof Pap 314-C:
27–44
Jones DL, Robertson DM, McFadden PL (1975) A palaeomagnetic
study of Precambrian dyke swarms associated with the Great
Dyke of Rodesia. Trans Geol Soc S Afr 78:57–65
Junnila RM, Young GM (1995) The Paleoproterozoic upper Gowganda
Formation, Whitefish Falls area, Ontario, Canada: subaqueous
deposits of a braid delta. Can J Earth Sci 32:197–209
Karlstrom KE, Flurkey AT, Houston RS (1984) Stratigraphy and depo-
sitional setting of Proterozoic metasedimentary rocks of southeast-
ern Wyoming: record of an early Proterozoic Atlantic-type cratonic
margin. Geol Soc Am Bull 94:1257–1294
Kasting JE (2004) When methane made climate. Sci Am 291:78–85
Kasting JE (2005) Methane and climate during the Precambrian era.
Precambrian Res 137:119–129
Kirschvink JL, Gaidos EJ, Bertani E, Beukes NJ, Gutzmer J, Maepa
LN, Steinberger RE (2000) Paleoproterozoic snowball earth:
extreme climatic and geochemical global change and its biological
consequences. Proc Natl Acad Sci 7:1400–1405
Kopp RE, Kirschvink JL, Hilburn IA, Nash CZ (2005) The Paleopro-
terozoic snowball earth: a climatic disaster triggered by the evolu-
tion of oxygenic photosynthesis. Proc Natl Acad Sci USA 102:
11131–11136
Kurtz DD (1981) Early Proterozoic diamictites of the Black Hills,
South Dakota. In: Hambrey MJ, Harland WB (eds) Earth’s pre-
Pleistocene glacial record. Cambridge University Press, Cambridge,
pp 800–802
Lahtinen R, Garde AA, Melezhik VA (2008) Paleoproterozoic evolu-
tion of Fennoscandia and Greenland. Episodes 31:20–28
Lindsey DA (1969) Glacial sedimentation of the Precambrian
Gowganda Formation, Ontario, Canada. Geol Soc Am Bull
80:1685–1704
Long DGF (1974) Glacial and paraglacial genesis of conglomeratic
rocks of the Chibougamau Formation (Aphebian), Chibougamau,
Quebec. Can J Earth Sci 11:1236–1252
Long DGF (1981) Glaciogenic rocks in the early Proterozoic
Chibougamau Formation of northern Quebec. In: Hambrey MJ,
Harland WB (eds) Earth’s pre-Pleistocene glacial record.
Cambridge University Press, Cambridge, pp 817–820
Long DGF (2004) The tectonostratigraphic evolution of the Huronian
basement and subsequent basin fill: geological constraints on
impact models of the Sudbury event. Precambrian Res 129:203–223
Mapeo RBM, Armstrong RA, Kampunzu AB, Modisi MP, Ramokate
LV, Modie BNJ (2006) A ca. 200 Ma hiatus between the lower and
upper Transvaal Groups of southern Africa: SHRIMP U-Pb detrital
zircon evidence from the Segwagwa Group, Botswana: implications
for Palaeoproterozoic glaciations. Earth Planet Sci Lett 244:
113–132
Marmo J, Kohonen J, Sarap€a€a O, Aik€as O (1986) Sedimentology, and
stratigraphy of the lower Proterozoic Sariola and Jatuli groups in the
Koli-Kaltimo area, eastern Finland. In: Laajoki K, Paakkola J (eds)
Geol Surv Finl, Spec Pap 5: 11–28
Marmo JS, Ojakangas RW (1984) Lower Proterozoic glaciogenic
deposits, eastern Finland. Geol Soc Am Bull 98:1055–1062
Martin DMcB (1999) Depositional setting and implications of
Paleoproterozoic glaciomarine sedimentation in the Hamersley
Province, Western Australia. Geol Soc Am Bull 111:189–203
Martin DMcB, Li ZX, Nemchin AA, Powell CM (1998a) A pre-2.2 Ga
age for giant hematite ores of the Hamersley Province, Australia.
Econ Geol 93:1084–1090
Martin DMcB, Clendenin CW, Krapez B, McNaughton NJ (1998b)
Tectonic and geochronological constraints on late Archaean and
Palaeoproterozoic stratigraphic correlation within and between the
Kaapvaal and Pilbara cratons. J Geol Soc Lond 155:311–322
Martini JEJ (1977) A copper-bearing horizon in the Pretoria Group in
the north-eastern Transvaal, Abstracts, Geocongress 77,
Johannesburg, South Africa. Geological Society of South Africa,
Johannesburg, pp 57–59
Martini JEJ (1979) A copper-bearing bed in the Pretoria Group in
Northeastern Transvaal. Geol Soc S Afr Spec, Pub 6:65–72
Melezhik VA (2006) Multiple causes of earth’s earliest global glacia-
tion. Terra Nova 18:130–137
Melezhik VA, Fallick AE (1996) A widespread positive d13Ccarb anom-
aly at around 2.33–2.06 Ga on the Fennoscandian Shield: a para-
dox? Terra Nova 8:141–157
Melezhik VA, Sturt BA (1994) General geology and evolutionary
history of the early Proterozoic Polmak-Pasvik-Pechenga-
Imandra/Varzuga-Ust’Ponoy Greenstone Belt in the north-eastern
Baltic Shield. Earth Sci Rev 36:205–241
Mertanen S, Halls HC, Vuollo JL, Pesonen LJ, Stepanov VS (1999)
Paleomagnetism of 2.44 Ga mafic dykes in Russian Karelia, eastern
Fennoscandian Shield – implications for continental
reconstructions. Precambrian Res 98:197–221
Miall AD (1983) Glaciomarine sedimentation in the Gowganda Forma-
tion (Huronian), northern Ontario. J Sediment Res 53:477–491
Moore JM, Tsikos H, Polteau S (2001) Deconstructing the Transvaal
Supergroup, South Africa: implications for Palaeoproterozoic
palaeoclimate models. J Afr Earth Sci 33:437–444
Morey GB (1973) Stratigraphic framework of Middle Precambrian
rocks in Minnesota. In: Young GM (ed) Huronian stratigraphy and
sedimentation. Geol Assoc Can Spec Pap 12: 211–249
Negrutza VZ (1984) Early Proterozoic stages of evolution of the east-
ern Baltic Shield. Nedra, Leningrad, p 270 (in Russian)
Negrutsa TF, Negrutsa VZ (1981a) Early Proterozoic Yanis-J€arvitilloids, South Karelia, USSR. In: Hambrey MJ, Harland WB
(eds) Earth’s pre-Pleistocene glacial record. Cambridge University
Press, Cambridge, pp 681–682
Negrutsa TF, Negrutsa VZ (1981b) Early Proterozoic Sarioli tilloids in
the eastern part of the Baltic Shield, USSR. In: Hambrey MJ,
Harland WB (eds) Earth’s pre-Pleistocene glacial record.
Cambridge University Press, Cambridge, pp 683–686
Nelson DR, Trendall AF, Altermann W (1999) Chronological
correlations between the Pilbara and Kaapvaal cratons. Precambrian
Res 97:165–189
Ojakangas RW (1985) Evidence for early Proterozoic glaciation: the
dropstone unit-diamictite association. In: Laajoki K, Paakkola J
(eds) Proterozoic exogenic processes and related metallogeny.
Geol Surv Finl Bull 331: 51–72
2 7.2 Huronian-Age Glaciation 1107
Ojakangas RW (1988) Glaciation: an uncommon mega-event as a key
to intracontinental and intercontinental correlation of early Protero-
zoic basin fill, North American and Baltic cratons. In: Kleinspehn
KL, Paola C (eds) New perspectives in basin analysis. Springer,
Berlin, pp 431–444
Ojakangas RW, Marmo JS, Heiskanen KI (2001a) Basin evolution of
the Paleoproterozoic Karelian Supergroup of the Fennoscandian
(Baltic) Shield. Sediment Geol 141–142:255–285
Ojakangas RW, Morey GB, Southwick DL (2001b) Paleoproterozoic
basin development and sedimentation in the Lake Superior region,
North America. Sediment Geol 141–142:319–342
Pettijohn FJ (1943) Basal Huronian conglomerates of Menominee and
Calumet districts, Michigan. J Geol 51:387–397
Pettijohn FJ (1970) The Canadian Shield – a status report, 1970. In:
Baer AJ (ed) Basins and geosynclines of the Canadian Shield. Geol
Surv Can, Pap 70–40: 239–255
Papineau D, Mojzsis SJ, Coath CD, Karhu JA, Keegan KD (2005) Multi-
ple sulfur isotopes of sulfides from sediments in the aftermath of
Paleoproterozoic glaciations. Geochem Cosmochim Acta 69:188–212
Papineau D, Mojzsis SJ, Schmitt AK (2007) Multiple sulfur isotopes
from Paleoproterozoic Huronian interglacial sediments and the rise
of atmospheric oxygen. Earth Planet Sci Lett 255:188–212
Pavlov AA, Kasting JF (2000) Mass-independent fractionation of sulfur
isotopes in Archean sediments: strong evidence for an anoxic
Archean atmosphere. Astobiology 2:27–41
Pickard AL (2003) SHRIMP U-Pb zircon ages for the Palaeopro-
terozoic Kurumn Iron Formation, Northern Cape Province, South
Africa: evidence for simultaneous BIF deposition on Kaapvaal and
Pilbara cratons. Precambrian Res 125:275–315
Polteau S, Moore JM, Tsikos H (2006) The geology and geochemistry
of the Palaeoproterozoic Makganyene diamictite. Precambrian Res
148:257–274
Potgieter GJ (1992) Tektonisme langs die noordoostelike rand van die
Bosveldkompleks, Suid-Afrika, Ph.D. thesis, University of Pretoria,South Africa, p 214
Puchtel IS, Hofmann AW, Mezger K, Schipansky AA, Kulikov VS,
Kulikova VV (1996) Petrology of a 2.41 Ga remarkably fresh
komatiitic basalt lava lake in Lion Hills, central Vetreny Belt, Baltic
Shield. Contrib Miner Petrol 124:273–290
Puchtel IS, Haase KM, Hofmann AW, Chauvel C, Kulikov VS, Garbe-
Sch€onberg C-D, Nemchin AA (1997) Petrology and geochemistry
of crustally contaminated komatiitic basalts from the Vetreny Belt,
southeastern Baltic Shield: evidence for an early Proterozoic mantle
plume beneath rifted Archean continental lithosphere. Geochim
Cosmochim Acta 61:1205–1222
Puffett WP (1969) The Reany Creek Formation, Marquette County,
Michigan. US Geol Surv Bull 1274:25p
Rainbird RH, Donaldson JA (1988) Nonglaciogenic deltaic deposits in
the early Proterozoic Gowganda Formation, Cobalt Basin, Ontario.
Can J Earth Sci 25:710–724
Robertson JA (1964) Geology of Scarfe, Mack, Cobden, and Striker
township. Ont Dept Mines Geol Rep 20:1–89
Robertson JA (1986) Geology of township 149 and township 150.
Ontario Dept Mines, Geol Rep 57:1–162
Roscoe SM (1969) Huronian rocks and uraniferous conglomerates in
the Canadian shield. Geol Surv Can Pap 68–40:205
Roscoe SM, Card KD (1993) The reappearance of the Huronian in
Wyoming: rifting and drifting of ancient continents. Can J Earth Sci
30:2475–2480
Reuschel M, Strauss H, Lepland A, Melezhik VA, Cartigny P, Kaufman
AJ (2009) Multiple sulfur isotope measurements from the 2.44 Ga
Seidorechka Sedimentary Formation. Geochim Cosmochim Acta
73:A1091
Salop LJ (1983) Geological evolution of the earth during the Precam-
brian. Springer, Berlin, p 459
Schenk PE (1965) Precambrian glaciated surface beneath the
Gowganda Formation, Lake Timagami, Ontario. Science 149:
176–177
Shields GA (2005) Neoproterozoic cap carbonates: a critical appraisal
of existing models and the plumeworld hypothesis. Terra Nova
17:299–310
Sims PK, Peterman ZE (1983) Evolution of Penokean foldbelt, Lake
Superior region, and its tectonic environment. In: Medaris LG Jr
(ed) Early Proterozoic geology of the Great lakes region. Geol Soc
Am, Memoir160: 3–14
Strand KO, Laajoki K (1993) Palaeoproterozoic glaciomarine sedimen-
tation in an extensional tectonic setting: the Honkala Formation,
Finland. Precambrian Res 64:253–271
Sturt BA, Melezhik VA, Ramsay DM (1994) Early Proterozoic regolith
at Pasvik, NE Norway: palaeotectonical implications for the Baltic
Shield. Terra Nova 6:618–633
Sumner DY, Bowring SA (1996) U-Pb geochronologic constraints on
deposition of the Campbellrand Subgroup, Transvaal Supergroup,
South Africa. Precambrian Res 79:25–36
Thomson R (1957) The Proterozoic of the Cobalt area. Roy Soc Can
Special Pub 2:40–45
Trendall AF (1976) Striated and faceted boulders from the Turee
Creek Formation – evidence for a possible Huronian glaciation on
the Australian continent. Geol Surv West Austr, Ann Rep, 1975:
88–92
Trendall AF (1979) A revision of the Mount Bruce Supergroup. Geol
Surv West Austr, Ann Rep, 1978: 63–71
Trendall AF, Compston W, Williams IS, Armstrong RA, Arndt NT,
McNaughton NJ, Nelson DR, Barley ME, Beukes NJ, De Laeter JR,
Retief EA, Thorne AM (1990) Chronological comparison of the
volcano-sedimentary sequences of the Kaapvaal and Pilbara cratons
between 3.2 and 2.4 Ga, Abstract, Third Archaean Symposium,
Perth, p 81–83
Trendall AF, De Later JR, Nelson DR (1995) Chronology of Gondwana
BIFs: progress report from recent zircon U-Pb results, Abstract,
Third Australian Conference on Geochronology, Perth, pp 1–3
Vallini DA, Cannon WF, Schulz KJ (2006) Age constraints for
Paleoproterozoic glaciation in the Lake Superior region: detrital
zircon and hydrothermal xenotime ages for the Chocolay Group,
Marquette Range Supergroup. Can J Earth Sci 43:571–591
Vajner V (1974) The tectonic development of the Namaqua mobile belt
and its foreland in parts of the Northern Cape. Precambrian
Research Unit Bulletin, vol 14. University of Cape Town, p 201
Van Schmus WR (1976) Early and middle Proterozoic history of the
Great lakes area, North America, Royal Society of London Philo-
sophical Transactions, Ser A2280, No. 1298: 605–628
Veizer J, Clayton RN, Hinton RW (1992) Geochemistry of Precam-
brian carbonates: IV. Early Paleoroterozzoic (2.25 � 0.25 Ga) sea-
water. Geochim Cosmochim Acta 56:875–885
Visser JNJ (1971) The deposition of the Griquatown glacial member in
the Transvaal Supergroup. Trans Geol Soc S Afr 74:187–199
Visser JNJ (1999) Lithostratigraphy of the Makganyene Formation
(Postmasburg Group). S Afr Comm Stratigr, Lithostratigr Ser
34:1–7
Vogel DC, Vuollo JI, Alapieti TT, James RS (1998) Tectonic, strati-
graphic, and geochemical comparison between ca. 2500–2440 Ma
mafic igneous events in the Canadian and Fennoscandian Shields.
Precambrian Res 92:89–116
Walraven F, Martini J (1995) Zircon Pb-evaporation age
determinations of the Oak Tree Formation, Chuniespoort Group,
Transvaal Sequence: implications for Transvaal-Griqualand West
basin correlations. S Afr J Geol 98:58–67
Wanless RK, Eade KE (1975) Geochronology of Archean and Protero-
zoic rocks in the southern district of Keewatin. Can J Earth Sci
12:95–114
1108 V.A. Melezhik et al.
Whitmeyer SJ, Karlstrom KE (2007) Tectonic model for the Protero-
zoic growth of North America. Geosphere 3:220–259
Williams GE, Schmidt PW (1997) Paleomagnetism of the Paleopro-
terozoic Gowganda and Lorrain formations, Ontario: low
paleolatitudes for Huronian glaciation. Earth Planet Sci Lett
153:157–169
Williams H, Hoffman PF, Lewry JF, Monger JWH, Rivers T (1991)
Anatomy of North America: thematic geologic portraits of the
continent. Tectonophysics 187:117–134
Young GM (1966) Huronian stratigraphy of the McGregor Bay area,
Ontario: relevance to the paleogeography of the Lake Superior area.
Can J Earth Sci 3:203–210
Young GM (1968) Sedimentary structures in Huronian rocks of
Ontario. Palaeogeogr Palaeocl Palaeoecol 4:125–153
Young GM (1970) An extensive early Proterozoic glaciation in North
America. Palaeogeogr Palaeocl Palaeoecol 7:85–100
Young GM (1973) Origin of carbonate-rich early Proterozoic Espanola
Formation, Ontario, Canada. Geol Soc Am Bull 84:135–160
Young GM (1975) Geochronology of Archean and Proterozoic rocks in
the southern district of Keewatin: discussion. Can J Earth Sci
12:1250–1254
Young GM (1981a) Diamictites of the early Proterozoic Ramsey Lake
and Bruce Formations, north shore of Lake Huron, Ontario,
Canada. In: Hambrey MJ, Harland WB (eds) Earth’s pre-
Pleistocene glacial record. Cambridge University Press,
Cambridge, pp 813–816
Young GM (1981b) The early Proterozoic Gowganda Formation,
Ontario, Canada. In: Hambrey MJ, Harland WB (eds) Earth’s pre-
Pleistocene glacial record. Cambridge University Press, Cambridge,
pp 807–812
Young GM (1983) Tectono-sedimentary history of early Proterozoic
rocks of the northern Great lakes region. In: Medaris LG Jr (ed)
Early Proterozoic geology of the Great lakes region. Geol Soc Am
Memoir 160: 15–32
Young GM (1988) Proterozoic plate tectonics, glaciations and iron-
formations. Sediment Geol 58:127–144
Young GM (1991) The geological record of glaciation: relevance to the
climatic history of the earth. Geosci Can 18:100–108
Young GM (2002) Stratigraphic and tectonic settings of Proterozoic
glaciogenic rocks and banded iron-formations; relevance to the
snowball earth debate. J Afr Earth Sci Middle East 35:451–466
Young GM (2004) Earth’s earliest extensive glaciations: tectonic
setting and stratigraphic context of Paleoproterozoic glaciogenic
deposits. In: Jenkins GS, McMenamin MAS, McKay CP, Sohl L
(eds) The extreme Proterozoic: geology, geochemistry and climate.
American Geophysical Union, Washington, DC. Geophys
Monogr146: 161–181
Young GM, Church WR (1966) The Huronian system in the Sudbury
district and adjoining areas of Ontario: a review. Proc Geol Assoc
Can 17:65–82
Young GM, Long DGF (1976) Ice-wedge casts from the Huronian
Ramsay Lake Formation (>2300 m.y. old) near Espanola, Ontario,
Canada. Palaeogeogr Palaeocl Palaeoecol 19:191–200
Young GM, McLennan SM (1981) Early Proterozoic Padlei Formation,
Northwest Territories, Canada. In: Hambrey MJ, Harland WB (eds)
Earth’s pre-Pleistocene glacial record. Cambridge University Press,
Cambridge, pp 790–794
Young GM, Long DGF, Fedo CH, Nesbitt HW (2001) The Paleopro-
terozoic Huronian basin: product of a Wilson cycle accompanied
by Glaciation and Meteorite impact. Sediment Geol 141–142:
233–254
Young GM, Nesbitt HW (1985) The Gowganda Formation in the
southern part of the Huronian outcrop belt, Canada: stratigraphy,
depositional environments and regional tectonic significance. Pre-
cambrian Res 29:265–301
Young GM, von Brunn V, Gold DJC, Minter WEL (1998) Earth’s
oldest reported glaciation: physical and chemical evidence from
the Archean Mozaan Group (2.9 Ga) of South Africa. J Geol
106:523–538
2 7.2 Huronian-Age Glaciation 1109
7.3 The Palaeoproterozoic Perturbation of the GlobalCarbon Cycle: The Lomagundi-Jatuli Isotopic Event
Victor A. Melezhik, Anthony E. Fallick, Adam P. Martin, Daniel J. Condon,Lee R. Kump, Alex T. Brasier, and Paula E. Salminen
7.3.1 The Global Carbon Cycle and Its PrincipalReservoirs and Fluxes
Victor A. Melezhik, Anthony E. Fallick,Lee R. Kump, and Alex T. Brasier
On Earth, carbon cycles through the land, ocean, atmo-
sphere, living and dead biomass and the planet’s interior.
The global carbon cycle can be divided into the tectonically
driven geological cycle and the biological/physicochemical
cycles. The former operates over millions of years, whereas
the latter operate over much shorter time scales (days to
thousands of years). Within the geological cycle, atmo-
spheric carbon dioxide concentration is controlled by the
balance between weathering, biological drawdown, size of
sedimentary reservoir, subduction, metamorphism and vol-
canism over time periods of hundreds of millions of years.
The Earth’s crust represents a major carbon reservoir
containing 9 � 1022 g of C (e.g. Sundquist 1993). The
ocean (4 � 1019 g) together with reactive marine sediments
(3 � 1018 g) are the next major reservoir of carbon. The
terrestrial biosphere (6 � 1017 g), active (1 � 1018 g) and
old (5 � 1017 g) soils, and fossil fuels (4 � 1018 g), alto-
gether contain 6 � 1018 g of C (Sundquist 1993). The atmo-
sphere represents the smallest of the major reservoirs of the
Earth, 8 � 1017 g of C.
The atmosphere presently exchanges its C with the ocean
at the rate of 7 � 1016 g of C per year, and the exchange
between the biosphere and atmosphere is estimated at a simi-
lar rate of 6 � 1016 g of C per year (Sundquist 1993). The
exchange between the Earth’s crust and the three exogenic
reservoirs is about 2 � 1014 g of C per year. The present
global rate of CO2 emission from volcanoes is estimated at
about 4–5 � 1013 g of C per year (Gerlach 1991).
Details of biological carbon fixation are given in Chap.
7.6. Under steady state conditions, the requirement for mass
and isotope balance within the global carbon cycle
(Broecker 1970; Schidlowski et al. 1983; Summons and
Hayes 1992) leads to a relationship between d13C of the
input flux of carbon (din) and that of carbonate sequestered
in sediment (dcarb), given to reasonable approximation by:
dcarb ¼ din þ forgDc (1)
where Dc is the isotopic difference between concurrently
sequestered organic and inorganic carbon, and forg is the
fraction of carbon being sequestered which is organic. As
noted by Des Marais et al. (1992), din describes the13C/12C
ratio of carbon which enters the Earth’s near-surface
reservoirs (atmosphere, hydrosphere and biota) through the
processes of volcanism, metamorphism and weathering. For
timescales >100 Myr, din is commonly taken at around
�6‰, the average value for crustal carbon (Holser et al.
1988), which equates to “a major isotopic composition sig-
nature for the mantle” (Deines 2002). dcarb represents the
weighted-average isotopic composition of oxidised carbon
buried in carbonates, and Schidlowski (1988) has drawn
attention to the high frequency with which values close to
0.5 � 2.5‰ (V-PDB) are found throughout the last 3.5
billion years of Earth’s history. With Dc (difference between
dcarb and dorg) influenced by enzymatic processes, particu-
larly by ribulose bisphosphate carboxylase-oxidase
(RUBISCO) involved in photosynthesis, and therefore gen-
erally around 25‰, forg (fraction of carbon buried in reduced
form) is constrained to 0.2 (�~0.1), for timescales
>100 Myr.
It is commonly assumed that din and Dc remain constant
throughout Earth’s history (e.g. Schidlowski 1988;
Schidlowski et al. 1983). Within this context, dcarb can
deviate from the average value of 0.5 � 2.5 ‰ due to
V.A. Melezhik (*)
Geological Survey of Norway, Postboks 6315, Sluppen, NO-7491
Trondheim, Norway
Centre for Geobiology, University of Bergen, Allegaten 41,
Bergen N-5007, Norway
e-mail: [email protected]
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_3, # Springer-Verlag Berlin Heidelberg 2013
1111
fluctuation of forg, which is fraction of carbon buried as
organic matter. Consequently, two extreme scenarios can
be envisaged if forg approximates zero or reaches 1.
The first scenario can be expected if biological productivity
is completely shut down (‘biopump’ failure), carbonate
sedimentation continued, and din is entirely dominated by
mantle carbon with d13C values at c. �6 ‰ (canonical
mantle value, e.g. Mattey 1987): then d13Ccarb ~ �6 ‰.
The second scenario would result if all carbon entering
global surface environment were to be sequestered and
buried in the reduced from as organic carbon (forg ¼ 1).
Considering Eq. 1 the highest dcarb can then be up to
about +20 ‰.
Recognising the inability of the conventional (Broecker)
steady state approach adequately to describe the global
carbon cycle during specific episodes of Earth history,
Rothman et al. (2003) proposed a dynamic systems
approach to elucidate the behaviour of the carbon cycle,
specifically focussing on the Neoproterozoic and the
Shuram-Wonoka carbon isotope excursion at the
Proterozoic-Cambrian boundary. They invoked an oceanic
reservoir of suspended and dissolved organic carbon
between 102 and 103 times larger than at present, with
consequentially a greater average age and so ocean resi-
dence time: its properties changed only slowly but it was
interactive with the inorganic carbon reservoir. The eventual
transfer of most of this organic carbon to the carbonate
pool as the proposed reservoir terminally diminished
(driven by a combination of factors including enhanced
remineralisation, ocean ventilation, and faecal pellet-
assisted transport to the seafloor) resulted in a prominent
isotope excursion characterised by low d13Ccarb. Whereas
Rothman et al. (2003) argued that a shift in Neoproterozoic
d13C of 10 ‰ (from �5 ‰ to +5 ‰) for an inorganic
reservoir of modern size would require the equivalent of
only 4% of the present atmospheric inventory of molecular
oxygen, others have doubted the validity of the model
(e.g. Bristow and Kennedy 2008) on the basis of the avail-
able oxidant budget. Of course, the existence of a large
reservoir of oceanic organic carbon (and Rothman et al.
2003 modelled one containing 32 � 1018 moles of organic
carbon – ten times the present inventory of oceanic inor-
ganic carbon) presupposes its creation, with concomitant
increase in d13Ccarb during the buildup, prior to the
Neoproterozoic (assuming the organic matter is effectively
removed from the dynamic carbon cycle by its long resi-
dence time). Perhaps applying the dynamic systems
approach to appropriate geological sequences older than
740 Ma would allow further investigation of this and pro-
vide a critical test of the undoubtedly imaginative and
thought-provoking approach, whose singular advantage in
this context is that it does not assume a steady state.
1112 V.A. Melezhik et al.
7.3.2 Historical Overview
Victor A. Melezhik, Anthony E. Fallick,Alex T. Brasier, and Lee R. Kump
The earliest systematic measurements of carbon isotopes in
sedimentary carbonates in the late 1960s and early 1970s
(Galimov et al. 1968; Schidlowski et al. 1975) recorded
unusually 13C-rich sedimentary carbonates (d13C up to
+13 ‰). Although unrecognised at that time, this discovery
was one of the greatest perturbations of the global carbon
system, later termed the Lomagundi-Jatuli positive isotopic
excursion (Melezhik et al. 2005a). Because of the unusually
high enrichment in 13C and a limited database at that time,
such enrichment was first considered as a local organic
carbon burial phenomenon (Schidlowski et al. 1976), or to
represent an environment where the organic part of the
carbon cycle was absent and the isotopic composition of
chemically-precipitated carbonates was governed by equi-
librium within the CO2–HCO3�–CO3
2� system (Galimov
et al. 1968). The double name of the excursion derives
from two geographically distant areas where 13C-rich sedi-
mentary carbonates were first discovered: the Jatulian-age
rocks in Karelia, eastern Fennoscandia (Galimov et al. 1968;
Schidlowski et al. 1975), and the Lomagundi Group in
Zimbabwe, western Africa (Schidlowski et al. 1975).
In the late 1980s, to the Galimov-Schidlowski observ-
ations were added two new occurrences of 13C-rich sedimen-
tary carbonates, and all together these were proposed to
represent a positive isotopic excursion of global signifi-
cance. The excursion was ascribed to an enhanced accumu-
lation of organic carbon, which drove oxidised carbon
(marine carbonate) isotopically heavy, and led to a rise in
atmospheric oxygen (Baker and Fallick 1989a, b).
J. Karhu conducted intensive studies in several
Palaeoproterozoic basins on the Fennoscandian Shield and
provided the first reliable geochronological constraint on the
duration of this isotopic excursion (Karhu 1993; Karhu and
Holland 1996). It was placed between 2200 and 2060 Ma,
with a suggested duration of 140 Myr. Although a few new
age constraints were subsequently obtained (Karhu et al.
2008; Melezhik et al. 2007), the previous estimate has
remained largely unmodified ever since.
The apparent absence in the geological record of organic-
rich sedimentary rocks necessary for balancing this excur-
sion of 140 Myr duration was highlighted as a “paradox”
by Melezhik and Fallick (1996). They also linked the excur-
sion to 12 other events, which all seemed to be global in
nature. Amongst the most remarkable links is the association
of 13C-rich carbonates with “red beds”, stromatolites,
Ca-sulphates and other evaporites. Shields (1997) removed
the “paradox” by invoking a model of the stratified ocean of
Keith (1982). Similarly, Aharon (2005) appealed to a redox-
stratified ocean model, decoupling the P and C cycles. Such
a model predicts not only formation of 13C-rich shallow-
water shelf carbonates but also the accumulation of
authigenic 13C-poor carbonates within deeper parts of the
stratified ocean (Keith 1982). A similar model was invoked
again by Bekker et al. (2008), but these authors did not
address the absence of 13C-poor carbonates, which are com-
mon by-products of intensive remineralisation of organic
carbon in anoxic environments. Since such 13C-poor primary
sedimentary carbonates are not reported in the geological
record, perhaps yet to be discovered, the “paradox” of
Melezhik and Fallick (1996, 1997) seems to remain.
Yudovich et al. (1991) acknowledged the absence of
geological evidence for buried carbon to compensate13C-rich carbonates and linked the excursion to
methanogenic diagenesis. They assumed that during this
time, stromatolite-forming cyanobacterial mats were subject
to anaerobic diagenesis in shallow-water environments,
which caused methane production and its rapid escape to
the atmosphere. Hayes and Waldbauer (2006) elaborated on
this approach and invoked fermentative and methanogenic
diagenesis in deeper levels of the sediment column as the
response to increasing O2 and SO42� concentration in the
ocean. However, fermentative diagenesis commonly
produces a “noisy” d13C pattern with a large range in
d13C (e.g. Watson et al. 1995); this has not been observed
so far in studied 13C-rich carbonate successions (Karhu
1993; Melezhik and Fallick 1996; Melezhik et al. 1999a,
2005b; Bekker et al. 2001, 2003a, b; Brasier et al. 2011).
However, such an isotopic pattern is a feature of younger
Palaeoproterozoic organic-rich successions, which overlie
the 13C-rich carbonates in the Fennoscandian Shield and
elsewhere (e.g. Yudovich et al. 1991; Melezhik et al.
1999b; Maheshwari et al. 2010). This feature is associated
with the period of global-scale enhanced accumulation of
organic matter (Salop 1982; Condie et al. 2001) known now
as the Shunga Event (Melezhik et al. 2005a).
“Black shales” have been recently reported to be
associated with 13C-rich Lomagundi-Jatuli carbonate rocks,
and were considered as a missing sink compensating the
Lomagundi-Jatuli positive excursion (Bekker et al. 2008;
Maheshwari et al. 2010; Master et al. 2010). However,
where precise depositional ages are available, it is apparent
that such black shales occur either at the very end (between
2083 � 6 and 2050 � 30 Ma in Francevillian basin;
Gancarz 1978; Horie et al. 2005) or in the aftermath of the
V.A. Melezhik (*)
Geological Survey of Norway, Postboks 6315, Sluppen, NO-7491
Trondheim, Norway
Centre for Geobiology, University of Bergen, Allegaten 41,
Bergen N-5007, Norway
3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1113
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_3, # Springer-Verlag Berlin Heidelberg 2013
1113
Lomagundi-Jatuli positive excursion (after 2060–2050 Ma
in the Per€apohja and Pechenga belts (Perttunen and Vaasjoki
2001; Melezhik et al. 2007, respectively)). Other examples
of suggested “considerable” temporal overlaps between13C-rich carbonates and organic-bearing shales involve
many contingents (“ifs”) and circumstantial evidence and
are not yet supported by reliable age constraints (e.g.
Sengoma Argillite Formation in Botswana, Bekker et al.
2008). The 2080–2050 Ma Late “Lomagundi-Jatuli” black
shales are very unlikely to represent a compensating sink for
the entire excursion and cannot be considered as the cause
for its initiation at >2200 Ma, as well as for its being
sustained over a period of c. 140 Myr. Thus, again, the
“paradox” remains unresolved.
An interesting model recently advanced by Kirschvink
et al. (2009) invokes a reduced rate of organic carbon
recycling as a main driver of isotopically heavier carbonate
carbon. The model infers a c. 2300–2056 Ma transitional
period with oxygen content below that required for respira-
tion. Consequently, the Photosystem-II-generated O2 would
have been largely unavailable for remineralisation of
dissolved organic carbon, thus profoundly shifting the burial
ratio of organic/inorganic carbon. This model somewhat
echoes models by Hayes and Waldbauer (2006) and Fallick
et al. (2008) in the sense that these two earlier hypotheses also
explored important biological changes in response to devel-
opment of the O2-rich biosphere. However, in contrast, the
Kirschvink model still faces the “paradox”: organic-rich
sedimentary rocks reflecting a high organic/inorganic carbon
ratio are yet to be found. On the one hand, a common “defen-
sive” approach that organic carbon-rich shales once existed
and were continuously accumulated over period of more than
120 Ma but were subsequently subducted or eroded leaves
many geologists puzzled (why would all the organic-rich
rocks be preferentially subducted or eroded?) and seems to
have no testable implications, and so dubious credentials as
to being considered strictly scientific. On the other hand, we
acknowledge that most of the sediments that were deposited
during this (or any other) time have been eroded; perhaps
preservation is the oddity. Could it be that Lomagundi-Jatuli
carbonates are anomalously well-preserved for tectonic
reasons (e.g. accumulated in subsiding basins, and then
infolded and relatively rarely uplifted)?
In an attempt to better quantify the mass-balance prob-
lem, Melezhik et al. (1999a, 2005b) focussed on the global
pattern of the Lomagundi-Jatuli excursion, and possible
roles of local factors in amplification of a global signal.
However, their suggested d13C of around +5 ‰ as an appar-
ent background value is left unsubstantiated in view of
currently limited robust age constraints through the known
Palaeoproterozoic 13C-rich formations (Melezhik et al.
2007). This issue was revisited by Frauenstein et al. (2009)
when they considered the carbon isotopic composition
of sedimentary carbonates from several formations in
the Transvaal Group (Silverton, Lucknow, Rooinekke
and Duitschland), including those deposited within the
Lomagundi-Jatuli time-interval. They concluded that
intercalations of 13C-enriched (+10 ‰) and normal marine
(c. 0 ‰) carbonates between and within formations and
specific horizons cannot be explained by frequent and drastic
fluctuation of global d13C and must be governed by either
basinal or regional factors. It remains unclear whether or not
the Duitschland Formation, in which most of the isotopic
fluctuations have been observed, represents part of the
Lomagundi-Jatuli story (Bekker et al. 2001).
Another puzzle related to the apparent problem of
balancing masses of reduced and oxidised carbon during
the Lomagundi-Jatuli excursion comes from North Amer-
ica where Bekker et al. (2003a) recorded, in a supposedly
marine transgressive succession, carbonate d13C up to
+28 ‰. It is apparent that d13C ~ +28 ‰ as a global
signal would require the fraction of organic carbon buried
to be above 1 (in other words, burial of more organic
carbon than existed): this calls for an explanation (but
see also Rothman et al. 2003). A rapid d13C rise from
+10 ‰ to +16 ‰ and decline back to +10 ‰ within 2 m
in the Tulomozero Formation, in the Onega basin of the
Fennoscandian Shield (e.g. Melezhik et al. 1999a)
represents less of a puzzle but an explanation involving
rapid restructuring of the global oceanic carbon budget is
arguably unrealistic. Thus, the issue of global d13C signal
versus local enhancement remains unresolved and calls
for further research.
Melezhik et al. (1999a) emphasised that the Lomagundi-
Jatuli positive isotopic excursion of d13Ccarb was not
followed by a negative isotopic shift significantly below
0 ‰, as has usually been observed in younger isotopic
events, reflecting an overturn of major marine carbon
reservoir. Further, they speculated that the absence of
the negative shift may be indicative of constant forg, imply-
ing that perhaps other mechanisms/forces drove the excur-
sion. Isotopic evidence for massive oxidation of organic
matter towards the end of the Lomagundi-Jatuli recently
recognized by Kump et al. (2011) suggests that this may
not be the case.
Figure 7.32 illustrates the geographic locations of13C-rich (>5 ‰) sedimentary carbonates occurring within
the Lomagundi-Jatuli time-interval (c. 2322–2052 Ma),
and confidently proves the global nature of the excursion
regardless of its driving force(s). Below we revisit some
key areas and provide a brief review of major achievements
and remaining problems and highlight the significance
and potential of regional/basinal data for a better under-
standing of this unprecedented perturbation of the global
carbon cycle, one of the greatest in Earth’s history.
We specifically address intraformational internal d13Cfluctuations and their possible causes during the
Lomagundi-Jatuli interval.
1114 V.A. Melezhik et al.
7.3.3 Review of Available Radiometric AgesConstraining the Lomagundi-JatuliPositive Isotopic Excursionof Carbonate Carbon
Adam P. Martin and Daniel J. Condon
Constraining the initiation, termination and inferentially,
duration of the Lomagundi-Jatuli event is particularly chal-
lenging due to the limited dataset of robust radio-isotopic
dates that can be confidently related to specific d13C data. In
his study of Palaeoproterozoic sedimentary carbonates in
Fennoscandia, Karhu (1993) adopted d13C values >+3 ‰as enriched (with respect to normal sedimentary carbonates)
though he used >+4 ‰ in practice for discrimination
between Palaeoproterozoic normal and 13C-rich carbonates
(Fig. 4 in Karhu 1993).
Establishing chronological constraints on non-fossili-
ferous sedimentary horizons relies upon radio-isotopic dat-
ing of intercalated igneous units, and/or assumptions about
minimum and maximum age constraints from basement
lithologies, detrital minerals (e.g. zircon), diagenetic over-
growths and/or metamorphic assemblages. Radio-dates have
been determined for stratigraphic units associated with
sediments recording the Lomagundi-Jatuli event from all
continents except Antarctica utilising a range of
radioactive-decay schemes and analytical techniques with
varying degrees of precision and relevance (given advances
in chronological technique). The most widely applied chro-
nometer is U-Pb applied to zircon and other uranium-bearing
accessory minerals; however, other decay systems (e.g.
Re-Os applied to organic rich shales) are increasingly
being utilised (see Condon and Bowring 2011, for a review
of methodologies and a discussion of limitations/strengths of
the different systems).
In order to clearly delineate the onset and termination of
the Lomagundi-Jatuli event, we review the current dataset of
pertinent geochronological constraints divided into three
categories: (1) pre-Lomagundi-Jatuli, (2) coeval with
Lomagundi-Jatuli, and (3) post-Lomagundi-Jatuli (Fig. 7.33).
Plotting the data as in Fig. 7.33 allows us to integrate
geographically disparate datasets but is predicated on the
assumption that the high-13C signals are globally correlated.
This simplistic approach eliminates the need to assign a date
to a d13C value, which often involves inference. The assign-
ment of a published age into one of the three categories was
most usually done explicitly by the authors from whom the
published age was referenced, or rarely the inference was
based upon the stratigraphic position of the dated unit below,
coeval, or above, sedimentary units known to represent the
Lomagundi-Jatuli event.
Initiation of the Lomagundi-Jatuli Event
The youngest maximum age constraints on the initiation of
the Lomagundi-Jatuli event come from:
1. A 2306 � 9 Ma age (U-Pb SHRIMP) on detrital zircon in
the Sturgeon Quartzite in the Marquette Range Supergroup
of North America (Vallini et al. 2006; datum 24 on
Fig. 7.33), that stratigraphically underlies the Saunders
Formation with d13C values�+3.1 ‰ (Bekker et al. 2006);
2. A U-Pb detrital zircon date of 2317 � 6 Ma from the
Enchantment Lake Formation (datum 26 on Fig. 7.33),
Marquette Range Supergroup, which underlies the Kona
Dolomite with d13C values �+9.5 ‰ (Bekker et al.
2006), and
3. A 2316 � 7 Ma age (Re-Os, authigenic pyrite, Hannah
et al. 2004) from the Rooihoogte Formation (datum 25 on
Fig. 7.33), Transvaal Supergroup, that underlies the
Silverton Formation, which records d13C values up to
+10 ‰ (Frauenstein et al. 2009).
The oldest age constraints that demonstrably post-date
the initiation of the Lomagundi-Jatuli event include:
1. Deposition of the Per€apohja Belt, Finland, had com-
menced prior to c. 2221 � 5 Ma based upon the age of
the Laurila mafic sill (U-Pb ID-TIMS baddeleyite; datum
23 on Fig. 7.33) (Perttunen and Vaasjoki 2001) that
intrudes a stratigraphic succession including carbonate
beds with d13C values � +4 ‰;
2. Deposition of the Gordon Lake Formation with d13Cvalues � +4 ‰, Huronian Supergroup, which pre-dates
intrusion of the Nissiping Intrusions dated at 2217 � 9
Ma (U-Pb ID-TIMS baddeleyite and rutile; datum 22 on
Fig. 7.33); and
3. A 2206 � 9 Ma date (U-Pb zircon; datum 20 on
Fig. 7.33) from a diabase dyke that cross-cuts the
Lomagundi-Jatuli bearing Sericite Schist Formation
(Fig. 42b), Kuusamo Belt, Finland.
Thus the initiation of the Lomagundi-Jatuli event is
constrained between c. 2310 Ma and c. 2220 Ma, a time
gap of c. 90 Myr (Fig. 7.33).
Termination of the Lomagundi-Jatuli Event
The youngest age constraints that are demonstrably coeval
with the Lomagundi-Jatuli event include:
1. A 2115 � 6 Ma age (U-Pb ID-TIMS, zircon; Pekkarinen
and Lukkarinen 1991) from diabase bodies (Datum 16
on Fig. 7.33) that are inferred to feed lavas of the Koljola
A.P. Martin (*)
NERC Isotope Geosciences Laboratory (NIGL), Keyworth,
Nottingham, UK
3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1115
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_3, # Springer-Verlag Berlin Heidelberg 2013
1115
Formation (Karhu 1993). The Koljola Formation is
overlain by the Viistola Formation containing sedi-
ments with d13C values �+4 ‰ (Karhu 1993) in the
Kiihtelysvaara area, Finland;
2. A 2142 � 4 Ma age (U-Pb ID-TIMS) on a rhyolite
(datum18 on Fig. 7.33) that is overlain by the Denault
and Abner Formations and underlain by the Uve Forma-
tion, both of which have d13C values �+4 ‰ (Melezhik
et al. 1997).
Constraints on the termination of the Lomagundi-Jatuli
event include:
1. In the Kuusamo Belt, Finland, there is a drop in d13Cvalues from +10.6 ‰ in the Dolomite Formation down
to +4.2 ‰ in the Limestone Formation (Fig. 44b) and
separating these two formations is the Amphibole Schist
Formation (Karhu 1993). Silvennoinen (1991) has
suggested that the youngest generation of diabase
intrusions in the Kuusamo Belt, such as the Viipus sill,
are the feeders to the metavolcanics in the Amphibole
Schist Formation. The Viipus sill is dated at 2078 � 8Ma
(U-Pb ID-TIMS, zircon, datum 15 on Figs. 7.33 and 44b)
and would therefore inferentially constrain the decline in
d13C and thus the termination of the Lomagundi-Jatuli
event;
2. Datum 14 on Fig. 7.33 is from the Kolosjoki Sedimentary
Formation that records d13C values between +1 ‰ and
+2.5 ‰ (Melezhik et al. 2007), overlying the Kuetsj€arvi
Sedimentary Formation that records d13C values�+9 ‰.
Detrital zircons from volcaniclastic conglomerate in
the middle part of the Kuetsj€arvi Volcanic Formation
and from volcaniclastic greywackes at the base of
the Kolosjoki Sedimentary Formation are dated at
2058 � 6 Ma (U-Pb ID-TIMS; Melezhik et al. 2007);
petrographically in both cases clastic material was
sourced from the Kuetsj€arvi Volcanic Formation, thus its
age is inferred to be c. 2060 Ma. The Kuetsj€arvi Volcanic
Formation is stratigraphically between the underlying
Kuetsj€arvi Sedimentary Formation rocks that record the
Lomagundi-Jatuli event, and the isotopically normal
rocks of the Kolasjoki Sedimentary Formation.
3. A c. 2052 Ma age on detrital zircons (U-Pb ID-TIMS) of
the Il’mozero Sedimentary Formation (datum 11; Martin
et al. 2010), Imandra-Varzuga Greenstone Belt, Kola
craton. The Il’mozero Sedimentary Formation is thought
to be the lithostratigraphic equivalent of the Kolosjoki
Sedimentary Formation (see Chap. 7.3.4) and work on the
carbon isotopes of the Imandra-Varzuga Belt is ongoing
(see below).
The termination of the Lomagundi-Jatuli event is infer-
entially dated at c. 2080 Ma in the Kuusamo Belt, Finland
(datum 15), which is supported by dates of c. 2060 Ma from
the Pechenga Greenstone Belt (datum 13) and the Imandra-
Varzuga Belt (datum 12) on detrital zircons and related
volcanic formations. The duration of the Lomagundi-Jatuli
event can be seen from Fig. 7.33 to endure at least 160 Myr.
Global Timing of the Lomagundi-Jatuli Event
Assigning published radio-isotopic dates from units
associated with the Lomagundi-Jatuli event (Fig. 7.33) into
the three categories, pre-, syn and post-strata with d13C values
�+4 ‰, delineates the initiation and termination of the
Lomagundi-Jatuli event and provides an estimate of its dura-
tion. Importantly, this global compilation has not revealed any
reversals (i.e. sections where 13C-rich sediments are coeval
and/or younger than sections with normal d13C), further
supporting the first order inference that the Lomagundi-Jatuli
event is globally synchronous (Fig. 7.33).
A second order of complexity can be introduced to the
chronology by assigning age constraints to specific units
recording Lomagundi-Jatuli d13C values worldwide. This
is typically represented in the classic age versus d13Cplots (Baker and Fallick 1989a, b; Karhu 1993; Karhu
and Holland 1996). The challenges in producing this
diagram are:
1. The sedimentary units which record d13C values often do
not contain minerals suitable for single grain U-Pb
chronology.
2. When a robust age constraint is provided, it is often
several units removed from the unit recording the d13Cvalues of interest, providing at best a maximum/mini-
mum age.
3. Often there is only one reliable age constraint stratigra-
phically above (or below) the unit recording the
d13C values of interest, begging the question of an appro-
priate upper or lower age limit.
Figure 7.34 is one interpretation of the Lomagundi-Jatuli
age versus d13C plot; it tries to make the fewest assumptions
about the depositional age of the unit(s) recording the
Lomagundi-Jatuli event. This plot uses only U-Pb and Re-
Os radiometric dates from published sources (all data and
references are in Table 7.1). The conservative age
constraints made in constructing Fig. 7.34 produce a large,
positive d13C curve representing the Lomagundi-Jatuli
event. The classic Lomagundi-Jatuli curve (Fig. 1 in Karhu
and Holland 1996) is superimposed on the data set in
Fig. 7.34. The termination and peak of the Lomagundi-Jatuli
curve matches well between the two datasets (Fig. 7.34). The
initiation of the event is constrained between 2221 � 7 Ma
(Perttunen and Vaasjoki 2001) and 2316 � 7Ma (Re-Os age
on diagenetic pyrite from South Africa, Hannah et al.
(2004); Fig. 7.34), whereas its termination is c. 2058 � 6Ma
(Melezhik et al. 2007). Thus the duration of the Lomagundi-
Jatuli event is at least c. 160 Myr, but may be as long as
c. 260 Myr.
1116 V.A. Melezhik et al.
7.3.4 Lomagundi-Jatuli Excursion as Seen fromthe Fennoscandian Shield Record
Victor A. Melezhik, Anthony E. Fallick,Alex T. Brasier, and Paula E. Salminen
High-13C Palaeoproterozoic sedimentary carbonate forma-
tions of the Fennoscandian Shield are numerous and some
are amongst the best-studied rocks of this age in the world.
More than 60 formations that accumulated in different depo-
sitional settings are known to date (Fig. 7.35) and represent
potentially valuable material for addressing some important
aspects of the Lomagundi-Jatuli paradox. The Fennoscandian
Shield also provides the best available geochronological
constraints for the duration, internal structure and termination
of the event (Karhu 1993, 2005; Karhu et al. 2008; Melezhik
et al. 2007; Martin et al. 2010). Detailed research on 13C-rich
carbonate formations from the Lomagundi-Jatuli time period
conducted in the Pechenga, Onega and Kalix areas and some
other selected areas is reviewed below.
The Pechenga Greenstone Belt
In the Pechenga Greenstone Belt, the Lomagundi-Jatuli pos-
itive isotopic excursion is recorded in the Kuetsj€arvi Sedi-
mentary Formation, a 150-m-thick succession, which was
deposited on basaltic andesites of the Ahmalahti Formation
and is overlain by alkaline-series volcanic rocks of the
Kuetsj€arvi Volcanic Formation. The formation comprises
red siliciclastic rocks, dolostones, minor limestones, and
dolomitic travertine (Fig. 7.36). All accumulated in an intra-
plate, shallow-water, lacustrine depositional system, which
was partially influenced by seawater (see Chap. 4.2; and
Melezhik and Fallick 2005). The Kuetsj€arvi high d13Ccarbonates are devoid of organic carbon, are commonly red
or pale pink in colour (Fig. 7.36a–d), and thus accumulated
in oxic environments. The succession bears numerous
features of subaerial exposure episodes, erosion and
redeposition (Fig. 7.36a–e). Rocks contain flat-laminated
stromatolites and evidence of mud desiccation and evaporite
mineral growth (Melezhik and Fallick 2005; Melezhik et al.
2004). Other sedimentological details of the Kuetsj€arvi Sed-
imentary Formation are presented in Chap. 6.2.2.
Volcanic rocks of the Kuetsj€arvi Sedimentary Forma-
tion were dated by U-Pb techniques (zircon) at 2058 � 6
Ma. This currently provides a minimum age constraint for
the deposition of 13C-rich Kuetsj€arvi dolostones, as well
as for the end of the Lomagundi-Jatuli excursion
(Melezhik et al. 2007).
The Kuetsj€arvi Sedimentary Formation underwent meta-
morphic alteration with grade ranging from biotite-actinolite
(~330�C) to epidote-amphibolite (~420�C) along a 100 km
strike-length (Petrov and Voloshina 1995; Fig. 7.37a). The
succession was sampled from surface outcrops as well as from
two drillholes, including the Kola Superdeep Drillhole
(Fig. 7.37a); both drillholes intersected the entire formational
thickness.
The d13C range in the least-altered dolostone and lime-
stone whole-rock samples from surface outcrops and drillhole
X core is from +5 ‰ to +9.6 ‰ (+7.4 � 0.7‰ on average,
n ¼ 167). Travertine deposits show a considerable deposi-
tional d13C variation from �6.1 ‰ to +7.7 ‰ (Melezhik
et al. 2005b; Fig. 7.37b, c). Metamorphic alteration under
lower temperature epidote-amphibolite conditions resulted
in 13C depletion with d13C ranging between �1 ‰ and
þ5 ‰ (sampling sites 1, 7–10, Fig. 7.37a). This was mainly
due to the metamorphic reaction between dolomite and
quartz, and the formation of tremolite, metamorphic calcite
(calcite2, Fig. 7.38a) and13C-rich CO2:
Dolomite þ Quartz ! Tremolite þ Calcite213C� depleted� �þ CO2" 13C� enriched
� �
A similar degree of 13C-depletion was measured from
microcored metamorphic calcite (calcite2) and whole-rock
dolostone samples obtained from the Kola Superdeep
Drillhole at a depth of 5,717–5,642 m, metamorphosed
under high-temperature epidote-amphibolite conditions
(Fig. 7.39a, b). However, microcored micritic dolomite relicts
still retain d13C (þ7.1 � 0.6 ‰ n ¼ 38, Fig. 7.38a, b;
Melezhik et al. 2003) close to the least altered values
measured from the whole-rock dolostones from the biotite-
actinolite metamorphic zone, whereas microcored pre-
metamorphic calcite (calcite1) samples show a sizable deple-
tion (+6.5 � 0.5 ‰ n ¼ 6, Fig. 7.39b). The overall
Superdeep Drillhole section displays highly fluctuating and
generally 13C-depleted values with respect to those
documented in the drillhole X section from the biotite-
actinolite zone (Fig. 7.39c). This highlights the importance
of screening isotopic data for metamorphic overprint, but the
practice is not common.
In contrast with metamorphic isotope resetting, diagenetic
alteration (e.g. early dolomicrite versus late dolospar) resulted
in only modest d13C lowering of the order of 1 ‰ or less
(Fig. 7.38b–d). This is also the case with the isotopic compo-
sition of dolomite from caliche crust: there is less than 1 ‰difference between early micrite and late caliche dolospar.
V.A. Melezhik (*)
Geological Survey of Norway, Postboks 6315, Sluppen, NO-7491
Trondheim, Norway
Centre for Geobiology, University of Bergen, Allegaten 41,
Bergen N-5007, Norway
3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1117
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_3, # Springer-Verlag Berlin Heidelberg 2013
1117
Interestingly, the travertine dolomite retains its low d13Cvalues although thinly intercalated with 13C-rich dolomite
(Fig. 7.38d).
When metamorphically altered samples are excluded, the
stratigraphic d13C profile displays a smooth negative excur-
sion from around +8.5 ‰ to +6 ‰ characterised by a limited
scatter (Fig. 7.39c). This would not be expected for diagenetic
(i.e. associated with methanogenesis) 13C-rich carbonate, and
does not suggest obvious influence of local factors unless they
were long-term and totally dominant over global factors.
There are several d13C positive outliers recorded in the
upper part of the drillhole X section (Fig. 7.39c). Here, low
Sr isotope ratios (0.70406–0.70486) suggest invasion of
seawater into the basin, which corresponds to a smooth
drop in d13C from +7 ‰ to +6 ‰ over 10 m of thickness
(Melezhik et al. 2005b). However, the lower part of this
negative d13C excursion contains one sharp departure from
þ7.2 ‰ toþ8.7 ‰ and back to 7.0 ‰ within a c. 3-m-thick
interval (Fig. 7.39c). The 1.5 ‰ rise and 1.7 ‰ drop within
a 3-m-thick section would, if global, require rapid and con-
siderable restructuring of the global carbon reservoir. As
such a scenario is arguably unrealistic, the anomaly in ques-
tion was very likely driven by local factors (but see also
Dickens 1999). There are three other positive sharp (though
of smaller magnitude; 0.5–1 ‰) departures superimposed
on the overall negative trend in the uppermost part of the
section. These might have a similar origin governed by local
factors.
The end of Kuetsj€arvi positive d13C excursion was
constrained by dating overlying volcanic rocks at 2058 � 6
Ma, which currently corresponds to the end of Lomagundi-
Jatuli excursion (Melezhik et al. 2007). Consequently, it
remains unresolved as to whether or not the smooth negative
shift in the upper part of the Kuetsj€arvi Sedimentary Forma-
tion represents a d13C decline towards the end of
Lomagundi-Jatuli event, or was caused by the invasion of
seawater as suggested by the Sr isotope data. In the former
case, these four positive spikes superimposed on the nega-
tive trend very likely represent evidence of local 13C
enhancement, though the cause(s) is yet to be identified.
However, if the negative trend was the result of the seawater
invasion of a 13C-rich lacustrine system, then not only these
four positive spikes, but all d13C > +6 ‰ might reflect the
isotopic composition of lake water, and be the result of 13C
enhancement by local factors.
SummaryThe overview on the Kuetsj€arvi 13C-rich dolostones
demonstrates that:
1. Diagenetic alteration results only in modest (<1 ‰)
resetting of the carbon isotope system.
2. Caliche carbonates are equally as rich in 13C as sedimen-
tary carbonates and do not show a sizable incorporation
of 13C-depleted, soil-derived components.
3. The high-temperature greenschist metamorphic trans-
formation through dolomite + quartz reaction caused13C-depletion in bulk carbonates by up to 5 ‰ (less
than 2 ‰ in relict dolomicrite) and resulted in a very
“noisy” isotopic pattern.
4. The least-altered d13C values exhibit a stratigraphic profile
displaying a smooth negative trend from +11.5 ‰ to +6 ‰.
5. The smooth stratigraphic drop in d13C from +7 ‰ to
+6 ‰ in the upper part of the section likely corresponds
to invasion of seawater into the lacustrine basin as
suggested by low Sr isotope ratios (0.70406–0.70486).
6. At the end of the Kuetsj€arvi d13C excursion, there are
several positive short-term (within 3-m-thick section)
spikes with amplitude up to 1.7 ‰, which are difficult
to explain by restructuring of the global carbon reservoir.
7. Such spikes may indicate the interplay between global
and local factors in formation of 13C-rich Kuetsj€arvi
carbonates, though the quantification of these factors
requires specifically targeted research and isotopic com-
parison between several distant sections.
8. The Kuetsj€arvi Volcanic Formation has been dated at
2058 � 6 Ma (U-Pb ID-TIMS method, Melezhik et al.
2007) providing a maximum age constraint for the termi-
nation of the Lomagundi-Jatuli event in the Pechenga Belt.
The Imandra/Varzuga Greenstone Belt
In the Imandra/Varzuga Greenstone Belt 13C-rich carbonates
are associated with the Umba Sedimentary Formation, a 50-
to 120-m-thick succession resting with depositional contact
on komatiitic basalts of the Polisarka Volcanic Formation
and overlain by alkaline series rocks of the Umba Volcanic
Formation.
The formation is composed of dolostones, marls, shales
and quartzitic sandstones (for details see Chaps. 4.1, 6.1.3
and 6.1.4). Siliciclastic lithofacies occur mainly in the upper
part of the formation and include thin rhythmically- and
lenticular-bedded greywacke, parallel-laminated sandstone-
siltstone-shale, and massive quartz sandstone. 13C-rich car-
bonate lithofacies vary greatly in thickness (20–100 m) and
their physical appearance ranges from variegated marls
through fine-grained (micritic) dolostones to dolarenites
and dolorudites. Marls and dolostones are thinly parallel-
bedded, whereas dolarenites are often pink in colour and
exhibit vague, thick, horizontal bedding and distinct rhyth-
mic bedding (Fig. 7.36f) with numerous submarine erosional
features including variable-scale erosional channels. Beds of
dolomite-cemented ultramafic breccia (Fig. 7.36g), and
lenses of jaspers and variegated cherts with barite (up to
6 wt% Ba) are common.
A deep, low-energy, marine basin occasionally affected
by submarine hydrothermal processes defines the overall
depositional framework of the Umba Sedimentary
1118 V.A. Melezhik et al.
Formation carbonate rocks (Melezhik and Predovsky 1982;
Melezhik 1992). The observed lithofacies variations and
sedimentological features of the carbonate rocks suggest
their resedimented (clastic) origin. Siliciclastic (greywacke
and shale) sedimentation was affected by weak tidal
currents. The late stage of clastic sedimentation (quartz
sandstone) apparently occurred in a non-marine setting
(details are presented in Chaps. 4.1, 6.1.3 and 6.1.4).
Detrital zircons in the Il’mozero Sedimentary Formation
have been dated at c. 2052 Ma by the U-Pb ID-TIMS
technique (Martin et al. 2010). An inferred source for the
detrital zircon in the Il’mozero Sedimentary Formation is the
stratigraphically underlying volcanic rocks of the Umba
Volcanic Formation. Isotopically “normal” carbonate rocks
form part of the Il’mozero Sedimentary Formation
(Melezhik and Fallick 1996) and provide a maximum age
constraint to the deposition of 13C-rich Umba dolostones and
the termination of the Lomagundi-Jatuli excursion.
The Umba Sedimentary Formation underwent metamor-
phic alteration with the grade ranging from chlorite-epidote-
actinolite to biotite-epidote-actinolite along a 150 km strike-
length (Petrov and Voloshina 1995). Isotopic research of
these rocks is in its infancy (e.g. Melezhik and Fallick
1996; Pokrovsky and Melezhik 1995). The succession was
sampled from surface outcrops as well as from a single
drillhole that intersected the upper part of the formation
(Fig. 7.40). The entire d13C range obtained from whole-
rock samples is from �0.6 ‰ to +5.4 ‰ (n ¼ 25). A
drillcore-based (hole 337, Fig. 7.40b, d) stratigraphic profile
displays a considerable and erratic d13C fluctuation between
�1.8 ‰ and +5.4 ‰, whereas d18O (V-SMOW) ranges
between 10.7 ‰ and 14.6 ‰ (n ¼ 20). An important pecu-
liarity of the Umba Formation dolostones is a significant
variation of d13C within individual sections as well as
between sections (Fig. 7.40c, d). The overall d13C average
based on 45 samples is +3.2 � 2.1 ‰. This is significantly
lower than that of the Kuetsj€arvi Sedimentary Formation
(+7.4 � 0.7 ‰, n ¼ 167). However, these two formations
occur within the common North Transfennoscandian Green-
stone Belt and represent supposedly chronologically near-
correlative units (2058 � 6 Ma vs. ~2052 Ma, respectively).
Although, the formations in question underwent similar
grades of metamorphic alteration under low-temperature
greenschist facies conditions, lower d18O and its positive
covariation with d13C in the Umba dolostones suggest a
possible metamorphic resetting of both isotope systems in
the Umba rocks (Fig. 7.40e, hole 337).
Summary1. The great variation of d13C within and between the
sections, if syndepositional in nature, indicates possible
interplay between local and global factors governing
C-isotopic composition of ambient water bicarbonate.
2. The significant difference in average d13C of the Umba
carbonate rocks (+3.2 � 2.1 ‰) with respect to that of
the supposedly chronologically correlative Kuetsj€arvi
carbonate rocks (+7.4 � 0.7 ‰) accompanies their
drastic differences in depositional settings (open marine
versus lacustrine, respectively); the nature of this link
(coincidental or causal) is worth investigation, and pos-
sible metamorphic overprint should be taken into
account.
3. Detrital zircons from the Il’mozero Sedimentary Forma-
tion have been dated at c. 2052 Ma (U-Pb ID-TIMS
method, Martin et al. 2010) providing a maximum age
constraint for the termination of the Lomagundi-Jatuli
event in the Imandra/Varzuga Greenstone Belt.
The Onega Basin
In the Onega basin (see Chap. 4.3), situated at the south-
eastern margin of the Fennoscandian Shield (Fig. 3. 10a), the
Lomagundi-Jatuli positive isotopic excursion is recorded in
the Tulomozero Formation. This is a 680-m-thick succession
of stromatolitic dolostones, magnesite, dissolution/collapse
breccias, sandstones, siltstones and mudstones
(Fig. 7.36h–k). Most of the rocks are red in colour or
variegated (Fig. 7.36h–k), and thus were accumulated in
oxidised environments. They were deposited in varied plat-
form settings ranging from fluvial and playa to sabkha and
intertidal environments (Melezhik et al. 2000). Evaporitic
features and evaporite mineral growth are abundant and
occur throughout the entire formational thickness and across
an area of more than 2,000 km2. They are present as
dissolution-collapse breccias (Fig. 7.36i) as well as Ca-
sulphates that were entirely or partially pseudomorphed by
quartz, calcite and dolomite. Such features are present in
playa brown mudstones and fenestral stromatolitic sheets
(Fig. 7.36k); in sabkha and supratidal desiccated stromato-
litic sheets and in intertidal lenticular-bedded siltstone-
mudstone (for details see Chaps. 7.8.2 and 7.5). Halite
casts were reported from playa brown mudstone (Melezhik
et al. 2000). Recently, a 3,500-m-deep drillhole (Onega
parametric drillhole, Fig. 7.41a) intersected a c. 200-m-
thick bed of halite at the base of the Tulomozero Formation
(2,940–2,740 m) overlain by a c. 290-m-thick interval of
massive anhydrite interbedded with magnesites and
siltstones (Morozov et al. 2010; for details see Chap. 7.5).
The formation was imprecisely dated at 2090 � 70 Ma
(Pb-Pb age on dolomite; Vasileva et al. 2000). The suc-
cession underwent deformation and prehnite-pumpellyite
to low-temperature greenschist facies metamorphism
(Volodichev 1987) at around 1890 Ma. The metamorphic
parageneses of the greenschist facies dolostones are
defined by the following reactions:
3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1119
3dolomite þ 4quartz þ 1H2O ! 1talc þ 3calcite213C� depleted� �þ 3CO2" 13C� enriched
� �
3dolomite þ K� feldspar þ H2O ! phlogopite
þ 3calcite213C� depleted� �þ 3CO2 "
13C� enriched� �
Isotopic research on the Tulomozero Formation has a long
history. It began in the 1960s and led to the first observation
of 13C-rich Palaeoproterozoic sedimentary carbonates
(Galimov et al. 1968). Later, a series of specifically targeted
investigations was carried out (Yudovich et al. 1991; Karhu
1993; Melezhik et al. 1999a, 2000, 2001, 2005c). The suc-
cession was sampled from surface outcrops as well as from
several drillholes intersecting the entire formational thick-
ness (Fig. 7.41a). Based on previous work (Melezhik et al.
1999a, 2005c), the entire d13C range obtained from the least-
altered dolostone and limestone whole-rock samples from
surface outcrops and drillhole core is from +3.5 ‰ to
+18.0 ‰ (+10.4 � 2.6 ‰ on average, n ¼ 438).
A drillcore-based (holes 7 and 9, Fig. 7.41b) stratigraphic
profile displays a considerable d13C fluctuation at the base
(from +10 ‰ to +17.1 ‰), whereas the following part of the
section represents an overall smooth decreasing trend from
c. +12 ‰ to +7.5 ‰, and then a rise to +10 ‰ with a few
positive outliers (Fig. 7.41b). The stratigraphic top is remi-
niscent of the lower part of the section in that it is marked by
a wide d13C range from +6.1 ‰ to +15.4 ‰. Although
detailed study revealed no apparent significant diagenetic/
metamorphic resetting of the carbon isotope system in the
Tulomozero Formation (e.g. Melezhik et al. 1999a), taking a
conservative stand, one may assume that all values depleted
in 13C with respect to the average d13C curve (red lines in
Fig. 7.41b) might have experienced a variable degree of
post-depositional alteration. However, this cannot be applied
to several sharp, positive departures of around 4 ‰ to 7 ‰occurring in the lower, middle and upper parts of the section;
diagenetic and metamorphic alterations would commonly
drive carbonates isotopically low. There is no indication
that anaerobic remineralisation (i.e. methanogenesis) has
been involved in production of 13C-rich fluids and formation
of these exceptionally high d13C carbonates. Nor can these
multiple sharp rises and falls within a few metres be
explained by the restructuring of global carbon reservoir.
Hence, such anomalies would require a local 13C enhance-
ment by as yet unidentified processes.
Two drilled sections, which are located 100 km apart
(7 and 9 vs. 4699 and 5177), demonstrate similar d13Chistograms showing multi-modal distributions (Fig. 7.41c).
These two sections also exhibit comparable d13C strati-
graphic trends (shown by red lines in Fig. 7.41b) in the
lower and middle parts of section, whereas the stratigraphic
top appears isotopically different. This implies that upper-
most carbonates from these two successions, separated by a
distance of 100 km, might have incorporated carbon derived
from isotopically different sources. If these carbonates are
from time-equivalent successions, then one of them (or
perhaps both) does not reflect the global d13C signal.
Hence, local factors of 13C enhancement cannot be ruled
out. This is certainly the case for the 7 ‰ sharp positive
spike recorded in drillhole 7 (Fig. 7.41b).
A d13C stratigraphic profile, summarising the total isoto-
pic dataset obtained from cores (7, 9, 4699 and 5177),
suggests that an average curve can be drawn through the
middle part of the succession, whereas high d13C scatter in
upper and lower parts cannot as confidently be averaged
(Fig. 7.41d). A histogram, summarising the same dataset
shows three- or perhaps even four-modal distribution
(Fig. 7.41e), consistent with either different age- or process-
related subsets being involved. However, available palaeode-
positional reconstruction suggests four major types of envi-
ronment with different degrees of basinal restriction
(Melezhik et al. 2000, 2001). When isotope data are plotted
against inferred depositional settings, the broad range of d13Cbecomes clustered, and shows a strong dependence on
palaeoenvironment (Fig. 7.41f). The playa carbonates are
most enriched, whereas those from the intertidal settings
with sporadic evaporites exhibit the lowest values clustered
within a single symmetrical mode. The positive correlation
of 87Sr/86Sr with d13C (Fig. 7.41g) corroborates the environ-
mental dependence of isotopic composition of the
Tulomozero carbonates (see also Kuznetsov et al. 2010).
The carbonates from more restricted basinal settings are the
most enriched in 13C and most influenced by 87Sr-rich conti-
nental waters. However, the bulk of the most 13C-enriched
carbonates from the restricted basinal environments are
associated with the lower part of the succession, whereas
the least enriched ones are from open marine settings in the
upper part. Hence, the discrimination diagram shown in
Fig. 7.41f represents a composite depositional and strati-
graphic trend. Confident discrimination between the two
trends requires specifically targeted future research.
Summary1. The Tulomozero Formation dolostones record the
greatest enrichment in 13C with d13C ¼ +18 ‰, which,
if global, would correspond to an unlikely forg ¼ 0.96 on
a steady state model, but see Rothman et al. (2003).
2. The formation has a large array of shallow-water
stromatolites, abundant Ca-sulphates and halite.
3. d13C displays an overall smooth, upward, stratigraphic
decline from c. +17 ‰ to c. +7 ‰ over a c. 700-m-long
section.
4. Several sharp positive departures with a magnitude of up
to 7 ‰ punctuate this stratigraphic trend at different
depth intervals and some spikes (2 and 3, Fig. 7.41b)
1120 V.A. Melezhik et al.
can be roughly correlated in two areally separated
sections, whereas others (1 and 4) cannot.
5. Sharp isotope fluctuations of high magnitude cannot eas-
ily be explained by restructuring of global carbon reser-
voir (but see e.g. Dickens 1999) and were very likely
governed by local depositional factors.
6. Strong dependence of d13C on palaeoenvironments
together with its positive correlation with 87Sr/86Sr
corroborates the above inference.
7. The constraint on causes for all sharp positive departures
remains relatively understudied and forms an important
subject for future research.
The Kalix Greenstone Belt
The Palaeoproterozoic Kalix Greenstone Belt is located at
the northern end of the Bothnian Bay in Sweden. It comprises
Lower, Middle and Upper groups. The Lower group
represents a c. 3,000-m-thick unit that rests on the Archaean
basement and comprises subaerially erupted basalts
interbedded with fluviatile conglomerates. The succession
was deposited in an intraplate rift setting (Lager and Loberg
1990). It was truncated by a break-up unconformity, above
which is the Middle group. This is a 1,200-m-thick succes-
sion of dolostones, arenites, and volcaniclastic and mafic
volcanic rocks that were accreted in markedly variable depo-
sitional environments associated with transition from a
marine-influenced rift to a rimmed carbonate shelf/platform.
The overlying Upper group is a more than 2,000-m-thick unit
of deep-water, Corg-bearing shales deposited on the drowned
carbonate platform of the Middle group in response to tec-
tonically enhanced subsidence (Wanke and Melezhik 2005).
Depositional ages of the Kalix Belt rocks are poorly
constrained. Two suites of granitoids dated at 1890–1860
and 1800–1770 Ma (Ski€old 1987) cut the supracrustal rocks
of the Kalix Belt (Ohlander et al. 1992), thus providing an
upper age limit for deposition. Karhu (1993) suggested, and
Melezhik and Fallick (2010) confirmed, that the Middle
group carbonates are isotopically similar to the upper
Rantamaa Formation dolostones (Kivalo Group, Per€apohja
Belt in Finland) whose maximum age limit was constrained
at 2090 � 70 Ma (Huhma et al. 1990).
The Lomagundi-Jatuli isotopic event is recorded in the
Middle group dolostones (Karhu 1993; Melezhik and Fallick
2010). A detailed sedimentological study of a c. 800-m-thick
section of the group, presented in Wanke and Melezhik
(2005), was informally subdivided into Lower, Middle and
Upper formations (Fig. 7.42).
The Lower formation rests unconformably on the Lower
group terrestrial rocks with pillow basalts at the base that
signify a flooding event. The remainder of the formation
comprises interbedded arenites, stromatolitic dolostones,
mafic tuffs and minor amygdaloidal basalts totalling 200 m
in thickness. A large morphological array of intertidal stro-
matolitic build-ups is a characteristic feature of the forma-
tion (Fig. 7.36l–n). The depositional framework is defined
by irregular repetition of supratidal, intertidal and subtidal
facies suggesting frequent relative sea level fluctuations.
Three short-term phases of emergence and non-marine depo-
sition were documented in the upper part of the succession.
This is expressed by abundant desiccation features, tepees
and subaerially erupted lavas (Wanke and Melezhik 2005).
The Middle formation is a c. 500-m-thick pile of mafic
volcanic rocks. They rest conformably on a mixed
dolostone-siliciclastic lithofacies of the Lower formation
and comprise mainly subaerially erupted amygdaloidal
lavas with minor mafic lava breccias and tuff-cemented
dolostone breccias.
The Upper formation is separated from underlying sub-
aerial lavas by a local unconformity and a flooding event. Its
base comprises a thin unit of mafic pillow lava, whereas the
following 380-m-thick succession is composed of mafic
volcaniclastic rocks succeeded by biohermal stromatolitic
dolostones with thin arenite beds. The rocks were deposited
on a passive margin within a rimmed shelf/platform
environment.
All dolostones comprising the Middle group are enriched
in 13C. The dolostones show a metamorphic grade progres-
sively increasing down section from greenschist to epidote-
amphibolite facies (Fig. 7.42a). The metamorphic paragene-
sis of greenschist facies Kalix dolostone is defined by the
following reaction:
3dolomite þ 4quartz þ 1H2O ! 1talc þ 3calcite213C� depleted� �þ 3CO2" 13C� enriched
� �
All four minerals were observed. Relict quartz suggests
incomplete reaction between dolomite and quartz. The meta-
morphic paragenesis in the epidote-ampibolite facies
dolostones is a product of the following reaction:
Dolomite þ Quartz þ 1H2O ! 1tremolite þ 3calcite213C� depleted� �þ 7CO2" 13C� enriched
� �
In the studied case, quartz deficiency with respect to dolo-
mite for the production of tremolite resulted in the formation
of the tremolite + calcite2 + dolomite paragenesis in addition
to tremolite + calcite2. In both metamorphic facies rocks,
calcite3 is present filling thin, extensional joints; this is
associated with the latest (retrograde) event of the tectono-
metamorphic history (Melezhik and Fallick 2010).
The carbonate-bearing succession was sampled from sur-
face outcrops, and d13C was obtained from 236 whole-rock
and 43 microcored samples (Fig. 7.42a; Melezhik and
Fallick 2010). The overall range of measured d13C values
is between �0.3 ‰ and +8.5 ‰ (+5.5 � 1.6 ‰ on average,
3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1121
n ¼ 279). A d13C histogram exhibits a weakly bimodal
distribution and perhaps suggests the presence of two
subsets, age- or process-related (Fig. 7.42b). A d13C versus
Mg/Ca plot suggests three subsets (Fig. 7.42c). Two discrete
subsets are represented by dolostone. Group 1 dolostone has
d13C ranging between +6.6 ‰ and +7.5 ‰ (high d13Cdolostone), whereas Group 2 has d13C in the range of
+3.5 ‰ to +5.6 ‰ (low d13C dolostone). The third subset
(Group 3) includes dolostones which were variably
calcitised due to dolomite + quartz reaction, and have d13Cranging from �0.3 ‰ to +4.3 ‰. Group 3 dolostone
exhibits a significant, positive d13C-Mg/Ca correlation,
whereas Groups 1 and 2 dolostones do not (Fig. 7.42c).
Group 3 dolostones containing >10 wt% SiO2 show also a
significant, but negative, correlation between SiO2 and d13C,
whereas those with <10 wt% SiO2 do not (Fig. 7.42c). The
SiO2-d13C negative and d13C-Mg/Ca positive correlations
were attributed to metamorphic depletion in 13C, whose
magnitude was controlled by the amount of quartz, which
reacted with dolomite and water, producing 13C-depleted
calcite2 and 13C-rich CO2 (Melezhik and Fallick 2010).
However, in this subset, d13C obtained from whole-rock
samples containing <10 wt% SiO2 do not show dependence
on quartz content, and correspond closely to d13C measured
from microcored dolospar, and so were considered to repre-
sent the least altered values in Group 3 (Melezhik and
Fallick 2010).
Interestingly, a few dolostone clasts from the middle part
of the section show d13C ranging between +6 ‰ and
+8.3 ‰, which is around 2–4 ‰ higher than the isotopic
value of host dolostones that represent the inferred strati-
graphic trend (Fig. 7.42a). If such clasts were derived from
elsewhere-located strata but deposited synchronously with
the clast-hosting dolostones, then this is a case of a non-
global signal: the matter is worth pursuing.
Detailed sedimentological and isotopic research in the
Kalix Belt revealed a d13C stratigraphic trend, which differs
in terms of its magnitude and internal structure compared to
those obtained from the Pechenga Belt and the Onega Basin.
Within a 600-m-thick succession of alternating volcanic,
volcaniclastic, siliciclastic and dolomitic rocks, the least
altered dolostone samples show a gentle oscillation between
+2 ‰ and +4 ‰ throughout the stratigraphy with a second-
order positive excursion from +4 ‰ through +8 ‰, and
gradually back again to +4 ‰ in the c. 150-m-thick unit in
the middle and upper parts of the succession. This second-
order excursion coincides with the transition from a marine-
influenced rift to a passive margin setting (Wanke and
Melezhik 2005; Melezhik and Fallick 2010). If the long-
distance lithological and isotopic correlation of the Kalix13C-rich carbonate succession with the dated Rantamaa For-
mation in Per€apohja Belt (Karhu 1993; Melezhik and Fallick
2010) is correct, then the Kalix d13C section represents the
termination of the Lomagundi-Jatuli isotopic event.
Summary1. The 600-m-thick Kalix succession shows a gentle oscilla-
tion between +2 ‰ and +4 ‰ throughout the stratigraphy.
2. The second-order positive excursion of around +4 ‰(corresponding change in forg is from0.4 to 0.6)might reveal
internal structure towards the end of the Lomagundi-Jatuli
excursion; it may or may not reflect the global signal, and
more detailed work in chronostratigraphically equivalent
sections representing the end of the Lomagundi-Jatuli
event is needed to make progress in this area.
Other Fennoscandian Examples
Karhu (1993) measured C- and O-isotopes in numerous
limestone and dolostone occurrences in the eastern part
of the Fennoscandian Shield. He performed and discussed
229 analyses that include samples obtained from c.
2500–1900 Ma carbonate formations. The total database
suggested d13C ranging between �4 ‰ and +16 ‰ and a
bimodal distribution with strong maxima at c. 1 ‰ and c.
10 ‰ (Karhu 1993). Another outcome of this research
was a provisional d13C evolution curve spanning the c.
2500–1900 Ma time interval (Fig. 7.34). In his summary,
Karhu (1993) divided the d13C curve into five stages. The
first stage precedes the onset of the Lomagundi-Jatuli excur-
sion and is characterised by rare carbonate sedimentation
documented in Finnish Lapland and in the Imandra/Varzuga
Belt; d13C ranges between �4 ‰ and �1.5 ‰. The second
stage, the rise of d13C from c. 0 ‰ to +8 ‰, is hypothetical
in that it was not confidently documented on the
Fennoscandian Shield. The third stage includes most of the
carbonate formations showing systematic enrichment in 13C
with d13C fluctuating between +8 ‰ and +12.5 ‰ within a
c. 2200–2100 Ma time span. Karhu pointed out that carbon-
ate rocks of this stage do not represent a single stratigraphic
unit; instead they form several units that were accumulated
in connection with prograding rifting of, and sedimentation
episodes on, the epicontinental platform(s). Their deposi-
tional settings vary from supratidal (abundant mudcracks)
to intertidal (Karhu 1993). The fourth stage records an
approximately 10 ‰ drop in d13C for carbonate formations
occurring in numerous depositional sites between c. 2110
and 2060 Ma, all showing intertidal to subtidal features
(Fig. 7.36o–s). The youngest stage five represents the period
after c. 2060 Ma, and is typified by d13C ranging between
�3 ‰ and +3 ‰ documented mainly within the
Svecofennian domain (Fig. 3.3). In a later compilation,
Karhu (2005) reported that the start of the excursion remains
poorly constrained between c. 2320 Ma and 2210 Ma,
whereas its end is well defined between c. 2100 Ma and
2050 Ma.
The three most densely sampled and well-dated sections
in the Kuusamo and Per€apohja belts and Kiihtelysvaara area
1122 V.A. Melezhik et al.
(Fig. 7.43) present several salient features. In all three areas
d13C exhibits considerable fluctuation exceeding 5 ‰between the formations and, importantly, within the
formations. The fluctuations are not associated with post-
depositional resetting (e.g. Karhu 1993), and hence are likely
syn-depositional. The Rantamaa Formation (Fig. 7.43a) is
marked by a rapid drop in d13C from +11.4 ‰ (lower part of
the section) to c. +3 ‰ within the c. 200-m-thick succession
(Karhu 1993); Papineau et al. (2005) reported 2 ‰fluctuations superimposed on the generally declining trend.
However, the overall continuous d13C decrease in the
Per€apohja Belt at the end of the Lomagundi-Jatuli event is
c. 11 ‰, which can be constrained within c. 50Myr based on
available precise radiometric dates (Fig. 7.43a). In all three
sections shown in Fig. 7.43, the transition from 13C-rich
dolostones to those with d13C < +6 ‰ to + 4 ‰ is coinci-
dent with the first appearance of Corg-bearing rocks.
Summary1. The Rantamaa Formation dolostones suggest an abrupt
end to the Lomagundi-Jatuli excursion with rapid decline
in d13C equivalent to a steady state change in forg from c.
0.7 to 0.2 and the first appearance of Corg-bearing sedi-
mentary rocks.
2. The second-order positive 4 ‰ oscillation recorded in the
Kalix section is not seen in the Per€apohja section, but it
may be present in the Kiihtelysvaara section.
3. Intraformational d13C fluctuations of the order of 5 ‰recorded within the Dolomite Formation (Kuusamo) and
Kivalo Group (Per€apohja) may or may not reflect
restructuring of the global carbon reservoir; this cannot
be resolved until reliable information on the stratigraphic
position of the individual samples is provided.
4. There is a clear indication that d13C stratigraphic trends
recorded in three dated sections cannot be easily correlated.
d13C Distribution Patterns of the Lomagundi-JatuliAge Sedimentary Carbonates in theFennoscandian ShieldFigure 7.44 summarises d13C data obtained from various
formations/basins in the form of histograms. The currently
available database (909 analyses) suggests a four-modal
distribution with the overall d13C range between �2 ‰and +18 ‰ with two major modes at c. +4 ‰ and c.
+8 ‰, and two smaller ones, at c. +13 ‰ and c. +16 ‰(Fig. 7.44a). The d13C histograms representing different
formations exhibit different distribution patterns and
appear as several disparate populations (Fig. 7.44b). The
Finnish data were obtained from various formations whose
depositional age was constrained at between 2,200 and
2,060 Ma (Karhu 1993, 2005). Hence the Finnish histo-
gram represents the most complete dataset from the
Fennoscandian Shield in terms of age-coverage, and thus
can be used as a reference diagram. In contrast, each other
individual histogram shown in Fig. 7.44b summarises data
obtained either from a single formation or from two suc-
cessive formations at a single depositional site. Each of
these histograms/formations exhibits a d13C range differ-
ent from the others. From their comparison, it becomes
apparent that the observed d13C range correlates with
formational thickness. For example, the thickest forma-
tion, which is the Tulomozero Formation (c. 500 m in
thickness), and two combined successive formations
(c. 600 m in thickness) in the Kalix Belt, each exhibit a
much larger d13C range (from +5 ‰ to +18 ‰, and from
�1 ‰ to +8 ‰) with respect to that of the Kuetsj€arvi
Sedimentary Formation (from +5 ‰ to +9 ‰), which is
only c. 80 m thick (Fig. 7.44b).
In the absence of dates constraining deposition of these
successions, it remains unresolved whether or not such iso-
topic differences between various sites reflect stratigraphic
trends. It is tempting to infer that the very tight Kuetsj€arvi
histogram with a total range +5 ‰ to +9 ‰ (and c. 70% of
the data between +7 ‰ and +8 ‰) might reflect a relatively
short time-span ‘snapshot’ of part of the Lomagundi-Jatuli
excursion.
The d13C histograms from the two thickest (and roughly
equal in thicknesses) formations, the Tulomozero and Kalix
carbonate successions, display only partial overlap at
between +5 ‰ and +8 ‰. Interestingly, in both cases,
such a range characterises the uppermost part of the
successions (Fig. 7.41b, d vs. Fig. 7.42), which pass upward
into younger, organic-rich formations recording the Shunga
event (Melezhik et al. 1999a, b; Wanke and Melezhik
2005; Melezhik and Fallick 2010). This suggests that
d13C > +8 ‰ is not present in Kalix and d13C < +5 ‰ is
not recorded in Tulomozero, and suggests that regardless of
the relative age of the formations in question, the Kalix data
record a syndepositional negative inflection down to +2 ‰within the Lomagundi-Jatuli d13C curve; this corroborates
an earlier proposal by Melezhik and Fallick (2010).
The Finnish reference histogram covers almost the entire
d13C span documented in the Tulomozero Formation apart
from values ranging between +15.1 ‰ and +18 ‰. This
suggests that either the Finnish dataset does not represent
the entire time-range of the excursion, or the Tulomozero
data do not reflect a global d13C marine signal and represent
locally modified d13C values. In fact, it is not only the
Finnish data, but also those of the Kuetsj€arvi and Kalix
carbonate successions, as well as other Fennoscandian
formations in Russia and Norway (Melezhik and Fallick
1996), that do not exceed +15.1 ‰. Hence, the second
option, namely, locally modified d13C, seems to be the
most plausible explanation for the extreme values recorded
by the Tulomozero formation carbonates. Such an inference
is also in agreement with the d13C–depositional environment
plot presented in Fig. 7.41f.
3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1123
7.3.5 The Lomagundi-Jatuli Isotopic Excursion:Unresolved Problems and Implicationsof FAR-DEEP Core for Future Work
Victor A. Melezhik, Anthony E. Fallick,Alex T. Brasier, and Lee R. Kump
The main achievements from studies of the Lomagundi-
Jatuli d13Ccarb positive excursion since the discovery of
anomalously high 13C/12C carbonates (Galimov et al. 1968;
Schidlowski et al. 1975) have been the realisation of its
global significance (Baker and Fallick 1989a, b) and the
constraint on its minimum duration of c.140 Myr (Karhu
1993; Karhu and Holland 1996). Several major unresolved
issues include the following:
1. Robust and precise time constraints for the onset of the
excursion.
2. Internal structure of the d13C curve.
3. Uncovering the true values of the global d13C marine
signal throughout the duration of the excursion.
4. Role of local factors/processes in modification of the
global d13C marine signal.
5. Mechanisms for the onset and termination of the
excursion.
Resolving issues 1 and 2 in particular, by employing U-Pb
techniques on magmatic minerals, appears to be challenging.
This is because the apparent onset and a significant part of the
excursion coincide in time with the c.2.45–2.2 Ga magmatic
shut/slowdown (e.g. Condie et al. 2009), hence magmatic
zircons/monazites are largely unavailable for radiometric dat-
ing. Another possible direction for making progress in this
area is in situ U-Pb dating of diagenetic xenotime, which is a
relatively common mineral in siliciclastic sedimentary rocks.
It appears as diagenetic overgrowths on detrital zircons
forming shortly after sediment deposition. The method has
potential for dating sedimentary sequences of all ages but is
considered to be especially valuable for refining the Precam-
brian time scale (McNaughton et al. 1999; Fletcher et al.
2000; Vallini et al. 2002). It has a great potential for providing
radiometric dating of the Lomagundi-Jatuli excursion on the
Fennoscandian Shield: all FAR-DEEP drillholes intersecting
supposedly Lomagundi-Jatuli-time, 13C-rich carbonate suc-
cession (see Chaps. 6.1.3, 6.2.2, 6.3.1 and 6.3.2 describing
Holes 4A, 5A, 10A, 10B and 11A) contain a large proportion
of siliciclastic sedimentary rocks with abundant clastic zircon.
Issues 3 and 4 can be elucidated and resolved by com-
parison of d13C stratigraphic curves in time-equivalent
successions accumulated in settings with variable palaeo-
tectonic and palaeonevironmental conditions. However,
such exercises could only be achieved if synchronous
deposition of the compared formations is confidently
proven. Strictly speaking, such a task represents a tremen-
dous challenge even if cutting-edge radiometric dating
technology is successfully employed. This is because:
(1) even analytical uncertainties of 1–5 Ma would not
allow to discriminate confidently between depositionally
and time-dependent d13C trends, and (2) rocks suitable
for high-precision radio-isotopic dating are lacking in the
key stratigraphic intervals. Nevertheless, the problem might
be partially resolved by detailed petrographic and geo-
chemical investigation of successions exhibiting extremely
high d13C, and those showing considerable d13C fluctua-
tions, or solitary positive and negative d13C spikes (see
Figs. 7.39c and 7.41b as examples). When a local process
(es) is operating and its(their) sedimentological (textural
and structural), mineralogical and geochemical expres-
sion/fingerprints are confidently identified, such knowledge
could potentially be employed in discrimination between
local and global d13C signals in other successions which
may retain only selectively preserved information. FAR-
DEEP cores offer an excellent opportunity for such
detailed sedimentological (textural and structural), miner-
alogical and geochemical research. In addition, FAR-DEEP
and previously obtained cores should enable us to compare/
contrast several d13C stratigraphic trends in successions
with a spatial separation basin-wide in the Pechenga Belt
(Fig. 7.37a) and Onega Basin (Fig. 7.41a).
Issue 5, namely mechanisms for the onset and termination
of the excursion, represents a fundamental problem, which
has been vigorously debated for over two decades (see section
7.3.2 Historical Overview). Almost all “modellers” have
agreed upon the reasonable assumption that a high fractional
organic carbon burial rate is the most plausible mechanism
that drives contemporaneous sedimentary carbonate isotopi-
cally heavy. This can be achieved in various ways, but the
remaining unresolved problem with this approach is that
organic-rich sediments compensating for high-d13C over the
140 million year interval have yet to be discovered. Euxinia
with decoupled phosphate and organic carbon burial is
another of the most frequently applied models. In principle,
the existence of such conditions can be tested by redox prox-
ies (e.g. Fe, U, Cr and Mo isotopes). Various proxies are
discussed to varying degrees in Chap. 7.10, and FAR-DEEP
cores are likely suitable to carry out such investigations.
Assuming it is practically possible to investigate this hypoth-
esis, the remaining problem is to how to sustain such
conditions over a period of at least 140 million years. Here,
the application of geochemical proxies is of limited use.
V.A. Melezhik (*)
Geological Survey of Norway, Postboks 6315, Sluppen, NO-7491
Trondheim, Norway
Centre for Geobiology, University of Bergen, Allegaten 41,
Bergen N-5007, Norway
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_3, # Springer-Verlag Berlin Heidelberg 2013
1124
1124 V.A. Melezhik et al.
Instead, a comprehensive geotectonic and palaeogeographic
approach is needed for understanding of global land-ocean
mass distribution and continent motions. Perhaps, the answer
lies at least partly in the proposed c. 2.45–2.2 Ga global shut/
slowdown of tectonic activity, global subduction and “frozen”
immobile continents (Condie et al. 2009), and possible
euxinic oceans accumulating organic-rich rocks not preserved
(or yet found) in geological record.
The possible role of a putative magmatic shutdown (or,
more likely, slowdown) during the interval 2.45–2.2 Ga on
the subsequent carbonate carbon isotopic excursion has been
addressed by Condie et al. (2009). They identify several
plausible changes to major carbon transfer processes – lead-
ing to either reduction or enhancement of atmospheric CO2
levels – and point out that supercraton breakup would create
basins suitable for organic carbon burial. Their discussion,
however, is not sufficiently quantitative to allow estimation
of any concomitant change to din, so perhaps this is a topic forfuture work. Interestingly, we are not aware of any major
change in the carbon cycle (and so dcarb) at contrasting times
of very high plate velocities, e.g. ~2.7 Ga.
In a thoughtful and constructive review of this Chapter,
Graham Shields has suggested that consideration be given to
possible changes in D. Radical reorganisation of biospheric
operation in the aftermath of the first global glaciation
should not be casually dismissed. It follows trivially from
Eq. 1 that for dcarb ¼ +10 ‰ (a reasonable estimate for
the peak of the excursion), forg ¼ 0.2 and din ¼ �5, then
D ¼ 75 ‰. Where diffusive transport of CO2 into the cell
increases in importance, the carbon fixed as organic matter
will be relatively depleted in 13C, but such extreme meta-
bolic isotopic fractionation as D ¼ 75 seems unlikely. Per-
haps the major objection to the excursion being forced by D,whilst din and forg remain relatively constant, is why the
carbon cycle should have operated for most of the preceding
billion years and a significant fraction of the subsequent two
billion years in such a way as to have dcarb ~0 � 2.5 ‰. One
need not be an adherent of strict uniformitarianism to be
concerned about hypothesising a perhaps 150 Myr interval
with such an extreme change to D. It is certainly the case thatmore modest variations in species-specific carbon isotope
fractionations are known. House et al. (2003) report values
ranging from 0.2 ‰ to 26.7 ‰ for Archaea and other ther-
mophilic prokaryotes and highlight the potential of growth-
dependent effects. However, in the absence of cogent
arguments as to the likely mechanisms of change, further
discussion of variation in D seems unconstrained and per-
haps best left for future study.
There are two notable exceptions to the rule that
models explaining the Lomagundi-Jatuli event require
the accumulation of organic-rich rocks. The first model
not to include such organic carbon burial was presented
Fig. 7.32 World distribution of Palaeoproterozoic rocks (compiled
by Aivo Lepland) showing geographic locations of c. 2200–2060 Ma13C-rich sedimentary carbonates associated with the Lomagundi-Jatuli
positive excursion of carbonate carbon isotopes. Data are from
Schidlowki et al. (1975), McNaughton and Wilson (1983), Baker
and Fallick (1989a, b), Zagnitko and Lugovaja (1989), Karhu
(1993), Zlobin (1993), Melezhik and Fallick (1996), Melezhik et al.
(1997), Buick et al. (1998), Hofmann and Davidson (1998), Lindsay
and Brasier (2002), Bekker et al. (2003a, b, 2006), Maheshwari et al.
(1999, 2010)
3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1125
by Hayes and Waldbauer (2006). They linked the accu-
mulation of 13C-rich carbonates to the onset of fermenta-
tive and methanogenic diagenesis in deeper levels of the
sediment column, as the response to increasing O2 and
SO42� concentrations in the ocean. The effect of this step
function in microbial ecosystem dynamics was formation
of 13C-rich diagenetic carbonates with the absence of
concomitant variations in the organic matter isotopic
record and without requiring enhanced burial of organic
matter. The model can be tested by identifying various
carbonate phases commonly associated with methane gen-
eration, its recycling and oxidation. The existing studies
(Melezhik et al. 2003; Brasier et al. 2011), of acknowl-
edged limitations, have so far failed to identify any
carbonate phases that validate the proposed mechanism.
FAR-DEEP cores (see Chaps. 6.1.3, 6.2.2, 6.3.1 and 6.3.2)
contain a large array of carbonate phases formed during
the course of diagenesis and hence represent valuable
material for comprehensive research in the search for a
signal from methanogenic diagenesis.
The second model not to rely solely on enhanced organic
carbon burial was presented by Fallick et al. (2008). Simi-
larly to the “methanogenetic” model, they invoked impor-
tant biological change as the response to development of
the O2-rich hydrosphere. The proposed response was the
forced shift of anaerobic microorganisms, which recycle
organic matter, from the previously anoxic water column
and sediment/water interface (from where CO2 and CH4
Fig. 7.33 Age (Ma) of formations within sections recording the
Lomagundi-Jatuli positive d13C isotope excursion (Lomagundi-Jatuli).
Note that there is no y-axis scale; data are spaced in the vertical
direction for clarity only. Green colour denotes formations formed
prior to the Lomagundi-Jatuli event. Red represents formations formed
coeval with the Lomagundi-Jatuli event. Blue colour marks formations
that postdated the Lomagundi-Jatuli event. The dashed boxes highlighttime periods where no radiometric ages are available. The numbers
from 1 to 41 correspond to data in Table 7.1. The x-axis error bars
correspond to errors in ages reported in Table 7.1
1126 V.A. Melezhik et al.
Table
7.1
Global
stratigraphic
unitsthat
record
theLomagundi-Jatulieventandtheirageconstraintswithreferences(D
ataforeach
regionarelisted
instratigraphic
order)
Region
Form
ation/
Stratigraphy
d13C
Ref.to
d13C
No.in
Fig.7.33
No.in
Fig.7.34
(d13C)
No.in
Fig.7.34
(AgeMa)
Age
(Ma)
Error
(Ma)
Ref.to
age
Mineral
dated
Rock
type
dated
Dating
method
min.
max.
S.Africa
Transvaal
BushveldComplex
––
–12
10(U
)2054
238
Zircon
Ign.
SHRIM
P
Transvaal
Houtenbek
Fm.
�3.3
�0.5
4–
––
Transvaal
Silverton
+2
+10
12
10
––
–
Transvaal
Rooihoogte
Fm.
––
–25
7(S);10(L)
2316
715
Pyrite
Aut-Sed.
Re-Os
Transvaal
DuitschlandFm.
�8+10
67
––
–
Transvaal
Oak
TreeFm.
––
–37
2582
15
23
Zircon
Ign.
SHRIM
P
Transvaal
Lucknow
Fm.
+1.5
+10
12
––
–
Transvaal
Mooidraai
Fm.
+0.5
+2
3–
––
Transvaal
Rooinekke
�8�1
12
––
–
Transvaal
Kuruman
Fm.
––
–33
2465
533
Zircon
Ign.
SHRIM
P
N.America
Great
Lakes
NipissingIntrusions
––
–22
4(U
)2217
910
Bdy&
Rt
Ign.
ID-TIM
S
Great
Lakes
GordonLakeFm.
+5
+8.2
54
––
–
Great
Lakes
Espanola
Fm.
�4�0
.85
––
–
Great
Lakes
Basem
ent
––
–31
4(L)
2450
25
19
Zircon
Ign.
ID-TIM
S
Great
Lakes
KonaDolomite
+5
+9.5
5–
––
Great
Lakes
Enchantm
entLake
Fm.
––
–26
2317
641
Zircon
Sed.
SHRIM
P
Great
Lakes
Randville
Fm.
�0.4
+3.1
5–
––
Great
Lakes
SturgeonQuartzite
––
–24
2306
941
Zircon
Sed.
SHRIM
P
Great
Lakes
Bad
River
Dolomite
�0.5
+2.4
5–
––
Great
Lakes
Sunday
Quartzite
––
–40
2647
541
Zircon
Sed.
SHRIM
P
Great
Lakes
Bremen
Creek
GraniteGneiss
––
–7
19(U
)1982
513
Zircon
Ign.
ID-TIM
S
Great
Lakes
Denham
,TroutLake
�1.2
+2.5
519
––
–
Great
Lakes
Basem
ent
––
–14
19(L)
2076
58
Zrn
&
Bdy
Ign.
ID-TIM
S
Labrador
Trough
Nim
ishFm.
––
–3
22(U
)1878
211
Zircon
Ign.
ID-TIM
S
Labrador
Trough
FlemingFm.
00
25
––
–
(continued)
3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1127
Table
7.1
(continued)
Region
Form
ation/
Stratigraphy
d13C
Ref.to
d13C
No.in
Fig.7.33
No.in
Fig.7.34
(d13C)
No.in
Fig.7.34
(AgeMa)
Age
(Ma)
Error
(Ma)
Ref.to
age
Mineral
dated
Rock
type
dated
Dating
method
min.
max.
Labrador
Trough
Denault&
Abner
Fm’s.
�2+4
25
22
––
–
Labrador
Trough
MistamiskFm.
––
–18
11(U
);22
(L)
2142
49
?(Zrn)
Ign.
ID-TIM
S
Labrador
Trough
UveFm.
+5
+8
25
11
––
–
Labrador
Trough
Alder
Fm.
+8
+12
25
––
–
Labrador
Trough
DunphyFm.
+16
+16
25
19
11(L)
2169
437
Zircon
Ign.
ID-TIM
S
Labrador
Trough
Basem
ent
––
–41
2654
528
Zrn
&
Mnz
Ign.
ID-TIM
S
Scandinavia
Onega
Suisaarian
Fm.
––
–6
17(U
)1972
17
34
Zircon
Ign.
SHRIM
P
Kiihtelysvaara
Pet€ aikk€ oFm.
+6
+0.5
18
17
––
–
Kiihtelysvaara
Viistola
Fm.
+4
+12
18
––
–
Kiihtelysvaara
Koljola
Fm.
––
–16
17(L)
2115
630
Zircon
Ign.
ID-TIM
S
Kuusamo
Lim
estoneDolomite
+2
+7
18
15
––
–
Kuusamo
Amphibole
Schist
––
–15
12(U
);15
(L)
2078
839
Zrn
&
Ttn
Ign.
ID-TIM
S
Kuusamo
Dolomite
+8
+11.5
18
12
––
–
Kuusamo
Siltstone
+12
+12
18
20
8(U
);12(L)
2206
939
Zrn,Ttn,
Bdy
Ign.
ID-TIM
S
Kuusamo
SericiteSchist
+7
+8
18
8–
––
Kuusamo
Conglomerate
––
–27
8(L)
2405
639
Zircon
Ign.
ID-TIM
S
Per€ apohja
V€ ayst€ aj€ aFm.
�10
17
10
13(U
)2050
831
Zircon
Ign.
ID-TIM
S
Per€ apohja
Rantamaa
Fm.
+2
+11
17
13
––
–
Per€ apohja
Kvartsim
aaFm.
+9
+10
17
––
–
Per€ apohja
Laurila
Sill
––
–23
5(U
);13(L)
2221
531
Bdy
Ign.
ID-TIM
S
Per€ apohja
PalokivaloFm.
+5
+11
17
5–
–
Per€ apohja
Sompuj€ arviFm.
+8
+8
17
––
–
Per€ apohja
Elij€ arvigranite
––
–28
5(L)
2433
431
Zircon
Ign.
ID-TIM
S
Kalix
Granitoid
––
–4
14(U
)1891
743
Zircon
Ign.
ID-TIM
S
Kalix
Upper
Fm.
+3
+8
24
14
––
–
1128 V.A. Melezhik et al.
Kalix
Lower
Fm.
+2
+4.5
24
––
–
Pechenga
Pilguj€ arviVolcanic
Fm.
––
–5
21(U
)1970
516
Zircon
Ign.
ID-TIM
S
Pechenga
Kolosjoki
Sedim
entary
Fm.
+1
+2.5
26
13
21
9(U
);21(L)
2058
626
Zircon
Sed.
ID-TIM
S
Pechenga
Kuetsj€ arviVolcanic
Fm.
––
–9(U
);21(L)
2058
626
Zircon
Sed./Ign.
ID-TIM
S
Pechenga
Kuetsj€ arvi
Sedim
entary
Fm.
+5.5
+9
26
9–
–
Pechenga
General’skaya
Intrusion
––
–35
9(L)
2505
1.6
1Zircon
Ign.
ID-TIM
S
Imandra-
Varzuga
Il’m
ozero
Sedim
entary
Fm.
––
–11
–2052
Zircon
Ign.
Imandra-
Varzuga
Seidorechka
Volcanic
Fm.
––
–29
3(U
)2442
1.7
1Zircon
Ign.
ID-TIM
S
Imandra-
Varzuga
Seidorechka
Sedim
entary
Fm.
�1�6
27
3–
––
Imandra-
Varzuga
PanaTundra
(gabbro-norite)
––
–34
3(L)
2504
3.7
1Zircon
Ign.
ID-TIM
S
Australia
YilgarnCraton
YelmaFm.
�1+3
20
820
20(S)
2017
15
14
Zircon
Sed.
SHRIM
P
YilgarnCraton
MaraloouFm.
––
–2
16(U
)1843
10
36
Mnz
Ign.
SHRIM
P
YilgarnCraton
JohnsonCairn
Fm.
�0.4
�0.4
20
––
–
YilgarnCraton
JuderinaFm.
+5.5
+8
20
16
PilbaraCraton
JuneHillVolcanics
––
–1
23(U
)1795
742
Zircon
Ign.
SHRIM
P
PilbaraCraton
Duck
Creek
Dolomite
�3+2
20
23
––
–
PilbaraCraton
Wooly
Dolomite
�5+2
20
918
18(S);23(L)
2031
629
Zircon
Ign.
SHRIM
P
PilbaraCraton
KazputFm.
�6.5
+2
20
––
–
PilbaraCraton
Sill
––
–21
6(U
)2208
10
29
Bdy
Ign.
SHRIM
P
PilbaraCraton
Meteorite
Bore
Mem
ber
�1.5
+1
20
6–
––
PilbaraCraton
Woongarra
Rhyolite
––
–30
2449
32
Zircon
Ign.
ID-TIM
S
PilbaraCraton
Brockman
IronFm.
––
–32
2454
332
Zircon
Ign.
SHRIM
P
PilbaraCraton
Wittenoom
Fm.
�2+1.5
20
36
21(U
);2(S)
2561
840
Zircon
Ign.
SHRIM
P
PilbaraCraton
Caraw
ineDolomite
�1.5
+2
20
1
PilbaraCraton
Marra
Mam
baIron
Fm.
––
–38
1(L)
2597
540
Zircon
Ign.
SHRIM
P
(continued)
3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1129
Table
7.1
(continued)
Region
Form
ation/
Stratigraphy
d13C
Ref.to
d13C
No.in
Fig.7.33
No.in
Fig.7.34
(d13C)
No.in
Fig.7.34
(AgeMa)
Age
(Ma)
Error
(Ma)
Ref.to
age
Mineral
dated
Rock
type
dated
Dating
method
min.
max.
India
Rajasthan
UdaipurFm.
�5+1
35
––
–
Rajasthan
Jham
arkotraFm.
�1.5
+5.5
35
––
–
S.America
Sao
Francisco
?–
––
2059
58
22
Zircon
Sed.
ID-TIM
S
Sao
Francisco
?–
––
17
2125
421
Zircon
Sed.
ID-TIM
S
Sao
Francisco
FechodoFunil
+5.5
+7.5
7–
––
Sao
Francisco
Cercadinho
+3
+5.5
7–
––
Sao
Francisco
Gandarela
�1.5
+0.5
7–
––
Sao
Francisco
Moeda
––
–39
2606
47
22
Zircon
Sed.
ID-TIM
S
SSyndepositionalageconstraint,Uupperageconstraint,Llowerageconstraint,Bdy
baddeleyite,Mnz
monazite,Rtrutile,T
tntitanite,Ign.igneous,Sed.sedim
ent,Aut-Sed.authogenicsedim
ent,
ID-TIM
SU-PbIsotopeDilutionThermal
IonizationMassSpectrometry,SH
RIM
PU-PbSensitiveHighResolutionIonMicroprobe.Referencescited:(1)Amelin
etal.(1995),(2)Barleyet
al.
(1997),(3)Bau
etal.(1999),(4)Bekker
etal.(2004),(5)Bekker
etal.(2006),(6)Bekker
etal.(2001),(7)Bekker
etal.(2003c),(8)Buchan
etal.(1996),(9)Clark
(1984),(10)Corfuand
Andrews(1986),(11)Findlayetal.(1995),(12)Frauensteinetal.(2009),(13)GoldichandFischer
(1986),(14)Halilovicetal.(2004),(15)Hannah
etal.(2004),(16)Hanskietal.(1990),(17)
Karhu(2005),(18)Karhu(1993),(19)Kroghetal.(1984),(20)LindsayandBrasier
(2002),(21)Machadoetal.(1992),(22)Machadoetal.(1996),(23)Martinetal.(1998),(24)Melezhik
and
Fallick
(2010),(25)Melezhik
etal.(1997),(26)Melezhik
etal.(2007),(27)Melezhik
etal.(1999),(28)MortensenandPercival
(1987),(29)M€ ueller
etal.(2005),(30)Pekkarinen
and
Lukkarinen
(1991),(31)Perttunen
andVaasjoki(2001),(32)Pickard(2002),(33)Pickard(2003),(34)Puchteletal.(1998),(35)Purohitetal.(2010),(36)Rasmussen
andFletcher
(2002),(37)
Rohonet
al.(1993),(38)Scoates
andFriedman
(2008),(39)Silvennoinen
(1991),(40)Trendallet
al.(1998),(41)Valliniet
al.(2006),(42)Wilsonet
al.(2010),(43)Wilsonet
al.(1987).
1130 V.A. Melezhik et al.
would readily have escaped to the atmosphere), to deep
within the sediments, to escape the rise of deadly poisonous
dioxygen. The effect of such microbial ecosystem dynam-
ics was the creation of a new locus for organic matter
recycling and fixation of 12C-rich by-products (CO2 and
CH4) in three newly created reservoirs: (1) diagenetic
concretions (which are seemingly largely absent earlier in
the geological record); (2) disseminated carbonate cements
resulting from biological remineralisation of organic mat-
ter within the sediment column; and (3) the accumulation
of sediment-associated methane clathrate-hydrates (Fallick
et al. 2011). This model can be tested by looking for
temporal links between the onset of the Lomagundi-Jatuli
excursion and the appearance of widespread 12C-rich dia-
genetic carbonates, including concretions; note that burial
diagenetic cements where the carbon is remobilised
thermally rather than biologically are not included in such
a test.
Another remaining issue to be resolved is the mechanism
for the termination of the excursion. More d13Corg data
covering the Lomagundi-Jatuli-age interval would surely
help to shed light on the merits of competing hypotheses
and on the inherent assumptions, and this ought to be
addressed in future studies. Currently, there seems to be a
dearth of data on d13Corg, and the issue of its thermal alter-
ation (i.e. via H/C ratio; Strauss et al. 1992) has not been
adequately addressed (e.g. Karhu 1993; Bekker et al. 2001,
2003a, b, 2006). In future studies, precautions should be also
taken whether contrasted/compared d13Corg and d13Ccarb
values represent coupled or decoupled carbon reservoirs;
the issues recently discussed by Bekker et al. (2008). One
should avoid contrasting/comparing d13Corg and d13Ccarb
values obtained from carbonate (commonly poor in organic
carbon) and organic-bearing shale (commonly devoid of
carbonates) successions whose “chronostratigraphic” corre-
lation is based on a lithostratigraphic tool.
Fig. 7.34 Variability in d13C in Palaeoproterozoic carbonate
formations based upon U-Pb and Re-Os data presented in Table 7.1.
The vertical bars represent the reported range of d13C. The numbers
correspond to formations listed in Table 7.1. The horizontal bars show
the age uncertainty. The solid grey curve represents the median ages
� 1s of the data set in Table 7.1. The dashed black line is the
Lomagundi-Jatuli curve of Karhu and Holland (1996). I-V denote
five-stage-evolution of the d13C composition of Palaeoproterozoic sed-
imentary carbonates in the Fennoscandian Shield (From Karhu 1993,
2005). See text for a discussion
3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1131
Fig. 7.35 Geographic distribution (red circles) of c. 2330–2060 Ma13C-rich sedimentary carbonates associated with the Lomagundi-Jatuli
positive excursion of carbonate carbon isotopes in Fennoscandia
(Data are from Yudovich et al. (1991), Tikhomirova and Makarikhin
(1993), Karhu (1993), Melezhik and Fallick (1996), Melezhik et al.
(2005b), and Melezhik and Fallick (unpublished). Geological map
modified by A. Lepland from Koistinen et al. (2001))
1132 V.A. Melezhik et al.
Fig. 7.36 Rock images illustrating 13C-rich carbonates of the
Fennoscandian Shield. Kuetsj€arvi Sedimentary Formation: (a)
Variegated, crudely-bedded, resedimented dolostone with platy siltstone
clasts. (b) Beige and pale brown, massive or thickly-bedded resedimented
dolostone that was deposited on a travertine crust (arrowed) that
precipitated on an exposure surface. (c) Pale pink, dolomite-cemented,
quartz sandstone capped by thin travertine crust (black arrow) and pink,calichified, partially dissolved dolomicrite (red arrow) that in turn is
veneered by white travertine; note the large, rounded clasts of white
dolostone surrounded by platy fragments of pink dolostone in the sand-
stone bed. (d) White and pale pink travertine crusts precipitated on
exposure surfaces within massive redeposited dolostone, which is
intersected by a sub-vertical travertine vein. (e) Quartz-sandstone-filled
extensional cracks in pale yellow, massive dolarenite.Umba SedimentaryFormation: (f) Pale pink and purple, parallel-bedded dolarenite
3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1133
Fig. 7.36 (continued) (g) Variegated dolostone with fragments of
altered ultramafic volcanic rocks. Tulomozero Formation: (h) Pink,
columnar mini-stromatolites accreted on an uneven surface of brownish
dolarenite and overlain by flat-laminated, dolomitic stromatolite; note
bleaching along the contact zone. (i) Dissolution-collapse breccia
composed of fragments of pink and brown mudstone cemented by
white dolospar. (j) Dark brown dolomarl overlain by brick-coloured
dolomarl with syn-sedimentary faulting and deformation; note the
dolomite-replaced (white) evaporitic crust occurring along the contact.
Core diameter in (i) and (j) is 4 cm
1134 V.A. Melezhik et al.
Fig. 7.36 (continued) (k) Dolomite-replaced Ca-sulphate nodule
(bright) and plastically deformed layers with enterolithic structure
interbedded with brown and pink mudstone laminae. Middle group:(l) Dome-like dolomitic stromatolite (pale yellow) in mafic tuff (paleand dark brown) containing white, bedded and massive dolomicritic
bed and lenses; coin is 1.5 cm in diameter. (m) Stromatolitic bioherm
accreted on intraformational dolostone breccia with patches of brown
mafic tuff material; note that the mafic tuff bed above the bioherm
contains stone (dolostone) rosette. (n) Stromatolitic biostrome
(arrowed) accreted on and overlain by laminated clayey dolostone,
passing upwards into indistinctly bedded dolostone; scale-bar is
10 cm
3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1135
Fig. 7.36 (continued) Rantamaa Formation: (o, p) Stromatolitic
biostromes composed of columnar (o) and domal (p) stromatolites;
individual biostromes are separated from each other by dark coloured,
clayey, laminated dolostone. (q) Fine-pebble dolostone conglomerate
erosively overlying gritty greywacke. (r) Cross- and parallel-bedded
dolomite-cemented quartz sandstone with abundant rounded, platy
clasts of pale yellow dolostone; coin is 2 cm in diameter. (s)
Rhythmically-bedded (ribbon), light- and dark-coloured dolostone suc-
cession; width of the view is c. 15 m (Photographs courtesy of Kauko
Laajoki (d), Vesa Perttunen (p–r), Eero Hanski (q–s). Photographs
(a–c, e–n) by Victor Melezhik. Sample (h) courtesy of Pavel
Medvedev)
1136 V.A. Melezhik et al.
Fig. 7.37 The Pechenga Greenstone Belt. (a) Outline of the
Pechenga Greenstone Belt showing the position of the Kuetsj€arviSedimentary Formation, sampled and drilled sites, logged sections,
and metamorphic zones. Metamorphic zones and mineral equilibrium
metamorphic temperatures are from Petrov and Voloshina (1995). (b)
Histogram of d13Ccarb for Kuetsj€arvi Sedimentary Formation
carbonate rocks based on sampling from surface outcrops and
drillhole X cores (Data are from Melezhik and Fallick 2001;
Melezhik et al. 2005b). (c) Histograms of d13Ccarb for Kuetsj€arviSedimentary Formation carbonate rocks from individual sampling
sites located within different metamorphic zones (Data are from
Melezhik and Fallick 2001; Melezhik et al. 2005b)
3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1137
Fig. 7.38 Different carbonate phases of Kuetsj€arvi Sedimentary
Formation dolostones, and their carbon and oxygen isotopic
compositions (Modified from Melezhik and Fallick 2003; Melezhik
et al. 2004). (a) Tremolite (Tr)-calcite2 (Cal2)-dolomicrite (DM)
rock from a high-temperature epidote-amphibolite zone; sample
collected from the Kola Superdeep Drillhole core. (b) Mottled,
flat-laminated, dolomitic stromatolite composed of pale pinkdolomicrite (DM), quartz (Q) and dolospar filling voids and
fenestrae (FDS); sample collected from drillhole X core. (c) Sandy,
allochemical dolostone consisting of detrital quartz (dark greygrains) and dolomicrite intraclasts (DM) cemented by dolospar
(DS); sample collected from drillhole X core
1138 V.A. Melezhik et al.
Fig. 7.38 (continued) (d) Thin caliche profile underlain by laminated
dolomitic travertine (3, 5–9, 11), deposited on dolospar-cemented
sandstone (1, 2). The caliche is composed of red, iron-stained, non-
laminated dolomicrite (CDM), pale grey, cloudy dolospar (CDS)
cementing dissolution cavities, and dolomicrospar (CDMS); sample
collected from a quarry. Numbers on the scanned thin section denote
drilling/sampling sites
3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1139
Fig. 7.39 Comparison of carbon isotopes in Kuetsj€arvi Sedimentary
Formation dolostones from different metamorphic zones. (a)
Lithostratigraphic section through the Kola Superdeep Drillhole;
boundaries of the metamorphic zones between the Superdeep
Drillhole (Glagolev et al. 1987; Petrov and Voloshina (1982)) and
the surface (Modified from Petrov and Voloshina 1995) are inferred.
(b) Plots of d13C versus metamorphic facies based on data from
surface outcrops, and core from drillhole X and Kola Superdeep
Drillhole. (c) d13C stratigraphic profile based on core from drillhole
X and Kola Superdeep Drillhole (Panels (a) and (b) modified from
Melezhik et al. (2003); (c) based on data from Melezhik et al.
(2003) and Melezhik et al. (2005b))
1140 V.A. Melezhik et al.
Fig. 7.40 Carbon-isotopic composition of Umba Sedimentary Forma-
tion dolostones. (a) Geographic position of the Imandra/Varzuga
Greenstone Belt (IVGB). (b) Outline of the the Imandra/Varzuga
Greenstone Belt showing position of the Umba Sedimentary Forma-
tion, sampling and drilling sites, and logged sections; site 2 – drillhole
337, site 4 – FAR-DEEP Hole 4A. (c) Histogram of d13Ccarb for
showing d13C variations in carbonate rocks along strike of the Umba
Sedimentary Formation based on sampling from surface outcrops and
drillhole 337 core (Isotopic data are from Melezhik and Fallick 1996).
(d) d13C stratigraphic profile through the upper part of the formation
based on the drillhole 337 core. (e) A d13C versus d18O plot based on
the drillhole 337 core (Isotopic data from Melezhik and Fallick 1996)
3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1141
Fig. 7.41 Carbon-isotopic composition of Tulomozero Formation
dolostones. (a) Geological map of the Onega Basin (Modified by
Aivo Lepland from Koistinen et al. 2001) showing drillhole positions.
(b) Generalised lithological columns of the Tulomozero Formation
with d13C stratigraphic profiles (Modified from Melezhik et al. 2000,
2005c); A to H denote lithostratigraphic members and 1–4 sharp
positive departures of d13C from background values. (c) Core-based
d13C histograms from two correlative sections (drillholes 5177/4699
and 7/9) located c. 100 km apart
1142 V.A. Melezhik et al.
Fig. 7.41 (continued) (d) Combined d13C stratigraphic profile
through the Tulomozero Formation based on two sets of drillholes
located c. 100 km apart (5177/4699 – red dots, and 7/9 – blackdots); note that d13C data have been used for the correlation of the
drilled sections (by eye as the best fit), and no robust lithological
criteria are currently available. (e) d13C histogram combining data
obtained from drillholes 5177/4699 and 7/9. (f) d13C histograms
versus inferred depositional environments (Isotope data are from
Melezhik et al. 1999a; environmental interpretations are from
Melezhik et al. 1999a, 2000). (g) 87Sr/86Sr versus d13C plot for
Tulomozero Formation carbonates (Reproduced from Melezhik
et al. 2005c)
3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1143
Fig. 7.42 Carbon-isotopic composition of carbonate rocks from the
Kalix Greenstone Belt. (a) d13C stratigraphic profile through the Mid-
dle group carbonate rocks (Modified from Melezhik and Fallick 2010).
(b) d13C histogram for Middle group carbonate rocks. (c) d13C versus
Mg/Ca and SiO2 plots for three groups of dolostones (Modified from
Melezhik and Fallick 2010)
1144 V.A. Melezhik et al.
Fig. 7.43 The best dated sections containing 13C-rich dolostones in Finland (Modified from Karhu (1993), which also contains references for
isotopic and age data)
3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1145
Fig. 7.44 d13C histograms of Lomagundi-Jatuli age carbonate rocks
from the Fennoscandian Shield. (a) Histogram summarising
published up-to-date analyses (Data are from Yudovich et al.
1991; Karhu 1993; Melezhik and Fallick 1996, 2010; Kortelainen
1998; Melezhik et al. 1999a, 2005a, b). (b) Individual d13Chistograms for selected areas
1146 V.A. Melezhik et al.
References
Aharon P (2005) Redox stratification and anoxia of the early Precam-
brian oceans: implications for carbon isotope excursions and oxida-
tion events. Precambrian Res 137:207–222
Amelin YV, Heaman LM, Semenov VS (1995) U-Pb geochronology of
layered mafic intrusions in the eastern Baltic Shield; implications
for the timing and duration of Paleoproterozoic continental rifting.
Precambrian Res 75:31–46
Baker AJ, Fallick AE (1989a) Heavy carbon in two-billion-year-old
marbles from Lofoten-Vesteralen, Norway: implications for the
Precambrian carbon cycle. Geochim Cosmochim Acta
53:1111–1115
Baker AJ, Fallick AE (1989b) Evidence from Lewisian limestones for
isotopically heavy carbon in two thousand million year old sea
water. Nature 337:352–354
Barley ME, Pickard AL, Sylvester PJ (1997) Emplacement of a large
igneous province as a possible cause of banded iron formation 2.45
billion years ago. Nature 385:55–58
Bau M, Romer Rolf L, Lueders V, Beukes Nicolas J (1999) Pb, O, and
C isotopes in silicified Mooidraai Dolomite (Transvaal Supergroup,
South Africa); implications for the composition of Paleoproterozoic
seawater and “dating” the increase of oxygen in the Precambrian
atmosphere. Earth Planet Sci Lett 174:43–57
Bekker A, Kaufman AJ, Karhu JA, Beukes NJ, Swart QD, Coetzee LL,
Eriksson KA (2001) Chemostratigraphy of the Paleoproterozoic
Duitschland Formation, South Africa: implications for coupled
climate change and carbon cycling. Am J Sci 301:261–285
Bekker A, Karhu JA, Eriksson KA, Kaufman AJ (2003a) Chemostra-
tigraphy of Paleoproterozoic carbonate successions of theWyoming
Craton: tectonic forcing of biogeochemical change? Precambrian
Res 120:279–325
Bekker A, Sial AN, Karhu JA, Ferreira VP, Noce CM, Kaufman AJ,
Romano AW, Pimentel MM (2003b) Chemostratigraphy of
carbonates from the Minas Supergroup, Quadrilatero ferrıfero
(Iron Quadrangle), Brazil: a stratigraphic record of Early Protero-
zoic atmospheric, biogeochemical and climatic change. Am J Sci
303:865–904
Bekker A, Holmden C, Patterson W, Coetzee LL, Beukes NJ (2004)
Chemostratigraphy of early Paleoproterozoic carbonates of South
Africa, GSA Abstracts with Program 36: 341
Bekker A, Karhu JA, Kaufman AJ (2006) Carbon isotope record for the
onset of the Lomagundi carbon isotope excursion in the Great Lakes
area, North America. Precambrian Res 148:145–180
Bekker A, Holdmen C, Beukes NJ, Kenig F, Eglinton B, Patterson WP
(2008) Fractionation between inorganic and organic carbon during
the Lomagundi (2.22–2.1 Ga) carbon isotope excursion. Earth
Planet Sci Lett 271:278–291
Brasier AT, Fallick AE, Prave AR, Melezhik VA, Lepland A (2011)
Coastal sabkha dolomites and calcitised sulphates preserving the
Lomagundi-Jatuli carbon isotope signal. Precambrian Res
189:193–211
Bristow TF, Kennedy MJ (2008) Carbon isotope excursions and the
oxidant budget of the Ediacaran atmosphere and ocean. Geology
36:863–866
Broecker WS (1970) A boundary condition on the evolution of atmo-
spheric oxygen. J Geophys Res 75:3553–3557
Buchan KL, Halls HC, Mortensen JK (1996) Paleomagnetism, U-Pb
geochronology, and geochemistry of Marathon Dykes, Superior
Province, and comparison with the Fort Frances swarm. Can
J Earth Sci 33:1583–1595
Buick IS, Uken R, Gibson RL, Walmach T (1998) High-d13CPalaeoproterozoic carbonates from Transvaal Supergroup, South
Africa. Geology 26:875–878
Clark T (1984) Geologie de la Region du lac Cambrien. Territoire du
Nouveau-Quebec, ET 83–02:37
Condie KC, Des Marais DJ, Abbot D (2001) Precambrian superplumes
and supercontinents: a record of black shales, carbon isotopes, and
paleoclimates. Precambrian Res 106:239–260
Condie KC, O’Neill C, Aster RC (2009) Evidence and implications for
a widespread magmatic shutdown for 250My on Earth. Earth Planet
Sci Lett 282:294–298
Condon DJ, Bowring SA (2011) A user’s guide to Neoproterozoic
geochronology. The geological record of Neoproterozoic
glaciations. In: Arnaud E, Shields G, Halverson G (eds) Geological
Society, London, Memoir 36:135–149
Corfu F, Andrews AJ (1986) A U-Pb age for mineralized Nipissing
Diabase, Gowganda, Ontario. Can J Earth Sci 23:107–109
Deines P (2002) The carbon isotope geochemistry of mantle xenoliths.
Earth Sci Rev 58:247–278
Des Marais DJ, Strauss H, Summons RE, Hayes JM (1992) Carbon
isotope evidence for the stepwise oxidation of the Proterozoic
environment. Nature 359:605–609
Dickens GR (1999) Carbon cycle – the blast in the past. Nature
401:752–753
Fallick AE, Melezhik VA, Simonson B (2008) The ancient anoxic
biosphere was not as we know it. In: Dobretsov N, Kolchanov N,
Rozanov A, Zavarzin G (eds) Biosphere origin and evolution.
Springer, New York, pp 169–188
Fallick A, Melezhik VA, Simonson B (2011) On Proterozoic
ecosystems and the carbon isotopic composition of carbonates
associated with Banded Iron Formations. In: Neves L et al (eds)
Modelacao de Sistemas Geologicos. Univeridade de Coimbra,
Coimbra, pp 57–71
Findlay RM, Parrish RR, Birkett TC, Watanabe DH (1995) U-Pb ages
from the Nimish Formation and Montagnais glomeroporphyritic
gabbro of the central New Quebec Orogen, Canada. Can J Earth
Sci 32:1208–1220
Fletcher IR, Rasmussen B, McNaughton NJ (2000) SHRIMP U–Pb
geochronology of authigenic xenotime and its potential for dating
sedimentary basins. Aust J Earth Sci 47:845–859
Frauenstein F, Veizer J, Beukes N, Van Niekerk HS, Coetzee LL (2009)
Transvaal Supergroup carbonates: implications for Paleoproterozoic
d18O and d13C records. Precambrian Res 175:149–160
Galimov IM, Kuznetsova NG, Prokhorov VS (1968) The problem of
the composition of the Earth’s ancient atmosphere in connection
with results of isotopic analyses of carbon from Precambrian
carbonates. Geochemistry 11:1376–1381 (in Russian)
Gancarz AJ (1978) U-Pb age (2.05 � 109 years) of the Oklo uranium
deposit. In: Proceedings of the natural fission reactors. Annual
international atomic energy agency conference, Vienna,
pp 513–520
Gerlach TM (1991) Present-day carbon dioxide emissions from
volcanos. Earth Space 4:5–14
Glagolev AA, Kazansky VI, Prokhorov KV, Rusinov VL, Maslennikov
VA, Voronovsky SN, Ovchinnikov LN (1987) Zonality and age of
metamorphism. In: Kozlowsky YeA (ed) The superdeep well of the
Kola Peninsula. Springer, Berlin, pp 166–198
Goldich SS, Fischer LB (1986) Air-abrasion experiments in U-Pb
dating of zircon. Chem Geol 58:195–215
Halilovic J, Cawood PA, Jones JA, Pirajno F, Nemchin AA (2004)
Provenance of the Earaheedy Basin; implications for assembly of
the Western Australian Craton. Precambrian Res 128:343–366
Hannah JL, Bekker A, Stein HJ, Markey RJ, Holland HD (2004)
Primitive Os and 2316 Ma age for marine shale; implications for
Paleoproterozoic glacial events and the rise of atmospheric oxygen.
Earth Planet Sci Lett 225:43–52
Hanski EJ, Huhma H, Smol’kin VF, Vaasjoki M (1990) The age of
ferropicritic volcanites and comagmatic Ni-bearing intrusions at
3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1147
Pechenga, Kola Peninsula, U.S.S.R. Geol Surv Finl Bull
62:123–133
Hayes JM,Waldbauer JR (2006) The carbon cycle and associated redox
processes through time. Philos Trans R Soc B 361:931–950
Hofmann HJ, Davidson A (1998) Paleoproterozoic stromatolites,
Hurwitz Group, Quartzite Lake area, Northwest Territories,
Canada. Can J Earth Sci 35:280–289
Holser WT, Schidlowski M, Mackenzie FT, Maynard JB (1988) Geo-
chemical cycles of carbon and sulphur. In: Gregor CB, Garrels RM,
Mackenzie FT, Maynard JB (eds) Chemical cycles in the evolution
of the Earth. Wiley, New York, pp 106–173
Horie K, Hidaka H, Gauthier-Lafaye F (2005) U-Pb geochronology and
geochemistry of zircon from the Franceville series at Bidoudouma,
Gabon, Goldschmidt conference abstract, Geochim Cosmochim
Acta 69: p A11
House CH, Schopf JW, Stetter KO (2003) Carbon isotopic fractionation
by Archaeans and other thermophilic prokaryotes. Org Geochem
34:345–357
Huhma H, Cliff RA, Perttunen V, Sakko M (1990) Sm-Nd and Pb
isotopic study of mafic rocks associated with early Proterozoic
continental rifting: the Per€apohja schist belt in northern Finland.
Contrib Mineral Petrol 104:369–379
Karhu JA (1993) Palaeoproterozoic evolution of the carbon isotope
ratios of sedimentary carbonates in the Fennoscandian Shield.
Geol Surv Finl Bull 371:1–87
Karhu JA (2005) Paleoproterozoic carbon isotope excursion. In:
Lehtinen M, Nurmi PA, R€am€o OT (eds) Precambrian geology of
Finland – key to the evolution of the Fennoscandian Shield.
Elsevier, Amsterdam, pp 669–680
Karhu JA, Holland HD (1996) Carbon isotopes and the rise of atmo-
spheric oxygen. Geology 24:867–879
Karhu J, KortelainenNM,HuhmaH, PerttunenV, Sergeev SS (2008) The
end of the Paleoproterozoic carbon isotope excursion: new time
constraints, Abstract, 33rd geological congress, Oslo, 6–14 Aug 2008
Keith ML (1982) Violent volcanism, stagnant oceans and some
inferences regarding petroleum, strata-bound ores and mass extinc-
tion. Geochim Cosmochim Acta 46:2621–2637
Kirschvink JL, Nash CZ, Raub TD, Raub TMD, Kopp RD, Hilburn IA
(2009) The Lomagundi-Jatuli event, Lois Pasteur, and the geologi-
cal record of aerobic respiration, Goldschmidt Conference Abstract,
Davos, Geochim Cosmochim Acta 73: p A622
Koistinen T, Stephens MB, Bogatchev V, Nordgulen Ø, Wennestr€omM,
Korhonen J (Comps.) (2001) Geological map of the Fennoscandian
Shield, Scale 1:2 000 000, Espoo/Trondheim/Upsala/Moscow
Kortelainen N (1998) Depositional environment, carbon isotope com-
position and trace element geochemistry of Palaeoproterozoic car-
bonate rocks in the Per€apohja Belt. Master’s thesis, Department of
Geology, University of Helsinki, p 97 (in Finnish)
Krogh TE, Davis DW, Corfu F (1984) Precise U-Pb zircon and
baddeleyite ages for the Sudbury area. In: Pye GE, Naldrett AJ,
Giblin PE (eds) The geology and ore deposits of the Sudbury
Structure, Ont Geol Surv, Spec Vol 1:431–446
Kump LR, Junium Ch, Arthur MA, Brasier A, Fallick AE, Melezhik
VA, Lepland A, Crne AE, Luo G (2011) Isotopic evidence for
massive oxidation of organic matter following the great oxidation
event. Science 334:1694–1696
Kuznetsov AB, Melezhik VA, Gorokhov IM, Melnikov NN,
Konstantinova GV, Kutyavin EP, Turchenko TL (2010) Sr isotopic
composition of Paleoproterozoic 13C-rich carbonate rocks: the
Tulomozero Formation, SE Fennoscandian Shield. Precambrian
Res 182:300–312
Lager I, Loberg B (1990) Sedimentological and basin analysis ore
prospecting methods in the North Bothnian greenstone belt, STU-
Projekt 86-03967P, Final report, Economic Geology, Lulea Techni-
cal University, Lulea, p 112 (in Swedish)
Lindsay JF, Brasier MD (2002) Did global tectonics drive early bio-
sphere evolution? Carbon isotope record from 2.6 to 1.9 Ga
carbonates of Western Australian basins. Precambrian Res
114:1–34
Machado N, Noce CM, Ladeira Eduardo A, Belo de Oliveira O (1992)
U-Pb geochronology of Archean magmatism and Proterozoic meta-
morphism in the Quadrilatero Ferrifero, southern Sao Francisco
Craton, Brazil. Geol Soc Am Bull 104:1221–1227
Machado N, Schrank A, Noce CM, Gauthier G (1996) Ages of detrital
zircon from Archean-Paleoproterozoic sequences; implications for
greenstone belt setting and evolution of a Transamazonian foreland
basin in Quadrilatero Ferrifero, Southeast Brazil. Earth Planet Sci
Lett 141:259–276
Maheshwari A, Sial AN, Chittora VK (1999) High-d13C Paleopro-
terozoic carbonates from the Aravalli Supergroup, Western India.
Int Geol Rev 41:949–954
Maheshwari A, Sial AN, Gaucher C, Bossi J, Bekker A, Ferreira VP,
Romano AW (2010) Global nature of the Paleoproterozoic
Lomagundi carbon isotopic excursion: a review of occurrences in
Brazil, India and Uruguay. Precambrian Res 182:274–299
Martin DM, Clendenin CW, Krapez B, McNaughton NJ (1998) Tec-
tonic and geochronological constraints on late Archaean and
Palaeoproterozoic stratigraphic correlation within and between the
Kaapvaal and Pilbara cratons. J Geol Soc Lond 2:311–322
Martin AP, Condon DJ, Prave AR, Melezhik VA, Fallick A (2010)
Constraining the termination of the Lomagundi-Jatuli positive iso-
tope excursion in the Imandra-Varzuga segment (Kola Peninsula,
Russia) of the North Transfennoscandian Greenstone Belt by high-
precision ID-TIMS. AGU fall meeting, Abstracts, San Francisco,
13–17 Dec
Master S, Bekker A, Hofmann A (2010) A review of the stratigraphy
and geological setting of the Paleoproterozoic Magondi Super-
group, Zimbabwe – type locality for the Lomagundi carbon isotope
excursion. Precambrian Res 182:254–273
Mattey DP (1987) Carbon isotopes in the mantle. Terra Cognita
7:31–37
McNaughton NJ, Wilson AF (1983) 13C-rich marbles from the Prote-
rozoic Einasleigh Metamorphics, northern Queensland. J Geol Soc
Aust 30:175–178
McNaughton NJ, Rasmussen B, Fletcher IR (1999) SHRIMP uranium-
lead dating of diagenetic xenotime in siliciclastic sedimentary
rocks. Science 2:78–80
Melezhik VA (1992) Early Proterozoic sedimentary and rock-forming
basins of the Baltic Shield. Nauka (Science), St. Petersburg, p 258
(in Russian)
Melezhik VA, Fallick AE (1996) A widespread positive d13Ccarb anom-
aly at around 2.33–2.06 Ga on the Fennoscandian Shield: a para-
dox? Terra Nova 8:141–157
Melezhik VA, Fallick AE (1997) A widespread positive d13C anomaly
at around 2.33–2.06 Ga on the Fennoscandian Shield – reply. Terra
Nova 9:148–151
Melezhik VA, Fallick AE (2001) Palaeoproterozoic travertines of vol-
canic affiliation from a 13C-rich rift lake environment. Chem Geol
173:293–312
Melezhik VA, Fallick AE (2003) d13C and d18O variations in primary
and secondary carbonate phases: several contrasting examples from
Palaeoproterozoic 13C-rich metamorphosed dolostones. Chem Geol
201:213–228
Melezhik VA, Fallick AE (2005) The Palaeoproterozoic, rift-related,
shallow-water, 13C-rich, lacustrine carbonates, NW Russia – Part I:
sedimentology and major element geochemistry. Trans R Soc Edinb
Earth Sci 95:393–421
Melezhik VA, Fallick AE (2010) On the Lomagundi-Jatuli carbon
isotopic event: the evidence from the Kalix Greenstone Belt,
Sweden. Precambrian Res 179:165–190
1148 V.A. Melezhik et al.
Melezhik VA, Predovsky AA (1982) Geochemistry of early Protero-
zoic lithogenesis. Nauka (Science), Leningrad, p 208 (in Russian)
Melezhik VA, Fallick AE, Clark T (1997) Two billion year old isoto-
pically heavy carbon: evidence from the Labrador Trough, Canada.
Can J Earth Sci 34:271–285
Melezhik VA, Fallick AE, Medvedev PV, Makarikhin VV (1999a)
Extreme 13Ccarb enrichment in ca. 2.0 Ga magnesite-stromatolite-
dolomite-‘red beds’ association in a global context: a case for the
world-wide signal enhanced by a local environment. Earth Sci Rev
48:71–120
Melezhik VA, Fallick AE, Filippov MM, Larsen O (1999b) Karelian
shungite – an indication of 2000 Ma-year-old metamorphosed oil-
shale and generation of petroleum: geology, lithology and geo-
chemistry. Earth Sci Rev 47:11–40
Melezhik VA, Fallick AE, Medvedev PV, Makarikhin VV (2000)
Palaeoproterozoic magnesite–stromatolite–dolostone–‘red bed’
association, Russian Karelia: palaeoenvironmental constraints on
the 2.0 Ga positive carbon isotope shift. Norsk Geologisk Tidsskrift
80:163–186
Melezhik VA, Fallick AE, Medvedev PV, Makarikhin V (2001)
Palaeoproterozoic magnesite: lithological and isotopic evidence
for playa/sabkha environments. Sedimentology 48:379–397
Melezhik VA, Fallick AE, Smirnov YuP, Yakovlev YuN (2003) Frac-
tionation of carbon and oxygen isotopes in 13C-rich Palaeopro-
terozoic dolostones in the transition from medium-grade to high-
grade greenschist facies: a case study from the Kola Superdeep
Drillhole. J Geol Soc Lond 160:71–82
Melezhik VA, Fallick AE, Grillo SM (2004) Subaerial exposure
surfaces in a Palaeoproterozoic 13C-rich dolostone sequence from
the Pechenga Greenstone Belt: palaeoenvironmental and isotopic
implications for the 2330–2060 Ma global isotope excursion of13C/12C. Precambrian Res 133:75–103
Melezhik VA, Fallick AE, Hanski E, Kump L, Lepland A, Prave A,
Strauss H (2005a) Emergence of the aerobic biosphere during the
Archean-Proterozoic transition: challenges for future research. Geol
Soc Am Today 15:4–11
Melezhik VA, Fallick AE, Kuznetsov AB (2005b) The Palaeopro-
terozoic, rift-related, shallow-water, 13C-rich, lacustrine carbonates,
NW Russia – Part II: global isotope signal recorded in the lacustrine
dolostones. Trans R Soc Edinb Earth Sci 95:423–444
Melezhik VA, Fallick AE, Rychanchik DV, Kuznetsov AB (2005c)
Palaeoproterozoic evaporites in Fennoscandia: implications for sea-
water sulphate, d13C excursions and the rise of atmospheric oxygen.
Terra Nova 17:141–148
Melezhik VA, Huhma H, Condon DJ, Fallick AE, Whitehouse MJ
(2007) Temporal constraints on the Paleoproterozoic Lomagundi-
Jatuli carbon isotopic event. Geology 35:655–658
Morozov AF, Hakhaev BN, Petrov OV, Gorbachev VI, Tarkhanov GB,
Tsvetkov LD, Erinchek YuM, Akhmedov AM, Krupenik VA,
Sveshnikova KYu (2010) Rock-salts in Palaeoproterozoic strata of
the Onega depression of Karelia (based on data from the
Onega parametric drillhole). Trans Acad Sci 435(2):230–233 (in
Russian)
Mortensen JK, Percival JA (1987) Reconnaissance U-Pb zircon and
monazite geochronology of the Lac Clairambault area, Ashuanipi
Complex, Quebec. Geol Surv Can Pap 87–2:135–142
M€ueller SG, Krapez B, Barley ME, Fletcher IR (2005) Giant iron-ore
deposits of the Hamersley Province related to the breakup of
Paleoproterozoic Australia; new insights from in situ SHRIMP
dating of baddeleyite from mafic intrusions. Geology 33:577–580
Ohlander B, Lager I, Loberg BEH, Schoberg H (1992) Stratigraphical
position and Pb-Pb age of lower Proterozoic carbonate rocks from
the Kalix greenstone belt, northern Sweden. Geologiska F€oreningeni Stockholm F€orhandlingar 114:317–322
Papineau D, Mojzsis SJ, Coath CD, Karhu JA, McKeegan KD (2005)
Multiple sulphur isotopes of sulfides from sediments in the
aftermath of Paleoproterozoic glaciations. Geochim Cosmochim
Acta 69:5033–5060
Pekkarinen L, Lukkarinen H (1991) Paleoproterozoic volcanism in the
Kiihtelysvaara-Tohmaj€arvi district, eastern Finland. Geol Surv Finl
Bull 357:1–30
Perttunen V, Vaasjoki M (2001) U-Pb geochronology of the Per€apohjaSchist Belt, northwestern Finland. Geol Surv Finl Spec Pap
33:45–84
Petrov VP, Voloshina ZM (1982) Metamorphism. In: Gorbunov GI (ed)
The Imandra/Varzuga zone of the Karelides. Nauka (Science),
Leningrad, pp 192–212 (in Russian)
Petrov VP, Voloshina ZM (1995) Regional metamorphism of the
Pechenga area rocks. In: Mitrofanov FP, Smolkin VF (eds)
Magmatism, sedimentogenesis and gedodynamics of the Pechenga
Palaeorift. Kola Science Centre, Apatity, pp 164–182 (in Russian)
Pickard AL (2002) SHRIMP U-Pb zircon ages of tuffaceous mudrocks
in the Brockman iron formation of the Hamersley range, Western
Australia. Aust J Earth Sci 49:491–507
Pickard AL (2003) U-Pb SHRIMP U-Pb zircon ages for the Palaeopro-
terozoic Kuruman iron formation, Northern Cape Province, South
Africa; evidence for simultaneous BIF deposition on Kaapvaal and
Pilbara cratons. Precambrian Res 125:275–315
Pokrovsky B, Melezhik VA (1995) Variations of carbon and oxygen
isotope composition in Palaeoproterozoic carbonate rocks of the
Kola Peninsula. Stratigr Geol Correl 3:45–53
Puchtel LS, Arndt NT, Hofmann AW, Haase KM, Kroener A, Kulikov
VS, Kulikova VV, Garbe-Schoenberg CD, Nemchin AA (1998)
Petrology of mafic lavas within the Onega Plateau, central Karelia;
evidence for 2.0 Ga plume-related continental crustal growth in the
Baltic Shield. Contrib Mineral Petrol 130:134–153
Purohit R, Sanyal P, Roy AB, Bhattacharya SK (2010) 13C enrichment
in the Palaeoproterozoic carbonate rocks of the Aravalli Super-
group, northwest India: influence of depositional environment.
Gondwana Res 18:538–546
Rasmussen B, Fletcher IR (2002) Indirect dating of mafic intrusions by
SHRIMP U-Pb analysis of monazite in contact metamorphosed
shale; an example from the Palaeoproterozoic Capricorn Orogen,
Western Australia. Earth Planet Sci Lett 197:287–299
Rohon ML, Vialette Y, Clark T, Roger G, Ohnenstetter D, Vidal P
(1993) Aphebian mafic-ultramafic magmatism in the Labrador
Trough (New Quebec); its age and the nature of its mantle source.
Can J Earth Sci 30:1582–1593
Rothman DH, Hayes JM, Summons RE (2003) Dynamics of the
Neoproterozoic carbon cycle. Proc Natl Acad Sci 100:8124–8129
Salop LI (1982) Geological evolution of the Earth in the Precambrian.
Nedra, Leningrad, p 343 (in Russian)
Schidlowski M (1988) A 3,800-million-year isotopic record of life from
carbon in sedimentary rocks. Nature 333:313–318
Schidlowski M, Eichmann R, Junge CE (1975) Precambrian sedimen-
tary carbonates: carbon and oxygen isotope geochemistry and
implications for the terrestrial oxygen budget. Precambrian Res
2:1–69
Schidlowski M, Eichmann R, Junge CE (1976) Carbon isotope geo-
chemistry of the Precambrian Lomagundi carbonate province,
Rhodesia. Geochim Cosmochim Acta 40:449–455
Schidlowski M, Hayes JM, Kaplan IR (1983) Isotopic inference of
ancient biochemistries: carbon, sulphur, hydrogen and nitrogen.
In: Schopf JW (ed) Earth’s earliest biosphere: its origin and evolu-
tion. University Press, Princeton, pp 149–186
Scoates JS, Friedman RM (2008) Precise age of the platiniferous
Merensky Reef, Bushveld Complex, South Africa, by the U-Pb
zircon chemical abrasion U-Pb ID-TIMS technique. Econ Geol
Bull Soc Econ Geologists 103:465–471
Shields G (1997) A widespread positive d13C anomaly at around
2.33–2.06 Ga on the Fennoscandian Shield – Comment. Terra
Nova 9:148
3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1149
Silvennoinen A (1991) Pre-Quaternary rocks of the Kuusamo and
Rukatunturi map-sheet areas. Explanation to the maps of Pre-
quaternary rocks, sheets 4524 + 4542. Geological map of Finland
1:100,000, Geological Survey of Finland, Espoo, p 63 (in Finnish
with English summary)
Ski€old T (1987) Aspects of the Proterozoic geochronology of northern
Sweden. Precambrian Res 35:161–167
Strauss H, Des Marais DJ, Hayes JM, Summons RE (1992)
Concentrations of organic carbon and maturities and elemental
composition of kerogen. In: Schopf WJ, Klein C (eds) The Protero-
zoic biosphere: a multidisciplinary study. Cambridge University
Press, New York, pp 95–99
Summons RE, Hayes JM (1992) Principles of molecular and isotopic
biogeochemistry. In: Schopf JW, Klein C (eds) The Proterozoic
biosphere. Cambridge University Press, Cambridge, pp 83–93
Sundquist ET (1993) The global carbon dioxide budget. Science
259:934–941
Tikhomirova M, Makarikhin VV (1993) Possible reasons for the d13Canomaly of lower Proterozoic sedimentary carbonates. Terra Res
5:244–248
Trendall AF, de Laeter JR, Nelson DR, Hassler SW (1998) Precise
zircon U-Pb ages from the Marra Mamba Iron formation and
Wittenoom formation, Hamersley Group, Western Australia. Aust
J Earth Sci 45:137–142
Vallini D, Rasmussen B, Krapez B, Fletcher IR, McNaughton NJ
(2002) Obtaining diagenetic ages from metamorphosed sedimen-
tary rocks: U-Pb dating of unusually coarse xenotime cement in
phosphatic sandstone. Geology 30:1083–1086
Vallini DA, Cannon WF, Schulz KJ (2006) Age constraints for
Paleoproterozoic glaciation in the Lake Superior region; detri-
tal zircon and hydrothermal xenotime ages for the Chocolay
Group, Marquette Range Supergroup. Can J Earth Sci
43:571–591
Vasileva IM, Ovchinnikova GB, Melezhik VA, Gorokhov IM,
Kusnetsov AB, Gorokhovsky BM (2000) Pb-Pb isotopic age of
the Tulomozero dolostones, N. Lake Onega area, All-Russian
conference. General issues of Precambrian subdivision, Abstract
Volume, Kola Science Centre, Apatity, pp 53–54
Volodichev OI (1987) Metamorphism. In: Sokolov VA (ed) Geology of
Karelia. Nauka (Science), Leningrad, pp 152–175 (in Russian)
Wanke A, Melezhik VA (2005) Palaeoproterozoic sedimentation and
stromatolite growth in an advanced intracontinental rift associated
with the marine realm: a record of the Neoarchaean continent
breakup? Precambrian Res 140:1–35
Watson RS, Trewin NH, Fallick AE (1995) The formation of carbonate
cement in the Forth and Balmoral Fields, northern North Sea: a case
for biodegradation, carbonate cementation and oil leakage during
early burial. In: Harley AJ, Prosser DJ (eds) Characterization of
deep marine clastic systems, vol 94, Geological Society Special
Publication, pp 177–200
Wilson MR, Sehlstedt S, Claesson L, Smellie JAT, Aftalion M,
Hamilton PJ, Fallick AE (1987) J€orn: an early Proterozoic intrusivecomplex in a volcanic-arc environment, north Sweden. Precambrian
Res 36:201–225
Wilson JP, Fischer WW, Johnston DT, Knoll AH, Grotzinger JP, Walter
MR, McNaughton NJ, SimonM, Abelson J, Schrag DP, Summons R,
Allwood A, Andres M, Gammon C, Garvin J, Rashby S, Schweizer
M,WattersWA (2010) Geobiology of the late Paleoproterozoic Duck
Creek formation, Western Australia. Precambrian Res 179:135–149
Yudovich YE, Makarikhin VV, Medvedev PV, Sukhanov NV (1991)
Carbon isotope anomalies in carbonates of the Karelian Complex.
Geochem Int 28:56–62
Zagnitko VN, Lugovaja IP (1989) Isotope geochemistry of carbonate
and ferrous-siliceous rocks of the Ukraine Shield, Naukova Dumka,
Kiev, p 316 (in Russian)
Zlobin VL (1993) Geology of carbonate-gneiss complex of the Anabar
Shield. Ph.D. thesis, Institute of the Lithosphere, Russian Academy
of Sciences, Moscow, p 26 (in Russian)
1150 V.A. Melezhik et al.
7.4 An Apparent Oxidation of the Upper Mantleversus Regional Deep Oxidation of TerrestrialSurfaces in the Fennoscandian Shield
Kyle S. Rybacki, Lee R. Kump, Eero J. Hanski, and Victor A. Melezhik
7.4.1 Introduction
Why the Fennoscandian Shield?
Part of the Palaeoproterozoic Karelian igneous and sedimen-
tary rocks of the Fennoscandian Shield were erupted and
deposited during the “Great Oxidation Event” (GOE). The
drillcores collected for the Fennoscandia Arctic Russia –
Drilling Early Earth Project (FAR-DEEP) allow detailed
geological and geochemical sampling through this very
dynamic time in geologic history. One of the unusual
characteristics of the Palaeoproterozoic volcanic rocks in
the eastern part of the Fennoscandian Shield is the presence
of highly oxidised lava flows (Fig. 7.45), suggestive of a link
to the GOE, either cause or effect. The most oxidised volca-
nic rocks are found in the Jatulian system deposited within
the time interval of 2.3–2.06 Ga (see Fig. 7.46). The age and
sampling structure of the FAR-DEEP cores permit the test-
ing and assessment of two competing hypotheses for the
origin of the highly oxidised volcanic rocks of the
Fennoscandian Shield: an apparent increase in the oxidation
state of the upper mantle from which the lavas were erupted,
or subsequent deep oxidative weathering of the lavas as a
result of the GOE, or the combined effect of both. The rocks
sampled by the FAR-DEEP cores allow the comparison of
primary and secondary mineralogical and diagenetic details,
which may not be present in outcrop. In addition to
investigating the origin of the highly oxidised rocks, other
questions can be addressed because of the exquisite preser-
vation of the rocks sampled in the FAR-DEEP cores. Specif-
ically, are there discernable physical and/or chemical
differences in weathering profiles developed on lava flows
before and after the GOE, and can palaeo-water tables in the
shield be identified through the use of redox proxies? The
FAR-DEEP cores also sample igneous rocks erupted during
the proposed magmatic activity shutdown/slowdown
between 2.45 and 2.2 Ga (Condie et al. 2009). Overall, the
FAR-DEEP cores are conducive to detailed geochemical
analysis and potential insight into a poorly understood inter-
val in Earth’s history.
The “Great Oxidation Event”
The approximate date of 2.45 Ga for the GOE is associated
with the disappearance of mass-independent fractionation of
sulphur isotopes (MIF-S; Bekker et al. 2004; Farquhar et al.
2000, 2010; Guo et al. 2009), the close association of
diamictites and redbeds (Evans et al. 1997), as well as the
positive isotopic compositions of chromium and iron within
banded iron formations (BIFs; Frei et al. 2009). When oxy-
gen concentrations become sufficiently high in the atmo-
sphere, the reduced sulphur compounds generated by
photochemical reactions in the atmosphere and fractionated
mass-independently become re-oxidised and homogenised
isotopically with oxidised sulphur, and then deposited at the
surface (Farquhar et al. 2010). The termination of MIF-S in
the geologic rock record occurred at approximately 2.45 Ga
(Bekker et al. 2004; Kasting 2006). Recently, this interpre-
tation was challenged by Watanabe et al. (2009) who
demonstrated that modest MIF-S can be the result of pro-
cesses other than photochemistry, namely from hydrother-
mal interactions between sulphur and organic matter;
multiple such fractionations could generate substantial MIF
if the fractionated products were isolated. With that caveat,
in this chapter we assume that Earth’s atmosphere was
primarily anoxic prior to 2.45 Ga, and oxic thereafter.
The atmospheric concentration of oxygen during, and
subsequent to, the GOE is poorly understood. Models
using Fe- and Mg-retention in palaeosols have been
implemented to better quantify atmospheric oxygen
concentrations (Murakami et al. 2011). Iron and magnesium
K.S. Rybacki (*)
Department of Geosciences, The Pennsylvania State University,
University Park, PA 16802, USA
e-mail: [email protected]
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_4, # Springer-Verlag Berlin Heidelberg 2013
1151
retention, with respect to immobile elements Ti, Al, and Zr,
increased between approximately 2.5 and 2.1 Ga. By com-
bining Fe2+ oxidation kinetics with Fe retention, Murakami
et al. (2011) calculated that pO2 increased linearly on a
logarithmic scale between 2.5 and 2.0 Ga from less than
10�6 atm to greater than 10�3 atm. Note that a linear
increase on a logarithmic scale is an exponential increase,
so Murakami et al. (2011) mischaracterise this as a gradual
increase. Murakami et al. (2011) also modeled the effects
that significant swings in atmospheric temperature, from 0�Cto 40�C, would have on pO2 levels during the Palaeopro-
terozoic. These models exhibited rapid oscillations in pO2
between 2.4 and 2.3 Ga. Quantifying Palaeoproterozoic
atmospheric oxygen levels is very important; however,
defining what caused the rise of oxygen in the first place is
currently more pressing.
Hypotheses for the “Great Oxidation Event”
The rise of atmospheric oxygen between 2.45 and 2.32 Ga
(Bekker et al. 2004) was a major turning point in Earth’s
history. Today, it is not clear what caused the GOE (Canfield
2005; Catling and Claire 2005; Farquhar et al. 2010; Holland
2009). Was it the evolution and proliferation of oxygen-
producing organisms, a decrease in oxygen sinks, or a com-
bination of the two?
Increase in O2 Production
One hypothesis proposed for the rise of oxygen is an
increase in O2 production. Only one major source is known
to exist for oxygen, oxygenic photosynthesis. Kopp et al.
(2005) propose that cyanobacterial evolution coincided with
the GOE at approximately 2.45 Ga. They propose that
increased terrestrial weathering rates associated with the
Huronian and Makganyene glaciations would increase the
flux of phosphorous into the ocean. In the quiescent period
between glaciations, the additional flux of phosphorous
would stimulate cyanobacteria development, which would
increase oxygen production and eventually lead to the
oxygenation of Earth’s atmosphere.
Evidence for the oldest stromatolites is found in the
3.49 Ga Dresser Formation in northwestern Australia, and
the oldest hydrocarbon fluid inclusions are hosted in rocks
approximately 3.23 Ga old; however, their structure is not
conclusively indicative of photosynthesis (Buick 2008). The
earliest known molecular cyanobacteria biomarkers, i.e.
organic compounds with a specific structure that can be
related to a particular source organism, and fossil
morphologies of photosynthetic organisms are thought to
be present in rocks approximately 2.7 Ga old (Battistuzzi
et al. 2004; Buick 1992, 2008; Kasting 2008), but a post
2.7 Ga origin for the biomarkers cannot be disregarded
(Rasmussen et al. 2008). The liquid oil present within the
Matinenda fluid inclusions from the Huronian Supergroup in
Canada, trapped before peak metamorphism at ca. 2.2 Ga,
contain steranes and 2a-methylhopanes, which may be
linked to photosynthetic cyanobacteria (Buick 2008). The
synthesis of sterols (the sterane precursors) in existing
organisms requires the presence of free O2 (Summons et al.
2006), and 2a-methylhopanes are primarily limited to
cyanobacteria and therefore used as an oxygenic photosyn-
thesis biomarker (Summons et al. 1999). One question still
remains: if photosynthesising organisms evolved approxi-
mately 2.7 Ga ago, then why is there a lag in the MIF-S
record of approximately 0.25 Ga (Kopp et al. 2005)? In
general, the lag between the evolution of photosynthetic
organisms and the rise of atmospheric oxygen is accredited
to overcoming oxygen sinks such as the oxidation of organic
carbon and reduced minerals.
Decrease in O2 Sinks
Another hypothesis for the rise of oxygen is a permanent
decrease in the primary sink(s) for oxygen during the
Archaean–Proterozoic Transition (Holland 1984; Kasting
et al. 1993; Kump and Barley 2007; Kump 2008). Major
sinks for O2 include reduced volcanic gases (e.g. H2, CH4,
CO, and H2S), reduced metamorphic fluids and gases, and
the weathering of continental landmasses where sedimentary
organic-carbon and reduced mineral species, such as pyrite
(FeS2), and other ferrous-silicate minerals are exposed
(Catling and Claire 2005; Farquhar et al. 2010). A change
in the composition of reduced volcanic and metamorphic
gases could greatly affect the balance of O2 production and
consumption. A compositional change in the volcanic gases
emanated from volcanic sources via the progressive oxida-
tion of the upper mantle might result in a significant decrease
in the size of the volcanic gas sink of O2 (Holland 1984;
Kasting et al. 1993; Kump et al. 2001; Kump and Barley
2007; Gaillard et al. 2011). This could be accomplished
through the gradual oxidation of the upper mantle (Kasting
et al. 1993; Kump et al. 2001) in conjunction with a shift
from dominantly submarine to subaerial volcanism (Kump
and Barley 2007). Quantitative evidence is presented in
Gaillard et al. (2011) in favour of a shift from submarine to
subaerial volcanism. This evidence includes the subsequent
change in the volcanic gas budget (as originally proposed by
Kump and Barley (2007)), as well as the validation of the
calculated sensitivity of volcanic fluid composition to tem-
perature and pressure of Li and Lee (2004). A decrease in
oxygen sinks, specifically an oxidised upper mantle, might
have effectively contributed to the rise of atmospheric
1152 K.S. Rybacki et al.
oxygen, and the oxidised subaerial volcanics observed
within the Fennoscandian Shield may provide verification
of this hypothesis.
Brief Overview of Mantle Redox and ItsEvolution
Today, the redox state of the mantle, as recorded by ferric/
ferrous iron ratios in mid-ocean ridge and plume-related
basaltic magmas or the oxygen fugacity of volcanic gases,
falls generally in the range of two log units below to one unit
above the fayalite-magnetite-quartz (FMQ �2 < mantle
redox state < FMQ þ1) oxygen buffer (e.g. Bezos and
Humler 2005; Christie et al. 1986; Oskarsson et al. 1994;
Symonds et al. 1994). In island arc environments, where
hydrated oceanic crust is subducted into the mantle, the
oxygen fugacity of the lavas may rise to a higher level than
in oceanic ridge and intra-plate settings, but still remains
usually less than 2–3 log units above the FMQ buffer (e.g.
Rowe et al. 2009).
Researchers of mantle redox evolution can, in the most
general sense, be divided into two schools of thought – those
who favour the gradational oxidation of the mantle through
time (e.g. Holland 1984; Kasting et al. 1993; Kump et al.
2001), and those who believe that the mantle redox state
evolved very early in Earth’s history and has essentially
remained unchanged ever since (e.g. Canil 2002; Delano
2001; Frost and McCammon 2008; Lee 2005; Li and Lee
2004). The differing opinions originate from the use and
interpretation of different palaeoredox proxies.
Gradual Oxidation of the Mantle
Early hypotheses pertaining to the oxygenation of the atmo-
sphere proposed that a change in the redox state of the
mantle, from reduced to oxidised, would decrease the volca-
nic gas sink for atmospheric oxygen thus allowing its grad-
ual build up (Holland 1984; Kasting et al. 1993). These
works eventually gave rise to the Kump et al. (2001) hypoth-
esis that the oxidation and subsequent subduction of
oxidised oceanic lithosphere could result in the oxidation
of the upper mantle and/or the development of slab
graveyards at the core-mantle boundary. This hypothesis
accounts for the decrease in oxygen sink necessary for the
GOE, and the lag time between the evolution of photosyn-
thetic organisms and the rise of atmospheric oxygen.
Oxidised crustal material has, theoretically, been avail-
able since the advent of plate tectonics (Kasting et al. 1993;
Kump et al. 2001), and the oxidation of the upper mantle
could be achieved most effectively through the subduction
of oxidised crustal material (e.g. subduction of BIFs,
hydrated crust, etc.), the oxidation of ferrous iron, or hydro-
gen loss to space (Fig. 7.47a, b; Kasting et al. 1993; Kump
et al. 2001). The hypothesis is that subduction of oxidised
crustal material, coupled with the subsequent loss of H2 to
space would reduce the overall free oxygen demand
(Fig. 7.47c). If this cycle were to be repeated continually,
and assuming the upper mantle is not allowed to mix with
the underlying, more reduced lower-mantle reservoir, then
over time the oxidation state of the upper mantle would
theoretically evolve to a more oxidised state.
In an oxygen-poor atmosphere crustal material can be
oxidised via serpentinisation (Fig. 7.47b; Kasting et al.
1993). In the following reaction related to serpentinisation
of olivine, ferrous iron is systematically oxidised to ferric
iron by water, producing magnetite, quartz, and hydrogen
that is subsequently lost to space:
3Fe2SiO4 Fayaliteð Þ þ 2H2O ! 2Fe3O4 Magnetiteð Þþ 3SiO2 Quartzð Þ þ 2H2
In the Precambrian, the lack of a well stratified atmo-
sphere would aid in the escape of hydrogen to space. Once
atmospheric oxygen levels reached approximately 1 % pres-
ent atmospheric level (PAL), then the rate of hydrogen
escape to space would have decreased rapidly due to the
accumulation of ozone (O3) within Earth’s upper atmo-
sphere. Ozone rapidly oxidises hydrogen forming water,
thus preventing its subsequent escape to space (Catling and
Claire 2005; Holland 1984; Kasting et al. 1993).
The assimilation of oxidised oceanic lithosphere is not a
process that occurs solely within the upper mantle
(Fig. 7.47c). Cold, dense subducted oceanic lithosphere
may sink all the way to the core-mantle boundary with
penetration precluded due to the strong density contrast
between the outer core and lower mantle (Fig. 7.47d).
Through time, the accumulation of oceanic slabs at the
core-mantle boundary would result in a “slab graveyard”
(Kellogg et al. 1999; Van der Hilst and Karason 1999).
This material will eventually be re-assimilated due to the
increased temperatures and pressures at the core-mantle
boundary. In the absence of strong thermal convection
cells, a slab graveyard would produce a pocket of oxidised
magma relative to the surrounding mantle. This oxidised
material can be expedited to Earth’s surface via a mantle
plume for eruption (i.e. the “upside-down” Archaean mantle
of Kump et al. 2001).
Mantle plumes are the most plausible mechanism of
transporting the oxidised melt material from the core-mantle
boundary to Earth’s surface (Fig. 7.47d). If the assimilated
material becomes more buoyant than the surrounding mate-
rial, it would rise to the lithosphere-mantle boundary. At this
4 7.4 An Apparent Oxidation of the Upper Mantle Versus Regional Deep Oxidation 1153
location the less buoyant, hotter material would thermally
erode the overlying oceanic or continental lithosphere
(e.g. Fennoscandia) eventually resulting in the eruption of
intermediate to ultramafic lava flows. This could occur in a
rift situation (e.g. similar to the Afar rift, Rio Grande rift, or
Midcontinent rift) or a localised hotspot (e.g. Hawaiian
Islands, Yellowstone).
Unchanged Mantle Redox State
The lack of change within palaeoredox proxy datasets, with
respect to time, suggests that the oxidation state of the mantle
has remained unchanged since the Archaean (Frost and
McCammon 2008). Oxidised surface materials of the Earth
would have been available for subduction into the mantle
since the inception of plate tectonics (Kasting et al. 1993;
Kump et al. 2001), but the long-term effects of subduction
processes on the oxidation state of the mantle are still poorly
understood (Hirschmann 2009). Palaeoredox proxies suggest
that the mantle has essentially been at FMQ since the
Archaean (Berry et al. 2008; Canil 2002; Delano 2001;
Frost and McCammon 2008; Li and Lee 2004).
Palaeoredox studies using vanadium focus primarily on
mafic to ultramafic rocks with MgO contents between 8 and
12 wt.% (Li and Lee 2004). This narrow window is due to
phase equilibrium constraints and partitioning coefficients
resolved through careful experimental petrology. At low
MgO concentrations, less than 8 wt.%, clinopyroxene is
the dominant mineral phase. Since the partitioning
coefficients of vanadium are not well constrained due to
the varying chemistry of clinopyroxenes, the measured
V/Sc ratios used in determining palaeoredox conditions of
the melt may not reflect original fO2 conditions of the melt.
Ternary plots of V, Cr, and MgO (e.g. Delano 2001), as
well as binary plots of V/Sc vs. MgO (e.g. Li and Lee 2004),
V vs. MgO (e.g. Lee et al. 2003), V vs. Al2O3 (e.g. Canil
2002), and Cr vs. MgO (e.g. Delano 2001) all display
overlapping data points throughout geologic time around
the FMQ buffer. This pattern is interpreted as evidence for
an early oxidation of the upper mantle; however, this may
not necessarily be the case. For example, within the Cr vs.
MgO plots of Delano (2001), the data points are slightly
offset from one another through time. In other words, the
data collected from rocks 3900–3600 Ma old appear to be
slightly more oxidised than the 2900–2400Ma old rocks, and
both are more reduced when compared to modern MORB.
This pattern, although very subtle, suggests that the mantle
did indeed evolve to a more oxidised state throughout geo-
logic time, and merits further investigation. The data
presented in Delano (2001), coupled with the tight �0.5
and �0.3 error bars of Canil (2002) and Lee and Li (2004),
respectively, perhaps suggest that only a slight shift in the
redox state of the mantle may be necessary to decrease the
volcanic oxygen sink enough to allow the accumulation of
atmospheric oxygen. The highly oxidised lavas of the
Kuetsj€arvi Volcanic Formation from the Pechenga Green-
stone Belt may provide valuable insight into the redox
evolution through this pivotal time in Earth’s history, specif-
ically in testing the oxidised upper mantle hypothesis.
7.4.2 Kuetsj€arvi Volcanic Formation
Field Descriptions
The Kuetsj€arvi Volcanic Formation is the second of the four
volcanic formations of the Pechenga Group, with a thickness
between 800 and 2,000 m (Fig. 7.46). It is made up of mostly
subaerially erupted lavas varying in composition from
picrites to trachydacites forming amygdaloidal lava flows,
fluidal lavas and various lava breccias. The U-Pb age of
2058 � 2 Ma obtained from volcaniclastic conglomerates
in the middle part of the Kuetsj€arvi Volcanic Formation
(corresponding to the top of FAR-DEEP Core 6A) constrains
the minimum age of the volcanic succession (Melezhik et al.
2007). Only a brief summary of the unique physical and
geochemical properties observed within the Kuetsj€arvi Vol-
canic Formation is provided here. A more complete litho-
logic and petrographic description of the Kuetsj€arvi
Volcanic Formation can be found in Chaps. 4.2, 6.2.3,
6.2.4, and 6.2.5 of this book series.
The Kuetsj€arvi Volcanic Formation clearly differs from the
other volcanic units in the Pechenga area in being more
oxidised with Fe2O3 contents ranging between 6.34 and
21.3 wt.%. The ratio of oxidised iron to total iron (Fe3+/SFe)has a wide range in the Kuetsj€arvi volcanic rocks, being oftenhigher than 0.3, while in other volcanic formations within the
Pechanga complex it commonly falls below 0.3 (Fig. 7.48; see
Chap. 6.2.6). The volcanic rocks from the overlying Kolosjoki
Volcanic Formation and underlying Ahmalahti Formation
display a maximum at Fe3+/SFe of ca. 0.25.The difference in the oxidation states of iron may partly
be explained by the geotectonic environment during erup-
tion. The eruptions of the two oldest volcanic units of the
Pechenga Group, the Ahmalahti Formation and Kuetsj€arviVolcanic Formation, are interpreted as subaerial, but the two
youngest volcanic units, the Kolosjoki and Pilguj€arvi Volca-
nic Formations, are interpreted as being primarily submarine
(Hanski and Smolkin 1989; Melezhik et al. 2007). However,
the difference in the present redox states between the
Ahmalahti and Kuetsj€arvi volcanic rocks is striking, even
though both represent subaerial volcanism, and thus needs to
be explained.
1154 K.S. Rybacki et al.
The Geochemistry of Iron in Kuetsj€arvi VolcanicRocks
FAR-DEEP Holes 5A, 6A, 7A and 8B recovered more than
400 m of highly oxidised lavas in the Kuetsj€arvi VolcanicFormation with the most oxidised volcanic rocks being
found in the middle part of formation in Core 6A
(Fig. 7.46). This anomaly may represent evidence for rela-
tively oxidised mantle material (e.g. contaminated by
subducted BIFs or oxidised oceanic slabs). However, the
question still remains: do these lavas represent a change in
the oxidation state of the upper mantle, a result of magma
redox evolution upon ascent, or the simply secondary oxida-
tion during palaeo-weathering?
Oxidised Surface and GroundwaterThe highly oxidised lavas observed in the Fennoscandian
Shield could also be the result of a secondary process such as
the circulation of oxidising water. After the GOE, surface
waters on Earth would be an effective oxidising agent if they
were in equilibrium with the oxygen-rich atmosphere. Rocks
exposed to these waters on Earth’s surface and shallow
subsurface would become oxidised through interacting
with them. The oxidising fluid hypothesis for the origin of
the highly oxidised Kuetsj€arvi lavas is supported by the
observation of decoupled niobium/thorium (Nb/Th) and nio-
bium/uranium (Nb/U) ratios within depleted mantle derived
rocks (Collerson and Kamber 1999). According to Collerson
and Kamber (1999), the noticeable difference in Nb/U and
Nb/Th ratios around 2.0 Ga ago could be the result of
preferential recycling of U during oxidative weathering,
thus denoting the advent of oxidising surficial conditions.
Under oxidising conditions, the development of chemical
and physical weathering profiles, and in some cases com-
plete soil development, at the surface is to be expected;
however, these profiles may not be preserved due to their
preferential removal via physical erosion. Furthermore, the
concentrated movement and percolation of these oxidised
surface waters within fractures in the subsurface would
result in the oxidation of minerals in close proximity to the
fracture. If the bedrock were pervasively fragmented, or
even brecciated, the oxidation of redox sensitive minerals
would be much more extensive and may even result in the
development of weathering rinds such as those observed in
the Fennoscandian Shield (Fig. 7.49a; see also Chap. 6.2.3).
To further complicate matters, the presence or absence of a
diverse biosphere would have significant effects on physical
and chemical weathering processes.
Implications for FAR-DEEP Research
Some volcanic rocks in the Fennoscandian Shield, specifi-
cally the Kuetsj€arvi Volcanic Formation, are extremely
oxidised (Figs. 7.45 and 7.49a–e). The enriched ferric iron
content of the Kuetsj€arvi Volcanic Formation may be pri-
mary or secondary in origin. These rocks, and field area,
provide a unique opportunity to study and test the two
hypotheses proposed to explain their anomalously high fer-
ric iron contents: (1) an oxidised upper mantle or (2) the
deep oxidative weathering of terrestrial surfaces.
Within the Kuetsj€arvi Volcanic Formation and many
other Jatulian volcanic rocks the absence of primary mag-
matic silicate minerals (Fig. 7.49f), in addition to the
oxide phase alteration, limits the use of mineral composition
data for estimating the redox state of the parent magma.
However, primary clinopyroxene is preserved in olivine-
clinopyroxene-phyric picritic basalt from the Umba Volca-
nic Formation (Fig. 7.49g). The rocks of the Umba Volcanic
Formation are equally highly-oxidised (see Chap. 3.4) and
their minimum age has been constrained at ca. 2052Ma
(U-Pb on zircon; Martin et al. 2010). Thus, they represent
a time-correlative volcanic succession to the Kuetsj€arvi
Volcanic Formation. No Jatulian-age rocks seems to contain
petrographic evidence of primary hydrous phenocryst phases
that could indicate high water contents and corresponding
high oxygen fugacities of the parental magmas (cf. Kelley
and Cottrell 2009).
If the present iron redox states of the rocks do actually
represent the parental magmatic material primary redox
state, then the measured Fe3+/SFe ratios would require
logfO2 values varying from the NNO buffer (FMQ +1) to 3
log units above the HM buffer (FMQ +8), as calculated with
the method of Jayasuriya et al. (2004). Conversely, if the
high Fe3+/SFe ratios are due to post-crystallisation oxidationassociated with haematisation and formation of magnetite
within the rocks, then the primary redox state of the magma
cannot be inferred.
Kelley and Cottrell (2009) demonstrate that Fe3+/Fe in
fresh, undegassed basaltic glasses are greater in subduction
zone magmas than MORB and OIB. This observation
correlates well with H2O and trace-element tracers of
slab-derived fluids used to interpret the effects of subduc-
tion on the oxidation state of arc magmas. Examples of
elevated Fe3+/SFe include the hydrous basaltic andesites in
western Mexico as well as the subduction-related basalts
from the Stromboli volcano, Italy, which correspond to
redox conditions of approximately FMQ +4 (Lange and
Carmichael 1990), and FMQ +9 (Cortes et al. 2006), respec-
tively. The Stromboli volcano basalts have whole-rock Fe3+/
SFe ratios of approximately 0.90. Cortes et al. (2006) pro-
pose that the measured oxygen fugacities are due to the
crystallisation of a highly oxidised melt resulting from
magma chamber degassing.
Physical processes, namely volcanic degassing, can have
a major effect on the final chemical composition of volcanic
rocks (e.g. Burgisser and Scaillet 2007; Oppenheimer et al.
2011). Burgisser and Scaillet (2007) found that the redox
4 7.4 An Apparent Oxidation of the Upper Mantle Versus Regional Deep Oxidation 1155
state of erupted magmas and their coeval gases are not the
same. Varying the initial sulphur content within a H-O-S
system, and consequentially the initial fH2, of a magma at
depth readily affects the final redox state of a magma as it
rises to the surface (Burgisser and Scaillet 2007; their
Fig. 2). For example, a magma with an initial sulphur con-
tent of 253 ppm and fH2 ¼ 1 has a calculated redox compo-
sition equivalent to NNO +1.5 (FMQ +3.5) when fH2O is
initially fixed at 1 kbar. As the magma rises to the surface,
the redox state of the magma continually becomes more
reducing ending somewhere between NNO �0.4 (FMQ
+0.6) and NNO +0.4 (FMQ +1.4) depending upon initial
gas concentrations. Conversely, the redox state of a magma
with lower initial sulphur and fH2 contents rising to the
surface first increases and then gradually decreases to a
final oxidation state that is either higher or lower than the
initial redox state of the magma depending upon the initial
gas content (Burgisser and Scaillets 2007; their Fig. 2).
Magmas with the same initial gas content, but different
initial sulphur and fH2 contents, acquire similar final redox
states. Magma at depth with initial gas contents of 0.1, 1, and
5 wt.% result in a final magma oxidation state at the surface
of NNO �0.1 [FMQ +0.9], �0.6 [FMQ +0.4], and �1.0
[FMQ], respectively. Oppenheimer et al. (2011) propose
that within the magma body of Erebus volcano, Antarctica,
CO2 is partitioned into liquid form from its mantle source. In
turn, the liquid CO2 is expedited through the system
oxidising the lowermost magma body (FMQ +1). When
the liquid CO2 reaches the shallow lava lake close to the
surface, it is converted into CO2 gas due to the overall
reduction in pressure. This CO2 gas is then expelled from
the lava lake in the form of bubble eruptions as well as
continuous diffusion across the magma-air interface, thus
causing the reduction in the redox state of the magma in
the shallow lava pool to approximately FMQ.
Alternatively, these rocks are highly oxidised because
they interacted with oxidising fluids during or after eruption
and emplacement. Other examples of ancient haematised
basalts have been observed around the world, for example,
in the Pilbara Craton (Kato et al. 2009). The Archaean rocks
of the Pilbara Craton are interpreted to have been oxidised
via percolation of O2-rich groundwater through a shear zone
prior to 2760 Ma when the basalts were exposed to the
surface after an orogeny at 2900 Ma, approximately
700 Ma after the initial eruption of the basalts estimated at
3460 Ma, and almost 300 Ma prior to the GOE. Kato et al.
(2009) propose two scenarios for the origin of oxygenated
groundwater. One possibility is the development of
oxygenated water bodies though intense cyanobacterial
activity not in contact with the anoxic atmosphere. Another
possibility is that the goundwaters equilibrated with an
atmosphere containing approximately 1.5 % PAL oxygen.
Kato et al. (2009) clearly state that further study is needed to
determine which of the two scenarios is correct.
In the Fennoscandia Shield, oxidation predates regional
metamorphism and is thus pre-orogenic and may be related
to the same spatially associated surfical processes that led to
red bed formation (see Chap. 6.2.3). A better analogue might
be found in a continental setting with subaerial, vesicular
lava eruptions such as those in the Mesoproterozoic Mid-
Continental Rift System in North America. In these lavas,
opaque minerals are commonly haematised (Cornwall
1951). Annells (1972) described olivine grains in otherwise
fresh basalts, where they were replaced by saponite and
haematite, thus forming an analogue to the chlorite- and
haematite-replaced olivine phenocrysts in magnesian lavas
of the Kuetsj€arvi Volcanic Formation (Fig. 7.49f).
Deciphering secondary redox overprinting from the primary
magma redox state is very difficult and thus calls for the use
of palaeo-redox proxies which are believed to not be
affected by secondary weathering processes. The detailed
study of source magma oxidation state and palaeosol devel-
opment, coupled with the use of palaeo-redox proxies
immune to weathering, diagenesis, and metamorphism will
allow for the direct comparison of primary versus secondary
redox states.
Magmatic Palaeoredox ProxiesIn situ determination of Fe+2/Fe+3 in primary magmatic
clinopyroxene of the Umba volcanic rocks (Fig. 7.49g) has
great potential to address the redox state of the mantle. In
general, vanadium (V) can be used to evaluate the redox
state of the mantle through time since it exists in three
valence states, V5+, V4+, and V3+, on Earth (Lee et al.
2003). For example, V-Al2O3, V-MgO, and V-Sc systemat-
ics have been utilised to estimate the relative oxygen
fugacities of mafic volcanic rocks (Canil 2002; Frost and
McCammon 2008; Lee 2005; Lee et al. 2003; Li and Lee
2004). In highly oxidised magmas, vanadium behaves as an
incompatible element because V3+ is preferentially
incorporated into crystals. In other words, vanadium-bearing
minerals which crystallised in equilibrium with a magma
that has a low fO2 would have a vanadium content much
greater than the same mineral crystallised from a similar
magma source with a high fO2 (Canil 2002; Lee et al. 2003).
Studies using V-Al2O3, V-MgO, and V/Sc as redox prox-
ies of magmatic eruptions have not revealed significant
differences between mid-ocean ridge, ocean island, and
island arc basalts; therefore, it has been concluded that the
mantle source regions of these basalts are indistinguishable
from each other, having oxygen fugacities of FMQ � 0.5
(Canil 2002; Frost and McCammon 2008; Lee 2005; Lee
et al. 2003; Li and Lee 2004; Mallmann and O’Neill 2009).
The more oxidising conditions observed in island arc
1156 K.S. Rybacki et al.
magmas, as indicated by Fe3+/Fe2+ ratios, are the result of
late-stage processes and thus not reflective of the mantle
source region. Thus, V is a better proxy for the redox state
of the mantle in the geologic past.
PalaeosolsPalaeosols possibly contain the most promising palaeo-
atmospheric proxy record since they form in direct contact
with the atmosphere. However, their low preservation poten-
tial and correspondingly limited abundance in the geologic
record limits their usage. Drillcores collected during FAR-
DEEP potentially sample several Precambrian palaeosols
spanning the GOE. The detailed geochemical analysis of
multiple palaeosols during this significant time will provide
invaluable insight into the dominant weathering conditions
during the GOE.
Well-drained palaeosols generally exhibit specific trends
with respect to mobile and immobile elements. In the most
general terms, the measured concentrations of mobile
cations (i.e. Ca, Mg, Na, K and Mn) will typically decrease
upwards within a palaeosol, while immobile cations (i.e. Ti,
Al and Zr) should display little to no variation vertically.
These general observations suggest that the mobile cations
are greatly affected during palaeosol development, and that
immobile cations are not. Using this assumption, immobile
element profiles can be used to determine if palaeosols have
been subjected to further post-depositional alteration (e.g.
tau plots and compaction calculations; Brantley and White
2009).
Redox sensitive elements, such as iron, are valuable
because they can behave as both a mobile and immobile
element depending upon the atmospheric conditions during
palaeosol development (Driese 2004). Since palaeosols form
in direct contact with the atmosphere, the oxidation state of
the iron within iron-bearing mineral phases should reflect the
chemistry of the atmosphere. For palaeosols that formed in
equilibrium with an oxygen-poor atmosphere, the dominant
iron species should be ferrous iron. Iron will presumably be
lost from the palaeosol profile because ferrous iron is solu-
ble, and it will behave like a mobile element. Alternatively,
palaeosol profiles in equilibrium with an atmosphere that
had sufficient oxygen (~1 % present atmospheric levels) to
oxidise the ferrous iron to ferric iron will exhibit minimal
loss of iron from the profile (Holland 1984). This is because
iron will behave as an immobile element because the soluble
ferrous iron will re-precipitate instantaneously into insoluble
iron oxides and oxy-hydroxide mineral phases. The precipi-
tation of ferric iron mineral phases theoretically inhibits iron
from being lost from the system. For these reasons, examin-
ing iron retention and loss within palaeosol profiles is an
effective way to estimate palaeo-atmospheric conditions.
This is of course assuming that secondary alteration pro-
cesses have not significantly altered the chemistry of the iron
within the palaeosol.
The use of multiple palaeo-atmospheric redox proxies
is an excellent way to address if palaeosols have been
altered by secondary alteration processes, and a clearer
picture of the geochemical evolution of a palaeosol should
emerge.
Alteration by Groundwater and HydrothermalFluidsSubaerially erupted magmas that degas result in the forma-
tion of rocks with ubiquitous cavities of variable shape
formed by the entrapment of gas or vapour bubbles during
the solidification of lavas. These cavities, or vesicles, com-
monly remain unfilled until the lava succession subsides
below the groundwater table. Under such conditions the
vesicles begin to be filled by various low-temperature,
hydrothermal minerals such as calcite, quartz, chalcedony,
zeolites and others. Hence, chemical and isotopic composi-
tion of amygdaloidal minerals represents a proxy for the
composition of the groundwater from which they
precipitated. The Kuetsj€arvi Volcanic Formation rocks are
an outstanding example of eruption and formation in subaer-
ial environments. They contain abundant amygdales, which
are well represented in FAR-DEEP Cores 6A and 7A
(Fig. 7.49e, h–m). The Kuetsj€arvi amygdales exhibit very
diverse mineralogy with chlorite, haematite, quartz
(Fig. 7.49e), axinite, and epidote predominant (Fig. 7.49h).
Many amygdales show a composite infill consisting of epi-
dote, axinite, adularia, calcite, minor pyrite and chalcopyrite
(Fig. 7.49i), which are all indicative of the evolving chemis-
try of groundwater. Some large amygdales are composed of
finely-crystalline, homogeneous quartz resembling chalce-
dony (Fig. 7.49j), whereas others, smaller in size, consist of
bands of quartz and haematite with micron-scale rhythmic
alternation (Fig. 7.49k, l). Many amygdales contain sphene
and allanite, which were likely precipitated from hydrother-
mal fluids (Fig. 7.49m). Consequently mineralogical diver-
sity of the Kuetsj€arvi amygdales enables addressing
groundwater and/or hydrothermal fluid geochemistry. Oxy-
gen (quartz), iron (haematite), boron (axinite) and sulphur
(sulphides) isotope systems could potentially be used to
elucidate the temperature, the composition and redox-state
of mineral-precipitating fluids and perhaps even evolution.
Moreover, some amygdaloidal minerals can be dated by the
U-Pb (sphene and allanite) and U-Th-He (haematite)
techniques thus providing time constraints for rock alteration
involving Palaeoproterozoic groundwaters.
Summary
The FAR-DEEP drillcores sample rocks from a very
dynamic time in geologic history. They allow for detailed
geological and geochemical sampling of the highly oxidised
rocks from the Fennoscandian Shield, which may offer
4 7.4 An Apparent Oxidation of the Upper Mantle Versus Regional Deep Oxidation 1157
definitive evidence when testing the oxidised upper mantle
versus regional deep oxidative weathering of terrestrial
surfaces hypotheses. The use of vanadium and chromium
redox proxies, and the examination of cation mobility and
Fe3+/Fe2+ within weathering profiles will be the most effec-
tive methods to unravel the very complex, but compelling
geochemical evolution of the ocean and atmosphere as
recorded in the Fennoscandian Shield.
1158 K.S. Rybacki et al.
Fig. 7.45 Field outcrop of highly-oxidised, fluidal and fragmented
trachydacitic lava with bands of haematite and magnetite (black). Thelava marks the uppermost surface of the Kuetsj€arvi volcanic successionbeneath the Kolosjoki Sedimentary Formation. A high iron content
(ca. 25 wt.%) and its oxidation state may reflect a combined effect of
various primary (magmatic) and secondary (postvolcanic hydrothermal
alteration or surface oxidation) processes (Photograph by Victor
Melezhik)
4 7.4 An Apparent Oxidation of the Upper Mantle Versus Regional Deep Oxidation 1159
Fig. 7.46 Lithological column of the Northern Pechenga Group,
position of FAR-DEEP drillholes, and stratigraphic profile of Fe3+/SFe (%) through the middle part of the Kuetsj€arvi Volcanic Formation
intersected by FAR-DEEPHole 6A. The Fe3+/SFe (%) diagram is based
on unpublished analyses by R. Kontio performed at the Department of
Geosciences, University of Oulu, Finland. FeO was determined by
potassium permanganate titration and Fe2O3 was calculated from the
difference between FeO and total Fe measured by XRF. Superscripts
denote radiometric ages from (1) Amelin et al. (1995), (2)Melezhik et al.
(2007), (3) Hannah et al. (2006) and (4) Hanski (1992)
1160 K.S. Rybacki et al.
Fig. 7.47 (a) A cartoon cross-section of Earth illustrating potential
sources and processes of oxidised magma available for eruption to
the surface (Modified from Albarede and van der Hilst 1999). The
assimilation of oxidised crustal material in the upper mantle through
slab dewatering could potentially create an oxidised upper
mantle (above dashed line). On the other hand, if subducted material is
4 7.4 An Apparent Oxidation of the Upper Mantle Versus Regional Deep Oxidation 1161
Fig. 7.47 (continued) detached from the subducted slab and sinks to
the core-mantle boundary (e.g. “slab graveyard”), it can then be
reassimilated creating an oxidised magma body relative to the
surrounding mantle (Albarede and Van der Hilst 1999; Kump
et al. 2001). The oxidation of iron, in addition to other redox
sensitive phases, at the water-rock interface can take place both in
the absence or presence of oxygen ((b) Catling and Claire 2005;
Holland 1984; Kasting et al. 1993). In the absence of free oxygen,
ferrous iron is oxidised by the dissociation of water thus producing
magnetite (Fe3O4) and hydrogen, but when free oxygen is present,
ferrous iron – in the form of ferrous oxide (FeO) and iron sulphide
(Fe2Sx) mineral phases – is oxidised by the free oxygen and
produces haematite (Fe2O3). In addition to haematite, the oxidation
of iron sulphide mineral phases will produce hydrogen and soluble
sulphate (Catling and Claire 2005). The oxidation of organic carbon
by free oxygen would produce CO2.. When this oxidised crustal
material is subducted, it can be recycled quickly and erupted in
the form of a volcanic arc (c) or detach and sink to the core-
mantle boundary where it is slowly reassimilated and can later be
uplifted and melted via a mantle plume, producing an oxidised
parent magma eruption to the surface ((d) Albarede and van der
Hilst 1999). Magma with a fO2 of greater than FMQ +1 is defined as
an oxidised magma, while reduced mantle is hypothesized to be
equal to the FMQ buffer
Fig. 7.48 Oxidation state of iron in volcanic rocks from the Pechenga
Belt, based on data from the literature and FAR-DEEP project.
Histograms of ferric/total iron ratios are illustrated for four formations
being, from oldest to youngest, the Ahmalahti Formation (a),
Kuetsj€arvi (b), Kolosjoki (c), and Pilguj€arvi (d) Volcanic formations.
The dashed vertical line at Fe3+/SFe is arbitrarily chosen to illustrate
the uniqueness of the relative ferric iron content of the Kuetsj€arviFormation (>0.3) when compared to other formations (<0.3)
1162 K.S. Rybacki et al.
Fig. 7.49 Volcanic rocks of the Kuetsj€arvi Volcanic Formation. (a)
Volcaniclastic conglomerate containing rounded fragments of dacitic
lava with concentric, haematite-rich (red, brown) bands suggesting thatthe haematisation predates the metamorphism; the upper part of Core6A. (b) Field outcrop of top surface of rhyodacitic lava-flow exhibiting
haematised contraction joints which are cross-cut by quartz veins. (c)
Cross-section view through a series of thin lava flows with
haematisation (brown bands) occurring on either side of each flow
contact. (d) Field outcrop of a single flow of amygdaloidal basaltic
lava with fragmented top affected by haematisation (dark coloured)
4 7.4 An Apparent Oxidation of the Upper Mantle Versus Regional Deep Oxidation 1163
Fig. 7.49 (continued) (e) Microcrystalline, trachydacitic lava with
ubiquitous, variably flattened amygdales composed of chlorite (green),haematite (brown) and quartz and calcite (white); themiddle part of Core6A. (f) Photomicrograph in reflected light of olivine phenocrysts
replaced by chlorite and haematite and chromite crystals replaced by
haematite in picritic basalt from the upper part of Core 6A. (g) Photomi-
crograph in transmitted, polarised light of olivine-clinopyroxene-phyric
picritic basalt from the Umba Volcanic Formation, showing a fresh,
twinned and zoned, euhedral clinopyroxene phenocryst together with a
serpentinised olivine phenocryst in a fine-grained groundmass. (h) Field
outcrop of large vesicles filled by violet axinite rimmed by green epidote
in basaltic lava flow; the flow corresponds to the uppermost part of Core6A. (i) Composite amygdale composed of red adularia, white calcite,
green epidote and violet axinite; the middle part of Core 7A
1164 K.S. Rybacki et al.
Fig. 7.49 (continued) (j) Large amygdales composed of finely-
crystalline quartz in basaltic lava flow from the middle part of Core
7A. (k, l) Photomicrographs (transmitted, non-polarised light) showing
ultra-fine alternation of quartz and haematite bands in amygdales from
trachyandesitic lava; the middle part of Core 6A. (m) Photomicrograph
(transmitted, non-polarised light) of quartz-calcite amygdale with core
filled with allanite from trachydacitic lava; the middle part of Core 6A(Photographs (a–e, h–j) by Victor Melezhik, photographs (f, g, k–m)
by Zhen-Yu Luo. Sample (g) courtesy of the Mineralogical Museum of
the Geological Institute of the Kola Science Centre)
4 7.4 An Apparent Oxidation of the Upper Mantle Versus Regional Deep Oxidation 1165
References
Albarede F, van der Hilst RD (1999) New mantle convection model
may reconcile conflicting evidence. EOS Trans Am Geophys Union
80:535–539
Amelin YuV, Heaman LM, Semenov VS (1995) U-Pb geochronology
of layered mafic intrusions in the eastern Baltic Shield: implications
for the timing and duration of Palaeoproterozoic continental rifting.
Precambrian Res 75:31–46
Annells R (1972) Proterozoic flood basalts of eastern Lake Superior:
the Keweenawan volcanic rocks of the Mamainse Point area,
Ontario. Department of Energy, Mines and Resources, Ottawa, p 51
Battistuzzi F, Feijao A, Hedges SB (2004) A genomic timescale of
prokaryote evolution: insights into the origin of methanogenesis,
phototrophy, and the colonization of land. BMC Evol Biol 4:44
Bekker A, Holland H, Wang P, Rumble D, Stein H, Hannah J, Coetzee
L, Beukes N (2004) Dating the rise of atmospheric oxygen. Nature
427:117–120
Berry AJ, Danyushevsky LV, O’Neill HSC, Newville M, Sutton SR
(2008) Oxidation state of iron in komatiitic melt inclusions
indicates hot Archaean mantle. Nature 455:960–963
Bezos A, Humler E (2005) The Fe3+/Fe ratios of MORB glasses and
their implications for mantle melting. Geochim Cosmochim Acta
69:711–725
Brantley SL, White AF (2009) Approaches to modeling weathered
regolith. Rev Mineral Geochem 70:435–484
Buick R (1992) The antiquity of oxygenic photosynthesis: evidence
from stromatolites in sulphate deficient Archaean Lakes. Science
255:74–77
Buick R (2008) When did oxygenic photosynthesis evolve? Philos
Trans R Soc B Bio Sci 363:2731–2743
Burgisser A, Scaillet B (2007) Redox evolution of a degassing magma
rising to the surface. Nature 445:194–197
Canfield DE (2005) The early history of atmospheric oxygen: homage
to Robert M. Garrels. Annu Rev Earth Planet Sci 33:1–36
Canil D (2002) Vanadium in peridotites, mantle redox and tectonic
environments: Archean to present. Earth Planet Sci Lett 195:75–90
Catling DC, Claire MW (2005) How Earth’s atmosphere evolved to an
oxic state: a status report. Earth Planet Sci Lett 237:1–20
Christie DM, Carmichael ISE, Langmuir CH (1986) Oxidation states of
mid-ocean ridge basalt glasses. Earth Planet Sci Lett 79:397–411
Collerson KD, Kamber BS (1999) Evolution of the continents and
atmosphere inferred from Th-U-Nb systematics of the depleted
mantle. Science 283:1519–1522
Condie KC, O’Neill C, Aster RC (2009) Evidence and implications for
a widespread magmatic shutdown for 250My on Earth. Earth Planet
Sci Lett 282:294–298
Cornwall HR (1951) Ilmenite, magnetite, hematite, and copper in lavas
of the Keweenawan series [Michigan]. Econ Geol 46:51–67
Cortes JINA, Wilson M, Condliffe E, Francalanci L (2006) The occur-
rence of forsterite and highly oxidizing conditions in basaltic lavas
from Stromboli volcano, Italy. J Petrol 47:1345–1373
Delano JW (2001) Redox history of Earth’s interior since 3900 Ma:
implications for prebiotic molecules. Orig Life Evol Biosph
31:311–341
Driese SG (2004) Pedogenic translocation of Fe in modern and ancient
Vertisols and implications for interpretations of the Hekpoort
paleosol (2.25 Ga). J Geol 112:543–560
Evans D, Beukes N, Kirschvink J (1997) Low-latitude glaciation in the
Palaeoproterozoic era. Nature 386:262–266
Farquhar J, Bao H, Thiemens M (2000) Atmospheric influence of
Earth’s earliest sulfur cycle. Science 289:756–758
Farquhar J, Zerkle AL, Bekker A (2010) Geological constraints on the
origin of oxygenic photosynthesis. Photosynth Res 107:11–36
Frei R, Gaucher C, Poulton SW, Canfield DE (2009) Fluctuations in
Precambrian atmospheric oxygenation recorded by chromium
isotopes. Nature 461:250–253
Frost DJ, McCammon CA (2008) The redox state of Earth’s mantle.
Annu Rev Earth Planet Sci 36:389–420
Gaillard F, Scaillet B, Arndt NT (2011) Atmospheric oxygenation
caused by a change in volcanic degassing pressure. Nature
478:229–232
Guo Q, Strauss H, Kaufman AJ, Schroder S, Gutzmer J, Wing B, Baker
MA, Bekker A, Jin Q, Kim ST, Farquhar J (2009) Reconstructing
Earth’s surface oxidation across the Archean-Proterozoic transition.
Geology 37:399–402
Hannah JL, Stein HJ, Zimmerman A, Yang G, Markey RJ, Melezhik
VA (2006) Precise 2004 � 9 Ma Re‐Os age for Pechenga black
shale: comparison of sulfides and organic material. Geochim
Cosmochim Acta 70:A228
Hanski EJ (1992) Petrology of the Pechenga ferropicrites and
cogenetic, Ni-bearing gabbro-wehrlite intrusions, Kola Peninsula,
Russia. Geol Survey Finland Bull 367:192
Hanski EJ, Smolkin VF (1989) Pechenga ferropicrites and other early
Proterozoic picrites in the eastern part of the Baltic Shield. Precam-
brian Res 45:63–82
Hirschmann M (2009) Ironing out the oxidation of Earth’s mantle.
Science 325:545–546
Holland HD (1984) The chemical evolution of the atmosphere and
oceans. Princeton University Press, Princeton, p 587
Holland HD (2009) Why the atmosphere became oxygenated: a pro-
posal. Geochim Cosmochim Acta 73:5241–5255
Jayasuriya KD, O’Neill HSC, Berry AJ, Campbell SJ (2004) A
Mossbauer study of the oxidation state of Fe in silicate melts. Am
Mineral 89:1597–1609
Kasting JF (2006) Ups and downs of ancient oxygen. Nature
443:643–645
Kasting JF (2008) The primitive earth. In: Wong JT-F, Lazcano A (eds)
Prebiotic evolution and astrobiology. Landes Biosciences, Austin,
pp 1–8
Kasting JF, Eggler DH, Raeburn SP (1993) Mantle redox evolution
and the oxidation state of the Archean atmosphere. J Geol
101:245–257
Kato Y, Suzuki K, Nakamura K, Hickman AH, Nedachi M, Kusakabe
M, Bevacqua DC, Ohmoto H (2009) Hematite formation by
oxygenated groundwater more than 2.76 billion years ago. Earth
Planet Sci Lett 278:40–49
Kelley KA, Cottrell E (2009) Water and the oxidation state of subduc-
tion zone magmas. Science 325:605–607
Kellogg LH, Hager BH, Van der Hilst RD (1999) Compositional
stratification in the deep mantle. Science 283:1881–1884
Kopp RE, Kirschvink JL, Hilburn IA, Nash CZ (2005) The Paleopro-
terozoic snowball Earth: a climate disaster triggered by the evolu-
tion of oxygenic photosynthesis. Proc Natl Acad Sci USA
102:11131–11136
Kump LR (2008) The rise of atmospheric oxygen. Nature 451:277–278
Kump LR, BarleyME (2007) Increased subaerial volcanism and the rise
of atmospheric oxygen 2.5 billion years ago. Nature 448:1033–1036
Kump LR, Kasting JF, Barley ME (2001) Rise of atmospheric oxygen
and the “upside-down” Archean mantle. Geochem Geophys
Geosyst 2: Paper No. 2000GC0114, p 10
Lange RA, Carmichael ISE (1990) Hydrous basaltic andesites
associated with minette and related lavas in western Mexico. J
Petrol 31:1225–1259
Lee CTA (2005) Similar V/Sc systematics in MORB and arc basalts:
implications for the oxygen fugacities of their mantle source
regions. J Petrol 46:2313–2336
Lee CTA, Brandon AD, NormanM (2003) Vanadium in peridotites as a
proxy for paleo-fO2 during partial melting: prospects, limitations,
and implications. Geochim Cosmochim Acta 67:3045–3064
1166 K.S. Rybacki et al.
Li ZXA, Lee CTA (2004) The constancy of upper mantle fO2 through
time inferred from V/Sc ratios in basalts. Earth Planet Sci Lett
228:483–493
Mallmann G, O’Neill HSC (2009) The crystal/melt partitioning of V
during mantle melting as a function of oxygen fugacity compared
with some other elements (Al, P, Ca, Sc, Ti, Cr, Fe, Ga, Y, Zr and
Nb). J Petrol 50:1765–1794
Martin AP, Condon DJ, Prave AR, Melezhik VA, Fallick A (2010)
Constraining the termination of the Lomagundi-Jatuli positive iso-
tope excursion in the Imandra-Varzuga segment (Kola Peninsula,
Russia) of the North Transfennoscandian Greenstone Belt by high-
precision ID-TIMS, AGU, San Francisco, 13–17 Dec 2010
Melezhik VA, Huhma H, Condon DJ, Fallick AE, Whitehouse MJ
(2007) Temporal constraints on the Paleoproterozoic Lomagundi-
Jatuli carbon isotopic event. Geology 35:655–658
Murakami T, Sreenivas B, Sharma SD, Sugimori H (2011) Quantifica-
tion of atmospheric oxygen levels during the Paleoproterozoic using
paleosol compositions and iron oxidation kinetics. Geochim
Cosmochim Acta 75:3982–4004
Oppenheimer C, Moretti R, Kyle PR, Eschenbacher A, Lowenstern JB,
Hervig RL, Dunbar NW (2011) Mantle to surface degassing of
alkalic magmas at Erebus volcano, Antarctica. Earth Planet Sci
Lett 306:261–271
Oskarsson N, Helgason O, Steinthorsson S (1994) Oxidation state of
iron in mantle-derived magmas of the Icelandic rift zone. Hyperfine
Interact 91:733–737
Rasmussen B, Fletcher IR, Brocks JJ, Kilburn MR (2008) Reassessing
the first appearance of eukaryotes and cyanobacteria. Nature
455:1101–1104
Rowe MC, Kent AJR, Nielsen RL (2009) Subduction influence on
oxygen fugacity and trace and volatile elements in basalts across
the Cascade Volcanic Arc. J Petrol 50:61–91
Summons RE, Jahnke LL, Hope JM, Logan GA (1999) 2-
Methylhopanoids as biomarkers for cyanobacterial oxygenic photo-
synthesis. Nature 400:554–557
Summons RE, Bradley AS, Jahnke LL, Waldbauer JR (2006) Steroids,
triterpenoids and molecular oxygen. Philos Trans R Soc B Biol Sci
361:951–968
Symonds RB, Rose WI, Bluth GJS, Gerlach TM (1994) Volcanic-gas
studies; methods, results, and applications. Rev Mineral Geochem
30:1–66
Van der Hilst RD, Karason H (1999) Compositional heterogeneity in
the botton 1000 kilometers of Earth’s mantle: toward a hybrid
convection model. Science 283:1885–1888
Watanabe Y, Farquhar J, Ohmoto H (2009) Anomalous fractionations
of sulfur isotopes during thermochemical sulfate reduction. Science
324:370–373
4 7.4 An Apparent Oxidation of the Upper Mantle Versus Regional Deep Oxidation 1167
7.5 Abundant Marine Calcium Sulphates:Radical Change of Seawater Sulphate Reservoirand Sulphur Cycle
Harald Strauss, Victor A. Melezhik, Marlene Reuschel, Anthony E. Fallick,Aivo Lepland, and Dmitry V. Rychanchik
7.5.1 Introduction
Harald Strauss
The modern (pre-industrial) ocean is characterised by a
concentration of dissolved sulphate of 28 mM with little
variability in its horizontal or vertical distribution. This
homogeneity is a consequence of the long residence time
of sulphate of some 25 Ma in comparison to the present
ocean mixing time of 1,000–2,000 years (e.g. Holland 1984).
The major source for oceanic sulphate is continental
weathering of evaporitic gypsum and anhydrite and of reduced
sulphur-bearing minerals under our present-day oxygen-rich
atmosphere, with pyrite (FeS2) being the most prominent form
of reduced sulphur. Although this reaction can occur as a
purely inorganic reaction, it can/will equally well be mediated
by sulphide oxidising bacteria (e.g. Canfield 2001). The
resulting sulphate is delivered to the ocean via rivers providing
an integrated signal of oxidative chemical weathering for any
given catchment. Only in the direct discharge area of large
river systems can a horizontal and/or vertical gradient in
sulphate concentration be observed (e.g. Cameron et al. 1995).
Under Earth surface conditions (i.e. below 100�C) two
principal pathways mediate the transfer of dissolved oceanic
sulphate from the ocean into the sediment. The first one is the
precipitation of calcium sulphates under evaporitic conditions.
It should be noted that also barite and celestite precipitate in
the marine realm (e.g. Paytan et al. 2002), albeit representing a
minor contribution tomarine sediments.More importantly, yet
unconstrained in respect tomass-balance considerations, is the
incorporation of oceanic sulphate into marine carbonates with
concentrations ranging from a few 10 to several 1,000 ppm (e.
g. Busenberg and Plummer 1985; Staudt and Schoonen 1995;
Grossman et al. 1996). This carbonate-associated sulphate has
become an important proxy for reconstructing the sulphur
isotopic composition of oceanic sulphate in the geologic past
(see discussion in the subsequent sections). The second path-
way is (microbially driven) sulphate reduction and subsequent
formation of pyrite.
At present, no major evaporite formation occurs on Earth.
Consequently, bacterial sulphate reduction and resulting
pyrite formation represents the major sink for marine
dissolved sulphate in the modern ocean. Sulphate reduction,
mediated by strictly anaerobic bacteria (e.g. Canfield 2001), is
coupled to the mineralisation (recycling) of organic matter.
And in fact, next to aerobic respiration, bacterial sulphate
reduction is second in importance with respect to the
recycling of sedimentary organic matter in the marine realm
(e.g. Jørgensen 1982). Note, however, that this process
requires anoxic environmental conditions.
From these introductory remarks, it is evident that the abun-
dance of dissolved oceanic sulphate is intimately linked to the
abundance of oxygen in the environment. Moreover, microbial
redox processes appear to be important for the source but more
so as a sink function for oceanic sulphate. Finally, evaporitic
sulphate is considered to be a faithful recorder of climatic (i.e.
evaporitic) conditions through Earth history (e.g. Ziegler
1990). As a side aspect and accepting a uniformitarian view
towards climate belt distribution, evaporite deposits have been
utilised in reconstructing temporal variations in palaeo-
geography throughout the past 2.5 billion years (Evans 2006).
This chapter aims at discussing the reason(s) for a radical
change in seawater sulphate abundance in the early Palaeo-
proterozoic. Chemical and physical evidence is presented,
some of it already from the newly acquired FAR-DEEP drill
cores, that indicates the existence of a sizeable marine
sulphate reservoir already in the Palaeoproterozoic ocean.
However, before turning to the distant past, we will set the
stage by looking at the systematics of the younger Phanero-
zoic sulphur cycle and the temporal evolution of oceanic
sulphate abundance. Being much better understood, it serves
well to formulate the principal questions for critically
evaluating the Precambrian archive of seawater sulphate.
H. Strauss (*)
Institut f€ur Geologie und Pal€aontologie, Westf€alische Wilhelms-
Universit€at, Corrensstr. 24, 48149 M€unster, Germany
e-mail: [email protected]
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_5, # Springer-Verlag Berlin Heidelberg 2013
1169
7.5.2 The Global Sulphur Cycle DuringPhanerozoic Time
Harald Strauss and Marlene Reuschel
From a geological perspective, a rather simplified view of the
global sulphur cycle has emerged (Fig. 7.50) that is largely
constrained by two important observations: (a) being a redox-
sensitive element with valence states ranging from �2 to þ6,
sulphur participates in redox reactions (whether inorganic or
microbially mediated), and (b) the two thermodynamically
stable sedimentary minerals are pyrite (FeS2) with sulphur at a
mean valence state of �1 and a calcium sulphate, whether
anhydrite (CaSO4) or gypsum (CaSO4 � 2 H2O), with sul-
phur at a valence state of þ6. Hence, the principal processes,
both input and output functions of the global sulphur cycle are
redox dependent.
Mass balance considerations indicate that oceanic sulphate
concentration is a dynamic balance between sources and
sinks. Currently, no large-scale evaporite formation occurs
on Earth, resulting in an oceanic sulphate budget that is
greatly out of balance (in particular when considering the
additional anthropogenic contribution). However, ample evi-
dence indicates the recurring formation of evaporite deposits
during Phanerozoic time (e.g. Zharkov 1984; Warren 1999),
and hence changes in environmental conditions including the
abundance of oceanic sulphate (e.g. Hardie 1996).
At times without substantial evaporite formation, pyrite
formation and burial represents the principal sink for oceanic
sulphate. The overall reaction can be written as:
2 Fe2O3 þ 16 Ca2þ þ 16 HCO3� þ 8 SO4
2� !4 FeS2 þ 16 CaCO3 þ 8H2O þ 15O2
Sulphate reducing bacteria are strictly anaerobic
organisms. This sets certain environmental limits for this
prime sink function for oceanic sulphate. Under modern
oxic conditions in the atmosphere–ocean system, this
microbially mediated reaction occurs almost exclusively
within the marine sediments, more precisely in the pore
water realm below the zone where aerobic respiration has
depleted the sedimentary column in oxygen. The quantita-
tive importance of this process is determined by the avail-
ability of reactive organic matter at and the delivery of
dissolved sulphate to the actual site of reduction in the
sedimentary pore space. Only in few places where
the water column exhibits anoxic conditions below a
chemocline (such as the Black Sea), resulting from a net
sink in oxygen (i.e. O2 consumption >O2 replenishment),
bacterial sulphate reduction is possible/occurs above the
sediment-water interface and hydrogen sulphide as the prin-
cipal product of this reaction is produced in the water col-
umn. Given sufficiently available reactive iron, the
extremely reactive hydrogen sulphide will be captured and
archived as sedimentary iron sulphide.
The products of both sink functions, either the oxidised
(evaporite precipitation) or the reduced (bacterial sulphate
reduction and subsequent iron sulphide formation) sulphur
are archived in the sedimentary rock record. Consequently,
evidence for a proposed increase in the abundance of oce-
anic sulphate and a radical change in the global sulphur
cycle, has to come foremost from the sedimentary rock
record. This seemingly simple and straight forward
approach, however, turns out to be rather difficult. The
temporal distribution of evaporite deposits is irregular in
time and space. It is highly fragmentary and incomplete,
due to the necessity of distinct environmental conditions
for their formation and the ease at which respective deposits
are being eroded. Hence, even a qualitative assessment of
the temporal evolution of oceanic sulphate abundance,
let alone a quantitative one that is solely based on evaporite
occurrences, appears susceptible to significant errors. This
pertains to the Phanerozoic and even more so to the Precam-
brian. But apart from the erosional aspect, the unequivocal
marine nature of preserved evaporite deposits needs to be
established. Only then constraints on the temporal record of
oceanic sulphate concentration and the global sulphur cycle
can be firmly placed.
Recent accounts of evaporite formation have been
presented, among others, by Schreiber et al. (2007).
Although we generally tend to associate evaporite deposits
with a marine origin, it is clear that evaporites also form in
non-marine continental settings. Yet, it appears that the
definition of easily applicable criteria that unequivocally
allow to distinguish a marine from a non-marine origin for
a given evaporite deposit are difficult to develop. Size,
sediment texture, mineralogy and geochemical/isotopic
parameters of the evaporite deposit itself have all been
discussed, yet a conclusive answer appears difficult.
A rather simplistic approach can be developed from con-
sidering the chemical composition of the modern global
ocean. In respect to its major dissolved ions (Na+, Mg2+,
Ca2+, K+, Cl�, SO42�), the modern ocean exhibits an
extremely conservative composition that is expressed in a
homogenous lateral and vertical distribution of ocean salin-
ity at 35‰. Even considering temporal variations in salinity
(e.g. Hay et al. 2006) and chemical composition (e.g. Hardie
1996), evaporite deposits that would form from a homoge-
nous seawater composition at a given time should be roughly
identical in its mineralogy and chemical composition. Here,
H. Strauss (*)
Institut f€ur Geologie und Pal€aontologie, Westf€alische Wilhelms-
Universit€at, Corrensstr. 24, 48149 M€unster, Germany
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_5, # Springer-Verlag Berlin Heidelberg 2013
1170
1170 H. Strauss et al.
evaporite composition will be primarily determined by solu-
bility and the degree of evaporation. Still, differences in
facies development and lateral/vertical architecture of a
given evaporite deposit may result. Differences could be
expected (Schreiber et al. 2007) depending on whether
evaporite formation occurred in basins that developed on a
tectonically stable continental platform (laterally consistent
lamination) or whether it occurred in tectonically active
settings, such as active rift systems (mechanically reworked
and/or highly deformed evaporites).
In contrast, non-marine evaporite deposits display a
rather variable chemical and mineralogical composition
(Warren 1989) that is strongly dependent on the highly
differing composition of brines that develop in continental
settings (e.g. the presence of sodium carbonate in many non-
marine deposits). Consequently, time equivalent evaporite
deposits would be grossly different in their chemical and
mineralogical composition. Hence, evidence of comparable
mineralogical, chemical, and isotopic composition for
multiple time equivalent evaporite deposits would be an
excellent way to distinguish marine from non-marine evap-
orite deposits. This approach fails, however, when only a
single evaporite deposit exists for a given time interval
and/or when the original evaporitic minerals have been
replaced subsequently during diagenesis (pseudomorphic
growth). In particular, in such a case, evidence from the
individual evaporite deposit can and should be supplemented
by evidence from other stratigraphically related sediments
(e.g. carbonate rocks within the same succession). Detailed
facies analysis of these might then allow a more robust
reconstruction of the depositional environment, including
the evaporitic unit.
Acknowledging the difficulties in directly constraining
oceanic sulphate abundance from the pure presence of
evaporite deposits, their size, sediment texture and/or miner-
alogy, researchers have turned to geochemical proxy signals.
Of particular importance is the stable sulphur (and oxygen)
isotopic composition of sulphate. Key to this is the fundamen-
tal observation that the two principle transfer functions of
sulphate from the ocean into the sedimentary archives (i.e.
evaporite precipitation and bacterial sulphate reduction plus
iron sulphide precipitation) are associated with characteristic
fractionations of their stable sulphur isotopic composition.
While bacterial sulphate reduction is characterised by a sub-
stantial shift in the sulphur isotopic composition discriminat-
ing against the heavy 34S isotope (e.g. Canfield 2001),
evaporite precipitation is accompanied by only a negligible
isotope effect (e.g. Claypool et al. 1980). Consequently, the
sulphur isotopic composition of ambient oceanic sulphate is
archived in evaporitic minerals, and evaporites can be
(and have traditionally been) utilised for reconstructing the
temporal evolution in seawater sulphate sulphur isotopic com-
position (e.g. Claypool et al. 1980; Strauss 1997). Again,
based on the very homogenous sulphur isotope record of
modern oceanic sulphate, it can be expected that multiple
time equivalent evaporite deposits exhibit a uniform isotopic
signature. This would then be representative for global sea-
water sulphate at that time (e.g. Claypool et al. 1980). In more
recent years (e.g. Kampschulte and Strauss 2004), and
acknowledging the fact that preserved evaporite occurrences
provide a poor time resolution and a highly fragmentary
sulphur isotope record, researchers have turned to another
proxy for recasting the sulphur isotopic composition of
ancient seawater sulphate: carbonate-associated sulphate.
During carbonate precipitation, the sulphate ion is substituted
into the calcite crystal lattice at quantities of a few 10 to
several 1,000 ppm. This way, the sulphur (and oxygen) isoto-
pic compositions of ambient oceanic sulphate are archived in
the sedimentary rock record. Furthermore, sulphate is
incorporated into marine barite, and Paytan et al. (1998,
2004) provide a high-resolution sulphur isotope record
based on marine barite for the Cenozoic and the Cretaceaous.
Considering that input and output functions into/from the
global sulphur cycle are characterised by diagnostic sulphur
isotope values, we can apply the simplified view of the
global sulphur cycle (Fig. 7.50) for recasting shifts in the
operation of the global sulphur cycle through time. The basis
for a respective analysis of temporal variations in the global
sulphur cycle is an isotope mass balance:
d34Sinput ¼ f sulphided34Ssulphide þ 1� f sulphide
� �d34Ssulphate
with fsulphide reflecting the proportional fraction of sulphur
being buried in the sedimentary record as sulphide (i.e. the
output function of reduced sulphur), and the different d34Svalues representing the isotopic compositions of the differ-
ent forms of sulphur in and out of the reaction chamber,
i.e. the oceanic dissolved sulphate reservoir. Based on
the sulphur isotope mass balance, modeling of the respective
sulphur isotope time series for sedimentary sulphate and
sulphide (Garrels and Lerman 1984; Kump 1989; Simon
et al. 2007; Prokoph et al. 2008; Godderis and Veizer
2000) govern our understanding of oceanic sulphate concen-
tration and evaporite deposition during the Phanerozoic. In
simple terms, during times when the sulphur isotopic com-
position of seawater sulphate is high (such as in the early
Palaeozoic), significant amounts of dissolved sulphate were
transferred out of the oceanic reservoir and stored in the
sedimentary realm as biogenic pyrite. Conversely, less posi-
tive d34S values (such as in the Permo-Carboniferous) reflect
a smaller proportion of pyrite burial, but more so a higher
proportion of pyrite weathering to the sulphate input from
continental weathering. Given appropriate environmental
conditions, like in the Permo-Carboniferous, a resulting
high sulphate abundance would be a favorable precondition
for evaporite deposition.
5 7.5 Abundant Marine Calcium Sulphates: Radical Change of Seawater Sulphate Reservoir 1171
The validity of this approach can be tested by comparing
the temporal evolution of seawater sulphate concentration
as derived from modeling the isotope time series with
data for sulphate occurrences and/or dissolved sulphate
concentrations in fluid inclusions. And even though the
temporal distribution of the latter is sparse and unevenly
distributed over the past 545 Ma, respective data, i.e.
sulphate concentration in fluid inclusions, agree reasonably
well (e.g. Horita et al. 2002).
Provided a marine origin has been established for a given
evaporite deposit, the sulphate sulphur isotopic composition
provides a proxy signal not only for the sulphur isotopic
composition of ancient oceanic sulphate but with it also
acts as a reflection of temporal variations in the global
sulphur cycle. Based on the observations pertaining to the
Phanerozoic, we will now analyse the record of Precambrian
seawater sulphate and the implications for the evolution of
the global sulphur cycle through time.
1172 H. Strauss et al.
7.5.3 Isotopic Evidence for PrecambrianOceanic Sulphate Abundance
Harald Strauss, Marlene Reuschel, Anthony E.Fallick, Victor A. Melezhik, and Aivo Lepland
For discussing Precambrian oceanic sulphate, its origin and
abundance, and in particular the proposed substantial rise in
oceanic sulphate abundance during the Palaeoproterozoic,
we will consult the respective sulphur isotope records
(d34S) of Precambrian sedimentary sulphate and sulphide
(Fig. 7.51).
The record of Archaean sulphate occurrences commences
with barite deposits broadly 3.5 Ga in age in Western
Australia and southern Africa and 3.2 Ga barites in India
(Huston and Logan 2004). Despite the generally accepted
view of an oxygen-free Archaean atmosphere (e.g. Holland
1999, 2006; but see Ohmoto 1999; Ohmoto et al. 2006, for a
different view), these barite occurrences were regarded by
many as representing local oxygen oases that “obviously”
allowed locally limited oxidative sulphur cycling and
subsequent evaporitic precipitation of original calcium
sulphate minerals (as initially discussed by Lambert et al.
1978). Alternative views invoking a hydrothermal origin,
have been proposed for the Australian deposits by Buick
and Dunlop (1990) and van Kranendonk et al. (2008). It has
long been noted that all Archaean barite occurrences exhibit
a rather narrow range in d34S between +3 and 6‰. Such a
narrow range in sulphur isotopic composition has always
been considered as evidence for a marine evaporite deposit,
referring to the equally homogenous isotopic nature of most
Phanerozoic examples and the homogenous isotopic compo-
sition of modern oceanic sulphate. Multiple sulphur isotope
research over the past 10 years (e.g. Farquhar et al. 2000;
Bao et al. 2007), however, completely changed this view.
Most barites (and by inference the dissolved oceanic
sulphate), but more so the abundant sedimentary sulphides
of Archaean and early Palaeoproterozoic age (<2400 Ma),
display clearly mass-independently fractionated sulphur
isotopes. This distinct signature is regarded as reflecting
the photochemical cycling of volcanogenic sulphur dioxide
in an oxygen-free atmosphere. Initially expressed by
Farquhar et al. (2000), the mass-independent sulphur isotope
signature recorded in Archaean and early Palaeoproterozoic
sedimentary sulphates and sulphides provide evidence for
the atmospheric origin of these sedimentary sulphur
compounds. Moreover, these authors argued for a strong
atmospheric influence on the global sulphur cycle, decidedly
different from the later one that is governed by oxidative
weathering of sulphides and microbial turnover of oceanic
sulphate. With respect to the Archaean barite occurrences of
Western Australia, South Africa and India, it is interesting to
note their largely comparable multiple sulphur isotopic com-
position (i.e., d34S, D33S, and D36S). Deviations were noted
by Ueno et al. (2008) for a subset of barite samples from
Western Australia. In addition to their bedded and vein-type
barites displaying mass-independent sulphur isotope frac-
tionation, hence, an atmospheric signature, these authors
report the presence of an additional, likely magmatic
sulphate that displays mass-dependent sulphur isotope
values.
Little information is available with respect to the sulphate
sulphur isotopic composition of Meso- and Neoarchaean age.
Domagal-Goldman et al. (2008) reported d34S and D33S
values for carbonate-associated sulphate from nine samples
ranging in age between 3000 and 2500 Ma. Trace amounts
of sulphate were extracted from stromatolitic and non-
stromatolitic carbonates, one dolomitic black chert and one
siltstone. The authors acknowledge that the sulphate yield was
rather low and that it may represent a mixture of primary
sulphate and sulphate from pyrite oxidation. Based on this
assessment, the implication of these results in respect to a
potential seawater signature remains to be validated and data
will, hence, not be considered during further discussion.
In contrast to the limited sulphate sulphur isotope
record, a voluminous record of traditional (d34S) and a
sizeable record of multiple (D33S, D36S) sulphur isotope
data exist for Precambrian sedimentary sulphides. These
have formulated our basic understanding concerning the
Precambrian global sulphur as a whole with more recent
reviews, e.g. provided by Strauss (2002), Canfield (2004)
or Lyons and Gill (2010). While numerous critical
issues are still waiting to be uncovered, some conclusions
pertaining to the present discussion about oceanic sulphate
abundance can be drawn from these multiple sulfide sul-
phur isotope records.
Starting with the discovery paper published by Farquhar
et al. (2000), multiple sulphur isotope research performed on
Precambrian sedimentary sulphides over the past 10 years
clearly reveal a distinct change from mass-independently
fractionated sulphur (MIF-S) in the Archaean and early
Palaeoproterozoic to solely mass-dependently fractionated
sulphur (MDF-S) isotopes thereafter (Fig. 7.52a; see also
Chap. 7.1). The distinct change occurred earlier than 2.32 Ga
ago (Bekker et al. 2004) and is archived in key sections in
North America (Papineau et al. 2007) and southern Africa
(Guo et al. 2009). The clearly structured temporal record of
the MIF-S signature, i.e. its ubiquitous presence in rocks
older than 2.4 Ga and its complete absence thereafter,
suggested a strong link to the absence/presence of atmo-
spheric oxygen. It is now commonly accepted that MIF-S
H. Strauss (*)
Institut f€ur Geologie und Pal€aontologie, Westf€alische Wilhelms-
Universit€at, Corrensstr. 24, 48149 M€unster, Germany
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_5, # Springer-Verlag Berlin Heidelberg 2013
1173
5 7.5 Abundant Marine Calcium Sulphates: Radical Change of Seawater Sulphate Reservoir 1173
is a result of UV-induced photochemical reaction of
volcanogenic sulphurous compounds (most notably sulphur
dioxide) in the atmosphere. Associated with this process is a
mass-independent fractionation of the sulphur isotopes,
provided the ambient level of oxygen in the atmosphere
does not exceed 10�5 of the present atmospheric level
(PAL; Pavlov and Kasting 2002). Hence, the presence of
the MIF-S signature in the sedimentary sulphur reservoir on
the early Earth (>2.4 Ga) attests to a low abundance of
atmospheric oxygen. Consequently, dissolved sulphate in
the early oceans was likely of atmospheric origin and, in
strong contrast to the modern ocean, not a product of conti-
nental oxidative weathering and riverine delivery.
The termination of the MIF-S signature in the early
Palaeoproterozoic is attributed to an increase in atmospheric
oxygen from <10�5 to >10�2 PAL (Fig. 7.52c; Pavlov and
Kasting 2002). Profound changes in Earth surface
environments resulted as a consequence of this rise in atmo-
spheric oxygen, among them most likely an increase in the
concentration of oceanic sulphate. Evidence for this is
apparent from the record of mass-dependently fractionated
sulphur recorded in sedimentary sulphides.
As noted earlier, the microbial turnover of sulphate is
associated with a mass-dependent fractionation of the sul-
phur isotopes (e.g. Canfield 2001). More precisely, sulphate
reducing bacteria discriminate against the heavy 34S isotope.
The resulting sulphide (subsequently archived as iron sul-
phide in the sediment) displays variable but frequently neg-
ative d34S values. The magnitude of isotopic fractionation
depends on the type of organism as much as on the physico-
chemical boundary conditions. Data from natural habitats
indicate that the isotopic fractionation between sulphate and
sulphide is on average between 10‰ and 30‰, yet smaller
but more importantly also larger fractionations have been
reported (e.g. Wortmann et al. 2001; Werne et al. 2003;
Canfield et al. 2010). Culture experiments revealed a maxi-
mum isotopic fractionation of 45‰ (e.g. Detmers et al.
2001), although a fractionation of up to 66‰ has recently
been reported for a single strain of sulphate reducers (Sim
et al. 2011). An isotopic fractionation larger than 45‰ has
also been proposed on theoretical grounds (Brunner and
Bernasconi 2005). On the other hand, greatly attenuated
sulphur isotopic fractionations appear to be characteristic
for low-sulphate conditions (cf. Habicht et al. 2002). This
has been considered to support the view of a low-sulphate
Archaean and early Palaeoproterozoic ocean followed by a
protracted evolution towards higher sulphate concentrations
(Kah et al. 2004).
Although being highly dependent on environmental
conditions, the magnitude in isotopic fractionation between
sulphate and sulphide (Fig. 7.52b) may provide at least some
information in respect to oceanic sulphate abundance. Only
small deviations from 0‰ have been recorded for Archaean
sedimentary sulphides older than 2.7 Ga (e.g. Strauss 2002,
2003). Except for the early Archaean barite occurrences,
evidence for the sulphur isotopic composition of sedimentary
sulphate (of presumed marine origin) is lacking. Accepting
the sulphur isotopic composition (d34S) of these barites as
representative for early Archaean seawater sulphate, the
apparently small magnitude in isotopic fractionation between
sulphate and sulphide and the lack of a substantial deviation
from 0‰ observed for most of Archaean time have been
interpreted as evidence for bacterial sulphate reduction in an
ocean with an extremely low abundance of sulphate of
<200 mM (Fig. 7.52d; Habicht et al. 2002). Alternatively,
bacterial sulphate reduction was of little importance for the
sulphur cycling in Archaean Earth surface environments (e.g.
Strauss 2003). The only exception showing a substantial
fractionation in sulphur isotopes are microcrystalline
sulphides associated with some of the Archaean barite
deposits (e.g. Shen et al. 2001, 2009; Philippot et al. 2007).
Respective fractionations suggest microbial turnover of
sulphate and/or elemental sulphur, although details are being
discussed rather controversially. Between 2.7 and 2.4 Ga, the
total variability in d34S for sedimentary sulphide amounts
to more than 30‰ including d34S values as low as �19.9‰(e.g. Grassineau et al. 2001). These have been interpreted as
evidence for microbial turnover of sulphate, even without any
evidence for the sulphur isotopic composition of seawater
sulphate. Undisputed evidence for bacterial sulphate reduc-
tion, based on strongly negative d34S values as low as
�34.7‰ (e.g. Bekker et al. 2004) appears in the sedimentary
rock record around 2.3 Ga ago. Acknowledging a sulphur
isotopic composition between +10 and +20‰ (e.g. Schr€oder
et al. 2008; Guo et al. 2009) as representative for oceanic
sulphate at that time, a maximum isotopic fractionation
between 40 and 50‰ approaches a magnitude mostly
observed in younger and modern marine settings. While the
small isotopic fractionation associated with most Archaean
sediments has been considered as (somewhat circumstantial)
evidence for a low-sulphate ocean (<200 mM), early Palaeo-
proterozoic sedimentary rocks exhibiting a substantial sulphur
isotopic fractionation suggest a significantly higher sulphate
concentration (certainly in the lower mM range). Considering
the timing, this increase in sulphate abundance and therefore
sulphate availability for microbially driven redox processes
appears to be linked to the rise in atmospheric oxygen
abundance.
Most recently, Reuschel et al. (2012) provided a first
account for the sulphur isotopic composition of sulphate
from the c. 2.1 Ga Tulomozero Formation, Onega Basin,
Russia. Ex-situ (n ¼ 9) and in-situ (n ¼ 91) sulphur isotope
analyses of carbonate-associated sulphate, breccia-hosted
sulphate and pseudomorphs after Ca-sulphate containing
anhydrite and barite inclusions yielded a rather narrow
range in d34S between þ7.8 and þ11.3‰ (accepting one
1174 H. Strauss et al.
outlier atþ15.8‰). Samples were spread throughout a strat-
igraphic thickness of more than 500 m. In considering the
different forms of sulphate present and the homogenous
distribution in isotopic composition that was confirmed by
different analytical approaches, this range appears to be
characteristic for the seawater sulphate at the time of depo-
sition of the Tulomozero evaporites. More so, considering
the abundant calcium sulphate (present as pseudomorphs)
rather than chloride suggests widespread evaporite deposi-
tion and a modern-style evaporite sequence with carbonate
followed by sulphate and then chloride. Considering this
evidence in combination with a homogenous sulphate sul-
phur isotopic composition, the authors concluded that the
early Palaeoproterozoic ocean represented already a sizeable
sulphate reservoir of �2.5 mM.
5 7.5 Abundant Marine Calcium Sulphates: Radical Change of Seawater Sulphate Reservoir 1175
7.5.4 Multiple Sulphur Isotope Evidence forthe Early Palaeoproterozoic Rise in OceanicSulphate Abundance
Harald Strauss and Marlene Reuschel
Detailed geochemical evidence for the time of rising oceanic
sulphate abundance and the causal relationship with the rise in
atmospheric oxygen is provided by Guo et al. (2009). These
authors provide a coherent dataset of multiple sulphur
isotopes (d34S, D33S, D36S) obtained from carbonate-
associated sulphate extracted from early Palaeoproterozoic
carbonate rocks from the Duitschland Formation, Transvaal
Supergroup, deposited on the Kaapvaal Craton, South Africa,
supplemented with multiple sulphur isotope data for pyrite
and carbonate carbon and oxygen isotope results from the
very same samples (see Fig. 7.3 in Chap. 7.1). This high-
resolution multiple sulphur isotope data clearly reveal the
termination of MIF-S, in parallel with an increase in the
magnitude of mass-dependent sulphur isotope fractionation.
This latter change to strongly positive d34S values for
carbonate-associated sulphate in the upper Duitschland For-
mation (caused by the kinetic isotope effect associated with
bacterial sulphate reduction) indicates the enhanced microbial
turnover of an emerging sulphate reservoir. A concomitant
rise in d13C for carbonate carbon points to an increase in
organic carbon burial and a respective release of oxygen to
the early Palaeoproterozoic atmosphere (cf. Hayes et al. 1983,
but see Hayes and Waldbauer 2006, and Fallick et al. 2008,
2011, for alternative views). This oxygen release must have
enhanced oxidative weathering on the continents, flushing
dissolved sulphate from pyrite oxidation into the early
Palaeoproterozoic ocean. Moreover, increasing sulphate
availability would have stimulated bacterial sulphate reduc-
tion. The fact that the sedimentary sulphides studied by Guo
et al. (2009) exhibit equally positive d34S values, resulting in asmall isotopic difference between parental sulphate and
resulting sulphide, suggests a rather limited sulphate reservoir
and its nearly complete microbial turnover.
From the multiple isotope study of Guo et al. (2009) it can
be firmly concluded that as a consequence of the Great Oxi-
dation Event, the abundance of oceanic sulphate increased in
the early Palaeoproterozoic. This dissolved sulphate, although
still lower in abundance than today, resulted undoubtedly
from oxidative continental weathering, marking an important
turning point in the global sulphur cycle. Moreover, these
results for the Duitschland Formation clearly underline the
coupling of the carbon and sulphur isotope cycles along this
transition from a largely anoxic to an oxygen-dominated
world.
H. Strauss (*)
Institut f€ur Geologie und Pal€aontologie, Westf€alische Wilhelms-
Universit€at, Corrensstr. 24, 48149 M€unster, Germany
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_5, # Springer-Verlag Berlin Heidelberg 2013
1176
1176 H. Strauss et al.
7.5.5 Physical Evidence for Abundant OceanicSulphate in the Palaeoproterozoic
Victor A. Melezhik, Anthony E. Fallick,Harald Strauss, Aivo Lepland, andDmitry V. Rychanchik
A recent comprehensive compilation of Precambrian evap-
orite occurrences was presented by Evans (2006) with the
objective of utilising these sediments and their distinct
depositional conditions, i.e. evaporitic conditions
prevailing in the arid latitude belt between 15� and 35�, inhis assessment of orbital obliquity and Earth’s geomagnetic
field throughout the Proterozoic. The record extends back
to 2.25 Ga, Evans’ oldest example of an evaporite occur-
rence of (reconstructed) basin-wide scale, for which reli-
able palaeomagnetic information can be gathered. A total
of 21 deposits, each of a (reconstructed) great thickness and
basin-wide scale are listed in Table 7.2, and the reader is
referred to Evans (2006) for further details and respective
references. As a general conclusion from this compilation,
it can be noted that thick successions of preserved, (near-)
original evaporitic minerals such as anhydrite, gypsum and
halite are present only in several of the Neoproterozoic and
early Mesoproterozoic occurrences whereas the older
Mesoproterozoic and Palaeoproterozoic deposits have gen-
erally been identified on the basis of pseudomorphic
minerals after sulphate or chloride minerals or replacement
by magnesite. However, respective pseudomorphic mineral
replacements are present throughout thick sedimentary
successions of sometimes large areal extend, and the
subsequent section will provide a well illustrated account
of the physical evidence for abundant sulphates of
Palaeoproterozoic age.
For the c. 2.1 Ga Tulomozero Formation on the
Fennoscandian Shield, Melezhik et al. (2005) documented
evidence for former evaporites in a rock record 500 m thick
and covering an area of >2,000 km2. A large variety of
dolomite and silica pseudomorphs after sulphates were
reported from a wide range of facies, and probable halites
from inferred playa-lake deposits in the Tulomozero Forma-
tion (Fig. 7.53).
Pear-shaped fans of vertically radiating dolomite crystals
grown upward on an erosional surface and draped by the
next overlying mudstone layer represent a case of
syndepositional origin of bottom-grown former sulphate
blades (Fig. 7.53a). Another example of syndepositionally-
grown former sulphates is a series of sub-spherical nodules
that nucleated on an erosional surface, separated by and
draped with thin clayey material (Fig. 7.53b). On a bedding
plane, the nodular masses form an irregular, clay-draped,
hummocky surface.
Dolomite and quartz pseudomorphs after gypsum crystals
are the most abundant expression of former sulphates. They
occur in a variety of shapes. Rhomboidal and prismatic
crystals (typical for gypsum) hosted by dark brownmudstones
can be aligned sub-parallel to the bedding plane (Fig. 7.53c, d)
or randomly distributed in brown mudstone (Fig. 7.53e). The
prismatic crystals aligned sub-parallel to the bedding plane
were interpreted to represent originally gypsum hopper
crystals, formed by evaporation at the brine-air interface,
which settled down on the sediment surface (Melezhik et al.
2005). Crystals replaced by quartz retain numerous relicts of
anhydrite as microinclusions (Fig. 7.53f).
Dolomite and quartz pseudomorphs after twinned swal-
lowtail crystals and crystal rosettes (of probable gypsum) are
abundant in red and brown mudstones (Fig. 7.53g, h).
Although the twinned swallowtail crystals have been
deformed during compaction and, in places, tectonically
rotated from their primary vertical position, their preserva-
tion remains compatible with many examples reported from
younger unmetamorphosed rocks (e.g. Kendall 1984;
Demicco and Hardie 1994).
Lenticular or discoidal, single or coalescent crystals prob-
ably originally of gypsum, either aligned sub-parallel to the
bedding plane (Fig. 7.53i), or randomly distributed in dolo-
mitic marls with the sediment entrained within the crystals
(Fig. 7.53i–l) are features characteristic of displacively
grown gypsum crystals in modern marine and non-marine
evaporitic environments (Demicco and Hardie 1994).
Rounded nodules and nodular masses probably originally
of sulphate resembling “chicken-wire” anhydrite occur in
variegated claystones and mudstones (Fig. 7.53m–o). Such
nodular masses often contain relicts of soft-sediment
deformed mudstones (Fig. 7.53 m, o), implying formation
in unconsolidated sediments. Associated sedimentary rocks
are 0.5–2 mm dolomite-pseudomorphed sulphates–mud
couplets with former nodular sulphates and enterolithic
structure (Fig. 7.53p). This resembles bedded evaporites
reported from many supratidal gypsum pans of modern
sabkhas (Demicco and Hardie 1994). The bedded evaporites
imply evaporation at the brine-air interface of an ephemeral
or perennial brine pool (Hardie and Shinn 1986).
Dolomite- and quartz-pseudomorphed nodules are abun-
dant throughout the Tulomozero Formation (Fig. 7.53r–v).
Some nodules retain primary sedimentary structures of the
host sediments, thus having grown replacively; many show a
significant differential compaction with respect to the host
laminites (by a factor of 4), suggesting that the nodules
V.A. Melezhik (*)
Geological Survey of Norway, Postboks 6315 Sluppen, NO-7491
Trondheim, Norway
Centre for Geobiology, University of Bergen, Allegaten 41, Bergen
N-5007, Norway
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_5, # Springer-Verlag Berlin Heidelberg 2013
1177
5 7.5 Abundant Marine Calcium Sulphates: Radical Change of Seawater Sulphate Reservoir 1177
formed during early diagenesis (Fig. 7.53q, s). Many such
early diagenetic nodules and concretionary layers have par-
tially eroded/dissolved surfaces draped by mudstones
(Fig. 7.53w), implying that they formed in close proximity
to the surface, probably under shallow, ephemeral gypsum
pan conditions (Demicco and Hardie 1994 and references
therein). Thus the combined observational evidence indi-
cates that many of the Tulomozero Formation sulphates
formed syndepositionally, either during sedimentation or,
slightly thereafter and well before burial (Melezhik et al.
2005).
Gee and Grey (1993) and El Tabakh et al. (1999)
published evidence for evaporites of roughly the same
age from Western Australia. Quartz pseudomorphs after
evaporitic calcium sulphate minerals (Fig. 7.54a–c) are
widespread in a succession of shallow-marine to tidal-flat
stromatolitic carbonates and siliciclastic sedimentary rocks
of the Bubble Well Member (Juderina Formation, Windplain
Group) from the Palaeoproterozoic Yerrida Basin, Western
Australia (Gee and Grey 1993; El Tabakh et al. 1999;
Pirajno et al. 1998, 2004). Stromatolitic carbonates from
the Bubble Well Member were found to be isotopically
heavy with d13C values ranging from þ5 to þ9‰ (Russell
1992; Lindsay and Brasier 2002), characteristic of the
Lomagundi-Jatuli isotopic excursion developed worldwide
at c. 2200–2060 Ma (see Chap. 7.3). A depositional age of
2170 � 64Ma for the Bubble Well Member is provided by a
Pb-Pb isochron obtained from a sample of stromatolitic
carbonate (Woodhead and Hergt 1997), consistent with
deposition during the period of the Lomagundi-Jatuli isoto-
pic excursion. Evaporite pseudomorphs with rare remnant
calcium sulphate crystals occur in association with variably
silicified carbonate rocks including stromatolitic carbonates,
laminated microbialite, and dolarenites. Evaporitic calcium
sulphate crystal mush domes have developed in some car-
bonate beds, consistent with supratidal conditions
(Fig. 7.54c). Evidence of primary gypsum and anhydrite
has been inferred from silica-replaced pseudomorphs of
great morphological variety, including botryoidal nodules
(Fig. 7.54a), needle-shaped laths, bladed crystals with pyra-
midal terminations, swallow-tail forms and rosettes
(Fig. 7.54b) (Gee and Grey 1993; El Tabakh et al. 1999).
Disruption of stromatolite laminae by growth of evaporite
crystals indicates nucleation from interstitial brines prior to
the lithification of stromatolites, possibly shortly after sedi-
mentation or during early diagenesis (El Tabakh et al. 1999).
Several Palaeoproterozoic formations in North America
are known to contain evidence of former Ca-sulphates.
Pseudomorphs after gypsum were reported by Pope and
Grotzinger (2003) from the c. 1.9 Ga Stark Formation
(northwest Canada). The c. 2.3–2.22 Gordon Lake Forma-
tion (Lake Huron, Canada) contains barite beds, silicified
and pristine anhydrite and gypsum nodules and layers
(Cameron 1983; Chandler 1988; Bekker et al. 2006).
Molds after anhydrite nodules and gypsum crystals were
documented in the c. 2.15 Ga Lower Nash Fork Formation
(Wyoming, USA) by Bekker and Eriksson (2003) and
Bekker et al. (2003). Pseudomorphs after gypsum and anhy-
drite in sandstones and 13C-rich dolostones are known in the
c. 2.3–2.22 Kona Dolomite (Michigan, USA) (Fig. 7.55).
In Zimbabwe, thinly bedded anhydrite-bearing dolomites
and argillites, and sulphate pseudomorphs are relatively
common in the c. 2.15 Ga Norah Formation, Deweras
Group, and also occur in the overlying Lomagundi Group
(Master et al. 2010). Schr€oder et al. (2008) reported
pseudomorphed marine sulphate evaporites containing relict
Ca-sulphate from the c. 2.2–2.1 Ga Lucknow Formation,
Transvaal Supergroup, South Africa, and argued for sul-
phate concentrations of >2.5 mM. In all cases sulphate
occurrences are associated with 13C-rich carbonates (for
references see Schr€oder et al. 2008).An occurrence of anhydrite beds and veins has been
known since the 1970s in the Fedorovo Formation (Aldan,
Russia) as evidence of Archaean sulphates (Vinogradov
et al. 1976). This formation has recently been dated as
Palaeoproterozoic (Velikoslavinsky et al. 2003), and
associated carbonates have been shown to be enriched in13C (Guliy and Wada 2003), consistent with their accumula-
tion during the Lomagundi-Jatuli isotopic excursion.
Finally, thick-bedded anhydrites (Morozov et al. 2010;
Krupenik et al. 2011a) were recently discovered when a c.
3,500-m-deep drillhole in the Onega Basin in the eastern
Fennoscandian Shield intersected 13C-rich dolostones of
Lomagundi-Jatuli age at a depth of 2,115 m (335 m thick)
followed by massive anhydrite and anhydrite-magnesite
rocks (c. 100 m thick), nodular shale interbedded with
massive anhydrite (190 m thick) and a c. 194-m-thick
halite formation (70–75% halite, 12–20% anhydrite, 10–15%
magnesite) containing large blocks (up to 1 m) of bedded,
coarse-grained anhydrite and magnesite (Fig. 7.56a–c). The
salts (Fig. 7.56d–e) appear to have formed prior to or syn-
chronously with 13C-rich dolostones of presumed
Lomagundi-Jatuli excursion age (see Melezhik et al. 2011).
The sulphur isotopic composition of this anhydrite ranges
from þ4‰ to þ8‰ (Krupenik et al. 2011b), which is
slightly lower than the results obtained from the coeval
Tulomozero Formation (Reuschel et al. 2012) and the
Lucknow Formation (Schr€oder et al. 2008).The substantial anhydrite occurrence in the Onega Basin,
Fennoscandian Shield, and the abundant occurrences of pseu-
domorphic sulphate evaporites in numerous and thick sedi-
mentary successions of Palaeoproterozoic age post-dating the
Great Oxidation Event, provide unequivocal evidence for a
sizeable oceanic sulphate reservoir that emerged as a conse-
quence of progressive continental oxidative weathering in the
aftermath of Earth’s atmospheric oxidation.
1178 H. Strauss et al.
7.5.6 A Radical Change of the SeawaterSulphate Reservoir: Implication of the FAR-DEEP Core
Harald Strauss and Aivo Lepland
FAR-DEEP intersected a crucial time interval in Earth history
that witnessed some of the most pronounced changes
in geology, climatic conditions, the chemical state of the
atmosphere–ocean system, and the evolution of life. As
outlined above, significant changes occurred in the redox
state during the early Palaeoproterozoic. Most importantly,
unequivocal evidence points to a substantial rise in atmo-
spheric oxygen abundance some 2.3 Ga ago (prominently
termed the Great Oxidation Event, cf. Holland 1999, 2006).
Although detailed causal relationships are still under discus-
sion, one of the immediate consequences of a substantial
increase in atmospheric oxygen would be the onset of oxida-
tiveweathering on the continents. Under the newly established
oxic surface conditions, redox sensitive minerals would
become unstable, most prominently iron sulphide. Water sol-
uble sulphate resulting from this oxidative breakdown on the
continents would be flushed from the catchments into the
rivers and ultimately delivered to the ocean. Consequently,
oceanic sulphate abundance would increase.
Physical and chemical evidence presented in the previous
sections suggest that the Palaeoproterozoic ocean represented
already a sizeable marine sulphate reservoir, although a pro-
posed minimum concentration of 2.5 mM is still substantially
below the sulphate concentration in the modern ocean. Yet,
ample evidence indicates that evaporites precipitated from
the Palaeoproterozoic ocean, including abundant sulphates
suggesting a modern-style evaporite deposition, given the
right environmental conditions. Based on our present knowl-
edge, the sedimentary succession on the Fennoscandian Shield
represents the prime example for abundant sulphate precipita-
tion from a seemingly homogenous oceanic sulphate reservoir.
FAR-DEEP Holes 10A, 10B, and 11A intersected the
Tulomozero Formation, a thick evaporitic succession that
bears evidence for this change in the chemical composition of
the ocean in the aftermath of the Great Oxidation Event. This
corematerial provides a unique opportunity to study all aspects
of Palaeoproterozoic evaporite precipitation through detailed
petrographic and geochemical work at high spatial resolution.
Fig. 7.50 Simplified view of the global sulphur cycle (After Strauss 1997)
H. Strauss (*)
Institut f€ur Geologie und Pal€aontologie, Westf€alische Wilhelms-
Universit€at, Corrensstr. 24, 48149 M€unster, Germany
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_5, # Springer-Verlag Berlin Heidelberg 2013
1179
5 7.5 Abundant Marine Calcium Sulphates: Radical Change of Seawater Sulphate Reservoir 1179
Fig. 7.51 The sulphur isotopic composition of Precambrian sedimentary sulphate and sulphide (After Thomazo et al. 2009)
1180 H. Strauss et al.
Fig. 7.52 Geochemical proxy signals showing (a) the termination of
mass-independent sulphur isotope fractionation, (b) a change in magni-
tude in isotope fractionation between sulphate and sulphide, (c) the
temporal evolution of atmospheric oxygen abundance as percent of
present atmospheric level (PAL), and (d) the temporal change in seawa-
ter sulphate concentration as percent of present oceanic level (POL)
5 7.5 Abundant Marine Calcium Sulphates: Radical Change of Seawater Sulphate Reservoir 1181
Table 7.2 Precambrian (pre-Ediacaran; >600 Myr) evaporite basins (Adapted from Evans 2006)
Evaporite basin Age (Myr)a Volume (km3)a
Skillogalee, Australia ~770 25,000
Curdimurka, Australia ~785 50,000
Kilian-Redstone River, Canada ~770 30,000
Minto Inlet, Canada ~800 90,000
Duruchaus, Namibia ~800 15,000
Copperbelt, Central Africa ~830 (?) 25,000
Centralian, Australia ~830 140,000
Borden, Canada ~1200 15,000
Char/Douik, West Africa ~1200 (?) 8,000
Belt, USA, Canada ~1460 10,000
Discovery, Australia ~1500 �2,800
Balbirini, Australia ~1610 2,500
Lynott, Australia ~1635 3,000
Myrtle, Australia ~1645 13,000
Mallapunyah, Australia ~1660 5,000
Corella, Australia ~1740 2,000
Stark, Canada ~1870 30,000
Rocknest, Canada ~1950 1,000
Juderina, Australia ~2100 1,000
Tulomozero, Russia ~2100 1,000
Chocolay, Canada-USA ~2250 4,500aNote: data provided by Evans (2006) with more information available in this article
1182 H. Strauss et al.
Fig. 7.53 Ca-sulphates in the c. 2.1 Ga Tulomozero Formation from the
Onega Basin, southeastern Fennoscandia. (a) Upward-radiating, dolomite-pseudomorphed sulphate crystal blades forming pear-shaped bodies grown
on dolarenite substrate and draped bymud layer (red arrowed); large,whitearrows show growth direction. (b) Bed of individual and coalesced former
sulphate nodules formed on the clayey dolostone passing downwards intobedded dolarenite; red arrows mark mud-draped, irregular surface of the
nodular bed overlain by dolarenite. (c) Scanned thin section with dolomite-
pseudomorphed, prismatic crystals of probable gypsum aligned sub-parallel
to the lamination of black, haematite-rich mudstone. (d) Polished core with
dolomite-pseudomorphed, prismatic crystals of probable gypsum in dark
brown mudstone; note that the crystals are aligned sub-parallel to the
bedding plane. (e) Polished core with quartz-pseudomorphed, discoidal
Ca-sulphate crystals in dark brown mudstone. (f) Back-scattered electron
image of rectangular, cleaved anhydrite crystals (white) from quartz-
pseudomorphed discoidal crystals shown in (e)
5 7.5 Abundant Marine Calcium Sulphates: Radical Change of Seawater Sulphate Reservoir 1183
Fig. 7.53 (continued) (g) Polished core slab with dolomite-
pseudomorphed rosettes of gypsum distorting lamination; many show
twinned swallowtail forms. (h) Polished core-slab showing dolomite-
pseudomorphed, tabular crystals of probable gypsum; some show
twinned swallowtail forms. (i) Sawn core of dolomitic marl with
dolomite-pseudomorphed, displacively-grown gypsum nodules and
crystals having a discoidal or lenticular morphology flattened normal
to c-axis; sulphate growth caused plastic distortion of primary lamina-
tion. (j) Sawn core of dolomitic marl crowded with dolomite-
pseudomorphed, displacively-grown, discoidal gypsum crystals
1184 H. Strauss et al.
Fig. 7.53 (continued) (k) A combination of lenticular, discoidal,
twinned swallowtail forms and irregular masses of gypsum entrapping
host sediment. (l) Cluster of dolomite-pseudomorphed gypsum crystals
of different shapes growing at the contact between dark brown mud-
stone and laminated siltstone; note that gypsum crystals distort primary
lamination. (m) Polished core-slab of dolomite-pseudomorphed, nodu-
lar masses resembling ‘chicken-wire’ anhydrite in soft-sediment
deformed, pink mudstone. (n) Sawn core exhibiting dolomite-
pseudomorphed, enterolithic layer of probable nodular gypsum or
anhydrite
5 7.5 Abundant Marine Calcium Sulphates: Radical Change of Seawater Sulphate Reservoir 1185
Fig. 7.53 (continued) (o) Scanned thin section showing ‘chicken-
wire’ anhydrite in soft-sediment deformed, pink mudstone. (p) Scanned
thin section showing dolomite-pseudomorphed sulphate–mud couplets
with enterolithic structure; probable bedded Ca-sulphate. (q) Scanned
thinsection of dolomite-pseudomorphed sulphate–mud couplets with
enterolithic structure and a nodule (red arrowed)
1186 H. Strauss et al.
Fig. 7.53 (continued) (r) Scanned thin section showing a dolomite-
pseudomorphed, replacively grown sulphate nodule in former bedded
evaporite; plastically-deformed host sediment laminae with enterolithic
structure (red arrows) continue across the nodule (black arrows)though showing significant differential compaction. (s) Scanned thin
section exhibiting a silica-pseudomorphed sulphate nodule in red mud-
stone showing prominent differential compaction. (t) Polished core-
slab with silica pseudomorphed, coalesced former sulphate nodules
containing abundant relicts of Ca-sulphate. (u) Polished core slab
with silica-pseudomorphed, Ca-sulphate crystals and nodules emplaced
into dark brown mudstone with a clotted fabric; both crystals and
nodules contain abundant relicts of Ca-sulphate (v) Polished core slab
with silica-pseudomorphed, sulphate nodules coalesced into nodular
mass retaining abundant relicts of Ca-sulphate. (w) Scanned thin sec-
tion of mud and dolomite-pseudomorphed sulphate laminae with
enterolithic structure and nodules; some nodules display partial ero-
sion/dissolution (red arrowed) and draping by dark mudstones
(Photographs (a, b, d–i, n, o, r, s, and w) reproduced from Melezhik
et al. (2005) with permission of Blackwell Publishing Ltd., photographs
(c, j–m, p, q, t–v) by Victor Melezhik)
5 7.5 Abundant Marine Calcium Sulphates: Radical Change of Seawater Sulphate Reservoir 1187
Fig. 7.54 Polished rock slab showing gypsum crystals in carbonate
from the c. 2.2 Ga Yerrida Basin, Western Australia. (a) Polished slab
of sandstone-siltstone with flaser and lenticular bedding (lower half)
overlain by red, laminated mudstone with botryoidal quartz, interpreted
as replacements of anhydrite nodules (white). (b) Polished slab
showing crystal habits of silica-replaced Ca-sulphate evaporates
including needle-shaped laths, bladed crystals, swallow-tail forms and
rosettes in dolostone. (c) Outcrop photograph of a crystal mush dome
containing abundant silicifield Ca-sulphate pseudomorps in a carbonate
matrix; knife length is 7 cm (Photographs by Aivo Lepland, rock slabs
were made available by Kathleen Grey)
1188 H. Strauss et al.
Fig. 7.55 Pseudomorphed Ca-sulphates from Kona Dolomite,
Michigan, USA. (a) Dolomite-pseudomorphed apparent gypsum
crystals in pink and white dolarenite alternating with quartz sandstone
(grey) and siltstone (dark grey) passing upward into dolostone breccia.
(b) Weathered-out gypsum crystals in sandstone (Samples courtesy of
Bouke Zwaan, photographs by Victor Melezhik)
5 7.5 Abundant Marine Calcium Sulphates: Radical Change of Seawater Sulphate Reservoir 1189
Fig. 7.56 Halite, magnesite and anhydrite in the Tulomozero formation
recovered by the Onega parametric drillhole. (a) Lithological column of
the Onega Basin formations drilled by the Onega hole (Modified after
Morozov et al. 2010; Krupenik et al. 2011a). (b) Unsawn cores of
massive anhydrite; scale-bar in centimetres
1190 H. Strauss et al.
Fig. 7.56 (continued) (c) Sawn cores demonstrating a massive texture
of anhydrite from 2,516 to 2,507 m depth interval. (d) Core exhibiting
halite (dark grey) partially dissolved by drilling fluid; the halite is
capped by an anhydrite bed with cavities left after dissolved halite.
(e) Core of halite partially dissolved by drilling fluid; note numerous
inclusions of anhydrite and magnesite. (f) Sulphur isotopic composition
of massive anhydrite; data are from Krupenik et al. (2011b), in red, and
from Rychanchik and Fallick (unpublished), in black (Sample courtesy
of the Institute of Geology, Karelian Science Centre, photographs by
Dmitry Rychanchik)
5 7.5 Abundant Marine Calcium Sulphates: Radical Change of Seawater Sulphate Reservoir 1191
References
Bao H, Rumble D, Lowe DR (2007) The five stable isotope
compositions of Fig Tree barites: implications on sulfur cycle in
ca. 3.2 Ga oceans. Geochim Cosmochim Acta 71:4868–4879
Bekker A, Eriksson KA (2003) Paleoproterozoic drowned carbonate
platform on the southeastern margin of the Wyoming Craton: a
record of the Kenorland Breakup. Precambrian Res 120:327–364
Bekker A, Karhu JA, Eriksson KA, Kaufman AJ (2003) Chemostra-
tigraphy of Paleoproterozoic carbonate successions of theWyoming
Craton: tectonic forcing of biogeochemical change? Precambrian
Res 120:279–325
Bekker A, Holland HD, Wang PL, Rumble D, Stein HJ, Hannah JL,
Coetzee LL, Beukes NJ (2004) Dating the rise of atmospheric
oxygen. Nature 427:117–120
Bekker A, Karhu JA, Kaufman AJ (2006) Carbon isotope record for the
onset of the Lomagundi carbon isotope excursion in the Great Lakes
area, North America. Precambrian Res 148:145–180
Brunner B, Bernasconi SM (2005) A revised isotope fractionation
model for dissimilatory sulfate reduction in sulphate reducing bac-
teria. Geochim Cosmochim Acta 69:4759–4771
Buick R, Dunlop JSR (1990) Evaporitic sediments of early Archean age
from the Warrawoona Group, North Pole, Western Australia. Sedi-
mentology 37:247–277
Busenberg E, Plummer LN (1985) Kinetic and thermodynamic factors
controlling the distribution of SO42� and Na+ in calcites and
selected aragonites. Geochim Cosmochim Acta 49:713–725
Cameron EM (1983) Evidence from early Proterozoic anhydrite for
sulphur isotopic partitioning in Precambrian oceans. Nature
304:54–56
Cameron EM, Hall GEM, Veizer J, Krouse HR (1995) Isotopic and
elemental hydrogeochemistry of a major river system: Fraser River,
British Columbia, Canada. Chem Geol 122:149–169
Canfield DE (2001) Biogeochemistry of sulphur isotopes. In: Valley
JW, Cole DR (eds) Stable isotope geochemistry, Reviews in miner-
alogy and geochemistry, vol. 53. Geological Society of America,
Washington, DC, pp 607–633
Canfield DE (2004) The evolution of the Earth surface sulfur reservoir.
Am J Sci 304:839–861
Canfield DE, Farquhar J, Zerkle AL (2010) High isotope fractionations
during sulfate reduction in a low-sulfate euxinic ocean analog.
Geology 38:415–418
Chandler FW (1988) Diagenesis of sabkha-related, sulphate nodules in
the Early Proterozoic Gordon Lake Formation, Ontario, Canada.
Carbon Evapor 3:75–94
Claypool GE, Holser WT, Kaplan IR, Sakai H, Zak I (1980) The age
curves of sulfur and oxygen isotopes in marine sulfate and their
mutual interpretation. Chem Geol 28:190–260
Demicco RV, Hardie LA (1994) Sedimentary structures and early
diagenetic features of shallow marine carbonate deposits. Society
of Sedimentary Geologists, Tulsa, p 265
Detmers J, Br€uchert V, Habicht KS, K€uver J (2001) Diversity of sulfur
isotope fractionations by sulfate-reducing prokaryotes. Appl Envi-
ron Microbiol 67:888–894
Domagal-Goldman SD, Kasting JF, Johnston DT, Farquhar J (2008)
Organic haze, glaciations and multiple sulphur isotopes in the Mid-
Archean Era. Earth Planet Sci Lett 269:29–40
El Tabakh M, Grey C, Pirajno F, Schreiber BC (1999) Pseudomorphs
after evaporitic minerals interbedded with 2.2 Ga stromatolites of
the Yerrida basin, Western Australia: origin and significance. Geol-
ogy 27:871–874
Evans DAD (2006) Proterozoic low orbital obliquity and axial-dipolar
geomagnetic field from evaporite palaeolatitudes. Nature 444:51–55
Fallick AE, Melezhik VA, Simonson B (2008) The ancient anoxic
biosphere was not as we know it. In: Dobretsov N, Kolchanov N,
Rozanov A, Zavarzin G (eds) Biosphere origin and evolution.
Springer, New York, pp 169–188
Fallick AE, Melezhik VA, Simonson B (2011) On Proterozoic
ecosystems and the carbon isotopic composition of carbonates
associated with banded iron formations. In: Neves L et al (eds)
Modelacao de Sistemas. Geologicos, Universidade de Coimbra,
Portugal, pp 57–71
Farquhar J, Bao H, Thiemens M (2000) Atmospheric influence of
Earth’s earliest sulphur cycle. Science 289:756–758
Garrels RM, Lerman A (1984) Coupling of the sedimentary sulfur and
carbon cycles – an improved model. Am J Sci 284:989–1007
Gee RD, Grey K (1993) Proterozoic rocks on the Glengarry 1:250,000
sheet: stratigraphy, structure and stromatolite biostratigraphy. Geol
Surv West Aust Rep 41:30
Godderis Y, Veizer J (2000) Tectonic control of chemical and isotopic
composition of ancient oceans: the impact of continental growth.
Am J Sci 300:434–461
Grassineau NV, Nisbet EG, Bickle MJ, Fowler CMR, Lowry D,
Mattey DP, Abell P, Martin A (2001) Antiquity of the biological
sulphur cycle: evidence from sulphur and carbon isotopes in 2700
million year old rocks of the Belingwe belt, Zimbabwe. Proc R Soc
Lond B268:113–119
Grossman EL, Mii H-S, Zhang C, Yancey TE (1996) Chemical varia-
tion in Pennsylvanian brachiopod shells – diagenetic, taxonomic,
microstructural, and seasonal effects. J Sed Res 66:1011–1022
Guliy VN, Wada H (2003) Macro and microvariations of isotopic
composition of carbon and oxygen of carbonates from the Precam-
brian of the Aldan Shield. Geochemistry 5:482–491 (in Russian)
Guo Q, Strauss H, Kaufman AJ, Schr€oder S, Gutzmer J, Wing B, Baker
MA, Bekker A, Jin Q, Kim S-T, Farquhar J (2009) Reconstructing
Earth’s surface oxidation across the Archean-Proterozoic transition.
Geology 37:399–402
Habicht KS, Gade M, Thamdrup B, Berg P, Canfield DE (2002)
Calibration of sulfate levels in the Archean ocean. Science
298:2372–2374
Hardie LA (1996) Secular variation in seawater chemistry: an explana-
tion for the coupled secular variation in the mineralogies of marine
limestones and potash evaporites over the past 600 m.y. Geology
24:279–283
Hardie LA, Shinn EA (1986) Carbonate depositional environments
modern and ancient, part 3: tidal flats. Colo Sch Mines Q 81:1–74
Hay WW, Migdisov A, Balukhovsky AN, Wold CN, Fl€ogel S,
S€oding E (2006) Evaporites and the salinity of the ocean during
the Phanerozoic: implications for climate, ocean circulation and
life. Palaeogeogr Palaeoclimatol Palaeoecol 240:3–46
Hayes JM,Waldbauer JR (2006) The carbon cycle and associated redox
processes through time. R Soc Lond Philos Trans B Biol Sci B
361:931–950
Hayes JM, Kaplan IR, Wedeking KW (1983) Precambrian organic
geochemistry, preservation of the record. In: Schopf JW (ed)
Earth’s earliest biosphere: its origin and evolution. Princeton
University Press, Princeton, pp 93–134
Holland HD (1984) The chemical evolution of the atmosphere and
oceans, Princeton series in geochemistry. Princeton University
Press, Princeton, p 582
Holland HD (1999) When did the Earth’s atmosphere become oxic?
A reply. Geochem News 100:20–22
Holland HD (2006) The oxygenation of the atmosphere and oceans.
Philos Trans R Soc Lond B Biol Sci B361:903–915
Horita J, Zimmermann H, Holland HD (2002) Chemical evolution of
seawater during the Phanerozoic: implications from the record of
marine evaporates. Geochim Cosmochim Acta 66:3733–3756
Huston DL, Logan GA (2004) Barite, BIFs and bugs: evidence for the
evolution of the Earth’s early hydrosphere. Earth Planet Sci Lett
220:41–55
Jørgensen BB (1982) Mineralization of organic matter in the sea bed –
the role of sulphate reduction. Nature 296:643–645
1192 H. Strauss et al.
Kah LC, Lyons TW, Frank TD (2004) Low marine sulphate and
protracted oxygenation of the Proterozoic biosphere. Nature
431:834–838
Kampschulte A, Strauss H (2004) The sulphur isotopic evolution of
Phanerozoic seawater based on the analysis of structurally
substituted sulphate in carbonates. Chem Geol 204:255–286
Kendall AC (1984) Evaporites. In: Walker RG (ed) Facies models, vol 1,
2nd edn, Geoscience Canada reprint series. Geological Association of
Canada, Toronto, pp 259–296
Krupenik VA, Akhmedov AM, Sveshnikova KYu (2011a) Section of
the Onega structure based on data from OPH (Onega parametric
hole). In: Glushanin LV, Sharov NV, Shchiptsov VV (eds) The
Onega Palaeoproterozoic structure (geology, tectonics, deep struc-
ture and minerageny). Karelian Science Centre, Petrozavodsk, pp
172–189 (in Russian)
Krupenik VA, Akhmedov AM, Sveshnikova KYu (2011b) Carbon,
oxygen and sulphur isotopic composition of rocks from Ludicovian
and Jatulian Super-Horizons. In: Glushanin LV, Sharov NV,
Shchiptsov VV (eds) The Onega Palaeoproterozoic structure
(geology, tectonics, deep structure and minerageny). Karelian Sci-
ence Centre, Petrozavodsk, pp 250–255 (in Russian)
Kump LR (1989) Alternative modeling approaches to the geochemical
cycles of carbon, sulfur and strontium isotopes. Am J Sci
289:390–410
Lambert IB, Donnelly TH, Dunlop JSR, Groves DI (1978) Stable
isotopic compositions of early Archaean sulphate deposits of prob-
able evaporitic and volcanogenic origins. Nature 276:808–810
Lindsay JF, Brasier MD (2002) Did global tectonics drive early bio-
sphere evolution? Carbon isotope record from 2.6 to 1.9 Ga
carbonates of western Australian basins. Precambrian Res 114:1–34
Lyons TW, Gill BC (2010) Ancient sulfur cycling and oxygenation of
the early biosphere. Elements 6:93–99
Master S, Bekker A, Hofmann A (2010) A review of the stratigraphy
and geological setting of the Palaeoproterozoic Magondi Super-
group, Zimbabwe – type locality for the Lomagundi carbon isotope
excursion. Precambrian Res 182:254–273
Melezhik VA, Fallick AE, Rychanchik DV, Kuznetsov AB (2005)
Palaeoproterozoic evaporites in Fennoscandia: implications for sea-
water sulphate, the rise of atmosheric oxygen and local amplifica-
tion of the d13C excursion. Terra Nova 17:141–148
Melezhik VA, Fallick AE, Rychanchik DV, FAR-DEEP Drilling Team
(2011) Abundant marine sulphate in the Palaeoproterozoic: models
come and go, but the rock record endures. In: Goldschmidt confer-
ence, Prague, 14–19 Aug 2011
Morozov AF, Hahaev BN, Petrov OV, Gorbachev VI, Tarkhanov GB,
Tsvetkov LD, Erinchek YuM, Akhmedov AM, Krupenik VA,
Sveshnikova KYu (2010) Rock-salts in Palaeoproterozoic strata of
the Onega depression of Karelia (based on data from the Onega
parametric drillhole). Commun Acad Sci 435(2):230–233 (in
Russian)
Ohmoto H (1999) When did the Earth’s atmosphere become oxic?
Geochem News 93:12–27
Ohmoto H, Watanabe Y, Ikemi H, Poulson SR, Taylor BE (2006)
Sulphur isotope evidence for an oxic Archaean atmosphere. Nature
442:908–911
Papineau D, Mojzsis SJ, Schmitt AK (2007) Multiple sulphur isotopes
from Paleoproterozoic Huronian interglacial sediments and the rise
of atmospheric oxygen. Earth Planet Sci Lett 255:188–212
Pavlov AA, Kasting J (2002) Mass-independent fractionation of sul-
phur isotopes in Archean sediments: strong evidence for an anoxic
Archean atmosphere. Astrobiology 2:27–41
Paytan A, Kastner M, Campbell D, Thiemens MH (1998) Sulfur isotope
composition of Cenozoic seawater sulfate. Science 282:1459–1462
Paytan A, Mearon S, Cobb K, Kastner M (2002) Origin of marine barite
deposits: Sr and S isotope characterization. Geology 30:747–750
Paytan A, Kastner M, Campbell D, Thiemens MH (2004) Seawater
sulfur isotope fluctuations in the Cretaceous. Science
304:1663–1665
Philippot P, van Zuilen M, Lepot K, Thomazo C, Farquhar J,
Van Kranendonk MJ (2007) Early Archaean microorganisms
preferred elemental sulfur, not sulfate. Science 317:1534–1537
Pirajno F, Occhipinti SA, Swager CP (1998) Geology and tectonic
evolution of Palaeoproterozoic Bryah, Padbury and Yerrida
(formerly Glengarry Basin), Western Australia: implications for
the history of the south-central Capricorn Orogen. Precambrian
Res 90:119–140
Pirajno F, Jones JA, Hocking RM, Halilovic J (2004) Geology and
tectonic evolution of Palaeoproterozoic basins of the eastern Capri-
corn Orogen, Western Australia. Precambrian Res 128:315–342
Pope MC, Grotzinger JP (2003) Paleoproterozoic Stark Formation,
Athapuscow Basin, Northwest Canada; record of cratonic-scale
salinity crisis. J Sed Res 73:280–295
Prokoph A, Shields GA, Veizer J (2008) Compilation and time-series
analysis of a marine carbonate d18O, d13C, 87Sr/86Sr and d34Sdatabase through Earth history. Earth Sci Rev 87:113–133
Reuschel M, Melezhik VA, Lepland A, Fallick AE, Strauss H (2012)
Isotopic evidence for a sizeable seawater sulfate reservoir at 2.1 Ga.
Precambrian Res 192:78–88
Russell J (1992) Investigation of the potential of Pb-Pb radiometric
dating for the direct age determination of carbonates. Ph.D. thesis,
University of Oxford
Schreiber BC, Babel M, Lugli S (2007) Introduction and overview. In:
Schreiber BC, Lugli S, Babel M (eds) Evaporites through space and
time, vol 285, Geological Society Special Publications. Geological
Society, London, pp 1–13
Schr€oder S, Bekker A, Beukes NJ, Strauss H, van Niekerk HS (2008)
Rise in seawater sulphate concentration associated with the
Paleoproterozoic positive carbon isotope excursion: evidence from
sulphate evaporites in the 2.2–2.1 Gyr shallow-marine Lucknow
Formation, South Africa. Terra Nova 20:108–117
Shen Y, Buick R, Canfield DE (2001) Isotopic evidence for microbial
sulphate reduction in the early Archaean era. Nature 410:77–81
Shen Y, Farquhar J, Masterson A, Kaufman AJ, Buick R (2009)
Evaluating the role of microbial sulfate reduction in the early
Archean using quadruple isotope systematics. Earth Planet Sci
Lett 279:383–391
SimMS, Bosak T, Ono S (2011) Large sulfur isotope fractionation does
not require disproportionation. Science 333:74–77
Simon L, Godderis Y, Buggisch W, Strauss H, Joachimski MM (2007)
Modeling the carbon and sulfur isotope compositions of marine
sediments: climate evolution during the Devonian. Chem Geol
246:19–38
Staudt WJ, Schoonen MAA (1995) Sulfate incorporation into sedimen-
tary carbonates. In: Vairavamurthy MA, Schoonen MAA (eds)
Geochemical transformations of sedimentary sulfur. American
Chemical Society, Washington, pp 332–345
Strauss H (1997) The isotopic composition of sedimentary sulfur
through time. Palaeogeogr Palaeoclimatol Palaeoecol 132:97–118
Strauss H (2002) The sulfur isotopic composition of Precambrian
sedimentary sulfides – seawater chemistry and biological evolution.
In: Altermann W, Corcoran P (eds) Precambrian sedimentary
environments: a modern approach to ancient depositional systems,
Special publications of the international association of
sedimentologists, vol 33. Blackwell Science, Oxford, pp 67–105
Strauss H (2003) The early Archean sulfur cycle as evident from sulfur
isotopes. Precambrian Res 126:349–361
Thomazo C, Pinti D, Busigny V, Ader M, Hashizume K, Philippot P
(2009) Biological activity and the Earth’s surface evolution:
insights from carbon, sulfur, nitrogen and iron stable isotopes in
the rock record. C R Palevol 8:665–678
5 7.5 Abundant Marine Calcium Sulphates: Radical Change of Seawater Sulphate Reservoir 1193
Ueno Y, Ono S, Rumble D, Maruyamas S (2008) Quadruple sulfur
isotope analysis of c. 3.5 Ga Dresser Formation: new evidence for
microbial sulfate reduction in the early Archean. Geochim
Cosmochim Acta 72:5675–5691
Van Kranendonk MJ, Philippot P, Lepot K, Bodorkos S, Pirajnoa F
(2008) Geological setting of Earth’s oldest fossils in the ca. 3.5 Ga
Dresser Formation, Pilbara Craton, Western Australia. Precambrian
Res 167:93–124
Velikoslavinsky SD, Kotov AB, Sal’nikova EB, Glebovitsky VA,
Kovach VP, Zagarnaya NY, Belyaevsky NA, Yakovleva SZ,
Fedoseenko AM (2003) The U–Pb age of the Fedorov sequence of
the Aldan granulite–gneiss megacomplex, the Aldan Shield.
Commun Russ Acad Earth Sci 393:1151–1155 (in Russian)
Vinogradov VI, Reimer TO, Leites AM, Smelov SB (1976) The oldest
sulfates in the Archean Formations of the South African and the
Aldan Shields, and the evolution of the Earth’s oxygen atmosphere.
Lithol Miner Resour 11:407–420 (in Russian)
Warren JK (1989) Evaporite sedimentology. Prentice Hall Inc,
Englewood Cliffs, p 285
Warren J (1999) Evaporites: their evolution and economics. Oxford
University Press, Oxford, p 438
Werne JP, Hollander DJ, Lyons TW, Sinninghe Damste JS (2003)
Organic sulphur biogeochemistry: recent advances and future
research directions. In: Amend JP, Edwards KJ, Lyons TW
(eds) Sulfur biogeochemistry: past and present, Geol Soc Am
Spec Pap 379: 135–150, Geological Society of America, Boulder
Woodhead JD, Hergt JM (1997) Application of the ‘double spike’
technique to Pb-isotope geochronology. Chem Geol 138:
311–321
Wortmann UG, Bernasconi SM, B€ottcher ME (2001) Hypersulfidic
deep biosphere indicates extreme sulfur isotope fractionation
during single-step microbial sulfate reduction. Geology 29:
647–650
Zharkov MA (1984) Paleozoic salt bearing formations of the world.
Springer, Berlin, p 395
Ziegler PA (1990) Geological atlas of Western and Central Europe, 2nd
edn. Shell Internationale Petroleum Mij, B.V., The Hague
1194 H. Strauss et al.
7.6 Enhanced Accumulation of Organic Matter:The Shunga Event
Harald Strauss, Victor A. Melezhik, Aivo Lepland, Anthony E. Fallick,Eero J. Hanski, Michael M. Filippov, Yulia E. Deines, Christian J. Illing,Alenka E. Crne, and Alex T. Brasier
7.6.1 Introduction
Harald Strauss
A number of sedimentary formations deposited globally
around 2.0 Ga ago are characterised by high abundances of
organic carbon. These formations often contain occurrences of
highly concentrated, matured organic material representing
metamorphosed oil, now pyrobitumen. Apart from their com-
mon names pyrobitumen or anthraxolite, different terminol-
ogy has been used for these rocks within the pertinent
literature, including shungite, thucolite, or Precambrian
“coal”. Given their long and frequently complex geologic
history, these sedimentary formations exhibit a variable and
sometimes substantial degree of metamorphic (thermal) over-
print. Consequently, many of them show undisputable signs
of thermal mobilisation, migration and likely loss of hydro-
carbons/bitumen. This includes the so-called shungite rocks on
the Fennoscandian Shield.
The term “shungite” was originally introduced by
Inostranzev (1885, 1886) to describe a black, lustrous sub-
stance containing c. 98 wt.% C that occurs in the form of
veins and layers in Palaeoproterozoic sedimentary succes-
sion in the Onega Basin near the Russian village Shunga in
Karelia. During the course of 200 years of investigation, the
original meaning of this term has been frequently and arbi-
trarily modified, thus leading to misunderstandings and con-
fusion (reviewed in Filippov 2000; 2002; Melezhik et al.
2004; Filippov and Melezhik 2007). Most of the confusion
resulted from Borisov’s (1956) classification, which served
industrial purposes, was solely based on the organic carbon
content in any Palaeoproterozoic Corg-bearing rocks known
from the Onega Basin, and considered neither the nature of
the host lithology nor the nature of the organic matter itself.
In this contribution we adopt the original definition of
shungite as proposed by Inostranzev (1885, 1886), which
equates with the terms “pyrobitumen” or “anthraxolite”
(petrified hydrocarbons). The Palaeoproterozoic rocks in
the Onega Basin (e.g. the Zaonega Formation) display vari-
able contents of organic carbon, which occurs both as resid-
ual kerogen and migrated pyrobitumen (shungite). These
two forms of carbon are mixed in various proportions;
hence, the term “shungite” is not used in the current contri-
bution to describe rocks. Instead the common term organic
carbon-rich (or organic carbon-bearing) rocks has been
utilised. In light of the original definition of the term
“shungite”, this chapter will start with a concise treatment
of the global record of Palaeoproterozoic organic-rich
sediments, thereby focusing on major occurrences on the
Fennoscandian Shield.
H. Strauss (*)
Institut f€ur Geologie und Pal€aontologie, Westf€alische Wilhelms-
Universit€at M€unster, Corrensstr. 24, 48149 M€unster, Germany
e-mail: [email protected]
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_6, # Springer-Verlag Berlin Heidelberg 2013
1195
7.6.2 World-Wide Record of PalaeoproterozoicCarbonaceous Sediments Representing theShunga Event with Emphasis on theFennoscandian Shield
Aivo Lepland, Eero J. Hanski, Michael M. Filippov,and Victor A. Melezhik
The Shunga Event is defined as an episode of enhanced
accumulation of organic matter that is inferred to be global
and synchronous (Melezhik et al. 1999a; Melezhik et al.
2004). Numerous sedimentary formations that were all
deposited broadly around 2000 Ma ago are characterised
by high abundances of organic carbon (Table 7.3;
Fig. 7.57). The most remarkable among these are the
Palaeoproterozoic organic-rich rocks in Russian Karelia,
particularly those comprising the Zaonega Formation. They
represent both the source rock and reservoir of an ancient
petrified oil field. The formation contains several distinct
stratigraphic intervals that show concentrations of total
organic carbon in excess of 25 wt.%. In fact, the Shunga
Event is named after the eponymous village in Karelia where
a spectacular outcrop of the Zaonega Formation contains a
dm-thick layer of nearly pure pyrobitumen (total organic
carbon content >98 wt.%).
The three best-studied sedimentary successions in Rus-
sian Fennoscandia that record the Shunga Event are the
Pilguj€arvi, Il’mozero and Zaonega formations (Table 7.3).
The Pilguj€arvi Sedimentary Formation in the Pechenga
Greenstone Belt consists of turbiditic carbonaceous sand-
stone-siltstone-mudstone units deposited in a deep-water
continental slope environment (Bekasova 1985; Akhmedov
and Krupenik 1990). Tuffs and tuffites are common in the
upper part of the sedimentary section and are used as evi-
dence that the boundary with overlying mafic-ultramafic
volcanic rocks (the Pilguj€arvi Volcanic Formation) is transi-
tional (Hanski 1992). These characteristics thus indicate that
carbonaceous Pilguj€arvi sediments accumulated in a volca-
nically active continental slope setting.
The Il’mozero Sedimentary Formation in the Imandra/
Varzuga Greenstone Belt comprises mostly greywacke,
dolostone, chert and black shale, and represents a succession
deposited along an initially clastic-dominated shelf environ-
ment that subsequently shallowed, allowing the establish-
ment of a carbonate platform represented by stromatolitic
dolostones (Melezhik and Predovsky 1982; Melezhik 1992).
Although thin tuffites are present in places in the lower part
of the formation and a layer of mafic tuff occurs locally
within black shales in the upper part of the succession (see
Chap. 4.1), the bulk of the Il’mozero sediments accumulated
without any volcanic influence.
The Zaonega Formation in the Palaeoproterozoic Onega
Basin consists of organic-poor siltstone and shale in the
lower part and Corg-rich greywacke, dolostone, mudstone,
chert and mafic tuff in the upper part (Galdobina 1987;
Filippov 1994). The Corg-rich upper part is interlayered
with several mafic lava flows and intersected by sills
indicating an apparent association of Corg-rich sediments
and magmatic rocks. Melezhik et al. (1999a) suggested
that the Zaonega succession accumulated in a rift-bound
basin that experienced volcanic and submarine hydrothermal
activity; this activity may have contributed to the delivery of
nutrients thereby favouring the enhanced productivity of
organic matter. Synchronous pyroclastic volcanism will
have contributed to its rapid burial.
Figure 7.58 shows the occurrence of black shales (or
black schists as they are called when metamorphosed under
a high metamorphic grade) in Finland based on field
observations and interpretation of aerogeophysical mea-
surements. These rocks are widely distributed within the
youngest Karelian successions in eastern and northern
Finland, but are also found in the Svecofennian Domain
among the supracrustal belts surrounding the Central
Finland Granitoid Complex. It is interesting to note that
among the first attempts to utilize carbon isotopes to argue
for the biogenic origin of Precambrian sedimentary graphitic
carbon was made by Rankama (1948) from the Sveco-
fennian 1.89–1.91 Ga island arc-related Tampere Belt
(Fig. 7.58). In addition, graphite-bearing schists occur
among highly metamorphosed turbidites of the Lapland
Granulite Belt (Kola Orogen), displaying a carbon isotope
composition consistent with the biogenic origin of the sedi-
mentary carbon (Korja et al. 1996).
Among Karelian Complex rocks, black shales are
associated both with Ludicovian (marine Jatulian) and
Kalevian pelitic sedimentary successions. Ludicovian black
shales were deposited together with dolomitic sediments on
Jatulian sandstones in a deepening epicontinental basin and
have been described, for example, from the North Karelia and
Kuusamo belts (e.g. Pekkarinen 1979; Pet€aikk€o, Juuanj€arvi,Petonen, Liikasenvaara and Siulionpalo formations in
Fig. 7.58). More extensive accumulations occur in central
Lapland where black shale-bearing sedimentary rocks of the
Matarakoski Formation are cut by the 2057 � 8 Ma Keivitsa
mafic layered intrusion (see Fig. 7.58) (Mutanen and Huhma
2001), providing evidence that these sedimentary rocks are
penecontemporaneous with the black shales in the Onega
Basin and older than those in the Pechenga Greenstone Belt.
In central Lapland, graphite-bearing tuffaceous schists are
also found in the Porkonen Formation dominated by different
kinds of banded iron formations (Paakkola 1971). The
A. Lepland (*)
Geological Survey of Norway, Postboks 6315 Sluppen, NO-7491
Trondheim, Norway
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_6, # Springer-Verlag Berlin Heidelberg 2013
1196
1196 H. Strauss et al.
c. 2.0 Ga Porkonen Formation is Ludicovian in age, but its
correlation with Ludicovian sedimentary rocks is not straight-
forward because it is lithologically different and part of the
allochthonous Kittil€a Group (Hanski and Huhma 2005).
The most voluminous black shale formations in Finland
are found among Kalevian sedimentary rocks, which are
divided into two tectonostratigraphic units: the autochtho-
nous-parautochthonous Lower Kaleva, containing turbiditic
conglomerates and breccias, quartz wackes, graywackes and
black shales and locally also banded iron-formations, and
the allochthonous Upper Kaleva comprising mainly deep-
marine turbiditic greywackes, phyllites and black shales
deposited on ophiolitic complexes (e.g. Kontinen 1987;
Lahtinen et al. 2010). The depositional ages of the Kalevian
sedimentary rocks are still poorly constrained; Lower
Kalevian sediments were likely deposited later than c.
1.98 Ga, the age of the youngest mafic dyke generation in
the Archaean basement, while detrital zircon ages of c.
1.95–1.92 Ga determine the maximum age of sedimentation
for Upper Kalevian rocks (Lahtinen et al. 2010). The latter
ages were interpreted as indicating potential sedimentary
sources in the Kola Orogen for the Upper Kaleva, while
the Lower Kaleva had mainly an Archaean provenance
(Lahtinen et al. 2010).
Much attention has been focused in recent years on the
Lower Kalevian black shales in the Talvivaara area, Kainuu
Belt, due to the presence of a world-class Ni-Cu-Zn deposit
in these rocks. The host rocks of the deposit are described
by Loukola-Ruskeeniemi and Heino (1996) and Loukola-
Ruskeeniemi (2011). The thickness of the folded black shale
formation reaches 400 m, but the original thickness likely
varied between 20 and 120 m. The C and S contents are high
with median values of 7.6 and 9.0 wt.%, respectively
(Loukola-Ruskeeniemi 2011). Apart from being rich in
base metals, the Talvivaara black shales have typically
high Mn concentrations (>0.8 wt.%) and locally contain
phosphorus-rich horizons. Originally the black shales were
organic-rich mud deposited on the sea-floor under apparent
anoxic conditions, and were later enriched in metals by
hydrothermal input (Loukola-Ruskeeniemi and Heino
1996). Detrital zircon and Sm-Nd iotope data are compatible
with a major Archaean provenance (Lahtinen et al. 2010).
Other black shales assigned to the Lower Kaleva include
those forming part of the turbidite sequences of the Martimo
Formation in the Per€apohja Belt (Perttunen and Hanski
2003) and the Haukipudas Formation in the Kiiminki Belt
(Fig. 7.58) with the latter being closely associated with
submarine mafic volcanism (Honkamo 1985).
In the Outokumpu area, C- and S-rich black shales are
intimately associated with ophiolitic serpentinites, dolomite-
rich rocks, calc-silicate rocks, metasomatic quartz-rich rocks
and Outokumpu-type Cu-Co-Zn ores (Loukola-Ruskeeniemi
1999, 2011). The Upper Kalevian black shales were depos-
ited on sea-floor exposing ultramafic mantle rocks, which
were later thrust together onto the craton margin (Peltonen
2005). The Outokumpu black shales are equally high in C
and S as the Talvivaara black shales but low in base metals
and Mn (Loukola-Ruskeeniemi 1999, 2011).
Other major occurrences of Early Palaeoproterozoic Corg-
rocks are present outside the Fennoscandian Shield and
pertinent data have been compiled in Table 7.3. The reader
is referred to the original literature for further information.
The majority of Palaeoproterozoic Corg-rich successions,
including the Zaonega Formation are insufficiently dated
(Table 7.3) to determine unequivocally if these carbona-
ceous sediments indeed represent a broadly synchronous,
relatively short-lived event (<50 Ma) or a long-lasting
period (>50 Ma) when conditions prevailed for high pri-
mary productivity and/or organic matter burial and/or pres-
ervation. In fact, the Corg-rich sediments may have
accumulated at different times during an extended period
in response to local basinal conditions such as fluctuating
supply of nutrients and sediment loads.
In summary, the palaeoenvironmental interpretations
derived fromoccurrences on theFennoscandianShield indicate
that Palaeoproterozoic carbonaceous sediments accumulated in
different depositional settings. Processes and factors specific to
each area, such as localised volcanic/hydrothermal activity,
likely promoted primary productivity and/or enhanced burial/
preservation. A frustrating dearth of radiometric age data
hinders making robust stratigraphic correlations of these
formations across the Fennoscandian Shield and to other
Palaeoproterozoic organic-rich units elsewhere. Nevertheless,
the existing lithostratigraphic constraints combined with the
few available age data are permissive of models that assume a
global synchroneity in the genesis of the Shunga Event and its
temporal distinctness from the Lomagundi-Jatuli Event.
In light of the amount of organic matter that was depos-
ited originally, both in respect to a single sedimentary basin
and even more so with regard to the presumed globally
synchronous deposition during the so called Shunga Event,
specific environmental conditions are required. These per-
tain to the aspect of primary production, here specifically the
availability of nutrients, but also to conditions that favour the
preservation of the deposited organic matter. Both aspects
will be addressed in the following sections.
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1197
7.6.3 The Shunga Event: A Tale of Productivityand Preservation of Organic Matter in the EarlyPalaeoproterozoic Ocean
Harald Strauss and Christian J. Illing
The early Palaeoproterozoic is characterised by distinct,
massive perturbations of the global carbon cycle. The seem-
ingly coeval global appearance of sediments that are extre-
mely rich in organic carbon represents an unusual (if not
unprecedented) production/accumulation of organic matter,
a massive perturbation of the global carbon cycle that has
collectively been termed the Shunga Event. Moreover, based
on current age information the Shunga Event was initiated
directly in the aftermath of yet another apparent massive
perturbation of the global carbon cycle, the Lomagundi-Jatuli
Event. Although regional/local amplifications have been
addressed (e.g. Melezhik et al. 1999b), the presence of posi-
tive to even strongly positive carbonate carbon isotope values
in coeval sedimentary successions worldwide, being the
prime characteristic of the Lomagundi-Jatuli Event, have
been viewed as an expression of the enhanced deposition of
organic matter (but see Hayes and Waldbauer 2006, or
Fallick et al. 2008, 2011 for alternative interpretations).
Our quest to unravel this rather puzzling succession of
major events that are all demonstrably related to the produc-
tion and deposition (and recycling) of organic material, but
foremost discussing the Shunga Event, can be guided by two
seemingly simple questions:
1. What was the nature of primary productivity during the
early Palaeoproterozoic?
2. What were the depositional conditions/requirements that
led to the unusual accumulation of organic matter?
While the former question tackles an issue of global impor-
tance, the latter will be discussed in the light of the sedimen-
tary rocks preserved on the Fennoscandian Shield and cored
by the FAR-DEEP.
The Nature of Primary Productivity in the EarlyPalaeoproterozoic Ocean
Constraining the geologic record of primary productivity
(or even identifying the primary producers themselves)
relies on two different lines of evidence (for a recent review,
see Knoll et al. 2007): body fossils and chemofossils in
sedimentary rocks, i.e. molecular biomarkers and the stable
isotopic composition of carbon. However, exploiting
respective evidence archived in the sedimentary rock record
is strongly dependent on preservation of the host sediment.
Post-depositional processes during diagenesis and metamor-
phism tend to obscure or completely erase such evidence of
ancient life. More so, discussing primary productivity in the
ancient ocean requires very specific evidence that unambig-
uously proves/identifies the nature of primary producers in
the Palaeoproterozoic marine realm.
Guided by Charles Lyell’s “first principle” (i.e., “The
Present is the Key to the Past”; Lyell 1830) and nestled
into the environmental framework some 2.0 billion years
ago, the question can be re-phrased in the sense whether
primary producers at that time in Earth history were eukary-
otic or prokaryotic (see Chap. 7.8.3), and whether primary
productivity in the marine realm was exclusively based
on photosynthetic autotrophic carbon fixation as we know
it today (i.e. Calvin Cycle RUBISCO-type oxygenic
photosynthesis).
Unambiguous fossils of eukaryotes occur in sediments as
old as 1.85 Ga (see Chap. 7.8.3; Zhang 1986; Peng et al.
2009). Whether these organisms formed part of the marine
benthos and/or were part of the prevailing phytoplankton
remains to be determined. In addition to preserved body
fossils, steranes as presumed eukaryotic molecular fossils
have been reported from Proterozoic and even Neoarchaean
sedimentary successions (e.g. Brocks et al. 1999; Dutkiewicz
et al. 2006; 2007). More recently however, the syngeneity of
these molecular fossils has been questioned (Brocks et al.
2003; Rasmussen et al. 2008), in particular for the older and
more thermally mature Neoarchaean sediments. This casts
some doubt about their significance for reconstructing the
temporal evolution of eukaryotes. Thus, little firm evidence
exists that photosynthetic eukaryotes played a significant role
(if existing at all) for primary productivity in the marine
realm during the time of the Shunga Event.
In the absence of clear evidence for eukaryotic life on
Earth, primary productivity in the early Palaeoproterozoic
ocean and before was likely governed by prokaryotic
organisms. Acknowledging that the Great Oxidation Event
at 2.4 Ga reflects the first time in Earth’s history when
oxygen production outcompeted oxygen consumption, it is
generally assumed that oxygenic photosynthesis evolved
earlier than this (e.g. Blankenship et al. 2007), even though
this important biological innovation is poorly constrained in
time. More so, it is considered that early oxygenic photosyn-
thesis would have been related to cyanobacteria and these
were, thus, inhabitants of the Palaeoproterozoic ocean. The
oldest unambiguous cyanobacteria-like fossils have been
reported from 1.9 Ga old rocks of the Belcher Group in
Canada (Hofmann 1976; Golubic and Hofmann 1976). Fos-
sil evidence suggests that many Proterozoic examples were
part of benthic shallow water communities, probably also
representing the architects of many Proterozoic stromatolites
H. Strauss (*)
Institut f€ur Geologie und Pal€aontologie, Westf€alische Wilhelms-
Universit€at, Corrensstr. 24, 48149 M€unster, Germany
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_6, # Springer-Verlag Berlin Heidelberg 2013
1198
1198 H. Strauss et al.
(Grotzinger and Knoll 1999). But they must have also been
constituents of the Proterozoic phytoplankton, hence, impor-
tant contributors to marine primary production (e.g. Knoll
et al. 2007). The importance of cyanobacteria in the marine
realm is further suggested by the presence of distinct
biomolecules in the sedimentary rock record. More precisely
the presence of 2-Methylhopanes in organic-rich shales
suggests that cyanobacteria could have been the primary
producers during the Proterozoic (e.g. Summons et al.
1999). Again, the significance of this Proterozoic biomarker
record but more so an extension of this record into the
Neoarchaean suffers from the same doubts put forward in
respect to evidence for the advent of sterane biosynthesis, i.e.
the aspect of syngeneity versus modern contamination
(Brocks et al. 2003). Still, molecular evidence for
cyanobacteria coupled to the observation of an increase in
the atmospheric oxygen abundance some 2.4 Ga ago (the
Great Oxidation Event, c.f. Holland 2002, 2006) would be
consistent with the conclusion that oxygenic photosynthesis
was the key process for primary productivity in the early
Palaeoproterozoic.
The uniqueness of this conclusion, and hence, the ubiq-
uity of oxygenic photosynthesis in space and time, could
be questioned when considering evidence for anoxygenic
photoautotrophy under sulphidic water column conditions
during the later Palaeoproterozoic. Based on biomarker evi-
dence from the 1.64 Ga Barney Creek Formation, McArthur
Basin, Australia, Brocks et al. (2005) concluded that anaer-
obic phototrophic sulphide-oxidising bacteria thrived in a
stratified ocean where euxinic conditions reached well into
the photic zone. Free hydrogen sulphide was available as an
electron donor during carbon fixation and biosynthesis in the
photic zone of a water body. This required that a stable
chemocline separated an upper oxic from a lower sulphidic
water column, and that this chemocline was located in the
photic zone, i.e. the upper few tens of meters of the water
body (this specific scenario is termed photic-zone-anoxia).
Consequently, however, this geochemical fingerprint attests
to the fact that specific physico-chemical conditions allowed
at least for an additional and substantially different form of
primary productivity in this environment. There is limited
evidence available for photic-zone-anoxia and anoxygenic
photosynthesis in the geologic record. Whether this reflects
only limited significance for global primary productivity is
difficult to assess.
No doubt, the presence of molecular fossils in Precam-
brian rocks is strongly affected by thermal alteration of their
sedimentary host rock. Coupled to an ongoing discussion
about the syngeneity of molecular fossils in ancient rocks
versus them being modern contaminants (e.g. Brocks et al.
2003), questions remain about the reliability of biomarkers
for any conclusions drawn about evolutionary trends in
primary production and oxygenic photosynthesis. As an
alternative, an established carbon isotope record for the
Precambrian (e.g. Hayes et al. 1983; Schidlowski 1988;
Strauss et al. 1992; Shields and Veizer 2002) can be
inspected for evidence of primary productivity via autotro-
phic carbon fixation.
Modern day marine primary productivity is governed by
autotrophic carbon fixation through oxygenic photosynthesis:
CO2 þ H2O ! CH2Oþ O2
with CH2O representing a simplified expression of primary
biomass. This process is associated with a distinct fraction-
ation in carbon isotopes. The carbon isotopic composition is
expressed as d13C ¼ ([(13C/12C)sample/(13C/12C)standard]�1)*
1,000 in per mil (‰), relative to the Vienna Pee Dee Belem-
nite (VPDB) standard. The reaction product, i.e. the resulting
biomass, is characterised by depletion in the heavy stable
carbon isotope 13C. Hence, organic matter carries a negative
d13C value. The magnitude of this isotopic fractionation, i.e.
the difference between the isotopic composition of carbonate
(in the modern ocean around 0 ‰) and organic carbon
(mostly between �30 ‰ and �20‰), is dependent on
growth rate, physiology of the photosynthetic organism, and
the ratio between external, i.e. atmospheric, and cell-internal
concentration of carbon dioxide (for details, see reviews
by Hayes 1993; Des Marais 2001). But this magnitude in
isotopic fractionation is diagnostic for autotrophic carbon
fixation, and researchers have traced the isotopic composition
of the inorganic carbon source and the organic carbon product
through time in search for evidence of metabolic pathways of
primary productivity (e.g. Hayes et al. 1983; Schidlowski
et al. 1983; Schidlowski 1988; Strauss et al. 1992; Shields
and Veizer 2002; Eigenbrode and Freeman 2006; Thomazo
et al. 2009).
Carbon isotopes allow reconstructing the operation of the
global carbon cycle. The Phanerozoic carbon cycle and its
respective isotope records (Fig. 7.59; Veizer et al. 1999;
Hayes et al. 1999) are well constrained for explaining the
isotope effects associated with the respective carbon flows
between the inorganic and organic carbon pools. Marine
carbonates have been analysed in order to constrain the
isotopic composition of the inorganic carbon substrate avail-
able for photosynthetic carbon fixation, i.e. atmospheric
carbon dioxide. The carbon isotopic composition of sedi-
mentary organic carbon is viewed to result from marine
primary productivity (details discussed, e.g. in Hayes et al.
1999; Des Marais 2001). Two observations can be made:
(1) both isotope records vary in a seemingly sympathetic
manner thereby exhibiting a more or less constant average
isotopic difference between both records of some 25–30 ‰,
and (2) large fluctuations in the absolute d13C values exist
for both carbonate and organic carbon. There are numerous
details archived in these isotope time series, but it is beyond
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1199
the scope of this presentation to provide an in-depth discus-
sion (see Veizer et al. 1999 and Hayes et al. 1999 for
intricate details pertaining to these isotope records). How-
ever, a single important conclusion can be drawn based on
the magnitude of the largely invariable difference between
inorganic (i.e. oxidised) and organic (i.e. reduced) carbon.
The magnitude in isotopic fractionation and the sympathetic
secular variation of both isotope records attest that marine
primary production was largely governed by a single process
throughout the entire Phanerozoic. Considering the modern
world as an analogue, this process was autotrophic carbon
fixation via oxygenic photosynthesis.
Guided by this view, researchers have evaluated the Pre-
cambrian carbon isotope records (Fig. 7.60). These isotope
time series display a considerable variation in d13C that
results from temporal changes in the initial carbon isotopic
composition (carbonate precipitation and biosynthesis) but
also exhibit substantially more “noise” resulting from post-
depositional alteration of these ancient sedimentary rocks.
Different processes during diagenesis and metamorphism
affect both the carbonate carbon and the organic carbon in
different ways (Fig. 7.61a). During diagenesis, the carbon
isotopic composition of marine carbonates frequently
changes to more negative d13C values (e.g. Irwin et al.
1977). This is a consequence of the incorporation of13C-depleted carbon dioxide resulting from the respiration
of sedimentary organic matter during precipitation of diage-
netic carbonate cements in the pore water realm. This micro-
bial turnover of organic matter can happen under oxic
as well as anoxic conditions, but is frequently coupled to
the anaerobic process of bacterial sulphate reduction (see
Chap. 7.5). In contrast, 13C-enriched carbonates point to bac-
terial methanogenesis (i.e. production of 13C-depleted CH4
that is subsequently lost to the environment), again in the
diagenetic realm where the remaining 13C-enriched carbon
dioxide (a by-product of methanogenesis) is precipitated as
isotopically positive carbonate cement. Thermal alteration of
sedimentary organic matter tends to remove 13C-depleted
compounds (e.g. Hayes et al. 1983). As a consequence, the
remaining residual organic matter becomes more and more
fragmentary and its carbon isotopic composition changes
to less negative, 13C-enriched d13C values. Finally, a strong
metamorphic overprint might ultimately even lead to partial
isotopic equilibrium between carbonate and organic carbon
where the isotopic difference between both carbon species
decreases (e.g. Schidlowski et al. 1979). Acknowledging that
different diagenetic and/or metamorphic reactions result in
variable changes in d13C, one feature becomes apparent,
notably that post-depositional processes generally lead to a
change in the isotopic difference between carbonate and
organic carbon that – initially – resulted from isotopic frac-
tionation associated with biosynthesis.
The apparent “noise” related to post-depositional pro-
cesses is in contrast to an observation in the carbon isotope
time series for the Phanerozoic as well as the Precambrian
where both d13Ccarb and d13Corg values change in parallel
way (Fig. 7.61b), i.e. both isotopic compositions become
more positive (13C-enriched) or more negative (13C-
depleted). Here, the isotopic difference between both carbon
species does not change. This suggests that the isotopic
composition of the carbon source that is common to both
carbon species, notably atmospheric carbon dioxide,
changed as a consequence of processes affecting the global
carbon cycle. Such perturbations have been interpreted to
reflect temporal changes in the fractional burial of organic
carbon, based on a simple carbon isotope mass balance:
d13Cinput ¼ forganicd13Corganic þ 1� forganic
� �d13Ccarbonate
with forganic reflecting the proportional fraction of carbon
being buried in the sedimentary record as organic carbon,
and the different d13C values representing the isotopic
compositions of organic (i.e. reduced) and carbonate
(i.e. oxidised) carbon. In the modern world (Fig. 7.61b),
the isotopic compositions of carbonate and organic carbon
result in an forganic value of 0.2, i.e. 20 % of global carbon
burial occurs as organic matter and 80 % as carbonate.
Following this reasoning, the enhanced burial of 13C-
depleted organic carbon (i.e. forganic >0.2) will inevitably
change the isotopic composition of the entire carbon cycle
and subsequently formed carbonate and organic carbon will
display 13C-enriched, more positive d13C values.
Returning to the Precambrian carbon isotope records
(Fig. 7.60), two different signals govern the isotopic
variability. Some of the more extreme negative and maybe
also positive d13C values reflect post-depositional changes
that occurred in the diagenetic realm. However, despite
sometimes intense post-depositional alteration, the record
contains additional, largely unaltered primary signatures
that reflect the isotopic fractionation associated with biosyn-
thesis. Comparing both carbon isotope time series (for a
review, see e.g. Des Marais 2001), it becomes apparent that
an overall isotopic difference between carbonate and organic
carbon of 20–30 ‰ suggests that autotrophic carbon fixation
was the prime metabolic pathway for primary productivity
throughout much of Precambrian time. More so, the appar-
ent isotopic difference between carbonate and organic car-
bon would even be consistent with oxygenic photosynthesis
as the key process. Superimposed on this are more positive
d13Ccarb values that characterise the Lomagundi-Jatuli
Event between c. 2.2 and 2.0 Ga (see Chap. 7.3). The
appearance of such positive carbonate carbon isotopes in
coeval sedimentary successions worldwide suggests this to
be a global phenomenon, despite claims for regional/local
amplifications of the isotopic signature (e.g. Melezhik et al.
1999b). Following isotope mass balance consideration, the
presence of strongly positive carbonate carbon isotope
values suggests the enhanced deposition of organic matter
1200 H. Strauss et al.
(but see Hayes and Waldbauer 2006, and Fallick et al. 2008,
2011 for alternative interpretations; see also Chap. 7.3).
In apparent contrast are strongly negative d13Corg values
(more negative than �40 ‰) that were measured for
Neoarchaean sedimentary organic matter, suggesting the
activity of methanotrophic bacteria using methane as the
principal carbon source for biomass production (Hayes
1994; Eigenbrode and Freeman 2006). In similar contrast
to the majority of data are 13C-depleted Neoarchaean and
early Palaeoproterozoic carbonates associated with banded
iron formations that either reflect specific environmental
conditions or represent a diagenetic or a metamorphic fea-
ture (e.g. Beukes et al. 1990; Winter and Knauth 1992;
Fallick et al. 2011).
Considering the carbon isotope records for the time of the
Shunga Event around ~2.0 Ga (Fig. 7.60), two observations
are discernible. First, rocks display a substantial isotopic
variability in their d13Corg signature. This includes strongly13C-enriched as well as 13C-depleted values that could well
result from post-depositional diagenetic and/or metamorphic
processes. Focusing specifically on carbonaceous rocks
from the Zaonega and overlying Kondopoga formations
on the Fennoscandian Shield, these display organic carbon
isotope values ranging from �45 ‰ to �17 ‰ (Fig. 7.62;
Melezhik et al. 1999a, 2009). Moreover, d13Corg values dis-
play a clear bimodal distribution with maxima at�36 ‰ and
at �28 ‰. However, the presence of organic matter within
the stratigraphic column that records the Shunga Event on the
Fennoscandian Shield is highly complex and includes
sediments that might have escaped substantial mobilisation
of organic matter, different generations of migrated bitumen,
cross-cutting veins filled with bitumen, and redeposited
organic matter (e.g. Melezhik et al. 2009). Considering the
different rock types and different appearance of organic mat-
ter within the stratigraphy, the range in d13Corg would be
consistent with a common biological source but would also
suggest modifications of the d13Corg signature during post-
depositional thermal alteration, migration and redeposition.
Second, the carbonate carbon isotopic record at c. 2.0 Ga
displays a substantial number of d13Ccarb values around
0 ‰, but also includes 13C-enriched as well as strongly13C-depleted carbon isotope data as low as �18 ‰(for compilations, see Shields and Veizer 2002; Fallick
et al. 2008; Prokoph et al. 2009). d13Ccarb values around
0 ‰ suggest an operation of the global carbon cycle compa-
rable to most parts of Earth history, i.e. governed by autotro-
phic carbon fixation. The strongly positive carbonate carbon
isotope data appear to record the decline of the Lomagundi-
Jatuli Event (e.g. Karhu 1993). In contrast, many of the
negative d13Ccarb values were measured on iron carbonates
(Winter and Knauth 1992) for which a formation from a
stratified water body was proposed. This observation points
to microbial reworking of organic matter in the anoxic part of
a stratified water body, possibly by sulfate-reducing bacteria.
In addition, the negative carbonate carbon isotopic signature
could result from microbial heterotrophic reworking of sedi-
mentary organic matter and subsequent incorporation of
resulting carbon dioxide during diagenetic carbonate forma-
tion within the sediment (see Fig. 7.61a). For the Corg-rich
rocks, this becomes apparent in concretionary carbonates
from the Zaonega Formation where d13Ccarb values range
from �26 ‰ to �5‰ (Melezhik et al. 1999a).
Based on the carbon isotopic signature of organic matter
deposited at the time of the Shunga Event, its biological
origin can be assumed. Considering the substantial amount
of organic matter deposited on the Fennoscandian Shield
(Melezhik et al. 2009), let alone worldwide during the time
of the Shunga Event, places some as yet not quantified
constraints on the chemical composition of the ocean. Most
critical is the aspect of nutrient availability. Looking at
modern marine surface waters, the primary production
requires the availability of biologically metabolisable nitro-
gen and phosphorus as macro-nutrients (Redfield 1958) and
a suite of metals (most prominently Fe) as micro-nutrients.
With respect to the macro-nutrients in the modern world,
phosphorus is derived from continental weathering and the
biological nitrogen demand is satisfied with dissolved nitrate
(Schlesinger 1997). The latter in particular, i.e. a sufficient
supply of biologically accessible nitrogen (e.g. as nitrate)
requires the establishment of the respective nitrogen cycling.
Geochemical indications for the onset of aerobic nitrogen
cycling during late Neoarchaean time have been published
(Garvin et al. 2009; Thomazo et al. 2011).
Even considering a biological origin for the organic mat-
ter deposited during the Shunga Event, and accepting
constraints in respect to nutrient requirements, establishing
a firm understanding of the prevailing biochemical process
of primary production during this time interval based on the
carbon isotopic composition represents a challenge. More-
over, the question needs to be assessed whether the environ-
mental conditions might have played a key role, in addition
to primary production, for the enhanced deposition and/or
preservation of this unprecedented amount of organic matter.
The lack of a full understanding, and the challenge for
future research, results from the highly complex appearance
of organic matter within Corg-bearing rocks on the Fenno-
scandian Shield. The overall variability in isotopic fraction-
ation between inorganic carbon source (archived in the
carbonate rock record) and organic product (exemplified by
sedimentary organic carbon) is well within the range of
autotrophic carbon fixation and may even be consistent
with oxygenic photosynthesis. However, considering the
variation in d13Corg, any more detailed interpretations
require an assessment of the environmental conditions dur-
ing deposition and any subsequent alteration that occurred
during post-depositional processes. Both aspects can be
addressed by looking at the rock succession itself, which
will be done in the subsequent section.
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1201
7.6.4 Giant Palaeoproterozoic Petrified OilField in the Onega Basin
Victor A. Melezhik, Anthony E. Fallick, MichaelM. Filippov, Yulia E. Deines, Alenka E. Crne,Aivo Lepland, Alex T. Brasier, and Harald Strauss
One of the specific features of the Shunga Event is the wide-
spread occurrence of both autochthonous organic matter and
migrated former bitumen (Fig. 7.63). Such occurrences were
reported from c. 1.8–2.0 Ga rocks in several places, including
Greenland (Bondesen et al. 1967), North America (Gunflint,
Onwatin, Matineda andMichigamme pyrobitumen; Mancuso
et al. 1989), Africa (Francevillian pyrobitumen; Gauthier-
Lafaye and Weber 1989) and Fennoscandia (Karelian
shungite; Inostranzev 1885, 1886). Another feature of the
event is the generation of oil on a scale previously unprece-
dented, with an estimated original petroleum potential com-
parable to modern supergiant oil fields (Mossman et al. 2005;
Melezhik et al. 2009). The Onega Basin in Russian
Fennoscandia (Fig. 7.64) contains one such supergiant
petrified oil field, though it is uniquely preserved including
source rocks, oil migration pathways, evidence of oil traps,
subaqueous and surface oil seeps (Melezhik et al. 2009).
Recentlymany of these features have been targeted by several
FAR-DEEP drillholes (see Chaps. 6.3.3 and 6.3.4); hence new
valuable material is offered for research into this Palaeopro-
terozoic petrified oil field.
Source Rocks
Inferred source rocks for the Onega petrified oil field are
confined to the Zaonega Formation, which is one of the nine
sedimentary-volcanic successions comprising the Onega
synclinorium (for details see Chap. 4.3). The minimum age
of the Zaonega Formation is constrained at c. 1.98 Ga by
several whole-rock and mineral Sm-Nd, Re-Os and Pb-Pb
isochrons obtained from a differentiated mafic intrusion in
the overlying volcanic succession of the Suisari Formation
(Puchtel et al. 1992, 1998, 1999). The maximum age is
constrained at 2.090 � 0.07 Ga, which is the age obtained
by Ovchinnikova et al. (2007) for the underlying Tulo-
mozero dolostones using the Pb-Pb technique. Preliminary
Re-Os data on Corg-rich rocks from the Zaonega Formation
suggest an age of ~2.05 Ga (Hannah et al. 2008). The
formation was deformed and underwent greenschist facies
metamorphism during the 1.8 Ga Svecofennian orogeny.
The Zaonega Formation is a c. 1,500-m-thick succession
with an areal extent of 9,000 km2 on the present-day surface
(Figs. 7.64 and 7.65). It consists of organic carbon and
sulphide-bearing greywackes, siltstones, mudstones, calcare-
ous greywacke and shales, marls, mudstone- and pyrobitumen-
rich mudstone-supported breccias, mafic lavas and tuffs, and
subordinate limestones, dolostones, and cherts (Fig. 7.66).
The greywacke-siltstone-shale is rhythmically bedded and
comprises Bouma sequences, hence the siliciclastic rocks
were deposited from turbidity currents. Breccias apparently
have a variable origin. In part, they represent mass-flow
deposits associated with slope-slide movements, whereas
othersmay be related to seafloor explosive eruptions associated
with formation of peperite (see below for some details).
Sedimentary dolostones (in situ chemically precipitated
or resedimented) are laminated and comprise laterally per-
sistent beds, whereas limestones and massive dolostones
occurring as lenses are likely diagenetically formed
(Melezhik et al. 1999b). Cherts occur as concretions and
up to 6-m-thick intervals. Whether the cherts are seafloor
hydrothermal deposits or represent pervasively silicified
sediments, remains to be proven. The lower part of the
formation shows enrichment in sodium whereas the upper
part has a “potassic” character (Filippov et al. 1994).
Detailed sedimentological description of the Zaonega For-
mation succession is presented in Chaps. 6.3.3 and 6.3.4.
Overall, the Zaonega Formation was very likely deposited in
a rift system associated with mafic volcanism within an
active continental margin setting (see Chaps. 3.3 and 6.3.3).
The Zaonega succession contains a high proportion of
igneous rocks, namely mafic lavas, tuffs and gabbro sills,
whose volume ranges from c. 35 % in sections drilled by
FAR-DEEP Holes 12A, 12B and 13A to as high as 60 % in a
section intersected by the 3,500-m-deep Onega parametric
drillhole (Fig. 7.65b). In the latter case approximately 50 %
of sedimentary rocks are rich in Corg (>5 wt.% total organic
carbon). Both lavas and sills show interaction with the host
sediments. Solid-state recrystallisation and formation of
hornfels have not been observed at contact zones with gab-
bro, though primary layering/bedding shows a considerable
soft-sediment modification and obliteration. In places,
organic matter-rich rocks in contact with gabbro sills and
dykes show well-developed columnar joints oriented per-
pendicular to the contact (Fig. 7.66h). Although such
features are rarely described in the published literature,
similar prismatic columnar joints, caused by combustion of
organic matter in bituminous sediments, were reported from
the Hatrurium basin in Israel (Grapes 2006 and references
therein). Both sill and lava flows are associated with the
formation of peperite (Fig. 7.66i, j) (Biske et al. 2004),
implying that gabbro sills were emplaced into, and lavas
V.A. Melezhik (*)
Geological Survey of Norway, Postboks 6315 Sluppen, NO-7491
Trondheim, Norway
Centre for Geobiology, University of Bergen, Allegaten 41, Bergen
N-5007, Norway
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_6, # Springer-Verlag Berlin Heidelberg 2013
1202
1202 H. Strauss et al.
were extruded onto, unconsolidated wet sediments. Forma-
tion of peperite in the Zaonega Formation as the result of wet
sediment-magma interaction may have an important
palaeoenvironmental implication in that peperite is often
associated with explosive eruptions and generation of signif-
icant hydrothermal systems (Skilling et al. 2002 and
references therein). The onset of such a hydrothermal system
may explain the “sodic” nature and hydrothermal/metaso-
matic alteration of the lower part of the formation with
potential derivation of Na either from seawater or from a
partially dissolved, thick halite bed present at the base of the
Tulomozero Formation underlying the Zaonega Formation
(Fig. 7.65b). Unusual breccias with a pyrobitumen-rich-
matrix (Fig. 7.66f) may be the result of an explosive eruption
associated with the formation of peperite.
The organic carbon content, accounting residual kerogen
and both allochthonous and migrated bitumen, ranges
between 0.1 and c. 99 wt.% (Filippov and Golubev 1994;
Kupryakov 1994; Melezhik et al. 1999a). The average car-
bon content in the Zaonega Formation is poorly constrained
and has been estimated at c. 25 wt.% (e.g. Melezhik et al.
1999a). This value is apparently biased towards Corg-rich
lithologies. A Corg histogram based on data published up to
1999 shows a four-modal distribution with the major mode
at c. 30 wt.% (Fig. 7.67a). This mode, however, does not
apply for major lithologies, but instead represents Corg-rich
rocks, which are commercially exploited and were over-
sampled in earlier studies with respect to Corg-low rocks.
Subsequently, Melezhik et al. (2009) suggested a more mod-
est average value of 10 wt.%, which was influenced by the
reconnaissance analytical work on newly-obtained FAR-
DEEP core material (Fig. 7.67b). The accurate, weighted
average content of organic matter has yet to be constrained.
Similarly to the Corg content of the Zaonega Formation
sedimentary rocks, their carbon isotopic composition varies
greatly. Galdobina et al. (1986) suggested that the carbon
was mantle-derived. However, published d13Corg values
range between �45 ‰ and �17 ‰ and hence are consistent
with a biological source of carbon (Melezhik et al. 1999a).
Published data exhibit a bimodal distribution with modes at
c. �27 ‰ and �36 ‰ (Fig. 7.67c), while the stratigraphic
trend shows that the lower part of the formation is
characterised by d13Corg fluctuating around �25 ‰ and
sharply shifting to �42 ‰ in the middle part (Fig. 7.68)
(Melezhik et al. 1999a). Reconnaissance analytical work
on new material obtained from FAR-DEEP cores
(Fig. 7.67d) and the Onega parametric hole (Fig. 7.67e)
suggest a somewhat similar, though not identical d13Corg
distribution pattern with the higher d13C mode at the canon-
ical value of �25 ‰; there is also a small but intriguing
number of samples with higher d13C values, but it is not
clear at present whether these represent the primary compo-
sition or the effect of secondary processes (e.g. methane loss
during heating); the frequency distribution of these higher
values is flat (see Fig. 7.67a). The d13Corg stratigraphic trend
is shown in Fig. 7.68, though only d13Corg data have been
used for the correlation of all drilled sections as there are no
independent lithological criteria currently available. The
agreement in absolute values of d13C as well as the patterns
of change across the three independent data sets is impres-
sive, and more consistent with global (or at least basinal)
driving forces rather than local effects (e.g. low values
caused by a contribution to biomass from methanotrophs).
A full understanding of the substantial variation of d13Corg as
well as the source(s) for the extremely low d13Corg values
require more specifically targeted work. Currently, it
remains unclear which values characterise the isotopic com-
position of primary biomass (�34 ‰ was suggested by
Melezhik et al. 1999a), and which are the result of microbial
reworking and/or thermal alteration.
The isotopic composition of the associated carbonate
rocks does not offer additional robust constraints on the
carbon cycle until their nature is confidently reconstructed.
Currently available data (Yudovich et al. 1991; Melezhik
et al. 1999a) suggest that their isotopic composition ranges
between �25 ‰ and +8 ‰. A somewhat similar range
(�17 ‰ to +4 ‰) was obtained from 29 FAR-DEEP
archive samples. To explain this range, Yudovich et al.
(1991) invoked involvement of CO2 produced through
methanogenesis (high d13Ccarb) and methanotrophy (low
d13Ccarb). Melezhik et al. (1999a) linked the formation of
carbonates with low d13Ccarb to organic matter recycling via
bacterial sulphate reduction, which was supported by high
abundances of isotopically heterogeneous diagenetic
sulphides (d34S ¼ �22 ‰ to +31 ‰; Shatzky 1990); the
carbonate rocks with near zero d13C values were inferred
to represent the isotopic composition of seawater bicarbon-
ate. However, the prominent stratigraphic shift of d13Corg
from c. �25 ‰ to �42 ‰ remains unexplained.
Estimated Oil Reserve
The current, relatively inaccurate estimate of the organic
matter content in different lithologies of the Zaonega For-
mation ranges between 0.1 and 99 wt.% total organic carbon
(Melezhik et al. 1999a), which includes rocks with both in
situ and migrated organic matter. A conservative estimate
suggests an average of 10 wt.% total organic carbon content
(Melezhik et al. 1999a).
Hunt (1996) estimates that 1 wt.% total organic carbon in
ancient rocks represents a reasonable cut-off for oil source
rocks, and 0.5 wt.% for gas source rocks. Similarly, Tissot
and Welte (1984) considered c. 1 wt.% total organic carbon
as the minimum value for effective hydrocarbon generation
and expulsion from oil-prone organic matter. Neruchev et al.
(1998) assumed that extractable hydrocarbon forms
17–37 % out of total organic carbon.
Based on this, a conservative estimate of the amount of
liquid hydrocarbon in the Onega Basin can be made. Further
considerations include:
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1203
• The >10,000 km2 areal extent of the Zaonega Formation
sedimentary rocks.
• The thickness of the sedimentary rocks: c. 900 m.
• The present average total organic carbon content: 10 wt.%,
which prior to diagenetic alteration would have been
at least 10 % higher (e.g. Arthur and Sageman 1994).
An area of 10,000 km2 with a thickness of 900 m would
equate to 9 � 1012 m3 of source rocks. Choosing 10 wt.%
total organic carbon and 25 % extractable hydrocarbon, one
cubic meter would yield 5 l of petroleum (assuming that the
specific gravity of shale is 2.5 g/cm3 and that of petroleum is
0.8 g/cm3). Hence, the original petroleum potential
(450 � 1011 l or 280 � 109 US barrels) would translate to
a modern supergiant oil field (>5 � 109 US barrels, e.g.
http://www.britannica.com).
Evidence for Oil Generation and Its Timing
The total organic carbon distribution pattern for the Zaonega
Formation summarised in Melezhik et al. (1999a) exhibits
four modes (Fig. 7.67a) suggesting the involvement of dif-
ferent processes in its accumulation. The first mode
(0–15 wt.%) characterises tuffites, greywackes, siltstones,
dolostones, limestones and cherts; all rock types except
some carbonate and cherts beds/layers retain primary bed-
ding and/or lamination (Fig. 7.66a–c). These lithologies
represent background sedimentary rocks comprising the
bulk of the Zaonega Formation. Although the organic matter
is mainly bound to residual kerogen, many beds are enriched
in pyrobitumen, originally petroleum. The pyrobitumen
occurs as intergranular infill in dolostones (Fig. 7.66k),
small droplets and bedding-parallel films or as extensive
impregnation in sandstone beds. In the latter case, clast
particles float in pyrobitumen matrix implying oil generation
prior to cementation (Fig. 7.73f).
The secondmode (15–45wt.%) in the total organic carbon
distribution histogram includes mainly massive or brecciated
rocks rich in Corg (Fig. 7.66m), mudstones exhibiting no or
indistinct layering, and few laminated pyrobitumen-rich
mudstones. The massive or brecciated Corg-rich rock is
locally termed maksovite (Filippov 2002) (Si- and Corg-rich
rocks with fluidal pyrobitumen matrix) and will be consid-
ered in a separate section “Organosiliceous rocks” below.
Mudstones with no or indistinct layering represent thermally-
and chemically-modified organic-rich sediments that are
typically involved in the formation of peperite. Whether the
maksovite and the thermally-modified sediments are geneti-
cally related or not, remains to be studied.
The rocks forming the third mode at 45–75 wt.% organic
carbon occur only at the type locality at Shunga where they
are exposed in a small quarry and an adit (Fig. 7.69). Here,
they appear as black, semilustrous, massive rocks with
conchoidal fracture; some varieties are indistinctly bedded
or show weak parting (Fig. 7.66n–p). Up to ten beds of such
organic-rich rocks occur within a c. 5-m-thick siltstone-
dolostone-chert section. The beds are from 0.2 to 1 m thick
with the lowermost unit resting on a bedded siltstone-
mudstone, while the uppermost is overlain by either a
dolostone or a chert. The organic-rich beds are separated
from each other by thin interlayers of grey dolostone. Pri-
mary lamination in these organic-rich rocks is obscured due
to high Corg content. A substantial portion of organic matter
is represented by pyrobitumen, which might have migrated
in from other source rocks. The rocks grouped in mode three
were termed the oil shales (Melezhik et al. 1999a). Interest-
ingly, FAR-DEEP Hole 13A (see Chap. 6.3.4), located c.
200 m to the south-west of the Shunga quarry (Fig. 7.69), did
not intersect the “Shunga-type” Corg-rich rocks shown in
Fig. 7.66n–p, thus implying their limited lateral extent.
The fourth mode, with the highest Corg concentration
(c. 95 wt.% on average), belongs to vein-pyrobitumen
(Fig. 7.66q) and represents former petroleum trapped in
different environments.
Time Constraint on Oil Generation
The precise absolute timing of oil generation remains
unknown apart from a preliminary Re-Os age of c. 2050 Ma
(Hannah et al. 2008) obtained from organic material
collected from the “mode-three” rocks (Fig. 7.66n–p) at
Shunga. The relative age, however, can be constrained
based on the interaction between source rocks, generated
oil, breccias containing pyrobitumen clasts and breccias
with pyrobitumen-rich matrix, mafic lavas and gabbro sills.
The gabbro sill and associated peperite, occurring in the
lower part of the drill section (Hole 12B, 490–410 m,
see Chap. 6.3.3), contain abundant contraction joints filled
with pyrobitumen, chlorite, calcite and sulphides. The pyro-
bitumen occurs in different morphological forms, including
cylindrical, platy, globular as well as cauliflower-like and
graphic morphologies: all are emplaced into a chlorite or
calcite matrix (Fig. 7.70a–e). The presence of pyrobitumen
in contraction joints of gabbro and peperite suggests that the
liquid hydrocarbon was generated either synchronously with
or prior to the emplacement of the mafic magma into non-
lithified sediments.
Similar to the gabbro, basaltic lava flows (e.g. Hole 12A,
94.5–56 m) contain pyrobitumen veins, chlorite-pyrobitumen
veinlets and vesicles filled with pyrobitumen (Fig. 7.70f–h).
Lava-related peperite injected into Corg-rich sediments shows
columnar joints, which are cemented by pyrobitumen, chlo-
rite and calcite (Fig. 7.70i–j). Thus, both examples suggest
that the oil had already been generated prior to or synchro-
nously with the emplacement of lava flows.
1204 H. Strauss et al.
The Zaonega Formation contains numerous breccias
occurring at different depths. Several breccia bodies are
composed of different-size particles and fragments of pure
pyrobitumen (Fig. 7.70k, l), soft-sediment deformed and
angular fragments of greywacke, siltstone, carbonate rocks,
cherts as well as pyrobitumen-rich sandstone (Fig. 7.70l, m);
in some cases, clasts are supported by a pyrobitumen-rich
matrix (Fig. 7.70n). Incorporation of pure pyrobitumen
material (primarily oil) and pyrobitumen-rich sediments
(primarily oil shale) into syndepositional, slope-slide or
peperite-associated explosive breccias strongly suggests
that the generation of liquid petroleum was an early phe-
nomenon in the burial process.
In general, oil can be expelled from organic-rich rocks,
usually shales containing >1 wt.% total organic carbon
(Tissot and Welte 1984; Hunt 1996) when they are buried
and subjected to increasing temperatures and pressures.
There are three stages to hydrocarbon formation during
sediment burial termed diagenesis (<50–60 �C), catagenesis(from >50–60 �C to <150–200 �C) and metagenesis
(>150–200 �C) (Fig. 7.71a). Diagenesis involves the
biological, chemical, and physical alteration of organic
material before heating begins to affect it. Catagenesis
corresponds to the burial stage when, at about 60 �C, oilbegins to form in the source rock due to the thermo-
genic breakdown (cracking) of organic matter. This special
environment is called the “oil window” (Hunt 1996). The
third stage, metagenesis, corresponds to high temperature
(>200 �C) alteration. It is also known as the “gas window”.
Thus, oil is generated and expelled from the source
rocks when the latter pass through the “oil window”, a
temperature-dependent interval in the subsurface repre-
senting the period of time during which organic matter
is thermogenically transformed into hydrocarbons. Areas
where Earth’s crust is thin have high thermal gradients,
while areas with a thick crust have a lower geothermal
gradient. In areas with “normal” thermal gradient of
25 �C/km (Geothermal gradients 2011), the “oil window”
with a temperature interval of 60–120 �C corresponds to a
burial depth of c. 2–4 km (Fig. 7.71b). However, some areas
have much higher heat flows (e.g. 150 �C/km, Jiracek et al.
1986) because of deep fault zones, rifting, magmatic
intrusions, or active tectonic forces. In such areas, the oil
window may exist at shallower depths. The rift-bound
Zaonega succession with abundant mafic intrusions and
lava flows apparently represented one such environment
with an enhanced geothermal gradient.
The maximum thickness of the Zaonega Formation
intersected by the 3,500-m-deep Onega parametric hole
(Krupenik et al. 2011a) is c. 1,500 m (Fig. 7.68). There is a
clear indication that generated oil (pyrobitumen now) was
already involved in syn-sedimentary slumping at a depth of
c. 250 m in FAR-DEEP Hole 12B (Fig. 6.116cy, cz in Chap.
6.3.3). The referred depth corresponds to themiddle part of the
1,500-m-thick Zaonega section (Fig. 7.68), implying that the
lower part of the succession (inferred potential source rocks)
were buried at a depth of c. 750 m, and produced oil at a much
shallower depth compared to basins with normal geothermal
gradient (e.g. 25 �C/km). Perhaps coincidentally, this depth
also corresponds to that at which there is a distinct change in
organic carbon d13C, from around �25 ‰ to less negative
values (see Fig. 7.68). Several lines of evidence indicate that
the amount of generated and migrated hydrocarbons consider-
ably increased while burial advanced: the presence of a
syndepositional breccia containing pyrobitumen-rich clasts at
a depth of c. 202 m in Hole 12B (Fig. 7.70k–m) – also coinci-
dent with an inflection in the carbon isotope stratigraphic
pattern, a massive oil spill at a depth of c. 150 m represented
by the massive pyrobitumen-rich rocks (up to 40 wt.% Corg;
for details see section “Organosiliceous rocks”), mafic lava
flows intruding into seafloor-oil seeps at a depth of c. 100 m,
and upwards (Fig. 7.66j), and abundant breccias with
pyrobitumen-rich matrix in the upper part of FAR-DEEP
Hole 13A section (see Chap. 6.3.4).
Oil Migration Pathways
After expulsion from the source rock, the oil/gas (lighter
than water) migrates upwards through permeable rocks
(sandstones) or fractures until being stopped by a seal,
a tight, non-permeable layer of rock, like a shale. There is
plentiful evidence of liquid hydrocarbon migration
throughout the Zaonega Formation. In most cases, former
oil migration pathways appear as 0.1- to 5 cm-wide,
pyrobitumen-filled veinlets and veins cutting different
lithologies (Figs. 7.70f and 7.72a–d). The veins were mostly
developed in open, ductile or semiductile, extensional
cracks, which show semiplanar and parallel walls (Figs. 7.70f
and 7.72b, d). The veinlets may occur as a few millimetre-
thick, single, solitary, wall-parallel joints, while others form
a stockwork-like system (Fig. 7.72e). In places, pyrobitumen
veinlets can be traced to source layers (Fig. 7.72a, d, f).
Some veins show a ptygmatic appearance suggesting a
later compaction (Fig. 7.72a). Many veins and veinlets are
zoned parallel to their walls (Fig. 7.72g–i), implying multi-
phase hydrocarbon migration. The walls are smooth and may
even have transitional contacts with the host rocks
suggesting only partial lithification. In zoned veins, margins
are commonly more enriched in pyrobitumen while the
central parts are composed of pyrobitumen-rich sediment-
mush (Fig. 7.72h, i) resembling sand dykes compositionally
and texturally. The multiphase hydrocarbon migration is
also supported by the presence of multiple cross-cutting
pyrobitumen veinlets, which occur within a thicker pyrobit-
umen vein (Fig. 7.72j). Some extensional pyrobitumen-filled
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1205
joints were reactivated resulting in formation of brittle
extentional cracks parallel and perpendicular to the original
vein, and filled with quartz and/or calcite (Fig. 7.72k).
In places, dolostone-hosted pyrobitumen veins contain
vesicles. The vesicles are slightly elongated spheres
and exhibit a bimodal size distribution (c. 50 � 30 and c.
5 � 3 mm). In parts of the veins, elongated vesicles
are aligned into a concentric pattern (Fig. 7.72o). Small-
sized vesicles mostly remain unfilled, and their walls
show a desiccated texture (Fig. 7.72n). Larger vesicles are
filled with chlorite, calcite, pyrite, sphalerite and galena
(Fig. 7.72l–n) or with silicate minerals (Fig. 7.72p). The
overall significance of the vesicular pyrobitumen veins in
dolostones remains to be ascertained. Very likely, such
pyrobitumen was originally liquid oil which was solidified
and degassed during the course of thermal cracking.
Published d13Corg values (n ¼ 4) obtained from bulk
samples of vein pyrobitumen in the Zaonega Formation ranges
between �30.3 ‰ and �27 ‰ (Filippov and Golubev 1994;
Melezhik et al. 1999a) and cluster on the total d13Corg histo-
gram within the isotopically heavy mode (Fig. 7.67c–e). This
is heavier than the presumed biological source signature (the
suggested �34 ‰ by Melezhik et al. 1999a). However, it is
unknown whether the isotopic composition of the veinlet
pyrobitumen reflects a local or a distant source, and whether
or not sources were homogeneous.
Oil Traps
Reliable information is not available on whether or not
large-scale traps existed in the Onega Basin prior to
Svecofennian deformation and metamorphism. The most
voluminous pyrobitumen masses (originally oil) have
been found in drilled sections trapped within brecciated
dolostones. Here, the pyrobitumen fills either the space in
between the dolostone fragments or fractures of various
scales (Fig. 7.73a–c). In several stratigraphic intervals the
pyrobitumen fills intergranular space in crystalline dolo-
stones (Fig. 7.66k). A substantial volume of pyrobitumen was
trapped in jointed sandstones and siltstones (Fig. 7.73d, e).
In some coarse-grained, graded beds, the oil migrated
towards interbed space where it occurs now as sub-
millimeter thick films (Fig. 7.73f). However, the most spec-
tacular oil traps preserved in the Zaonega Formation are
interlayer and interbed openings with the best examples at
Shunga village (Fig. 7.69a, b). Here, the former liquid petro-
leum occurs in interbed openings located above a bed of
organic-rich rocks (55–65 wt.%; Melezhik et al. 1999a) and
below a thick dolostone bed capped by a thick unit of chert
(Figs. 7.69a, b and 7.73g–i). The thickness of the interbed
openings ranges from a few to 62.5 cm and is filled with a
solid pyrobitumen: a homogeneous, massive and black
organic substance with conchoidal fracture (Fig. 7.73h),
locally termed shungite (Filippov 2002). The pyrobitumen
also fills intergranular space in crystalline dolostones which
cap the pyrobitumen layer (Fig. 7.66k).
d13Corg of the pyrobitumen layer (oil trap, Fig. 7.73h),
intergranular pyrobitumen, and from bulk organic matter
(kerogen + pyrobitumen) in the rocks below the oil trap
(Fig. 7.73i) at Shunga, ranges between �37.6 ‰ and
�37.2‰ (Melezhik et al. 1999a), thus clustering in the
total d13Corg histogram in the isotopically lighter mode
(Fig. 7.67c–e).
Organosiliceous Rocks or Maksovite
The Type Locality MaksovoA typical representative of the organosiliceous rocks is the
Maksovo deposit (for location, see Fig. 7.63), from which
these rocks are exploited as a flux substitute in cast iron
production. The organosiliceous rocks were locally termed
maksovite (Filippov 2002): Si- and Corg-rich rocks with
fluidal pyrobitumen matrix. The maksovite occurs in the
type locality as black, mat, massive cryptocrystalline rocks
with conchoidal fracture (Fig. 7.74a). Well-pronounced
columnar joints are always developed in the maksovite
when it is in contact with gabbro (Fig. 7.74b–d). The
rocks are opaque in transmitted light. The total organic
carbon content ranges between 16 and 53 wt.%. X-ray dif-
fractometry suggests that the organic carbon-free mass is
mainly composed of quartz with subordinate sericite, calcite,
chlorite and feldspar (Filippov et al. 1994). One of the most
important genetic characteristics of the maksovite is its
structural organisation. Although the rocks are massive,
they often exhibit heterogeneity at different scales. On the
macro-scale, such heterogeneity is expressed by the pres-
ence of rare mm-size, rounded and elongated pyrobitumen-
rich particles, sulphides and shale fragments dispersed in a
pyrobitumen-rich matrix (Figs. 7.66m and 7.74e). More
important and more characteristic is a micro-structural
heterogeneity of the matrix expressed by the irregular dis-
persion of micron-size quartz particles (with minor clasts
of other minerals and sedimentary rocks) supported by a
pyrobitumen mass with fluidal structures (Fig. 7.74f). In
the following text, this microstructural pattern is referred to
as “maksovite-type matrix”. Because the two major
constituents of the rocks are quartz and pyrobitumen, such
rocks were earlier termed the “organosiliceous rocks”
(Melezhik et al. 2004). Although currently a tendency exists
to apply the term maksovite to any massive rock rich in
organic matter, the rocks with the above-mentioned struc-
tural, textural, geochemical and mineralogical charac-
teristics have been reported so far only from the middle
part of the Zaonega Formation (Figs. 7.65b and 7.68).
1206 H. Strauss et al.
The maksovite forms cupola-like or lensoidal bodies but
occurs also as veins that cross-cut bedded siltstones. Geo-
logical aspects of maksovite were studied in detail in the
Maksovo deposit, which was intensely drilled during
the 1960s and quarried ever since. The results are published
in Russian, mainly local, journals and booklets (e.g.
Kupryakov and Mikhailov 1988) with an English summary
presented in Melezhik et al. (2004). At Maksovo, the
maksovite structurally forms an asymmetrical, flat, cupola-
shaped or lensoidal body exposed in the core of a small
antiform. The body has an ellipsoidal outcrop pattern with
axes of 700 m and 500 m (Fig. 7.75). It is composed mainly
of massive, jointed and brecciated varieties of maksovite.
Although the spatial distribution of the massive and
brecciated rocks is complex, the massive maksovite is
mainly associated with the footwall and the core of the
body (Fig. 7.76a, profile C–D). Syndepositional maksovite
breccias occur along the upper margin of the maksovite body
(Fig. 7.76a). Similar vertical distribution was documented
for so-called “vuggy” maksovite, though it may also occur
inside the lens (Fig. 7.76a).
The greatest thickness of the body is c. 120 m though it
decreases to less than 35 m towards the periphery of the lens
(Fig. 7.75). The foot-wall and hanging-wall sedimentary
strata comprise basaltic tuff, volcaniclastic siltstone, grey-
wacke, dolostone, limestone and chert. A greywacke bed
located beneath the maksovite lens is enriched in carbonate
minerals. Both the host rocks and the maksovite lens are
intruded by a gabbro sill (Fig. 7.75). Maksovite shows well-
developed columnar joints at the contact with the gabbro
(Fig. 7.74b–d) whereas the host sedimentary rocks do not.
The lateral contact relationships as well as the upper and
lower contacts of the maksovite body with the country rocks
are not described in accessible published scientific literature.
Examination of numerous drillcores suggests that the lower
contact is sharp and depositional. The nature of the upper
contact of the maksovite body remains poorly constrained.
A core collected from drillhole 202 at a depth of 36.8 m
corresponds to the upper contact of the maksovite lens
and appears to be a syndepositional maksovite breccia
(Fig. 7.74g). The breccia consists of fragments of silica-
rich and pyrobitumen-rich rocks in a maksovite-type matrix
with a fluidal structure. Pyrobitumen-rich clasts have a
rounded shape and show diffuse boundaries against the
matrix, from which they are distinguished by a lower
content of rounded quartz particles and higher content of
pyrobitumen mass. In contrast, the sandstone clasts exhibit a
variable shape and are enveloped by fluidal pyrobitumen
matrix (Fig. 7.74g). These fragments contain numerous
vugs and vesicles filled with pyrobitumen (former petro-
leum). The latter implies that these rocks went through the
“oil window” at depth, and hence were not incorporated into
the breccias from surface or near-surface sediments.
Massive maksovite represents the main lithology in the
Maksovo deposit. The rock is black and dark grey, mat,
cryptocrystalline (Fig. 7.74a, e) and opaque in transmitted
light. Microscopically the massive maksovite exhibits a
structural heterogeneity expressed by the dispersion of
micron-size, rounded, quartz particles, and larger platy,
pyrobitumen-rich particles, as well as rounded, partially
disintegrated siltstone and sandstone fragments in a pyro-
bitumen framework showing fluidal structures (Fig. 7.74 h).
The texture is described as indicative of disintegration
caused by fluidisation and multi-phase migration in a non-
lithified state (Melezhik et al. 2004).
Massive maksovite rocks may in places contain abundant
vugs and were previously termed vuggy maksovite
(Melezhik et al. 2004). Although they prefentially occur in
the upper part of the lens, there are also small pockets of
such rocks in the northwestern margin of the lens (Fig. 7.76a,
profile C–D). The spheroidal vugs are 3–5 mm in diameter
and were formed due to shrinkage whereas larger irregular
vugs resemble gas/fluid-escape structures (Melezhik et al.
2004). The smaller vugs are filled with pyrobitumen
(Fig. 7.67g, i) whereas in the larger ones, quartz lines walls
of the vugs and pyrobitumen occupies their centres. Quartz
filling the vugs is commonly characterised by a concentric
microfabric and globular structure; the pyrobitumen infill
displays quartz-filled syneresis cracks.
On both flanks of the lens, the massive maksovite passes
into a type termed the ‘cryptic’ breccia (Melezhik et al.
2004), a black, mat rock with lustrous specks and cryptic,
flaser-like and fluidal structures. The term ‘cryptic’ breccia
was applied to describe the structure which is distinguished
by chaotically distributed mm- to cm-size, angular and
flame-shaped, partially dispersed fragments of microcrystal-
line quartz in a maksovite-type matrix (Fig. 7.74i), resulting
in three main types of fabric: soft-sediment brecciated,
flaser-like and irregular fluidal. Many quartz fragments
exhibit concentric growth and thus apparently were origi-
nally formed hydrothermally. The matrix represents a typi-
cal maksovite-type material; it is composed of micron-size,
elongated and rounded quartz, and minor chlorite and silt-
stone particles distributed in a pyrobitumen framework.
The ‘cryptic’ breccia is specifically marked by syneresis
cracks occurring from micro- to macro-scale. The syneresis
cracks have been observed in both quartz and pyrobitumen
that infill voids (Fig. 7.74j, k). Ubiquitous syneresis cracks
indicate that at least part of the silica was originally a
gel. The ‘cryptic’ breccia contains abundant, mm-size,
differently-shaped quartz fragments with spectacular
micron-scale concentric structures (Fig. 7.74l–o). Concen-
tric microfabrics suggest involvement of hydrothermal pro-
cesses. These peculiar fragments are commonly elongated in
parallel with the fluidal fabric. In many cases the concentric
microfabrics are obliterated along marginal parts of the
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1207
fragments and replaced by a rim of homogeneous silica of
irregular thickness (Fig. 7.74l–o).
The overall textural development observed in the ‘cryp-
tic’ breccia was explained by Melezhik et al. (2004) as the
dispersion and fragmentation caused by multiple fluidisation
processes of organosiliceous mush.
Jointed maksovite is distinguished from the massive vari-
ety by joints (Fig. 7.74o, p), whereas cryptocrystallinity is
retained and the principal mineralogical and chemical
compositions remain unchanged (Filippov et al. 1994). The
rocks form lenses tens of metres thick with no preferential
spatial distribution with respect to the body. One- to two-cm-
long joints form polygonal systems, a combination of con-
centric and radial systems, or a system of orthogonal joints.
The jointed pattern is similar to that resulting from syneresis
and was considered to represent shrinkage cracks formed by
the spontaneous expulsion of water or other liquids from the
organosiliceous gel during ageing (Melezhik et al. 2004).
The open joints are filled with pure pyrobitumen or
pyrobitumen-rich material (Fig. 7.74o, p). Joints filled with
pure pyrobitumen (former liquid oil) or pyrobitumen-rich
material are indicative of fracturing before the hydrocarbon
was solidified, hence prior to metamorphism.
The quartz-cemented maksovite breccia tends to occur in
the upper part of the body. Angular maksovite fragments,
ranging in size from 1 to 5 cm, are cemented by white
crystalline quartz (Fig. 7.74q). The volume of quartz may
reach 30 % in zones of intensive, postdepositional, tectonic
brecciation. Many fragments exhibit in situ brecciation with
limited rotation and tectonic modification, implying that the
brecciation was associated with an extensional, decom-
pressional regime.
Based on bulk analysis, the isotopic composition of
organic carbon from the maksovite exhibits a considerable
spread between �40 ‰ and �25 ‰, overlaps with the total
d13Corg range measured from Zaonega Formation sedimen-
tary rocks, and similarly exhibits a bimodal distribution
(Fig. 7.77a). Processes causing such fluctuations have not
been identified. Unpublished in situ laser-combustion
analyses (for a description of the method, see Bruneau
et al. 2002) suggest a similar spread though with three
outliers, two at around �66 ‰ and one at �46 ‰(Fig. 7.77b); these low values remain to be verified by in
situ-based ion-probe analysis. However, d13C of around
�45 ‰ was also measured by conventional methods for a
migrated pyrobitumen from Shunga (see Fig. 7.77a). The
data in Fig. 7.77b show a pronounced skew to values lower
than the dominant mode at �29 ‰, which coincidentally
covers the same range as the trough between the two promi-
nent modes in Fig. 7.77a. A new study using high spatial
resolution ion microprobe techniques (SIMS) is advocated,
because confirmation of spatially-resolved very low d13Cwould strongly implicate methanotrophy and an
interpretation that the broad, moderately low isotopic mode
at d13C of ~�32 ‰ to �38 ‰ might then reflect metabolic
products of a mixed community of microorganisms with
about one quarter of the biomass produced by
methanotrophs and the rest by autotrophic carbon fixation
(at �25 ‰). One might even speculate that the apparent
tendency in Fig. 7.77a for migrated organic carbon to be
predominantly of low d13C could be related to chemical
differences between the two categories of biomass.
Regardless of their lithological characteristics and types
briefly outlined above, the maksovite at the Maksovo deposit
is composed of two basic components (Table 7.4). These are
39–77 wt.% silica, mainly in the form of quartz, and
16–53 wt.% Corg in the form of pyrobitumen with a fraction
of kerogen. The rocks are low in sodium (Na2O <0.06 wt.%
on average) but contain a sizable amount of potassium (K2O
> 0.8 wt.% on average). At Maksovo, the maksovite body
exhibits vertical and lateral zoning in terms of total organic
carbon content and SiO2/Al2O3 ratio, with the rocks most
enriched in pyrobitumen and silica located within the central
part of the lens (Fig. 7.76b, c, profile E–F).
When maksovite chemical composition is calculated on
organic carbon-free basis, the aluminosilicate residue exhi-
bits a highly siliceous nature (Fig. 7.78) with an average
SiO2 content of 83.2 � 8.5 wt.% (n ¼ 107). The total
organic carbon content ranges between 15.3 and 54.2 wt.%
and has an average of 34.2 � 9.2 wt.% (n ¼ 107).
FAR-DEEP Hole 12BSeveral lenses of organosiliceous rocks were intersected by
FAR-DEEP Hole 12B, which is located c. 1.5 km to the
north of the Maksovo deposit (Fig. 7.64). Whether or not
such lenses represent a time-equivalent section to the
maksovite body at Maksovo, remains unproven. We infer
that the intersected lenses are represented by maksovite per
se based on their great compositional, and macro- and
microstructural similarities with the maksovite from the
type locality.
Although the Hole 12B site was less densely drilled, the
available holes collectively allow to reconstruct the
maksovite body in three-dimensional space (Fig. 7.79).
Maksovite occurs in Core 12B within the 156.1–132.9 m
interval as five discrete stratiform bodies of variable thick-
ness and one crosscutting vein (Fig. 7.80). The lowermost
body (156.1–138.9 m) is the thickest of all and composed of
monotonous massive maksovite (Fig. 7.80a). The body
appears to be an up to c. 45-m-thick (in drillhole c-19),
ellipsoidal lens, which extends laterally from west to east
over a distance of c. 350 m, whereas its north–south extent is
unknown, though exceeding 200 m. The lower contact of the
maksovite with massive limestone is depositional with a
c. 1-cm-thick, chlorite-rich layer in between (Fig. 7.80b).
The upper contact with organic-rich shale is tectonically
1208 H. Strauss et al.
modified, though the body appears to be stratigraphically
conformable with the hosting sedimentary strata in three-
dimensional space (Fig. 7.79).
Although the maksovite comprising the main body
appears to be rather massive, a macrostructural heterogene-
ity is present and expressed by rare and irregularly-scattered,
rounded fragments of pyrobitumen-enriched rocks, a few
millimetres in diameter (Fig. 7.74e). Microstructural pattern
of the maksovite is similar to that of the Maksovo deposit.
The framework (matrix) is mainly composed of rounded and
elongated particles of quartz, ranging from 1 to 5 mm in size
and having mammillated surface, and larger fragments of
partially disintegrated siltstone and sandstone; pyrite grains,
cubes and lumps, and phlogopite “balls” are also present as
minor components. All these particles are embedded in
pyrobitumen displaying together with elongated mineral
and rock particles a fluidal microfabric (Fig. 7.80c–f). This
“fluidal” pyrobitumen-rich matrix envelopes, and is plasti-
cally injected in between larger rock fragments and pyrite
nodules, implying a nonsolidified nature of this sandy
pyrobitumen, apparently originally sandy oil-mush. Rock
fragments embedded in the sandy pyrobitumen matrix
range in size from a few microns up to 1 cm. Variably
sized sandstone and siltstone fragments show partial disinte-
gration and produce “rip-offs” of smaller clasts or individual
mineral grains incorporated into the pyrobitumen-rich
matrix (Fig. 7.80e). In some clasts, pores are filled with
pyrobitumen (originally oil) while in other clasts, they are
filled mainly with kerogen (Fig. 7.80e). In both cases, rock
clasts do not show a visible cementation, and thus represent
kerogen- or oil-rich sediments prior to their incorporation
into the maksovite-type, sandy, bitumen mush (originally
sand-oil mush). There are also micron-size “balls” and
“plates” of pyrobitumen-rich (up to 80 vol.%) material
containing tiny, rounded, clastic particles of silicate and
aluminosilicate minerals. Pyrobitumen-rich “balls” com-
monly exhibit early soft-sediment internal folding, whereas
pyrobitumen plates retain unmodified horizontal lamination
(Fig. 7.80d). The overall microstructural pattern of the
maksovite can be explained by multiple transport of non-
solidified sandy oil-mush involving internal disintegration.
Four other maksovite beds, two nearly consecutive at the
depth of 138.6–137.7 m, one thin bed at 137.24–137.05 m,
and the uppermost bed at 136.6–136 m are breccias. The
lowermost bed is composed of maksovite showing partial
soft-sediment disintegration into large clumpswith remaining
space filled with black, organic-rich mudstone. These rocks
lie on laminated siltstone and mudstone containing fragments
of maksovite, and are affected by soft-sediment deformation
(Fig. 7.80g). Although these sediments experienced a signifi-
cant syndepositional deformation, their bedding is grossly
coherent with general stratification. Compositionally and
microstructurally, the partially disintegrated maksovite and
the maksovite fragments embedded into laminated siltstone
are identical (Fig. 7.80h–k) to the maksovite occurring in the
main body. The similar pyrobitumen-rich matrix shows a
fluidal fabric, plastic deformation and multiple transport.
Pyrobitumen-supported, micron-size clasts are represented
by rounded quartz grains with mammillated surface, partially
disintegrated siltstones, sericite particles, pyrite grains, and
platy, pyrobitumen-rich (80 vol.%) fragments with inherited
and preserved parallel lamination (Fig. 7.80k). Rounded,
pyrobitumen-rich clasts exhibit an earlier generation of soft-
sediment deformation (Fig. 7.80h). Some pyrobitumen-rich
“balls” show a concentric structure with distinct cores
enriched in pyrobitumen (Fig. 7.80i). There are abundant
“balls” composed of platy aggregates of phlogopite (Fig.
7.80j); these apparently represent former oil-clay “balls”.
The uppermost maksovite breccia-bed is somewhat simi-
lar to the lowermost one though soft-sediment disintegration
did not affect the uppermost portion of the bed, which,
however, experienced syndepositional, soft-sediment defor-
mation resulting in formation of a wavy topography of the
bedding plane (Fig. 7.80l). Uneven topography was evened
by deposition of a graded sandstone and then buried beneath
organic-rich, laminated siltstone and shale, thus preserving
crucial evidence that the maksovite bed was deposited on a
seafloor.
The maksovite vein was intersected by FAR-DEEP Hole
12B at a depth interval of 133.7–132.9 m (Fig. 7.80m–o).
Both contacts are well preserved and cut host-rock bedding
at oblique angles. The hanging-wall contact is straight
(Fig. 7.80m, o), whereas in the footwall of the lens, the
host sedimentary rocks were injected with small maksovite
off-shoots (Fig. 7.80m, n). Macro- and microstructural
patterns of the maksovite vein are similar to those described
above for the massive maksovite.
Based on bulk analysis, the isotopic composition of
organic carbon of the maksovite is homogeneous, ranges
between �26 ‰ and �24 ‰ (n ¼ 4), and overlaps only
with the most 13C-rich maksovite from the Maksovo deposit.
However, the isotopic composition obtained from the
maksovite vein shows a significant depletion in 13C (d13Corg
¼ �29.2 ‰) and plots within the low-d13C mode
(Fig. 7.77a). The d13Corg values reported from maksovite
intersected by the Onega parametric hole overlaps with the13C-depleted end of the Maksovo maksovite (Fig. 7.77a).
The isotopic differences in maksovite from geographically
different locations remain unexplained.
The chemical composition of the stratiform maksovite
intersected by FAR-DEEP Hole 12B (Table 7.5) is some-
what similar but not identical to that from the Maksovo
deposit. In both cases, the silica and organic carbon in the
form of pyrobitumen are the major components, though
Al2O3, Na2O and K2O abundances measured in Core 12B
are within the maximum range reported from the Maksovo
deposit (Table 7.4). If the maksovite chemical composition
is calculated on organic carbon-free basis, the
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1209
aluminosilicate residue exhibits a highly siliceous nature
(74–79 wt.%, n ¼ 3), whereas maksovite from the vein is
less siliceous (60 wt.%). Interestingly, microprobe analyses
indicate that even pyrobitumen from the maksovite matrix
contains a fraction of silica though in a finely-dispersed form
that cannot be resolved by electron microscopy.
Formation of Maksovite/Organosiliceous Rocks:A Seafloor Hydrocarbon Expulsion?The genesis of the maksovite remains unresolved. There
are several main issues to be considered: (1) silica source,
(2) pyrobitumen source, (3) silica-pyrobitumen mingling
process, (4) emplacement/depositional mechanism. Several
prominent features allow a clear distinction between the
maksovite and the sedimentary rocks known in the Zaonega
Formation. There are many features which make maksovite
different from normal bedded/laminated sediments, which
were deposited mechanically or precipitated chemically
from the water column: (1) Maksovite is massive and devoid
of any lamination or bedding (Figs. 7.66m, 7.74a, e, and
7.80a). It also differs from diagenetically formed concre-
tionary beds and lenses. (2) It was deposited on the seafloor
and interacted with unlithified sediments (Fig. 7.80l), and
occurs as veins, thus showing an intrusive nature as well
(Fig. 7.80m–o). Maksovite differs compositionally from any
chemically precipitated (e.g. chert) or mechanically depos-
ited (e.g. sandstone, greywacke, siltstone) rocks of the
Zaonega Formation. (3) It is composed mainly of quartz (c.
50 wt.% SiO2) as detrital grains and authigenic precipitate,
and organic carbon in the form of pyrobitumen (c. 30 wt.%).
Maksovite is distinguished from common sedimentary rocks
microstructurally. (4) Its pyrobitumen-rich matrix exhibits a
fluidal fabric and multiple-phase plastic deformation, disin-
tegration and dispersion of pyrobitumen-rich material and
non-lithified quartz-rich sandstone and siltstone fragments
(Fig. 7.80d–f, h, i, k).
Based on studied examples from Phanerozoic oil fields
(Hedberg 1974; Gretener 1969), it was suggested that the
maksovite body at Maksovo may represent a relict diapiric
structure (Filippov and Romashkin 1994). However, feature
(2) is in conflict with the diapiric origin unless the diapir
penetrated the entire succession and extruded onto the sea-
floor. The model has not been supported by an actual
observation of diapirs per se, and supposes penetration of
sedimentary strata, unless such diapirs are incipient. The
existing densely-drilled example, the Maksovo deposit,
demonstrates a conformable relationship between the mak-
sovite lens and the overlying strata (Fig. 7.75), and thus does
not support the diapir model. This is difficult to verify in
tectonically modified, deformed and faulted strata.
An alternative suggestion, a remnant of a mud volcano
(Melezhik et al. 2004), is not entirely consistent with the
feature (3); examples, described in the literature, report
neither mud containing a considerable amount of organic
carbon nor liquid hydrocarbon or bitumen (e.g. Lancea et al.
1998; Stadnitskaia et al. 2008). The model is verifiable if
feeder-channels to maksovite bodies are found; this, how-
ever, would require specially targeted intense drilling.
Another possible model invokes a hydrothermal system
that was supposedly initiated by heat produced during the
emplacement of mafic intrusive bodies (Fig. 7.81). Such heat
might have created the necessary temperature gradient for
early oil generation, subsequent thermal oil-to-gas cracking,
and initiation of shallow-seated, sub-surface, hydrothermal
circulation (Fig. 7.81). Silica, hydrothermally leached from
mafic rocks (or sediments), might mingle/mix with hydro-
carbon and gas (primarily CO2, CH4) extracted from the host
sedimentary rocks, and a gas-rich oil-silica-H2O fluid carry-
ing also sediment particles would have migrated into perme-
able beds (reservoir). Increased lithostatic pressure during
the course of subsequent basin subsidence may force gas-
rich oil-silica-H2O fluids to move either laterally within the
reservoir or vertically along zones of weakness. In the first
case, gas-rich oil-silica-H2O-sediment mush would have
formed stratiform maksovite beds entrapped within sedi-
mentary strata. In the second case, the result would be
crosscutting maksovite veins. If veins reached the seafloor,
the oil-silica-H2O-rich fluid with dispersed sediment clasts
(mush) may extrude forming a cupola-like or flat lensoidal
body interacting with unlithified sediments. During the
course of migration/transport, the oil-silica-H2O-sediment
mush might have experienced several stages of partial lithi-
fication, as well as fluidisation processes leading to the
formation of several generations of micro- and macro-
brecciated rocks with a pyrobitumen-rich fluidal matrix.
The model can be verified by identifying hydrothermally
leached, Si-depleted rocks (silica source) in the Zaonega
Formation, and components containing magmatic water in
the maksovite. In FAR-DEEP Hole 13A, one of the mafic
lava flow tops (depth 115.33 m) shows a considerable deple-
tion in SiO2 (c. 35 wt.%) with respect to inner parts of other
flows (47–50 wt.% at depth of 129–119 m, see Appendix
41), thus suggesting that Si was leached and liberated. FAR-
DEEP Holes 12A and 13A intersected turbiditic shales with
chert nodules (Fig. 6.116s, u, v in Chap. 6.3.3) and a thick
chert interval (39.3–33.2 m, Hole 13A). The chert nodules
indicate that the silica was largely available and mobile
during diagenesis. The thick chert interval, if it represents
chemical precipitation from ambient water, would imply that
the silica was delivered to the basin perhaps through seafloor
hydrothermal activity.
The presence of silica-Corg rocks as fragments with mul-
tiple micron-size, concentric zonation (Fig. 12l–n) in the
marginal maksovite breccia is indicative of hydrothermally
formed material. In addition, there is direct observational
evidence for a contribution of siliceous material to
1210 H. Strauss et al.
maksovite from sedimentary siltstone (e.g. Fig. 7.74h, from
the Maksovo quarry). The question arises, then, as to
whether there is any evidence for specifically magmatic
siliceous material associated with maksovite, which might
directly link the processes of mafic volcanism with hydro-
carbon generation and migration. One approach to address
this question is to consider silicate oxygen isotope ratios
(18O/16O reported as d18O in ‰ relative to V-SMOW); the
underlying assumption is the ‘central dogma’ of Taylor
and Sheppard (1986), stating that: “All relatively 18O-
rich or 18O-depleted silicate melts. . ...d18O � +4.5, or
d18O � +7.5 . . . must have in part been derived from, orhave exchanged with, a precursor material that once upon a
time resided on or near the Earth’s surface”. That is,
magmatically crystallised silicate should have d18O between
5 ‰ and 7.5 ‰, whereas hydrothermal silica (and chert) and
sedimentary rocks (greywacke, siltstone, etc.) should have
d18O > 7.5 ‰, and perhaps substantially so (by several per
mil). To investigate this, a pilot project on the oxygen
isotopic composition of bulk maksovite silicates (mainly
quartz) was carried out on samples from the Maksovo
deposit and FAR-DEEP Core 12A and 12B.
Following the measurement of d13C of bulk organic car-
bon (as described in Melezhik et al. 1999a), the non-
carbonaceous fraction was isolated from a separate aliquot
by intensive low temperature oxygen plasma ashing, and
d18O determined on the silicate residue by laser fluorination
(Macaulay et al. 2000). Data are presented in Table 7.6; d18Ovaries from +16.2 ‰ to +22.3 ‰, as expected for hydro-
thermal products and sedimentary rocks (e.g. Hoefs 2009)
and there is no evidence to support a significant contribution
from magmatic silicate, though this has not been definitively
excluded. Alternative silica sources, compatible with the
evidence from microscopy and oxygen isotopes, are
suggested by the presence of partially disaggregated
fragments of sandstone and siltstone in the maksovite,
which supply micron-size quartz grains to the pyrobitumen
matrix (Fig. 7.80e, k). Such silica-rich sand-silt material can
be inherited from sedimentary rocks in the stratigraphic
sequence and dispersed in oil-gas-water-sediment mush
(as emulsions?) during the course of its expulsion onto the
seafloor. Admittedly, very highly silica-rich clastic sedimen-
tary rocks do not seem to be present in the Zaonega Forma-
tion (Appendices 36, 39, 42). When all available chemical
analyses are calculated on an organic-free basis, no sand-
stone, siltstone or shale from the Zaonega Formation
matches the compositions seen in maksovites (Fig. 7.78),
and most clastic sedimentary rocks containing >5 wt.%
organic carbon are enriched in SiO2 compared to those
containing <5 wt.% organic carbon (Fig. 7.78). However,
incorporation of sedimentary rocks into the mush need not
have been wholesale. It has been noted that there is an
abundance of micron-size quartz grains within the
pyrobitumen matrix, so a size-separation effect during
sediment disaggregation and fluidisation of the mush is
feasible, possibly coupled with more exotic separation
mechanisms (e.g. density, electrostatic, surface tension
etc.). This idea is compatible with the observation of
aggregates of flaky phlogopite in the “clay balls” of
Fig. 7.80j. There is the possibility of colloidal or hydro-
thermal silica with silicon derived ultimately from the igne-
ous system. Depending on the timing of closure to oxygen
isotope exchange with aqueous fluids, this putative compo-
nent could perhaps have lost its igneous d18O signature (of
between 5 ‰ and 7.5 ‰) for a lower temperature or hydro-
thermal one (say d18O � 16 ‰), thus merging with the
range of values expected for sedimentary rocks which have
been through a weathering cycle (Fig. 7.82).
Despite several unresolved problems with the sources,
the origin and mechanism of emplacement, the available
sedimentological data briefly outlined above strongly indi-
cate that the maksovite signifies events of massive expulsion
of hydrocarbon-rich material onto the seafloor. The
consequences of such hydrocarbon debouchement on water
geochemistry and seafloor and water-column microbial life
might be significant, and remains to be studied.
Clastic Pyrobitumen and Surface Oil Seeps
A spectacular occurrence of pyrobitumen in the form of
inclusions and redeposited clasts occurs in a c. 80-m-thick
bed corresponding to the middle part of the 500-m-thick
Kondopoga Formation, which lies unconformably on Suisari
Formation basalts (see Chap. 4.3). The clastic pyrobitumen
has been reported to represent a surface oil seep: the first
occurrence ever reported from the Palaeoproterozoic
(Melezhik et al. 2009).
In the Kondopoga Formation, the most common
lithologies are grey volcanoclastic sandstones, siltstones
and mudstones (Fig. 7.83a–c), which form 2- to 15-cm-thick
rhythms with three- to four-units, corresponding to the A-D
units of a Bouma sequence deposited from turbidity currents
in a distal part of the basin (Melezhik et al. 2009). The
rhythmically bedded succession contains several tightly
spaced, c. 2-m-thick and 10-m-long lenses composed of
polymict breccia. Large clasts (up to 60 cm) of siltstones,
mudstones and sandstones, and fragmented and redeposited
carbonate nodules are embedded in a greywacke matrix and
show a limited degree of sorting and variable roundness.
Sedimentological features are consistent with massflow
deposition in a channelised environment. The average Corg
content in all lithologies is c. 1 wt.%. The rocks are generally
devoid of sulphides (Filippov and Golubev 1994). Within
thickly bedded turbiditic succession, minor sulphides occur
as dissemination in some pyrobitumen clasts. Sulphides
become more abundant in the uppermost, thinly-bedded
rocks where they appear in the form of Fe- and Cu-sulphides.
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1211
The presence of chalcopyrite suggests that at least some of
the sulphides are not primary. The general lack of sulphides
and abundant ankerite concretions in the lower and middle
part of the Kondopoga Formation are consistent with a fresh-
water lacustrine depositional system (Melezhik et al. 2009).
Both rhythmically bedded greywacke-siltstone and
massflow polymict breccias contain clasts and inclusions of
pyrobitumen (Fig. 7.83c–h) identified as former oxidised
petroleum (Mishunina 1979; Melezhik et al. 2009). In
bedded greywacke-siltstone, the pyrobitumen occurs as
1- to 10-mm-thick lenses in cross-sections, and ‘pancakes’
reaching 50 cm in diameter on the bedding surface
(Fig. 7.83e–f). In response to post-depositional compaction
or to shrinkage and desiccation, the pyrobitumen inclusions
show an intensive polygonal or linear cracking (Fig. 7.83f–h)
suggesting their transport to the depositional site in only
partially solidified state. Similarly, partial injection ofmaterial
from pyrobitumen inclusions into host sediments (Fig. 7.83g)
implies emplacement of pyrobitumen in the sedimentary
strata in a plastic state. Some pancake-shaped pyrobitumen
inclusions are surrounded by a dense impregnation of host
rocks by thin films of lustrous pyrobitumen (Fig. 7.83h),
which were apparently droplets of liquid hydrocarbon
squeezed out during post-depositional compaction. A few
pyrobitumen inclusions are densely coated at the margins by
foreign particles (Fig. 7.83i).
Massflow breccias and associated thick, massive sand-
stone beds contain the greatest concentration of the
pyrobitumen clasts and inclusions. These vary in size and
form. The ‘pancake’ type inclusions, 1–50 cm in diameter,
are abundant in the sandstones (Fig. 7.83d). Some inclusions
show partial injection into the host sediments. Millimetre-
size, angular pyrobitumen clasts are ubiquitous in both
sandstones and breccias (Fig. 7.83c). In addition, the
breccias commonly contain large pyrobitumen inclusions
0.5–5 cm in size, which retain a spherical shape. Variable
degree of roundness, compaction and injection into the host
rocks indicate that the bitumen was originally in the state of
variable degrees of solidification.
Sedimentological features and stratigraphic position of
source rocks and preserved former oil reservoirs presented
in Melezhik et al. (2009) suggest that the bitumen was
derived from surface oil seeps located near the lake. The
seep-derived hydrocarbon was affected by oxidation and
variable degree of solidification prior to erosion, transport
and redeposition into the lake by turbidity systems. There is
also indication of oil seeping through lake sediments
(Melezhik et al. 2009).
Fourteen samples of clastic pyrobitumen collected through-
out the Kondopoga succession exhibit d13Corg values cluster-
ing tightly between�36.0 ‰ and�35.4 ‰. These values plot
within the isotopically light mode of Zaonega Formation
organic matter (Fig. 7.77a) and indicate a single, isotopically
homogeneous source, which contrasts with the wide d13Corg
range documented for interbed-trapped pyrobitumen,
maksovite pyrobitumen and the entire organic matter in the
Zaonega Formation (Melezhik et al. 2009). The documented
isotopic difference remains unexplained and enigmatic.
Finally, the Kondopoga Formation demonstrates another
unique feature of the Palaeoproterozoic Shunga Event,
namely the preservation of the most ancient surface oil
seeps. Such seeps demonstrate that some or all Zaonega oil
reservoir seals were breached and oil was spilled onto the
surface. Since oil seeps are a very common attribute of
almost every major petroleum-producing province in the
world (e.g. Clarke and Cleverly 1990), the Kondopoga oil
seeps highlight the scale of oil generation and migration in
the Onega Basin.
1212 H. Strauss et al.
7.6.5 Possible Driving Forces for the Onsetof the Shunga Event and Implication of FAR-DEEP Core for the Shunga Event
Victor A. Melezhik, Aivo Lepland, Harald Strauss,and Anthony E. Fallick
Three FAR-DEEP drill holes intersected collectively more
than 700 m of Corg-rich rocks from the Zaonega Formation
recording the Shunga Event in the Onega Basin on the
Fennoscandian Shield. These will serve to unravel several
shortcomings in understanding the significance of the
unprecedented accumulation of Corg-rich rocks in the Early
Palaeoproterozoic and its causal relationship to other major
environmental upheavals during this time interval.
The overall driving forces for the onset of the Shunga
Event remain to be elucidated, but have to explain either
unusual environmental conditions for an unprecedented
magnitude in primary productivity, such as an extraordi-
narily high supply of nutrients, or favourable conditions for
the accumulation and preservation of an extremely high
percentage of the primary productivity.
Given a peak in the record of recognised ancient passive
margins at 1.9–2.0 Ga (Bradley 2008), the availability of
wide shelf areas and formation of epeiric basins with
enhanced sedimentation rates (Eriksson et al. 2005) may
have been important prerequisites for the accumulation of
the Palaeoproterozoic carbonaceous successions. Nutrients
supporting the high productivity in such epeiric basins would
have been supplied either by upwelling or by riverine input
from continental weathering. The latter, however, would
have been attenuated during time intervals post-dating the
early Palaeoproterozoic glaciation due to the changes in
atmospheric CO2 and weathering-enhanced feedbacks in
the carbon-silicate cycle (Berner 1993). Evidence for intra-
plate magmatism at c. 2.1 Ga (Ernst and Bleeker 2010) is
consistent with continental breakup. Newly formed basins
were likely to experience abundant sediment and nutrient
supply from erosion-prone nearby uplifted continental
margins, and hence may have served as depocentres for
carbonaceous sediments. Basinal configurations and sedi-
mentation regimes were also influenced by formation of the
global 2.1–1.8 Ga collisional belts (Zhao et al. 2002). Lastly,
extensive deposition of black shales may also be related to an
oceanic superplume event, which results in displacement of
seawater by oceanic plateaus and consequently a sea level
rise, and increase in hydrothermal activity and input of CO2
and CH4 into the ocean and the atmosphere (Condie 2004).
Most crucial in this respect is the precise timing of the
onset of the Shunga Event. A trial project on radiometric
dating of the Zaonega Corg-rich rock by employing the Re-
Os isotopic system was successful and yielded an age of
2.05 Ga (Hannah et al. 2008). Other studies successfully
employed this technique for dating sedimentary pyrite of
Palaeoproterozoic age (e.g. Hannah et al. 2003). This
indicates that the Re-Os technique can potentially provide
time constraints on the deposition of organic matter and
sulphides in the Zaonega Formation. However, the deposi-
tional (sedimentation and resedimentation from multiple tur-
bidity currents) and postdepositional (diagenesis, regional
and contact metamorphism) history of organic matter and
sulphides of the Zaonega Formation is complex as evident
from petrographic, geochemical and isotopic data (e.g.
Shatzky 1990; Melezhik et al. 2009). Hence, the crucial
prerequisite for a successful employment of the Re-Os tech-
nique is a careful selection of the least altered material based
on multidisciplinary research aimed at deciphering a sedi-
mentological control on deposition (selecting background
sediments) and a petrographic, geochemical and isotopic
control on postdepositional history (selecting the least altered
sulphides and kerogen). Although all these studies are yet to
emerge, the available FAR-DEEP archive data and initial
research indicate that several promising intervals for utilising
the Re-Os method exist in the uppermost and middle parts of
Core 13A at Shunga, and in the middle part of Core 12B at
Tetjugino (for details see Chaps. 6.3.3 and 6.3.4). In addition,
geochronologic studies of phosphorites in the Zaonega For-
mation containing xenotime and monazite and several
generations of apatite (for details see Chap. 7.8.2) may also
help to better constrain the age of the Shunga Event.
Another highly challenging problem is the significance of
and causes for the large d13Corg variations and the well-
pronounced stratigraphic trend from isotopically heavy to
isotopically light organic carbon (d13Corg as low as �42‰)
in the Zaonega succession. A respective understanding needs
to be based on firm constraints for the carbon isotopic frac-
tionation associated with primary production (including the
initial isotopic composition of primary biomass) as well as the
effects of microbial reworking in the sediment (i.e., the isoto-
pic composition of bacterial biomass and/or metabolic
products). Since the Zaonega Formation contains both
reduced (kerogen) and oxidised (carbonates) forms of carbon,
the task in hand can be achieved by comparison of the carbon
isotopic composition in coeval kerogen-carbonate pairs.
However, it is vital to make a confident distinction between
(1) carbonates precipitated from seawater and synchronously
with the organic matter (primary carbonates), (2) carbonates
resedimented from older sedimentary successions, and (3)
carbonates formed diagenetically in the pore water realm. If
V.A. Melezhik (*)
Geological Survey of Norway, Postboks 6315 Sluppen, NO-7491
Trondheim, Norway
Centre for Geobiology, University of Bergen, Allegaten 41, Bergen
N-5007, Norway
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_6, # Springer-Verlag Berlin Heidelberg 2013
1213
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1213
the primary carbonates are useful for tracking the isotopic
composition of primary biomass and alteration process
through comparison of their d13C with those of the coeval
kerogen, the d13C of diagenetic carbonates can assist in
deciphering microbial recyclers of the primary biomass.
Given the specific organic matter maturation and oil
generation process in the Zaonega Formation involving
the emplacement of magmatic bodies into wet, uncon-
solidated, Corg-rich sediments, and circulation of oil/
bitumen-rich hydrothermal fluids, formation of unusual car-
bonaceous structures can be expected. Previous studies have
provided evidence for the non-graphitizing nature of the
Zaonega carbonaceous matter and occurrence of various
structural varieties including stacked graphene and fullerenes
(Khavari-Khorasani and Murchison 1979; Kovalevski et al.
2001; Buseck et al. 1992, 1997). The occurrence of fullerenes
in the Zaonega Formation reported by Buseck et al. (1992)
could not be confirmed by subsequent studies (Mossman
et al. 2003; Ebbesen et al. 1995), possibly due to their
heterogeneous distribution. The link between the formation
of unusual carbonaceous structures and geologic history,
specifically the nature of organic matter maturation, oil/
bitumen solidification and fluid circulation is poorly
established currently, but likely holds the key for tracking
the structural evolution of carbon. Considering the recent
interest in carbon based nano-materials, detailed studies on
natural carbonaceous structures and their origin in the
Zaonega Formation appear warranted.
TheZaonega Formation ismarkedly different from its other
counterparts in the world by the intensive volcanism occurring
synchronously with sedimentation and by the greatest accu-
mulation of organic matter. On the Fennoscandian Shield, the
formation shares one thing in common with other Corg-rich
successions: the deposition occurred during the transition from
rifting to drifting and initial dispersion of the shield. Conse-
quently, there exists a potential for the investigation of possible
causal links between the accumulation of organic matter,
depositional conditions and the role of volcanism in nutrient
supply. This can be potentially achieved by comparison of the
Zaonega Formation with other successions, which formed
synchronously but accumulated in different depositional
settings. FAR-DEEP Cores 12A, 12B and 13A bears such
potential.
1214 H. Strauss et al.
Table
7.3
Palaeoproterozoic
c.2100–1900Maform
ationscontainingorganic-richsedim
entsthat
may
record
theShungaEventin
Fennoscandia
andelsewhere
Geologic
unit
Age(G
a)Lithology
Thickness
(m)
Corgcontent
(%)average/
max
Types
ofOM,d1
3C
(‰PDB)
Metam
orphic
grade
References
FennoscandianShield
ZaonegaFm.OnegaBasin,Karelia
<1.98�
27
Greywacke,siltstone,shale,dolostone,
chert
300–1,800
5/80
Kerogen
Greenschist
1,2,3,4
�17.4
to�4
3.5
Pyrobitumen
�27.1
to�4
4.4
Pilguj€ arviSed.F
m.,PechengaBelt,
Kola
2.06–2.0
Greywacke,siltstone,shale
1,000
2/11
Greenschist
5
Mennel,Ansemjoki,Kalloj€ arvi
fms.,PechengaBelt,Kola
<1.97to
>1.94
Greywacke,siltstone,shale,chert,tuff
~6,000
1/9
Greenschistto
amphibolite
5,6
Il’m
ozero
Sed.Fm.,Im
andra/
VarzugaBelt,Kola
<2.06
Greywacke,siltstone,shale
800–1,000
1/10
Greenschist
5
Sovaj€ arviFm.,Kuolaj€ arviBelt,
Karelia
<2.06
Siltstone,shale,dolostone,chert
>1,000
1.5/23
Greenschistto
amphibolite
5
SoanlahtiFm.,Savvo-Ladoga
Zone,Karelia
2.1–1.9
Black
shale
~1,500
5/50.4
Kerogen,
pyrobitumen
Greenschist
7,8
�26.1
to�4
1.0
MatarakoskiFm.,Central
Lapland
Belt,Finland
~2.05
Phyllite,black
schist,mafictuffand
tuffite,dolostone,BIF
100–500
�15.6
to�4
3.6
Greenschist
9,10
Porkonen
Fm.,Central
Lapland
Belt,Finland
~2.0
BIF,phyllite,black
schist,mafictuff,
chert
~300–400
Greenschist
42
Liikasenvaara
Fm.,KuusamoBelt,
Finland
~2.05
Mafictuff,dolomite,black
schist
>250
�16.6
to�1
7.0
Greenschist
9,11
SiuliunpaloFm.,Salla
Belt,
Finland
~2.05
Micaschist,graphiteschist,jaspilite,
dolostonee
200
Greenschist
12
Petonen
Fm.,Kuopio
Belt,Finland
<2.06
Dolostone,black
schist
53
�30.1
Amphibolite
9,13
Pet€ aikk€ oFm.(“MarineJatuli”),
NorthKarelia
Belt,Finland
<2.06
Dolostoneshale/schist
50
Kerogen
Greenschistto
amphibolite
9,14,15
�17.6
to�2
0.0
MartimoFm.,Per€ apohja
Belt,
Finland
1.97–1.90
Greywacke,phyllite,black
shale
0.2–11.5
Amphibolite
16
Haukupudas
Fm.,KiiminkiBelt,
Finland
Cutby1.83Ga
granites,<2.1
Greywacke,phyllite,black
shale,mafic
volcanics
2.7/25
Amphibolite
17,18
Talvivaara
Fm.,KainuuBelt,
Finland
1.97–1.90
Phyllite,black
shale
Cgrafaver.7–8
�24.7
to�2
8.3
Lower
tomiddle
amphibolite
19,20,21,
22,23
Outokumpuassemblage,North
Karelia
Belt,Finland
1.97–1.90
Micaschist,black
schist,serpentinite,
carbonaterock,skarnrock,quartzrock
�20.5
to�3
0.1
middle
Amphibolite
20,22,23,
24
Mounanahorst,Gabon
FB,FCandFD
fms.,Franceville
Basin,Gabon
~2.10
Black
shales
400–1,000
5/>
20
Kerogen
Greenschist
25,26,27,
28
�22.6
to�4
6.2
Pyrobitumen
�42.2
to�4
5.4
(continued)
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1215
Table
7.3
(continued)
Geologic
unit
Age(G
a)Lithology
Thickness
(m)
Corgcontent
(%)average/
max
Types
ofOM,d1
3C
(‰PDB)
Metam
orphic
grade
References
Voronezhmassiv
(KMA)
Tim
Fm.,OscolSeries
2.3–2.0
Black
shale
600
5/36.3
Kerogen
andgraphite
Greenschistto
epidote-
amphibolite
29,30
�27.4
to�3
1.1
NorthAmericanCraton(N
ain
Craton)
CodIslandFm.,Mugford
Group,
N.Labrador
1.88–2.0
Shale,chertdolostone,siltstone
Kerogen
Greenschist
31
31.8
to�3
2.2
WyomingCraton
NashFork
Fm.,LibbyCreek
Group
2.2–2.1
Dolostone,heteroliticsiliciclastics-
dolomite,carbonaceousshale
~1,700
20.4
Carbonaceousmatter
Greenschist
32
�6.9
to�3
0.8
KetilidianOrogen,Greenland
Grænsesø
Fm,VallenGroup;
Foselv
Fm,SortisGroup
1.85–2.13
Carbonaceousshale
150–600;
1,000
Kerogen
and
pyrobitumen
Greenschist
33,34
�22.3
to�2
3.0
Graphite
32.1
to�3
2.6
AravalliCraton
Jham
arkotraFm.,Aravalli
Supergroup
2.2–1.9
Black
shale
14
Kerogen
Greenschistto
amphibolite
35
�13.1
to�2
9.7
PineCreek
Orogen,N.Australia
Whites
Fm.
1.85–2.02
Black
shale
~1,000
3.9/30
Kerogen,carbon
coatingsonmineral
grains
Greenschist
36,37,38,
�15.8
to�3
1.3
KaapvaalandZim
babweCratons
UmfuliFm.,PiriwiriGroup,
Zim
babwe
~2.1
Graphitic
phyllite,argillite,greywacke
~1,000
Greenschistto
granulite
39
SengomaArgillite/Silvertonfm
s.,
PretoriaGroup,Botswana/S.
Africa
~2.15
Graphitic
phyllite,argillite,greywacke
500–700
23.5
Kerogen
and
pyrobitumen
Greenschist
40
�13.5
to�2
4.5
SaoFranciscoCraton
BarreiroFm.,Minas
Supergroup,
Brazil
~2.1
Graphitic
phyllite
4.4
Carbonaceousmatter
Greenschist
41
�26.6
1.Buseck
etal.1997;2.Filippov2002;3.Galdobinaetal.1984;4.Melezhik
etal.1999;5.Melezhik,etal.1988;6.Avedisyan
1995;7.HazovandHazova1982;8.Biske1997;9.Karhu1993;
10.Lehtonen
etal.1998;11.Silvennoinen
1972;12.Manninen
1991;13.Lukkarinen
2008;14.Pekkarinen
1979;15.Pekkarinen
andLukkarinen
1991;16.Perttunen,andHanski2003;
17.Ahtonen
1996;18.Honkam
o1985;19.Loukola-RuskeeniemiandHeino1996;20.Loukola-Ruskeeniemi1999;21.Loukola-Ruskeeniemiet
al.1991;22.Loukola-Ruskeeniemi1991;
23.Loukola-Ruskeeniemi2011;24.Taran
etal.2011;25.Bonhommeetal.1982;26.Cortialetal.1990;27.Gauthier-LafayeandWeber
1989;28.Weber
etal.1983;29.Sozinovetal.1988;
30.ZakrutkinandZhmur1989;31.Wilton1996;32.Bekkeretal.2003a;33.Bondesen
etal.1967;34.Chadwicketal.2001;35.Papineauetal.2009;36.McK
irdy1974;37.McK
irdyandIm
bus
1992;38.Worden
etal.2008;39.Masteret
al.2010;40.Bekker
etal.2008;41.Bekker
etal.2003b;42.Paakkola
andGeh€ or
1988
1216 H. Strauss et al.
Table 7.5 Geochemistry of the maksovite, archive samples from FAR-DEEP Hole 12B
Depth, m SiO2 TiO2 Al2O3 Fe2O3 MgO CaO Na2O K2O MnO P2O5 Stot Corg
140.45 40.9 0.37 5.73 2.56 0.82 0.09 0.40 1.96 <0.01 0.06 1.1 40.7
147.58 42.5 0.32 4.96 2.68 0.81 0.07 0.33 1.77 <0.01 0.04 1.7 39.7
155.02 37.1 0.36 5.62 2.98 1.94 0.08 0.24 1.86 <0.01 0.05 1.7 40.7
Table 7.4 Geochemistry of the main types of maksovite from the Maksovo deposit (Data are from Filippov 2002)
SiO2 TiO2 Al2O3 Fe2O3 FeO MgO CaO Na2O K2O Stot Corg
Massive maksovite (n ¼ 21)
Average 54.4 0.23 3.74 1.49 0.55 0.59 0.17 0.06 1.34 0.90 36.6
Min. 41.8 0.14 2.10 0.37 0.14 0.21 0.07 0.03 0.58 0.11 22.2
Max. 71.0 0.30 5.09 2.96 1.72 1.67 0.56 0.12 2.24 1.90 50.4
Jointed maksovite (n ¼ 11)
Average 47.0 0.25 4.16 1.13 0.42 0.57 0.08 0.02 1.25 0.38 44.6
Min. 38.6 0.19 3.11 0.35 0.27 0.43 0.01 0.03 0.94 0.13 31.9
Max. 61.0 0.38 5.60 2.06 0.52 0.88 0.14 0.52 1.56 0.93 53.3
Vuggy maksovite (n ¼ 10)
Average 60.6 0.20 3.16 1.17 1.06 0.54 0.13 0.05 0.96 0.73 31.4
Min. 44.6 0.14 2.27 0.21 0.45 0.26 0.01 0.02 0.60 0.20 22.8
Max. 70.1 0.32 4.00 4.12 4.04 1.65 0.43 0.10 1.74 2.65 44.3
Quartz-cemented maksovite breccia (n ¼ 23)
Average 62.4 0.18 2.96 1.07 0.42 0.44 0.09 0.04 0.79 0.38 31.0
Min. 47.7 0.10 2.04 0.30 0.14 0.21 0.01 0.01 0.48 0.10 15.6
Max. 76.8 0.26 4.14 3.14 0.87 0.87 0.29 0.08 1.51 1.32 46.8
Table 7.6 Oxygen and organic carbon isotopic composition of maksovite and associated sedimentary rocks (Fallick, Melezhik, Brasier and
Lepland, unpublished)
Hole Depth (m) Rock Mineral d18O (‰) d13C (‰)a
12A 8.23 Chert bed Quartz residue 19.5 �33.1
12A 24.33 Chert nodule Silicate/oxide component 19.2 �36.6
12A 48.63 Maksovite vein Silicate residue 20.3 �36.7
12B 147.58 Maksovite bed Silicate residue 22.3 �25.1
12B 251.7 Chert nodule Silicate/oxide component 16.2 �23.4
12B 412.78 Organic-rich shale Silicate residue 17.5 �21.9
13A 36.24 Chert Quartz residue 18.9 �30.9
13A 98.65 Organic-rich shale Silicate/oxide component 19.8 �37.4
13A 130.0 Organic-rich shale Silicate residue 19.6 �35.4
13A 139.56 Siltstone Silicate residue 15.6 �34.5
268 96.5 Maksovite breccia Quartz residue, grey, coarse grained 21.0
268 96.5 Maksovite breccia Quartz residue, grey, fine grained 18.4
268 96.5 Maksovite breccia Quartz residue, black and white, coarse grained 20.3
268 96.5 Maksovite breccia Bulk organic carbon �24.4
The oxygen isotope ratios were measured by the laser fluorination method of Macaulay et al. (2000)ad13C was measured from bulk organic carbon.
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1217
Fig. 7.57 Reported occurrences of c. 2100–1900 Ma black shales shown on the map of global distribution of Paleaeoproterozoic rocks. See
Table 7.3 for details on geological units containing black shales
1218 H. Strauss et al.
Fig. 7.58 Simplified geological map of Finland (Modified from Koistinen et al. 2001) showing the distribution of Palaeoproterozoic black shales
(After Arkimaa et al. 2000)
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1219
Fig. 7.60 Temporal variations in d13Ccarb (Source of data: Prokoph
et al. 2009) and d13Corg (Source of data: Strauss and Moore 1992;
Thomazo et al. 2009) for the time interval 3000 to 1500 Ma. d13Corg
data for shungite from Fennoscandia are marked in red (Source of data:Melezhik et al. 1999a, 2009). Plotted d13Ccarb and d13Corg values have
not been screened for diagenetic and metamorphic alterations
Fig. 7.59 Phanerozoic secular variations in d13Ccarb and d13Corg shown as moving averages (Redrawn after Hayes et al. 1999)
1220 H. Strauss et al.
Fig. 7.61 Variations in d13C during diagenesis (a) and as a consequence of changes in the fractional burial of organic carbon (b). See text for
further explanations
Fig. 7.62 Variations in d13Corg for shungite-bearing rocks (Redrawn after Melezhik et al. 2009)
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1221
Fig. 7.63 Lustrous pyrobitumen, termed shungite by Inostranzev
(1885, 1886) after the village of Shunga at the discovery point. The
sample collected from an old adit represents metamorphosed
Palaeoproterozoic oil which was trapped in an interbed-opening. Sam-
ple is from the mineralogical museum in the Institute of Geology in
Petrozavodsk, Russia (Photograph courtesy of Alexander Romashkin)
1222 H. Strauss et al.
Fig. 7.64 Geological map of the Onega Basin with locations of FAR-DEEP and other drillholes relevant to discussion presented in the chapter
(The geological map is modified by Aivo Lepland from Koistinen et al. (2001))
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1223
Fig. 7.65 Lithological composition of the Zaonega Formation based
on sections intersected by the Onega parametric drillhole (After
Morozov et al. 2010; Krupenik et al. 2011a) and FAR-DEEP Holes
12A, 12B and 13A (Compiled by Victor Melezhik and Alenka Crne).
A tentative correlation of the drilled sections is indicated by redrectangles. Drillhole positions are shown in Fig. 7.64
1224 H. Strauss et al.
Fig. 7.66 Main lithological features of Zaonega Formation rocks
(core diameter in all plates is 5 cm unless specified otherwise).
(a) Corg-rich, rhythmically bedded greywacke-shale from FAR-DEEP
Core 13A. (b) Beds of Corg-rich shale (black) with thin greywacke
interlayers from FAR-DEEP Core 12B. (c) Parallel-bedded greywacke
(grey), clayey siltstone (yellow) and Corg-rich mudstone (black);FAR-DEEP Core 12B. (d) Laminated siltstone with three layers of
calcareous greywacke containing black, spherical fragments of
Corg-rich mudstone apparently representing entrapped former tar
“balls” from seafloor oil seeps; FAR-DEEP Core 12B
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1225
Fig. 7.66 (continued) (e) Large calcite concretion in Corg-rich siltstone;
FAR-DEEP Core 12B. (f) Polymict breccia composed of unsorted
fragments of black chert, carbonate rocks (pale grey) and smaller fragments
of other sedimentary rocks emplaced into pyrobitumen-rich matrix; an
inferred product of explosive eruption associated with formation of
peperite; FAR-DEEPCore 13A. (g) Several thinmafic lava flows separated
by black, Corg-rich mudstone; FAR-DEEP Core 13A. (h) Polygonal cracks
(columnar joints in three-dimensional space) in organosiliceous rocks
(locally termedmaksovite) at the contact with a gabbro body; theMaksovo
quarry. (i) Peperite composed of dark brown fragments of mafic lava with
pyritised and calcitised sedimentary matrix; FAR-DEEP Core 13A. (j)
Close-up view of peperite composed of fragment of mafic lava in
pyrobitumenmatrix (P); note pyrobitumen filling the vesicle core (enlarged
in white rectangle and arrowed); FAR-DEEP Core 13B
1226 H. Strauss et al.
Fig. 7.66 (continued) (k) Back-scattered electron image of dolostone
composed of zoned, euhedral crystals closely packed in hypidiotopic
mosaic with remaining intergranular space occupied by black
pyrobitumen (former petroleum); FAR-DEEP Core 12B, 194.67 m.
(l) Back-scattered electron image of sandy siltstone with pyrobitumen-
rich (black) matrix (former petroleum); FAR-DEEP Core 12B, 408.4 m.
(m) Dark-grey organosiliceous rock/maksovite with massive appear-
ance and scattered rounded clasts of pyrobitumen-rich rocks (grey),shales (black) and pyrite (bright); FAR-DEEP Core 12B
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1227
Fig. 7.66 (continued) (n) Cross-section view of semilustrous rock rich
in migrated pyrobitumen and residual kerogen (c. 55 wt.% total organic
carbon) with well-developed parting; the adit at Shunga. (o) A hand
specimen of semimat rock rich in migrated pyrobitumen and residual
kerogen (c. 50 wt.% total organic carbon) with faint bedding; the adit at
Shunga. (p) A hand specimen of semilustrous, massive rock rich in
migrated pyrobitumen and residual kerogen (c. 65 wt.% total organic
carbon) with faint bedding; the adit at Shunga. (q) A hand specimen of
lustrous pyrobitumen (shungite) containing 99 wt.% total organic car-
bon and representing metamorphosed oil which was trapped in interbed
opening; the adit at Shunga (Photographs (a–n, and q) by Victor
Melezhik, (o and p) reproduced from Melezhik et al. (1999a) with
permission of Elsevier)
1228 H. Strauss et al.
Fig. 7.67 Histograms showing distribution pattern of total organic
carbon contents and d13C of organic matter in sedimentary rocks of
the Zaonega Formation. (a) Histogram of total organic carbon content
showing four-modal distribution; the diagram summarises data
published up to 1999 with references to data source in Filippov and
Golubev (1994), Kupryakov (1994) and Melezhik et al. (1999a).
(b) Histogram of total organic carbon content showing bimodal distri-
bution; the diagram is based on archive samples collected from
FAR-DEEP Cores 12A, 12B and 13A. (c) d13C histogram showing
bimodal distribution; the diagram summarises data published up to
1999 with references to data source in Filippov and Golubev (1994)
and Melezhik et al. (1999a). (d) d13C histogram with bimodal distribu-
tion based on archive samples obtained from FAR-DEEP Cores 12A,
12B and 13A. (e) d13C histogram with bimodal distribution based on
samples obtained from the Onega parametric drillhole (Krupenik et al.
2011b)
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1229
Fig. 7.68 d13Corg stratigraphic profiles through the Zaonega Formation
(Compiled by Victor Melezhik and Michael Filippov) based on data
obtained from the Onega parametric drillhole (Data are from Krupenik
et al. 2011a, b), drillhole 5190 (Melezhik et al. unpublished data) and
FAR-DEEP Holes 12A and 12B (http://far-deep.icdp-online.org). Note
that d13Corg data have been used for the correlation of the drilled sections
(by eye as the best fit), and no independent lithological criteria are
currently available. Drillhole positions are shown in Fig. 7.64
1230 H. Strauss et al.
Fig. 7.69 Shungite type locality at Shunga village. (a) Geological map
of the Shunga area (Simplified after Ryabov 1948). (b) Lithological
sections through strata with occurrence of lustrous shungite
(pyrobitumen). The AB section is simplified after Gorlov (1984), the
CD section is simplified after Ryabov (1948); section position is located
on (a). (c) Lithological section of the Zaonega Formation based on FAR-
DEEP Core 13A; part of the section correlated with the AB and CD
section is indicated by red rectangle, and hole position is shown in (a)
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1231
Fig. 7.70 Pyrobitumen occurrences in igneous rocks and peperite-
associated explosive breccias. (a) Gabbro from a zone close to the
contact with organic-rich sedimentary rock; note abundant sulphide
amygdales and chlorite-filled joints; FAR-DEEP Core 12B, 413.7 m.
(b) Photomicrograph in reflected light showing a gabbro with veinlet
filled with pyrobitumen that occurs as “pencils” (bright) emplaced in
chlorite matrix; FAR-DEEP Core 12B, 417.21 m. (c) Back-scattered
electron image of graphic intergrowth between pyrobitumen (black)and calcite (bright) occurring in gabbro-hosted veinlet; FAR-DEEP
Core 12B, 417.21 m
1232 H. Strauss et al.
Fig. 7.70 (continued) (d) Back-scattered electron image of pyrobitumen “roses” (black) in calcite (bright) occurring in gabbro-hosted veinlet;
FAR-DEEP Core 12B, 417.21 m
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1233
Fig. 7.70 (continued) (e) Photomicrograph in reflected light showing gabbro-hosted veinlet filled with pyrobitumen that occurs as hollow
“pencils” (bright) emplaced in chlorite matrix (pale grey); FAR-DEEP Core 12B, 417.21 m
1234 H. Strauss et al.
Fig. 7.70 (continued) (f) Pyrobitumen-rich vein rimmed with pyrite
(yellow) in mafic lava flow; joints in lava are also filled with pyrite;
FAR-DEEP Core 12B, 53.7 m. (g) Photomicrograph in reflected light
showing pyrobitumen (bright) emplaced in chlorite matrix (pale grey)occurring in veinlet hosted by mafic lava; FAR-DEEP Core 12B,
93.12 m. (h) Photomicrograph in reflected light showing vesicles in
mafic lava filled with pyrobitumen that occurs as alternating concentric
bands composed of radial and massive pyrobitumen; FAR-DEEP Core
12B, 58.54 m
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1235
Fig. 7.70 (continued) (i) Photomicrograph in reflected light showing
peperite: a black shale containing fragment of mafic lava with contrac-
tion joints. (j) Detailed view (red rectangle in “i”) of contraction joints
filled with pyrobitumen (bright) and quartz (grey). Both images are
from FAR-DEEP Core 12A, 46.86 m
1236 H. Strauss et al.
Fig. 7.70 (continued) (k) Explosive breccia composed of
pyrobitumen fragments (marked as P1) in two generations of soft-
sediment deformed mudstone matrix (marked as P2 and P3);
FAR-DEEP Core 12B. (l) Polymict breccias (mass-flow or explosive)
composed of unsorted clasts of various sedimentary rocks and
pyrobitumen (red arrowed); note that clasts have a variable degree of
roundness and some are soft-sediment deformed (yellow arrowed);FAR-DEEP Core 12B. (m) Photomicrograph in reflected light showing
a clast of silica-rich rock with pyrobitumen-filled voids from polymict
breccia; FAR-DEEP Core 12B
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1237
Fig. 7.70 (continued) (n) Pyrobitumen-rich explosive breccias
associated with gabbro emplacement; matrix is composed of
disintegrated sedimentary material cemented by pyrobitumen-rich
substance (P1), whereas clasts are greywacke, rhythmically-bedded
siltstone-mudstone, sulphides as well as pyrobitumen-rich rocks (P2);
the latter also occur in the core as a seemingly intact layer (P3)
interbedded with shale (S) and containing pyrobitumen veinlet
(white-arrowed), and fragments of mudstone (red-arrowed) and
pyrobitumen (P4); FAR-DEEP Hole 12B (All photographs by Victor
Melezhik)
1238 H. Strauss et al.
Fig. 7.71 Maturation of kerogen during burial processes. (a) Oils and gas windows in the burial sequence (Modified from http://
oilandgasgeology.com). (b) Oil and gas windows versus thermal gradient and depth (Modified from http://www.CliffsNotes.com)
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1239
Fig. 7.72 Oil migration pathways in the Zaonega Formation. (a) Sub-
vertical, pyrobitumen-rich, ptygmatic veinlets cross-cutting massive
dolostone; note that the pyrobitumen (originally oil) was sourced
from organic-rich shale interlayers; FAR-DEEP Core 12B. (b) Vertical,
pyrobitumen-rich (originally oil) vein cross-cutting dolostone and
shale; FAR-DEEP Core 12A. (c) Bedding-parallel, pyrobitumen-rich
veins in dolostone; FAR-DEEP Core 12A. (d) Sub-vertical and
bedding-parallel, pyrobitumen-rich veinlets (bright brown) sourced
from pyrobitumen-rich layer (bright brown) at the base of organic-
rich shale (dark brown) above pale grey, massive dolostone; FAR-
DEEP Core 12A
1240 H. Strauss et al.
Fig. 7.72 (continued) (e) Pyrobitumen-rich, branching veins (palebrown) cross-cutting organic-rich siltstone; FAR-DEEP Core 12B,
278.09 m. (f) Pyrobitumen-rich vein (bright) cross-cutting calcareous
greywacke and sourced from the upper part of kerogen-rich mudstone
bed; FAR-DEEP Core 12B, 196.12 m. (g) Zoned, pyrobitumen-rich
(bright) vein with wall-parallel banding in calcareous greywacke;
FAR-DEEP Core 12B, 195.15 m. All images are photomicrographs
taken in reflected light
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1241
Fig. 7.72 (continued) (h) Symmetrically-zoned vein with
pyrobitumen-rich margins (bright) with faint, wall-parallel banding,
and kerogen-rich mudstone (black) core cross-cutting calcareous
greywacke; FAR-DEEP Core 12B, 196.12 m. (i) Zoned vein with
pyrobitumen-rich margin (bright), calcite core (pale grey) and partiallytransitional contact with calcareous greywacke (top); FAR-DEEP Core
12B, 196.12 m. (j) A fragment of pyrobitumen-rich (bright) vein
showing multiphase, syndepositional brecciation and cementation;
FAR-DEEP Core 12B, 102.06 m. (k) Late, quartz-filled, extensional
crack in pyrobitumen-bearing mudstone vein (sedimentary dyke?) in
clayey dolostone; FAR-DEEP Core 12B, 242.42 m. All images are
photomicrographs taken in reflected light
1242 H. Strauss et al.
Fig. 7.72 (continued) (l) Back-scattered electron image of vesicular
pyrobitumen (dark grey) vein hosted by dolostone; vesicles are filled
with galena, pyrite, dolomite or remain unfilled (grey and black).(m) Detailed view showing composition and shape of vesicles; note
peculiar relationship of paired galena-filled vesicles (bright); somewhat
similar relationship between pyrite-filled (pale grey) and unfilled vesi-
cle (black). (n) Detailed view of unfilled vesicle with desiccated walls.
FAR-DEEP Core 12B, 242.42 m
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1243
Fig. 7.72 (continued) (o) A fragment of pyrobitumen (bright) veinwith flattened vesicles showing concentric pattern; FAR-DEEP Core
12B, 242.42 m. (p) A fragment of pyrobitumen-rich (bright) vein with
vesicles filled with chlorite; FAR-DEEP Core 12B, 240.37 m (Both
images are photomicrographs taken in reflected light. All photographs
by Victor Melezhik)
1244 H. Strauss et al.
Fig. 7.73 Oil traps in the Zaonega Formation. (a, b) Pyrobitumen
(originally oil) filling space between dolostone fragments; FAR-DEEP
Core 13A. (c) Pyrobitumen filling joints in fractured dolostone;
FAR-DEEP Core 13A. (d) Late compaction fracture in greywacke filled
with pyrobitumen (originally oil); FAR-DEEP Core 12B
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1245
Fig. 7.73 (continued) Late compaction fracture in greywacke filled
with pyrobitumen (originally oil); FAR-DEEP Core 12B. (e) Rhythmi-
cally bedded greywacke-mudstone with late compaction joints in mud-
stone layers filled with pyrobitumen (originally oil); FAR-DEEP Core
12A, 95.9 m. (f) Back-scattered electron image of pyrobitumen-rich
(black), graded sandstone with pyrobitumen accumulation (black, orig-inally oil) within interbed space; large clast from syndepositional
breccias in FAR-DEEP Core 12B, 408.4 m
1246 H. Strauss et al.
Fig. 7.73 (continued) (g) Brown pyrobitumen (originally oil) occur-
ring in space between oil-shale (c. 55 wt.% total organic carbon,
Melezhik et al. 1999a) at the bottom and massive dolostone on top;
knife for scale is 21 cm long. (h) Close-up view of the pyrobitumen
shown in (g); note conchoidal fracture with brown jarosite films
(98.4 wt.% total organic carbon, Melezhik et al. 1999). (i) Close-up
view of the oil-shale shown in (g); note partings developed parallel to
stratification (Photographs were taken from an adit near Shunga village.
Photographs (a–f, h–i) by Victor Melezhik, (g) modified from
Melezhik et al. (1999a)
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1247
Fig. 7.74 Organosiliceous rocks (maksovite) from the Maksovo
deposit in the Zaonega Formation. (a) Black, mat, massive maksovite
with quartz-filled joints. (b) Columnar joints in maksovite developed
at the contact with gabbro body; section parallel to column axis.
(c) Columnar joints in maksovite in the section perpendicular to the
column axis. (d) Maksovite with columnar joints cemented by pyrite.
Photographs were taken from the Maksovo quarry (location is shown in
Fig. 7.64)
1248 H. Strauss et al.
Fig. 7.74 (continued) (e) Massive maksovite with scattered rounded
fragments of pyrobitumen-rich material (dark grey, arrowed) and
smaller fragments of shale (black) and sulphide (bright); FAR-DEEPCore 12B, 150.3 m. (f) Photomicrograph in reflected light of maksovite
exhibiting its microstructural fabric, which is expressed by rounded
quartz particles “floating” in pyrobitumen matrix (bright); FAR-DEEPCore 12B, 145.09 m. (g) Syndepositional maksovite breccia from the
upper contact of the maksovite body in the Maksovo deposit; breccia is
composed of rounded fragments of pyrobitumen-rich material
(arrowed) and dark grey clasts of siliceous rocks with white “droplets”of pyrobitumen (originally oil); note that pyrobitumen matrix
envelopes clasts (upper part) and exhibits a fluidal structure with
alternating brighter and darker bands. Drillhole 202, 36.8 m (location
is shown in Fig. 7.75)
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1249
Fig. 7.74 (continued) (h) Back-scattered electron image of maksovite
matrix showing rounded, partially disintegrated (lower left clast margin)
siltstone clasts, which supply grainy particles to pyrobitumen-rich
(black) matrix. Inset in the lower right corner is a photomicrograph in
reflected light emphasising the fluidal structure of the pyrobitumen-rich
(bright) matrix. Sample collected from the Maksovo quarry
1250 H. Strauss et al.
Fig. 7.74 (continued) (i) Maksovite breccias containing disintegrated
and dispersed quartz masses (grey) in siliceous, pyrobitumen-rich
matrix; note voids (arrowed area) in the quartz fragment filled with -
finely-crystalline, quartz-pyrobitumen material. (j) Void-infill in
quartz; quartz filling the void shows concentric pattern of alternating
darker (Si-pyrobitumen substance) and brighter (pure SiO2) bands; the
remaining space is filled with pyrobitumen (black) showing shrinkage
joints cemented with late quartz. (k) Close-up view of void-infill; note
shrinkage joints in quartz-infill cemented by pyrobitumen (bright).
(l–n) Pyrobitumen-rich (black) maksovite matrix with clasts of silica-
pyrobitumen material; clasts show concentric pattern of alternating
micron-size bands whose colour ranges from black to light grey
depending on silica/pyrobitumen ratio; note that along margins, the
fine-scale concentric bands were totally obliterated and replaced by
later generation of quartz. Photomicrographs in reflected light (i, k) are
taken from drillcore 202 (36.8 m), and back-scattered electron images
(j, l, m, n) from drillcore 203 (53.9 m) in the Maksovo deposit
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1251
Fig. 7.74 (continued) (o) Jointed maksovite with joints filled with
lustrous pyrobitumen; the Maksovo deposit, drillhole 208, 40.2 m, (p)
Brecciated maksovite with cracks cemented by pyrobitumen-rich
material (black); the Maksovo deposit, drillhole 201, 11.9 m. (q)
Quartz-cemented maksovite breccias; Maksovo quarry (All
photographs by Victor Melezhik)
1252 H. Strauss et al.
Fig. 7.75 Geological map of the Maksovo deposit, with cross- and longitudinal profiles through the maksovite lens (Modified after Kupryakov
1994)
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1253
Fig. 7.76 Cross- and longitudinal profiles through the maksovite lens
of the Maksovo deposit (modified after Kupryakov 1994). (a) Vertical
and lateral distribution of different types of maksovite. (b) Vertical and
lateral distribution of the Corg content. (c) Vertical and lateral distribu-
tion of SiO2/Al2O3 ratio
1254 H. Strauss et al.
Fig. 7.77 Carbon-isotope
composition of organic carbon
from Zaonega Formation
sedimentary rocks, maksovite,
migrated pyrobitumen and
Kondopoga surface oil seep. (a)
d13C of the total organic carbon
from Zaonega Formation
sedimentary rocks and migrated
Shunga pyrobitumen compared
with pyrobitumen and total
organic carbon of maksovite
(Data are from Filippov and
Golubev (1994), Melezhik et al.
(1999a), FAR-DEEP archive
samples (http://far-deep.icdp-
online.org), drillhole 5190
(Melezhik et al., unpublished
data) and the Onega parametric
hole (Krupenik et al. 2011b). Data
for Kondopoga oil seep are from
Melezhik et al. (2009)). (b) Laser-
based d13C of pyrobitumen from
Maksovo maksovite (drillhole
203, 53.9 m) are from V.
Melezhik and A. Fallick
(unpublished data)
Fig. 7.78 SiO2 content in
maksovite, compared with that of
background sedimentary rocks of
the Zaonega Formation calculated
to total organic carbon-free basis
(Compiled by V. Melezhik and Y.
Deines). Note that the maksovite
extremely rich in SiO2, and
sedimentary rocks with >5 wt.%
total organic carbon show a
significant enrichment in SiO2
with respect to those containing
<5 wt.% total organic carbon
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1255
Fig.7.79
Drillhole-based
sectionsshowinglateralandverticallithological
variationoftheZaonegaForm
ationandthree-dim
ensional
view
ofthemaksovitebodyintersectedbyFAR-D
EEP
Hole
12A
and12B(Compiled
byY.Deines
andV.Melezhik)
1256 H. Strauss et al.
Fig. 7.80 Organosiliceous rocks (maksovite) in the Zaonega Forma-
tion intersected by FAR-DEEP Hole 12B. (a) Massive, cryptocrystal-
line maksovite representing bulk of the lens. (b) Lower contact of the
maksovite lens with massive dolostone; note that the two lithologies are
separated by a chlorite-rich bed (arrowed). (c) Photomicrograph in
reflected light showing pyrobitumen-rich (white), fluidal matrix with
scattered grains and rounded fragment of quartz (grey), and pyrite
crystals (bright); depth of 154.1 m. (d) Photomicrograph in reflected
light showing pyrobitumen-rich (white) matrix with scattered quartz
grains, elongated fragments of siltstone (grey) and platy fragments of
pyrobitumen-rich (bright) rocks retaining parallel lamination and
quartz grains (grey); note the rounded sandstone fragment that is
attached to one such “plate”; depth of 143.42 m
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1257
Fig. 7.80 (continued) (e) Rounded, partially disintegrated clasts of
sandstones “enveloped” by pyrobitumen-rich matrix with fluidal fabric;
note that the rock clasts in upper left corner contain intergranular
pyrobitumen (originally oil) whereas other clasts are rich in kerogen
(black); depth of 140.45 m. (f) Nodular pyrite “enveloped” by fluidal,
pyrobitumen-rich matrix with scattered quartz grains and sandstone
fragments; depth of 140.45 m. Photomicrographs are taken in
reflected light
1258 H. Strauss et al.
Fig. 7.80 (continued) (g) Maksovite breccia consisting of slumped
and partially disintegrated massive maksovite (grey with brownish hue)in black mudstone matrix, and passes downward to soft-sediment
deformed siltstone and mudstone with two fragments of massive
maksovite (left side). (h) Photomicrograph in reflected light showing
maksovite matrix with a large clast of pyrobitumen-rich (bright)maksovite with early soft-sediment deformation; depth of 138.55 m.
(i) Two generations of maksovite-type matrix occur in a large
maksovite clast emplaced into third generation. The early generation
is represented by pyrobitumen-rich rounded core (bright) enveloped bythe second generation with fluidal fabric; photomicrograph in reflected
light, depth of 137.88 m. (j) Back-scattered electron image of a “clay
ball” in maksovite-type matrix; the “ball” is composed of flaky sericite
and pyrobitumen (black); depth of 138.55 m
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1259
Fig. 7.80 (continued) (k) Maksovite pyrobitumen matrix supporting
quartz particle, rounded and with mammillated surface, partially
disintegrated clasts of sandstone and siltstone (red-arrowed), K-feldspar
partially replaced by quartz (yellow-arrowed), sericite aggregate (green-arrowed), platy fragment of pyrobitumen-rich (black) material (blue-arrowed); back-scattered electron image, depth of 138.55 m
1260 H. Strauss et al.
Fig. 7.80 (continued) (l) Upper contact of the maksovite breccias appearing as the surface with uneven topography buried with graded sandstone
and laminated siltstone-shale; this provides crucial evidence that the maksovite was deposited/emplaced on the seafloor
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1261
Fig. 7.80 (continued) (m) Maksovite vein (M) cross-cutting
laminated siltstone-mudstone (S); note that the lower contact is com-
plicated by a small-scale injection (red-arrowed) of maksovite into
mudstone, whereas the upper contact is straight (yellow-arrowed). (n)Detailed image of the lower contact demonstrating that the maksovite
(M) was injected into the black, organic-rich mudstone. (o) Detailed
image of the upper contact demonstrating that the maksovite (M) vein
cross-cut laminated siltstone-mudstone. These images provide crucial
evidence that the maksovite has and intrusive, allochthonous nature
(All photographs by Victor Melezhik)
1262 H. Strauss et al.
Fig. 7.81 A cartoon illustrating formation of organosiliceous mush (maksovite) invoking subsea hydrothermal circulation induced by emplace-
ment of gabbro into unconsolidated organic-rich sediments and formation of peperites (Compiled by V. Melezhik, A. Lepland and Y. Deines)
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1263
Fig. 7.82 Oxygen isotope data from maksovite and sedimentary rocks
of the Zaonega Formation versus natural oxygen isotope reservoirs.
Data for natural oxygen isotope reservoirs are from Taylor (1974),
Onuma et al. (1972), Sheppard (1977), Graham and Harmon (1983)
and Hoefs (2009) (Data for maksovite and Zaonega Formation rocks
are presented in Table 7.6)
1264 H. Strauss et al.
Fig. 7.83 Main lithological features of Kondopoga Formation rocks
preserving surface oil seeps, Kondopoga aggregate quarry. (a) Bedded
greywacke (bright, pale grey), clayey siltstone (dark grey, brownish)and mudstone (black); note loading structures at the base of bright
greywacke beds (b) Rhythmically bedded, massive, graded greywacke
and laminated siltstone. (c) Laminated siltstone with mudstone clasts
(black) and angular fragments of pyrobitumen (bright)
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1265
Fig. 7.83 (continued) (d) Cluster of pyrobitumen clasts occurring on the bedding surface of massive greywacke bed; note that the shrinkage
joints are filled with quartz stained by brown jarosite
1266 H. Strauss et al.
Fig. 7.83 (continued) (e) Rounded clast composed of lustrous pyrobitumen. (f) Pancake-like inclusion of pyrobitumen with polygonal shrinkage
joints filled with quartz stained by brown jarosite. (g) Pyrobitumen inclusion showing injection into host sandstone; coin is 2 cm in diameter
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1267
Fig. 7.83 (continued) (h) A fragment of pyrobitumen clast with diffuse
margin expressed as impregnation of host siltstone with thin films of
pyrobitumen. (i) Pyrobitumen clast impregnated by fragments of foreign
particles (Photographs (a–c, and h) by Victor Melezhik, photograph (f)
reproduced from Melezhik et al. (1999) with permission of Elsevier, and
photographs (d, e, g and i) reproduced from Melezhik et al. (2009))
1268 H. Strauss et al.
References
Ahtonen N (1996) Geochemistry, mineralogy and petrophysics of black
shales in the Northern Ostrobotnia schist belt with emphasis on ore
prospecting. Unpublished M.Sc thesis, Department of Geology and
Astronomy, University of Oulu, p 78, (in Finnish)
Akhmedov AM, Krupenik VA (1990) Turbidity conditions of sedimen-
tation and pyrite formation in the early Proterozoic Pechenga basin.
Sov Geol 11:51–60 (in Russian)
Arkimaa H, Hyv€onen E, Lerssi J, Loukola-Ruskeeniemi K, Vanne J
(2000) Proterozoic black shales and aeromagnetic anomalies in
Finland, Scale 1: 1 000 000. Geological Survey of Finland, Espoo
Arthur MA, Sageman BB (1994) Marine black shales: depositional
mechanism and environments of ancient deposits. Annu Rev Earth
Planet Sci 22:499–551
Avedisyan AA (1995) Lithologic-geochemical features of
metasediments of the “productive unit” and Southern zone. In:
Mitrofanov FP, Smolkin VF (eds) Magmatism, sedimentogenes
and geodynamics of the Pechenga Palaeoriftogenic structure. Kola
Research Centre, Apatity, pp 124–138 (in Russian)
Bekasova NB (1985) Pechenga palaeogeography in early Pilguj€arvitime of the early Proterozoic. Lithol Miner Resour 1:127–137 (in
Russian)
Bekker A, Karhu JA, Eriksson KA, Kaufman AJ (2003a) Chemostra-
tigraphy of Paleoproterozoic carbonate successions of theWyoming
Craton: tectonic forcing of biogeochemical change? Precambrian
Res 120:279–325
Bekker A, Sial AN, Karhu JA, Ferreira VP, Noce CM, Kaufman AJ,
Romano AW, Pimentel MM (2003b) Chemostratigraphy of
carbonates from the Minas Supergroup, Quadrilatero Ferrifero,
Brazil: a stratigraphic record of early Proterozoic atmospheric,
biogeochemical and climatic change. Am J Sci 303:865–904
Bekker A, Holmden C, Beukes NJ, Kenig F, Eglington B, PattersonWP
(2008) Fractionation between inorganic and organic carbon during
the Lomagundi (2.22–2.1 Ga) carbon isotope excursion. Earth
Planet Sci Lett 271:278–291
Berner RA (1993) Weathering and its effect on atmospheric CO2 over
Phanerozoic time. Chem Geol 107:373–374
Beukes NJ, Klein C, Kaufman AJ, Hayes JM (1990) Carbonate petrog-
raphy, kerogen distribution and carbon ad oxygen isotope variations
in an early Proterozoic transition from limestone to iron-formation
deposition. Econ Geol 85:663–690
Biske NC (1997) Carbon isotopic composition of graphite in Northern
Ladoga area. In: Geology, petrography and geochemistry of Pre-
cambrian rocks of Karelia. Karelian Research Centre,
Petrozavodzk, pp 37–40 (in Russian)
Biske NC, Romashkin AE, Rychanchik DV (2004) Petrosavodsk
peperite-structures of Lebestchina. In: Geology and mineral
deposits. Proceedings of the Institute of Geology, Karelian
Research Centre, vol 7, pp 193–199, Karelian Research Centre,
Petrozavodsk (in Russian)
Blankenship RE, Sadekar S, Raymond J (2007) The evolutionary
transition from anoxygenic to oxygenic photosynthesis. In:
Falkowski PG, Knoll AH (eds) The evolution of primary producers
in the Sea. Elsevier, Boston, pp 21–35
Bondesen E, Pedersen KR, Jørgensen O (1967) Precambrian organisms
and the isotopic composition of organic remains in the Ketilidian of
south-west Greenland. Grønlands Geologiske Undersøgelse Bulle-
tin 67:1–41
Bonhomme M, Gauthier-Lafaye F, Weber F (1982) An example of
lower Proterozoic sediments: the Francevillain in Gabon. Precam-
brian Res 18:87–102
Borisov PA (1956) The Karelian Shungites. Karelian Book Publisher,
Petrozavodsk, p 92 (in Russian)
Bradley DC (2008) Passive margins through earth history. Earth Sci
Rev 91:1–26
Brocks JJ, Logan GA, Buick R, Summons RE (1999) Archean molecu-
lar fossils and the early rise of eukaryotes. Science 285:1033–1036
Brocks JJ, Love GD, Snape CE, Logan GA, Summons RE, Buick R
(2003) Release of bound aromatic hydrocarbons from late Archean
and Mesoproterozoic kerogens via hydropyrolysis. Geochim
Cosmochim Acta 67:1521–1530
Brocks JJ, Love GD, Summons RE, Knoll AH, Logan GA, Bowden SA
(2005) Biomarker evidence for green and purple sulphur bacteria in
a stratified Palaeoproterozoic sea. Nature 437:866
Bruneau PMC, Ostle N, Davidson DA, Grieve IC, Fallick AE (2002)
Determination of rhizosphere 13C pulse signals in soil thin sections
by laser ablation isotope ratio mass spectrometry. Rapid Commun
Mass Spectrom 16:2190–2194
Buseck PR, Tsipursky SJ, Hettich R (1992) Fullerenes from the geo-
logical environment. Science 257:215–217
Buseck PR, Galdobina LP, Kovalevski VV, Rozhkova NN, Valley JW,
Zaidenberg AZ (1997) Shungites: the C-rich rocks of Karelia,
Russia. Can Mineralogist 35:1363–1378
Chadwick B, Claeys P, Simonson B (2001) New evidence for a large
Palaeoproterozoic impact: spherules in a dolomite layer in the
Ketilidian orogen, South Greenland. J Geol Soc 158:331–340
Clarke RH, Cleverly RW (1990) Leakage and post accumulation
migration. In: England WA, Fleet AJ (eds) Caprocks and seals.
Geol Soc Lon, Spec Pub 59:265–271
Condie KC (2004) Precambrian superplume events. In: Eriksson PG,
Altermann W, Nelson DR, Mueller WU, Catuneanu O (eds) The
Precambrian Earth: tempos and events, development in Precam-
brian geology, vol 12. Elsevier B.V., Amsterdam, pp 163–173
Cortial F, Gauthier-Lafaye F, Lacrampe-Couloume G, Oberlin A,
Weber F (1990) Characterization of organic matter associated
with uranium deposits in the Francevillian Formation of Gabon
(Lower Proterozoic). Org Geochem 15:73–85
Des Marais DJ (2001) Isotopic evolution of the biogeochemical carbon
cycle during the Precambrian. Rev Mineral Geochem 43:555–578
Dutkiewicz A, Volk H, George SC, Ridley J, Buick R (2006)
Biomarkers from Huronian oil-bearing fluid inclusions: an uncon-
taminated record of life before the great oxidation event. Geology
34:437–440
Dutkiewicz A, George SC, Mossman DJ, Ridley J, Volk H (2007) Oil
and its biomarkers associated with the Palaeoproterozoic Oklo
natural fission reactors, Gabon. Chem Geol 244:130–154
Ebbesen TW, Hiura H, Hedenquist JW, De Ronde CEJ, Andersen A,
Often M, Melezhik VA (1995) Origin of fullerenes in rocks. Sci-
ence 268:1634–1635
Eigenbrode JL, Freeman KH (2006) Late Archean rise of aerobic
microbial ecosystems. Proc Natl Acad Sci USA 103:15759–15764
Eriksson PG, Catuneanu O, Sarkar S, Tirsgaard H (2005) Patterns of
sedimentation in the Precambrian. Sediment Geol 176:17–42
Ernst R, Bleeker W (2010) Large igneous provinces (LIPs), giant dyke
swarms, and mantle plumes: significance for breakup events within
Canada and adjacent regions from 2.5 Ga to the Present. Can J Earth
Sci 47:695–739
Fallick AE, Melezhik VA, Simonson B (2008) The ancient anoxic
biosphere was not as we know it. In: Dobretsov N, Kolchanov N,
Rozanov A, Zavarzin G (eds) Biosphere origin and evolution.
Springer Science+Business Media, New York, pp 169–188
Fallick AE, Melezhik VA, Simonson B (2011) On Proterozoic
ecosystems and the carbon isotopic composition of carbonates
associated with banded iron formations. In: Neves L et al (eds)
Modelacao de Sistemas Geologicos. Universidade de Coimbra,
Portugal, pp 57–71
Filippov MM (1994) Current views on the organic precursor of
shungite rocks. In: Filippov MM (ed) The organic matter of
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1269
Karelian Shungite rocks (genesis, evolution and the methods
of study). Karelian Research Centre, Petrozavodsk, pp 16–24,
(in Russian)
Filippov MM (2000) Formational model of shungite deposits in the
Onega Synclinorium. Doctoral thesis, St. Petersburg University,
p 310 (in Russian)
Filippov MM (2002) Shungite rocks of the Onega structure. Karelian
Research Centre, Petrozavodsk, p 280 (in Russian)
Filippov MM, Golubev AI (1994) Carbon isotope composition of
shungite rocks. In: Filippov MM (ed) The organic matter of Kare-
lian Shungite rocks (genesis, evolution and the methods of study).
Karelian Research Centre, Petrozavodsk, pp 32–43 (in Russian)
Filippov MM, Romashkin AE (1994) Shungite rocks. Karelian
Research Centre, Petrozavodsk, p 89 (in Russian)
Filippov MM, Golubev AI, Romashkin AE, Rychanchik DV (1994)
Mineral constituent of shungite-bearing rocks: primary composi-
tion, sources and relation to shungite matter. In: Filippov MM (ed)
The organic matter of Karelian Shungite rocks (genesis, evolution
and the methods of study). Karelian Research Centre, Petrozavodsk,
pp 78–93 (in Russian)
Filippov MM, Melezhik VA et al (eds) (2007) Atlas of textures and
structures of shungite rocks of the Onega Synclinorium.
Scandinavia, Petrozavodsk, p 80 (in Russian)
Galdobina LP (1987) The Ludicovian horizon. In: Sokolov VA (ed)
The geology of Karelia. Nauka, Leningrad, pp 59–67 (in Russian)
Galdobina LP, Schidlowski M, Matzigkait U, Sokolov VA (1984)
Isotopic study of early Proterozoic shungite, Karelia. In: Abstract
volume II, 27th international geological congress, Moscow, p 292
(in Russian)
Galdobina LP, Kalinin YK, Kupryakov SV (1986) Endogenic origin of
the Karelian shungite rocks. Abstract, the 2nd all-union symposium
on carbon geochemistry, GEOKHI, Moscow, pp 79–81 (in Russian)
Garvin J, Buick R, Anbar AD, Arnold GL, Kaufman AJ (2009) Isotopic
evidence for an aerobic nitrogen cycle in the latest Archean. Science
323:1045–1048
Gauthier-Lafaye F, Weber F (1989) The Francevillian (Lower Protero-
zoic) uranium ore deposits of Gabon. Econ Geol 84:2267–2285
Geothermal Gradients (2011) http://www.CliffsNotes.com. Accessed
3 May 2011
Golubic S, Hofmann HJ (1976) Comparison of Holocene and mid-
Precambrian Entophysalidaceae (Cyanophyta) in stromatolitic
algal mats: cell division and degradation. J Paleontol 50:1074–1082
Gorlov VI (1984) The Onega shungite (geology, genesis and economic
potential), D.Sc. thesis, Leningrad, p 20 (in Russian)
GrahamCM, Harmon RS (1983) Stable isotope evidence on the nature of
crust-mantle interaction. In: Hawkesworth CJ, Norry MJ (eds) Conti-
nental basalts and mantle xenoliths. Shiva, Nantwich, pp 20–45
Grapes RH (2006) Pyrometamorphism. Springer, Berlin/Heidelberg,
p 275
Gretener PE (1969) Fluid pressure in porous media – its importance in
geology, a review. Bull Can Petrol Geol 17:255–295
Grotzinger JP, Knoll AH (1999) Proterozoic stromatolites: evolution-
ary mileposts or environmental dipsticks? Annu Rev Earth Planet
Sci 27:313–358
Hannah JL, Stein HJ, Bekker A, Markey RJ, Holland HD (2003)
Chondritic initial 187Os/188Os in Paleoproterozoic shale (seawater)
and the onset of oxidative weathering. Geochim Cosmochim Acta
67:A-34
Hannah JL, Stein HJ, Zimmerman A, Yang G, Melezhik VA, Filippov
MM, Turgeon SC, Creaser RA (2008) Re-Os geochronology of a
2.05 Ga fossil oil field near Shunga, Karelia, NWRussia. In: Abstract,
the 33rd international geological congress, Oslo, 6–14 Aug 2008
Hanski EJ (1992) Petrology of the Pechenga ferropicrites and
cogenetic, Ni-bearing gabbro-wehrlite intrusions, Kola Peninsula,
Russia. Bull Geol Surv Finl 367:1–192
Hanski E, Huhma H (2005) Central Lapland greenstone belt. In:
Lehtinen M, Nurmi PA, R€am€o OT (eds) Precambrian geology of
Finland – key to the evolution of the Fennoscandian Shield.
Elsevier, Amsterdam, pp 139–194
Hayes JM (1993) Factors controlling 13C contents of sedimentary organic
compounds: principles and evidence. Mar Geol 113:111–125
Hayes JM (1994) Global methanotrophy at the Archean-Proterozoic
transition. In: Bengtson S (ed) Early life on Earth. Columbia
University Press, New York, pp 220–236
Hayes JM,Waldbauer JR (2006) The carbon cycle and associated redox
processes through time. R Soc Lond Philos Trans Ser B Biol Sci
B361:931–950
Hayes JM, Kaplan IR, Wedeking KW (1983) Precambrian organic
geochemistry, preservation of the record. In: Schopf JW (ed)
Earth’s earliest biosphere: its origin and evolution. Princeton
University Press, Princeton, pp 93–134
Hayes JM, Strauss H, Kaufman AJ (1999) The abundance of 13C
in marine organic matter and isotopic fractionation in the global
biogeochemical cycle of carbon during the past 800 Ma. Chem Geol
161:103–125
Hazov RA, Hazova VI (1982) Area of Pitk€aranta (Northern Ladoga).
In: Sokolov VA (ed) Geology of shungite bearing volcanic and
sedimentary rocks of Palaeoproterozoic of Karelia. Karelian
Research Centre, Petrozavodzk, pp 63–75 (in Russian)
Hedberg HD (1974) Relation of methane generation to under
compacted shales, shale diapirs, and mud volcanoes. Bull Am
Assoc Petrol Geologists 58:661–673
Hoefs J (2009) Stable isotope geochemistry, Sixth revised and updated
edition. Springer, Berlin, p 201
Hofmann HJ (1976) Precambrian microflora, Belcher Islands, Canada:
significance and systematic. J Palaeontol 50:1040–1073
Holland HD (2002) Volcanic gases, black smokers, and the great
oxidation event. Geochim Cosmochim Acta 66:3811–3826
Holland HD (2006) The oxygenation of the atmosphere and oceans.
Philos Trans R Soc Lond B Biol Sci B361:903–915
Honkamo M (1985) On the Proterozoic metasedimentary rocks of the
Northern Pohjanmaa schist area, Finland. Bull Geol Surv Finl
331:117–129
Hunt JM (1996) Petroleum geochemistry and geology. W.H. Freeman
& Co, New York, p 743
Inostranzev AA (1885) Geology. General lecture course for students of
the St.-Petersburg University, vol 1, 2nd edn. St-Petersburg Univer-
sity, St-Petersburg (in Russian)
Inostranzev AA (1886) Once more on shungite. Mining J 2:35–45
(in Russian)
Irwin H, Curtis C, Coleman M (1977) Isotopic evidence for the source
of diagenetic carbonate during burial of organic-rich sediments.
Nature 269:209–213
Jiracek GR, Ander ME, Holcomber HT (1986) Magnetotelluric
soundings of crustal conductive zones in major continental
rifts. In: Riecker RE (ed) Rio grande rift: tectonics and
magmatism. American Geophysical Union, Washington, DC, pp
209–222
Karhu JA (1993) Palaeoproterozoic evolution of the carbon isotope
ratios of sedimentary carbonates in the Fennoscandian Shield.
Bull Geol Surv Finl 371:1–87
Khavari-Khorasani G, Murchison DG (1979) The nature of Karelian
shungite. Chem Geol 26:165–182
Knoll AH, Summons RE, Waldbauer JR, Zumberge J (2007) The
geological succession of primary producers in the Oceans. In:
Falkowski PG, Knoll AH (eds) The evolution of primary producers
in the Sea. Elsevier, Boston, pp 133–163
Koistinen T, Stephens MB, Bogatchev V, Nordgulen Ø, Wennestr€omM,
Korhonen J (comps.) (2001) Geological map of the Fennoscandian
Shield, Scale 1:2 000 000. Espoo/Trondheim/Upsala/Moscow
1270 H. Strauss et al.
Kontinen A (1987) An early Proterozoic ophiolite – the Jormua
mafic–ultramafic complex, northeastern Finland. Precambrian Res
35:313–341
Korja T, Tuisku P, Pernu T, Karhu J (1996) Field, petrophysical and
carbon isotope studies on the Lapland Granulite Belt: implications
for deep continental crust. Terra Nova 8:48–58
Kovalevski VV, Buseck PR, Cowley JM (2001) Comparison of carbon
in shungite rocks to other natural carbons: an X-ray and TEM study.
Carbon 39:243–256
Krupenik VA, Akhmedov AM, Sveshnikova KYu (2011a) The section
of the Onega structure based on the Onega parametric drillhole. In:
Glushanin LV, Sharov NV, Shchiptsov VV (eds) The Onega
Palaeoproterozoic structure (Geology, tectonics, deep structure
and minerageny). Karelian Research Centre, Petrozavodsk, pp.
172–189 (in Russian).
Krupenik, V.A., Akhmedov, A.M., and Sveshnikova, K.Yu., (2011b)
Isotopic composition of carbon, oxygen and sulphur in the
Ludicovian and Jatulian rocks. In: Glushanin LV, Sharov NV,
Shchiptsov VV (eds) The Onega Palaeoproterozoic structure
(Geology, tectonics, deep structure and minerageny). Karelian
Research Centre, Petrozavodsk, pp. 250–255 (in Russian).
Kupryakov SV (1994) Geology and genesis of shungite rocks at
Zazhogino. In: Filippov MM (ed) The organic matter of Karelian
Shungite rocks (genesis, evolution and the methods of study).
Karelian Research Centre, Petrozavodsk, pp 93–98 (in Russian)
Kupryakov SV, Mikhailov VP (1988) Zazhogino deposit of shungite
rocks. In: News in geology of the North-West of the RSFSR.
Science (Nauka), Moscow, pp 9–86 (in Russian)
Lahtinen R, Huhma H, Kontinen A, Kohonen J, Sorjonen-Ward P
(2010) New constraints for the source characteristics, deposition
and age of the 2.1–1.9 Ga metasedimentary cover at the western
margin of the Karelian Province. Precambrian Res 176:77–93
Lancea S, Henrya P, Le Pichona X, Lallemanta S, Chamleyb H,
Rostekc F, Faugeresd J-C, Gonthierd E, Olue K (1998) Submersible
study of mud volcanoes seaward of the Barbados accretionary
wedge: sedimentology, structure and rheology. Mar Geol
145:255–292
Lehtonen M, Airo M-L, Eilu P, Hanski E, Kortelainen V, Lanne E,
Manninen T, Rastas P, R€as€anen J, Virransalo P (1998) The stratig-
raphy, petrology and geochemistry of the Kittil€a greenstone area,
northern Finland. A report of the Lapland volcanite project. Geol
Surv Finl, Rep Invest 140:1–144 (in Finnish with English summary)
Loukola-Ruskeeniemi K (1991) Geochemical evidence for a hydrother-
mal origin of sulphur, base metals and gold in Proterozoic
metamorphosed black shales, Kainuu and Outokumpu areas,
Finland. Miner Deposita 26:152–164
Loukola-Ruskeeniemi K (1999) Origin of black shales and the
serpentinite-associated Cu-Zn-Co ores at Outokumpu, Finland.
Econ Geol 94:1007–1028
Loukola-Ruskeeniemi K (2011) Graphite- and sulphide-bearing schists
in the Outokumpu R2500 drill core: comparison with the Ni-Cu-Co-
Zn-Mn-rich black schists at Talvivaara, Finland. Geol Surv Finl,
Spec Pap 51:229–252
Loukola-Ruskeeniemi K, Heino T (1996) Geochemistry and genesis of
the black shale-hosted Ni-Cu-Zn deposit at Talvivaara, Finland.
Econ Geol 91:80–110
Loukola-Ruskeeniemi K, Heino T, Talvitie J, Vanne J (1991) Base-
metal-rich metamorphosed black shales associated with Pro-
terozoic ophiolites in the Kainuu schist belt, Finland: a genetic
link with the Outokumpu rock assemblage. Miner Deposita
26:143–151
Lukkarinen H (2008) Pre-Quaternary rocks of the Siilinj€arvi and
Kuopio map-sheet areas. Explanation to the maps of Pre-Quaternary
rocks, sheets 3331 and 3242. Geological Survey of Finland, Espoo,
p 228 (in Finnish with English summary)
Lyell C (1830) Principles of geology. John Murry, London, p 586
Macaulay CM, Fallick AE, Haszeldine RS, Graham CM (2000)
Methods of laser-based stable isotope measurement applied to dia-
genetic cements and hydrocarbon reservoir quality. Clay Mineral
35:313–322
Mancuso JJ, Kneller WA, Quick J (1989) Precambrian vein
pyrobitumen: evidence for petroleum generation and migration
2 Ga ago. Precambrian Res 44:137–146
Manninen T (1991) Volcanic rocks in the Salla area, northeastern
Finland. A report of the Lapland volcanite project. Geol Surv Finl,
Rep Invest 104:1–97 (in Finnish with English summary)
Master S, Bekker A, Hofmann A (2010) A review of the stratigraphy
and geological setting of the Palaeoproterozoic Magondi Super-
group, Zimbabwe – type locality for the Lomagundi carbon isotope
excursion. Precambrian Res 182:254–273
McKirdy DM (1974) Organic geochemistry in Precambrian research.
Precambrian Res 1:75–137
McKirdy DM, Imbus SW (1992) Precambrian petroleum: a decade of
changing perceptions. In: Schidlowski M, Golubic S, Kimberley
MM, McKirdy DM, Trudinger PA (eds) Early organic evolution:
implications for mineral and energy resources. Springer, Berlin,
pp 176–192
Melezhik VA (1992) Early Proterozoic sedimentary and rock-forming
basins of the Baltic Shield. Nauka (Science), Leningrad, p 256 (in
Russian)
Melezhik VA, Predovsky AA (1982) Geochemistry of early Protero-
zoic lithogenesis. Nauka (Science), St. Leningrad, p 208 (in
Russian)
Melezhik VA, Basalaev AA, Predovsky AA, Balabonin NL, Bolotov
VI, Pavlova MA, Gavrilenko BV, Abzalov MZ (1988) Carbona-
ceous deposits of the earliest stages of earth evolution (geochemis-
try and depositional environments on the Baltic Shield). Nauka
(Science), Leningrad, p 197 (in Russian)
Melezhik VA, Fallick AE, Filippov MM, Larsen O (1999a) Karelian
shungite an indication of 2000 Ma-year-old metamorphosed oil-
shale and generation of petroleum: geology, lithology and geo-
chemistry. Earth Sci Rev 47:1–40
Melezhik VA, Fallick AE, Medvedev PV, Makarikhin VV (1999b)
Extreme 13Ccarb enrichment in ca. 2.0 Ga magnesite-stromatolite-
dolomite-“ red beds” association in a global context: a case for the
world-wide signal enhanced by a local environment. Earth Sci Rev
48:71–120
Melezhik VA, Filippov MM, Romashkin AE (2004) A giant Palaeopro-
terozoic deposit of shungite in NW Russia: genesis and practical
applications. Ore Geol Rev 24:135–154
Melezhik VA, Fallick AE, Filippov MM, Lepland A, Rychanchik DV,
Deines YE, Medvedev PV, Romashkin AE, Strauss H (2009) Petro-
leum surface oil seeps from a Palaeoproterozoic petrified giant
oilfield. Terra Nova 21:119–126
Mishunina ZA (1979) Petroleum genesis in carbonate formations. In:
Nalivkin VD, Ivanchuk PK, Dvali MF (eds) Problems in the geol-
ogy of oil and gas. Nedra, Leningrad, pp 139–143 (in Russian)
Morozov AF, Hakhaev BN, Petrov OV, Gorbachev VI, Tarkhanov GB,
Tsvetkov LD, Erinchek YuM, Akhmedov AM, Krupenik VA,
Sveshnikova KYu (2010) Rock-salts in Palaeoproterozoic strata of
the Onega depression of Karelia (based on data from the Onega
parametric drillhole). Trans Acad Sci 435(2):230–233 (in Russian)
Mossman D, Eigendorf G, Tokaryk D, Gauthier-Lafaye F, Guckert KD,
Melezhik VA, Farrow CEG (2003) Testing for fullerenes in geo-
logic materials: Oklo carbonaceous substances, Karelian shungites,
Sudbury Black Tuff. Geology 31:255–258
Mossman DJ, Gauthier-Lafaye F, Jackson SE (2005) Black shales,
organic matter, ore genesis and hydrocarbon generation in the
Paleoproterozoic Franceville Series, Gabon. Precambrian Res
137:253–272
Mutanen T, Huhma H (2001) U-Pb geochronology of the Koitelainen,
Akanvaara and Keivitsa layered intrusions and related rocks. In:
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1271
Vaasjoki M (ed) Radiometric age determinations from Finnish
Lapland and their bearing on the timing of Precambrian volcano-
sedimentary sequences. Geol Surv Finl, Spec Pap 33:229–246
Neruchev SG, Rogosina EA, Shimansky BK (1998) Reference on
petroleum and gas geochemistry. Nedra, St. Petersburg, p 575 (in
Russian)
Onuma N, Clayton RN, Mayeda TK (1972) Oxygen isotope
cosmothermometer. Geochim Cosmochim Acta 36:169–188
Ovchinnikova GV, Kusnetzov AB, Melezhik VA, Gorokhov IM,
Vasileva IM, Gorokhovsky BM (2007) Pb-Pb age of Jatulian car-
bonate rocks: the Tulomozero Formation in south-eastern Karelia.
Stratigr Geol Correl 4:20–33
Paakkola J (1971) The volcanic complex and associated manganiferous
iron formation of the Porkonen-Pahtavaara area in Finnish Lapland.
Bull Geol Surv Finl 247:1–83
Paakkola J, Geh€or S (1988) The lithofacies associations and sedimen-
tary structures of the iron-formations in the early Proterozoic Kittil€agreenstone belt, northern Finland. Geol Surv Finl, Spec Pap
5:213–238
Papineau D, Purohit R, Goldberg T, Pi D, Shields G, Bhu HR, Steele A,
Fogel ML (2009) High productivity and nitrogen cycling after the
Paleoproterozoic phosphogenic event in the Aravalli Supergroup,
India. Precambrian Res 171:37–56
Pekkarinen L (1979) The Karelian formations and their depositional
basement in the Kiihtelysvaara-Tohmaj€arvi district, eastern
Finland. Bull Geol Surv Finl 301:1–141
Pekkarinen LJ, Lukkarinen H (1991) Paleoproterozoic volcanism in the
Kiihtelysvaara – Tohmaj€arvi district, eastern Finland. Bull Geol
Surv Finl 357:1–30
Peltonen P (2005) Ophiolites. In: Lehtinen M, Nurmi PA, R€am€o OT
(eds) Precambrian geology of Finland – key to the evolution of the
Fennoscandian Shield. Elsevier, Amsterdam, pp 237–278
Peng Y, Baoa H, Yuan X (2009) New morphological observations for
Paleoproterozoic acritarchs from the Chuablinggou Formation,
North China. Precambrian Res 168:223–232
Perttunen V, Hanski E (2003) Pre-Quaternary rocks of the Koivu and
T€orm€asj€arvi map-sheet areas. Explanation to the geological map of
Finland 1:100 000, pre-Quaternary rocks, sheets 3631 and 2633.
Geological Survey of Finland, Espoo, p 88 (in Finnish with English
summary)
Prokoph A, Shields GA, Veizer J (2009) Compilation and time-series
analysis of a marine carbonate d18O, d13C, 87Sr/86Sr and d34Sdatabase through Earth history. Earth Sci Rev 87:113–133
Puchtel IS, Zhuravlev DZ, Ashikhmina NA, Kulikov VS, Kulikova VV
(1992) Sm-Nd age of the Suisarian suite on the Baltic Shield. Trans
Russ Acad Sci 326:706–711 (in Russian)
Puchtel IS, Arndt NT, Hofmann AW, Haase KM, Kr€oner A, KulikovVS, Kulikova VV, Garbe-Sch€onberg C-D, Nemchin AA (1998)
Petrology of mafic lavas within the Onega plateau, central Karelia:
evidence for the 2.0 Ga plume-related continental crustal growth in
the Baltic Shield. Contrib Mineral Petr 130:134–153
Puchtel IS, Br€ugmann GE, Hofmann AW (1999) Precise Re-Os mineral
isochron and Pb-Nd-Os isotope systematics of a mafic-ultramafic
sill in the 2.0 Ga Onega plateau (Baltic Shield). Earth Planet Sci
Lett 170:447–461
Rankama K (1948) New evidence on the origin of pre-Cambrian
carbon. Bull Geol Soc Am 59:389–416
Rasmussen B, Fletcher IR, Brocks JJ, Kilburn MR (2008) Reassessing
the first appearance of eukaryotes and cyanobacteria. Nature
455:1101–1105
Redfield AC (1958) The biological control of chemical factors in the
environment. Am Sci 46:205–221
Ryabov NI (1948) Description of the Karelian shungite deposits.
Unpublished report 1314, Karelian geological expedition, p 51
(Russian)
Schidlowski M (1988) A 3,800-million-year isotopic record of life from
carbon in sedimentary rocks. Nature 333:313–318
Schidlowski M, Appel PWU, Eichmann R, Junge CE (1979) Carbon
isotope geochemistry of the 3.7. 109 yr-old Isua sediments, West
Greenland: implications for the Archaean carbon and oxygen
cycles. Geochim Cosmochim Acta 43:189–199
Schidlowski M, Hayes JM, Kaplan IR (1983) Isotopic inferences of
ancient biochemistries: carbon, sulfur, hydrogen, and nitrogen. In:
Schopf JW (ed) Earth’s earliest biosphere: its origin and evolution.
Princeton University Press, Princeton, pp 149–186
Schlesinger WH (1997) Biogeochemistry – an analysis of global
change, 2nd edn. Academic, San Diego, p 588
Shatzky GV (1990) Isotope composition of sulphides from the
Zazhogino shungite deposit. Lithol Mineral Deposits 1:20–28 (in
Russian)
Sheppard SMF (1977) The Cornubian batholiths, SW England: D/H
and 18O/16O studies of kaolinite and other alteration minerals.
J Geol Soc Lond 133:573–591
Shields G, Veizer J (2002) Precambrian marine carbonate isotope
database: version 1.1. Geochem Geophys Geosyst 3:1–12
Silvennoinen A (1972) On the stratigraphic and structural geology of the
Rukatunturi area, northeastern Finland. Geol Surv Finl Bull 257:1–48
Skilling IP, White JDL, McPhie J (2002) Peperite: a review of
magma–sediment mingling. J Volcanol Geotherm Res 114:1–7
Sozinov NA, Chistyakova NN, Kazantsev VA (1988) Metallogenic
black shales of the Kursk Magnetic anomaly. Nauka (Science),
Moscow, p 149 (in Russian)
Stadnitskaia A, Ivanov MK, Poludetkina EN, Kreulen R, van Weering
TCE (2008) Sources of hydrocarbon gases in mud volcanoes from
the Sorokin Trough, NE Black Sea, based on molecular and carbon
isotopic compositions. Mar Petrol Geol 25:1040–1057
Strauss H, Moore TB (1992) Abundances and isotopic compositions of
carbon and sulphur species in whole rock and kerogen. In: Schopf
JW, Klein C (eds) The Proterozoic biosphere: a multidisciplinary
study. Cambridge University Press, Cambridge, pp 709–797
Strauss H, Des Marais DJ, Summons RE, Hayes JM (1992) The carbon-
isotopic record. In: Schopf JW, Klein C (eds) The Proterozoic
biosphere: a multidisciplinary study. Cambridge University Press,
Cambridge, pp 117–127
Summons RE, Jahnke LL, Hope JM, Logan GA (1999) 2-
Methylhopanoids as biomarkers for cyanobacterial oxygenic photo-
synthesis. Nature 400:554–557
Taran LN, Onoshko MP, Mikhailov ND (2011) Structure and composi-
tion of organic matter and isotope geochemistry of the Palaeopro-
terozoic graphite and sulphide-rich metasedimentary rocks from the
Outokumpu deep drill hole, eastern Finland. In: Kukkonen IT (ed)
Outokumpu deep drilling project 2003–2010. Geol Surv Finl, Spec
Pap 51:219–228
Taylor HP (1974) The application of oxygen and hydrogen isotope
studies to problems of hydrothermal alteration and deposition.
Econ Geol 69:843–883
Taylor HP, Sheppard SMF (1986) Igneous rocks: I. Processes of
isotopic fractionation and isotope systematics. Rev Mineral
16:227–271
Thomazo C, Pinti D, Busigny V, Ader M, Hashizume K, Philippot P
(2009) Biological activity and the Earth’s surface evolution:
insights from carbon, sulfur, nitrogen and iron stable isotopes in
the rock record. C R Palevol 8:665–678
Thomazo C, Ader M, Philippot P (2011) Extreme 15N-enrichments in
2.72-Gyr-old sediments: evidence for a turning point in the nitrogen
cycle. Geobiology 9:107–120
Tissot BP, Welte DH (1984) Petroleum formation and occurrence.
Springer, Berlin, p 699
Veizer J, Ala D, Azmy K, Bruckschen P, Buhl D, Bruhn F, Carden GAF,
Diener A, Ebneth S, Godderis Y, Jasper T, Korte C, Pawellek F,
1272 H. Strauss et al.
Podlaha OG, Strauss H (1999) 87Sr/86Sr, d13C and evolution of
Phanerozoic seawater. Chem Geol 161:59–88
Weber F, Schidlowski M, Arneth JD, Gauthier-Lafaye F (1983) Carbon
isotope geochemistry of the lower proterozoic Francevillian series
of Gabon (Africa). Terra Cognita 3:220
Wilton DHC (1996) Palaeoproterozoic, 1.88–2.0 Ga, organic matter
from the Mugford Kaumajet mountain group, northern Labrador.
Precambrian Res 77:131–141
Winter BL, Knauth LP (1992) Stable isotope geochemistry of
cherts and carbonates from the 2.0 Ga Gunflint iron formation:
implications for the depositional setting, and the effects of
diagenesis and metamorphism. Precambrian Res 59:283–313
Worden KE, Carson CJ, Scrimgeour IR, Lally JH, Doyle N (2008)
A revised Palaeoproterozoic chronostratigraphy for the central Pine
Creek Orogen, northern Australia: evidence from SHRIMP U-Pb
zircon geochronology. Precambrian Res 166:122–144
Yudovich YE, Makarikhin VV, Medvedev PV, Sukhanov NV (1991)
Carbon isotope anomalies in carbonates of the Karelian Complex.
Geochem Int 28:56–62
Zakrutnin VE, Zhmur SI (1989) Carbonaceous-rich formation in the
Kursk magentic anomaly, lower Proterozoic. Rostov University,
Rostov on Don, p 125 (in Russian)
Zhang Z (1986) Clastic facies microfossils from the Chuanlinggou
Formation (1800 Ma) near Jixian, North China. J Micropalaeontol
5:9–16
Zhao G, Cawood PA, Wilde A, Sun M (2002) Review of global
2.1–1.8 Ga orogens: implications for a pre-Rodinia supercontinent.
Earth Sci Rev 59:125–162
6 7.6 Enhanced Accumulation of Organic Matter: The Shunga Event 1273
7.7 The Earliest Phosphorites: Radical Change in thePhosphorus Cycle During the Palaeoproterozoic
Aivo Lepland, Victor A. Melezhik, Dominic Papineau,Alexander E. Romashkin, and Lauri Joosu
7.7.1 Introduction to the Phosphorus Cycle
Phosphate is an essential and growth-limiting nutrient
required by all forms of life, as it is a key component of
many important macro-molecules. These macro-molecules
are involved in energy transport, information storage, and
structural support functions include membrane lipids, proteins,
and nucleic acids. The global phosphorus cycle, which includes
only dissolved and solid phases without any gaseous
components, is strongly influenced by biological processes
(Gulbrandsen 1969; Jahnke 1992; F€ollmi 1996). Continental
weathering and riverine discharges are the most important
sources delivering both particulate and dissolved phosphate
into the oceans (Froelich et al. 1982; F€ollmi 1995). Long-
term changes in the phosphorus cycle, such as variations in
sources, concentration of dissolved seawater phosphate, forma-
tion of phosphorite deposits, and sequestration in biomass, are
linked with other biogeochemical cycles and track major
changes in Earth’s environmental conditions (Sheldon 1980;
Baturin 1982; Papineau 2010; Planavsky et al. 2010). Biologic
influence upon the phosphorus cycle can be traced back to the
early Archaean (Blake et al. 2010). Ancient biologic processing
of phosphate is inferred from the oxygen isotope ratios of some
phosphates in 3200–3500 Ma sediments that are similar to
those of modern marine biogenic phosphates (Blake et al.
2010).
Phosphate minerals are common constituents in Archaean
sedimentary rocks, particularly in the banded iron formations
(Trendall and Blockley 1970; Ewers andMorris 1981; Dymek
and Klein 1988; Pecoits et al. 2009; Planavsky et al. 2010;
Papineau et al. 2011) though the concentrations are generally
low (P2O5 < 1 %). Low phosphorus concentrations charac-
terise the sedimentary rock record until c. 2000 Ma in the
Palaeoproterozoic Era when phosphate-rich deposits suddenly
appeared worldwide in several sedimentary successions
(Fig. 7.84; Yudin 1996; Bekker et al. 2003; Melezhik et al.
2005; Papineau 2010). Many of these Palaeoproterozoic phos-
phatic deposits have been described as phosphorites though
the term phosphorite has not been clearly defined in the
literature (Bentor 1980). Different lower limits for P2O5 con-
tent (either 20 %, 18 %, 15 % or 10 %) are used to define a
phosphorite, whereas in many studies, including this contri-
bution, the term is geochemically not precisely defined, but
rather refers to a sedimentary rock that contains abundant
phosphate (Cook and Shergold 1986a).
7.7.2 Formation of Phosphorites:Phosphogenesis
Formation of phosphorites by direct precipitation from the
water column in areas of elevated dissolved phosphate
concentrations such as in upwelling zones has been proposed
by earlier workers (Kazakov 1938). Later studies have,
however, shown that direct precipitation on the seafloor
may occur in restricted areas such as hardgrounds experiencing
low or no sedimentation (F€ollmi and Garrison 1991), but
overall, this mechanism is considered insignificant (Bentor
1980; Cook and Shergold 1986b; Compton et al. 2000) due
to slow kinetics of carbonate fluorapatite (main authigenic
phosphate mineral) precipitation (F€ollmi 1996), and inhibiting
effects of dissolved magnesium (Martens and Harris 1970).
Instead, the shallow levels of the sediment column, close to
the sediment water interface within the oxic to suboxic diage-
netic zone have shown to be the main sites of phosphogenesis
due to elevated interstitial dissolved phosphate and availabil-
ity of nucleation templates (Lamboy 1993; Jarvis et al. 1994;
Savenko 2010). Desorption of scavenged phosphate fromMn-
and Fe-oxyhydroxides (Berner 1973) and release of phos-
phate from decomposing organic matter are the principal
sources of interstitial phosphate in the oxic-suboxic diage-
netic environment. Whereas Mn- and Fe-oxyhydroxides can
A. Lepland (*)
Geological Survey of Norway, Postboks 6315 Sluppen, NO-7491
Trondheim, Norway
e-mail: [email protected]
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_7, # Springer-Verlag Berlin Heidelberg 2013
1275
be important initial carriers and concentrators of phosphate
in areas experiencing submarine hydrothermal venting and
significant continental runoff (Glenn et al. 1994), the
mineralisation of organic matter is by far the most important
phosphate source for phosphogenesis (Berner et al. 1993;
Krajewski et al. 1994).
Phosphogenesis is, however, not restricted to the oxic-
suboxic zone, but commonly occurs in the sulphidic diage-
netic environment of organic-rich sediments where a variety
of microorganisms including sulphate-reducing and sulphide-
oxidising bacteria decompose organic matter, and control the
uptake and release of phosphate (Burnett 1977; Schulz and
Schulz 2005; Bailey et al. 2007; Arning et al. 2009;
Tribovillard et al. 2010). Diagenetic pyrite formed bymicrobial
sulphate reduction represents one of the most common acces-
sory mineral in phosphorites (Baturin 1982; McArthur 1985;
Xiao and Knoll 1999; Gorshkov et al. 2000). Ultimately, one
important condition for phosphogenesis is ample phosphate
supply achieved through the presence of vast amounts of
decomposable organic matter within the sediment column
(Bentor 1980; Lucas and Prevot-Lucas 2000). Areas supporting
high primary productivity and accumulation of organic-rich
sediments on the continental shelves and slopes, and in estuarine
and deltaic environments are thus the preferred sites of
phosphogenesis (O’Brien and Veeh 1980; Bremner and Rogers
1990; Ruttenberg andBerner 1993). Changes inweathering and
sedimentation rates, ocean circulation and oxygen levels, nutri-
ent supply, and climate and tectonic regime influence primary
productivity and accumulation of organic-rich sediments, and
control temporal trends in phosphogenesis on a geologic time-
scale (Papineau 2010). Compilation of phosphorite data through
the geological record allows recognition of at least four impor-
tant periods of phosphogenesis through Earth history:
Palaeoproterozoic, late Neoproterozoic-Cambrian, late
Palaeozoic and Cretaceous-Holocene (Notholt et al. 1989;
Papineau 2010).
7.7.3 Palaeoproterozoic Phosphorites
Worldwide occurrences of sedimentary phosphate deposits
around 2000 Ma (Fig. 7.84) suggest that phosphogenesis had
a common underlying cause and was related to global event(s)
in the sequence of tectonic and environmental perturbations
(for details see Chap. 1.1) following the rise of atmospheric
oxygen at c. 2300 Ma (Bekker et al. 2004). Formation and
preservation of phosphate-rich rocks in the Palaeoproterozoic
occurred only where local conditions of deposition were
adequate. Phosphogenesis was likely related to the overall
supply of phosphorus to the oceans from continental
weathering, which is intimately connected to tectonic and
climatic perturbations. The geodynamic setting that preceded
the Palaeoproterozoic phosphogenic event included a surge in
tectonic activity that appears to have occurred during the
assembly of large continental landmasses in the Neoarchaean
(possibly one or more supercontinents) and their wide-spread
rifting and break-up in the earliest Palaeoproterozoic (Aspler
and Chiarenzelli 1998; Bleeker 2003; Zhao et al. 2003; Barley
et al. 2005). The final separation and dispersion of such large
continental landmasses was completed around
2100–2000 Ma, when a plume-related event occurred
(Heaman 1997; Barley et al. 2005; Halls et al. 2008). Break-
up of these ancient large pieces of continental crust may have
been caused by the development of large igneous provinces
(Ernst and Bleeker 2010) and led to the creation of new rift-
bound Palaeoproterozoic sedimentary basins.
Such extensional basins were likely to experience abun-
dant sediment and phosphorus (and other nutrient) supply
from freshly rearranged, erosion-prone nearby landmasses,
and had the potential for accumulating thick organic- and
phosphate-rich successions that can be preserved in the rock
record. Sediment and nutrient discharges were also influenced
by changes in chemical weathering rates over geological time
scales. Intensive chemical weathering and elevated discharge
of phosphorus and other nutrients are expected to occur dur-
ing post-glacial periods due to perturbation of the atmospheric
CO2 levels and related carbon-silicate cycle (Berner 1993;
Godderis et al. 2007). High nutrient supply to marine basins
during such periods may have had a stimulating effect on
primary productivity and in the accumulation of organic- and
phosphate-rich sediments. Continental discharges likely had
most influence upon phosphogenesis in rift basins, whereas
upwelling processes may have controlled nutrient supply,
primary productivity and phosphogenesis in epeiric seas
along passive continental margins. A compilation of
Palaeoproterozoic phosphate deposits was presented in
Papineau (2010), and the scope of this contribution is to
describe the geological setting and nature of a selection of
Palaeoproterozoic phosphate-rich sedimentary rocks in the
context presented above.
C. 2000 Ma Lower Aravalli Group, Rajasthan,India
Stromatolitic phosphorites are common in the Jhamarkotra
Formation of the Palaeoproterozoic Lower Aravalli Group in
the north western Indian Shield. The minimum age of the
Aravalli Supergroup provided by intruding Darwal Granite is
1900 � 80 Ma (Choudhary et al. 1984), while the Pb-Pb
isochron of lower Aravalli carbonates has given an age of
1921 � 67 Ma (Sarangi et al. 2006). The Pb-Pb model ages
of galena from the basal Aravalli volcanic rocks indicate an
age of 2075–2150 Ma (Deb and Thorpe 2004). The Sm–Nd
model ages on Lower Aravalli komatiites and tholeiites sug-
gest active volcanism around 2.3–1.8 Ga (Ahmad et al.
2008). The Lower Aravalli Group unconformably overlies
the Archaean basement (Heron 1953), and its lowermost part
1276 A. Lepland et al.
comprises palaeosols, polymicitic and diamictic conglo-
merates, and sandstones of the Delwara Formation. These
coarse-grained metasedimentary rocks are overlain by
shallow-marine dolostones and carbonaceous shales of the
Jhamarkotra Formation, occurring in a series of sub-basins,
most within a 30 km radius of the city of Udaipur. The
depositional environment of the Lower Aravalli Group was
an active rift basin and sedimentation was transgressive (Roy
and Paliwal 1981). The rocks have experienced greenschist
facies metamorphism with evidence for amphibolite-facies
alteration in some areas (Choudhuri 1989).
The columnar stromatolitic phosphorites in the lower part
of the Jhamarkotra dolomite horizon are exceptionally well-
preserved (Fig. 7.85a, b), but in places they also occur as
brecciated and fragmented phosphorites. Carbonate
fluorapatite is the main constituent of stromatolite columns,
forming convex, dark grey laminae. These phosphatic
laminae occur in intimate petrographic relation with organic
matter, dolomite and calcite (Fig. 7.85c, d), and occasionally
with chert, sulphides, clay minerals, and iron oxides
(Banerjee 1971). The intercolumnar space is occupied by
dolomite, phosphorite fragments and minor quartz (Choudhuri
1989). The thickness of stromatolitic phosphorite units within
the Jhamarkotra dolomite varies from 5 to 35 m. The occur-
rence of Aravalli phosphorites within stromatolitic structures is
consistent with a biochemical formation mechanism. However,
the laminated nature of Aravalli phosphorites has also been
used to argue for their formation through direct authigenic
chemical precipitation (Choudhuri 1989). The whole-rock
P2O5 content is up to 37 wt.% (Banerjee 1971) making these
deposits one of the most significant economic phosphorites
from the Palaeoproterozoic.
Among the Lower Aravalli sub-basins, the occurrence of
abundant phosphorites is restricted to the ones that contain
isotopically normal (d13Ccarb ~ 0‰) stromatolitic dolostones.
On the other hand, the phosphorites are absent in basins in
which the dolostones are massive and isotopically heavy
(d13Ccarb up to 11‰) (Sreenivas et al. 2001; Purohit et al.
2010). This d13Ccarb contrast between phosphatic and non-
phosphatic Aravalli sub-basins has been interpreted in differ-
ent ways. Roy and Paliwal (1981), Sreenivas et al. (2001) and
Purohit et al. (2010) inferred that the observed contrasts in
isotopic composition and phosphorus content were caused by
differences in depositional setting, biologic productivity and
diagenetic environment. Phosphatic sub-basins with
stromatolites represent land-locked epicontinental self regions
that supported high cyanobacterial productivity whereas non-
phosphatic sub-basins with massive dolostones represent
shelf-bank and restricted hypersaline environments (Roy and
Paliwal 1981; Sreenivas et al. 2001; Purohit et al. 2010). In
contrast, Maheshwari et al. (2010) reported that 13C-rich, non-
stromatolitic, phosphorus-free dolostones were deposited
prior to the formation of isotopically “normal” phosphatic
stromatolites, thus questioning the palaeoenvironmental inter-
pretation outlined above. The relationship between Aravalli
phosphorites and isotopically heavy carbonates remains thus
unresolved due to the complicated structural and metamor-
phic history of the region, and lack of chronostratigraphic
markers for robust correlation between tectonically discon-
nected sub-basins (Maheshwari et al. 2010).
Fig. 7.84 Distribution of Paleaeoproterozoic rocks and geological units with their respective ages, containing phosphorites (Data from
compilation of Papineau (2010))
7 7.7 The Earliest Phosphorites: Radical Change in the Phosphorus Cycle During the Palaeoproterozoic 1277
Phosphorites of the PalaeoproterozoicFennoscandian Shield
Phosphate-rich horizons within a variety of lithologies
including siliciclastic and carbonate sediments, banded iron
formations and volcanic rocks have been reported from
numerous Palaeoproterozoic supracrustal successions in the
Fennoscandian Shield (Fig. 7.84) (Rehtij€arvi et al. 1979;Vaasjoki et al. 1980; Aik€as 1989). In the Finnish part of
the shield, phosphorites associated with the organic-rich
siliciclastic and carbonate sediments are commonly
uraniferous, and many of these occurrences have been
revealed and characterised in connection with uranium
prospecting (Laajoki and Saikkonen 1977; Aik€as 1980,
Fig. 7.85 Stromatolitic phosphorites from the Palaeoproterozoic
Lower Aravalli Group in India. (a, b) Outcrop photos of columnar
stromatolitic phosphorites in cross- (a) and bedding-parallel (b)
sections from the Jhamarkotra Formation; higher resistance to
weathering of apatite-rich columns results in higher relief compared
to dolomite. (c, d) Back-scattered electron images of the Aravalli
stromatolitic phosphorite illustrating different scales of laminations of
apatite (light grey) and dolomite (dark grey) in stromatolitic columns
(Images by Dominic Papineau)
1278 A. Lepland et al.
1981). However, fluorapatite bands associated with organic
and sulphide-rich banded iron formations in the Kainuu
Schist Belt are typically non-uraniferous (Laajoki 1975;
Geh€or 1994). The REE characteristics of these
phosphate-rich horizons are consistent with their formation
in a marine environment (Laajoki 1975; Rehtij€arvi 1983;Geh€or 1994). In the Finnish part of the Fennoscandian
Shield, c. 20 occurrences of Palaeoproterozoic phosphorites
with up to 24 wt.% P2O5 have been reported from 2100 to
1900 Ma old successions (Aik€as 1989). Possibly coeval
phosphorites in banded iron formations in Swedish Lapland
contain up to 18 % P2O5 (Parak 1973). Phosphorites have
also been reported from the Russian part of the Fennoscandian
Shield, and three of these occurrences are detailed below.
Il’mozero Sedimentary Formation, Imandra/Varzuga Greenstone Belt, Kola Peninsula, Russia
Studies of carbonate and silicate concretions in the early
Palaeoproterozoic Imandra/Varzuga Greenstone Belt revealed
that some of them contain up to 1.2 wt.% P2O5 (Melezhik and
Predovsky 1978, 1982; Melezhik 1992). These phosphate-
bearing concretions occur in the Il’mozero Sedimentary For-
mation (Fig. 7.86a), (for details see Chap. 4.1), for which its
maximum depositional age has been constrained to c. 2051 Ma
by dating clastic zircons derived from the underlying volcanic
formation (Martin et al. 2010).
The Il’mozero Sedimentary Formation consists from bot-
tom to top of Tuffite, Greywacke, Dolostone-Chert, and
Black Shale members (Melezhik and Predovsky 1982) and
rests with erosional contact on subaerially erupted alkaline
basalts, andesites and dacites. Various concretions including
the phosphate-bearing ones have been found in the Varzuga
section of the formation (Figs. 7.86b, c, e-g and 7.87a, b). In
this section, the formation starts with the Greywacke mem-
ber, which rests on mafic lava with brecciated top. The lower
part of the member comprises c. 5-m-thick, massive,
volcaniclastic sandstone, indistinctly-laminated grey silt-
stone (c. 25 m) and dark grey siltstone (c. 20 m). The grey
siltstone is organic-poor, whereas the dark grey variety is
relatively organic-rich (up to 0.2 wt.%; Melezhik 1992). The
grey siltstone contains scattered calcite concretions whereas
the dark grey siltstone does not. The concretions (up to
10 � 5 cm in size) have a pancake-like shape and show a
significant depletion in 13C with d13Ccarb ranging between
�4 ‰ and �20 ‰ (Melezhik and Fallick 1996), implying
decomposed organic matter as bicarbonate source for calcite
precipitation. Fine-grained clastic material and faint parallel
lamination of both grey and dark grey siltstone suggest
deposition of the lower part of the Greywacke member in a
deep basin below fair- and storm-weather wave base.
The middle and upper parts of the Greywacke member
represent two coarsening- and thickening-upward cycles.
The lower succession is a c. 35-m-thick unit of rhythmically
bedded siltstone and clayey siltstone with rare, thin
(0.5–1.5 cm), coarse-grained greywacke beds. These thin
greywacke beds have erosive bases and rapidly grade into
siltstone. Phosphate-bearing concretions occur in the upper
part of the cycle (Fig. 7.86c). The second succession (c. 45m)
starts with fine-grained greywacke-siltstone exhibiting irreg-
ular rhythmic bedding, which was disrupted by numerous
small-scale erosional channels filled with coarse-grained
greywacke (Fig. 7.86d). Both in-place and redeposited
phosphate-bearing silicate and calcite concretions are abun-
dant, particularly in the lower part of the succession. Both
cyclic successions deposited from turbidity currents in a
deep, clastic shelf environment below fair- and storm-
weatherwave base. A few-meters-thick andmassive greywacke
bed at the top of the Greywacke member signifies a
restructuring of the basin and marks a sequence boundary.
The Dolostone-Chert member is represented by a c.
70-m-thick unit of pale pink, crystalline dolostone with
limestone base that is followed by a c. 60-m-thick unit of
pervasively silicified stromatolitic dolostones with numer-
ous chert beds (Fig. 7.86c). The overall depositional trend
indicates basin shallowing, establishment of a carbonate
shelf, followed by onset of a carbonate platform inhabited
by stromatolite-forming cyanobacteria. There are no
phosphorus-rich rocks reported from the Dolostone-Chert
member (Melezhik and Predovsky 1982; Melezhik 1992).
The total thickness of the strata containing phosphate-
bearing concretions is c. 60 m. Concretions form c. 1.5 % of
the rock volume. Although concretions occur as irregular
clusters, they always formed within sandy layers, and close
to shale-greywacke boundaries (Melezhik 1992; Fig. 7.86e–g).
Concretions exhibit mainly flattened (pancake-like) or sub-
spherical shapes. Flattened concretions have been observed
coalesced in a train-like manner. Some concretions are zoned
(Fig. 7.86e), whereas others retain inherited sediment layering
and have a differential-compaction coefficient in the range of
2–4 (Fig. 7.86f, g), thus implying formation during an early
diagenetic stage (Melezhik 1992).
The greywackes that host concretions are composed of
albite and microcline (20–30 %), leucoxene (5–10 %), quartz,
haematite, titanomagnetite, epidote, magnetite, and andesite
clasts emplaced in chlorite matrix (30–40%). The concretions
comprise quartz, microcline, chlorite, apatite and titanite with
minor monazite, biotite, sericite, titanomagnetite and
7 7.7 The Earliest Phosphorites: Radical Change in the Phosphorus Cycle During the Palaeoproterozoic 1279
leucoxene. The outer rim (1–3 mm) of zoned concretions is
particularly enriched in apatite co-occurring with quartz and
chlorite and minor K-feldspar and monazite (Fig. 7.87a, b).
Some concretions contain up to 1 wt.% of organic matter
(Melezhik 1992).
The concretion-forming process is tentatively linked to
the decomposition of organic matter within initially Corg-
rich shale layers/beds leading to CO2 and phosphate release.
Relatively coarse-grained, sandy layers may have served as
conduits for phosphate-rich diagenetic fluids as well as
nucleation sites for apatite precipitation (Melezhik 1992).
Pilguj€arvi Sedimentary Formation, PechengaGreenstone Belt, Kola Peninsula, Russia
Phosphorus-rich tuffitic schist (5.12 wt.% P2O5), gritstones
and coarse-grained sandstones (up to 8.2 wt.% P2O5) in the
Pilguj€arvi Sedimentary Formation have been known since
the early studies by Akhmedov (1973) and Bekasova and
Dudkin (1981). The Pilguj€arvi Sedimentary Formation is the
thickest sedimentary unit of the North Pechenga Group of
the Pechenga Greenstone Belt (Fig. 7.88a; for details see
Chap. 4.2). The 2004 � 9 Ma Re-Os age of the formation
obtained on sedimentary pyrite and organic matter (Hannah
et al. 2006) is in agreement with the U-Pb zircon age of
1970 � 5 Ma reported from the felsic tuff beds within
overlying basaltic rocks of the Pilguj€arvi Volcanic Forma-
tion (Hanski 1992).
The Pilguj€arvi Sedimentary Formation rests with deposi-
tional contact on a c. 1,800-m-thick pile of tholeiitic pillow
lavas. It reaches a thickness of c. 1,000 m. However, this may
diminish to c. 500m if the space occupied by numerous gabbro
and differentiated gabbro-wehrlite intrusions is removed. The
formation consists largely of rhythmically bedded, Corg- and
sulphide-rich sandstone-siltstone-mudstone (Bekasova 1985)
deposited by turbidity currents in a deep-water continental
slope environment (Akhmedov and Krupenik 1990).
Lithological logs of numerous deep drillholes have shown
that the phosphate-bearing strata typically occur in association
with gritstone and coarse-grained sandstones. These
phosphate-rich, coarse-grained horizons with a thickness of
50–200m (Fig. 7.88b), and strike-length of c. 10 kmwere first
described from the central part of the formation (Bekasova and
Dudkin 1981). Later drilling data have shown that the
phosphate-bearing, coarse-grained lithofacies is wedging out
basin-dip over a distance of c. 400 m (Fig. 7.88b) and is
overstepped by rhythmically bedded, fine-grained siltstone-
shale of distal turbiditic facies (Fig. 7.89a). The phosphate-
bearing strata have been interpreted as part of either a deltaic
system (Bekasova 1985), submarine slope-slide facies
(Akhmedov and Krupenik 1990) or a long-term operating
fan system (Melezhik et al. 1998).
The phosphate-bearing lithofacies occurs as a series of
beds, lenses and small-scale channels (1–10 cm) of
gritstones and coarse-grained sandstones associated with
black, rhythmically bedded siltstone-shale (Fig. 7.89b).
The gritstone that hosts well-rounded phosphatic particles
is composed of clasts of bedded sandstones, laminated
shales, mafic lavas, vein quartz, quartzites, carbonate
rocks and quartz-muscovite schists (Fig. 7.89c, d). Because
of high abundance of pyrite and pyrrhotite (>50 vol.% in
places) with sizes from sub-mm to 20 mm, these gritstones
have locally been named “golden gritstones”. The phos-
phatic clasts range in size from 1 to 5 mm, form less
than 5 vol.% of the gritstone, and are randomly distributed
within beds. Some gritstone beds rapidly grade into coarse-
grained greywacke containing less phosphatic and sulphide
clasts but may include outsized, angular or rounded and
softly-deformed fragments of siltstone and mudstone
(Fig. 7.89e). Some gritstone beds also contain outsized,
softly-deformed fragments of bedded and laminated
siltstone-mudstone (Fig. 7.89f).
Bekasova and Dudkin (1981) considered the phosphatic
particles as concretions while acknowledging the redeposited
nature for some of them. However, the phosphatic particles
exhibit an allochthonous, redeposited origin, and in-place
phosphatic particles and/or “concretions” are yet to be
found. The gritstones are most enriched in phosphatic clasts
with respect to other lithologies (Table 7.7). Abundant
occurrences of relatively large phosphatic clasts in gritstones
and coarse-grained sandstones suggest selective sorting and
enrichment through high-energy, erosional reworking of
the primary sediment and winnowing of finer-grained
components. The formation of such intraclastic phosphorites
with the aid of reworking points to a semi-lithified nature of
the primary phosphatic sediment and fragmentation into clasts
during erosion and transport. The ubiquitous presence of other
locally derived clasts (e.g. shales, pyrite-cemented
greywacke, pyrite-rich carbonate, and pyrite that may be
concretional; Fig. 7.89d) indicates a short transport distance.
Gritstone-sandstone facies, proximal to provenance is most
enriched in P2O5 reaching 8 wt.% (Table 7.7). Systematic
sampling through the stratigraphy revealed that the relatively
distal facies shows maximum values at around 1.5 wt.% P2O5
(Fig. 7.90, drillhole 2400, 203 analyses), whereas P2O5 con-
tent in the most distal facies does not exceed 0.5 wt.%
(Fig. 7.90, drillhole 2900. 151 analyses).
Table 7.7 P2O5 content (wt%) in different lithologies from proximal
facies (Data from Bekasova and Dudkin 1981)
Lithology n P2O5 variation P2O5 average
Gritstone 45 0.11–8.2 2.4
Coarse-grained sandstone 27 0.11–2.1 0.56
Medium-grained sandstone 41 0.06–1.1 0.21
Siltstone 40 0.06–0.51 0.11
1280 A. Lepland et al.
Carbonate-fluorapatite (francolite) is the principle min-
eral phase within the phosphatic clasts (Bekasova nad
Dudkin 1981), but impurities such as pyrite, quartz, feldspar,
calcite, chlorite and organic matter are also present. Irregular
sand- to silt-sized siliciclastic laminae are observed in
some phosphatic clasts (Fig. 7.91a). This may indicate an
alternating accumulation of siliciclastic and phosphatic
components, and the origin of phosphorites as primary bed-
ded deposits. Alternatively, phosphate precipitation may
have occurred during diagenesis through replacement of
organic-rich sediment, and in such a case, the siliciclastic
laminae may represent the unaltered remnants of the original
sediment that were incorporated in diagenetic phosphates.
High abundance of pyrite in the majority of phosphatic clasts
(Fig. 7.91b, d, e, f) is consistent with phosphogenesis in a
sulphidic diagenetic environment where organic matter was
decomposed and phosphate liberated by sulphate-reducing
microorganisms, thus favouring the diagenetic precipitation
of Pilguj€arvi phosphates. Remnants of organic matter occur-
ring as fine disseminations and relatively large flaky particles
(Fig. 7.91c, d) are found in phosphatic clasts, giving them a
typical black colour.
Pyrite including framboidal and micronodular varieties is
the main impurity reaching up to 25 % in some phosphatic
clasts. Other clasts contain distinctly less pyrite in their outer
rims compared to the internal parts (Fig. 7.91b). Such deple-
tion in rims may reflect oxidative alteration of pyrite during
transport and/or within the top of the sediment column
where oxidants were available. Peculiar pyrite rings ranging
in diameter from 10 to 50 mm are present in many phosphatic
clasts (Fig. 7.91e, f). Although such pyrite occurrences
appear as rings in two-dimensional view, they most likely
represent coatings on unknown spherical substrate. This
unknown substrate has been replaced by phosphate in larger
rings (Fig. 7.91e) whereas some of the smaller rings contain
quartz inside (Fig. 7.91f). Carbonaceous residues within
some of the rings hint that organic material or microorganisms
may have been the original substrate for pyrite coatings.
Pyrite rings are not restricted to the phosphatic clasts, and
are also common in pyrite-cemented greywacke, chert and
carbonate clasts.
Zaonega Formation, Onega Palaeobasin,Karelia, Russia
Elevated phosphorus concentrations (up to 4.8 wt.% P2O5)
in sediments of the Zaonega Formation, Onega Palaeobasin,
Karelia (Fig. 7.92a) have been reported in earlier
publications (e.g. Golubev et al. 1984), but no dedicated
studies on phosphogenesis have been undertaken so far.
The minimum age of c. 1980 Ma of the Zaonega Formation
is constrained by several whole-rock and mineral Sm-Nd
and Pb-Pb isochrons obtained from the gabbro body in the
overlying volcanic succession of the Suisari Formation
(Puchtel et al. 1998, 1999). Considering that the Zaonega
Formation postdates the Lomagundi-Jatuli carbon isotope
excursion, the termination of which in Fennoscandia was
dated at 2060 Ma (Karhu 2005; Melezhik et al. 2007), the
accumulation of the Zaonega sediments can be constrained
to the 1980–2060 Ma time interval. The 1,500-m-thick
Zaonega succession (for details see Chap. 4.3) of siliciclastic,
carbonate and siliceous sedimentary rocks, and mafic tuff,
interlayered and intersected by mafic lavas and sills
(Galdobina 1987), is exceptionally rich in organic carbon
(Filippov 1994), and represents one of the earliest geological
manifestations of significant petroleum generation in Earth
history (Melezhik et al. 1999, 2009). Carbonaceous matter
occurs in rocks of the Zaonega Formation as autochtonous
kerogen residues and allochtonous (migrated) pyrobitumen.
Several intervals are pervasively impregnated with
pyrobitumen resulting in obliteration of sedimentary layering,
and in places leads to a completely massive appearance of the
host sedimentary rock. Numerous veins consisting largely of
pyrobitumen or containing also quartz and carbonate can be
observed throughout the Zaonega Formation as well as in
overlying sedimentary and volcanic rocks (Melezhik et al.
1999, 2009).
Intervals rich in phosphorus are common in the upper part
of the Zaonega Formation, which comprises a succession of
Corg- and sulphide-rich dolostone, chert, greywacke and
mudstones (Fig. 7.92b). Such intervals have been identified
in the FAR DEEP Cores 12A and 13A, and recently in
outcrops in the vicinity of the Hole 13A drilling site near
Shunga village. Phosphates occur as in-place concretionary
precipitates and cements in the Corg-rich, fine-grained
siliciclastic sediments and dolostones (Fig. 7.93a–c), and
as phosphatic clasts in gritstones and coarse-grained
sandstones (Fig. 7.93d–f). These phosphatic clasts are
interpreted to represent eroded and redeposited, phosphate-
cemented sediments and concretions.
Microcrystalline carbonate-fluorapatite (francolite) is
the main phosphate phase in phosphatic cements and
concretions and is typically intergrown with variable
amounts of calcite and carbonaceous material and minor
pyrite. The in-place phosphate precipitates are concentrated
in Corg-rich interlayers in bedded sediments (Fig. 7.93a). The
typical co-occurrence and intergrowth of phosphate
precipitates with calcite indicates a co-formation of these
phases that was likely triggered by diagenetic decomposition
of organic matter. Effective diagenetic alteration of organic
matter in the Zaonega Formation is likewise consistent with
the widespread occurrence of carbonate concretions with a
strongly 13C-depleted isotopic signature (Melezhik et al.
1999), which are particularly common in phosphate-rich
intervals.
7 7.7 The Earliest Phosphorites: Radical Change in the Phosphorus Cycle During the Palaeoproterozoic 1281
Both in-place phosphatic concretions/cements as well as
redeposited phosphorites in the form of phosphatic clasts in
coarse-grained sediments have been affected by secondary
alteration. The formation of phosphatic layers and lenses in
close association with phosphatic clasts of a clearly earlier
generation (Fig. 7.93d) may point to phosphate
remobilisation from clasts, and adjacent reprecipitation.
Considering that the phosphatic clasts originally formed in
the subsurface, likely in suboxic-anoxic diagenetic environ-
ment, their relocation to the sediment surface (possibly oxic)
due to erosion and redeposition may have caused chemical
instability and dissolution of some metastable phases leading
to locally elevated dissolved phosphate. Many phosphatic
particles have apparently been altered by late diagenetic and/
or hydrothermal fluids. These fluids caused partial dissolu-
tion of the earlier stage, impure phosphatic particles and
reprecipitation of clean, impurity-free apatite around the
clasts, and in vein systems cross-cutting both the clasts and
the host sediments (Fig. 7.93f). Minor phosphate phases
such as monazite, xenotime and autunite occur in veins
together with impurity-free apatite, but are not found in
impure phosphatic clasts. It is likely that REE, Y and U
required for crystallization of these phases were remobilised
from the early generation phosphate precipitates that had
a high trace-element content. However, some of the trace
elements may have been initially carried by the ubiquitous
organic matter, and the adsorbed elements may have been
remobilised through the activity of percolating fluids during
the alteration processes.
7.7.4 Significance of Phosphoritesin the Geologic Record and Implicationsof the FAR DEEP Material
The timing of the first appearance of globally significant
phosphorites in the Palaeoproterozoic rock record (Melezhik
and Fallick 1996; Bekker et al. 2003; Melezhik et al. 2005;
Papineau 2010) just after the rise of atmospheric oxygen
points to a genetic link between the phosphogenic episode
and the establishment of an aerobic Earth system. Several
factors, such as weathering rates, ocean circulation, supply
of nutrients, primary productivity, burial of organic matter,
and diagenetic mineralisation of organic matter influence
phosphogenesis. It appears that the first phosphorites coin-
cide in time (c. 2000 Ma) with the abundant formation
of 13C-depleted, diagenetic carbonate concretions in
siliciclastic rocks. The appearance of carbonate concretions
was interpreted to track the onset of effective recycling of
organic matter in the sedimentary column as the biospheric
response to increased oxygen concentrations (Fallick et al.
2008). However, the relative importance of individual
factors triggering the earliest phosphogenic episode remains
poorly understood and awaits further studies.
The ability of sedimentary phosphates to incorporate and
concentrate a variety of geologically important elements
makes them a valuable palaeoenvironmental archive. How-
ever, the potential of this unique archive in environmental
interpretations using trace-element (REE, Th, U) signatures
and isotope (O, C, S, Sr) ratios of phosphates have yet to be
systematically explored. The Palaeoproterozoic Fennoscandian
rocks including the FAR-DEEP material have experienced
metamorphic alteration typically under greenschist facies
conditions that may have blurred the primary geochemical
signatures. The reliable interpretation of the formation
conditions of phosphates and of the palaeoenvironment thus
depends on the ability to recognise the secondary overprints
and distinguish them from the primary characteristics.
The preliminary work undertaken on the FAR-DEEP
material indicates that different generations of co-occurring
phosphates can be petrographically distinguished. Detailed
geochemical and isotopic studies of co-occurring early and
late phosphates may help to assess the significance of meta-
morphic resetting, and thus allow the establishment of
criteria for recognising primary signatures. Geochemical
comparison of sedimentary phosphates from different basins
may likewise help in evaluating the significance of local
versus regional/global signals stored in phosphates. Timing
of Palaeoproterozoic phosphogenesis is generally poorly
established, but geochronologic studies of apatite, and par-
ticularly of monazite and xenotime (G€opel et al. 1994;
Barfod et al. 2003), common minerals in Fennoscandian
phosphorites, will likely help to better constrain the age of
the episode.
The typical association of sedimentary phosphates with
organic matter suggests that phosphates may contain and
uniquely preserve various biosignatures including
microfossils (Xiao and Knoll 1999) and biomarkers. More-
over, the precipitation of phosphates may be microbially
mediated involving sulphide-oxidising bacteria as suggested
for modern and Neoproterozoic phosphorites (Schulz and
Schulz 2005; Bailey et al. 2007). As the activity of
sulphide-oxidising bacteria requires fluctuating redox
conditions and periodic contact with oxic/suboxic bottom
waters (Schulz and Schulz 2005), a link between Palaeopro-
terozoic phosphorites and sulphide-oxidising bacteria, if
proven, may allow to assess the redox state of ancient sea-
floor. Considering that the Palaeoproterozoic phosphogenic
episode coincides with significant changes in the history of
life (see Chap. 7.8.3) including the oldest preserved
cyanobacterial fossils (Javaux and Benzerara 2009) and the
proposed advent of multicellular life (El Albani et al. 2010),
1282 A. Lepland et al.
7 7.7 The Earliest Phosphorites: Radical Change in the Phosphorus Cycle During the Palaeoproterozoic 1283
Fig. 7.86 (a) Locations of the Imandra/Varzuga Greenstone Belt
(IVGB) and Pechenga Greenstone Belt (PGB) in the Kola Peninsula,
Russia. (b) Geological position of the Il’mozero Sedimentary Forma-
tion in the IVGB, and location of the logged Varzuga section. (c)
Simplified lithological column through lower and middle parts of the
Il’mozero Sedimentary Formation with a P2O5 profile; based on data
from Melezhik and Predovsky (1982) and Melezhik (1992).
(d) Turbiditic siltstone with small-scale erosional channels filled with
greywacke that rapidly grades into siltstone. (e) Pancake-shape, zoned
concretions containing abundant apatite in dark grey and black rims
(see also Fig. 7.87a, b). (f, g) Sub-spherical phosphate-bearing
concretions retaining sedimentary layering with differential compac-
tion in the order of 2–3 times less with respect to layering of the host
greywacke (Photographs by Victor Melezhik)
1284 A. Lepland et al.
Fig. 7.87 (a) Thin-section scan of a zoned concretion from the
Il’mozero Sedimentary Formation, containing abundant apatite in the
outer, dark rim. (b) Back-scattered electron image from the phosphate-
rich outer rim of the concretion shown in (a). Abundant fine-grained
apatite (Ap) co-occurs with quartz (Qtz) and chlorite (Chl) and minor
K-feldspar (Kfs) and monazite (Mnz) (Images by Aivo Lepland)
7 7.7 The Earliest Phosphorites: Radical Change in the Phosphorus Cycle During the Palaeoproterozoic 1285
Fig. 7.88 (a) Outline of the Pechenga Greenstone Belt (for orientation
see Figs. 2.3, and 4.15), and the position of the Pilguj€arvi Sedimentary
Formation (black stripe in the middle); red arrows indicate position of
the longitudinal section shown in (b). (b) Drilling-derived longitudinal
and cross-sectional profiles through the Pilguj€arvi Sedimentary Forma-
tion, emphasising the submarine fan facies containing redeposited
phosphorite clasts (Modified from Melezhik et al. 1998)
1286 A. Lepland et al.
Fig. 7.89 Sedimentological features of phosphate-rich lithofacies.
(a) Rhythmically bedded greywacke-shale representing deep-water shelf
turbiditic facies (b) Cross-section view of rhythmically-bedded
greywacke-siltstone with several sandstone and gritstone beds (bright)
composed of clastic pyrite and phosphorite in greywacke matrix; whiterectangles show close-up image of the gritstone bed containing redepo-
sited phosphatic clasts (black clasts). (c) Polished slab of gritstone com-
posed of clasts of phosphorite (black) and sulphides in greywacke matrix
7 7.7 The Earliest Phosphorites: Radical Change in the Phosphorus Cycle During the Palaeoproterozoic 1287
Fig. 7.89 (continued) (d) Coarse-grained sandstone containing
rounded clasts of phosphorites (black). (e) Phosphorite-bearing
gritstone grading into sandstone with large fragments of siltstone
(grey) and mudstone (black). (f) Core demonstrating a gritstone
with numerous oversized clasts of greywacke, shale, quartz and phos-
phorite (black) (Photographs by Victor Melezhik)
1288 A. Lepland et al.
Fig. 7.90 Lithological sections demonstrating distal (drillhole 2900) and proximal (drillhole 2400) facies of the Pilguj€arvi Sedimentary
Formation with P2O5 profiles based on unpublished data of Victor Melezhik
7 7.7 The Earliest Phosphorites: Radical Change in the Phosphorus Cycle During the Palaeoproterozoic 1289
Fig. 7.91 Back-scattered electron images of phosphatic clasts from
the Pilguj€arvi Sedimentary Formation (a) Phosphatic clast (light gray)containing sand and silt-size detrital quartz grains (dark grey) that inplaces form irregular laminae. (b) Rounded phosphatic clast with
abundant authigenic pyrite (brighter phases within the clast) enveloped
by thin, discontinuous rim of secondary quartz; note that the sulphide
abundance decreases towards the edge of the clast. (c, d) Organic
matter (black) occurs in phosphatic clasts as thin stylolite-type layers
(c), relatively large flaky particles (d), and fine disseminations. (e, f)
Close-ups from phosphatic clasts reveal that authigenic pyrite (bright
phases) commonly forms rings (likely spheres in three-dimensional
space) with the diameter of up to 50 mm. Such pyrite rings contain in
the middle either the matrix phosphate (e), or quartz, particularly in
case of smaller rings (f) (Images by Aivo Lepland)
1290 A. Lepland et al.
Fig. 7.92 (a) Simplified geological map of the Palaeoproterozoic
Onega Basin showing the distribution of rocks of the Zaonega Forma-
tion, and positions of FAR-DEEP drill holes; (modified from Koistinen
et al. (2001)). (b) Simplified stratigraphic column of the Zaonega
Formation recorded in FAR-DEEP Cores 12A, 12B and 13A (compila-
tion by Alenka Crne and Aivo Lepland). The correlation between
drillholes is based on lithological and geochemical indicators, and
data from all three holes are used for the combined log (13A:
0–76.63 m, 12A: 76.63–161.9 m, 12B: 161.9–571.59 m).
Phosphorous-rich intervals occur in the upper part of stratigraphy
comprising dolostone, chert, greywacke and mudstone
7 7.7 The Earliest Phosphorites: Radical Change in the Phosphorus Cycle During the Palaeoproterozoic 1291
Fig. 7.93 Sedimetological features and mineral associations of
phosphorites in the Zaonega Formation. Red rectangles indicate areas
from which close-up images are provided (a) Thin section image of
bedded dolostone containing thin carbonaceous and phosphate-rich
layers (black layers). (b, c) Back-scattered electron images from the
phosphate-rich layer showing the association and intergrowing nature
of apatite (white) with calcite (light grey) and disseminated particles of
residual carbonaceous matter (black). Apatite and calcite shown on (c)
are both diagenetic phases presumably replacing a very organic-rich
original sediment in such layers. Apatite is also found in adjacent
dolostone layers as cement between rhombic dolomite (dark grey)crystals (b)
1292 A. Lepland et al.
Fig. 7.93 (continued) (d) Back-scattered electron (BSE) image of a
sediment containing beds of phosphatic lenses, layers and particles, (greyunits at the base and at the top), interbedded with finely laminated,
Corg-rich mudstone (dark grey-black unit in the middle). The differencein the backscatter response of individual phosphatic particles relates to
compositional heterogeneities and variable composition of impurities in
particles. The irregular surface at the top of the lower phosphatic bed is
erosional and draped by overlying mudstone. (e) Thin section image of a
gritstone-sandstone showing the concentration of phosphatic particles
(black) within layers as well as their random distribution throughout
sediment mass. (f) BSE image of a rounded impure phosphatic particle
and a later generation impurity-free apatite (white) occurring as a cement
between dolomite crystals and veinlets cutting through the phosphatic
particle (Images by Aivo Lepland)
7 7.7 The Earliest Phosphorites: Radical Change in the Phosphorus Cycle During the Palaeoproterozoic 1293
the biosignatures stored in phosphates as organic remnants
and geochemical tracers may provide useful information for
tracking biospheric evolution.
The rock record of the Fennoscandian Shield with
phosphate-rich sedimentary intervals in several Palaeopro-
terozoic supracrustal successions thus holds great promise
for assessing the evolution of life, ancient environmental
conditions and causes for the oldest-known phosphogenic
episode. Detailed petrographic, geochemical and isotopic
studies of sedimentary phosphates and the host sediments
in the FAR-DEEP cores, integrated with data from other
Palaeoproterozoic phosphate-rich successions, have the
potential to provide crucial information about the planetary
evolution during the beginning of aerobic Earth.
References
Ahmad T, Dragusanu C, Tanaka T (2008) Provenance of Proterozoic
basal Aravalli mafic volcanic rocks from Rajasthan, Northwestern
India: Nd isotopes evidence for enriched mantle reservoirs. Precam-
brian Res 162:150–159
Aik€as O (1980) Uraniferous phosphorite and apatite-bearing gneisses in
the Proterozoic of Finland. In: Ferguson J, Goleby AB (eds) Ura-
nium in the Pine Creek Geosyncline: proceedings of the Interna-
tional Uranium Symposium on the Pine Creek Geosyncline.
International Atomic Energy Agency, Vienna, pp 675–681
Aik€as O (1981) Proterozoic phosphorites in Finland. IGCP 156 News-
letter 9:21–27
Aik€as O (1989) Phosphate resources in early Proterozoic suprecrustal
rocks, Finland, with reference to the Baltic Shield. In: Notholt AJG,
Sheldon RP, Davidson DF (eds) Phosphate deposits of the world,
vol 2, Phosphate rock resources. Cambridge University Press,
Cambridge, pp 429–436
Akhmedov AM (1973) Phosphorus abundances in metasedimentary
rocks of the Pechenga complex. In: Predovsky AA (ed) Problems
of investigation and exploitation of natural resources of the North.
Kola Science Centre, Apatity, pp 66–74 (in Russian)
Akhmedov AM, Krupenik VA (1990) Turbidity conditions of sedimen-
tation and pyrite formation in the Early Proterozoic Pechenga Basin.
Soviet Geol 11:51–60 (in Russian)
Arning ET, L€uckge A, Breuer C, Gussone N, Birgel D, Peckmann J
(2009) Genesis of phosphorite crusts of Peru. Marine Geol
262:68–81
Aspler LB, Chiarenzelli JR (1998) Two Neoarchean supercontinents?
Evidence from the Paleoproterozoic. Sediment Geol 120:75–104
Bailey JV, Joye SB, Kalanetra KM, Flood BE, Corsetti FA (2007)
Evidence of giant sulphur bacteria in Neoproterozoic phosphorites.
Nature 445:198–201
Banerjee DM (1971) Precambrian stromatolitic phosphorites of
Udaipur, Rajasthan, India. Geol Soc Am Bull 82:2319–2330
Barfod GH, Otero O, Albarede F (2003) Phosphate Lu-Hf geochronol-
ogy. Chem Geol 200:241–253
Barley ME, Bekker A, Krapez B (2005) Late Archean to early
Paleoproterozoic global tectonics, environmental change and the
rise of atmospheric oxygen. Earth Planet Sci Lett 238:156–171
Baturin GN (1982) Phosphorites on the sea floor; origin, composition
and distribution, vol 33, Developments in sedimentology. Elsevier,
Amsterdam, p 343
Bekasova NB (1985) Pechenga palaeogeography in early Pilguj€arvitime of the early Proterozoic. Lithol Miner Resour 1:127–137 (in
Russian)
Bekasova NB, Dudkin OB (1981) Composition and nature of early
Precambrian Pechenga concretionary phosphorites (Kola Penin-
sula). Lithol Miner Resour 6:107–113 (in Russian)
Bekker A, Karhu JA, Eriksson KA, Kaufman AJ (2003) Chemostra-
tigraphy of Paleoproterozoic carbonate successions of theWyoming
Craton: tectonic forcing of biogeochemical change? Precambrian
Res 120:279–325
Bekker A, Holland HD, Wang PL, Rumble D, Stein HJ, Hannah JL,
Coetzee LL, Beukes NJ (2004) Dating the rise of atmospheric
oxygen. Nature 427:117–120
Bentor YK (1980) Phosphorites; the unsolved problems. Spec Publ Soc
Econ Paleontol Mineral 29:3–18
Berner RA (1973) Phosphate removal from sea water by adsorption on
volcanogenic ferric oxides. Earth Planet Sci Lett 18:77–86
Berner RA (1993) Weathering and its effect on atmospheric CO2 over
Phanerozoic time. Chem Geol 107:373–374
Berner RA, Ruttenberg KC, Ingall ED, Rao JL (1993) The nature of
phosphorus burial in modern marine sediment. In: Wollast R,
Mackenzie FT, Chou L (eds) Interactions of C, N, P and S biochem-
ical cycles and global change, vol 4, NATO ASI series, I. Springer,
Berlin, pp 365–378
Blake RE, Chang SJ, Lepland A (2010) Phosphate oxygen isotopic
evidence for a temperate and biologically active Archaean ocean.
Nature 464:1029–1032
Bleeker W (2003) The late Archean record: a puzzle in ca. 35 pieces.
Lithos 71:99–134
Bremner JM, Rogers J (1990) Phosphorite deposits on the Namibian
continental shelf. In: Burnett WC, Riggs SR (eds) Phosphate
deposits of the world; Neogene to modern phosphorites. Cambridge
University Press, New York, pp 143–152
Burnett WC (1977) Geochemistry and origin of phosphorite deposits
from off Peru and Chile. Geol Soc Am Bull 88:813–823
Choudhary AK, Gopalan K, Sastry CA (1984) Present status of the
geochronology of the Precambrian rocks of Rajasthan.
Tectonophysics 105:131–140
Choudhuri R (1989) Proterozoic phosphorites around Udaipur,
Rajasthan, India. In: Notholt AJG, Sheldon RP, Davidson DF
(eds) Phosphate deposits of the world, vol 2, Phosphate rock
resources. Cambridge University Press, Cambridge, pp 461–466
Compton J, Mallinson D, Glenn CR, Filippelli G, Foellmi K, Shields G,
Zanin Y (2000) Variations in the global phosphorus cycle. Spec
Publ Soc Sediment Geol 66:21–33
Cook PJ, Shergold JH (1986a) Proterozoic and Cambrian phosphorites:
an introduction. In: Cook PJ, Shergold JH (eds) Phosphate deposits
of the world, volume 1, Proterozoic and Cambrian phosphorites.
Cambridge University Press, Cambridge, pp 1–8
Cook PJ, Shergold JH (1986b) Proterozoic and Cambrian phosphorites
- nature and origin. In: Cook PJ, Shergold JH (eds) Phosphate
deposits of the world, volume 1, Proterozoic and Cambrian
phosphorites. Cambridge University Press, Cambridge, pp 369–386
Deb M, Thorpe RI (2004) Geochronological constraints in the Precam-
brian geology of Rajasthan and their metallogenic implications. In:
Deb M, Goodfellow WD (eds) Sediment-hosted lead–zinc sulphide
deposits. Narosa Publishing House, New Delhi, pp 246–263
Dymek RF, Klein C (1988) Chemistry, petrology and origin of banded
iron-formation lithologies from the 3800-Ma isua supracrustal belt,
West Greenland. Precambrian Res 39:247–302
El Albani A, Bengtson S, Canfield DE, Bekker A, Macchiarelli R,
Mazurier A, Hammarlund EU, Boulvais P, Dupuy J-J, Fontaine C,
Fuersich FT, Gauthier-Lafaye F, Janvier P, Javaux E, Ossa FO,
Pierson-Wickmann A-C, Riboulleau A, Sardini P, Vachard D,
1294 A. Lepland et al.
Whitehouse M, Meunier A (2010) Large colonial organisms with
coordinated growth in oxygenated environments 2.1 Gyr ago.
Nature 466:100–104
Ernst R, Bleeker W (2010) Large igneous provinces (LIPs), giant dyke
swarms, and mantle plumes: significance for breakup events within
Canada and adjacent regions from 2.5 Ga to the present. Can J Earth
Sci 47:695–739
Ewers WE, Morris RC (1981) Studies of the Dales Gorge Member of
the Brockman Iron Formation, Western Australia. Econ Geol
76:1929–1953
Fallick AE, Melezhik VA, Simonson BM (2008) The ancient anoxic
biosphere was not as we know it. In: Dobretsov N, Kolchanov N,
Rozanov A, Zavarzin G (eds) Biosphere origin and evolution.
Springer, New York, pp 169–188
Filippov MM (1994) Current views on the organic precursor
of shungite rocks. In: Filippov MM (ed) The organic matter of
Karelian Shungite Rocks: genesis, evolution and the methods of
study. Karelian Science Centre, Petrozavodsk, pp 16–24 (in
Russian)
F€ollmi KB (1995) 160 my record of marine sedimentary phosphorus
burial: coupling of climate and continental weathering under green-
house and icehouse conditions. Geology 23:859–862
F€ollmi KB (1996) The phosphorus cycle, phosphogenesis and marine
phosphate-rich deposits. Earth Sci Rev 40:55–124
F€ollmi KB, Garrison RE (1991) Phosphatic sediments, ordinary or
extraordinary deposits? The example of the Miocene Monterey
Formation (California). In: M€uller DW, McKenzie JA, Weissert H
(eds) Controversies in modern geology. Academic, London, pp
55–84
Froelich PN, Bender ML, Luedtke NA, Heath GR, DeVries T (1982)
The marine phosphorus cycle. Am J Sci 282:474–511
Galdobina LP (1987) Lyudikovian horizon. In: Sokolov VA (ed) The
geology of Karelia. Nauka, Leningrad, pp 59–67 (in Russian)
Geh€or S (1994) REE distribution in the phosphorite bands within the
Paleoproterozoic Tuomivaara and Pahtavaara iron-formations, cen-
tral and northern Finland. Geol Surv Finl Spec Paper 19:71–83
Glenn CR, F€ollmi KB, Riggs SR, Baturin GN, Grimm KA, Trappe J,
Abed AM, Galliolivier C, Garrison RE, Ilyin AV, Jehl C, Rohrlich
V, Sadaqah RMY, Schidlowski M, Sheldon RE, Siegmund H (1994)
Phosphorus and phosphorites: sedimentology and environments of
formation. Eclogae Geol Helv 87:747–788
Godderis Y, Donnadieu Y, Dessert C, Dupre B, Fluteau F, Francois
LM, Meert J, Nedelec A, Ramstein G (2007) Coupled modeling
of global carbon cycle and climate in the Neoproterozoic: links
between Rodinia breakup and major glaciations. C R Geosci 339:
212–222
Golubev AI, Akhmedov AM, Galdobina LP (1984) Geochemistry of
lower Proterozoic black shale complexes of the Kola-Karelian
Region. Science (Nauka), Leningrad, p 193 (in Russian)
G€opel C, Manhes G, Allegre CJ (1994) U-Pb systematics of phosphates
from equilibrated ordinary chondrites. Earth Planet Sci Lett
121:153–171
Gorshkov AI, Baturin GN, Bogdanova OY, Magazina LO (2000)
Sulfides in phosphorites of Africain Island; morphological
peculiarities and origin. Lithol Miner Resour 35:183–188
Gulbrandsen RA (1969) Physical and chemical factors in the formation
of marine apatite. Econ Geol 64:365–382
Halls HC, Davis DW, Stott GM, Ernst RE, Hamilton MA (2008) The
Paleoproterozoic Marathon Large Igneous Province: new evidence
for a 2.1 Ga long-lived mantle plume event along the southern
margin of the North American superior province. Precambrian
Res 162:327–353
Hannah JL, Stein HJ, Zimmerman A, Yang G, Markey RJ, Melezhik
VA (2006) Precise 2004 � 9 Ma Re Os age for Pechenga black
shale: comparison of sulfides and organic material. Geochim
Cosmochim Acta 70:A228
Hanski EJ (1992) Petrology of the Pechenga ferropicrites and
cogenetic, Ni-bearing gabbro-wehrlite intrusions, Kola Peninsula,
Russia. Geol Surv Finl Bull 367:1–192
Heaman LM (1997) Global mafic magmatism at 2.45 Ga: remnants of
an ancient large igneous province? Geology 25:299–302
Heron AM (1953) The geology of central Rajputana. Geol Surv India
Mem 79:1–389
Jahnke RA (1992) The phosphorus cycle. In: Butcher SS, Charlson RJ,
Orians GH, Wolfe GV (eds) Global biogeochemical cycles. Aca-
demic, San Diego, pp 301–315
Jarvis I, Burnett WC, Nathan Y, Almbaydin FSM, Attia AKM, Castro
LN, Flicoteaux R, Hilmy ME, Husain V, Qutawnah AA, Serjani A,
Zanin YN (1994) Phosphorite geochemistry: state-of-the-art and
environmental concerns. Eclogae Geol Helv 87:643–700
Javaux EJ, Benzerara K (2009) Microfossils. C R Palevol 8:605–615
Karhu JA (2005) Paleoproterozoic carbon isotope excursion. In:
Lehtinen M, Nurmi PA, R€am€o OT (eds) Precambrian geology of
Finland: key to the evolution of the Fennoscandian shield. Elsevier,
Amsterdam, pp 669–680
Kazakov A (1938) Phosphorite facies and genesis of natural
phosphates. Econ Geol Nonmetal Deposit 8:33–47
Koistinen T, Stephens MB, Bogatchev V, Nordgulen Ø, Wennestr€omM, Korhonen J (comps.) (2001) Geological map of the
Fennoscandian Shield, Scale 1:2 000 000, Espoo/Trondheim/
Upsala/Moscow
Krajewski KP, Vancappellen P, Trichet J, Kuhn O, Lucas J,
Martinalgarra A, Prevot L, Tewari VC, Gaspar L, Knight RI,
Lamboy M (1994) Biological processes and apatite formation in
sedimentary environments. Eclogae Geol Helv 87:701–745
Laajoki K (1975) Rare-earth elements in Precambrian iron formatins in
V€ayryl€ankyl€a South Puolanka area, Finland. Geol Soc Finl Bull
47:93–107
Laajoki K, Saikkonen R (1977) On the geology and geochemistry of the
Precambrian irons formations in V€ayryl€ankyl€a South Puolanka
area, Finland. Geol Surv Finl Bull 292:1–137
Lamboy M (1993) Phosphatization of calcium carbonate in
phosphorites; microstructure and importance. Sedimentology
40:53–62
Lucas J, Prevot-Lucas L (2000) Phosphorite and limestone, two inde-
pendent end-member products of the range of bio-productivity in
shallow marine environments. Spec Publ Soc Sediment Geol
66:117–125
Maheshwari A, Sial AN, Gaucher C, Bossi J, Bekker A, Ferreira VP,
Romano AW (2010) Global nature of the Paleoproterozoic
Lomagundi carbon isotope excursion: a review of occurrences in
Brazil, India, and Uruguay. Precambrian Res 182:274–299
Martens CS, Harris R (1970) Inhibition of apatite precipitation in the
marine environment by magnesium ions. Geochim Cosmochim
Acta 34:621–625
Martin AP, Condon DJ, Prave AR, Melezhik VA, Fallick A (2010)
Constraining the termination of the Lomagundi-Jatuli positive iso-
tope excursion in the Imandra-Varzuga segment (Kola Peninsula,
Russia) of the North Transfennoscandian Greenstone Belt by high-
precision ID-TIMS, AGU meeting, 13–17 Dec 2010
McArthur JM (1985) Francolite geochemistry; compositional controls
during formation, diagenesis, metamorphism and weathering.
Geochim Cosmochim Acta 49:23–35
Melezhik VA (1992) Early Proterozoic sedimentary and rock-forming
basins of the Baltic Shield. Nauka (Science), Leningrad, p 256 (in
Russian)
Melezhik VA, Fallick AE (1996) A widespread positive d13Ccarb anom-
aly at around 2.33–2.06 Ga on the Fennoscandian Shield: a para-
dox? Terra Nova 8:141–157
Melezhik VA, Predovsky AA (1978) Carbonate and silicate
concretions in Precambrian metasedimentary rocks of the Kola
Peninsula. Dokl USSR Acad Sci Geol 239(3):668–671 (in Russian)
7 7.7 The Earliest Phosphorites: Radical Change in the Phosphorus Cycle During the Palaeoproterozoic 1295
Melezhik VA, Predovsky AA (1982) Geochemistry of early Proterozoic
Lithogenesis. Nauka (Science), St. Petersburg, p 208 (in Russian)
Melezhik VA, Grinenko LN, Fallick AE (1998) 2000 Ma sulphide
concretions from the ‘Productive’ Formation of the Pechenga
Greenstone Belt, NW Russia: genetic history based on morphologi-
cal and isotopic evidence. Chem Geol 148:61–94
Melezhik VA, Fallick AE, Filippov MM, Larsen O (1999) Karelian
shungite: an indication of 2000 Ma-year-old metamorphosed oil-
shale and generation of petroleum: geology, lithology and geo-
chemistry. Earth Sci Rev 47:11–40
Melezhik VA, Fallick AE, Hanski EJ, Kump LR, Lepland A, Prave AR,
Strauss H (2005) Emergence of an aerobic biosphere during the
Archean-Proterozoic transition; challenges of future research. GSA
Today 15:4–11
Melezhik VA, Huhma H, Condon DJ, Fallick AE, Whitehouse MJ
(2007) Temporal constraints on the Paleoproterozoic Lomagundi-
Jatuli carbon isotopic event. Geology 35:655–658
Melezhik VA, Fallick AE, Filippov MM, Lepland A, Rychanchik DV,
Deines JE, Medvedev PV, Romashkin AE, Strauss H (2009) Petro-
leum surface oil seeps from Palaeoproterozoic petrified giant
oilfield. Terra Nova 21:119–126
Notholt AJG, Sheldon RP, Davidson DF (eds) (1989) Phosphate
deposit of the world, vol 2, Phosphate rock resources. Cambridge
University Press, Cambridge, p 563
O’Brien GW, Veeh HH (1980) Holocene phosphorite on the East
Australian continental margin. Nature 288:690–692
Papineau D (2010) Global biogeochemical changes at both ends of the
Proterozoic; insights from phosphorites. Astrobiology 10:165–181
Papineau D, De Gregorio BT, Cody GD, O’Neil J, Steele A, Stroud
RM, Fogel ML (2011) Young poorly crystalline graphite in the
>3.8-Gyr-old Nuvvuagittuq banded iron formation. Nat Geosci
4:376–379
Parak T (1973) Rare earths in the apatite iron ores of Lappland together
with some data about the Sr, Th and U content of these ores. Econ
Geol 68:210–221
Pecoits E, Gingras MK, Barley ME, Kappler A, Posth NR, Konhauser
KO (2009) Petrography and geochemistry of the Dales Gorge
banded iron formation; paragenetic sequence, source and
implications for palaeo-ocean chemistry. Precambrian Res
172:163–187
Planavsky NJ, Rouxel OJ, Bekker A, Lalonde SV, Konhauser KO,
Reinhard CT, Lyons TW (2010) The evolution of the marine phos-
phate reservoir. Nature 467:1088–1090
Puchtel IS, Arndt NT, Hofmann AW, Haase KM, Kr€oner A, KulikovVS, Kulikova VV, Garbe-Sch€onberg C-D, Nemchin AA (1998)
Petrology of mafic lavas within the Onega plateau, central Karelia:
evidence for the 2.0 Ga plume-related continental crustal growth in
the Baltic Shield. Contrib Mineral Petrol 130:134–153
Puchtel IS, Br€ugmann GE, Hofmann AW (1999) Precise Re-Os mineral
isochron and Pb-Nd-Os isotope systematics of a mafic-ultramafic
sill in the 2.0 Ga Onega plateau (Baltic Shield). Earth Planet Sci
Lett 170:447–461
Purohit R, Sanyal P, Roy AB, Bhattacharya SK (2010) 13C enrichment
in the Palaeoproterozoic carbonate rocks of the Aravalli Super-
group, northwest India: influence of depositional environment.
Gondwana Res 18:538–546
Rehtij€arvi P (1983) REE patterns for apatites from Proterozoic phos-
phatic metasediments, Finland. Bull Geol Soc Finl 55:77–82
Rehtij€arvi P, Aik€as O, M€akel€a M (1979) A middle Precambrian ura-
nium- and apatite-bearing horizon associated with the Vihanti zinc
ore deposit, western Finland. Econ Geol 74:1102–1117
Roy AB, Paliwal BS (1981) Evolution of lower Proterozoic epiconti-
nental deposits: stromatolite-bearing Aravalli rocks of Udaipur,
Rajasthan, India. Precambrian Res 14:49–74
Ruttenberg KC, Berner RA (1993) Authigenic apatite formation and
burial in sediments from non-upwelling, continental-margin
environments. Geochim Cosmochim Acta 57:991–1007
Sarangi S, Gopalan K, Roy AB, Sreenivas B, Das Sharma S (2006)
Pb–Pb age of the carbonates of Jhamarkotra formation constraining
the age of the Aravalli Supergroup, Rajasthan. J Geol Soc India
67:442–446
Savenko AV (2010) On the physicochemical mechanism of diagenetic
phosphorite synthesis in the modern ocean. Geochem Int
48:194–201
Schulz HN, Schulz HD (2005) Large sulfur bacteria and the formation
of phosphorite. Science 307:416–418
Sheldon RP (1980) Episodicity of phosphate deposition and deep ocean
circulation: a hypothesis. SEPM Spec Publ 29:239–247
Sreenivas B, Sharma SD, Kumar B, Patil DJ, Roy AB, Srinivasan R
(2001) Positive [delta]13C excursion in carbonate and organic
fractions from the Paleoproterozoic Aravalli Supergroup, North-
western India. Precambrian Res 106:277–290
Trendall AF, Blockley JG (1970) The iron formations of the Precam-
brian Hamersley group, Western Australia. Geol Surv West Aust
Bull 119:1–366
Tribovillard N, Recourt P, Trentesaux A (2010) Bacterial calcification
as a possible trigger for francolite precipitation under sulfidic
conditions. C R Geosci 342:27–35
Vaasjoki M, Aik€as O, Rehtij€arvi P (1980) The age of mid-Proterozoic
phosphatic metasediments in Finland as indicated by radiometric
U-Pb dates. Lithos 13:257–262
Xiao S, Knoll AH (1999) Fossil preservation in the Neoproterozoic
Doushantuo phosphorite Lagerst€atte, South China. Lethaia
32:219–238
Yudin NI (1996) Pre-Riphean phosphate genesis. Lithol Miner Res
31:286–292 (in Russian)
Zhao G, Sun M, Wilde SA (2003) Correlations between the Eastern
Block of the North China Craton and the South Indian Block of the
Indian Shield: an Archaean to Palaeoproterozoic link. Precambrian
Res 122:201–233
1296 A. Lepland et al.
7.8 Traces of Life
7.8.1 Introductory Remarks
Aivo Lepland (Editor)
When and how life on Earth started is still an open question.
Biochemical fingerprints stored in the ancient rock record
indicate the presence of traces of life back to some of the
oldest sedimentary rocks on the planet. The Earth has thus
harboured life throughout most of its geologic history, and
biological processes have contributed significantly to shap-
ing the environmental conditions on the surface of the
planet. Tracking the nature of ancient life using morpholog-
ical, mineralogical, chemical and isotopic proxies in the rock
record on Earth needs, however, to surmount a number of
obstacles. Most important are the effects of post-
depositional alteration of the sedimentary host rocks due to
exposure to metamorphic temperatures and pressures and
metasomatism during the protracted time before their pres-
ent exposure. Diagenetic and metamorphic overprints may
have resulted in recrystallisation of the original mineral
assemblages and deformation of the original textural
features in the sedimentary rocks, in many cases blurring
the biologic signatures and jeopardizing the reliable inter-
pretation of the nature of the lifeform.
Reliable biodiagnostic tools are of particular importance
for reconstructing the importance of life and biological pro-
cesses during the Great Oxygenation Event (GOE), and
related global environmental changes and events during the
Neoarchaean and Palaeoproterozoic. Although significant
advance has been made over the years in fine-tuning old,
and developing new, methods for deciphering life and its
nature in ancient rocks, fundamental questions that tie in
with the GOE, such as when the oxygen-producing photo-
synthesis started and how far back in Earth history eukary-
otic life can be tracked, are still without adequate answers.
However, the influx of fresh ideas and novel methods is
continuously improving our knowledge of what life and
biological processes may have been like during the deep
time when the aerobic Earth was established. These
advances in decoding microbial imprints and biochemical
signatures, combined with the improvements in assessment
of past environmental conditions, help to uncover links
between biologic and abiologic processes and evolution of
the entire Earth system. Elegant research undertaken over
the previous decades in tracing the biologic activity in
Neoarchaean and Palaeoproterozoic has revealed a variety
of ancient habitats ranging from pitch black environments of
sub-seafloor pillow lavas to shallow water stromatolite
settings with ample sun light, and hence considerable
biological diversity.
The following four chapters portray the ancient habitats,
describe state-of-the-art biodiagnostic methods, and provide
the current knowledge of the life record stored in
Neoarchaean and Palaeoproterozoic rocks with the focus
on the Fennoscandian findings. A range of biosignatures
including stromatolites, microfossils, bioalteration textures
of volcanic glass, and molecular biomarker and isotopic
tracers are treated in these chapters, and the potential of
the FAR-DEEP material in studying the unsolved biologic
problems is highlighted.
A. Lepland (Editor)
Geological Survey of Norway, Postboks 6315 Sluppen, NO-7491
Trondheim, Norway
e-mail: [email protected]
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_8, # Springer-Verlag Berlin Heidelberg 2013
1297
7.8.2 Palaeoproterozoic Stromatolites fromthe Lomagundi-Jatuli Interval of theFennoscandian Shield
Nicola McLoughlin, Victor A. Melezhik,Alex T. Brasier, and Pavel V. Medvedev
Introduction
Stromatolites are morphologically circumscribed sedimen-
tary growth structures with a primary lamination that is, or
may be, biogenic, and they form centimetre- to decimetre-
sized domes, cones, columns and planiform surfaces made
of carbonate (modified from Hofmann 2000). The aim of this
review is to describe and comprehensively illustrate
Fennoscandian stromatolites from the interval between
2220 and 2060 Ma with respect to their depositional
environments. These stromatolites formed in the course of
the Great Oxidation Event and the major perturbation of the
global carbon cycle known as the Lomagundi-Jatuli carbon
isotopic excursion (see Chap. 1.1 and Part 8). During this
time-period biota that had flourished in anoxic conditions
were forced to adapt to a new, oxic world. Abundant
stromatolites span this major biogeochemical transition,
and together with associated surrounding sediments, may
have captured important environmental information. In this
review we first discuss stromatolite definitions and the
criteria used to assess both their biogenicity and mechanisms
of accretion, before applying these to stromatolite
morphotypes from Fennoscandia. We then describe the
range of microfabrics and laminar geometries found in the
Fennoscandian stromatolites that reflect the complex inter-
play of physical, chemical and biological processes in their
accretion. This allows comparison of Fennoscandian stro-
matolite distribution, abundance and morphology from sev-
eral contrasting depositional settings to try and distinguish
the relative importance of various environmental controls.
The macro-morphologies and microfabrics of these
stromatolites along with their carbonate geochemistries
may also provide an archive of microbial evolution and
secular changes in seawater chemistry. Hence, the current
review also provides background to Chap. 7.3 that will
further address the significance of stromatolites and deposi-
tional environments in the formation of 13C-rich carbonates
during the Lomagundi-Jatuli isotopic excursion. Lastly, we
will outline research questions which may be addressed
using stromatolites of the FAR-DEEP drillcores.
What Is a Stromatolite?
The term “Stromatolith” was coined just over a century ago
by Ernst Kalkowsky in 1908 from the Greek words stroma
meaning bed, mattress or layer, and lithos meaning stone.
The term is now widely applied to laminated sedimentary
build-ups from throughout the geological record, and our
goal here is to briefly review what this term means, so that
we can usefully apply it to the examples that we later describe
from Fennoscandia. Kalkowsky first used the term stromato-
lite to describe laminated limestones from the Lower Triassic
Buntsandstein of the Harz Mountains, Germany. Kalkowsky
attributed the formation of these structures to simple plant-
like organisms, and the term was later popularized by Pia
(1927) as a type of fossil produced by calcium carbonate
precipitation. A more recent translation and interpretation
of this definition by Krumbein (1983) stated “stromatolitesare organogenic, laminated, calcareous rock structures, the
origin of which is clearly related to microscopic life, which in
itself must not be fossilised.” Since this early work, it has
become clear that prokaryotes, both photosynthetic and non-
photosynthetic, are the dominantmat-forming organisms that
contribute to stromatolite accretion (e.g. Monty 1972; Freytet
and Verrecchia 1998). Studies of modern analogues at, for
example, Shark Bay in Western Australia (e.g. Reid et al.
2003 and Fig. 7.94a–e), the Bahamas (e.g. Reid et al. 2000),
and Solar Lake, Sinai (e.g. Krumbein et al. 1977), have
helped to establish a connection between anaerobic decom-
position in mats, especially involving sulphate reducing bac-
teria, and carbonate precipitation. They have also shown that
eukaryotes including diatoms and algae are important
components of modern microbial mat systems. Whilst stud-
ies of modern analogues, including unlithified microbial
mats, have been very informative in elucidating mechanisms
of stromatolite growth, and have shown that the
macromorphologies of stromatolites have remained broadly
unchanged through geological time (compare Figs. 7.94 and
7.95), it should be appreciated that modern systems are
imperfect analogues for the Precambrian rock record. This
is especially the case for intervals of the Palaeoproterozoic
that had high seawater carbonate supersaturation, and prior to
the evolution of eukaryotes and metazoans.
Many papers have explored the definition and meaning of
the term stromatolite (e.g. Riding 2000, and refs therein).
These centre on whether the term should be used only in a
descriptive sense to refer to laminated sedimentary
structures, or whether it should be used genetically like the
Kalkowsky (1908) definition to refer to structures that are
N. McLoughlin (*)
Department of Earth Science and Centre for Geobiology, Allegaten 41,
Bergen N-5007, Norway
1298 N. McLoughlin et al.
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_8, # Springer-Verlag Berlin Heidelberg 2013
1298
demonstrably biogenic in origin. But herein lies the problem.
There are many types of related geological structures that
display some or all of the features of stromatolites formed in
a variety of environmental settings with varying and often
debatable degrees of biological involvement. Such structures
include: botryoidal crystal fans, hot-spring travertines, ambi-
ent temperature non-marine tufas, desert varnish crusts,
laminar caliche crusts, speleothems (including stalagmites,
stalactites and cave popcorn) and some types of sediment
deformation structures such as tepees. The possibility that
abiotic processes may solely account for stromatolite growth
was reasserted by Grotzinger and Rothman (1996) who used
the Kardar Paris Zhang (KPZ) equation of interface growth
to argue that the morphologies of some stromatolites can be
modelled by abiotic processes alone. This was a significant
study that re-invigorated the debate surrounding the
biogenicity of Precambrian stromatolites. It has been further
explored by laboratory experiments involving the subaerial
deposition of colloidal particles that produced columnar,
domal and branched stromatolites (McLoughlin et al.
2008). Taken together, this work cautions that the
biogenicity of stromatolites should not be inferred on the
basis of macro-morphology alone.
An alternative purely descriptive definition of a stromat-
olite was proposed by Semikhatov et al. (1979): “an
attached, laminated, lithified sedimentary growth structure,
accretionary away from a point or limited surface of initia-tion.” Many sedimentologists favour this non-genetic defini-
tion to describe a laminated sedimentary structure with
positive relief. This has resulted in the usage of terminology
such as “abiogenic stromatolite” to stress the absence of
compelling micro-textural or morphological evidence for
biological participation that is often lacking in many
examples due to diagenetic alteration. Strictly speaking,
however, this is a corruption of the original intention but
has the advantage of stressing the difficulties of inferring
biogenicity in many stromatolites. The term stromatoloid
has also been introduced by Buick et al. (1981) to refer to
laminated sedimentary structures the biogenicity of which is
uncertain. Genetic definitions, on the other hand, have been
proposed, for example, by Awramik and Margulis (1974) to
mean an “organosedimentary structure produced by sedi-ment trapping, binding, and/or precipitation as a result of
the growth and metabolic activity of micro-organisms, prin-
cipally cyanophytes.” The evidence necessary to substanti-
ate such a biogenic definition can be difficult to obtain in the
rock record, especially in Precambrian sedimentary
successions that have experienced diagenetic and metamor-
phic recrystallisation. In this chapter the term stromatolite is
therefore used in the non-genetic sense, but for palaeoenvir-
onmental reconstructions, it is still desirable to try to eluci-
date whether stromatolites are likely to be biotic or abiotic.
Some commonly used criteria, which might help here, are
therefore now discussed.
Stromatolite Biogenicity Criteria
There have been many attempts to develop stromatolite
biogenicity criteria, and we will introduce these before
applying them in our subsequent description of stromatolites
from the Fennoscandian Shield. Unfortunately, most of the
prescribed biogenicity criteria are so exacting that the major-
ity of Precambrian stromatolites of widely regarded biogenic
origin would fail to qualify. Here we give a brief critique of
the most widely applied biogenicity criteria drawing from
the classic study of Buick et al. (1981) and supporting points
(9–11) proposed by Hofmann (2000):
1. “The structures must occur in undoubted sedimentary or
metasedimentary rocks”. A viable sedimentary environ-
ment is a necessary first condition to demonstrate the
biogenicity of a stromatolite.
2. “It must be demonstrated that the structures aresynsedimentary”. It is necessary to exclude soft sediment
deformation (e.g. Lowe 1994) and/or later structural
deformation as contributing to the resulting morphology
(e.g. chevron folds, Fig. B95 in Wacey 2009).
3. “There should be a preponderance of convex upwards
structures.” This is a useful but very qualitative crite-
rion and is neither necessary nor sufficient to demon-
strate biogenicity. For example, abiotic self-organising
structures like stalagmites and agate crusts can exhibit
convex-upwards morphologies.
4. “Laminae should thicken over the crests of flexures.”This qualitative criterion is designed to exclude abiotic,
chemical crusts that are widely believed to exhibit
laterally uniform thickness, i.e. be isopachous (e.g.
Pope and Grotzinger 2000; Bartley et al. 2000). How-
ever, freshwater Phormidium stromatolites are com-
monly isopachous (Love and Chafetz 1988) and can be
differentiated from abiological speleothem cements by
being of uniform thickness around, for example, the
entire circumference of coated tree branches, whereas
abiological cement would be thicker on the undersides
due to geopetal effects. In other words, crusts do not
need to thicken – they may just maintain their thickness.
Further discussion of the different types of stromatolite
laminar geometries is given below, including the degree
of vertical inheritance between laminae and their lateral
continuity.
5. “If the structures are laminated, the laminations should
be wavy, wrinkled and/or have several orders of curva-ture.” Again this qualitative criterion is designed to
exclude abiogenic precipitated crusts that are thought
to be more uniform, but no limits are placed on the
extent of ‘crinkliness’ or ‘curvature’, which are also
controlled by sedimentary rheology and overprinted by
the degree of diagenetic modification.
6. “Microfossils or trace fossils should be present within
the structures.” This is far too rigid a criterion as the
preservation potential of microbial remains is extremely
8 7.8 Traces of Life 1299
low, and many recent, sub-fossilized stromatolites only
contain micro-organisms in their outermost layers. This
requirement is also in contradiction to the Krumbein
definition of a stromatolite and would exclude more
than an estimated 90 % of described fossil stromatolites
(Grey et al. 1999). Furthermore, the presence of
microfossils in a stromatolite does not confirm their
active role in formation of the structure, as they may
simply have been passively entombed by the accreting
mineral precipitate.
7. “Changes in composition of the microfossil assemblages
should be accompanied by morphological changes in
the stromatolite”. This is an extension of criterion 6 and
is extremely prescriptive as only a few instances are
known where this criterion is satisfied, for example,
Awramik and Semikhatov (1978) from the Gunflint
Chert, and Seong-Joo and Golubic (1999) from the
Mesoproterozoic of China. Studies of modern
stromatolites (e.g. Freytet and Verrecchia 1998) also
suggest similar fabrics can be produced by more than
one organism, such that morphological changes may not
accompany changes in microfossil assemblage.
8. “The fossils or trace fossils must be organised in a
manner indicating trapping, binding or precipitation of
sediment by the living micro-organisms”. Again this
would be desirable but is somewhat over-optimistic.
Tufted microbial filaments, fenestrae and micropores
created by the growth and decay of now absent microbes
would be useful but are only found when diagenesis is
minimal (e.g. Turner et al. 2000a).
9. “Brecciated mat chips accumulated in depressionsbetween convexly laminated mounds”. These are occa-
sionally seen and have been highlighted in discussions
surrounding early Archaean stromatolites, for example,
from the Strelley Pool Chert (e.g. Hofmann et al. 1999).
Such textures are most convincing if the mat fragments
show laminated internal fabrics or microtextures that
can distinguish them from fragments of brecciated sea-
floor crusts.
10. “Thin, rolled-up fragments as indications of the exis-tence of coherent flexible laminae that are reasonably
interpreted as microbial mats”. These are desirable and
occasionally seen, for example, in the Palaeoproterozoic
Hamersley Group of Western Australia (Simonson
and Carney 1999), and especially in silicified horizons
(e.g. Tice and Lowe 2004), but again this criterion
requires remarkable organic preservation.
11. “Distinct compositional differences between the
laminated growth structures and their surroundingmatrix, such as carbonate stromatolites set within ter-
rigenous detritus”. This can be a helpful criterion but
there are numerous exceptions. For instance, many of
the least controversial carbonate stromatolites are found
in carbonate settings, and especially those formed by
trapping and binding can have a similar composition to
their surrounding matrix. On the other hand, in terrige-
nous clastic setting, laminated carbonates are not always
a stromatolite: many caliche crusts would meet this
criterion (see Chap. 7.9.3 and Brasier 2011).
In short, a biogenic origin is most likely for stromatolites
that exhibit complex morphologies, laminae that show lat-
eral and vertical variability (see discussion of microfabrics
below) and organic-bearing microfabrics. In addition, the
case for a biogenic origin may be strengthened if it can be
shown that changes in both the macro- and microfeatures of
the stromatolite correlate with biologically significant envi-
ronmental gradients, such as light levels.
Mechanisms of Stromatolite Accretion
Given that the palaeoenvironmental reconstructions of the
Fennoscandian stromatolites described below requires an
understanding of the stromatolite microfabrics, we here out-
line current thinking on the three processes that contribute to
stromatolite accretion, summarising the laminar architecture
and microtextures that are thought to result. All three pro-
cesses described here might be found in the Palaeopro-
terozoic of Fennoscandia. A recent more detailed review of
stromatolite fabrics can be found in Riding (2008) with a
focus on Precambrian stromatolites.
Trapping and BindingIn recent stromatolites, the trapping and binding of
suspended sediment particles is an important accretionary
mechanism – so-called “coarse grained mat” in Fig. 7.96.
This process is controlled by: the slope of the accreting
stromatolite interface; the grain size and density of the
sediment particles; the size and motility of the microbes
(e.g. Riding 2000, and references therein). This process is
countered by the movement of sediment down stromatolite
flanks under the forces of gravity and current action that are
most vigorous in high-energy settings and for stromatolites
with steep flanks or significant relief. Sediment trapping and
binding is thought to be facilitated by biofilms that have
abundant ‘sticky’ extra-cellular polymer and by microbes
with high motility (e.g. Golubic et al. 2000). In addition,
early cementation is necessary to prevent loss of this detrital
sediment from the stromatolite flanks. These processes are
thought to create non-isopachous laminae that show irregu-
lar, uneven layering with low inheritance and limited lateral
continuity. In petrographic thin-section, preservation permit-
ting, the trapped and bound sediment grains should be visi-
ble, perhaps between the decayed remains of organic
1300 N. McLoughlin et al.
laminae that may have formed during more quiescent
periods.
Trapping and binding processes are commonly associated
with microbially-induced carbonate precipitation (see
below) that together result in so-called “Fine-Grained
Crusts” of Riding (2008) (see Fig. 7.96, lower right apex).
These exhibit micritic, clotted, peloidal or filamentous
microfabrics and are especially abundant in the
Neoproterozoic (e.g. Turner et al. 2000b). Additional sup-
port for microbial activity can come from included micro-
fossil remains, such as those found in the 1.8 Ga
stromatolites of the Gunflint (Awramik and Semikhatov
1978); or in the form of micro-pores interpreted to be the
moulds of decayed microbes (e.g. Bosak et al. 2004); or
palimpsest fabrics comprising vertically aligned, decayed
microbial filaments (e.g. Buick 1992). However, the mere
presence of such microbial remains is insufficient to demon-
strate full microbial control on precipitation, as it needs to be
shown that these remains were not just passively entombed,
but rather actively shaped the accreting structure. In some
stromatolites, the Fine Grained Crusts are interleaved with
so-called crystalline or Sparry Crusts (see below) creating
alternating light and dark layers conferring a streaky appear-
ance. These are interpreted to reflect the switching between
dark, fine-grained, essentially microbial-mediated layers and
light-coloured, possibly abiotically precipitated spar layers
that are common in many conical stromatolites termed
Conophyton (e.g. Grotzinger and Knoll 1999, and references
therein).
Abiotic Chemical PrecipitationCarbonate precipitation in microbial stromatolites can be
induced by the activities of living and decaying microbes
(e.g. Riding 2000) but also results from pore water supersat-
uration generated by abiotic processes such as degassing of
CO2 and evaporation. Abiotic chemical precipitation is
responsible for the growth of abiotic laminated deposits
such as agates, botryoids, crystalline crusts and cements.
These can sometimes be recognised by the expansion and
ultimate coalescing of neighbouring crystal fans or undulose
layers comprising precipitated crusts. These are termed
“Abiogenic Sparry Crusts” by Riding (2000) and shown on
the lower left apex of Fig. 7.96. In discussions concerning
the biogenicity of stromatolites, abiotic chemical precipita-
tion is widely assumed to produce isopachous laminae, i.e.
laminae of uniform thickness (e.g. Buick 1992). The syn-
thetic stromatolite experiments reported in McLoughlin
et al. (2008), however, together with several numerical
modelling studies (e.g. Grotzinger and Rothman 1996)
have shown that abiotic processes can also produce non-
isopachous laminae. This is well known, for example, from
meteoric cements formed in the phreatic zone, where void
spaces are water-filled resulting in the precipitiation of
isopachous laminae; whereas, cements formed in the vadose
zone have dripstone, i.e. non-isopacheous geometries. Thus
the assumption that abiotic precipitation exclusively results
in isopachous stromatolite laminae does not hold.
In the Precambrian rock record, stromatolitic growth by
abiotic, surface-normal chemical precipitation is suggested
by laminae that are composed of radial, fibrous crystal fans
(e.g. Fig. 4.c in Grotzinger and Knoll 1999) or so-called
“Sparry Crusts” (Riding 2008). These are characterised by
isopachous laminae that show extreme lateral continuity and
high degrees of vertical inheritance. Such structures are
believed to form by direct carbonate precipitation on the
seafloor and are devoid of clastic carbonate (e.g. Turner
et al. 2000b; Pope et al. 2000). They are particularly abun-
dant in Archaean and Palaeoproterozoic sequences and are
believed to be associated with times of high seawater car-
bonate supersaturation, often being associated with evapo-
ritic sequences (Pope et al. 2000). They have been termed
“chemical stromatolites” by Pope et al. (2000) who argued
that they are “largely abiotic in origin”.
A related stromatolite morphotype that is especially com-
mon in the early- to mid-Proterozoic is microdigitate stro-
matolite (lower left of Fig. 7.96). These are small, digitate,
laminated columns, typically less than 5 mm wide and less
than 20 mm high, that occur in densely packed layers and are
abundant in shallow peritidal settings. Their biogenicity has
been widely discussed, and given that the laminae show
radial fibrous fabrics, strong vertical inheritance and can be
traced laterally between several columns (e.g. Hofmann and
Jackson 1987), it is argued that chemical precipitation, rather
than aggradation of grains, plays the major role in their
formation. It has been proposed by some that this precipita-
tion is entirely abiotic (e.g. Grotzinger and Knoll 1999).
Biologically Induced Precipitation with a Focus onChemo- and Phototactic GrowthThe thickening of laminae across the crests of stromatolite
domes and cones is commonly believed to result from
microbially accelerated growth in an upwards direction due
to chemo- or phototaxis. Coniform stromatolites, in particu-
lar, are taken as indicators of phototactic microbial growth
by analogy to modern tufted and peak-shaped microbial
mats (e.g. Walter et al. 1976; Batchelor et al. 2004, and
lower centre of Fig. 7.96). It is envisaged that phototactic
biofilms strive to gain more light and that chemotactic
biofilms aim to elevate themselves in the benthic boundary
layer to access more nutrients. This accelerated upwards
growth is accomplished by microbial motility towards topo-
graphic highs termed “upslope diffusion” by Jogi and
Runnegar (2005). The exact geometries of the resulting
laminae, in particular, the degree of laminar thickening in
the axial zone, are likely controlled by: photic zone
conditions; the thickness of the benthic boundary layer;
8 7.8 Traces of Life 1301
and the nature of diffusive gradient both within and above
the accreting surface. Recent laboratory investigations of
reticulate cyanobacterial mats have found that it is cell
motility and not phototaxis that controls the shape of the
mat, because upwards growth was observed even when
illuminated from below (Shepard and Sumner 2010).
In addition to laminar geometries, there are other
microfabric clues that may indicate the phototactic growth
of stromatolites. Recent laboratory experiments growing
cyanobacteria-dominated microbial mats have identified
the presence of contorted laminae in the axial zone of coni-
cal aggregates and preservation of enmeshed oxygen
bubbles as indicators of oxygenic photosynthetic mat growth
(Bosak et al. 2009). Such features are common in the central
zones of well-preserved Proterozoic conical stromatolites,
from which these workers argue that oxygenic photosynthe-
sis first appeared in the Palaeoproterozoic. In outcrop, the
observation of sinuous growth axes of 850 Ma stromatolites
from the Bitter Springs Formation has also been argued to
represent heliotropism, i.e. microbial mat growth that tracks
the changing position of the sun (Vanyo and Awramik 1985) –
this of course requires that the stromatolites accreted and
lithified at a sufficiently fast rate to record such annual
variation.
Brief Review of Transition from Neoarchaean toProterozoic Stromatolites: Temporal Context forPalaeoproterozoic Examples from Fennoscandia
To provide an evolutionary context for the forthcoming dis-
cussion of stromatolites deposited during the Lomagundi-
Jatuli interval, a brief review of broad-scale trends in
stromatolites across the Archaean to Proterozoic is presented
below. Many well-studied Archaean stromatolites are
characterised by sparry microfabrics suggesting that they
formed predominantly by chemical precipitation. Stromato-
lite sequences with abundant Sparry Crusts (defined by Rid-
ing 2008) include: the Campbellrand-Malmani platform of
South Africa (e.g. Beukes 1987, and Fig. 7.95); the Steep
Rock Group of Ontario (e.g. Wilks and Nisbet 1985); and the
Carawine dolomite ofWestern Australia (e.g. Simonson et al.
1993). Many of these stromatolites are interbedded with
abiogenic seafloor precipitates, including herringbone calcite
and botryoidal crystal fans. Many of these late Archaean
sequences also include fenestrate microbialites, net-like
masses of irregular columns or tent-shapes, composed of
dark organic layers encased in irregular calcite cements
(Sumner 1997). This microbialite morphotype appears to be
largely restricted to the late Archaean (a detailed review
presented in Hofmann 2000).
There is a broad transition from chemically-precipitated,
sparry stromatolites in the Archaean to stromatolites that
show fine-grained, more microbially-influenced micro-
fabrics in the Neoproterozoic (Riding 2008). In the
intervening Palaeoproterozoic interval, which covers the
Fennoscandian examples described below, stromatolites
with hybrid fabrics (Fig. 7.96) are most common, and this
is argued to reflect a combination of chemical precipitation
and trapping and binding in their accretion (Riding 2008).
This broad transition has been suggested to represent a
long-term decline in seawater carbonate supersaturation
(e.g. Grotzinger and Kasting 1993), with a switch in impor-
tance from chemical precipitation to microbial trapping and
binding as the dominant mode of stromatolite growth. It
appears that stromatolite microfabrics are more sensitive
than their macro-morphology for recording these secular
changes in seawater chemistry, and we refer the reader to
Riding (2008) and references therein for a more comprehen-
sive explanation.
Overview of Stromatolites from theFennoscandian Shield
All of the stromatolites described below come from the
Lomagundi-Jatuli interval, a time of major global perturba-
tion in the carbon cycle that produced a marked positive
excursion in carbonate carbon isotope ratios. The duration of
the excursion, and consequently the time interval for forma-
tion of the stromatolites, has been constrained on the
Fennoscandian Shield to ca. 2220–2060 Ma (Karhu 2005;
Melezhik et al. 2007; Karhu et al. 2008). During this time,
mature intraplate rifts affected the entire shield that was
largely covered by shallow-water epeiric seas with a series
of carbonate platforms apparently providing an ideal setting
for the stromatolites that we will describe to flourish. Global
examinations of Proterozoic stromatolites with a 500 Ma
time-interval resolution have revealed maxima in diversity
and abundance between 2200 to 1650 Ma and 1350 to
675 Ma (e.g. Awramik 1992). Studies from different
continents have found a diversity maximum in the
Palaeoproterozoic of Eurasia, China and India between
2300 and 2000 Ma, whereas in Australia and North America,
the maxima in diversity and abundance appear to occur
somewhat later, between 2000 and 1800 Ma (Semikhatov
and Raaben 1994, 1996). These types of estimates are based
upon the classification of stromatolites at the species and
genera level, a concept that has been widely debated and is
discussed in more detail below. It has been estimated by
Semikhatov and Raaben (2000), for example, that 20 % of
the stromatolite species and genera described within the
current framework may be synonyms. This is thought to
result from preservational variability, incomplete descriptions
and uncertainties regarding the relative importance of differ-
ent characteristics used to distinguish groups and forms.
1302 N. McLoughlin et al.
Notwithstanding, the long-term first order trends in stromat-
olite abundance and diversity are worthy of consideration.
In the Fennoscandian Shield, a quantitative estimate of
the diversity of stromatolite taxa and their abundance in
connection with major depositional settings was made by
Melezhik et al. (1997). Although the study is somewhat
outdated, there have been no new stromatolite discoveries
reported since, and only two new radiometric dates have
been obtained (Melezhik et al. 2007; Karhu et al. 2008);
thus the original database remains valid. The study by
Melezhik et al. (1997) showed a clear maximum in stromat-
olite abundance between 2100 and 2060 Ma, while the
diversity maximum appears between 2330 and 2060 Ma.
The histograms (Figs. 3, 4, and 5 in Melezhik et al. 1997)
illustrate the rapid rise of stromatolite diversity at 2330 Ma
with a plateau occurring between 2330 and 2060 Ma, and an
abrupt decline in diversity at c. 2060 Ma. After 1800 Ma
there is currently no robust information available from the
Fennoscandian Shield.
It is suggested that the absence of stromatolites between
2500 and 2300 Ma on the Fennoscandian Shield may be due
to the deposition of predominantly clastic sediments,
followed by the onset of the Huronian glaciation at
c. 2440 Ma. The stromatolite expansion postdates the
Huronian glacial event and seems to be related to the
major phase of intracontinental rift development and forma-
tion of several carbonate platforms (see Chap. 3.3).
Palaeogeographically this coincides with the widespread
occurrence of dolomite-producing basins and the onset of
shallow-water conditions. Although these basins span two
different palaeotectonic settings, namely shallow-water car-
bonate platforms and numerous rift-bound lakes,
stromatolites became abundant in both. The rapid decline
in stromatolite abundance at around 2060 Ma corresponds in
time to initial separation of the late Archaean superconti-
nent, formation of the Kola ocean and Svecofennian sea, and
transition to marine conditions in most of the rifts fringing
the continent (Gaal and Gorbatschev 1987; Strand and
Laajoki 1999; Lahtinen et al. 2008). This has been dated
approximately to 2100 Ma (Korsman et al. 1999; Hanski
and Huhma 2005; Daly et al. 2006) when the numerous
shallow-water carbonate-producing basins were replaced
by relatively deep-water seas with siliclastic-dominated sed-
imentation (e.g. Wanke and Melezhik 2005). One can argue
that the immediately post-glacial waters (from which abun-
dant stromatolites are known) were likely to be quite alka-
line and therefore conducive to stromatolite accretion by
either abiotic or biotically-induced precipitation. But the
interval of stromatolite development spanned several
hundred million years, so this cannot be the sole reason.
A logical explanation for the decline or absence of stroma-
tolites following the Lomagundi-Jatuli interval is that
conditions were unsuitable for micro-organism growth.
This hypothesis can only be tested through examination of
the sedimentary rocks and the stromatolites themselves.
To encourage such work an overview of the main stro-
matolite horizons from the different depositional settings
associated with this interval of greatest stromatolite abun-
dance and diversity on the Fennoscandina Shield is
presented below. For reasons given earlier, we avoid all
taxonomic terminology and have rather classified the
stromatolites broadly by morphotype, and in Table 7.8, we
summarise how each of these morphotypes satisfies the
criteria that we gave for assessing biogenicity. The
stromatolites described here belong to the Tulomozero For-
mation of the Onega carbonate platform, lake deposits of the
Kuetsj€arvi Formation, and the Kalix rimmed carbonate shelf.
A c. 2100 Ma Shallow-Water Onega CarbonatePlatform; the Tulomozero Formation
The Tulomozero Formation is one of several units compris-
ing a large, open synform exposed on the northern coast of
Lake Onega and its numerous islands and peninsulas
(Fig. 4.34). The formation is 800 m thick and has been
studied by numerous workers in outcrop and several
drillcores (e.g. Sokolov 1963, 1987; Akhmedov et al.
2004). It is subdivided into several members (Fig. 7.97)
including: red, beige and variegated, 13C-rich dolostones
with interbedded arenites; siltstones and mudstones
containing abundant desiccation cracks; dissolution-collapse
breccias; and ubiquitous remnants of evaporites. A Pb-Pb
carbonate age of 2090 � 70 Ma obtained from dolostones is
interpreted as an age of sedimentation or early diagenesis
(Ovchinnikova et al. 2007). The formation accumulated in a
shallow-water carbonate platform that was subject to fre-
quent phases of emergence and evaporation (Fig. 7.97).
Detailed discussion of sedimentological aspects and deposi-
tional settings is provided in Melezhik et al. (1999, 2000,
2005b). The carbonate succession contains a varied suite of
stromatolitic and flat-laminated dolostones, and below we
present a short description of the best-preserved stromatolite
morphotypes supported by photographic images. These have
also been extensively described by Makarikhin (1992) and
Makarikhin and Kononova (1983).
Tulomozero Formation Morphotype 1: SpacedBioherms of Branched, Columnar StromatolitesTwo vertical stromatolite cycles are recognised in Members
B and C of the middle part of the Tulomozero dolomite
succession (Fig. 7.97): there is a transition from domed
bioherms comprising markedly divergent, branched colum-
nar stromatolites of morphotype 1, to more laterally contin-
uous bioherms comprising branched columnar stromatolites
of morphotype 2 described below. The domed bioherms of
8 7.8 Traces of Life 1303
morphotype 1 in the centres of Members B and C are
composed of tightly packed, branching columns, which are
10–15 mm in diameter and 20–30 cm high. They are most
abundant in the middle of Member B (e.g. Fig. 7.98a) where
they are surrounded by red, haematite-rich, structureless
dolostones. They form cupola-like build-ups 3–5 m wide
and 1.5 m high with steeply dipping margins. Here the
synoptic relief is estimated to be at least 20 cm. In the middle
of Member C, these biostromes are up to 1.5 m high and
extend up to 120 m laterally. The environments of deposition
are interpreted to be a combination of intertidal settings in
barred lagoons and ephemeral ponds in the supratidal zone.
Tulomozero Formation Morphotype 2: LaterallyContinuous Biostromes of Branched, ColumnarStromatolitesLaterally continuous biostromes up to 1.5 m high composed
of tightly packed branching columns (Fig. 7.98b) occur in
the centre of Member B (Fig. 7.97). Columns are 5–7 cm in
diameter and 25–30 cm high, with larger columns in the
centre of the biostromes becoming smaller towards the
margins (e.g. Fig. 7.103d; see also Melezhik et al. 2000).
The columns are often elongate, which Melezhik et al.
(2000) interpreted as caused by current action (Fig. 7.98b).
The laminae are gently convex, turning downwards at the
margins, but are indistinct due to recrystallisation. The depo-
sitional environment for this morphotype is interpreted to be
intertidal in protected bights.
Tabular biostromes also occur at the top of Members B
and C, consisting of closely spaced columns that are elon-
gated in plan view and up to 0.2 m high (Figs. 7.97 and 7.98).
The columns show beta modes of branching, bumpy or
ragged margins and steeply convex lamina profiles
(Fig. 7.98c). The columns are evenly and tightly spaced
and are slightly and uniformly inclined, up to 20� from
vertical. The infilling sediment is dolomite-rich. The incli-
nation likely indicates current direction given that the
columns are also elongate in plan view with ragged, eroded
margins. The depositional environment for this morphotype
is interpreted to be a barred lagoon to shallow-water subtidal
zones or ephemeral ponds in the supratidal zone.
Thin, less than 0.2-m-high biostromes of columnar
stromatolites also occur in Member H (Fig. 7.97). The
columns are closely spaced and connected to each other by
a number of bridges (Fig. 7.98d). Branching is less common
and the margins are ornamented by cornices and peaks.
Thin, less than 0.15-m-high biostromes of columnar
stromatolites occur in Member C. Here, the columns show
gamma modes of branching with moderate angles of diver-
gence. “Daughter” columns are short and do not evolve
upwards after branching. The laminae are smooth and have
gently convex profiles. The dark laminae show a clotted
fabric. The stromatolites are made of red, haematite-rich
dolomite that also fills the interstices between columns
(Fig. 7.98e, f).
Tulomozero Formation Morphotype 3:Flat-Laminated StromatolitesThis is the most abundant morphotype in the Tulomozero
Formation. Flat-laminated stromatolites forming low-relief
biostromes have been documented throughout the dolomite
sequence (Figs. 7.97 and 7.99). The laminae are wrinkled on
the 1–2 mm scale, to wavy on the 3–5 mm scale, and are
sometimes termed “blister stromatolites” due to syngenetic
brecciation of the laminae and the abundance of fenestrae
that, together, sometimes confer an indistinct, clotted fabric
(see, for example, Fig. 7.99a from Member C). The red
dolomite is haematite-rich and dessication cracks are widely
developed. The environment of deposition is interpreted to
be drained depressions and ephemeral ponds in the upper
tidal zone of a playa lake environment (Melezhik et al.
1999).
In the middle part of Member E (Fig. 7.97), another type
of stratiform deposit is found. Here tabular biostromes com-
prising columnar stromatolites (Fig. 7.99b, c) show a transi-
tion both upwards and laterally from separate columns to
merged columns forming stratiform layers. The laminae are
distinct and show a high degree of inheritance. The environ-
ment of deposition is interpreted to be an evaporative playa
lake.
Tulomozero Formation Morphotype 4: ColumnarBranched Mini-StromatolitesColumnar, branched mini-stromatolites 3–7 mm high and up
to 4 mm in diameter are found in Member A (Fig. 7.97). The
sub-cylindrical mini-columns are vertical or slightly inclined
with round to elliptical plan outlines. In Member A, the
columns are markedly divergent. Individual laminae are
relatively thick and form a gently convex lamina profile.
The interpreted depositional environment for this
morphotype is ephemeral ponds in the intertidal zone.
Tulomozero Formation Morphotype 5:Non-branching Mini-Columnar StromatolitesThis morphotype comprises non-branched, mini-columnar
stromatolites less than a cm wide, with individual columns
that expand upwards and are at most 3 cm high
(Fig. 9.100a–c). The stromatolites are characterised by
very fine laminations comprising clotted layers separated
by clear microspar, possible hybrid crusts (Fig. 7.100c).
These are separated by pale-grey, fine-grained dolostones
with oncolites, i.e. detached, laminated sub-spherical
structures. They are found in the centre and upper part of
Member G (Fig. 7.97). The environment of deposition is
interpreted to be intertidal to subtidal zone.
1304 N. McLoughlin et al.
Tulomozero Formation Morphotype 6: Large Non-branching Columnar StromatolitesLarge single, non-branching columnar stromatolites spaced
widely apart are restricted to the lower part of Member D
(Fig. 7.97). The columns are round in plan outline, up to
1.5 m high and 0.2 m in diameter, expanding slightly
upwards (Fig. 7.101a). The laminae show gently convex
profiles and consist of alternating light and dark bands
2 mm thick, composed of grey and pale grey dolomite in
clastic quartz. The microfabrics are not preserved due to
recrystallisation. This morphotype is rather rare but a typical
feature of the base of Member D. The interpreted deposi-
tional environment is intertidal to subtidal zone, close to the
shoreline and fully exposed to wave action.
Tulomozero Formation Morphotype 7: DomedBioherms Comprising Coalesced BulbousStromatolitesClosely spaced, domed biostromes 25–30 cm high and up to
60 cm across with flat bases, comprising coalesced bulbous
stromatolites, occur in the middle part of Member G and at
the base of Member H (Fig. 7.97). The diameter of individ-
ual bulbous stromatolites is 3–5 cm and for the coalesced
stromatolites up to 15–20 cm. The laminae are thin, have a
hemispherical profile and consist of alternating 0.5–1.0-mm-
thick, red and pink dolomite. The domed biostromes have
gently dipping margins and are separated by either pale grey,
laminated, fine-grained dolostone or red, haematite-rich
dolorudites containing oncolites, i.e. laminated, detached,
sub-spherical structures. The interpreted depositional envi-
ronment is sub-tidal.
Tulomozero Formation Morphotype 8:Hemispherical StromatolitesTabular biostromes less than 0.1 m in height comprise
cumulate hemispherical stromatolites, with common
laminae and moderate synoptic relief (Fig. 7.101b). The
stromatolites frequently coalesce and occur in the middle
part of Member C and at the top of Member D (Fig. 7.97).
The interpreted depositional environment is an intertidal
zone barred lagoon or ephemeral ponds in the supratidal
zone.
A c. 2060 Ma Rift-Bound Lake System; theKuetsj€arvi Sedimentary Formation
The Kuetsj€arvi Sedimentary Formation in the Pechenga
Greenstone Belt, NW Russia (Fig. 4.15), is a c. 150-m-
thick siliciclastic-carbonate succession formed in an
intracratonic rift setting. The formation is sandwiched
between two, 2-km-thick, subaerially erupted volcanic
units, and formed prior to 2058 Ma (Melezhik et al. 2007).
The lowermost part of the formation represents a distal
braidplain and braid delta siltstones, with sandstones
overstepped by lacustrine, variegated to mottled, fine-
grained siliciclastic rocks, ‘red beds’, dolostones containing
stromatolite sheets, and hydrothermal travertine deposits
(Melezhik and Fallick 2001; see Chap. 7.9.4). Desiccation
features are abundant, including tepees, caliche, surfical
silicified crusts, dissolution cavities and probable
pseudomorphed evaporites, suggesting repeated basin emer-
gence and apparent evaporitic conditions (Melezhik and
Fallick 2005; Melezhik et al. 2004). Although the formation
mostly accumulated in terrestrial environments, a significant
drop in 87Sr/86Sr recorded in the uppermost dolostone unit
indicates a short-term invasion of marine water to the
Kuetsj€arvi rift-bound lake just prior to the voluminous erup-
tion of the overlying volcanic rocks (Melezhik et al. 2005a).
Only the units (marked by A, B and C) in the upper part of
the succession (Fig. 7.102) contain carbonates of argued
microbial origin. The succession includes mainly non-
columnar stromatolites, although one occurrence of a colum-
nar stromatolite has been reported (Lybtsov 1979). The
thickness of the stromatolitic and flat laminated units ranges
from a few centimetres to 1 m. They are commonly
interbedded with travertine crusts and redeposited, impure
dolarenites and dolorudites.
Kuetsj€arvi Sedimentary Formation Morphotype 1:Stratiform Laminites (Non-columnar Stromatolite)The stratiform laminites are the most common morphotype
of argued microbial origin in dolostone-dominated units A
and B that were deposited in a shallow-water lacustrine
environment (Fig. 7.102). Unit A is a 48.3 m thick
dolostone-dominated succession consisting of interbedded
white, micritic dolostones and silty to sandy, allochemical
dolostones and abundant travertine crusts (Melezhik and
Fallick 2001, 2005). Stratiform laminites are rare and
occur as centimetre-thick sheets of an unknown lateral
extent. The sheets are irregularly distributed through the
stratigraphy (Fig. 7.102). Layering in the stratiform
laminites is expressed by alternation of 0.2–1.0-mm-thick
dolomicrite laminae and 0.1–0.3-mm-thick laminae
consisting of finely crystalline quartz and sparry dolomite
(Fig. 7.103a). These can be interpreted as hybrid crusts,
reflecting stromatolite accretion by both trapping and bind-
ing and (biotic?) chemical precipitation. Flat-laminations
and undulatory lamination with several orders of curvature
are common. Many laminae thicken over the crests of
flexures. Variably developed fenestrae and intensive syn-
sedimentary brecciation and buckling are widespread
(Fig. 7.103b–c). In places, stratiform laminites are discor-
dantly capped by a travertine crust. Some stratiform
laminites contain small crystals of authigenic albite and
8 7.8 Traces of Life 1305
1–4 mm spherical inclusions of sparry dolomite resembling
pseudomorphed evaporites.
Unit B is only 9 m thick (Fig. 7.102) and consists of
interbedded stratiform laminites, micritic, and sandy, sparry,
allochemical dolostones with abundant travertine crusts and
small-scale travertine mounds. All lithologies have either a
pink or variegated colour with a mottled appearance
(Fig. 7.103d, e). Similar to Unit A, the stratiform dolomitic
laminites form sheets, but these are developed in much
greater abundances and range in thickness from 10 cm up
to 1 m. The stratiform laminites display mainly undulatory
laminae (Fig. 7.103d), although some exhibit pseudo-
columnar or weakly-domed patterns (Fig. 7.103e). The lami-
nation is expressed by alternation of 0.2–0.5-mm-thick,
irregular laminae of micritic dolomite and thicker layers of
sparry dolomite containing carbonate intraclasts and quartz
detritus; chert- and dolospar-filled fenestrae are also com-
mon (Fig. 7.103f). In many cases, the lamination is highly
disrupted by desiccation features and formation of micro-
nodules of probable evaporites pseudomorphed by chert and
dolomite. Consequently, the microbial dolostones display
micro-brecciation and are the source of clasts in the
interbedded allochemical dolostones. The stromatolitic
dolostones also contain tepee structures, and stromatolitic
beds located above the tepees are affected by the develop-
ment of pedogenic dolocrete and silicrete containing abun-
dant silica sinters (Melezhik et al. 2004).
Kuetsj€arvi Sedimentary Formation Morphotype 2:Club-Like and Subspherical StromatoliteClub-like and subspherical stromatolites are rare and have
been documented only in the uppermost Unit C, accumula-
tion of which was influenced by incursion of seawater
(Melezhik et al. 2005a), though there is one exception at a
depth of c. 297 m (Fig. 7.102). This type of stromatolite
appears as solitary clubs and subspheres. In places, clubs and
undulose subspherical stromatolites form 0.5-m-high,
domed bioherms of unknown geometry and lateral extent.
Both solitary clubs and bioherms are developed within a
single unit and have accreted on and are capped by non-
columnar, flat-laminated stromatolites described above.
The solitary club-like stromatolites are up to 0.5 m in
height. The clubs’ base is only 5–10 cm in diameter, whereas
their head reaches 25 cm (Fig. 7.103g). The solitary
subspherical stromatolites are 10 cm in height
(Fig. 7.103h). Both types comprise steeply convex,
lensoidal, undulatory, 1–2-mm-thick, dolomicritic laminae
alternating with thicker laminae composed of detrital dolo-
mite and quartz, possibly indicating sediment trapping and
binding (Fig. 7.103g). Both micritic and detrital laminae are
highly disrupted. Fenestrae filled with sparry dolomite and
chert are widespread.
The undulose, subspherical stromatolites comprising
bioherms are 20–30 cm in height, and comprise alternating
lamina similar to ones described for solitary forms. These
morphotypes commonly nucleate as individual spheres that
are later overgrown by common laminae and continue
accreting as undulating stromatolites. The inter-sphere
space is filled with clasts of micritic dolomite, rounded
quartz grains and debris of flat-laminated microbial dolo-
mite, all poorly sorted and showing no stratification. The
latter is expressed by several 0.5–1-cm-thick, wavy lenses of
quartz sandstones.
Exotic Structures of Argued Microbial OriginThere are rare examples of build-ups resembling solitary
spheroidal stromatolites that are extremely flattened, and
these so-called pancake-like forms have a length of ~15 cm
and height of ~5 cm (Fig. 7.103i). These stromatolitic
“pancakes” are spatially associated with biohermal
stromatolites and have been accreted in marine-influenced
depositional system. They have a core composed of dark
brown calcareous siltstone overgrown by fine laminae. The
latter show compositional and micro-structural patterns simi-
lar to those described for club-like and subspherical
stromatolites. The microbial pancakes are encased in a
c. 10-cm-thick bed of flat-laminated stromatolite and covered
with dolarenite matrix-supported breccia. Angular, platy
micritic dolostone clasts, up to 10 cm in length, show uniform
imbrication, thus suggesting a strong unidirectional current.
A c. 2100–2000Ma Rimmed Carbonate Shelf; theKalix Greenstone Belt
The Palaeoproterozoic Kalix Greenstone Belt is located at
the northern end of the Bothnian Bay in Sweden (Fig. 3.1).
The belt constitutes a 5,800-m-thick succession of volcanic
and various sedimentary rocks that are informally subdivided
into three groups (Lager and Loberg 1990). The 3,000-m-
thick Lower group is comprised of subaerially erupted,
tholeiitic basalts interbedded with fluviatile conglomerates
deposited in an intraplate rift setting. The succession is
truncated by a break-up unconformity, which is overlain
by a 800-m-thick succession of the Middle group com-
posed of dolograinstones, stromatolitic dolostones, arenites,
volcaniclastic and mafic volcanic rocks. The Middle group
was deposited in a marine-influenced rift and near-shore
marine settings followed by a rimmed carbonate shelf. The
Upper group, which is more than 2,000 m thick, is composed
of deep-water shales deposited on the drowned carbonate
shelf/platform in response to tectonically enhanced subsi-
dence. The overall succession documents a Palaeoprotero-
zoic depositional history of the Fennoscandian Shield from
1306 N. McLoughlin et al.
rifting to passive margin development associated with dis-
persal of the Neoarchaean supercontinent (Fig. 7.104).
Only the Middle group succession contains carbonates of
argued microbial origin (Fig. 7.105). The studied succession
includes a varied suite of stromatolitic and flat-laminated
dolostones (described using Preiss’s (1976) classification).
The thickness of stromatolite and flat laminated units ranges
from a few centimetres to as much as 15 m. The stromatolitic
units are commonly composed of 1–3 varieties of stromato-
litic morphologies, and several modes of stromatolite occur-
rence have been recognised. Detailed discussion of
sedimentological aspects and depositional settings is
provided in Wanke and Melezhik (2005).
Kalix Greenstone Belt Morphotype 1: Parallel-Branching Columnar StromatoliteMorphotype 1 occurs in the Lower formation between 135
and 133.5 m in the section (Fig. 7.105a). It comprises
members of two cycles; each starts with a fine laminite
unit that is gradually or sharply overstepped by a columnar
stromatolite bed (Fig. 7.106a–c), both forming tabular
biostrome 60–70 cm in thickness. Columns are 1–5 cm in
diameter with beta- and alpha-parallel branching and
bumpy, eroded margins, evenly and tightly spaced, vertical
or slightly and uniformly inclined coherently with current
ripples in substrate. Gently to steeply convex laminae are
composed of alternating sparite and clotted micrite with
hybrid fabrics. Intervening material is rounded, tabular
dolostone clasts, mafic pyroclastic particles, dolosparite
and non-laminated dolomicrite. The stromatolites accreted
in an intertidal setting (Wanke and Melezhik 2005).
Kalix Greenstone Belt Morphotype 2: StratiformLaminitesThe stratiform laminites (Fig. 7.107) are the most common
morphotype of argued microbial origin documented in the
carbonates of the Kalix Greenstone Belt. They form low-
relief tabular biostromes and bioherms 1–120 cm thick.
Main structural and compositional motif is rhythmically
laminated, thick, grey micrite laminae with filmy
microfabrics, crinkled and scalloped upper boundaries
separated by a thin, dark grey film of siliciclastic silt and
mud. Abundant fenestrae are filled with dolospar. Incipient
and fully-developed tepee structures are common in
horizons with fenestrae. Many of the laminites show a
patchy appearance due to disruption by desiccation/diage-
netic processes (Fig. 7.107a). In peripheral parts of
biostromes and bioherms, the stratiform laminites show par-
tial erosion, fragmentation and redeposition in the form of
platy fragments forming stone rosettes (Fig. 7.107f).
Morphotype 2 stromatolites accreted in variable depositional
settings ranging from sabkha, intertidal through supratidal
carbonate and sand flats to subtidal environments (Wanke
and Melezhik 2005).
The stratiform laminites, which occur in the Lower
formation between 135 and 133.5 m (Fig. 7.106a), form
30–45-cm-thick, tabular biostromes that show a cyclic
development. Morphotype 2 forms the lower units in two
cycles, each overstepped by stromatolite beds of
morphotype 1. The stratiform laminites have either transi-
tional or sharp contacts with their basal substrate and
interbedded morphotype 1 stromatolites (Fig. 7.106a, b).
The substrate for the basal unit is a current-rippled dola-
renite. The stratiform laminites are composed of flat or
corrugated, cambered, warped or crumbled alternating
laminae of micrite with filmy microfabrics and sparite;
laminoid fenestrae (up to 9 cm) and “birdseyes” filled with
dolospar are abundant. Small-scale cumulate stromatolites
are a characteristic feature of the stratiform laminites. The
relief on successive growth interfaces is rather flat, although
locally, current ripples are present. In situ brecciation is
common and associated with incipient tepee structures
indicated by a slight upwarping of the carbonate laminae.
The stromatolites accreted (likely by biotic or abiotic chem-
ical precipitation) in an intertidal setting (Wanke and
Melezhik 2005).
Kalix Greenstone Belt Morphotype 3:Microcolumnar and Microdigitate StromatoliteThis morphotype occurs in the Lower formation between
158 and 171 m (Fig. 7.105a). It is developed on
subaqueously erupted mafic lava (Fig. 7.108a) and capped
by mafic tuff (Fig. 7.108b). The depositional setting is
interpreted to be an intertidal environment (Wanke and
Melezhik 2005). Macrofabrics are not pronounced on the
rock surface. The lack of lamination and rather massive
macro-appearance of the stromatolite bedforms superficially
resemble thrombolites, which have a clotted fabric (cf.
Aitken 1967). The stromatolites comprise a tabular
biostrome 15 m thick, occurring as several 0.3–0.5-m-thick
cycles separated by thin tuffitic beds. Microcolumnar and
microdigitate stromatolites are 1–5 mm in size, tightly and
unevenly packed, in places fragmented. They show beta-parallel branching, bumpy walls and are built of steeply
convex to parabolic, alternating micrite and sparite laminae.
In many beds, both the stromatolites and the intervening
sediments are enriched in fine-grained volcaniclastic mate-
rial. In such beds, dolospar-filled fenestrae are abundant.
Some fenestrae are enlarged by dissolution processes. The
microdigitate stromatolites and micrite ‘fingers’ may be
either unevenly distributed or occur as pockets within other
microstructures.
Kalix Greenstone Belt Morphotype 4:Microspherical StromatoliteThe microspherical stromatolites occur in the Lower forma-
tion between 158 and 171 m (Fig. 7.105a) where they
are closely associated and intergrown with morphotype 3.
8 7.8 Traces of Life 1307
In places, irregular, microspherical stromatolites and clotted
dolomicrite particles are successively overgrown by irregularly
branching microdigitate stromatolites (Fig. 7.108d). The
microspheres are 1–3 mm in diameter with beta-parallel
branching and bumpy walls. They are tightly and unevenly
packed, in places fragmented, and comprise alternating
sparitic and clotted micritic laminae, i.e. possible hybrid
crusts with a wavy and wrinkled appearance.
Kalix Greenstone Belt Morphotype 5: Subsphericalto Pillow-Shaped StromatoliteThis stromatolite morphotype is widespread. Examples are
particularly abundant in the Lower formation (Fig. 7.105a,
11–13, 20–21, 28.5, 49, and 170–173 m) where they are
developed in several 0.5–1.0-m-thick dolomicritic units,
which are sandwiched between beds of quartz arenites or
mafic volcaniclastic rocks (Fig. 7.109a–e). Some stromato-
litic bedforms are entirely encased in mafic tuff
(Fig. 7.109d). The stromatolites form various bioherms,
0.2–1.3 m thick and 0.3–5 m in diameter. They are
irregularly distributed in space and their internal
organisation is rather diverse. The domed bioherms may
form a series of isolated build-ups comprising two or more
pillow-like, successively growing stromatolitic bodies with
the smaller ones located at the base and the larger ones on
top of the bioherm (Fig. 7.109b–e). Several of these pillowed
bioherms may be laterally connected, thus forming a body
transitional to the domed biostrome but with a limited lateral
extent on the order of a few metres. In such linked bioherms,
the individual bodies show different shapes ranging from
club-like, single pillow- to multiple-pillowed forms
(Fig. 7.109a). Some bioherms are draped by desiccated
mud sheets (Fig. 7.109f) or by cracked dolomicritic beds
(Fig. 7.109g). Those enclosed within mafic tuff comprise a
series of tightly packed domal stromatolites. Such bioherms
have sharp contacts with both basal tuff bed and the
interfilling tuff material. When exhumed, the upper surface
exhibits multi-spheroidal topography (Fig. 7.109e). Numer-
ous stromatolitic micromounds and small non-laminated
dolomicrite lenses can be traced along strike on either side
of the bioherm (Fig. 7.109c, d). Morphotype 5 stromatolites
were formed in supratidal, intertidal and subtidal settings
(Wanke and Melezhik 2005).
Kalix Greenstone Belt Morphotype 6: Hat-ShapedStromatoliteThis stromatolite morphotype is localised in the Lower for-
mation between 172 and 173 m (Fig. 7.105a). It occurs as
randomly distributed bodies, 1–20 cm high and 0.1–0.8 m in
diameter, mainly encased into mafic tuff with rare smaller
examples in dolomicritic beds (Fig. 7.110a). The hat-shaped
stromatolites may form a series of isolated build-ups com-
prising either a single or several successively growing bod-
ies, separated from each other by a thin mafic tuff layer
(Fig. 7.110a). Numerous stromatolitic micro-mounds
encased into non-laminated dolomicrite lenses and mafic
tuff can be traced along strike on either side of the larger
bioherms (Fig. 7.110a). Such stromatolites are composed of
alternating 1–5-mm-thick, flat-laminated, undulatory,
pseudocolumnar micrite laminae and thin films of mud and
mafic tuff. The laminae are commonly highly disrupted and
in situ brecciated (Fig. 7.110c); many experienced erosion
and redeposition in the form of stone rosettes where platy
stromatolite fragments are cemented by green-grey, non-
laminated dolomicrite or mafic tuff (Fig. 7.110d). There
are many examples of incipient stromatolites composed of
2 or 3 lamina emplaced into either dolomicrite or mafic tuff
(Fig. 7.110e). Morphotype 6 stromatolites were formed in a
supratidal setting (Wanke and Melezhik 2005).
Kalix Greenstone Belt Morphotype 7: “Oversized”Spheroidal StromatoliteThe “oversized” spheroidal stromatolites, >1 m in diameter
and tightly spaced, occur in the Upper formation
(Fig. 7.105b, 127–129 m and Fig. 7.111) where they form
a tabular biostrome,>1 m thick and >30 m in diameter. The
stromatolite comprises flat-laminated, undulatory and wavy,
1–4-mm-thick, micritic laminae alternating with thin films
of mud. This may be separated by laminae composed of
silica and dolospar that overall confer a rhythmic-like
appearance. Several laminae show scalloped truncations
indicative of erosion. The interspheroidal space is filled
with intraclasts in the form of flaky conglomerate and
stone rosettes. The stromatolites have been assigned to an
intertidal carbonate flat setting (Wanke and Melezhik 2005).
Kalix Greenstone Belt Morphotype 8: Branchingand Coalescing Spheroidal StromatoliteThe coalesced spheroidal stromatolite are pink in colour,
tightly spaced, and rest directly in sharp contact on the
bioherm composed of morphotype 7 (Fig. 7.111b). They
form a tabular biostrome,>0.5 m thick and>30 m diameter.
Spheroidal forms show both branching and coalescing, and
are overgrown by a common envelope with irregular shapes
(Fig. 7.111d). They comprise flat-laminated, undulatory and
wrinkled, 1–2-mm-thick, micritic laminae alternating with
thin films of mud (perhaps suggesting accretion by both
chemical precipitation and trapping and binding). The
interspheroidal space is filled with flakestone. The larger
flakes are 1 mm in thickness and up to 3 cm in length,
cracked and aligned parallel to the bedding surface. The
smaller flakes are 2–5 mm in size and uniformly imbricated.
1308 N. McLoughlin et al.
Kalix Greenstone Belt Morphotype 9:Semispheroidal to Semicolumnar StromatoliteThis stromatolite morphotype is a feature of the Lower
formation (Fig. 7.105a, 129.5–130.5 m). The stromatolites
form a stack of domed, ellipsoidal bioherms 0.5–1.0 m thick
and 2–5 m in diameter with synoptic relief of 10–20 cm. The
bioherms are in sharp contact with surrounding micritic
dolostones. All contacts are commonly defined by thin
interlayers of mafic volcaniclastic material (Fig. 7.112a, b).
The space between individual bioherms is filled with
floatstones, mafic tuff and non-laminated dolomicrite, all
occurring as interfingering lenses (Fig. 7.112a, b). The
stromatolites occur as randomly inclined, tightly spaced
semi-spheroids and semi-columns, 1–5 cm in height, with
smooth walls and rare beta-branching. They are composed
of steeply convex, parabolic or rectangular, 1–2-mm thick,
micritic laminae alternating with thin films of mud;
dolospar-filled fenestrae are abundant. The depositional
environment of the bioherms has been assigned to a
supratidal setting (Wanke and Melezhik 2005).
Kalix Greenstone Belt Morphotype 10: SolitaryDomal and Spheroidal StromatolitesSolitary spheroidal stromatolites are abundant throughout
the entire section. They are commonly 20–50 cm in diameter
and occur with microbial dolostones composed of flat-
laminated stromatolites (Fig. 7.113a). The stromatolite
margins are smooth and well defined. Apparent synoptic
relief reaches 15 cm. The stromatolites comprise 1–5-mm-
thick, dolomicritic, cumulate laminae separated by thin films
of mafic volcaniclastic material.
Another type of solitary stromatolite occurs in the Lower
formation (Fig. 7.105a, 171.5–173 m) hosted by mafic tuffs
(Fig. 7.113b). The stromatolites form well-defined bodies,
which are 15–35 cm in length and 10–20 cm in height. They
comprise 1–10-mm-thick, gently convex, dolomicritic
laminites separated by thin films of mafic volcaniclastic
material. The depositional setting of solitary stromatolites
in the Kalix succession ranges from supratidal through inter-
tidal to subtidal (Wanke and Melezhik 2005).
On the Biogenicity and Accretion Mechanism ofthe Fennoscandian Stromatolites
The different stromatolite morphologies from various depo-
sitional settings described above have experienced
variable, though significant degrees of post-depositional re-
crystallisation. Consequently not all of the biogenicity
criteria outlined in this review can be readily applied to
these stromatolites. Table 7.8 summarises an effort to evalu-
ate the Fennoscandian stromatolites against the propsed
criteria – Table 7.8 does not include criteria 6, 7, 8, 10
concerning microfossils or trace fossils, because the preser-
vation is insufficient and no such structures have been
reported. Rather, Table 7.8 summarises the macroscopic
and microfabric characteristics that have been argued to be
indicative of a microbial origin. It also includes comments
on additional features that pertain to biogenicity.
Similar to the assessment of biogenicity, deciphering
the accretionary mechanisms of the Fennoscandian stro-
matolites is sometimes hampered by re-crystallisation.
Despite this, it is apparent that the Lomagundi-Jatuli interval
stromatolites include examples of both sparitic and micritic
fabrics, sometimes with laminae of ‘trapped’ clastic detritus.
Many can therefore be regarded as “hybrid crusts” with
laminae that formed through a combination of both chemical
precipiation and trapping and binding of micrite and clastic
sediment that may (in both cases) have been microbially
mediated. The quality of preservation of organic matter is
insufficient to convincingly identify microfossils, but many
examples of convex upwards structures and laminae that
thicken over crests perhaps point to a contribution from
photosynthetic microorganisms.
Significance of Stromatolites from theLomagundi-Jatuli Interval of Fennoscandia
Comparison of Stromatolites from DifferentDepositional SettingsStromatolite morphology, distribution and microfabrics are
the result of a complex interplay of chemical, physical and
biological variables as reviewed above. Various classifica-
tion systems have been proposed. Most recent studies adopt
the system of Groups being broadly equivalent to Genera,
and Forms taken to be equivalent to Species. The Groups are
based upon a combination of macro-morphological
characteristics, for instance, laminar geometry and the pres-
ence or absence of branching, and sometimes microstruc-
ture. The Forms are predominantly or exclusively defined on
the basis of microstructure. Here we avoided this taxonomic
style of classification and rather described the morphotypes
including microfabrics of the Fennoscandian stromatolites.
We now summarise our earlier descriptions of the selected
depositional environments and stromatolites of the
Fennoscandian Shield across the Lomagundi-Jatuli interval
to ask: what is the effect of depositional environment on the
stromatolite abundance and morphology? We then use these
findings to critically review the potential usefulness of stro-
matolite biostratigraphy across this interval.
First, in the case of the Kalix carbonate shelf/platform,
the evolving depositional settings have been constrained
using the distinctive succession of sedimentary structures.
Planar and herringbone cross-bedding, hummocky cross-
stratification, symmetrical and asymmetrical ripples,
8 7.8 Traces of Life 1309
desiccation cracks and tepees suggest a shallow-marine to
shoreline depositional setting influenced by tides, storms,
and repeated phases of emergence (Wanke and Melezhik
2005). Flat-laminated dolostones with desiccated laminae
have been commonly documented in supratidal settings.
All other stromatolite morphologies, in contrast, occur
across supratidal, intertidal and subtidal settings, and there-
fore cannot be taken to be indicative of any specific environ-
ment within this shallow-water marine basin. The Kalix
shelf/platform carbonate rocks are enriched in 13C, thus
recording the Lomagundi-Jatuli isotopic event. Published
d13C data obtained from the least altered samples range
between +2 ‰ and +8 ‰ (Melezhik and Fallick 2010).
The overall stratigraphic trend within a 600-m-thick succes-
sion is a gentle oscillation between +2 ‰ and +4 ‰ with
a second-order positive excursion up to +8 ‰ through a
c. 150-m-thick unit in the middle and upper parts of the
succession (Melezhik and Fallick 2010). The onset of the
second-order excursion coincides with the development of
pink spheroidal stromatolites (“oversized” and branching
and coalescing morphotypes, Fig. 7.111a–c) and the transi-
tion from a marine-influenced rift to a passive margin setting
(Wanke and Melezhik 2005).
Second, the Tulomozero Formation of the Onega carbon-
ate platform shows rather diverse shallow-water, evapora-
tive facies but the depositional environments have been
reconstructed with a lower level of confidence, due to lim-
ited observations of sedimentary structures in outcrop.
Similar to the Kalix carbonate platform, most of the flat-
laminated stromatolites (stratiform laminates) have been
observed in supratidal and sabkha-like environments
(Melezhik et al. 2000). In contrast, the diverse columnar
branching stromatolites have no obvious affinity to any
specific depositional setting. Large, solitary, non-branching
columnar stromatolites of Member D (Fig. 7.97) are
associated with influxes of seawater as indicated by the Sr-
isotopic data (Melezhik et al. 2005b; Kuznetsov et al. 2010)
and the onset of active hydrodynamic environments, as
suggested by tidal-channel, herringbone, cross-bedded
arenites. The branching, mini-columnar stromatolites of
Member H (Fig. 7.98d) are also embedded in transgressive
swash-zone, oolitic dolostone lithofacies with a marine Sr-
isotope signature. Both morphologies are unknown from
shallow-water evaporitic lithofacies. Similarly to the Kalix
stromatolitic dolostones, the Onega platform carbonate
rocks are enriched in 13C, but with even higher d13C values
ranging between +5 ‰ and +18 ‰ (Melezhik et al. 1999).
Although there is no one-to-one correlation between d13Cvalues and stromatolite morphologies, the most enriched
values are always bound to stratiform, intensely-desiccated
stromatolites from shallow water environments (Melezhik
et al. 2005b)
Third, the Kuetsj€arvi rift-bound lacustrine system
preserves a sedimentary succession containing only few
stromatolites that show limited diversities. It is possible
that this limited morphological diversity is a result of the
lacustrine depositional environment, although this is not
clear because there have been few direct comparisons of
marine and lacustrine stromatolites, except for example
Dean and Eggleston (1975). Stromatolites are, however, a
common feature of littoral zones in recent lakes that range in
chemistry from extremely dilute to hypersaline (Talbot and
Allen 1996), and significant morphological variability has
been reported from within the same bed, including: columns,
domes, and branching morphotypes (e.g. Awramik et al.
2000; Cohen et al. 1997). In the Archaean Meentheena
Formation of Western Australia, for example, also
interpreted to be a lake setting, both morphological
variability and stability were reported by Awramik and
Buchheim (2009). Notably, they observed that mm-scale
laminae produced by the grouping of thinner, dark–light
laminae are more common in this and other Archaean
lakes (e.g. Link et al. 1978) than in marine stromatolites.
This mm-scale laminated pattern has been also documented
in the Kuetsj€arvi lacustrine stromatolites (Fig. 7.103a, f).
However, similar types of microbial laminates have also
been found in the Onega and Kalix carbonate platform
successions (Figs. 7.98a, b and 7.99c), thus they are not
diagnostic of lacustrine stromatolites. We presume that the
limited abundance and even more limited morphological
diversity of the Kuetsj€arvi lacustrine stromatolites is the
result of the depositional and tectonic settings. All of the
carbonate lithofacies are impure (Melezhik and Fallick
2005), suggesting a significant influx of siliciclastic material
from rift shoulders into the shallow-water lake, thus
inhibiting stromatolite growth.
The Kuetsj€arvi lacustrine carbonates are enriched in 13C.
The entire d13C range in the least altered dolostone and
limestone samples is from +5 ‰ to +9.6 ‰, and the strati-
graphic d13C profile displays a smooth negative excursion
from +8.5 ‰ to +6 ‰ (Melezhik et al. 2005a). Similarly to
the Onega platform carbonates, there is no obvious link
between stromatolite morphology and enrichment in 13C.
However, a well-pronounced drop of d13C in the upper part
of the succession is coincidental with the influx of seawater
into the lacustrine basin (Melezhik et al. 2005a) and accre-
tion of club and spheroidal stromatolites (Fig. 7.103g–h).
In summary, from this overview of stromatolites accreted
in three contrasting depositional settings of the
Fennoscandian Shield, it is evident that non-columnar
stromatolites are common in all settings, whereas columnar
biohermal/biostromal and solitary morphotypes are less typ-
ical in the shallow-water, rift-bound lake. We can conclude
that the prevailing factors controlling stromatolite
1310 N. McLoughlin et al.
morphology in the Proterozoic of the Fennoscandian Shield
are the environment and tectonic setting. There is no obvious
universal correlation between stromatolite morphologies and
degree of 13C enrichment in coevally precipitated
carbonates. However, in each of the three depositional
settings, there are specific links between the most 13C-rich
(Onega, Kalix) and the relatively 13C-depleted (Pechenga)
values (see Chap. 7.3) and the appearance and/or dominance
of certain stromatolite morphotypes. Whether such links are
coincidental or causal, remains to be ascertained.
Stromatolite Biostratigraphy Across theLomagundi-Jatuli IntervalThere is an ongoing discussion concerning the potential
biostratigraphic usefulness of stromatolites (e.g. Semikhatov
and Raaben 2000) with the advocates being split into
polarised groups. One group argues that changes in stromat-
olite morphology and microstructure through geological
time reflect evolutionary changes in stromatolite-building
biotas and their environments, and that stromatolite mor-
phology can therefore be used as the basis for the correla-
tion, particularly for Precambrian sequences. This approach
was first pioneered in the U.S.S.R. as explained in Maslov
(1960) and reviewed by Semikhatov (1976). The other group
argues that this biostratigraphic approach breaks down,
because we cannot separate the macro- and microscopic
fingerprints of environmental versus microbiological
controls. For example, Knoll et al. (1989) described Protero-
zoic stromatolites from Spitsbergen where macrostructural
diversity exceeded the microtextural diversity and thus they
concluded that environmental parameters have the greatest
control over macro-morphology and that micro-textural
diversity is the best proxy for microbial diversity. Con-
versely, it has been noted that macro-morphologically simi-
lar stromatolites assigned to a particular Genus can yield
diverse microstructures and this led Hofmann (1977) to
suggest that macro-morphology and microstructure are par-
allel rather than hierarchical tools in stromatolite classifica-
tion. Thus Semikhatov and Raaben in their review in 2000
conclude that stromatolites are not suitable for the subdivi-
sion of Proterozoic stratigraphy, but rather provide
palaeontological characterisation of chronostratigraphic
units which have been defined by other methods and can
contribute to their correlation only within the limits of par-
ticular stratigraphic provinces. They caution that interpro-
vincial stromatolite-based correlations are of lower
reliability due to strong lateral variations in the taxonomic
composition of stromatolite assemblages; variability in the
time ranges of taxa at the interregional scale, as
demonstrated by chemostratigraphic studies; and provincial-
ism of time-dependent taxonomically distinct complexes.
In the Fennoscandian Shield, biostratigraphic correlation
of the Jatuli-age stromatolites has been attempted for geo-
graphically separated areas within the Karelian craton (e.g.
Makarikhin 1992). Although this exercise was argued to be
successful, it remains to be proven whether synchronously
deposited strata or rather similar depositional settings have
been correlated, especially in the absence of any precise
radiometric dates. Microdigitate stromatolites described
here as Tulomozero Formation morphotype 2 have received
particular attention and been suggested as potentially useful
for global interbasinal correlation across this interval
(Medvedev et al. 2005). Similar stromatolite morphologies
have been described from the 2173 � 80 Ma Juderina For-
mation of the Glengarry Group (Woodhead and Hergt 1997),
from the Yerrida Basin of Western Australia (e.g. Fig. 10 of
Grey 1994). Carbonates from this formation show a d13Cenrichment of up to +9 % (Lindsay and Brasier 2002) and
were deposited in a shallow marine, restricted-to-evaporitic
setting with high sulphate concentrations (El Tabakh et al.
1999). Medvedev et al. (2005) highlight several additional
examples of Palaeoproterozoic ministromatolite horizons
from North America that are morphologicaly comparable
to the Karelian microdigitate stromatolites. They conclude
that the “combination of lithological, geochemical, paleocli-matic and palaeontological [stromatolite] features emphasizes
time-specific characteristics of these successions and
provides the basis for interbasinal correlation among car-bonate successions deposited during the 2.2–2.1 Ga carbon
isotope excursion”. Crucially, they state that stromatolite
morphology must be considered together with the environ-
mental, lithological and geochemical evidence – a
realisation that stromatolite morphology alone is insuf-
ficient for stratigraphic correlation and a product of all
these factors.
Concluding Remarks and Implications for theFAR-DEEP Core
Several of the FAR-DEEP drillholes intersect the
Lomagundi-Jatuli interval and contain d 13C-rich carbonates
including stromatolites:
1. The shallow-water, evaporitic Tulomozero carbonate
platform was intersected by Holes 10A, 10B and 11A.
The core contains stromatolites (Fig. 7.114a–c) though
they are less abundant and morphologically less diverse
in comparison with those previously published and
summarised in this chapter (Fig. 7.98, 7.99, 7.100, and
7.101). This is probably because the FAR-DEEP
drillholes intersected much shallower carbonate
lithofacies accumulated in evaporitic settings, with a
8 7.8 Traces of Life 1311
considerable influx of siliciclastic material (Fig. 7.114d)
that suppressed the accretion of stromatolites (see Chaps.
6.3.1, and 6.3.2).
2. The deep-water Umba carbonate rampwas intersected by
Hole 4A (see Chaps. 6.1.3, and 6.1.4) and contains deep-
water dolostones that lack microbial laminates or
stromatolites (Fig. 7.114e). This is a surprise, because
nearly all previously studied 13C-rich Jatulian carbonate
successions of the Fennoscandian Shield contain
stromatolites. The cause again is argued to be an
unfavourable depositional setting: either too deep,
below sunlight penetration; or an unstable substrate due
to a hydrodynamically very active environment; or a
combination of these factors that together prevented
accretion of stromatolites.
3. The lacustrine rift-bound Kuetsj€arvi carbonate succes-sion was intersected by Hole 5A (see Chap. 6.2.2). The
retrieved succession of lacustrine carbonates contains
mainly non-columnar stromatolites with limited abun-
dance, and one subspherical morphotype (Fig. 7.114g–f).
Here again, accretion of stromatolites was apparently
suppressed by deposition in a shallow-water lake that
experienced frequent phases of emergence and erosion,
and with considerable influx of clastic material (Melezhik
et al. 2004).
The drilled successions could be used to further test the
biostratigraphic potential of stromatolite morphologies.
Holes 10A, 11A and 11B, which intersected different
lithofacies with respect to those containing previously
described abundant stromatolites in the Onega basin, have
a potential to test biostratigraphic correlation within a single
large basin. In contrast, Hole 5A contains lacustrine
stromatolites with morphologies that can be compared to
the Onega basin stromatolites and may be used to further
investigate the biostratigraphic potential at a regional scale.
The drilled successions could also be used to test the
hypothesis that local enhancement in 13C of the
Lomagundi-Jatuli event carbonates resulted from the photo-
synthetic uptake of 12C by cyanobacterial stromatolites in
restricted basins with high productivity (Melezhik et al.
1999, and also discussed in Chap. 7.3). This hypothesis has
been advanced on the basis of the observation that
cyanobacterial photosynthesis in modern freshwater lakes
and streams is known to cause 13C enrichment of dissolved
inorganic carbon in waters within cyanobacterial colonies
(and therefore of calcite precipitated from these waters) by
up to three per mil, but more commonly by less than one per
mil (Pentecost and Spiro 1990; Arp et al. 2001). Photosyn-
thetic effects are most pronounced in well illuminated slow
flowing or stagnant waters (Andrews et al. 1997; Arp et al.
2001) where carbon species are not quickly replaced by
dissolved inorganic carbon from refluxing water. Some
microbial carbonates from modern lake shores with slow
flowing or stagnant water reportedly have carbonate d13Cvalues as much as five per mil heavier than values obtained
from the mollusc shells that they encrust (Andrews et al.
1997). These processes produce local effects within
cyanobacterial colonies rather than changing the d13C com-
position of whole water bodies. However, even non-
microbial carbonates could be affected if post-depositional
carbon isotope equilibration with 13C-enriched microbial
carbonate occurred, especially if a large volume of microbial
carbonate was formed. This mechanism may thus provide
one explanation for some local d13C enrichment beyond the
global Lomagundi-Jatuli signal. Alternatively, other micro-
bial processes, particularly methanogenesis, may have
caused some local positive d13C shifts. The stromatolites
of the FAR-DEEP cores together with surface samples
provide an additional opportunity to further investigate
these scenarios.
1312 N. McLoughlin et al.
Table
7.8
Evaluationofthebiogenicty
criteriaintroducedin
thetextas
applied
tostromatolitemorphotypes
from
theLomagudi-JatuliintervaloftheFennoscandianShield.NAnotapplicable,
usually
because
thesefeaturesarenotpreserved
Biogenicity
criteria
Shape
1.
Sed.
rocks
2.Syn-
sedim
entary
3.
Convex
up
4.Lam
inae
thicken
over
crests
5.Wavy
wrinkly
laminae
9.Mat
chips
11.Comp.differences
bioherm
versusSedim
imen-
tary
matrix
Additional
comments,especially
microfabrics
Location
Tulomozero
form
ation,OnegaBasin
Morphotype
1,Fig.7.129
Spaced
bioherms-
branched
columnar
YY
YN
YN
Y
Morphotype
2,Fig.7.129
Continuous
biostromes,
branched
columnar
YY
YY
YN
NClotted
fabrics
sometim
es
preserved.Strongcurrent
alignment
Morphotype
3,Fig.7.130
Flatlaminated
YY
YSometim
esY
N?
Fenestrae
Morphotype
4
Columnar
branched,mini-
columnar
YY
YY
YN
?
Morphotype
5,Fig.7.131
Nonbranching
mini-columnar
YY
YN
NOncolites
NClotted
layersalternatewith
microspar,possible“hybridcrusts”
Morphotype
6,Fig.7.132
Large,non
branchingcolumnar
YY
YSlightly
Slightly
NY
Microfabrics
recrystallised
Morphotype
7
Bulbous
YY
YSometim
esSlightly
Oncolites
Y
Morphotype
8,Fig.7.132
Hem
ispherical
YY
YY
YN
Y
Kuetsj€ arvisedim
entary
form
ation,PechengaGreenstoneBelt
Morphotype
1,Fig.7.124
Stratiform
laminites
YY
YSometim
esY
Y?
Welldeveloped
fenestrae,
widespread
dessication,possible
“hybridcrusts”
Morphotype
2,Fig.7.124
Clublike
subspherical
YY
YSometim
esY
YY
Fenestrae,also
possible
trapping
andbinding
Exotic
structures,
Fig.7.124
“Pancakes”
YY
YN
YY
Y
Middle
group,KalixGreenstoneBelt
Morphotype
1,Fig.7.106
Parallelbranching
columnar
YY
YY
YN
YHybridcrusts
Morphotype
2,Fig.7.107
Stratiform
laminites
YY
NN
YY
YMicrobialandorabioticchem
ical
precipitation
Morphotype
3,Fig.7.108
Microcolumnar,
microdigitate
YY
YSometim
esY
NN
Morphotype
4,Fig.7.108
Microspherical
YY
YY
YN
YHybridcrusts
YY
YSlightly
NY
Y
(con
tinu
ed)
8 7.8 Traces of Life 1313
Table
7.8
(continued)
Biogenicity
criteria
Shape
1.
Sed.
rocks
2.Syn-
sedim
entary
3.
Convex
up
4.Lam
inae
thicken
over
crests
5.Wavy
wrinkly
laminae
9.Mat
chips
11.Comp.differences
bioherm
versusSedim
imen-
tary
matrix
Additional
comments,especially
microfabrics
Location
Morphotype
5,Fig.7.109
Subspherical
to
pillow
shaped
Stromatolitessometim
esencased
intuff
Morphotype
6,Fig.7.110
Hat
shaped
YY
YY
Disrupted
YY
Stromatolitesoften
encasedin
tuff
Morphotype
7,Fig.7.111
“Oversized”
spheroidal
YY
YN
YY
YRhythmiclayering,alternatingspar
andmicrite
Morphotype
8,Fig.7.111
Branchingand
coalescing,
spheriodal
YY
YY
YY
YChem
ical
precipitationand
trappingandbinding
Morphotype
9,Fig.7.112
Sem
ispheroidal
and
semicolumnar
YY
YY
NY
Y
Morphotype
10,
Fig.7.113
Solitary
domal
and
spheroidal
YY
YN
NY
Y
1314 N. McLoughlin et al.
8 7.8 Traces of Life 1315
Fig. 7.94 Images of selected modern stromatolites and microbial mats
(a–e) from Carbla-Foreshore, Shark Bay, Western Australia. (a)
Exposed intertidal domal stromatolites with an orange diatom-rich
coating. (b) Polished slice of a weakly laminated, porous carbonate
stromatolite bioherm. (c) Submerged intertidal stromatolites. (d) Darkbrown, unlithified, pustular microbial mat from the supratidal zone
grading laterally into (e) dark brown, unlithified pinnacle mat, with
polygonal network of ridges also from the supratidal zone. (f) Domal
carbonate strombolites from the shoreline of Lake Thetis, Western
Australia. (g) Submerged columnar stromatolites from Lake Clifton
Western Australia. (h) Calcified stromatolite showing multi-phase
growth from flat-laminated at base with organic-rich layers, upwards
into a disrupted interval followed by small-columnar stromatolites that
coalesce at the surface into a series of spherical forms, Lagoa Salada on
the eastern Brazilian coast. Scale bars: (a) and (c) 15 cm; (b, d, e) 5 cm;
(f, g) 50 cm (Photographs (a-g) by Nicola McLoughlin and (h) by
Victor Melezhik)
1316 N. McLoughlin et al.
Fig. 7.95 Images of selected Neoarchaean stromatolites from the
lower Transvaal Supergroup of South Africa. (a) Giant elongate stro-
matolite bioherms; green, 22-cm-long field book in circle for scale. (b)
“Wrinkled” surfaces of giant, elongate stromatolite bioherms. (c) A
series of stromatolite beds each starting with flat-laminate layers
overlain by tightly-packed, branching, columnar stromatolites of the
Reivilo Formation; irregular dolomitisation emphasized by brown
weathered staining. (d) Single large, columnar stromatolite with
smooth convex laminae, 40-cm-long hammer for scale. (e–f) Diverse
columnar stromatolites of the Reivilo Formation; dolomite shows
brown stain on weathered surfaces (All photographs of Victor Melezhik
with the exception of (d) courtesy of Ronny Schoenberg)
8 7.8 Traces of Life 1317
Fig. 7.96 Interpretative summary of Precambrian authigenic crusts
(taken from Riding 2008). Principal components: Sparry Crust (essen-
tially abiogenic precipitate), Fine-grained Crust (lithified microbial
mat), and allochthonous grains. Intermediates: Hybrid Crust, Coarse
Grained Crust, and Coarse Grained Mat. Apart from allochthonous
grains, each of the other five components and intermediates has at
some time been regarded as containing examples of stromatolites.
Examples of these stromatolitic deposits are indicated in red
(Reproduced with permission of Robert Riding)
1318 N. McLoughlin et al.
Fig. 7.97 Simplified and generalised lithostratigraphic columns of the Tulomozero Formation, illustrating stromatolite diversity and abundance
with the stratigraphic heights of subsequent illustrations indicated (Based on data from Melezhik et al. 2000)
8 7.8 Traces of Life 1319
Fig. 7.98 Branching, columnar stromatolites comprising domed
biostromes of morphotype 1 and laterally extensive biostromes of
morphotype 2. (a) Domed bioherm fromMember B comprising divergent
mini-columns ofmorphotype 1. (b) Plan view of tightly spaced, elongated
columns of morphotype 2 from Member B, axis of elongation from topright to lower left in this image. (c) Stromatolite columns of morphotype
2 with convex laminae from Member C showing bifurcate style of beta
branching. (d) Line drawing showing the marginal structure of columnar
stromatolites from morphotype 2 of Member H, with column walls
ornamented by cornices and peaks. (e) Plan and (f) cross-section views
of closely spaced, slightly elongated columnswith gently convex laminae;
morphotype 2, Member C, elongation in (e) from top left to lower right(Photographs (b, c, e and f) by Pavel Medvedev, (a and d) reproduced
from Makarikhin and Kononova (1983))
1320 N. McLoughlin et al.
Fig. 7.99 Morphotype 3. (a) Stratiform stromatolite from Member C
with low synoptic relief. The laminae are syngenetically brecciated and
contain fenestrae. (b) Mini-columnar stromatolites (lower part of
image) that merge laterally and upwards to form stratiform undulose
stromatolites from the middle part of Member B. The intervening
sediment is red sandstone. (c) Columnar mini-stromatolites from the
middle of Member E ((a) Reproduced from Melezhik et al. (1999) with
permission of Elsevier)
8 7.8 Traces of Life 1321
Fig. 7.100 Morphotype 5 from Member G. (a) Vertical section
showing unbranched mini-columns that expand upwards, with ragged
walls. The intervening pale grey sediment is oncolitic dolomite and
clastic quartz. (b) Individual column with smooth, gently convex
laminae and well defined margins; coin diameter is 1.5 cm. (c) Darkbrown, clotted laminae separated by variable amounts of clear
microspar viewed in transmitted light (Photographs by Pavel
Medvedev)
1322 N. McLoughlin et al.
Fig. 7.101 Morphotypes 6 and 8 (a) Large solitary stromatolite col-
umn of morphotype 6 from Member D in cross-section view. (b)
Hemispherical stromatolites of morphotype 8 from Member D in plan
view on the bedding surface show the coalescing of individual build-
ups (Photograph (b) by Pavel Medvedev, (a) reproduced from
Makarikhin and Kononova (1983))
8 7.8 Traces of Life 1323
Fig. 7.102 Lithological column of the Kuetsj€arvi Sedimentary Formation showing reconstructed depositional environments and position of
stromatolites (Modified from Melezhik and Fallick 2005). Units A (arrowed), B (arrowed) and C (uppermost dolostone) are described in the text
1324 N. McLoughlin et al.
Fig. 7.103 Non-columnar and club-like stromatolite from the
Kuetsj€arvi Sedimentary Formation morphotypes 1 and 2. (a) Flat-
laminated stromatolite from Unit A (see Fig. 7.133); dark grey
dolomicritic laminae may represent the microbial mat, whereas lighter
and thicker laminae consist of white dolosparite and quartz. Photomi-
crograph in reflected, plane-polarised light. (b) Brecciated and buckled,
dark grey microbial dolomicrite from Unit A, containing white
dolospar filling desiccation cracks and fenestrae. Photomicrograph in
transmitted, plane-polarised light. (c) Intensively brecciated, dark grey
microbial dolomite cemented by dolospar from Unit A. Photomicro-
graph in transmitted, plane-polarised light. (d) Polished slab showing a
vertical section from Unit B, containing mottled, non-columnar
stromatolites with pale pink, curly, blister and undulatory, microbial,
dolomicritic laminae separated by thin, irregular partings of silica sinter
(pale grey) and white dolospar
8 7.8 Traces of Life 1325
Fig. 7.103 (continued) (e) Polished slab showing a vertical section
from Unit B, constituting flat-laminated, weakly-domed and
pseudocolumnar stromatolites with pink, haematite-stained, microbial,
dolomicritic laminae separated by thin, irregular partings of silica sinter
(pale grey) and white dolospar. (f) Unit B dark grey stromatolitic
dolomicrite separated by laminae composed of dolospar (pale grey)
and quartz (white). Photomicrograph in transmitted, plane-polarised
light. (g) Solitary, club-like stromatolite morphotype 2 composed of
alternating white and pale grey, curly laminae; intervening sediment is
sandy dolorudite; the steeply convex laminae become progressively
more asymmetric suggesting that the right side of the stromatolitic
structure was abraded by current action
1326 N. McLoughlin et al.
Fig. 7.103 (continued) (h) Solitary subspherical stromatolite emplaced
in clayey dolorudite; the stromatolite comprises disrupted, lensoidal,
undulatory, dolomicritic laminae alternating with thicker laminae com-
posed of detrital dolomite and quartz; coin diameter is 1.5 cm. (i) Darkgrey, calcareous siltstone plate enveloped by alternating white and grey
laminae resembling microbial laminates, thus forming stromatolitic
“pancake”; the “pancake” emplaced into flat-laminated stromatolite
(marked by white bar) and covered with breccia having imbricated
dolostone clasts (marked by red bar) (Photographs (a, d–i) by Victor
Melezhik (b and c) reproduced from Melezhik and Fallick (2005) by
permission of the Royal Society of Edinburgh)
8 7.8 Traces of Life 1327
Fig. 7.104 A model illustrating evolution of depositional and tectonic
settings of the Kalix Greenstone Belt sequence (Modified from Wanke
and Melezhik 2005). (a) Intracontinental rift with volcanic infill (the
Lower group). (b) Marine-influenced, rifted passive margin with
volcanic rocks, siliciclastic and carbonate sediments (the Lower and
Middle formations of the Middle group). (c) Rimmed shelf (the Upper
formation of the Middle group)
1328 N. McLoughlin et al.
Fig. 7.105 Simplified and generalised lithostratigraphic columns of
the Kalix Greenstone Belt, illustrating stromatolite diversity and abun-
dance through the Middle group (Based on data from Wanke and
Melezhik 2005). (a) Section through the Lower formation with strati-
graphic heights of subsequent figures shown by red arrows. (b) Sectionthrough the Middle and Upper formation
8 7.8 Traces of Life 1329
Fig. 7.106 Morphotype 1 parallel-branching, columnar stromatolite
andmorphotype 2 stratiform laminites comprising a cyclic biostrome in
the Lower formation, height 133.5–135 m in Fig. 7.105a. (a) Cross-
section through cyclic biostrome of the flat laminites (1) interbedded
with columnar stromatolites (2); stromatolite columns in the lower bed
are slightly tilted from a vertical position
1330 N. McLoughlin et al.
Fig. 7.106 (continued) (b) A transitional contact between stratiform
laminites (2) and columnar stromatolites (3) in the cyclic biostrome
resting with a sharp contact on mud-draped (dark grey) dolarenite (1).(c) Cryptalgal laminites consisting of micritic laminae with straight,
crinkled or scalloped boundaries; scanned thin-section. (d) Detail
cross-section view of beta-branching, columnar stromatolites from
cyclic biostrome. (e) Cross-section showing wall (left margin) and
half of the stromatolite column composed of steeply convex, thick,
micritic laminae separated by thin laminae composed of silty material
with dolospar-filled fenestrae (white); photomicrograph in transmitted,
non-polarised light. (f) Dolorudite from intercolumnar space consisting
of subrounded tabular clasts of non-laminated dolomicrite and strati-
form laminites emplaced in fine-grained mafic, volcaniclastic material;
scanned thin-section (Photographs (a, b and e) by Victor Melezhik, (c,
d and f) reproduced from Wanke and Melezhik (2005) with permission
from Elsevier)
8 7.8 Traces of Life 1331
Fig. 7.107 Morphotype 2 stratiforml laminites. (a) Stratiform
laminites with a patchy appearance and nonfabric-related solution
cavities and cracks (white) filled with silica and clear dolospar; the
Lower formation, height 50 m in Fig. 7.105a. (b) The microstructure of
micritic dolostone from the Lower formation, height 17.5 m in
Fig. 7.105a. (c) Intensely desiccated micritic horizon with pyrite
(black); vertical cracks and bedding-parallel laminoid fenestrae are
filled with clear dolosparite and silica from the Lower formation, height
39.2 m in Fig. 7.105a
1332 N. McLoughlin et al.
Fig. 7.107 (continued) (d) Desiccated and dismembered stratiform
laminae (arrowed) in dolarenite containing large dolomicrite (grey)and quartz (white) clasts from the Upper formation, Fig. 7.105b,
126 m. (e) In situ brecciated cryptalgal laminites from the Lower
formation, Fig. 7.105a, 91.7 m. (f) Partially orientated, platy, laminated
dolomicrite clasts filling the intercolumnar space in a stromatolite
bioherm; the Upper formation, Fig. 7.136b, 126 m. (a, b, c, e, f) –
scanned thin-section, (d) – photomicrograph in transmitted non-
polarised light (All photographs reproduced fromWanke and Melezhik
(2005) with permission from Elsevier)
8 7.8 Traces of Life 1333
Fig. 7.108 Morphotype 3 microcolumnar and microdigitate stromato-
lite and morphotype 4 microspherical stromatolites of the Lower for-
mation, Fig. 7.105a, 170–158 m. (a) Mafic lava-breccia flow with plan
view of incipient pillow structure beneath a stromatolite build-up;
Fig. 7.105a, 152 m. (b) Mafic tuff layers (green, dark green) with
desiccated and in situ brecciated dolomicrite layer (pale yellow);Fig. 7.105a, 171 m. (c) Microdigitate stromatolites emplaced into
mafic material (dark green); non-fabric-related, solution-enlarged
cavities are filled with dolospar (white). (d) Microspherical stromatolite
successively overgrown by microdigitate stromatolites; the
intercolumnar space is filled with mafic tuff. (e) Two stromatolitic
columns coalescing into one; microfabric is defined by alternation of
laminae composed of dolomicrite and silty material (dark grey);fenestrae are filled with white dolospar
1334 N. McLoughlin et al.
Fig. 7.108 (continued) (f) Microdigitate stromatolite with parabolic
laminae and core composed of clotted dolomicrite. (g) Cross-section
through fenestral laminated carbonate consisting of irregular spherical
stromatolites and dolospar-filled fenestrae (white). (h) Microspherical
stromatolites with a complex internal structure; interspherical space is
filled with mafic tuff. (i) Microspherical stromatolites resembling
oolites: non-fabric-related, solution-enlarged cavities are filled with
dolospar (white). (b, c, d, g) – scanned thin-sections, (e, g, f, h, i) –
photomicrograph in transmitted non-polarised light (Photographs (a–c,
e–i) by Victor Melezhik, (d) reproduced from Wanke and Melezhik
(2005) with permission of Elsevier)
8 7.8 Traces of Life 1335
Fig. 7.109 Morphotype 5 Subspherical to pillow-shaped stromatolite
of the Lower formation. (a) Cross-section through domed stromatolitic
bioherm composed of several tightly packed, subspherical, pillow- and
head-shaped stromatolites; the bioherm is sandwiched between arenitic
sandstone (top) and dolomicrite (bottom); Fig. 7.105a, 54.7 m
1336 N. McLoughlin et al.
Fig. 7.109 (continued) (b) An isolated stromatolitic bioherm com-
prising several successive pillow-like forms overlain by a mafic tuff
bed (top) and underlain by laminated dolomicrite (bottom); Fig. 7.105a,49 m. (c) Two pillow-shaped stromatolites, one on top of the other,
embedded in mafic tuff containing lenses of laminated and massive
dolomicrite, and a smaller subspherical stromatolite (to the right of thelarger stromatolite); Fig. 7.105a, 172.3 m
8 7.8 Traces of Life 1337
Fig. 7.109 (continued) (d) Cross-section of a tabular bioherm com-
posed of several interlinked subspherical and pillow-shaped
stromatolites, which are entirely enclosed in mafic tuff (black); twosmall stromatolitic bodies are developed along strike to the left of thebioherm; Fig. 7.105a, 172.5 m. (e) Exhumed upper surface of
subspherical stromatolites; Fig. 7.105a, 9 m. (f) Plan-view of
desiccated mud, draping subspherical stromatolite; Fig. 7.105a,
9.0 m. (g) Plan view of intensely desiccated dolomicrite sheet which
developed above the top of the mafic tuff bed overlying bioherm shown
in (d) (Photographs (c and e) by Victor Melezhik, (a, b, d, f and g)
reproduced from Wanke and Melezhik (2005), (c) from Melezhik and
Fallick (2010), all with permission from Elsevier)
1338 N. McLoughlin et al.
Fig. 7.110 Morphotype 6 hat-shaped stromatolite of the Lower for-mation, Fig. 7.105a, 173 m. (a) Cross-section through a series of
laterally-linked hat-shaped stromatolites embedded in mafic tuff
containing massive and laminated dolomicrite lenses with incipient
stromatolites appearing as numerous cracked laminae. (b) Hat-shaped
stromatolites comprised of dolomicrite laminae separated by thin film
of mafic tuff
8 7.8 Traces of Life 1339
Fig. 7.110 (continued) (c) Intensively cracked and deformed laminae
separated by mafic tuff (dark brown and dark green) in a hat-shaped
stromatolite. (d) Locally-redeposited stromatolite laminae appear as
stone rosettes and platy fragments. (e) Incipient hat-shaped stromatolites
occurring as single laminae emplaced into mafic tuff
1340 N. McLoughlin et al.
8 7.8 Traces of Life 1341
Fig. 7.111 Morphotype 7 “oversized” spheroidal stromatolite, and
morphotype 8 branching and coalescing spheroidal stromatolite of the
Upper formation. (a) A large spheroidal stromatolite at the base of the
bioherm; the stromatolite comprised of flat and undulatory laminae;
Fig. 7.105b, 129.5 m. (b) Cross-section through a complex tabular
bioherm comprising large, beige, spheroidal stromatolites sharply
overstepped by smaller, pink branching and coalescing, spheroidal,
stromatolites; Fig. 7.105b, 130 m. (c) Planar, low-angle cross-stratified
dolarenite bed (1) erosionally overlain by ripple cross-laminated (2) and
wavy- to low-angle parallel-laminated (3) dolomicrite; several beds are
draped by mud (arrowed); a few beds are mud-draped; the Upper
formation, Fig. 7.105b, 129 m. (d) Detailed view of pink, branching
and coalescing, spheroidal stromatolites; Fig. 7.105b, 130 m
(Photographs (a, b and d) by Victor Melezhik, (c) reproduced from
Wanke and Melezhik (2005) with permission from Elsevier)
1342 N. McLoughlin et al.
Fig. 7.112 Morphotype 9 semispheroidal to semicolumnar stromato-
lite of the Lower formation, Fig. 7.105a, 127–130 m. (a) Cross-section
through stromatolitic build-up forming a stack of lensoidal bioherms
with thrombolitic i.e. clotted appearance; the bioherm is separated from
underlying and overlying dolostones by a thin screen of mafic tuff
(arrowed in red). (b) Inter-biohermal infill (left half of the photo)
composed of alternating lenses of mafic tuff (dark brown) and
dolomicrite
8 7.8 Traces of Life 1343
Fig. 7.112 (continued) (c) Cross-section through a bioherm composed
of small, tightly-packed, spheroidal and semicolumnar stromatolites.
(d) Small semi-spheroidal stromatolite (arrowed in black) showing
branching into two columnar stromatolites (arrowed in yellow).(e) Intercolumnar space filled with clasts of cryptalgal dolomite
emplaced into dolomicrite with clotted microfabric; photomicrograph
in transmitted non-polarised light (Photographs (b–e) by Victor
Melezhik, (a) reproduced from Wanke and Melezhik (2005) with
permission from Elsevier)
1344 N. McLoughlin et al.
Fig. 7.113 Morphotype 10domal and spheroidal
stromatolites of the Lower
formation. (a) Cross-section
through solitary spheroidal
stromatolite emplaced into
cryptalgal dolostones; black
mafic tuff at base; Fig. 7.105a,
86.5 m. (b) Two laterally-linked
spheroidal stromatolites
sandwiched between two mafic
tuff beds; Fig. 7.105a, 171 m
(Photograph (a) reproduced from
Wanke and Melezhik (2005), (b)
from Melezhik and Fallick (2010)
with permission from Elsevier)
8 7.8 Traces of Life 1345
Fig. 7.114 Stromatolitic and non-stromatolitic Jatulia 13C-rich
dolostones and some associated rocks retrieved by FAR-DEEP core
(a) Non-columnar stromatolitic dolostones composed of undulatory
laminae separated by a thin film of dark grey mud; Hole 10A, depth
of 325 m. (b) Layered-columnar stromatolitic dolostone (white)interleaved with red mudstone; Hole 11A, depth of 106.5 m, the core
axes is parallel to the bedding. (c) Pale pink dolostone layers composed
of indistinctly columnar stromatolite; individual stromatolitic layers are
separated by thin films of brown muddy siltstone; Hole 10A, depth of
327 m. (d) Mudstone-supported polymict breccia – a typical lithology
of the Tulomozero Formation – interbedded with stromatolitic
dolostones at the site of Hole 11A; depth of 183 m. All lithologies are
from the Tulomozero Formation and represent a shallow-water carbon-
ate platform. Core diameter in all images is 5 cm
1346 N. McLoughlin et al.
Fig. 7.114 (continued) (e) Indistinctly bedded, beige, redeposited,
deep-water ramp dolostones containing beds, lenses and channels of
altered ultramafic clasts (dark grey) from the Umba Sedimentary For-
mation; the dolostone contains no sign of microbial carbonates;
drillhole 4A, depth of 158–165 m, core diameter is 5 cm. (f) Lacustrine,
non-columnar stromatolite composed of white, beige, pale pink and
pink, dolomicritic, undulatory laminae, which are separated by pale
grey laminae consisting of quartz and dolospar; marine-influenced rift-
bound lake; the Kuetsj€arvi Sedimentary Formation, Hole 5A, depth of
33 m. (g) Lacustrine, grey-purple, subspherical stromatolites nucleated
on white dolarenite; the stromatolite sub-spheres are spaced in pale
grey clayey dolarenite containing patches of white dolarenite; the
Kuetsj€arvi Sedimentary Formation, Hole 5A, depth of 51.5 m; in the
core the stromatolites are inverted (Photographs by Victor Melezhik)
8 7.8 Traces of Life 1347
References
Aitken JD (1967) Classification and environmental signifi-
cance of cryptalgal limestones and dolomites, with
illustrations from the Cambrian and Ordovician of south-
western Alberta. J Sediment Petrol 37:1163–1178
Akhmedov AM, Panova EG, Krupenik VA, Sveshnikova
KY (2004) Devonian and Paleoproterozoic arid palaeobasins
on either side of the Baltic Shield-Russian platform boundary.
Trans St Petersburg Soc Natural 2(86):1–140 (in Russian)
Andrews JE, Riding R, Dennis PF (1997) The stable
isotope record of environmental and climatic signals in
modern terrestrial microbial carbonates from Europe.
Palaeogeogr Palaeoclimatol Palaeoecol 129:171–189
Arp G, Wedemeyer N, Reitner J (2001) Fluvial tufa
formation in a hard-water creek (Deinschwanger Bach,
Franconian Alb, Germany). Facies 44:1–22
Awramik SM (1992) The history and significance of
stromatolites. In: Schidlowski M, Golubic S, Kimberley MM,
Trudinger PA (eds) Early organic evolution: implications for
mineral and energy resources. Springer, Berlin/Heidelberg,
pp 435–449
Awramik SM, Buchheim HP (2009) A giant, late
Archaean lake system: the Meentheena Member (Tumbiana
Formation; Fortescue Group), Western Australia. Precam-
brian Res 174:215–240
Awramik SM, Semikhatov MA (1978a) (1978) The rela-
tionship between morphology, microstructure and microbiota
in three vertically intergrading stromatolites from the Gunflint
Iron Formation. Can J Earth Sci 16:484–495
Awramik SM, Margulis L (1974) Definition of stromato-
lite. Stromatolite Newsl 2:1–5
Awramik SM, Buchheim HP, Leggitt L, Woo KS (2000)
Oncoids of the Late Pleistocene Manix Formation, Mojave
Desert region, California. San Bernardino County Mus
Assoc Quart 47:25–31
Bartley JK, Knoll AH, Grotzinger JP, Sergeev VN (2000)
Lithification and fabric genesis in precipitated stromatolites
and associated peritidal carbonates, Mesoproterozoic
Billyakh Group Siberia. In: Grotzinger JP, James NP (eds)
Carbonate sedimentation and diagenesis in the evolving
precambrian world. Society of economic paleontologists
and mineralogists, special publication, 67. SEPM (Society
for Sedimentary Geology), Tulsa, Oklahoma, pp 59–73
Batchelor MT, Burne RV, Henry BI, Jackson MJ (2004)
A case for biotic morphogenesis of coniform stromatolites.
Phys A 337:319–326
Beukes NJ (1987) Facies relationships, depositional
environments and diagenesis in major Early Proterozoic
stromatolitic carbonate platform to basinal sequence,
Campbellrand Subgroup, Transvaal Supergroup, South
Africa. Sediment Geol 54:1–46
Bosak T, Souza-Egipsy V, Corsetti FA, Newman DK
(2004) Micrometer-scale porosity as a biosignature in car-
bonate crusts. Geology 32:781–784
Bosak T, Liang B, Sim MS, Petroff AP (2009) Morpho-
logical record of oxygenic photosynthesis in conical
stromatolites. Proc Natl Acad Sci 106:10939–10943
Brasier AT (2011) Searching for travertines, calcretes
and speleothems in deep time: processes, appearances,
predictions and the impact of plants. Earth Sci Rev
104:213–239
Buick R (1992) The antiquity of oxygenic photosynthe-
sis: evidence from stromatolites in sulphate-deficient
Archaean lakes. Science 255:74–77
Buick R, Dunlop JSR, Groves DI (1981) Stromatolite
recognition in ancient rocks: an appraisal of irregularly
laminated structures in an Early Archaean chert-barite unit
from North Pole, Western Australia. Alcheringa 5:161–181
Cohen AS, Talbot MR, Awramik SM, Dettman DL, Abell
P (1997) Lake level and paleoenvironmental history of Lake
Tanganyika, Africa, as inferred from late Holocene and
modern stromatolites. Geol Soc Am Bull 109:444–460
Daly JS, Balagansky VV, Timmerman MJ, Whitehouse
MJ (2006) The Lapland-Kola orogen: Palaeoproterozoic
collision and accretion of the northern Fennoscandian litho-
sphere. In: Gee DG, Stephenson RA (eds) European litho-
sphere dynamics, Geological Society of London Memoir 32:
579–598
Dean WE, Eggleston JR (1975) Comparative anatomy of
marine and freshwater algal reefs. reefs, Bermuda and
central New-York. Geol Soc Am Bull 86:665–676
El Tabakh M, Grey K, Pirajno F, Schreiber BC (1999)
Pseudomorphs after evaporitic minerals interbedded in 2.2
Ga stromatolites of the Yerrida basin, Western Australia:
origin and significance. Geology 27:871–874
Freytet P, Verrecchia EP (1998) Freshwater organisms
that build stromatolites: a synopsis of biocrystallization by
prokaryotic and eukaryotic algae. Sedimentology
45:535–563
Gaal G, Gorbatchev R (1987) An outline of the precam-
brian evolution of the baltic shield. Precambrian Res
35:15–52
Golubic S, Seong-Joo L, Browne KM (2000)
Cyanobacteria: architects of sedimentary structures. In: Rid-
ing RE, Awramik SM (eds) Microbial sediments. Springer,
Berlin/Heidelberg/New York, pp 57–67
Grey K (1994) Stromatolites from the Palaeoproterozoic
(Orosirian) Glenngarry Group, Glengarry Basin, Western
Australia. Alcheringa 18:275–300
Grey K, Hickman AH, Hofmann HJ, Van Kranendonk
MJ, Williams I (1999) Pilbara Archaean Stromatolite Excur-
sion, Field Guide, June 29th–July 6th 1999. Geological
Survey of Western Australia, p 37
1348 N. McLoughlin et al.
Grotzinger JP, Kasting JF (1993) New constraints on
Precambrian ocean composition. J Geol 101:235–243
Grotzinger JP, Knoll AH (1999) Stromatolites in Precam-
brian carbonates; evolutionary mileposts or environmental
dipsticks? Annu Rev Earth Planet Sci 27:313–358
Grotzinger JP, Rothman DH (1996) An abiotic model for
stromatolite morphogenesis. Nature 383:423–425
Hanski E, Huhma H (2005) Central Lapland Greenstone
Belt. In: Lehtinen M, Nurmi PA, R€am€o OT (eds) Precam-
brian geology of Finland: key to the evolution of the
Fennoscandian shield. Elsevier, Amsterdam, pp 139–193
Hofmann HJ (1977) On Aphebian stromatolites and
Riphean stromatolite stratigraphy. Precambrian Res
5:175–205
Hofmann HJ (2000) Archaean stromatolites as microbial
archives. In: Riding RE, Awramik SM (eds) Microbial
sediments. Springer, Berlin/Heildelberg, pp 315–327
Hofmann HJ, Jackson GD (1987) Proterozoic mini-
stromatolites with radial-fibrous fabric. Sedimentology 34:
963–971
Hofmann HJ, Grey K, Hickman AH, Thorpe RI (1999)
Origin of 3.45 Ga coniform stromatolites in Warrawoona
Group, Western Australia. Geol Soc Am Bull 111:1256–126
Jogi PM, Runnegar B (2005) Quantitative methods for
evaluating the biogenicity of fossil stromatolites. Astrobiol-
ogy 5:293
Kalkowsky E (1908) Oolith und stromatolilith in
norddeutschen Buntsandstein. Z Dtsch Geol Ges 60:68–125
Karhu JA (2005) Paleoproterozoic carbon isotope excur-
sion. In: Lehtinen M, Nurmi PA, R€am€o OT (eds) Precam-
brian geology of Finland: key to the evolution of the
Fennoscandian shield. Elsevier, Amsterdam, pp 669–680
Karhu J, Kortelainen NM, Huhma H, Perttunen V, Sergeev
SS (2008) The end of the Paleoproterozoic carbon isotope
excursion: new time constraints, abstract, 33rd geological
congress, 6–14 Aug 2008, Oslo
Knoll AH, Swett K, Burkhardt E (1989) Paleoenvir-
onmental distribution of microfossils and stromatolites
from the Upper Proterozoic Backlundtoppen Formation,
Spitsbergen. J Paleontol 63:129–145
Korsman K, Korja T, Pajunen M, Virransalo P, GGT/
SVEKAWorking Group (1999) The GGT/SVEKA transect:
structure and evolution of the continental crust in the
Paleoproterozoic Svecofennian orogen in Finland. Int Geol
Rev 41:287–333
Krumbein WE (1983) Stromatolites: the challenge of a
term in space and time. Precambrian Res 20:493–531
Krumbein WE, Cohen Y, Shilo M (1977) Stromatolitic
cyanobacterial mats. Limnol Oceanogr 22:635–656
Kuznetsov AB, Melezhik VA, Gorokhov IM, Melnikov
NN, Konstantinova GV, Kutyavin EP, Turchenko TL
(2010) Sr isotopic composition of Paleoproterozoic 13C-
richcarbonate rocks: the Tulomozero Formation, the south-
eastern Fennoscandian Shield. Precambrian Res 182:
300–312
Lager I, Loberg B (1990) Sedimentological and basin
analysis ore prospecting methods in the North Bothian
greenstone belt. STU-Project 86-03967P, Final Report. Eco-
nomic geology, Lulea Technical University, Lulea, p 112 (in
Swedish)
Lahtinen R, Garde AA, Melezhik VA (2008) Paleopro-
terozoic evolution of Fennoscandia and Greenland. Episodes
31:20–28
Lindsay JF, Brasier MD (2002) Did global tectonics drive
biosphere evolution? Carbon isotope record from 2.6 to 1.9
Ga carbonates of Western Australia. Precambrian Res
114:1–34
Link MH, Osborne RH, Awramik SM (1978) Lacustrine
stromatolites and associated sediments of the Pliocene Ridge
Route Formation, Ridge Basin, California. J Sediment Petrol
48:143–158
Love KM, Chafetz HS (1988) Diagenesis of laminated
travertine crusts, Arbuckle Mountains, Oklahoma. J Sedi-
ment Res 58:441–445
Lowe DR (1994) Abiological origin of described
stromatolites older than 3.2 Ga. Geology 22:387–390
Lybtsov VV (1979) Stromatolites of the Palaeopro-
terozoic Pechenga Complex, Kola Peninsula. Trans USSR
Acad Sci 247:419–421 (in Russian)
Makarikhin VV (1992) Lower Precambrian stromatolite
associations of Karelia. In: Schidlowski M, Golubic S,
Kimberley MM, Trudinger PA (eds) Early organic evolu-
tion: implications for mineral and energy resources.
Springer, Berlin, pp 463–467
Makarikhin VV, Kononova GM (1983) Early Proterozoic
phytolithes of Karelia. Nauka, Leningrad, p 180 (in Russian)
Maslov VP (1960) Stromatolites (genesis, study methods,
facies relation and geological significance). Proc Geol Inst
USSR Acad Sci 41:188 (in Russian)
McLoughlin N, Wilson LA, Brasier MD (2008) Growth
of synthetic stromatolites and wrinkle structures in the
absence of microbes – implications for the early fossil
record. Geobiology 6:95–105
Medvedev P, Bekker A, Karhu JA, Kortelainen N (2005)
Testing the biostratigraphic potential of early Paleopro-
terozoic microdigitate stromatolites. Rev Esp Micro-
paleontol 37:41–56
Melezhik VA, Fallick AE (2001) Palaeoproterozoic
travertines of volcanic affiliation from a 13C-richrift lake
environment. Chem Geol 173:293–312
Melezhik VA, Fallick AE (2005) Palaeoproterozoic, rift-
related, 13C-rich lacustrine carbonates, NW Russia, part I:
8 7.8 Traces of Life 1349
sedimentology and major element geochemistry. Trans R
Soc Edin Earth Sci 95:393–421
Melezhik VA, Fallick AE (2010) On the Lomagundi-
Jatuli carbon isotopic event: the evidence from the Kalix
Greenstone Belt, Sweden. Precambrian Res 179:165–190
Melezhik VA, Fallick AE, Makarikhin VV, Lyubtsov VV
(1997) Links between Palaeoproterozoic palaeogeography
and rise and decline of stromatolites: Fennoscandian Shield.
Precambrian Res 82:311–348
Melezhik VA, Fallick AE, Medvedev PV, Makarikhin
VV (1999) Extreme 13Ccarb enrichment in ca. 2.0 Ga magne-
site-stromatolite-dolomite-‘red-bed’ association in a global
context: a case for the world-wide signal enhanced by a local
environment. Earth Sci Rev 48:71–120
Melezhik VA, Fallick AE, Medvedev PV, Makarikhin VV
(2000) Palaeoproterozoic magnesite–stromatolite–dolostone–
‘red bed’ association, Russian Karelia: palaeoenvironmental
constraints on the 2.0 Ga positive carbon isotope shift. Nor
Geol Tidsskr 80:163–186
Melezhik VA, Fallick AE, Grillo SM (2004) Subaerial
exposure surfaces in a Palaeoproterozoic 13C-richdolostone
sequence from the Pechenga Greenstone Belt: palaeoenvir-
onmental and isotopic implications for the 2330–2060 Ma
global isotope excursion of 13C/12C. Precambrian Res
133:75–103
Melezhik VA, Fallick AE, Kuznetsov AB (2005a)
Palaeoproterozoic, rift-related, 13 C-rich, lacustrine
carbonates, NW Russia, part II: global isotope signal
recorded in the lacustrine dolostones. Trans R Soc Edin
Earth Sci 95:423–444
Melezhik VA, Fallick AE, Rychanchik DV, Kuznetsov
AB (2005b) Palaeoproterozoic evaporites in Fennoscandia:
implications for seawater sulphate, the rise of atmospheric
oxygen and local amplification of the d13C excursion. Terra
Nova 17:141–148
Melezhik VA, Huhma H, Condon DJ, Fallick AE,
Whitehouse MJ (2007) Temporal constraints on the Paleo-
proterozoic Lomagundi-Jatuli carbon isotopic event. Geol-
ogy 35:655–658
Monty C (1972) Recent algal stromatolitic deposits,
Andros Island, Bahamas. Preliminary report. Geol Rundsch
61:742–783
Ovchinnikova GV, Kuznetsov AB, Melezhik VA,
Gorokhov IM, Vasil’eva IM, Gorokhovskii BM (2007) Pb-
Pb age of Jatulian carbonate rocks: the Tulomozero Forma-
tion of southeastern Karelia. Stratigr Geol Correl
15:359–372 (in Russian)
Pentecost A, Spiro B (1990) Stable carbon and oxygen
isotope composition of calcites associated with modern
freshwater cyanobacteria and algae. Geomicrobiol J 8:17–26
Pia J 1927, Thallophyta, (1927) Thallophyta. In: Hirmer
M (ed) Handbuch der Pal€aobotanik, 1. R. Oldenbourg,
Munich, pp 31–136
Pope MC, Grotzinger JP (2000) Controls on fabric
development and morphology of tufas and stromatolites,
uppermost Pethei group 1.8 Ga, Great Slave Lake, NW
Canada. In: Carbonate sedimentation and diagenesis in
the evolving Precambrian world, vol 67, Society of econo-
mic paleontologists and mineralogists, special publication.
SEPM (Society for Sedimentary Geology), Tulsa,
Oaklahoma, pp 103–121
Pope MC, Grotzinger JP, Schreiber BC (2000) Evaporitic
subtidal stromatolites produced by in situ precipitation:
textures, facies associations, and temporal significance. J
Sediment Res 70:1139–1151
Preiss WV (1976) Basic field and laboratory methods
for the study of stromatolites. In: Walter MR (ed)
Stromatolites, vol 20, Developments in sedimentology.
Elsevier, Amsterdam, pp 5–13
Reid RP, Visscher PT, Decho AW, Stolz JF, Bebout BM,
Dupraz C, Macintyre IG, Paerl HW, Pinckney JL, Prufert-
Bebout L, Steppe TF, Des Marais DJ (2000) The role of
microbes in accretion, lamination and early lithification of
modern marine stromatolites. Nature 406:989–992
Reid PR, James NP, Macintyre IG, Dupraz CP, Buren RV
(2003) Shark Bay stromatolites: microfabrics and reinter-
pretation of origins. Facies 49:299–324
Riding R (2000) Microbial carbonates: the geological
record of calcified bacterial-algal mats and biofilms. Sedi-
mentology 47(Suppl 1):179–214
Riding R (2008) Abiogenic, microbial and hybrid
authigenic carbonate crusts: components of Precambrian
stromatolites. Geol Croatica 61(2–3):73–103
Semikhatov MA (1976) Experience in stromatolite stud-
ies in the USSR. In: Walter MR (ed) Stromatolites, vol 20,
Developments in sedimentology. Elsevier, Amsterdam,
pp 273–310
Semikhatov MA, Raaben ME (1994) Dynamics of the
global diversity of Proterozoic stromatolites. Article I:
Northern Eurasia, China and India. Stratigr Geol Correl 2
(6):492–513
Semikhatov MA, Raaben ME (1996) Dynamics of the
global diversity of Proterozoic stromatolites. Article II:
Africa, Australia, Northern America and general synthesis.
Stratigr Geol Correl 4:24–50
Semikhatov MA, Raaben ME (2000) Proterozoic stro-
matolite taxonomy and biostratigraphy. In: Riding RE,
Awramik SM (eds) Microbial sediments. Springer, Berlin,
pp 295–306
Semikhatov MA, Gebelein CD, Cloud P, Awramik SM,
Benmore WC (1979) Stromatolite morphogenesis: progress
and problems. Can J Earth Sci 16:992–1015
Seong-Joo L, Goulbic S (1999) Microfossil populations in
the context of synsedimentary micrite deposition and acicu-
lar carbonate precipitation: Mesoproterozoic Gaoyuzhuang
Formation, China. Precambrian Res 96:183–208
1350 N. McLoughlin et al.
Shepard RN, Sumner DY (2010) Unidirected motility of
filamentous cyanobacteria produces reticulate mats.
Geobiology 8:179–190
Simonson BM, Carney KE (1999) Roll-up structures:
evidence of situ microbial mats in Late Archaean deep
shelf environments. Palaios 14:13–24
Simonson BM, Schubel KA, Hassler SW (1993) Carbon-
ate sedimentology of the early Precambrian Hamersley
Group of Western Australia. Precambrian Res 60:287–335
Sokolov VA (1963) Geology and Lithology of the Early
Proterozoic Carbonate rocks of Karelia. Nauka (Science),
Moscow, p 185 (in Russian)
Sokolov VA (1987) The Jatulian Superhorizon. In:
Sokolov VA (ed) The geology of Karelia. Nauka (Science),
Leningrad, pp 51–59 (in Russian)
Strand K, Laajoki K (1999) Application of the
parasequence concept to the Paleoproterozoic record of the
northern Fennoscandian Shield. PrecambrianRes 97:253–267
Sumner DY (1997) Late Archaean calcite-microbe
interactions: two morphologically distinct microbial
communities that affected calcite nucleation differently.
Palaios 12:300–316
Talbot MR, Allen PA (1996) Lakes. In: Reading H (ed)
Sedimentary environments: processes, facies, and stratigra-
phy. Blackwell Science, Oxford, pp 83–124
Tice MM, Lowe DR (2004) Photosynthetic microbial
mats in the 3416-Myr-old ocean. Nature 431:549–552
Turner EC, James NP, Narbonne GM (2000a) Tapho-
nomic control on microstructure in early Neoproterozoic
reefal stromatolites and thrombolites. Palaios 15:87–111
Turner EC, Narbonne GM, James NP (2000b) Frame-
work composition of early NeoProterozoic calcimicrobial
reefs and associated microbialites, Mackenzie Mountains,
N.W.T., Canada. In: Grotzinger JP, James NP (eds) Carbon-
ate sedimentation and diagenesis in the evolving Precam-
brian world, vol 67, Society of economic paleontologists and
mineralogists, special publication. SEPM (Society for Sedi-
mentary Geology), Tulsa, Oklahoma, pp 103–121
Vanyo JP, Awramik SM (1985) Stromatolites and earth-
sun-moon dynamics. Precambrian Res 2:121–142
Wacey D (2009) Early life on Earth: a practical guide.
Springer Science + Business Media B.V, Dordrecht,
p 274
Walter MR, Bauld J, Brock TD (1976) Microbiology and
morphogenesis of columnar stromatolites (Conophyton,Vacerrilla) from hot springs in Yellowstone National Park.
In: Walter MR (ed) Stromatolites, vol 20, Developments in
sedimentology. Elsevier, Amsterdam/Oxford/New York,
pp 273–310
Wanke A, Melezhik VA (2005) Palaeoproterozoic sedi-
mentation and stromatolite growth in an advanced intracon-
tinental rift associated with the marine realm: a record of
the Neoarchaean continent breakup? Precambrian Res
140:1–35
Wilks ME, Nisbet EG (1985) Archaean stromatolites
from the Steep Rock Group, northwestern Ontario, Canada.
Can J Earth Sci 22:792–799
Woodhead JD, Hergt JM (1997) Application of ‘double
spike’ technique to Pb-isotope geochronology. Chem Geol
138:311–321
8 7.8 Traces of Life 1351
7.8.3 Palaeoproterozoic Microfossils
Emmanuelle J. Javaux, Kevin Lepot, Mark vanZuilen, Victor A. Melezhik, andPavel V. Medvedev
The Palaeoproterozoic Microfossil Record
The Palaeoproterozoic (2.5–1.6 Ga) era is a crucial time in
Earth history. Of particular importance with respect to the
evolution of life is the history of Earth’s oxygenation. Numer-
ous geochemical tracers, i.e. concentrations and/or isotopes of
redox-sensitive chemical elements, have been applied to sedi-
mentary rock successions straddling this important time win-
dow in order to constrain the appearance of free oxygen in
the atmosphere–ocean system. Time-series data for multiple
sulphur isotopes from carbonate-associated sulphates and
sulphides capture the loss of atmospheric mass-independent
sulphur isotope fractionation and concomitant increase of
marine sulphate reservoir as a sign of the first significant rise
in atmospheric oxygen on Earth in between two glacial units of
the Duitschland Formation, Transvaal Supergroup, South
Africa (Guo et al. 2009). This transition from a largely anoxic
to an aerobic world can be constrained between c. 2.48 and
2.32 Ga (Bekker et al. 2004; Hannah et al. 2004). Other
geochemical tracers suggest the appearance of free oxygen in
Earth’s atmosphere and/or surface ocean water columns even
earlier than 2.5–2.45Ga ago (Kaufman et al. 2007; Anbar et al.
2007; Rosing and Frei 2004; Nisbet et al. 2007; Buick 2008;
Ono et al. 2006). However, while geochemical tracers suggest
the appearance of atmospheric oxygen in the late Archaean,
they only provide a minimum age for the advent of oxygenic
photosynthesis. Previously advocated biomarker evidence for
the presence of cyanobacteria and eukaryotes at 2.7 Ga
(Summons et al. 1999; Brocks et al. 2003) was recently
reassessed as representing younger contaminants (Rasmussen
et al. 2008). Despite an unconstrained point in time for the
onset of oxygenic photosynthesis, it is clear that a significant
rise in atmospheric oxygen must have had profound con-
sequences for ocean chemistry and biology, and might have
opened new ecological niches for the diversifying biosphere.
The oldest microfossils diagnostic of cyanobacteria and of
eukaryotes have been found in Palaeoproterozoic rocks where
they are preserved most commonly in three dimensions in
cherts or flattened in two dimensions in shales (Javaux
and Benzerara 2009). In addition to microfossils, the
Palaeoproterozoic record of biological activities also includes
stromatolites, biomarkers, and isotopic fractionation of C and
S (discussed elsewhere in this volume), microbially induced
sedimentary structures in siliciclastics (“MISS”), macroscopic
carbonaceous compressions, mat rip-ups, and putative fossil
or trail impressions (e.g. Knoll 2003). The oldest unambigu-
ously identified microfossils of cyanobacteria occur in sedi-
mentary rocks from the Belcher Islands, Canada, dated at
1.9Ga. TheBelcherGroup comprises chert lenses and nodules
in silicified stromatolites growing in tidal and shallow subtidal
waters on a carbonate platform (Hofmann 1976; Golubic and
Hofmann 1976). The cherts contain three-dimensionally pre-
served filamentous and coccoidal (spheroidal) microfossils,
including fossilised colonies of microscopic pigmented cells.
The distribution and pattern of division of these later
microfossils (Eoentophysallis belcherensis) (Fig. 7.115a)
(colonies of coccoidal cells dividing by binary fission in
three planes inside preserved external envelopes and produc-
ing brown pigments at colony surfaces) permit relating them
to the living genera of cyanobacteria Entophysallis.
Eoentophysallis has been recorded in other Palaeoproterozoicand younger successions (Knoll and Golubic 1992). These
microfossils, together with sausage-shaped cysts (akinetes)
of cyanobacteria (Archaeoellipsoides) from the Francevillian
Series in Gabon dated at c. 2.1 Ga (Amard and Bertrand-
Safarti 1997), represent some of the oldest remains of
identified cyanobacteria. They demonstrate the diversification
of modern taxa of cyanobacteria in oxygenated carbonate
platforms during the Palaeoproterozoic.
Different communities contemporaneous of the Belcher
assemblages were discovered in stromatolitic and non-
stromatolitic cherts of the 1.88GaGunflint Formation,Ontario,
Canada (Barghoorn and Tyler 1965; Awramik and Barghorn
1977; Fralick et al. 2002). These illustrate the diversification of
the biosphere in middle Palaeoproterozoic oceans, with pro-
karyotic taxa being difficult to relate definitely to extant
organisms. No eukaryotes are unambiguously recognised,
although some taxa found in non-stromatolitic cherts like
Eosphaera (vesicles surrounded by a layer of smaller coccoids
in an envelope) and Leptotrichos (5–30 mm solitary coccoids)
could be prokaryotic or eukaryotic (Knoll 1996). In the stro-
matolitic cherts (Fig. 7.115b), the assemblage contains star-like
forms (Eoastrion), coccoids (Hurioniospora), problematic
forms (Kakabekia umbellata), filaments (Gunflintia), and bud-
ding tubular forms (Archaeorestis).Similar associations are known in the c. 2.1 Ga
Francevillian biota, Gabon (Amard and Bertrand-Sarfati
1976), the Balbirini Dolomite, Australia (Oehler 1977), the
1.8 Ga Duck Creek Dolomite, Western Australia (Knoll
et al. 1988), the Tyler Formation, Michigan (Cloud and
Morrison 1980), in the 1.7 Ga Frere Formation, Australia
(Walter et al. 1976), and in other coeval iron formations and
E.J. Javaux (*)
Department of Geology, University of Liege, 17 allee du 6 Aout B18,
4000 Liege, Belgium
1352 E.J. Javaux et al.
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_8, # Springer-Verlag Berlin Heidelberg 2013
1352
subtidal carbonates worldwide. These assemblages were
widely distributed in shallow habitats where iron-rich deep
waters mixed with oxygenated surface waters, before the
proposed late Palaeoproterozoic expansion of sulphidic sub-
surface waters (e.g., Shen et al. 2002). Recent investigations
of the 1.9 Ga Gunflint Formation (Planavsky et al. 2009) and
the slightly younger 1.8 Ga Duck Creek Formation, Western
Australia (Wilson et al. 2010) confirmed previous micro-
palaeontological investigations and provided geochemical
evidence supporting the presence of iron-oxidising bacteria
in these early ecosystems, representing an iron-driven com-
munity distinct from the photosynthetically dominated
assemblages found in shallow environments such as the
Belcher Supergroup.
Empty filamentous sheaths (e.g., species of the genera
Oscilliatoriopsis, Siphonophycus, Eomycetopsis, Tenuo-
filum, Taeniatum, Gunflintia) are abundant in most Protero-
zoic microbial mat assemblages and range from <1 mmto >10 mm (or >100 mm) in diameter (Knoll and Golubic
1992). The large sheaths are possibly attributed to
cyanobacteria, especially when they occur in shallow-water
sediments deposited in the photic zone and when their distri-
bution shows phototactism.When water chemistry of ancient
sedimentary environments is not well characterised, it is
difficult to distinguish these large sheaths from those of
Beggiatoa-like sulfur-oxidizing bacteria forming mats at
the sediment-water interface with a steep redox gradient
(Knoll and Golubic 1992). The narrow sheaths could also
represent the remains of Chloroflexus-type photosynthetic
bacteria, or to Leptothrix-type iron bacteria (such as
Gunflintia minuta, which grew in an iron-rich stromatolitic
environment of the Gunflint Formation) (Knoll and Golubic
1992). Similar uncertainty applies to other colonies of
microfossils, such as the possibly heterotrophic rod-shaped
Eosynechococcus or the coccoidal Myxococcoides, in the
absence of palaeoenvironmental information (Knoll and
Golubic 1992). The oldest unambiguous eukaryotic
microfossils are large organic-walled vesicles with striated
walls (Valeria lophostriata) (Fig. 7.115c, d). Scanning Elec-tron Microscopy shows that these striations are 1-mm-apart
ridges on the inner side of the vesicle (Javaux et al. 2004)
(Fig. 7.115d). They occur in the 1.8 Ga Chuanlinggou For-
mation (Chengcheng Supergroup, China) (Zhang 1986; Peng
et al. 2009; E. Javaux, pers. obs.), in the 1.65 GaMallapunyah
Formation (McArthur Supergroup, Australia; Javaux 2006),
and in most younger Proterozoic siliciclastic successions. In
Palaeoproterozoic and younger rocks, large, smooth,
organic-walled vesicles (up to a few 100 mm) are common
and may display a medial split suggestive of excystment
structures but the absence of any wall ornamentation
prevents their attribution to the eukaryotic domain. Ongoing
investigations of old drill cores and outcrop samples of
Palaeoproterozoic successions in Karelia, Russia, by P.
Medvedev and E. Javaux have led to the discovery of rare
large (up to>300 mm) carbonaceous vesicles (acritarchs) and
fragments of organic sheaths preserved in siltstones from the
upper members of the ~1.9 Ga Kondopoga Formation
(Fig. 7.115e, f). Poorly preserved and illustrated acritarchs
were first mentioned by Timofeev (1982). They could repre-
sent early protists or large prokaryotes such as cyanobacterial
colonial envelopes. Detailed studies of the wall ultrastructure
and chemistry may permit their identification in some cases
(Javaux et al. 2003, 2004; Javaux and Marshall 2006).
Among macroscopic carbonaceous compressions, the
coiled filaments Grypania spiralis from the 1.87 Ga
Negaunee Iron Formation, Michigan (Han and Runnegar
1992; redated by Schneider et al. 2002), have a diameter
up to 30 mm (across the whole coil). Samuelsson and
Butterfield (2001) have questioned the eukaryotic nature of
these structures, but observations by A. Knoll (in Knoll et al.
2006) suggest that it was a single organism and not a colony
or composite of much smaller prokaryotic filaments. While
the eukaryotic affinity of the Palaeoproterozoic Canadian
Grypania is still problematic (although no extant prokaryotic
filamentous sheath are known to reach such sizes), the
Mesoproterozoic Grypania from India show septa, terminal
cells, spiral nature and large size (up to 3.2 cm) suggesting
an eukaryotic affinity (or hypothetic gigantic cyanobacteria;
Kumar 1995). Recent reassessment of Palaeo- and Mesopro-
terozoic macroscopic coiled filamentous compressions
suggests a cyanobacterial affinity for some of them, while
others might be dubiofossils and more complex “tissue
grade” organisms (Sharma and Shukla 2009), or macroalgae
(Xiao and Dong 2006). Bedding-plane structures in
sandstones fromMontana and Western Australia (Horodyski
1982; Grey and Williams 1990; Yochelson and Fedonkin
2000) called Horodyskia moniliformis consist of 1–4-mm-
sized, spheroidal bodies connected by thin cylindrical strings
to form uniseriate, pearl-necklace-like structures up to 10 cm
long. These structures have been compared to seaweeds,
colonial metazoans, prokaryotic association, or non-
biological structures. Other traces consisting of U-shaped
ridges in sandstone of the 2.0–1.8 Ga Stirling Range Forma-
tion, Australia, first interpreted as animal traces (Bengtson
et al. 2007), could be produced by giant amoebae as
observed in recent sediments (Matz et al. 2008; Pawlowski
and Gooday 2008; Bengtson and Rasmussen 2009).
Recently, intriguing pyritic structures of up to 12 cm in
length and showing a central folded part and a radially
divided fringe have been discovered in the 2.1 Ga
Francevillian of Gabon (El Albani et al. 2010). Although
they may resemble mineral concretions at first, details of
their complex morphology, C and S isotopic chemistry
patterns, internal texture, and distribution in the hosting
black shale permit to interpret them as pyritised fossil
macrostructures. Unlike mineral concretions (such as “pyrite
flowers” and “frondescent pattern of pyrite discs”; Seilacher
2001, p. 50–52), they do not show a “fibrous cone-in-cone
structure” of pyritic layers radiating symmetrically from a
midline (typical of concretions growing completely or
8 7.8 Traces of Life 1353
partially in stiff mud) (Figs. 10 and 11 in Seilacher 2001) and
they exhibit evidence of precompactional folding (El Albani
et al. 2010, and their Fig. S13); hence they are interpreted as
originally flexible organic sheets, which were subsequently
pyritised during early diagenesis (El Albani et al. 2010).
Their macroscopic size suggests that they could represent
colonial microorganisms, but questions remains regarding
their taphonomy and biology.
Collectively, the unambiguous Palaeoproterozoic micro-
fossil record illustrates biospheric evolution in transitional
ocean chemistry having moderately oxygenated surface
waters but with subsurface waters changing from ferrugi-
nous to sulfidic in the late Palaeoproterozoic (Canfield 1998;
Planavsky et al. 2009; Wilson et al. 2010). The fossil
assemblages include iron-loving and other undetermined
filamentous and coccoidal prokaryotes, stem group
eukaryotes, and modern taxa of cyanobacteria. The
emerging picture is one of a changing and more complex
biosphere, in which the three domains of life, Archaea,
Bacteria and Eukarya, were diversifying in various ecologi-
cal niches (Fig. 7.116) marked by the diversification of
stromatolites, increasing abundance of biomarkers and
appearance of macroscopic problematic fossils or traces.
Besides these traces of life described above, the Palaeopro-
terozoic record also includes problematic structures. Indeed,
most early microfossils display simple morphologies such as
coccoids or filaments, which can be mimicked by abiotic
processes, producing pseudofossils. For example, the c.
2.05 Ga Kolosjoki Volcanic Formation in the Pechenga
Greenstone Belt is composed of pillowed lavas with thin
beds of chert containing simple, three-dimensionally pre-
served solitary spheres, 3–7 mm in diameter, or their clusters,
resembling coccoidal microfossils (Fig. 7.117) (Ivanova et al.
1988) whose biogenicity remains to be proven.
Before a microstructure can be accepted as a microfossil,
a series of techniques and multidisciplinary approaches need
to be employed to prove its endogenicity, syngenicity, and
biological origin, as well as to falsify an abiotic explanation
for the observed morphologies or chemistries. The next
paragraphs describe in situ analytical methods that can be
used to characterise possible microfossils. These micro to
nano-scale approaches should be complemented by macro-
scale observations and characterisation of the geological
context, as the environmental conditions will determine
the plausibility of ancient habitats and the conditions of
fossilisation.
Methods and Problems of Identification ofBiogenicity, Endogenicity and Syngenicity
IntroductionThe simple morphologies of microfossils in silicified and
unsilicified Proterozoic rocks possess a limited number of
attributes available for taxonomic characterisation. Many
characteristics of cell cultures can be modified during post-
mortem degradation. Elevated temperature, pressure and
strain can cause structures to flatten and ultimately loose
their original three-dimensional shape (Schopf and Klein
1992). The organic cell wall slowly converts to kerogen or
graphite (Fig. 7.118). This combination of morphologic and
chemical transformations that must have affected remnants of
life in ancient metamorphosed rocks has led to manymisinter-
pretations and to classifications such as ‘pseudofossil’ or
‘dubiofossil’ (Hofmann 2004). Typical artifacts that can
resemble microfossils include ambient inclusion trails
(Fig. 7.119a), self-assembled mineral nanostructures
(Fig. 7.119b–d), limonite-stained fluid inclusions, oil-filled
fluid inclusions, cavities, and most importantly extant endo-
lithic micro-organisms (Fig. 7.119e). For unambiguous rec-
ognition of a microfossil two crucial questions have to be
answered: (1) is the structure syngenetic (did it form when the
rock formed), or is it epigenetic (the result of a secondary
process such as hydrothermal fluid flow), and (2) is the struc-
ture indigenous (is it actually part of the rock) or allochtonous
(e.g. extant organisms colonising this rock)? Then the prob-
lem of biogenicity can be addressed.
Several in situ analytical techniques are now available
that enable the detailed micron-scale structural, isotopic and
chemical description of putative microfossils (Fig. 7.120).
Although these developments have greatly expanded and
improved this field of research, it is important to realize
that these techniques do not necessarily answer the question
of biogenicity. Furthermore, several of these techniques
have introduced a suite of analytical artefacts that have led
to misinterpretations. Here a brief account is given of some
in situ techniques, their use for microfossil characterisation
and their problems and potential pitfalls.
Raman SpectroscopyRaman spectroscopy is a highly effective technique for
identification of indigenous organic microstructures in
Archaean terrains and for ruling out opaque mineral
inclusions and post-metamorphic organic contamination
(Allwood et al. 2006; Tice et al. 2004; van Zuilen et al.
2006). It can be used effectively to determine the degree of
structural order in carbonaceous material (kerogen,
pyrobitumen, Fig. 7.121a, b). Inferences on metamorphic
grade can only be made, however, if proper care has been
taken to avoid alteration of the carbonaceous structure dur-
ing sample preparation. Polishing or crushing will cause
dislocations, and breakup of the crystal structure, leading
to a disordered Raman spectrum (Pasteris 1989). This prob-
lem can be circumvented by direct analysis of a cracked rock
surface (Tice et al. 2004), or by carefully pressing carbona-
ceous matter in a gold foil (Wopenka and Pasteris 1993).
Alternatively, carbonaceous matter in thin sections can be
studied by focusing the laser beam on the subsurface
1354 E.J. Javaux et al.
continuations of polished grains, or focus below the polished
surface on grains of interest that are fully embedded in
surrounding transparent mineral phases (Fig. 7.121c–e). If
these precautions are taken into account, carbonaceous
microfossils can effectively be visualized by hyperspectral
Raman mapping (Kudryavtsev et al. 2001; Schopf and
Kudryavtsev 2005; Schopf et al. 2002). It must be stressed,
however, that the Raman spectrum can also be derived from
abiologic forms of carbon such as graphitic coatings of fluid
inclusions (Pasteris and Wopenka 2002, 2003). It is there-
fore of crucial importance to combine Raman spectroscopy,
which cannot prove biogenicity, with other in situ tech-
niques that enable the morphological, ultrastructural, and
chemical characterisation of individual microstructures,
and with a good understanding of the environmental
conditions of their preservation and their taphonomy.
Transmission Electron Microscopy (TEM) andAnalytical TEM (ATEM)Transmission electron microscopy (TEM) and analytical
TEM (ATEM) permits to study in situ the style of perminer-
alisation (such as silicification), and the distribution, struc-
ture and evolution of organic matter (nm to mm-scale)
(e.g. Moreau and Sharp 2004; Lepot et al. 2009b). These
techniques require delicate preparation of samples, either by
embedding the sample in a resin and by ultra-thin sectioning
with a diamond knife, or by producing FIB sections
(Figs. 7.120 and 7.122a, b). Elemental analyses and elemen-
tal mapping (e.g. of organic C or S) with a resolution better
than 10 nm by Energy Dispersive X-ray Spectroscopy
(EDXS) can be coupled to the TEM observations (Lepot
et al. 2009b). Once in situ analyses have demonstrated the
endogenicity and syngenicity of potential carbonaceous
microfossils, other techniques can be used on extracted
isolated microfossils to investigate their taphonomy and
their biogenicity. Transmission electron microscopy (TEM)
can be used to differentiate kerogenous particles from
kerogenous hollow flattened vesicles (fossil cells or colonial
envelopes) and to reveal details of cell wall ultrastructure (e.
g. Javaux et al. 2004, 2010).
Secondary Ion Mass Spectrometry (SIMS)Secondary Ion Mass Spectrometry (SIMS) utilizes a Cs+
beam to ablate small 10–50 mm pits in polished rock
samples, and is therefore highly suitable for in situ carbon
isotope analysis of organic structures such as microfossils,
pyrobitumen, and kerogen (House et al. 2000; Rasmussen
et al. 2008; van Zuilen et al. 2006). An additional application
for microfossil research is the capability to construct 2-D
maps of heteroatom variability within a single microfossil
structure. A NanoSIMS instrument can reach a Cs+ beam
resolution of 0.05 mm, and therefore enables sub-micron
resolution elemental mapping. This technique has been
explored for the construction of elemental maps of C, N, S,
and C/N-ratio of individual microfossil structures (Oehler
et al. 2006, 2009). Several important problems have to be
overcome, however, in order to properly interpret SIMS ion
probe results. One of the most important is that ionization
efficiency depends on the structure of the material being
analyzed (matrix effect). This means that precise isotope
ratios or elemental concentrations can only be determined
if the measurement is compared to that of a standard material
which has the same structural characteristics as the sample.
Although some systematic studies on carbonaceous matter
have been carried out (Aleon et al. 2003; Sangely et al. 2005;
van Zuilen et al. 2006), more work is needed to precisely
characterise subtle, small-scale differences in isotopic and
elemental variation of various carbonaceous materials.
Synchrotron-Based Techniques (STXM, NEXAFS)Synchrotron-based X-ray analysis is a high-resolution (nm
to mm-scale), non-destructive technique that is well adapted
for studying the structure of minute and delicate organic
objects trapped in a mineralized matrix. Organic micro-
fossils in metamorphosed rocks consist of kerogen compris-
ing aggregates of individual graphene layers that are poorly
aligned, variously oriented, and disrupted by dislocations.
This disorder is in part the result of complex biologic pre-
cursor material that has introduced five or seven-member
rings in the graphene structure, as well as certain highly
resistant functional groups containing H and O and S.
These nano-scale characteristics of a carbonaceous material
can be effectively studied with synchrotron-based tech-
niques such as X-ray Absorption Near-Edge Spectroscopy
(XANES) in a Scanning Transmission X-ray Microscope
(STXM). This technique can be described as a transmission
microscope using a monochromated X-ray beam produced
by synchrotron radiation. The XANES spectrum of pure
graphitic carbon shows an absorption threshold of the
transition 1 s to p* states at 285.1 eV and a second threshold
of transitions 1 s to s* states at 292.8 eV. Poorly ordered
graphene displays additional peaks in the in the 285–290 eV
range (Brandes et al. 2008; Gago et al. 2001). Lepot et al.
(2008, 2009b) used STXM and XANES spectroscopy at the
carbon K-edge (e.g. Fig. 7.122c) to identify aromatic carbon
(285.2 eV p*, C ¼ C bonds, 292.6 eV 1 s to s*), thiophenegroups (285.7 eV, p*, C ¼ C bonds), aliphatic carbon
(~288 eV, sigma*, C-H bond), carboxyl functional groups
(288.6 eV, pi*, C ¼ O bond), thiophene, ketone or phenol
groups (287.3 eV) and hydroxylated aliphatic carbons
(~289.6 eV) in organic globules that occur in stromatolite
structures of the 2.7 Ga Tumbiana Formation, Fortescue
Group, Western Australia. Bernard et al. (2007) used this
technique to identify ketone and phenolic groups in the outer
wall of a carbonaceous fossil lycophyte megaspore from a
Triassic metasediment, Vanoise massif, Western Alps,
France. These studies show the great potential for studying
the metamorphosed organic-rich sediments of the Pechenga
8 7.8 Traces of Life 1355
greenstone belt, and particularly the Karelian shungites. It is
important to note that these are transmission-based
techniques and require preparation of ultra-thin sections
(c. 100–150 nm) of sample material. This can be achieved
by Focused Ion Beam (FIB, Figs. 7.122a and 8b) milling
(Wirth 2009), which preserves the intimate association
between organic fossils and their host mineral matrix and
eventually associated biominerals.
Combining these in situ techniques provides an unprece-
dented powerful approach to study microfossils in Archaean-
Palaeoproterozoic rocks. It will be possible to link intricate
geochemical variations on a sub-micron scale to geologic
observations on a macroscopic scale. In the case of
stromatolites, it will be possible to link the chemistry of
individual laminae to the overall morphology and associated
sedimentary setting. In the case of microfossils, it will be
possible to link the individual microstructure to the
surrounding mineral assemblages, and exclude post-
metamorphic artefacts. In both cases, combining such small
scale variations will reveal a degree of complexity that is
unlikely to be produced by abiologic processes, and will
lead to identification of ancient biological processes.
Implication of the FAR-DEEP Cores
As detailed above, many questions remain to be answered as
to the affinity of Palaeoproterozoic microfossils, including
the identification of the oldest eukaryotes, of cyanobacteria,
of akinete-forming cyanobacteria thriving in oxygenated
environments, and the distinction of other photosynthetic
bacteria or other prokaryotes. Palaeoproterozoic
microfossils have often been observed in single and/or lim-
ited stratigraphic horizons, thus limiting biostratigraphy.
The FAR-DEEP core provides a unique opportunity to
search for and identify microfossils and to correlate the latter
with other geochemical tracers and environmental
constraints on a large time frame. Microfossils in such a
stratigraphic succession would have been submitted to rela-
tively similar metamorphic conditions. This could enable us
to derive chemical information on the original precursors of
distinct morphological microfossils without the interference
of extremely different thermal alteration pathways. In addi-
tion, the use of drillcore limits the problem of surface con-
tamination by endolithic microorganisms.
The FAR-DEEP drillholes intersected several geologi-
cal formations spanning the time interval from 2.5 to 2.0 Ga
(Fig. 7.116), a crucial time for the diversification of the
early biosphere. One of the several objectives of FAR-
DEEP is to discover and characterise new microfossil evi-
dence for early cyanobacteria and eukaryotes, especially in
shallow-water fine-grained siliciclastic rocks, cherts,
phosphorites, and sedimentary infill in stromatolitic build-
ups. In Fennoscandia, poorly-illustrated putative
microfossils were reported from several Palaeoproterozoic
formations by Timofeev (1982). Previous research
(Ivanova et al. 1988) has identified putative coccoidal
microfossils (Fig. 7.123) in the c. 2.05 Ga cherts hosted
by pillowed lavas of the Kolosjoki Volcanic Formation.
Preliminary investigations by P. Medvedev and E. Javaux
of lacustrine and turbiditic siliciclastic sedimentary rocks
from the upper part of the c. 1.9 Ga Kondopoga Formation
have revealed rare carbonaceous vesicles (acritarchs)
(Fig. 7.115e, f) and fragments of organic sheaths. All
these findings require further detailed work along two
lines: (1) confirming previously made discoveries, and (2)
proving their biogenicity with implementation of sophisti-
cated analyses.
The FAR-DEEP core contains abundant rock types rele-
vant for micropalaeontological investigations. The 2.4 Ga
Seidorechka Sedimentary Formation (Fig. 7.124a) contains
“grey shales” originally deposited in shallow-water shelf
environments that predate the Huronian glaciation and the
Great Oxidation Event. In the Huronian-age Polisarka Sedi-
mentary Formation, shallow-water marine shales are
associated with glacial deposits, and form thin interlayers
in marine limestones (Fig. 7.124b, c). The c. 2.2–2.06 Ga
formations, following the Great Oxidation Event and
Lomagundi-Jatuli positive isotope excursion of carbonate
carbon isotopes, contain “grey shales” accumulated in a
variety of settings ranging from lacustrine (Kuetsj€arvi
Sedimentary Formation) to deeper marine (Umba Sedimen-
tary Formation) (Fig. 7.124d, e). The c. 2.05–2.00 Ga period
of global enhanced accumulation rich in organic carbon is
represented by a variety of lithofacies potentially suitable for
micropalaeontological analyses and include lacustrine and
shallow-water marine shales from the lower part of the
Zaonega and Kolosjoki Sedimentary Formations, respec-
tively (Fig. 7.124f, g). Chert layers from the c. 2.05 Ga
Kolosjoki Volcanic Formation, from which putative
coccoidal microfossils were described (Fig. 7.123), have
been drilled and core is available for research. Several
hundreds of metres of core have been obtained from c.
2.0 Ga Zaonega Formation turbiditic greywackes and shales
rich in organic carbon and containing chert layers and
nodules, which represent other promising lithologies
(Fig. 7.124h–k) for searching microfossils associated with
an unprecedented accumulation of organic matter and a
large-scale generation of petroleum known as the Shunga
Event. Complementary material is available from several
previously made drillholes and outcrop samples from
quarries. Respective material comes from redeposited
phosphorites from the c. 2.0 Ga Pilguj€arvi Sedimentary
Formation (Fig. 7.124l) and from the c. 1.9 Ga Kondopoga
Formation comprising lacustrine turbiditic shales with
authochtonous kerogen matter as well as pyrobitumen
redeposited from surface oil seeps (Fig. 7.124m).
1356 E.J. Javaux et al.
Fig. 7.115 Examples of Palaeoproterozoic microfossils. (a) The mat-
building colony of the cyanobacteria Eoentophysallis belcherensis,preserved in cherty stromatolites of the 1.5 Ga Bil’yakh Group, Siberia.
(b) 2-mm-wide filaments (Gunflintia minuta), possible large
cyanobacterial filaments, and 15 mm coccoids (Hurioniospora) fromcherty stromatolites of the Gunflint Formation, Canada. (c–d) Valerialophostriata, a protist with a wall ornamented with concentric
striations, pictures showing a half enrolled vesicle (c) and details of
concentric striations (d), from shales of the 1.65 Ga Mallapunyah
Formation, Australia, and extracted by acid maceration. (e-f) organic-
walled microfossils (acritarchs) from siltstones of the c. 1.9 Ga
Kondopoga Formation, Karelia, Russia; extracted by acid maceration
(Photograph (a) courtesy of Andrew Knoll, (b) by Kevin Lepot, and
(c–f) by Emmanuelle Javaux)
8 7.8 Traces of Life 1357
Fig. 7.116 Summary of main biological and geological records in the Precambrian and stratigraphic range of the FAR-DEEP cores
Fig. 7.117 A 2.05 Ga chert of the Kolosjoki Volcanic Formation from the Pechenga Greenstone Belt containing a cluster of uniform, empty
1358 E.J. Javaux et al.
Fig. 7.118 Simplified scheme showing the transformations of bio-
logic material during metamorphism. Cell wall material and
geopolymers precipitated from various degraded molecules will con-
vert to insoluble macromolecular carbonaceous material (kerogen),
obtain a higher degree of structural order and progressively lose
heteroatoms such as H, O, N, and S, until they ultimately transform
into crystalline graphite
8 7.8 Traces of Life 1359
Fig. 7.119 Examples of microfossil artifacts. (a) Filaments formed by
abiotic dissolution of silica in the presence of organic inclusions, 2.7 Ga
Maddina Formation (Reprinted from (Lepot et al. 2009a) with permis-
sion of John Wiley and Sons). (b) Grapes of TiO2 spheres within
volcanic ash, 2.7 Ga Tumbiana Formation, photograph by Kevin
Lepot. (c) Helicoidal silica filaments (mimicking spirochete bacteria)
formed by liquid crystal growth (Sokolov and Kievsky 2005) (Scanning
Electron Microscopy (SEM) image courtesy of Igor Sokolov). (d) SEM
image of septate-like silica-carbonate filaments, reproduced from
(Garcia-Ruiz et al. 2003), reprinted with permission from AAAS. (e)
Fossilised allochtonous endolithic microorganisms on a grain bound-
ary, metachert, 3.8 Ga Isua Supracrustal Belt (Westall and Folk 2003)
(SEM image courtesy of Frances Westall)
1360 E.J. Javaux et al.
Fig. 7.120 Examples of in situ analytical techniques for microfossil
research. Raman spectroscopy can be applied directly on structures
such as stromatolites and microfossils. Other techniques such as
NanoSIMS require a polished surface. FIB (Focused ion Beam) prepa-
ration of a thin foil (15 � 10 um, 100 nm thick) is necessary for STXM
and TEM techniques
8 7.8 Traces of Life 1361
Fig. 7.121 (a) First-order Raman spectra of three structurally differ-
ent carbonaceous materials. In highly crystalline graphite a D-peak is
absent and therefore the D/G intensity ratio (R1) is zero. A shungite
sample from drill core 12B has a much smaller crystal domain size than
graphite, as is evident from the high R1 ratio. In highly disordered
materials such as coal, however, the D-band is less intense but broad-
ened and a second disorder band is present (D0). For this reason it has
been argued (Beyssac et al. 2002) that in metamorphosed carbonaceous
matter another ratio (R2) should be used based on the peak-areas of the
D, D0 and G-bands. (b) Schematic of crystal domain size (La) and the
two commonly used parameters R1 and R2 for determining degree of
order in carbonaceous material. (c) Raman hyperspectral map of
graphite in a polished thin section of a metacarbonate rock from the
Isua Supracrustal Belt, Greenland. Point A is a spectrum of disordered
graphite taken at the surface, point B a spectrum of well-ordered
graphite below a thin chlorite cover. (d) Raman map of the surface,
generated from the integral intensity band at ~1,600 cm�1, showing
three micro-areas in which graphite is exposed to the surface. (e)
Raman map recorded with the fully focused beam adjusted ~2 mmbelow the surface and generated from broadening of the ~1,600 cm�1
band. The areas that are covered by chlorite are well ordered (small
FWHMs), while the areas exposed to the surface are poorly ordered
(large FWHMs) (Figures (c, d) and (e) are reproduced from Nasdala
et al. (2004) with permission of Prof. Lutz Nasdala)
1362 E.J. Javaux et al.
Fig. 7.122 Nano-scale in situ extraction and STXM chemical analy-
sis. (a–b) Sample foil extraction by Focused Ion Beam (FIB) milling
technique (Images from Kevin Lepot). (a) Viewing plane parallel to the
sample showing the thickness of the foil. (b) Viewing plane at 45� of
the sample showing the surface of the foil. Organic matter appears in
dark grey (arrowed) within the mineral matrix. (c) XANES carbon
K-edge spectra of the carbonate matrix and of two organic pools
found in FIB sections of 2.7 Ga stromatolites recorded using STXM
(spectra from Kevin Lepot). Spectra indicate highly aromatic carbon in
both pools and sulfur- and oxygen- bearing functional groups only in
cell-like structures (Lepot et al. 2009b). Chemical structures shown in c
symbolise aromatic and thiophene (S-bearing) groups
8 7.8 Traces of Life 1363
Fig. 7.123 Putative coccoidal microfossils in 2.05 Ga cherts of the
Kolosjoki Volcanic Formation from the Pechenga Greenstone Belt. (a)
Several separated, uniform, semitransparent spheres showing a darker
colour with respect to that of the host crystalline quartz; the colouration
may be caused by the presence of “dusty” organic matter. (b) Clusters
of dark grey and black spheres; some spheres in the upper right cornerexhibit thick, black outer rims and somewhat lighter cores. (c) A cluster
of spheres, which all show darker outer rims. (d) A cluster of uniform,
empty, semitransparent spheres showing a darker colour with respect to
that of the host crystalline quartz; the cluster resembles a colony of
coccoidal microfossils. (e) Several solitary spheres with some showing
a complex, multispheroidal structure. (f) Clusters and solitary spheres
with thick, black outer rims (Photographs by Victor Melezhik)
1364 E.J. Javaux et al.
Fig. 7.124 Various shales and cherts retrieved by the FAR-DEEP
drillholes as well as redeposited phosphorites obtained from previous
drilling operations represent an attractive target for micropalaeontological
studies. Pre-Huronian time: (a) Interbedded sandstone-shale couplets
accumulated in a shallow-marine, tide-dominated environment in the c.
2.4 Ga Seidorechka Sedimentary Formation, Imandra/VarzugaGreenstone
Belt, Hole 1A, depth of 167 m. Huronian time: (b) Rhythmically
interbedded sandstone-shale couplets resembling “varved” sediment
associated with glacial lithofacies from the Polisarka Sedimentary
Formation in the Imandra/Varzuga Greenstone Belt; Hole 3A, depth of
194.3 m. (c) Interbedded shale and limestone associated with glacial
sediments from the Polisarka Sedimentary Formation; Hole 3A, depth of
221 m. Great Oxidation Event and the Lomagundi-Jatuli Event time: (d)Interbedded sandstone-siltstone-shale deposited in a lacustrine environ-
ment from the c. 2.06 Ga Kuetsj€arvi Sedimentary Formation in the
Pechenga Greenstone Belt; Hole 5A, depth of 141 m. (e) Thinly laminated
marine siltstone-shale deposited in theUmbaSedimentary Formation in the
Imandra/Varzuga Greenstone Belt; Hole 5A, depth of 207 m
8 7.8 Traces of Life 1365
Fig. 7.124 (continued) Shunga Event time: (f) C. 2.0 Ga Zaonega
Formation turbiditic siltstone-shale deposited on top of the eroded
Jatulian carbonate platform succession; Hole 11A, depth of 43.5 m.
(g) C. 2.05 Ga shallow-marine, laminated, turbiditic siltstone-shale
from the Kolosjoki Sedimentary Formation in the Pechenga Greenstone
Belt; Hole 8A, depth of 141 m. (h) Thin chert bed from the pillowed
lava succession of the c. 2.05 Ga Kolasjoki Volcanic Formation in the
Pechenga Greenstone Belt; Hole 9A, depth of 83.4 m. (i) Early
diagenetic chert nodules (arrowed) in sandstone-shale from the c.
2.0 Ga Zaonega Sedimentary Formation in the Onega Basin; Hole
12A, depth of 22.2 m. (j) Chert beds (pale grey) interbedded with
shale from the Zaonega Sedimentary Formation in the Onega Basin;
Hole 12A, depth of 13.6 m (k) Turbiditic, rhythmically-bedded, deep-
water siltstone-shale of the c. 2.0 Ga Zaonega Sedimentary Formation
in the Onega Basin; Hole 12A, depth of 159 m
1366 E.J. Javaux et al.
References
Aleon J, Robert F, Chaussidon M, Marty B (2003) Nitrogen
isotopic composition of mantle macromolecular organic
matter in interplanetary dust particles. Geochim Cosmochim
Acta 67:3773–3783
Allwood AC, Walter MR, Kamber BS, Marshall CP,
Burch IW (2006) Stromatolite reef from the early Archaean
era of Australia. Nature 441:714–718
Amard B, Bertrand-Safarti J (1997) Microfossils in 2000
Ma old cherty stromatolites of the Franceville Groups,
Gabon. Precambrian Res 81:197–221
Anbar AD, Duan Y, Lyons TW, Arnold GL, Kendall B,
Creaser RA, Kaufman AJ, Gordon GW, Garvin J, Buick R
(2007) A whiff of oxygen before the great oxidation event?
Science 317:1903–1906
Awramik SM, Barghorn ES (1977) The Gunflint
microbiota. Precambrian Res 5:121–142
Barghoorn ES, Tyler SM (1965) Microorganisms from
the Gunflint chert. Science 147:563–577
Bekker A, Holland HD, Wang PL, Rumble D, Stein HJ,
Hannah JL, Coetzee LL, Beukes NJ (2004) Dating the rise of
atmospheric oxygen. Nature 427:117–120
Bengtson S, Rasmussen B (2009) New and ancient trace
makers. Science 323:346–347
Bengtson S, Rasmussen B, Krapez B (2007) The Paleo-
proterozoic megascopic Stirling biota. Paleobiology 33:
351–381
Bernard S, Benzerara K, Beyssac O, Menguy N, Guyot F,
Brown GE Jr, Goffe B (2007) Exceptional preservation of
fossil plant spores in high-pressure metamorphic rocks.
Earth Planet Sci Lett 262:257–272
Beyssac O, Goffe B, Chopin C, Rouzaud JN (2002)
Raman spectra of carbonaceous material in metasediments:
a new geothermometer. J Metamorph Geol 20:859–871
Brandes JA, Cody GD, Rumble D III, Haberstroh P,
Wirick S, Gelinas Y (2008) Carbon K-edge XANES
spectromicroscopy of natural graphite. Carbon 46:1424–1434
Brocks JJ, Buick R, Summons RE, Logan GA (2003) A
reconstruction of Archean biological diversity based on
molecular fossils from the 2.78–2.45 billion year old
Mount Bruce Supergroup, Hamersley Basin, Western
Australia. Geochim Cosmochim Acta 67:4321–4335
Buick R (2008) When did oxygenic photosynthesis
evolve? Philos Trans R Soc B 363:2731–2743
Canfield DE (1998) A new model for Proterozoic ocean
chemistry. Nature 396:450
Cloud PE, Morrison K (1980) New microbial fossils from
2 Gyr old rocks in northern Michigan. Geomicrobiol J
2:161–178
Fig. 7.124 (continued) (l) Polished slab of polymictic gritstone
containing clasts of resedimented phosphorites (black), sulphides
(pale yellow) and various sedimentary rocks from the c. 2.0 Ga
Pilguj€arvi Sedimentary Formation in the Pechenga Greenstone Belt;
an open pit mine. (m) Cross-section view of turbiditic greywacke with
rounded clasts of pyrobitumen (red arrows) derived from a surface oil
seep; the 1.9 Ga Kondopoga Sedimentary Formation from the Onega
Basin; the Kondopoga quarry (Photographs by Victor Melezhik)
8 7.8 Traces of Life 1367
Dutkiewicz A, George SC, Mossman DJ, Ridley J, Volk
H (2007) Oil and its biomarkers associated with the
Palaeoproterozoic Oklo natural fission reactors, Gabon.
Chem Geol 244:130–154
El Albani A, Bengtson S, Canfield DE, Bekker A,
Macchiarelli R, Mazurier A, Hammarlund EU, Boulvais P,
Dupuy J-J, Fontaine C, F€ursich FT, Gauthier-Lafaye F,
Janvier P, Javaux E, Ossa FO, Pierson-Wickmann A-C,
Riboulleau A, Sardini P, Vachard D, Whitehouse M,
Meunier A (2010) Large colonial organisms with coordi-
nated growth in oxygenated environments 2.1 Gyr ago.
Nature 466:100–104
Fralick P, Davis DW, Kissin SA (2002) The age of the
Gunflint Formation, Ontario, Canada: single zircon U–Pb
age determinations from reworked volcanic ash. Can J
Earth Sci 39:1085–1091
Gago R, Jimenez I, Albella JM (2001) Detecting with X-
ray absorption spectroscopy the modifications of the bond-
ing structure of graphitic carbon by amorphisation,
hydrogenation and nitrogenation. Surf Sci 482:530–536
Garcia-Ruiz JM, Hyde ST, Carnerup AM, Christy AG,
Van Kranendonk MJ, Welham NJ (2003) Self-assembled
silica-carbonate structures and detection of ancient
microfossils. Science 302:1194–1197
Golubic S, Hofmann HJ (1976) Comparison of Holocene
and mid-Precambrian Entophysalidaceae (Cyanophyta) in
stromatolitic algal mats: cell division and degradation. J
Paleontol 50:1074–1082
Grey K, Williams IR (1990) Problematic bedding-plane
markings from the Middle Proterozoic Manganese Sub-
group, Bangemall Basin, Western Australia. Precambrian
Res 46:307–327
Guo Q, Strauss H, Kaufman AJ, Schr€oder S, Gutzmer J,
Wing B, Baker MA, Bekker A, Jin Q, Kim S-T, Farquhar J
(2009) Reconstructing Earth’s surface oxidation across the
Archean-Proterozoic transition. Geology 37:399–402
Han T-M, Runnegar B (1992) Megascopic eukaryotic
algae from the 2.1-billion-year-old negaunee iron formation.
Science 257:232–235
Hannah JL, Bekker A, Stein HJ, Markey RJ, Holland
HH (2004) Primitive Os and 2316 Ma age for marine shale:
implications for Paleoproterozoic glacial events and the
rise of atmospheric oxygen. Earth Planet Sci Lett
225:43–52
Hofmann HJ (1976) Precambrian microflora, Belcher
Islands, Canada: significance and systematic. J Palaeontol
50:1040–1073
Hofmann HJ (2004) Archean microfossils and
abiomorphs. Astrobiology 4:135
Horodyski R (1982) Problematic bedding-plane markings
from the Middle Proterozoic Appekunny Argillite, Belt
Supergroup, northwestern Montana. J Paleontol 56:882–889
House CH, Schopf JW, McKeegan KD, Coath CD,
Harrison TM, Stetter KO (2000) Carbon isotopic
composition of individual Precambrian microfossils. Geol-
ogy 28:707–710
Ivanova LV, Chapina OS, Melezhik VA (1988) Discov-
ery of coccoidal microfossils in early Precambrian metamor-
phosed cherts. Commun USSR Acad Sci 303:210–211 (in
Russian)
Javaux EJ (2006) The early eukaryotic fossil record. In:
Jekely G (ed) Origins and evolution of eukaryotic
endomembranes and cytoskeleton, Advances in experimen-
tal medicine and biology. Landes Biosciences/Springer
series, Austin, pp 1–19
Javaux EJ, Benzerara K (2009) Microfossils. In:
Gargaud M, Mustin C, Reisse J, Vandenabeele-Trambouze O
(eds) Traces de vie presente ou passee: quels indices,
signatures ou marqueurs? vol 8, Comptes Rendus Palevol
Special Issue. Academie des Sciences/Elsevier Masson SAS,
Paris/Issy les Moulineaux cedex, pp 605–615
Javaux EJ, Marshall CP (2006) A new approach in
deciphering early protist paleobiology and evolution: com-
bined microscopy and microchemistry of single Proterozoic
acritarchs. Rev Palaeobot Palynol 139:1–15
Javaux EJ, Knoll AH, Walter MR (2003) Recognizing
and interpreting the fossils of early eukaryotes. Orig Life
Evol Biosph 33:75–94
Javaux EJ, Knoll AH, Walter MR (2004) TEM evidence
for eukaryotic diversity in mid-Proterozoic oceans.
Geobiology 2:121–132
Javaux EJ, Marshall CP, Bekker A (2010) Organic-
walled microfossils in 3.2-billion-year-old shallow-marine
siliciclastic deposits. Nature 463:934–938
Kaufman AJ, Johnston DT, Farquhar J, Masterson A,
Lyons TW, Bates S, Anbar AD, Arnold GL, Garvin J,
Buick R (2007) Late Archean biospheric oxygenation and
atmospheric evolution. Science 317:1900–1903
Knoll AH (1996) Archean and proterozoic paleontology.
In: Jansonius J, McGregor DC (eds) Palynology: principles
and applications, vol 1, American Association of Stratigraphic
Palynologists Foundation. American Association of Strati-
graphic Palynologists Foundation/Texas A&M University,
College Station, pp 51–80
Knoll AH (2003) Life on a young planet, the first three
billion years of evolution on Earth. Princeton University
Press, Princeton, 277 p
Knoll AH, Golubic S (1992) Living and Proterozoic
cyanobacteria. In: Schidlowski M, Golubic S, Kimberley MM
(eds) Early organic evolution: implications for mineral and
energy resources. Springer, Berlin, pp 450–462
Knoll AH, Strother PK, Rossi S (1988) Distribution and
diagenesis of microfofssils from the lower Proterozoic Duck
Creek Dolomite, Western Australia. Precambrian Res
38:257–279
Knoll AH, Javaux EJ, Hewitt D, Cohen P (2006) Eukary-
otic organisms in Proterozoic oceans. Philos Trans R Soc B
361:1023–1038
1368 E.J. Javaux et al.
Kudryavtsev AB, Schopf JW, Agresti DG, Wdowiak TJ
(2001) In situ laser-Raman imagery of Precambrian micro-
scopic fossils. Proc Natl Acad Sci 98:823–826
Kumar S (1995) Megafossils from the Mesoproterozoic
Rohtas Formation (the Vindhyan Supergroup), Katni area,
central India. Precambrian Res 72:171–184
Lepot K, Benzerara K, Brown GE Jr, Philippot P (2008)
Microbially influenced formation of 2724-million-year-old
stromatolites. Nat Geosci 1:118–121
Lepot K, Philippot P, Benzerara K, Wang G-Y (2009a)
Garnet-filled trails associated with carbonaceous matter
mimicking microbial filaments in Archean basalt.
Geobiology 7:6579–6599
Lepot K, Benzerara K, Rividi N, Cotte M, Brown GE,
Philippot P (2009b) Organic matter heterogeneities in 2.72
Ga stromatolites: alteration versus preservation by sulphur
incorporation. Geochim Cosmochim Acta 73:6579–6599
Matz MV, Frank TM, Marshall NJ, Widder EA, S€onke J(2008) Giant deep-sea protist produces bilaterian-like traces.
Curr Biol 18:1849–1854
Moreau JW, Sharp TG (2004) A transmission electron
microscopy study of silica and kerogen biosignatures in 1.9
Ga Gunflint microfossils. Astrobiology 4:196–210
Nasdala L, Smith DC, Kaindl R, Ziemann MA (2004)
Raman spectroscopy: analytical perspectives in mineralogical
research. In: Beran A, Libowitzky E (eds) Spectroscopic
methods in mineralogy, vol 6, EMU notes in mineralogy.
E€otv€os University Press, Budapest, pp 281–343
Nisbet EG, Grassineau NV, Howe CJ, Abel PI, Regelous
M, Nisbet RER (2007) The age of Rubisco: the evolution of
oxygenic photosynthesis. Geobiology 5:311–335
Oehler DZ (1977) Microflora of the middle Proterozoic
Balbirini Dolomite (McArthur Group) of Australia.
Alcheringa 2:269–309
Oehler DZ, Robert F, Mostefaoui S, Meibom A, Selo M,
McKay DS (2006) Chemical mapping of Proterozoic
organic matter at submicron spatial resolution. Astrobiology
6:838–850
Oehler DZ, Robert F, Walter M, Sugitani K, Allwood AC,
Meibom A, Mostefaoui S, Selo M, Thomen A, Gibson EK
(2009) NanoSIMS: insights to biogenicity and syngeneity of
Archean carbonaceous structures. Precambrian Res 173:
70–78
Ono S, Beukes NJ, Rumble D, Fogel ML (2006) Early
evolution of atmospheric oxygen from multiple-sulfur and
carbon isotope records of the 2.9 Ga Mozaan Group of
the Pongola Supergroup, Southern Africa. S Afr J Geol
109:97–108
Pasteris JD (1989) Insitu analysis in geological thin-
sections by laser Raman microprobe spectroscopy: a cau-
tionary note. Appl Spectrosc 43:567–570
Pasteris JD, Wopenka B (2002) Images of the Earth’s
earliest fossils? Nature 420:476–477
Pasteris JD, Wopenka B (2003) Necessary, but not
sufficient: Raman identification of disordered carbon as a
signature of ancient life. Astrobiology 3:727–738
Pawlowski J, Gooday AJ (2008) Precambrian biota: pro-
tistan origin of trace fossils? Curr Biol 19(1):R28–R30
Planavsky N, Rouxel O, Bekker A, Shapiro R, Fralick P,
Knudsen A (2009) Iron-oxidizing microbial ecosystems
thrived in late Paleoproterozoic redox-stratified oceans.
Earth Planet Sci Lett 286:230–242
Peng Y, Baoa H, Yuan X (2009) New morphological
observations for Paleoprotrtozoic acritarchs from the
Chuablinggou Formation, North China. Precambrian Res
168:223–232
Rasmussen B, Fletcher IR, Brocks JJ, Kilburn MR (2008)
Reassessing the first appearance of eukaryotes and
cyanobacteria. Nature 455:1101–1105
Rosing MT, Frei R (2004) U-rich Archean sea-floor
sediments from Greenland-indication of >3700 Ma oxy-
genic photosynthesis. Earth Planet Sci Lett 217:237–244
Samuelsson J, Butterfield NJ (2001) Neoproterozoic
fossils from the Franklin Mountains, northwestern Canada:
stratigraphic and palaeobiological implications. Precam-
brian Res 107:235–251
Sangely L, Chaussidon M, Michels R, Huault V (2005)
Microanalysis of carbon isotope composition in organic
matter by secondary ion mass spectrometry. Chem Geol
223:179–195
Schneider DA, Bickford ME, Cannon WF, Sculz KJ,
Hamilton MA (2002) Age of volcanic rocks and
syndepositional iron formations, Marquette Range Super-
group: implications for the tectonic setting of Paleopro-
terozoic iron formations of the Lake Superior region. Can J
Earth Sci 39:999–1012
Schopf JW, Klein C (eds) (1992) The Proterozoic bio-
sphere: a multidisciplinary study. Cambridge University
Press, Cambridge, p 1348
Schopf JW, Kudryavtsev A (2005) Three-dimensional
Raman imagery of precambrian microscopic organisms.
Geobiology 3:1–12
Schopf JW, Kudryavtsev A, Agresti DG, Wdowiak TJ,
Czaja AD (2002) Laser-Raman imagery of Earth’s earliest
fossils. Nature 416:73–76
Seilacher A (2001) Concretionmorphologies reflecting dia-
genetic and epigenetic pathways. Sediment Geol 143:41–57
Sharma M, Shukla Y (2009) Taxonomy and affinity of
early Mesoproterozoic megascopic helically coiled and
related fossils from the Rohtas Formation, the Vindhyan
Supergroup, India. Precambrian Res 173:105–122
Shen Y, Canfield DE, Knoll AH (2002) Middle Protero-
zoic ocean chemistry: evidence from the McArthur Basin,
northern Australia. Am J Sci 302:81–109
Sokolov I, Kievsky Y (2005) 3D design of self-assembled
nanoporous colloids. Stud Surf Sci Catal 156:433–442
8 7.8 Traces of Life 1369
Summons RE, Jahnke LL, Hope JM, Logen GA (1999) 2-
Methylhopanoids as biomarkers for cyanobacterial oxygenic
photosynthesis. Nature 400:554–557
Tice MM, Bostick BC, Lowe DR (2004) Thermal history
of the 3.5–3.2 Ga Onverwacht and Fig Tree Groups, Barberton
Greenstone Belt, South Africa, inferred by Raman microspec-
troscopy of carbonaceous material. Geology 32:37–40
Timofeev BV (1982) Microphytofossils of the early Pre-
cambrian. Nauka, Leningrad, p 128 (in Russian, summary in
English)
Van Zuilen MA, Chaussidon M, Rollion-Bard C, Marty B
(2006) Carbonaceous cherts of the Barberton Greenstone Belt,
South Africa: isotopic, chemical, and structural characteristics
of individual microstructures. Geochim Cosmochim Acta
71:655–669
Walter MR, Goode ADT, Hall WDM (1976) Microfossils
from a newly discovered Precambrian Stromatolitic Iron
Formation in Western-Australia. Nature 261:221–223
Westall F, Folk RL (2003) Exogenous carbonaceous
microstructures in early Archaean cherts and BIFs from the
Isua Greenstone Belt: implications for the search for life in
ancient rocks. Precambrian Res 126:313–330
Wilson JP, Fischer WW, Johnston DT, Knoll AH,
Grotzinger JP, Walter MR, McNaughton NJ, Simon M,
Abelson J, Schrag DP, Summons R, Allwood A, Andresh M,
Gammon C, Garvin J, Rasby S, Schweizerb M, Watters WA
(2010) Geobiology of the late Paleoproterozoic Duck
Creek Formation, Western Australia. Precambrian Res
179:135–149
Wirth R (2009) Focused Ion Beam (FIB) combined with
SEM and TEM: advanced analytical tools for studies of
chemical composition, microstructure and crystal structure
in geomaterials on a nanometre scale. Chem Geol
261:217–229
Wopenka B, Pasteris JD (1993) Structural characteristics
of kerogens to granulite-facies graphite: applications
of Raman microprobe spectroscopy. Am Mineral
78:533–557
Xiao S, Dong L (2006) On the morphological and ecologi-
cal history of Proterozoic macroalgae. In: Xiao S, Kaufman
AJ (eds) Neoproterozoic geobiology and paleobiology.
Springer, Dordrecht, pp 57–90
Yochelson EL, Fedonkin MA (2000) A new tissue-grade
organism 1.5 billion years old from Montana. Proc Biol Soc
Wash 113:843–847
Zhang Z (1986) Clastic facies microfossils from the
Chuanlinggou Formation (1800 Ma) near Jixian, North
China. J Micropalaeontol 5:9–16
1370 E.J. Javaux et al.
7.8.4 Seeking Textural Evidence of aPalaeoproterozoic Sub-seafloorBiosphere in Pillow Lavas of thePechenga Greenstone Belt
Nicola McLoughlin, Harald Furnes, Eero J. Hanski,and Hubert Staudigel
Introduction
Microorganisms that inhabit pillow lavas of the sub-seafloor
create distinctive granular and tubular cavities by microbial
etching of the volcanic glass (e.g. Fisk et al. 1998; Furnes
et al. 2001a; McLoughlin et al. 2009; Staudigel et al. 2006,
2008) and these have been termed bioalteration textures.
Their fossil record includes volcanic glass from the in-situ
oceanic crust and fragments of seafloor preserved in Phanero-
zoic ophiolites and Precambrian greenstone belts (for a com-
prehensive review, see Furnes et al. 2008). This suggests that
volcanic glass from the sub-seafloor is one of the oldest
habitats for microbial life on Earth. A survey of Precambrian
bioalteration textures known to date reveals a paucity of data
from the Proterozoic with only one reported example from the
�1950 Ma Jormua ophiolite of Finland (Furnes et al. 2005)
where carbon isotopic signatures consistent with microbial
bioalteration have been reported, but in the absence of com-
pelling tubular or granular bioalteration textures. Thus at
present, there is a nearly two-billion-year gap in the textural
record of bioalteration between ~2.5 billion year old late
Archaean textures from Wutai China (McLoughlin et al.
2010b) and the next oldest reported textural evidence at
~0.4 Ga from the Solund-Stavfjord ophiolite of Western
Norway (Furnes et al. 2002; Fliegel et al. 2011). The
Fennoscandian Shield offers stratigraphic horizons that may
help to plug this gap, for example, pillow lavas of the ~2.0 Ga
Pechenga Greenstone Belt in northwestern Russia that are
investigated herein. We will first review what is currently
know about bioalteration textures from the recent to Archaean
oceanic crust, before describing the results of textural
investigations of pillow lava from surface samples collected
from the Pechenga Belt. This chapter is intended to provide
the background to future investigations of pillow lava
sequences from the FAR-DEEP drillcores for microbial bio-
alteration textures. We provide a guide to assessing the
biogenicity of such alteration textures with a view to their
future application as tracers of the effects of Earth’s
oxygenation on sub-seafloor microbial environments.
Review of Bioalteration Textures in the In SituOceanic Crust
Bioalteration of volcanic glass by microorganisms that etch
into the glass was discovered in the 1990s and has been
investigated using a combination of petrographic, geochem-
ical and microbiological techniques. When sub-glacial vol-
canic breccias from Iceland were studied, bacteria were
found within pits on the surface of volcanic glass fragments.
These lead Thorseth et al. (1992) to propose that the
microbes modify the local fluid pH and thereby accelerate
dissolution of the glass. This phenomenon has subsequently
been experimentally investigated using volcanic and syn-
thetic glasses inoculated with lithoautotrophic and
organotrophic microbes that produce etch pits and surface
alteration rinds on the glass under laboratory conditions
(Thorseth et al. 1995; Staudigel et al. 1995, 1998; Daughneyet al. 2004). Since this early work, numerous studies have
documented the global occurrence of bioalteration textures
in volcanic glass from pillow lava rims and interpillow
breccias collected by the ocean drilling programme (e.g.
Fisk et al. 1998; Furnes et al. 2001b). These biotic alteration
textures are distinct from the products of abiotic alteration,
which produces smooth interfaces between the fresh and
altered glass, with banded palagonite, a mixture of clays
and iron oxyhydroxides, occurring along the alteration fronts
(Furnes et al. 2001a; Furnes et al. 2008). This contrasts with
biologically mediated alteration that produces ramified
interfaces between the fresh and altered glass (Fig. 7.125)
with tubular structures including annulated, branched and
spiraled morphologies propagating from the alteration front
into the fresh glass (e.g. McLoughlin et al. 2010a). Several
groups have worked to develop criteria for establishing the
biogenicity and antiquity of these textures (see reviews by
McLoughlin et al. (2007), and Staudigel et al. (2008)).
Briefly, the key lines of evidence from a variety of studies
that have come together to support a biological origin for
tubular and granular microtextures in volcanic glass, are:
1. Textural investigations have revealed the complex
morphologies of bioalteration textures found in volcanic
glass that have been argued to require biological pro-
cesses to create the range of morphologies seen, espe-
cially the more intricate, annulated, helicoidal and
branched tubular forms (e.g. Furnes et al. 2001a, 2008)
and these have recently been considered and classified
as trace fossils (Walton 2008; McLoughlin et al. 2009). In
addition, textural studies have highlighted the similarities
between bioalteration textures in volcanic glass and more
well-known examples of microbial microborings in
carbonate substrates such as shells (cf. Golubic et al.
1975), and also fungal microborings in silicates, for
instance in soils (cf. Smits 2006). Furthermore, textural
investigations of the distribution and abundance of
bioalteration textures in volcanic glass have documented
N. McLoughlin (*)
Department of Earth Science and Centre for Geobiology, Allegaten 41,
Bergen N-5007, Norway
8 7.8 Traces of Life 1371
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_8, # Springer-Verlag Berlin Heidelberg 2013
1371
putative evidence of biological behavior (e.g. Furnes et al.
2001a; Walton 2008) that are discussed further below.2. Culture independent sequencing studies have shown that
the microbial population inhabiting the sub-seafloor is
distinct from that found in both overlying seawater and
seafloor sediments and up to 3–4 times more abundant
(Mason et al. 2007; Santelli et al. 2008). Moreover,
biological staining has revealed that DNA is concentrated
at the interface between fresh and altered glass along the
edges of tubular and granular bioalteration traces (Torsvik
et al. 1998). In addition, it has been theoretically shown
that basaltic glass can yield sufficient energy to support
chemolithoautotrophic growth (Bach and Edwards 2003).
3. Controlled laboratory experiments confirm a role for
microorganisms in the dissolution of volcanic glass. It
has been found that enhanced, localised dissolution of
volcanic glass occurs in experiments inoculated with
microorganisms relative to abiotic controls (Thorseth
et al. 1995; Staudigel et al. 1998). It has not, however,
yet been possible to cultivate micro-organisms in the
laboratory that create extended tunnel shaped etch pits.
4. Micro-chemical mapping has documented thin linings less
than 1 mm wide rich in carbon, nitrogen, and phosphorus
localised along the margins of modern and ancient
bioalteration textures and these are interpreted to be
decayed cellular remains (Furnes andMuehlenbachs 2003).
5. Carbon isotope analyses measured upon disseminated
carbonates in pillow lava rims is 13C-poor, typically
between +3.9 ‰ to �16.4 ‰, compared to carbonate
within the unaltered pillow interiors, which has d13Cvalues of +0.7 ‰ to�6.9 ‰ and are comparable tomantle
values. This much greater range exhibited by the pillow
rims is interpreted to reflect the microbial oxidation of
organic matter that gives the more negative values and
perhaps also the loss of 12C-enrichedmethane fromArchea
to give the more positive values (e.g. Furnes et al. 2001b).
6. Partially fossilised, mineral-encrusted microbial cells
have been observed in or near etch pits on altered glass
surfaces and these pits show forms and sizes resembling
the associated microbes suggesting that they are involved
in pit formation (Thorseth et al. 1992, 2001, 2003).
Considering all of these morphological, chemical and
microbiological lines of evidence, a strong case for the
biogenicity of tubular and granular cavities in volcanic
glass has been advanced and a conceptual model of how
micro-organisms etch the volcanic glass has been developed
in a series of papers (Thorseth et al. 1992, 1995; Staudigel
et al. 1998; Furnes et al. 2008; Staudigel et al. 2008). This
process begins when circulating fluids in the sub-seafloor
introduce micro-organisms along fractures, into vesicles and
around the rims of glass fragments. These microbes progres-
sively etch the fresh glass creating micro-textures that radi-
ate away from the surface of initiation producing a ramified
interface that renews and increases the surface area of fresh
glass available to the microorganisms (Staudigel et al. 2004).
The exact biochemical mechanisms of how the micro-
organisms dissolve the glass are not fully understood but
may conceivably include the secretion of organic acids, or
the production of siderophores and complexing agents
(Staudigel et al. 2008). Dissolution may also be
accompanied by precipitation of fine-grained authigenic
minerals termed palagonite within the micro-textures and
fractures. These may entomb organic remains creating the
localised enrichment in carbon, nitrogen and phosphorus
along the margins of the bioalteration textures – this type
of early diagenetic environment is favourable to the preser-
vation of biosignatures in volcanic substrates. It is thought
that a consortia of microorganisms is involved in the
bioalteration process possibly involving heterotrophs and
chemolithoautotrophs, and that in addition to endoliths that
bore, microbes that dwell in fractures and vesicles may also
be fossilized in pillow lava sequences (Cavalazzi et al. 2011;
Peckmann et al. 2008). This type of cryptoendolithic
biosignature will not be discussed further here.
There have only been a small number of systematic stud-
ies to date that have investigated the controls on the distribu-
tion of ichnofossils in volcanic glass. Preliminary studies
have been undertaken to estimate the fraction of biotic versus
abiotic glass alteration with depth, temperature, permeability
and porosity in the oceanic crust (e.g. Furnes and Staudigel
1999; Furnes et al. 2001a). These studies have found that the
granular type is by far the most abundant and can be found at
all depths into the oceanic crust where fresh glass is pre-
served down to c. 550 m. In the upper c. 350 m of the oceanic
crust, the granular type is the most abundant, decreasing
steadily to become scarce at temperatures of c. 115 �C near
the currently known upper limits of hyperthermophilic life.
The tubular type, in contrast, constitutes only a minor frac-
tion of the total microbial alteration, at most c. 20 %, and
shows an abundance maximum at c. 120–130 m depth. In the
whole oceanic volcanic pile, the total percentage ofmicrobial
alteration increases with permeability and also with the pres-
ence of celadonite, which is suggestive of oxygenated waters
(e.g. Furnes and Staudigel 1999; Furnes et al. 2001a). This is
significant in the context of Earth’s oxygenation, suggesting
that the distribution and perhaps abundance of bioalteration
textures should change in response to progressive
oxygenation of the Earth system. Moreover, changes in the
redox state of fluids circulating in the sub-seafloor likely had
implications for the potential metabolisms available to
microbes involved in the bioalteration process and this will
be discussed further below. With respect to the timing of
microbial bioalteration in the oceanic crust, it is noteworthy
that both the 5.9 Ma Costa Rica Rift and the 110 Ma western
Atlantic oceanic sections show a similar maxima in the
amount of bioalteration as a percentage of the total alteration,
despite their very different ages (Fig. 11 in Furnes et al.
2001a). This suggests that a substantial portion of the
bioalteration occurs early in crustal history, but it is thought
to persist within the crust as long as hydrothermal fluid
1372 N. McLoughlin et al.
circulation continues (Staudigel et al. 2008). It should also be
borne in mind that taphonomic variables such as changes in
fluid flow and authigenic mineral precipitation, especially the
growth of clays that can involve significant changes in vol-
ume, will modify the preservation potential of the
bioalteration textures in different parts of the oceanic volca-
nic pile. The development of an ichnofabric index for volca-
nic glass, like that recently proposed by Montague et al.
(2010), a semi-quantitative measure of the textural products
of microbial activity in volcanic glass, will help to further
elucidate the controls on the distribution of microbial activity
in the oceanic crust.
The distribution and orientation of bioalteration textures at
the thin-section scale has been taken as recording evidence of
biological behaviour. For example, tubular borings do not
intersect but rather, subparallel tubes are sometimes seen to
abruptly change growth direction by up to 180� where they
meet another tube or fracture (Fig. 4c in Furnes et al. 2007;
Fig. 5a inWalton 2008). This is interpreted to reflect adjacent
micro-organisms sharing the substrate, whereas abiotic tubu-
lar structures might be expected to intersect. This sharing of
the substrate may also explain why in areas with a high
density of microtubes, they are subparallel to avoid inter-
secting, whereas in areas of lower density microboring,
the tubes show more anastamosing paths. Additionally, it
has been argued that this anastomosing distribution is evi-
dence of mining behaviour, with their paths being designed
to maximise the extraction of resources from the glass
(Walton 2008). It has also been observed that tubular
microborings sometimes appear to seek olivine
phenocrysts, which are a rich source of iron in the glass,
and to avoid plagioclase that is relatively poor in Fe
(Walton 2008) – alternatively the tubes may be exploiting
weaknesses in the glass surrounding these phenocrysts.
Lastly, it has been highlighted that spiral microtubes can
“wrap on and off” one another, tentatively interpreted as
evidence of one generation of tubes providing support to
another (Fig. 5 in McLoughlin et al. 2009). Taken together,
these observations are suggestive of biological behaviour
recorded bymicroorganisms exploiting structural weaknesses
and compositional heterogeneities in volcanic glass.
Techniques for Seeking Evidence ofBioalteration
Having outlined the morphological and chemical evidence
sought to identify bioalteration textures in sub-seafloor vol-
canic pillow lavas, the techniques used for deciphering these
traces are now reviewed:
Optical microscopy: is used to examines the morphology of
the putative bioalteration texture in 2-dimensions and, if
z-plane stacking is available, in 3-dimensions. It is also
used to understand the mineralogy of the enclosing rock
to assess the relative age of the candidate biosignature
with respect to, for example, phases of abiotic alteration
or metamorphism of the host glass. Optical microscopy is
the best tool for investigating the shape, size and distribu-
tion of bioalteration textures and has revealed a range of
morphologies that include, granular, simple tubes,
branched, spiralled and annulated tubes (e.g. McLoughlin
et al. 2009), also changes in distribution that may reflect
biological behaviour (see above).Scanning electron microscopy – with energy dispersive X-ray
spectroscopy (SEM-EDX): is used to examine the shape
and surface morphology of a putative bioalteration texture,
and is especially useful for looking at etch pits and
authigenic mineral growth in fresh glass dredged from
the seafloor. Accompanying element distribution maps
can be created using EDX and, for example, have been
used to investigate trace metal distributions in palagonite-
filled fractures containing the moulds of endolithic
microbes (e.g. Kruber et al. 2008; Thorseth et al. 2003).
Electron microprobe: is used for non-destructive analysis of
the chemical composition of a bioalteration texture
including the quantification of elements present at levels
as low as 100 ppm. For example, principal component
analysis of clay-mineral analyses from altered volcanic
glasses that contain microtubes have shown that these
clay minerals are distinct from clay minerals found in
abiotically altered zones that lack microtubes (Storrie-
Lombardi and Fisk 2004). Electron microprobe has also
been used to map elemental enrichments on the rims of
bioalteration textures, especially N, C, and P interpreted
to be the remains of decayed organic material (e.g. Furnes
et al. 2008, and references therein).
Confocal laser Raman micro-spectroscopy: generates spectra
that are diagnostic of different mineral and organic
polymorphs and can be used for rapidmineral identification.
In the case of volcanic alteration textures, this has been used
to confirm the presence and map the distribution of titanite
within the textures (e.g. Fig. 27 in Furnes et al. 2008) and to
seek traces of carbonaceous matter (Lepot et al. 2011).
Isotopic composition of carbon, iron and sulphur: carbon
isotope data comes either from disseminated carbonate in
pillow rims (e.g. Cockell et al. 2010; Furnes et al. 2001b),
or from organic matter associated with palagonite (e.g.
Kruber et al. 2008). The depleted carbonate values
obtained from pillow rims are contrasted with mantle
values obtained from the unaltered pillow cores (see
point 5 above), and the organic matter is comparable to
marine biomass. However, these d13C values are not
diagnostic of specific microbial metabolisms and cannot
be related to particular alteration textures. Iron isotope
studies meanwhile have produced equivocal
interpretations given that the d56Fe variations measured
in secondary Fe-bearing minerals may be explained by
either biotic or abiotic processes (Rouxel et al. 2003).
8 7.8 Traces of Life 1373
Sulphur isotopes measured on secondary sulphides in
altered glass appear more promising (Rouxel et al.
2008), although the current number of studies is small.
U-Pb dating by laser-ablation inductively coupled plasma
mass-spectrometry (LA-ICP-MS): bioalteration textures
in metavolcanic glass are most often preserved by titanite,
a common greenschist facies mineral, and U-Pb radio-
metric dating on the titanite can be used to obtain an age
of mineralisation for the bioalteration textures. The ana-
lytical method can be found in Simonetti et al. (2006) and
has been applied to early Archaean (Banerjee et al. 2007;
Fliegel et al. 2010b) and late Archaean (McLoughlin et al.
2010b) examples. This is an important measurement to
establish the antiquity on the bioalteration texture – see
below for more discussion.
Focussed ion beam milling – transmission electron micros-copy (FIB-TEM): FIB is used to mill a very thin, ~100 nm
wafer from a chosen site within a sample to target, for
example, the wall of a bioalteration textures or traces of
organic material. This can then be imaged by TEM at the
nanometre scale to reveal cellular and crystalline structures
(e.g. Cockell et al. 2010; Lepot et al. 2011). Electron
diffraction patterns can also be generated to identify crys-
talline phases (e.g. Fliegel et al. 2010a, and below).
Synchrotron X-ray Spectroscopy and Microscopy: uses theabsorption of x-rays to image samples at the micron to
nanometre scale and to investigate, for example, the redox
state or co-ordination chemistry of the sample. This tech-
nique has been applied to recent volcanic glasses to inves-
tigate the Mn oxidation state of Fe-Mn crusts containing
epilithic microorganisms (e.g. Templeton et al. 2009), also
to glasses of the Ontong Java plateau to investigate Fe and
C speciation (e.g. Benzerara et al. 2007).
(Nano)SIMS nanometre-scale secondary ion mass spectrom-etry: has been used to map sub-micron scale elemental
and isotopic variations in and around bioalteration
textures. For example, investigation of altered glass
dredged from an Arctic spreading ridge found rims
enriched in Mn surrounding microbe-shaped structures
in the altered glass (McLoughlin et al. 2011). SIMS has
also been used to investigate major, minor and trace
element variations in the host glass surrounding
bioalteration textures, but have not found a link between
particular element(s) and different textures.
Bioalteration Textures in PrecambrianMetavolcanic Glass
The fossil record of microbial alteration of volcanic glass
extends far beyond the oldest in situ oceanic crust that at most
is 170 Ma old (Fisk et al. 1999) to include sequences
of metavolcanic pillow lavas. This is possible in pillow
lava that have escaped strong deformation, and if the
microtextures were mineralised prior to metamorphism of
the host glass. The most frequently observed mineralising
phase is titanite (CaTiSiO4), a common greenschist facies
mineral, and titanite-filled bioalteration textures have been
found in Phanerozoic ophiolites and Archaean greenstone
belts (for comprehensive reviews, see Furnes et al. 2008;
Staudigel et al. 2008). The oldest examples yet found come
from Palaeoarchaean pillow lavas of the Barberton Green-
stone Belt, South Africa (Furnes et al. 2004; Banerjee et al.
2006; Fliegel et al. 2010b), and the Pilbara Craton, Western
Australia (Banerjee et al. 2007) and have shed new light on
discussions concerning the earliest evidence for life on Earth.
All known Precambrian examples of candidate bioalteration
textures are summarised in Table 7.9 along with the lines of
evidence that have been used to assess their biogenicity.
In comparison to bioalteration textures from the in-situ
oceanic crust, the titanite-mineralised Precambrian examples
show similar distributions and are comparable in size, some-
time being slightly thicker in average diameter (see Fig. 7,
Furnes et al. 2007, for quantitative size information). The
range of morphologies is more restricted in the mineralised
bioalteration textures, with the simple unbranched tubular
type dominating. Annulated, spiraled and branched tubes
(c.f. McLoughlin et al. 2009) are rare, and the granular type
is usually overprinted by recrystallisation of the host
glass. Some of the less convincing examples of titanite
bioalteration textures, from both the Phanerozoic and the
Precambrian, lack the classical morphology of tubular cluster
radiating from fractures in the metavolcanic glass. Contrast,
for example, the Barberton Greenstone Belt examples
illustrated in Fig. 7.126h, and i with the candidate
bioalteration textures of the Abitibi Greenstone Belt textures
(Fig. 6 in Bridge et al. 2010), which comprise bands and
strings of titanite, that are often deformed and may occasion-
ally show only short, tubular projections.
In the anoxic ocean of the late Archaean to early Protero-
zoic, it is hypothesised that any microorganisms dwelling
in the sub-seafloor must have employed anaerobic
lithoautotrophy or perhaps heterotrophic metabolisms (c.f.
Edwards et al. 2005). This is in contrast to the modern sub-
seafloor where the principal electron acceptor is dissolved
O2 coupled to the inorganic electron donors HS�, Fe2+, andMn2+ found in basaltic glass. Prior to the widespread
oxygenation of the oceans, hydrogen from fluid rock
interactions may have been much more significant in sus-
taining sub-seafloor microbial communities (c.f. Hellevang
2008). Potential metabolisms responsible for the anaerobic
bioalteration of volcanic glass therefore include:
methanogenesis involving dissolved carbonate and hydro-
gen; sulphate reduction utilising dissolved sulphate and
hydrogen; in addition to heterotrophy based upon organic
1374 N. McLoughlin et al.
matter sourced either from lithoautotrophic biomass or from
circulating fluids carrying biomass from the overlying water
column (see Table 1 in Edwards et al. 2005).
Antiquity of Bioalteration Textures
To demonstrate that candidate bioalteration textures are
syngenetic with the volcanic glass and are not later features
relies, firstly, upon fabric relationships. At the outcrop scale,
this involves mapping to show that the phases containing the
bioalteration textures are syn-eruptive and not younger veins
or dyke filling phases. At the thin-section scale, the
bioalteration textures themselves should also be seen to pre-
date cross-cutting fractures, veins and cements. They should
be concentrated along paths of early fluid migration and/or
weaknesses in the glass and occur as asymmetric masses
across fractures that are distinct from symmetric, abiotic
palagonite alteration fronts. Further support for their antiquity
can be gained from direct U-Pb dating of titanite by LA-ICP-
MS. For example, titanite-mineralised trace fossils from the
Pilbara Craton of Western Australia record a U-Pb age of
2.9 � 0.1 Ga (Banerjee et al. 2007) that is�400 Ma younger
than the accepted eruptive age of the 3.35Ga host pillow lavas
given by a U-Pb zircon age on an interbedded tuff (Nelson
2005). This titanite date corresponds to the age of a regional
metamorphic event related to the last phase of deformation
and widespread granite intrusion in the Northern Pilbara
Craton at 2.93 Ga (van Kranendonk et al. 2002). Thus the
U-Pb titanite age represents a minimum estimate for the
timing of titanite formation, especially given that metamor-
phic chlorite cross-cuts the titanite tubules and so must post-
date the titanite mineralisation. This implies a <400 Ma post
eruptive period, during which the bioalteration textures from
the Pilbara were mineralised; but does not exclude the possi-
bility that the textures formed soon after eruption, remained
hollow and were mineralised somewhat later. Theoretically,
as long as there is still fresh glass present and seawater
circulation continues, the microbial bioalteration may go on
for millions of years. In the case of titanite-mineralised
bioalteration textures from the Barberton Greenstone Belt of
South Africa, the time gap between eruption and
mineralisation of the textures, or the so-called “window for
bioalteration”, is estimated to be much smaller, �130 Ma
(Fliegel et al. 2010b). In Fig. 7.127, we compare the age of
eruption (yellow star) with the age of titanite formation (green
star) for pillow lava sequences in which bioalteration textures
have been reported and a titanite mineralisation age obtained,
with the “window for bioalteration” being shown as a black
arrow.
Distinguishing Abiogenic Microtunnels fromBioalteration Textures
Microtunnels and etch pits in volcanic glass may also con-
ceivably be formed by abiotic processes, and when
investigating candidate bioalteration textures, it is necessary
to exclude this possibility. Potential abiotic tunneling
mechanisms are explored in detail by McLoughlin et al.
(2010a), particularly the chemical dissolution of pre-existing
heterogeneities in volcanic glass such as radiation damage
trails, gas-escape structures, or fluid inclusion trails. How-
ever, bioalteration textures can be distinguished from these
structures, because they are restricted to sites that were
connected to early fluid circulation. Moreover, their shapes,
distribution, and the absence of intersections, in particular,
exclude an origin by the purely chemical dissolution of pre-
existing heterogeneities in the glass (criteria are presented
in Table 1 of McLoughlin et al. 2010a). Rather the charac-
teristics of bioalteration textures are best explained by
microbial dissolution involving, perhaps, cellular extensions
that provide a mechanism of localising and directing micro-
tunnel formation as observed, for example, in terrestrial soils
(see Staudigel et al. 2008, and references therein).
Microtextures known as ambient inclusion trails (AITs),
which are reported from cherts and authigenic minerals,
have also been compared to bioalteration textures (e.g.
Lepot et al. 2009, 2011). AITs are filamentous structures a
few microns wide, up to tens to hundreds of microns long,
polygonal in cross-section, with longitudinal striae and
sometimes, with a terminal inclusion (Tyler and Barghoorn
1963). These structures are hypothesised to form by locally
elevated fluid pressures that propel a mineral inclusion
through a relatively “soft” substrate, such as microcrystal-
line phosphorite or chert, leaving a microtube in its wake
(Tyler and Barghoorn 1963; Knoll and Barghoorn 1974).
This crystal migration is believed to create a microtunnel
through a poorly understood combination of mechanical
abrasion, pressure solution and chemical dissolution. The
exact mechanism(s) that form AITs remain unclear, but
they can be distinguished from bioalteration textures
because they apparently form by the migration of crystalline
or organic inclusions in sealed substrates, in contrast to
bioalteration textures that originate on the glass surface
and propagate from sites of fluid flow in permeable
substrates. In addition, AITs exhibit longitudinal striae, a
constant diameter, and polygonal cross-section, sometimes
with terminal inclusions (see Table 1 in McLoughlin et al.
2010a) – features that are not observed in bioalteration
textures. Lastly, it has been highlighted that AIT-type pro-
cesses are highly unlikely in volcanic glass because of the
8 7.8 Traces of Life 1375
absence of crystalline millstones, localised chemical solu-
tion agents, and elevated fluid pressures, necessary to drive
this process (Banerjee et al. 2007; McLoughlin et al. 2010a).
To date, AIT type structures have only been reported from
unusual volcanic settings, including silica gel-filled vugs in
lava flows (Lepot et al. 2009); and from mafic pyroclastic
volcanic glass shards rimmed by early cements that contain
organic carbon and sulphides argued to have migrated dur-
ing metamorphism (Lepot et al. 2011).
In the open system found in pillow lavaswith high volumes
of fluid circulation, dissolution to form tunnels requires a
mechanism of localising the chemical solution agent. The
only known mechanism is in the vicinity of microorganisms.
Early work by Thorseth et al. (1991) suggested individual
cells as the locus for dissolution, and later work has
hypothesised chains of cells tunneling into the glass (e.g.
Fig. 1 in Furnes et al. 2008). This cannot, however, fully
explain the production of extended tubular micro-cavities,
because the microorganism would find it difficult to access
circulating fluids and excrete waste products through a tunnel
with such a narrow diameter and extended length relative to
their size. This would seem to require some kind of biological-
pump and thus, cellular extensions similar to fungal hyphae
have been suggested by Staudigel et al. (2008) as a mecha-
nism for localising chemical dissolution in a tunnel.
Surface Sampling of the Pechenga GreenstoneBelt
General GeologyThe Pechenga Greenstone Belt consists of four sedimentary-
volcanic cycles (Fig. 7.128) that are subdivided into eight
lithostratigraphic units, with a total thickness in excess of
10 km. From oldest to youngest, they are: the Neverskrukk
Formation, the Ahmalahti Formation, the Kuetsj€arvi Sedimen-
tary and Volcanic formations, the Kolosjoki Sedimentary and
Volcanic formations, and the Pilguj€arvi Sedimentary and Vol-
canic formations (Melezhik and Sturt 1994). A description of
these units is presented in Chap. 4.2. The time span of this
sedimentary-volcanic sequence remains imprecisely
constrained between c. 2505 (Amelin et al. 1995) and
1970 Ma (Hanski et al. 1990). The volcanic rocks of the
lower stratigraphic levels range from subaerial basaltic
andesites to alkaline basalts, andesites and dacites, whereas
those of the Kolosjoki and Pilguj€arvi volcanic formations are
predominantly tholeiitic basalts erupted in submarine
environments (Melezhik and Sturt 1994). The volcanic rocks
of the Pilguj€arvi and Kolosjoki Volcanic formations consist
predominantly of submarine pillowed andmassive basalt lavas
and lava breccias (Fig. 7.129). The pillow lavas throughout the
Kolosjoki Volcanic Formation are moderately to highly vesic-
ular, whereas those of the basal to c. middle part of the
Pilguj€arvi Volcanic Formation are non-vesicular, indicative
of eruption in relatively shallow and deep water, respectively
(cf. Moore 1965). At a stratigraphic level c. 2,000 m above
base level of the Pilguj€arviVolcanic Formation, the pillows get
vesicular, and upwards pass into massive lava flows that again
are overlain by a 5–10-m-thick felsic tuff unit (Fig. 7.129). The
lava immediately above the felsic tuff consists of massive
flows, and further upwards they alternate with non-vesicular
pillow lava (Fig. 7.129). Interlayered with the massive lava
sequence below and above (c. 100 m) the felsic tuff unit, are
several minor (a few-cm-thick) felsic tuff beds.
The degree of regional metamorphism is lowest in the
central part of the Pechenga structure and increases towards
itsmargins, varying frompumpellyite tomedium-temperature
amphibolite facies (Petrov and Voloshina 1995). Even though
the rocks have been metamorphosed, their sedimentary and
volcanic features are in parts very well preserved (Fig. 7.129).
Metamorphic titanite in thin section 39-Pch-07 (see arrow
Fig. 6b) has yielded a U-Pb age of 1790 � 89 Ma by in-situ
LA-ICP-MS analysis (Fliegel et al. 2010a); this analysis was
not made on the candidate bioalteration textures but rather the
metamorphic host rock. This age has been interpreted to
record the main phase of regional metamorphism in the area
known as the Svecofennian orogeny (Korja and Heikkinen
2005) or alternatively, post-orogenic granitoid magmatism
has also been documented at ~1.8 Ga from the northern part
of the shield (e.g. Corfu and Evins 2002).
SamplingPillow lavas were sampled throughout the volcanic
sequences of the Kolosjoki and Pilguj€arvi Volcanic
formations during field work in 2005 and 2007, and the
stratigraphic position of the samples is shown in Fig. 7.128.
The main goal of the sampling was to look for bioalteration
textures in the original glassy margins of pillows and
interpillow hyaloclastite. The 2005 collection comprises 54
hand samples of which 49 are pillow margins and interpillow
hyaloclastite. The 2007 collection of 126 mini-core samples
was obtained using a hand-held mini-drill machine. The
locations of all the 2007 samples was photographed and
carefully positioned to transect the chilled margin of pillows;
the interpillow hyaloclastite between adjacent pillows, i.e.
material that originally consisted of fresh basalt glass; and as
a control, the crystalline part of pillows usually 5–20 cm
inside the chilledmargin. Thus, the total number of the pillow
rim and hyaloclastite samples collected in 2007 is 82, with 57
from the Pilguj€arvi Volcanic Formation and 25 from the
Kolosjoki Volcanic Formation. Candidate bioalteration
textures, termed “tubular textures” by Fliegel et al. (2010a),
have been identified in four samples coming from three
different stratigraphic levels in the Pilguj€arvi Volcanic
1376 N. McLoughlin et al.
Formation (Fig. 7.128). These samples will first be described
before we discuss the origins of the textures.
Previous Reports of Candidate BioalterationTextures from the Pechenga Greenstone Belt
A study by Fliegel et al. (2010a) reported rare tubular textures
15–20 mm in diameter and up to several 100 mm long in
prehnite–pumpellyite to lower greenschist facies
metavolcanic glass of the Pechenga Belt. Fliegel et al.
(2010a) focused on three thin sections from the same sample
set described here: 39-Pch-05, 44-Pch-07 and 117-Pch-07
that are also illustrated and described below. They described
the textures in the samples as septate with regular
compartments 5–20 mm across and drew attention to
branching, and what they interpreted as evidence of stopping
and avoidance behaviour. FIB-TEM investigations followed
by electron diffraction showed that some of the textures were
mineralised by orientated pumpellyite and that, on the
margins of the tubes, the pumpellyite is partially replaced
by mica and or chlorite. A thin, poorly crystalline Fe-phase,
probably precipitated out of solution, was also documented at
the interface between the pumpellyite and mica and or chlo-
rite. Synchrotron micro-energy dispersive X-ray was also
used to image elemental distributions along with scanning
transmission X-ray microscopy the C absorption bands. No
carbon was found along the boundaries of the tubes. Taken
together, this evidence was used to propose an origin for
these textures involving mineralised by pumpellyite of origi-
nally hollow tubes created by microbial activity in volcanic
glass (Fliegel et al. 2010a). The authors advanced the hypoth-
esis that these textures record a novel preservation mecha-
nism involving mineralisation of bioalteration textures by
pumpellyite rather than titanite as more commonly reported.
Moreover, the presence of segmentation in the textures was
emphasised and interpreted as once septate cell walls that
provided a template for metamorphic mineralisation (Fig. 22
in Fliegel et al. 2010a). However, it remains to be explained
how a thin organic membrane comprising relatively large
cellular segments >20 mm long and >10 mm across could
be preserved and seed metamorphic mineral growth within a
hollow tubular structure. More generally, these reported
Pechengamicrotextures differ in many aspects from previous
descriptions of titanite-mineralised bioalteration textures
from metavolcanic glasses found elsewhere, especially with
regard to their size, morphology and distribution. Here we
describe and illustrate the full range of tubular microtextures
found in the Pechenga pillow lavas.
Description of Alteration Textures from thePechenga Greenstone Belt
Two broad morphotypes of textures are recognised, and
these will be termed large, segmented textures, illustrated
in Figs. 7.130, 7.131, 7.132, and 7.133 and in Fliegel et al.
(2010a), and previously undescribed, narrower, curving
textures, see for example Figs. 7.131 and 7.133.
Large Morphotype: Metamorphic AlterationTexturesThe large morphotype can exceed 50 mm in diameter and
reach more than 200 mm in length (Figs. 7.130, 7.131,
7.132, and 7.133). Some examples show a near constant
diameter (e.g. Fig. 7.130), while others are strongly tapered
(e.g. Fig. 7f). They may be straight or curved, and are com-
monly segmented, with segments c.10 mm in length
(Fig. 7.130d). Many examples show ragged as opposed to
smooth margins (e.g. Fig. 7.130d). The distribution of these
textures has been incompletely described and includes at least
two modes of occurrence.
The first mode of occurrence is as dense zones around the
rims of volcanic glass fragments, especially in fragments
that are surrounded by carbonate and opaque mineral-rich
matrix phases (e.g. Fig. 7.131a–c). There are dense
overlapping masses fringing the glass fragments that grow
in from the matrix and, where chlorite becomes more abun-
dant, the pumpellyite masses break down and the large
morphotype becomes more distinct and individual seg-
mented textures become recognisable (e.g. Fig. 7.131c). In
some samples, it can be seen that these segmented textures
cross-cut the fabric in the chlorite groundmass and thus
postdate at least the early metamorphic growth of chlorite
(e.g. Fig. 7.131c). In glass fragments that are rimmed with
titanite (e.g. left hand side of fragment in Fig. 7.131a versus
right hand side), no textures are found.
The second mode of occurrence is associated with coarse,
cross-cutting carbonate-filled veins. Sometimes the textures
occur in dense patches adjacent, but not necessarily
connected, to the veins (e.g. Fig. 7.133c, d), and here the
elongate textures may overlap, often at right angles (e.g.
Fig. 7.133c, d). Sometimes the textures are connected to
these carbonate veins (e.g. Fig. 7.131d–f) and propagate at
high angles into the glass. These examples show a wide range
in widths and are often strongly tapered (e.g. Fig. 7.131d–f).
Both modes of occurrence are related to the occurrence of
carbonate, either in cross-cutting veins or in the vicinity of
carbonate-rich IPH matrix. This is taken here to suggest that
the growth of pumpellyite in these textures is associated with
the infiltration of a CO2-rich fluid.
8 7.8 Traces of Life 1377
Small Morphotype: Candidate Ambient InclusionTrailsThe smaller morphotype are less than 10 mm in diameter,
with many being much finer between 1–2 mm and generally
less than 100 mm in length (Fig. 7.132). Two modes of
occurrences have been observed:
The first mode involves small tubes radiating from a
central (botryoidal) quartz-carbonate-opaque phase-filled
area into chlorite, these often contain euhedral terminal
sulphides (identified in reflected light, e.g. Fig. 7.132d).
This mode of occurrence is most abundant in samples that
contain only small pockets of chlorite in large volumes of
carbonate-quartz-rich interpillow matrix (e.g. Fig. 7.132a).
The finding of terminal crystals in this morphotype (arrows
in Fig. 7.132d, e) suggests than an AIT-type process may be
involved in their formation.
The second mode of occurrence is more unusual and has
not been described before. It involves small tubes radiating
from rhombic structures and fine fractures in the sample
(Fig. 7.133). These occur in slides that also show extensive
iron staining along non-mineralised fractures, suggesting
the influence of recent surface-derived fluids. We postulate
that the rhombs are dissolved carbonate crystals possibly
related to weathering and that the fine tubes may be related
to biological activity. The antiquity of these traces needs to
be established, and testing with biological stains may help
to establish if this morphotype records recent microbial
activity. Until a recent origin can be excluded, this
morphotype is not relevant to discussions of the Protero-
zoic biosphere.
Summary of the Textural ObservationsThere are a number of observations that raise questions
about the biogenicity of these candidate bioalteration traces
from the pillow lavas of the Pechenga Greenstone Belt:
1. The large textures do not show the typical distribu-
tion pattern expected in bioalteration textures with
tubular cluster that radiates from “root zones” at original
fractures in the glass. Contrast, for example, Figs. 7.130,
7.131, 7.132, and 7.133 from Pechenga with Fig. 7.126h–i
that illustrates bioalteration textures from the Barberton
Greenstone Belt of South Africa. Rather in the Pechenga
samples we observe bands or zones of textures in the
vicinity of carbonate-bearing veins or inter-pillow matrix
that we suggested here are related to the infiltration of a
CO2-bearing fluid.
2. The large textures show a morphological spectrum from
dense masses of alteration in parts of the formerly glassy
rims of hyaloclastite fragments, that breaks down into
individual large, segmented textures further from the
carbonate source and where the matrix is more chlorite-
rich, e.g. Fig. 7.131a–c. We interpret the morphology of
these textures to be controlled by the composition and
fabric of the matrix, and to not record a biological
population. These textures can rather be compared to
the metamorphic growth of pumpellyite in metavolcanic
pillow lavas as reported, for example, in ophiolite
sequences of the Appalachian metamorphic belt (c.f.
Zen 1974).
3. The size and the size range of the textures are much
greater than that observed in previous reports, with
textures spanning 5 to >50 mm in diameter. This greatly
exceeds the ranges measured from other examples of
tubular bioalteration textures in (meta)volcanic glass
(c.f. Fig. 7 in Furnes et al. 2007).
4. The occurrence of the small morphotype (Fig. 7.131d–e)
interpreted to be artefacts that are related to an ambient
inclusion trail-type process, demonstrates that a range of
alteration processes have affected the Pechenga
metavolcanic glass. We note that comparable microtubu-
lar structures containing terminal sulphides and traces of
organic carbon have recently been described in 2.7 Ga
chloritised metavolcanic glass from West Australia
(Figs. 4 and 5 Lepot et al. 2011) and were also interpreted
as AITs.
In short, on the basis of textural evidence, we conclude
that the large tubular morphotype is not biological in origin
and rather, that their size, shape and distribution is more
consistent with an origin as abiogenic metamorphic alter-
ation products related to infiltration of a CO2-rich fluid. The
origin of the small tubular morphotype is uncertain, but the
occurrence of terminal inclusions in some suggests that an
AIT-type process was involved.
Implications for the FAR-DEEP Drill Core
The alteration textures described above from the pillow lavas
of the Pechenga Greenstone Belt are different in terms of their
size, size range, shape, distribution and mineralogy from all
other candidate bioalteration textures reported to date from
the Precambrian (Table 7.9). The current lines of evidence
used to argue for the biogenicity of titanite-mineralised
bioalteration textures in greenschist facies metavolcanic
glass do not apply to these textures from the Kolosjoki and
Pilguj€arvi Volcanic formations. Rather we find that the
textures reported by Fliegel et al. (2010a) and re-described
above occur in only narrow horizons in the pillow lava pile
(Figs. 4 and 5) apparently linked to sub-greenschist facies
metamorphic conditions and the influx of CO2-rich fluids.
Current models envisioned for the bioalteration of volcanic
glass cannot explain the textures found in the Pechenga Belt.
Before a biological origin for these textures can be
demonstrated, a more complete understanding of the meta-
morphic alteration processes affecting these rocks is required,
in addition to supporting geochemical evidence.We also note
the resemblance of some of the textures to abiotic ambient
inclusion trails, whilst acknowledging that this hypothesised
process is not fully understood in volcanic substrates.
1378 N. McLoughlin et al.
Table
7.9
Asummaryofallcandidatebioalterationtexturesin
ageorder,reported
todatefrom
thePrecambrian.Columnsreportthelithology(IPH
¼interpillowhyaloclastite),eruptionand
mineralisationage,andthelines
ofevidence
usedto
assess
biogenicity,nam
ely:morphology,mineralogy,presence\absence
ofC,N,Plinings,andthed1
3Ccarbisotopicvalues
ofdisseminated
carbonatefrom
thepillow
rims.A
colourcoded
assessmentofthebiogenicityisshown,with:green
¼strongcombined
evidence;orange¼
putativeevidence
(oneormore
lines
ofevidence
beingunavailableorinconclusive);andred¼
weakevidence,requiringfurther
investigation.In
thelaterpartsofthischapter,wewilldescribeindetailtexturesfoundinthe~2.0Gapillowlavas
ofthePechengaGreenstoneBelt,northwestern
Russia
Locality
Lithology
Eruptive
age(G
a)
Mineralisation
age(G
a)Morphologies
Mineralogy
C,N,Plinings
d13Ccarb
pillow
rims
Biogenicity
Reference
Joruma
Ophiolite,
Finland
IPH
1.95
Unknown
Texturesdestroyed
Titanite
C,N,Plinings
mapped
by
electronprobe
�14.1
to �3.3
‰
Furnes
etal.(2005)
Pechenga
GreenstoneBelt,
Russia
Pillow
rimsand
IPH
~2.0
1.790�
89
Segmentedtubular
microtextures,seebelow
Pumpellyite
some
exam
ples
NoC,N,P
linings
�25to
0‰
Fliegel
etal.(2010a)
Wutai,China
IPH
~2.52
1.81�
0.12
Raretubularclusters
Titanite
Nodata
Sparse
data
McL
oughlinet
al.(2010b)
Abitibi
GreenstoneBelt,
Canada
IPH
2.701
Unknown
Titanitebands,withrare,
small,possible
tubular
projections
Titanite
Nodata
None
Bridgeet
al.(2010)
Barberton
GreenstoneBelt,
South
Africa
Pillow
rimsand
IPH
3.47–3.45
3.342�
0.068
Segmented,tubular
microtextures
Titanite
Electronprobe
maps,C,N,P
linings
�16.4
to +3.9
‰
Furnes
etal.(2004);Fliegel
etal.(2010b);Banerjeeetal.
(2006)
PilbaraCraton,
WestAustralia
Pillow
rimsand
IPH
�3.350
2.921�
0.110
Segmented,tubular
microtextures
Titanite
Electronprobe
maps,C,N,P
linings
�1to
+3.0
‰Banerjeeet
al.(2007);
Furnes
etal.(2008)
8 7.8 Traces of Life 1379
1380 7.8 Traces of Life
Fig. 7.125 Transmitted light images of bioalteration textures in vol-
canic glass. (a) Fracture on right hand margin with banded palagonite at
core and from which unbranched tubular bioalteration textures propa-
gate into the fresh glass. Note also the igneous phenocryst on the left
hand margin and the fracture with granular alteration textures in the
lower left. (b) Dense zone of simple tubular textures propagating from
a fracture in the lower part of the image into the fresh glass. (c) A
formerly fluid-filled vesicle from which simple tubular textures radiate;
also a fracture (lower left) with granular alteration. (d) Fracture runningnorth–south in the centre of the image with banded material at the core
and dense fringing bands of granular alteration on the outside edge
from which tubular textures can be seen projecting into the unaltered
glass. (e) Enlargement showing mineralised tubular textures in the
lower part of the image, and unmineralised tubes with a smaller
diameter in the upper part of the image. (f) Extended depth of focus
image showing a spiral shaped tubular textures. (g) Spiral shaped
texture with an outer helix that shows very uneven spacing of the
whorls and a terminal swelling in glass that contains a fine network
of angular fractures and pyroxene microphenocrysts. (h) Extended
depth of focus image showing a fracture running north–south in the
centre of the image from which twisted filamentous structures radiate;
these are filled with micro-crystalline titanite and are contained within
glass now metamorphosed to zeolite minerals (Images (a–e) are from
hole 418A on the Bermuda Rise; (f–g) from sample CY-1-30 from the
Akaki River section of the Troodos ophiolite, Cyprus; and (h) from
sample 3-Al-00 from the 160 Ma Mirdita ophiolite of Albania where
the volcanic glass has been partially transformed to zeolite facies
minerals; figure modified from McLoughlin et al. (2010a))
◂
8 7.8 Traces of Life 1381
1382 7.8 Traces of Life
Fig. 7.126 Transmitted light images of bioalteration textures in
greenschist facies metavolcanic glass. (a) Low-magnification image
of metavolcanic glass composed largely of chlorite with bands of
titanite showing tubular projections. (b) Enlarged view showing curved
titanite band with tubular, in some cases segmented (arrowed)projections, note also the darker central band within the titanite,
interpreted to be an original fracture in the glass. (c) Interpillow
hyaloclastite sample with a chlorite-carbonate matrix containing
bands of titanite. (d) Enlargement showing the tubular textures com-
prising chains of titanite crystals. (e) Dense clusters of titanite (black
mineral) filled tubular textures in a chlorite, carbonate matrix. (f)
Enlargement showing curvi-linear tubes of variable diameters. (g)
Single tubular texture mineralised by micro-crystalline titanite. (h)
Titanite filled cluster of tubular textures radiating from a former frac-
ture in the original glass now comprising chlorite, quartz and epidote.
(i) High magnification image showing septate titanite filled tubular
textures, cross-cut by metamorphic chlorite from the groundmass
(arrowed). Dotted white circles in images (e–h) mark location of
laser pits for LA-MC-ICP-MS (Images (a–b) are from sample 114-
SS-00 from the ~440 Ma Solund Stavfjord ophiolite of W Norway;
(c–d) sample 13-WG-06 from the ~2.52 Ga Wutai Belt of China; (e–g)
sample 74-PG-04 from the 3.35 Ga Euro Basalt of the Pilbara Craton,
W. Australia; and (h–i) sample 29-BG-03 from the 3.46 Ga
Hooggenoeg Formation of the Barberton Greenstone Belt, S. Africa)
◂
8 7.8 Traces of Life 1383
Fig. 7.127 Geological timeline summarising the occurrences of pil-
low lavas containing candidate bioalteration textures. Orange starsrepresent chemical traces and yellow stars textural traces of
bioalteration found in (meta)volcanic glass of given eruptive ages;
green stars represent U-Pb titanite mineralisation ages of the given
localities, and the black arrows the time window for bioalteration.
References to each locality are given in the text
1384 N. McLoughlin et al.
Fig. 7.128 Simplified geological map showing the sampling area. Sections A and B show the volcanic stratigraphy of the Kolosjoki and
Pilguj€arvi Volcanic formations and the sample locations for the 2005 and 2007 collections (Map is modified from Melezhik and Sturt (1994))
8 7.8 Traces of Life 1385
Fig. 7.129 Field photographs of volcanic lithologies from the
Kolosjoki and Pilguj€arvi Volcanic formations. (a) Non-vesicular pillow
lava with drill holes showing the locations of samples 112-Pch-07 and
113-Pch-07, penetrating the chilled margin/interpillow hyaloclastite.
(b) The layer in the middle of the photo is the thick acid tuff (dated to
1970 � 5 Ma; Hanski et al. 1990). Below and above the tuff layer are
massive basalt lava flows. (c) Non-vesicular pillow lava at the base of
the Pilguj€arvi Volcanic Formation. (d) Highly vesicular pillow lava
from the upper part of the Kolosjoki Volcanic Formation. (e) Detail
from part of a pillow, showing the high concentration of vesicles along
the margin of the pillow. (f) Pillow lava breccia
1386 N. McLoughlin et al.
8 7.8 Traces of Life 1387
Fig. 7.130 Pillow rim sample containing large microtextures: (a)
Hand sample from which the thin section was made, pillow rim
(upper part) with a carbonate filled vug (dashed line) containing a
glass shard (arrowed); (b) Transmitted light image at low magnifica-
tion showing a chloritised glass shard in a carbonate-quartz-opaque-
filled vug, with microtextures propagating inwards from the margins of
the chloritised glass shard. The arrow (lower right) shows the band of
titanite that was dated by LA-ICP-MS. (c) Cross-polarised image of
fine-grained chlorite (dark green) cross-cut by coarse segmented
textures (light green-yellow). (d) Enlarged plane-polarised image
showing segmented microtextures with ragged margins. Sample 39-
Pch-05 also illustrated in Figs. 6–8 of Fliegel et al. (2010a). Strati-
graphic location shown in Fig. 7.129
1388 N. McLoughlin et al.
Fig. 7.131 Transmitted light image of the large microtextures. (a)
Metavolcanic glass fragment (green) surrounded by carbonate-quartz-
opaque matrix, dense alteration on the outer margin of the glass frag-
ment grades to the left into more chlorite rich areas where the individ-
ual large morphotype is visible. (b) The margin of the metavolcanic
glass fragment (IPH matrix upper part of image), band of titanite
“blobs” (lower part of image), segmented microtextures in centre. (c)
Cross-polarised image showing two segmented, large microtextures
(yellow-white) cross-cutting fabric in the fine grained chlorite matrix
(blue-grey). (d) Carbonate-filled vein (top right) with large, segmented,
tapered microtextures propagating into the chlorite matrix. (e) Cross-
polarised image at higher magnification showing further examples of
the large segmented textures. (f) Cross-polarised image revealing a
cross section through a vein coming out of the sample with strongly
taper, segmented microtextures radiating into the chlorite matrix (bluegrey) (Images: (a–c) from sample 88-Pch-07, an interpillow breccia;
(d–e) from sample 117-Pch-07, a pillow rim. Stratigraphic locations are
shown in Fig. 7.129)
8 7.8 Traces of Life 1389
Fig. 7.132 Transmitted light images of tubular textures, large (b–c)
and small (d–e) morphotype from a heavily silicified interpillow
hyaloclastite (IPH). (a) Low-magnification image largely comprising
IPH cement of opaques dispersed in carbonate (grey) with vugs of
quartz (white) and small areas of chlorite (green); (b) Higher magnifi-
cation image showing a chlorite-rich area of metavolcanic glass with
long, curving tubular structure with coarse segmentations; (c)
Enlargement showing two examples of the large microtextures. (d)
Opaque-rich area in centre from which fine tubes radiate into the
surrounding chlorite matrix; terminal sulphides are arrowed. (e) Finetubes propagating from quartz-carbonate cement into the chlorite,
suggestion of longitudinal striae, also a euhedral terminal sulphide
arrowed. Sample 43-Pch-05. Stratigraphic location is shown in
Fig. 7.129
1390 N. McLoughlin et al.
Fig. 7.133 (a) Pillow lava outcrop with a well-defined chilled margins
showing location of the mini-drill core sample; (b–e) Transmitted light
images of microtextures in a pillow rim sample. (b) Low-magnification
image of the thin section showing three areas (c, d, e) where candidate
biotextures have been found. The pillow core is to the left of the image
(note igneous phenocrysts) and the rim with carbonate veining and a
8 7.8 Traces of Life 1391
Several of the FAR-DEEP drill holes have intersected
sections of pillow basalts (9A – 100 m; 12A, 12B and 13A –
c. 90 m), which may provide the opportunity to further
address some of these questions. In particular, these drill
core sections could be used, firstly, to seek new evidence
to distinguish microbial alteration textures from abiotic alter-
ation products in the Palaeoproterozoic sub-seafloor. Sec-
ondly, to delimit the pressure, temperature, and XCO2
conditions of pumpellyite-grade metamorphism to better
document the type of textures formed in sub-greenschist
facies conditions and to compare these to the more well-
known titanite-mineralised textures found in greenschist
facies metavolcanic glasses. Lastly, if compelling evidence
for bioalteration can be found, then the FARDEEP drillcore
may allow investigation of a possible connection between the
rise of oxygen in the Proterozoic ocean, the style of fluid-rock
interactions in the oceanic crust, and the abundance and
distribution of bioalteration textures in the sub-seafloor.
References
Amelin YuV, Heaman LM, Semenov VS (1995) U-Pb
geochronology of layered mafic intrusions in the eastern Baltic
Shield: implications for the timing and duration of Palaeopro-
terozoic continental rifting. Precambrian Res 75:31–46
Bach W, Edwards KJ (2003) Iron and sulphide oxidation
within the basaltic ocean crust: implications for chemolitho-
autotrophic microbial biomass production. Geochim
Cosmochim Acta 67:3871–3887
Banerjee NR, Muehlenbachs K (2003) Tuff life: bioalter-
ation in volcaniclastic rocks from Ontong Java Plateau.
Geochem Geophys Geosyst 4. doi:10.1029/2002GC000470
Banerjee NR, Furnes H, Muehlenbachs K, Staudigel H,
de Wit MJ (2006) Preservation of microbial biosignatures in
3.5 Ga pillow lavas from the Barberton Greenstone Belt,
South Africa. Earth Planet Sci Lett 241:707–722
Banerjee NR, Simonetti A, Furnes H, Staudigel H,
Muehlenbachs K, Heaman L, Van Kranendonk MJ (2007)
Direct dating of Archean microbial ichnofossils. Geology
35:487–490
Benzerara K, Menguy N, Banerjee NR, Tyliszczak T,
Brown GE Jr, Guyit F (2007) Alteration of submarine basal-
tic glass from the Ontong Java Plateau: a STXM and TEM
study. Earth Planet Sci Lett 260:187–200
Bridge NJ, Banerjee NR, Mueller W, Muehlenbachs K,
Chacko T (2010) A volcanic habitat for early life preserved
in the Abitibi Greenstone belt, Canada. Precambrian Res
179:88–98
Cavalazzi B, Westall F, Cady SL, Barbieri R, Foucher F
(2011) Potential fossil endoliths in vesicular pillow basalt,
Coral Patch Seamount, Eastern North Atlantic Ocean. Astro-
biology 11:619–632
Cockell CS, van Calsteren P, Mosselmans JFW, Franchi
IA, Gilmour I, Kelley L, Olsson-Francis K, Johnson D, JC24
shipboard scientific party (2010) Microbial endolithic colo-
nization and the geochemical environment in young seafloor
basalts. Chem Geol 279:17–30
Corfu F, Evins PM (2002) Late Palaeoproterozoic mona-
zite and titanite U-Pb ages in the Archaean Suomujarvi
Complex, N-Finland. Precambrian Res 116:171–181
Daughney CJ, Rioux JP, Fortin D, Pichler T (2004) Lab-
oratory investigation of the role of bacteria in the weathering
of basalt near deep sea hydrothermal vents. Geomicrobiol J
21:21–31
Edwards KJ, Bach W, McCollom TM (2005)
Geomicrobiology in oceanography: microbe-mineral
interactions at and below subseafloor. Trends Microbiol
13:449–456
Fisk MR, Giovannoni SJ, Thorseth IH (1998) The extent
of microbial life in the volcanic crust of the ocean basins.
Science 281:978–979
Fisk MR, Staudigel H, Smith DC, Haveman SA (1999)
Evidence of microbial activity in the oldest ocean crust. EOS
80:F84–F85
Fliegel D, Wirth R, Simonetti A, Schreiber A, Furnes H,
Muehlenbachs K (2011) Tubular textures in pillow lavas
from a Caledonian west Norwegian ophiolite: a combined
TEM, LA-ICP-MS, and STXM study. Geochem Geophys
Geosyst 12. doi:10.1029/2010GC003255
Fliegel D, Wirth R, McLoughlin N, Simonetti A,
Furnes H, Staudigel H, Muehlenbachs K, Hanski E (2010a)
Septate tubular textures in 2.0 Ga pillow lavas from
the Pechenga Greenstone Belt, a nano-spectroscopic
approach to investigate their biogenicity. Geobiology
8:372–390
Fliegel D, Kosler J, McLoughlin N, Simonetti A, de Wit
MJ, Wirth R, Furnes H (2010b) In situ dating of Earth’s
oldest trace fossil at 3.34 Ga. Earth Planet Sci Lett 299:
290–298
Furnes H, Muehlenbachs K (2003) Bioalteration recorded
in ophiolitic pillow lavas. In: Dilek Y, Robinson PT (eds)
Ophiolites in Earth’s history, Geological Society of London
Special Publication 218:415–426
Fig. 7.133 (continued) more yellow colour to the right; (c) A central
carbonate-quartz vein running north–south with stubby microtextures
that overlap at high angles in the chlorite groundmass, (d) Cross-
polarised image showing an intergrowth of microtextures in the chlorite
matrix that intersect at high angles. On the right hand margin is a
carbonate vein containing “blobs” of titanite; (e) Fine tubular textures
propagating from a rhombic vug (dashed outline) and a thin fracture
(dashed line) lower right. Note the darker terminations of some of the
tubes suggesting a terminal infilling. Sample 44-Pch-07. See Fig. 7.129
for location of the sample
1392 N. McLoughlin et al.
Furnes H, Banerjee NR, Muehlenbachs K, Staudigel H,
de Wit MJ (2004) Early life recorded in Archean pillow
lavas. Science 304:578–581
Furnes H, Staudigel H (1999) Biological mediation in
ocean crust alteration: how deep is the deep biosphere?
Earth Planet Sci Lett 166:97–103
Furnes H, Staudigel H, Thorseth IH, Torsvik T,
Muehlenbachs K, Tumyr O (2001a) Bioalteration of basaltic
glass in the oceanic crust. Geochem Geophys Geosyst
2 (paper number GC000170)
Furnes H, Muehlenbachs K, Torsvik T, Thorseth IH,
Tumyr O (2001b) Microbial fractionation of carbon isotopes
in altered basaltic glass from the Atlantic Ocean, Lau Basin
and Costa Rica Rift. Chem Geol 173:313–330
Furnes H, Muehlenbachs K, Torsvik T, Tumyr O, Lang S
(2002) Bio-signatures in metabasaltic glass of a Caledonian
ophiolite, West Norway. Geol Mag 139:601–608
Furnes H, Banerjee NR, Muehlenbachs K, Kontinen A
(2005) Preservation of biosignatures in the metaglassy vol-
canic rocks from the Jormua ophiolite complex, Finland.
Precambrian Res 136:125–137
Furnes H, Banerjee NR, Staudigel H, Muehlenbachs K,
McLoughlin N, de Wit M, Van Kranendonk M (2007)
Bioalteration textures in recent to mesoarchean pillow
lavas: a petrographic signature of subsurface life in oceanic
igneous rocks. Precambrian Res 158:156–176
Furnes H, McLoughlin N, Muehlenbachs K, Banerjee
NR, Staudigel H, Dilek Y, de Wit M, Van Kranendonk M,
Schiffmann P (2008) Oceanic pillow lavas and hyaloclastites
as habitats for microbial life through time: a review. In:
Dilek Y, Furnes H, Muehlenbachs K (eds) Links between
geological processes, microbial activities, and evolution of
life. Springer Book Series, Berlin, pp 1–68
Golubic S, Perkins RD, Lukas KJ (1975) Boring
microorganisms and microborings in carbonate substrates.
In: Frey RW (ed) The study of trace fossils. Springer, Berlin,
pp 229–259
Hanski EJ, Huhma H, Smolkin VF, Vaasjoki M (1990)
The age of the ferropicritic volcanics and comagmatic
intrusions at Pechenga, Kola Peninsula, USSR. Bull Geol
Soc Finland 62:123–133
Hellevang H (2008) On the forcing mechanism for the
H2-driven deep biosphere. Int J Astrobiol 7:157–167
Knoll AH, Barghoorn ES (1974) Ambient pyrite in Pre-
cambrian chert: new evidence and a theory. Proc Natl Acad
Sci USA 71:2329–2331
Kruber C, Thorseth IH, Pedersen RB (2008) Seafloor
alteration of basaltic glass: textures, geochemistry and endo-
lithic microorganisms. G3 Geochem Geophys Geosyst 9.
doi:10.1029/2008GC002119
Korja A, Heikkinen P (2005) The accretionary
Svecofennian orogen – insight from the BABEL profiles.
Precambrian Res 136:241–268
Lepot K, Philippot P, Benzerara K, Wang Y (2009)
Garnet-filled trails associated with carbonaceous matter
mimicking microbial filaments in Archean basalt.
Geobiology 7:393–402
Lepot K, Benzerara K, Philippot P (2011) Biogenic ver-
sus metamorphic origins of diverse microtubes in 2.7 Gyr
old volcanic ashes; multi-scale investigations. Earth Planet
Sci Lett 312:37–47
Mason OU, Stingl U, Wilhelm LJ, Moeseneder MM, Di
Meo-Savoie CA, Fisk MR, Giovannoni SJ (2007) The phy-
logeny of endolithic microbes associated with marine
basalts. Environ Microbiol 9:2539–2550
McLoughlin N, Brasier MD, Wacey D, Green OR,
Perry RS (2007) On biogenicity criteria for endolithic
microborings on early earth and beyond. Astrobiology
7:10–26
McLoughlin N, Furnes H, Banerjee NR, Staudigel H
(2009) Ichnotaxonomy of microbial trace fossils in volcanic
glass. J Geol Soc Lond 166:159–169
McLoughlin N, Staudigel H, Furnes H, Eickmann B,
Ivarsson M (2010a) Mechanisms of microtunneling in rock
substrates: distinguishing endolithic biosignatures from abi-
otic microtunnels. Geobiology 8:245–255
McLoughlin N, Fliegel DJ, Furnes H, Staudigel H, Simonetti
A, Zhao GC, Robinson PT (2010b) Assessing the biogenicity
and syngenicity of candidate bioalteration textures in pillow
lavas of the 2.52 Ga Wutai Greenstone Terrane of China.
Chinese Sci Bull 55:188–199
McLoughlinN,WaceyD,Kruber C,KilburnMR, Thorseth
IH, Pedersen RB (2011) A combined TEM and NanoSIMS
study of endolithic microfossils in altered seafloor basalt.
Chem Geol. doi:10.1016/j.chemgeo.2011.07.022
Melezhik VA, Sturt BA (1994) General geology and
evolutionary history of the early Proterozoic Polmak-
Pasvik-Pechenga-Imandra/Varzuga-Ust’Ponoy Greenstone
Belt in the north-eastern Baltic Shield. Earth Sci Rev
36:205–241
Montague KE, Walton AW, Hasiotis ST (2010) Euendo-
lithic microborings in basaltic glass fragments in
hyaloclastites: extending the ichnofabric index to microbio-
erosion. Palios 25:393–399
Moore JG (1965) Petrology of deep-sea basalt near
Hawaii. Am J Sci 263:40–52
Nelson DR (2005) GSWA geochronology dataset. In:
Compilation of geochronology data, June 2005 update,
Western Australia Geological Survey, GSWA 178042
8 7.8 Traces of Life 1393
Peckmann J, BachW,BehrensK, Reitner J (2008) Putative
cryptoendolithic life in Devonian pillow basalt, Rheinisches
Schiefergebirge, Germany. Geobiology 6:125–135
Petrov VP, Voloshina ZM (1995) Regional metamor-
phism of the Pechenga area rocks. In: Mitrofanov FP,
Smol’kin VF (eds) Magmatism, sedimentogenesis and
geodynamics of the Pechenga Paleorift. Kola Science Cen-
tre/Russian Academy of Sciences, Apatity, pp 164–182 (in
Russian)
Rouxel O, Dobbek N, Ludden J, Fouquet Y (2003) Iron
Isotope fractionation during oceanic crust alteration. Chem
Geol 202:155–182
Rouxel O, Ono S, Alt J, Rumble D, Ludden J (2008)
Sulfur isotope evidence for microbial sulfate reduction in
altered oceanic basalts at ODP Site 801. Earth Planet Sci
Lett 268:110–123
Santelli CM, Orcutt BN, Banning E, Bach W, Moyer CL,
Sogin ML, Staudigel H, Edwards K (2008) Abundance and
diversity of microbial life in ocean crust. Nature
453:653–657
Simonetti A, Heaman LM, Chacko T, Banerjee NR
(2006) In situ petrographic thin section U–Pb dating of
zircon, monazite, and titanite using laser ablation-MC-ICP-
MS. Int J Mass Spectrom 253:87–97
Smits MM (2006) Mineral tunnelling by fungi. In: Gadd
GM (ed) Fungi in biogeochemical cycles. Cambridge Uni-
versity Press, Cambridge, pp 68–717
Staudigel H, Chastain R, Yaynos A, Bourcier R (1995)
Biologically mediated dissolution of glass. Chem Geol
126:147–154
Staudigel H, Yayanos A, Chastain R, Davies G,
Verdurmen EATh, Schiffman P, Bourcier R, De Baar H
(1998) Biologically mediated dissolution of volcanic glass
in seawater. Earth Planet Sci Lett 164:233–244
Staudigel H, Furnes H, Kelley K, Plank T, Muehlenbachs
K, Tebo B, Yayanos A (2004) The oceanic crust as a biore-
actor. In: Deep Subsurface Biosphere at Mid-Ocean Ridges,
vol 144, AGU Monograph. American Geophysical Union,
Washington, pp 325–341
Staudigel H, Furnes H, Banerjee NR, Dilek Y,
Muehlenbachs K (2006) Microbes and volcanos: a tale
from the oceans, ophiolites and greenstone belts. GSA
Today 16(10). doi: 10.1130/GSAT01610A.1
Staudigel H, Furnes H, McLoughlin N, Banerjee NR,
Connell LB, Templeton A (2008) 3.5 Billion years of glass
bioalteration: volcanic rocks as a basis for microbial life?
Earth Sci Rev 89:156–176
Storrie-Lombardi MC, Fisk MR (2004) Elemental abun-
dance distributions in sub oceanic basalt glass: evidence of
biogenic alteration. Geochem Geophys Geosyst 5.
doi:10.1029/2004GC000755, Art. No. Q10005
Templeton AS, Knowles EJ, Eldridge DL, Arey BW,
Dohnalkova AC, Webb SM, Bailey BE, Tebo BM, Staudigel
HS (2009) A seafloor microbial biome hosted within incipi-
ent ferromanganese crusts. Nat Geosci 2:872–876
Thorseth IH, Furnes H, Tumyr O (1991) A textural and
chemical study of Icelandic palagonite of varied composi-
tion and its bearing on the mechanism of the glass-
palagonite transformation. Geochim Cosmochim Acta 55:
731–749
Thorseth IH, Furnes H, Heldal M (1992) The importance
of microbiological activity in the alteration of natural basal-
tic glass. Geochim Cosmochim Acta 56:845–850
Thorseth IH, Torsvik T, Furnes H, Muehlenbachs K
(1995) Microbes play an important role in the alteration of
oceanic crust. Chem Geol 126:137–146
Thorseth IH, Torsvik T, Torsvik V, Daae FL, Pedersen
RB (2001) Diversity of life in ocean floor basalts. Earth
Planet Sci Lett 194:31–37
Thorseth IH, Pedersen RB, Christie DM (2003) Microbial
alteration of 0–30-Ma seafloor and sub-seafloor basaltic
glasses from the Australian Antarctic Discordance. Earth
Planet Sci Lett 215:237–247
Torsvik T, Furnes H, Muehlenbachs K, Thorseth IH,
Tumyr O (1998) Evidence for microbial activity at the
glass-alteration interface in oceanic basalts. Earth Planet Sci
Lett 162:165–176
Tyler SA, Barghoorn ES (1963) Ambient pyrite grains in
Precambrian cherts. Am J Sci 261:424–432
Van Kranendonk MJ, Hickman AH, Smithies RH, Nelson
DN, Pike G (2002) Geology and tectonic evolution of the
Archaean North Pilbara terrain, Pilbara Craton, Western
Australia. Econ Geol 97:695–732
Walton AW (2008) Microtubules in basalt glass from
Hawaii Scientific Drilling Project #2 phase 1 core and
Hilina slope,Hawaii: evidence of the occurrence and behaviour
of endolithic microorganisms. Geobiology 6:351–364
Zen E-An (1974) Prehnite- and pumpellyite-bearing
mineral assemblages, west side of the Appalachian meta-
morphic belt, Pennsylvania to Newfoundland. J Petrol
15:197–242
1394 N. McLoughlin et al.
7.8.5 Biomarkers and Isotopic Tracers
Roger E. Summons, Christian J. Illing,Mark van Zuilen, and Harald Strauss
Introduction
Molecular and isotopic fossils preserved in sedimentary rocks
are important sources of palaeobiological information and are
especially useful for reconstructing the histories and extent of
microbially-driven biogeochemical processes that are largely
inaccessible through other kinds of palaeontological and geo-
chemical records. Biogeochemical studies of organic matter-
rich sedimentary rocks deposited during the Phanerozoic Eon
afford particularly detailed records of phenomena such as
palaeoenvironmental settings (Brassell et al. 1983), past
climates and climate change (Brassell et al. 1986), ocean
redox and the history of ocean plankton (Damste et al. 2004;
Holba et al. 1998; Knoll et al. 2007). Going back further in
time, studies of Cambrian and Proterozoic sediments and
petroleum provide lines of evidence for the nature of early
photosynthetic communities (Moldowan and Talyzina 1998)
and the appearance of the first animals (Love et al. 2009).
However, the older the rocks one would like to examine, the
more difficult it becomes to extract reliable records.
Sediments are lost to uplift, erosion and subduction while
movement of basinal fluids, the effects of heating and ionising
radiation over long periods of time and other corrupting
influences degrade the record that remains (Dahl et al.
1988). Loss and/or replacement of residual carbon compounds
destroys the primary geochemical signature which, when
combined with analytical uncertainties, leads to doubt the
originality and syngeneity of biomarkers and isotopic fossils.
At this point, the presence of molecular and isotopic fossils
derived from more recent organisms becomes difficult to
discern unless we have the means to distinguish what is
original and what is contamination (e.g. Gerard et al. 2009).
Precambrian Sedimentary Organic Matter: TheSearch for Molecular Fossils
Organic matter preserved in sedimentary rock comprises
a predominant insoluble macromolecular component (kero-
gen) and a mobile, solvent soluble component (bitumen).
Biomarkers are hydrocarbons, which can be linked, directly
or indirectly, to biological sources; this makes them to
molecular fossils. Despite the fact that hydrocarbons can
be formed by some non-biological means such as Fischer-
Tropsch chemistry (high-temperature, metal-catalysed
reduction and polymerisation reactions of CO and H2) that
are hypothesised to occur in hydrothermal environments and
deep in the crust (McCollom et al. 1999; Sherwood Lollar
et al. 2002), it is difficult to envision processes to explain
how such relatively simple hydrocarbons could become
concentrated, polymerised and entrained in diverse sedimen-
tary rocks. The most parsimonious explanation for the origin
of the kerogens that are ubiquitous in sedimentary sequences
of all ages is that they result from biological processes. As
such, kerogen is, in itself, a potential fossil.
While kerogen is widespread, its complex, macromolec-
ular nature makes it very difficult to characterise and to
extract signals that could be diagnostic for particular
organisms or processes. At low metamorphic grades, kero-
gen can be broken down into tractable components, includ-
ing saturated hydrocarbon biomarkers, using pyrolysis or
hydropyrolysis (Love et al. 1995; Seifert 1978). Kerogens
from the Phanerozoic and Proterozoic have yielded valuable
molecular data when processed in this way (Bowden et al.
2006; Love et al. 2009). However, in older rocks, which tend
to be at the higher end of the metamorphic spectrum, kero-
gen pyrolysates are largely comprised of polyaromatic
hydrocarbons, which tend to carry less information than
their saturated counterparts (Brocks et al. 2003).
If they are preserved, the hydrocarbons present in the
bitumen component of organic matter in Palaeoproterozoic
and Archaean sedimentary rocks should carry an abundance
of geobiological information about microbial communities
in existence prior to, during and after the “Great Oxidation
Event” (Brocks and Summons 2003; Sessions et al. 2009).
This could be an exceptionally important source of informa-
tion to complement studies of inorganic proxies for surface
oxygenation. Studies of 1.4–1.7 Ga low metamorphic grade
sedimentary rocks from the McArthur Basin in northern
Australia (Jackson et al. 1988; Page and Sweet 1998), for
example, have revealed the presence of complex
distributions of carotenoid-derived hydrocarbons that are
unambiguously linked to purple and green sulphur bacteria
(Brocks et al. 2005; Brocks and Schaeffer 2008), which, in
turn, can only proliferate where sulphidic waters protrude
into the photic zone. Thus, these particular biomarkers are
informative about the organisms that produced them and
their favoured sedimentary environment. Other molecular
proxies can be indicative of organisms with particular phys-
iological capabilities such as the O2-dependent sterol bio-
synthesis of eukaryotes, oxygenic photosynthesis in the
case of the hopanoid-producing cyanobacteria, and aerobic
methane oxidation in alphaproteobacteria, which produce
distinctive 3-methylhopanoids and 4-methylsterols.
Molecules such as these have been used to infer the antiquity
of these physiologies and, therefore, provide time constraints
on the evolution of Earth’s oxygen budget.
R.E. Summons (*)
Department of Earth, Atmospheric and Planetary Sciences,
Massachusetts Institute of Technology, 77 Massachusetts Avenue,
E25-633 Cambridge, MA 02139, USA
8 7.8 Traces of Life 1395
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_8, # Springer-Verlag Berlin Heidelberg 2013
1395
Three major areas of uncertainty have been identified in
respect to interpreting the Precambrian molecular fossil
record. The first and foremost uncertainty concerns the
issue of contamination during acquisition, storage and
handling. This is important in all biomarker studies no mat-
ter what the age of the rocks is, but it is especially critical in
studies of the most ancient sediments where thermal matu-
rity tends to be high and hydrocarbon contents low. Potential
problems are best addressed using fastidious laboratory
techniques, appropriate blanks for each stage of the analysis
and a quantitative approach that allows one to demonstrate
that the ancient biomarker signal is well in excess of the
inevitable background of petroleum-derived hydrocarbons.
A second area of perennial uncertainty is the question of
whether or not hydrocarbon biomarkers hosted in ancient rocks
are indigenous. Doubt about the originality of these molecules
has always existed because of the exceptionally low elemental
hydrogen to carbon ratios of kerogen, the difficulty of exclud-
ing an origin from oil migrating at some time after deposition,
or even modern contaminants added during or after drill-core
retrieval (Brocks et al. 2008; Hoering 1965; Hoering
and Navale 1987; Leventhal et al. 1975). Though many lines
of evidence point towards the ancient origin of some
hydrocarbons in ancient rocks, it is very difficult to absolutely
rule out a younger origin for others (Brocks et al. 2003).
The most recent discussions of contamination and
syngeneity issues center on the findings of the potentially
oldest biomarker evidence for eukaryotes and oxygenic pho-
tosynthesis reported by Brocks et al. (1999) in ~2.7 Ga sedi-
mentary rocks from the Hamersley and Fortescue Groups,
Pilbara Craton, Western Australia. Resampling of these sed-
imentary rocks and in situ NanoSims analyses by Rasmussen
et al. (2008) revealed distinct carbon isotope results for
kerogens, pyrobitumen and the previously reported extract-
able hydrocarbons. These findings contrast the original
observations by Brocks et al. (1999) and prompted
Rasmussen et al. (2008) to cast doubt on the syngeneity of
the biomarkers and, hence, the potentially far-reaching
conclusions in respect to the appearance of eukaryotes (see
Chap. 7.8.3) and the advent of oxygenic photosynthesis (see
Chap. 8). However, the paper by Rasmussen et al. does not
specify the stratigraphic placement of any of the samples and,
in fact, the samples used for NanoSims measurements of the
d13C of the solid phases were different from those used to
measure the volatile ‘biomarkers’ (Brocks 2011). Also, sig-
nificant doubts surround the accuracy of the analytical
approaches. Hence, the work reported in Rasmussen et al.
(2008) cannot be said to falsify the previously published
work on Archaean biomarkers. Consequently, resolution of
this problem awaits the results of further work on different
and freshly drilled core. Given that early Palaeoproterozoic
biomarkers have been obtained from contaminant-free fluid
inclusions from Canada (Dutkiewicz et al. 1998, 2006;
George et al. 2008), hydrocarbon biomarker survival for
billions of years is clearly feasible.
Despite the difficulties of excluding hydrocarbon migra-
tion fromyounger sediments, there are several approaches that
can be used to assess the syngeneity of bitumens in Archaean
and Palaeoproterozoic sedimentary rocks. One approach is to
examine the spatial distribution of hydrocarbons in a sediment
in order to try to deduce presence and extent of overprinting by
external contaminants (Brocks 2011). However, the
experiments described in the latter paper were performed on
old existing samples which have been exposed to the atmo-
sphere for many years and where surficial contamination is to
be expected. Also, the ‘slice experiments’ that are described
are limited in the amount of sample used and afforded
extremely low yields of hydrocarbons. It is, therefore, not
surprising that practically no hydrocarbons could be detected
on the inside slices. While the approach of examining the
spatial distributions of hydrocarbons in fresh cores is an
exceedingly promising way to address question of syngeneity
in a robust way, experiments on exposed old cores cannot be
held to falsify all previous work done on rocks of this age.
Another approach is based on examination of correlations
between the biomarkers of interest and aspects of the rocks
chemistry that is unlikely to be affected by organic contami-
nation such as primary mineralogy or the stable isotopic
compositions of immobile components such as kerogen.
This approach was employed by Eigenbrode and others in
their study ofmethylhopane distributions in kerogenous rocks
from the Pilbara Craton inWestern Australia (Eigenbrode and
Freeman 2006; Eigenbrode et al. 2008) where they showed
that the relative abundances of 3-methylhopanes was strongly
correlated to the C-isotopic composition of kerogens in the
same samples. In their study of hydrocarbons in the
Griqualand Basin of South Africa,Waldbauer and co-workers
evaluated the maturity-dependent hydrocarbon distributions
and observed a striking contrast across a two-billion-year-
unconformity between the Late Palaeozoic Dwyka Formation
and theNeoarchaean sediments of theCampbellrand Platform
(Waldbauer et al. 2009). Further, the down-core distributions
of some steroid and triterpenoid hydrocarbons were stratigra-
phically correlated in two holes approximately 24 km apart.
One of the key techniques developed in the latter study
involved comparisons of hydrocarbons that were freely
extractable from the rocks (bitumen-1) with the hydrocarbons
that were only accessible after dissolution of the mineral
matrix (bitumen-2) (Sherman et al. 2007; Hallmann et al.
2011). Almost invariably, the hydrocarbons in bitumen-
2 show patterns that reflect the diagenetic influence of clay
minerals and suggesting that they had been intimately
associated with the clay fraction since deposition. Further
studies are underway to test this hypothesis on samples from
a range of Palaeozoic to Archaean sediments.
The third area of uncertainty is a more general one and
relates to the interpretation of the biomarker evidence itself.
Even if most biomarkers are linked to specific biological
sources, these links are not always fully understood or there
is more than one potential source for a given biomarker. An
1396 R.E. Summons et al.
example of this kind of ambiguity surrounds the origins of 2-
methylhopanoids, a biomarker proposed as a proxy for
cyanobacteria (Summons et al. 1999). More recently, Rashby
et al. (2007) found these hopanoids in anoxygenic phototrophs
and Welander et al. (2010) reported the gene encoding
the enzyme responsible for the characteristic methylation
at carbon-2 of the hopanoid ring system in organisms other
than cyanobacteria. This research demonstrates that only a
better understanding of the biosynthetic processes and
the genetic coding of the enzymatic reactions in microbes
can resolve respective uncertainties. The ongoing studies of
methylhopanoid biosynthetic pathways, including the origins
of 3-methylhopanoids, and the physiological functions of
hopanoid lipids offer much promise for a more nuanced inter-
pretation of the patterns of biomarker hydrocarbon occurrence
(Sessions et al. 2009). These are among the problems being
addressed by the vastly evolving field of geobiology where
geology, biochemistry, microbiology and genomics merge to
provide new and unexpected insights (e.g. Waldbauer et al.
2011)
Isotopic Biogeochemistry of PrecambrianSediments
Since the early days in stable isotope geochemistry, scholars
of Earth’s early evolution have exploited this field in their
quest to identify biologically driven processes and unravel the
conditions under which life on Earth emerged and evolved.
The base for this approach is the observation that biological
processes are associated with a (frequently substantial) isoto-
pic fractionation. Most prominently, researchers utilised the
carbon and sulphur isotopic systems (e.g. Hayes et al. 1983;
Strauss et al. 1992), with few supplementary studies on the
nitrogen isotopic composition of sedimentary organic matter
(Garvin et al. 2009; Godfrey and Falkowski 2010; Thomazo
et al. 2011). More recently, the application of non-traditional
stable isotopes (such as Fe, Mo, Cr, U) provides important
environmental constraints (see Chaps. 7.10.4, 7.10.5, and
7.10.6), which represent supporting evidence for the possible
presence of microbially driven processes.
The carbon isotopic composition is expressed as d13C ¼([(13C/12C)sample/(
13C/12C)standard] � 1)*1,000 in per mil (‰),
relative to the Vienna Pee Dee Belemnite (VPDB) standard.
A simplified view of the carbon-bearing compartments
within the global carbon cycle (Fig. 7.134) reveals that
these are characterised by distinct carbon isotopic
compositions. Starting from atmospheric carbon dioxide
with a pre-industrial d13C value of �6.5 ‰ (e.g. Trudinger
et al. 1999), a partitioning between photosynthetically pro-
duced organic matter and marine bicarbonate results in two
isotopically distinct carbon pools in the sedimentary realm.
Fresh organic matter exhibits a 13C-depleted carbon isotopic
composition resulting from a kinetic isotope effect associated
with the preferential biological turnover of 12CO2 during
primary production (e.g. DesMarais 2001). The consequence
of isotopic equilibrium between atmospheric carbon dioxide
and dissolved bicarbonate is a carbon isotope value for
dissolved inorganic carbon near 0 ‰ (e.g. Hoefs 2009). In
principle, both isotopic compositions are being preserved
during burial and/or lithification and, thus, archived in the
rock record. Thermal alteration during diagenesis and pro-
grade metamorphism will change the d13C of sedimentary
carbonaceous matter. Preferential breaking of the 12C–12C
bond over the 12C–13C bond during thermal cracking of
organic matter leads to the loss of isotopically light carbon
compounds and a 13C-enriched residue. In contrast, the car-
bon isotopic composition of marble resembles closely the
isotopic composition of its unmetamorphic precursor, i.e.
sedimentary carbonate (e.g. Baker and Fallick 1989a,
1989b; Melezhik et al. 2005). Finally, subduction of
carbon-bearing sediments recycle both, the reduced organic
and oxidised carbonate carbon, thereby homogenising the
carbon isotopic composition in the resulting carbon dioxide.
From this simplified view it becomes apparent that carbon
isotopic data for its oxidised and reduced forms, when exam-
ined in stratigraphic and palaeoenvironmental context and,
ideally, in paired samples, are useful for reconstructing aspects
of the global carbon cycle. Carbon isotopic data for carbonate
carbon and organic carbon through deep time and in sedimen-
tary rocks representing vastly different environments show a
continuous 20–30 ‰ separation (see Chap. 7.6). As this is in
the same range as the predominant biological fractionation
factor today (e.g. the Calvin-Benson cycle; a detailed discus-
sion of the various biological pathways involved in carbon
isotope fractionation is given in Zerkle et al. 2005, and
references therein), these data have been taken to imply the
operation of a biogeochemical carbon cycle for as far back as
the record of sedimentary rocks extends (Schidlowski 2001).
This includes the implication that the relative fluxes of car-
bonate precipitation and organic carbon burial have remained
constant over time (Des Marais 2001). Some large shifts to
negative organic d13C and positive carbonate d13C during the
late Neoarchaean and early Palaeoproterozoic, however, sug-
gest that perturbations in the global carbon cycle did occur (see
Chap. 7.3, and Baker and Fallick 1989a, 1989b; Karhu and
Holland 1996; Schidlowski 2001; Melezhik et al. 2007).
Acknowledging the strong similarity in the apparent iso-
topic fractionation between the inorganic carbon source
(represented by the carbonate carbon record) and the organic
carbon product (bulk rock sedimentary organic carbon and/
or kerogen), it appears logical to conclude that marine pri-
mary productivity in the surface waters of the early ocean
represents the prime input of organic matter in (reasonably)
well preserved Precambrian sedimentary rocks. Yet, no con-
clusion about the organism(s) responsible for ancient pri-
mary productivity can be drawn from the carbon isotopic
record alone. Moreover, because no distinct difference exists
in the organic carbon isotopic composition of biomass pro-
duced by oxygenic versus anoxygenic photosynthesis (for a
8 7.8 Traces of Life 1397
different view see Nisbet et al. 2007), the carbon isotopic
record(s) obtained from Precambrian sediments provide no
evidence for the onset of oxygenic photosynthesis and,
hence, do not reveal the starting point of this important
biologically driven process on Earth.
As mentioned above, thermal alteration changes the d13Cof sedimentary carbonaceous matter. In general, it can be
assumed that Archaean and Proterozoic carbonaceous
materials that have experienced lower-greenschist metamor-
phism (c. 300 �C) have obtained a d13C value that is enriched
relative to the original value at most by up to 3 ‰ (Des
Marais 1997; Hayes et al. 1983; Watanabe et al. 1997). Thus,
despite the fact that carbon isotopic records can be affected
by diagenesis and burial metamorphism, they appear to be
one of the most robust proxies for the existence of a
biological process on Earth. However, additional processes
during increasing metamorphism can influence the d13Cvalue and these include isotope exchange with carbonates
and isotope exchange with CO2-rich fluids (Kitchen and
Valley 1995; Robert 1988; Schidlowski et al. 1979). These
processes shift the d13C of sedimentary biological material
to significantly higher values and potentially lower the d13Cof carbonate. Consequently, this can complicate the inter-
pretation of the isotopic data, since the primary biological
and the carbonate reference signal are converging.
Just as in the modern marine realm, biomass resulting
from primary productivity was recycled in the ancient ocean
either in the water column and/or the marine sedimentary
column, possibly through a series of largely microbially
mediated processes. In the modern ocean, aerobic respiration
represents the key process responsible for recycling of
organic matter settling through the water column. Some-
where in the upper sedimentary column, the oxygen demand
for aerobic respiration outgrows the diffusive oxygen supply
from the overlying water column (e.g. Canfield et al. 2005).
This marks the spatial onset of anaerobic processes. Given
the present-day concentration of oceanic sulphate, bacterial
sulphate reduction represents the key anaerobic process for
the recycling of sedimentary organic matter in the contem-
porary marine realm. Considering the low abundances of
free atmospheric oxygen prior to the Great Oxidation
Event some 2.3 Ga ago, anaerobic processes likely thrived
in the water column and were largely responsible for the
remineralisation of organic matter (Fallick et al. 2008).
The sulphur isotopic composition is expressed as d34S ¼([(34S/32S)sample/(
34S/32S)standard] � 1)*1,000 in per mil (‰),
relative to the Vienna Canyon Diablo troilite (VCDT) stan-
dard. In the sedimentary realm (Fig. 7.135), dissolved oceanic
sulphate represents the largest sulphur pool. Its present sulphur
isotopic composition (d34S) is at +21 ‰ (e.g. B€ottcher et al.
2007).Without any significant isotopic fractionation and again
being a consequence of an isotopic equilibrium, this signature
is being transferred into evaporitic sulphate minerals (e.g.
Raab and Spiro 1991). Under anoxic conditions, dissolved
sulphate represents the prime terminal electron acceptor for
the microbial recycling (anaerobic respiration) of sedimentary
organic matter. This process of bacterial sulphate reduction is
a key factor inmarine environments. In themodern ocean, due
to the fact that sulphate-reducing bacteria are strictly anaero-
bic, this reaction linking the sulphur and carbon cycles occurs
in the sediment after oxygen has been exhausted during aero-
bic respiration. Under anoxic water column conditions, anaer-
obic processes could have thrived in the ocean waters.
Acknowledging a substantial sulphur isotopic fractionation
associated with bacterial sulphate reduction (for a review, see
Canfield 2001), researchers have applied sulphur isotope geo-
chemistry in order to trace the biological sulphur cycling back
into the distant past. As with carbon, it is the isotopic differ-
ence between the oxidised substrate (i.e. sulphate) and the
reduced product (i.e. sulphide) that is most informative about
the process itself. However, sulphur is more problematic in
this respect because of the ready solubility of most forms of
sulphate. While there is a substantial isotope record for Pre-
cambrian sedimentary sulphide (e.g. Strauss 2002), only frag-
mentary knowledge exists in respect to the sulphur isotopic
composition of seawater sulphate (e.g.Strauss 2004; Lyons
and Gill 2010; see Chap. 7.5). Carbonate-associated sulphate
is seen as an excellent recorder of this signal and the isotopic
separation of pyrite-sulphur and carbonate-associated sulphate
has proven useful for evaluating reservoir sizes and the effect
of ocean–atmosphere redox on the dynamics of the sulphur
cycle. Based on respective data it can be concluded that
sulphate reducing bacteria were ubiquitously active in the
marine realm at least as early as 2.3 Ga (e.g. Strauss 2002) if
not 2.7 Ga (e.g. Grassineau et al. 2001). Claims for an even
earlier presence of this metabolic pathway are based on
respective data for microcrystalline pyrite from the c. 3.5 Ga
barite occurrences inWestern Australia (e.g. Shen et al. 2009).
Alternatively (or additionally), Philippot et al. (2007) interpret
their sulphur isotope data for the same stratigraphic unit as
indicative of sulphur disproportionation.Detailed evidence for
the disproportionation of sulphur intermediates (such as ele-
mental sulphur or thiosulphate), requiring an oxidative step in
order to derive these intermediates from sulphide, is archived
in sediments as old as 1450Ma (Johnston et al. 2005). Similar
to carbon, numerous Precambrian sedimentary rocks and their
diagnostic sulphur isotopic signals have largely escaped their
obliteration during sediment diagenesis and subsequent meta-
morphism, providing a rich record for the existence of distinct
metabolic pathways within the global sulphur cycle.
Nitrogen, in the form of amino acid, protein and DNA, is
an essential element for life. There are two stable isotopes
(14N and 15N) and their relative abundances in atmospheric N2
are 99.6337 % and 0.3663 %, respectively (Rosman and
Taylor 1998). The nitrogen isotopic composition is expressed
as d15N ¼ ([(15N/14N)sample/(15N/14N)standard] � 1)*1,000, in
per mil (‰) relative to the atmospheric nitrogen standard
(Nier 1950). Unlike most other isotopic pairs, abiotic pro-
cesses and equilibrium reactions result in only minor fraction-
ation effects. The largest isotopic fractionations accompany a
1398 R.E. Summons et al.
few critical biological processes (Hoefs 2009) including
some at the heart of the nitrogen cycle such as nitrification
(i.e. the oxidation of ammonium to nitrate), denitrification
(i.e. the breakdown of nitrate to nitrogen) and the anaerobic
oxidation of ammonium (oxidation of ammonium with nitrate
to nitrogen and water). The process of nitrogen (N2) fixation is
energy-intensive but results in barely any fractionation. On
the other hand, if nitrate is not limited, the processes of
denitrification (Fig. 7.136) can cause very large isotope
fractionations (10–30 ‰; Hoefs 2009).
The nitrogen isotopic composition of sedimentary (organic)
nitrogen versus time (Fig. 7.137) provides insight for
reconstructing the principal nitrogen fixation pathways. A
nitrogen cycle that is dominated byN2-fixationwill be recorded
through d15N values around 0 ‰, the isotopic composition of
atmospheric nitrogen. A nitrogen isotopic composition
between 0 ‰ and +20 ‰ suggests a ‘modern style’ biological
cycling of nitrogen under aerobic conditions (e.g. Sigman and
Casciotti 2001). In contrast, repeated nitrification and denitrifi-
cation, the latter under suboxic to anoxic conditions, results in
the progressive enrichment of 15N (Fig. 7.136). The d15Nsignature is archived within sedimentary matter. For measure-
ment of the signature bulk sediment samples suit best.
In contrast to the isotopic compositions of carbon and
sulphur compounds, there is no sedimentary isotopic refer-
ence signal preserved for the atmospheric nitrogen reservoir.
Moreover, the concentration of nitrogen preserved in the
sediment is much more sensitive to alteration. Since nitrogen
is contained in functionalised molecules, it is very vulnera-
ble to degradation and fast recycling. It is also affected by
early diagenetic alteration (e.g. Vandenbroucke and Largeau
2007). Most biologically-fixed nitrogen is re-released from
the organic matter as part of the nitrogen cycling with only a
minor fraction that is preserved and buried in sedimentary
organic matter. Nitrogen, released from kerogen mainly in
the form of ammonium during dia- and catagenic processes,
is mobilised in pore fluids or brines (Illing and Ostertag-
Henning 2010). Ammonium can then substitute for potas-
sium in clay minerals (c.f. Williams et al. 1989), and since
the expulsion of nitrogen from the sedimentary organic
matter shows only small isotopic fractionation (a few
permil), clay minerals can record the isotopic signature of
the adjacently buried organic matter. However, during meta-
morphism and recrystallisation processes, the accompanying
liberation of ammonium from those clays can result in a
strong fractionation (Bebout and Fogel 1992; Jia 2006) and
this can readily corrupt the archived isotopic signature.
Beaumont and Robert (1999) have shown that the d15N of
kerogen in Archaean metasediments is several permil lower
than that found in the modern biosphere (c. +5 ‰). From this
they suggest an absence of nitrifying and denitrifying bacteria
and the presence of nitrogen fixing bacteria in the mildly
reducing Archaean oceans. In contrast, more recent work on
Archaean rocks has shown a much higher variation of the
isotopic composition of sedimentary nitrogen (Fig. 7.137)
suggesting biological nitrogen cycling (e.g. Garvin et al.
2009; Godfrey and Falkowski 2010; Thomazo et al. 2011).
Consequently, this has been interpreted to indicate the onset of
a modern-style nitrogen cycle ~300 Ma before the Great
Oxidation Event. Garvin et al. (2009) published an organic
nitrogen isotopic profile for the 2.5 Ga old Mount McRae
Shale in Western Australia and identified an anomaly that
complements other redox-sensitive proxies, all of which are
consistent with the hypothesis of Anbar et al. (2007) for a
‘whiff of oxygen’ prior to the Great Oxidation Event. It was
inferred that the Mount McRae Shale “section records an
episode of increased nitrification and denitrification” (Garvin
et al. 2009). Godfrey and Falkowski (2010) made a similar
observation in profiles of the Campbellrand-Malmani platform
in the Griqualand Basin in South Africa (~2.67–2.46 Ga).
Recently, Thomazo et al. (2011) reported “the first evidence
for the onset of the oxidative part of the nitrogen cycle” with
exceptionally high d15N values from metasediments of the
2.72 Ga old Tumbiana Formation that is attributed to ammo-
nium oxidisers limited by the availability of oxidants.
Methodology and Implications for the FAR DEEPCoresThe unique corematerial of FAR-DEEP covers almost 600Ma
of the critical interval across the Archaean-Palaeoproterozoic
transition. The fresh and un-weathered sample material offers
the opportunity to refine the existing stable isotope records
(C, N, S) for this unique interval and to improve our under-
standing of the temporal evolution these records. Thereby,
special emphasis will be given to the intervals covering the
Great Oxidation Event, the Lomagundi-Jatuli Event and the
Shunga Event (see Chaps. 7.3, 7.6, and 8).
Detection and syngeneity assessments of biomarkers that
require molecular oxygen for their biosynthesis (Summons
et al. 2006) will be an important primary objective. The
formation of organic-rich deposits such as those represented
by the Shunga sediments arguably requires a carbon cycle
with oxygenic photosynthesis at its base. Since the drilling
operation and core/sample handling was carefully performed
to minimise the risk of organic contamination, the recovered
sample material is suitable for molecular organic analyses.
This will allow us to search for biomarkers that are diagnostic
for individual organisms, metabolic pathways or environ-
mental conditions. The prime impediments to finding
biomarkers are excessive thermal maturity and exposure
to ionising radiation since both of these will destroy the
small molecules that we are seeking to detect. Hence,
samples will have to be analysed using a rigorous analytical
protocol initially developed by Sherman et al. (2007) and
Waldbauer et al. (2009) with further recent improvements. In
particular, the spatial distribution of hydrocarbons in both
bitumen-1 and bitumen-2 should be compared (Hallmann
et al. 2011). Patterns of hydrocarbons isolated from exposed
surfaces of cores may also be compared with those isolated
from the complementary internal sections of the cores
(Brocks 2011).
8 7.8 Traces of Life 1399
Fig. 7.134 Schematic view of the marine carbon cycle, identifying the main carbon pools, their typical range in isotopic composition and their
links to each other (Modified after Des Marais 1997)
Fig. 7.135 Schematic view of the four main marine sulphur pools (dissolved seawater sulphate, SO42�; magmatic sulfur, sedimentary sulfides
and sulfates) and their pertinent isotopic ranges (After Hoefs 2009; Clark and Fritz 1997)
1400 R.E. Summons et al.
Fig. 7.136 Schematic view of the marine nitrogen cycle, identifying
the main pools (N2(atm), atmospheric nitrogen; N2(sea); dissolved nitro-
gen; NO3�, nitrate; NO2
�, nitrite; NH4+, ammonium, all in seawater;
Norg, nitrogen fixed in living biomass, Nsed, sedimentary nitrogen, fixed
in organic matter or on minerals) and their range in isotopic
composition (based on Casciotti 2009; Shen et al. 2006; Sigman and
Casciotti 2001). The two sedimentary nitrogen pools are exemplary for
the two extreme cases of dominant fixation or dominant denitrification
without a nitrate limitation
Fig. 7.137 Nitrogen isotopic record for the time interval between
3500 and 1500 Ma before present. High d15N values in the rock record
around 2700 Ma point to the onset of the oxidative part of the nitrogen
cycle (Thomazo et al. 2011). The time interval covered by the
FAR-DEEP cores is symbolized by the black bar (Source of isotope
data: Garvin et al. 2009; Godfrey and Falkowski 2010; Thomazo et al.
2009, 2011)
8 7.8 Traces of Life 1401
References
Anbar AD, Duan Y, Lyons TW, Arnold GL, Kendall B,
Creaser RA, Kaufman AJ, Gordon GW, Scott C, Garvin J,
Buick R (2007) A whiff of oxygen before the great oxidation
event? Science 317:1903–1906
Baker AJ, Fallick AE (1989a) Evidence from Lewisian
limestones for isotopically heavy carbon in 2000 million-
year-old seawater. Nature 337:352–354
Baker AJ, Fallick AE (1989b) Heavy carbon in 2-billion-
year-old marbles from Lofoten-Vesteralen, Norway:
implications for the Precambrian carbon-cycle. Geochim
Cosmochim Acta 53:1111–1115
Beaumont V, Robert F (1999) Nitrogen isotope ratios of
kerogens in Precambrian cherts: a record of the evolution of
atmosphere chemistry. Precambrian Res 96:63–82
Bebout GE, Fogel ML (1992) Nitrogen-isotope
compositions of metasedimentary rocks in the Catalina
Schist, California: implications for metamorphic devolati-
lization history. Geochim Cosmochim Acta 56:2839–2849
B€ottcher ME, Brumsack H-J, Duerselen C (2007) The
isotopic composition of modern seawater sulfate: I. Coastal
waters with special regard to the North Sea. J Mar Sys
67:73–82
Bowden SA, Farrimond P, Snape CE, Love GD (2006)
Compositional differences in biomarker constituents of the
hydrocarbon, resin, asphaltene and kerogen fractions: an
example from the Jet Rock (Yorkshire, UK). Org Geochem
37:369–383
Brassell SC, Eglinton G, Maxwell JR (1983) The geo-
chemistry of terpenoids and steroids. Biochem Soc Trans
11:575–586
Brassell SC, Eglinton G, Marlowe IT, Pflaumann U,
Sarnthein M (1986) Molecular stratigraphy: a new tool for
climatic assessment. Nature 320:129
Brocks JJ (2011) Millimeter-scale concentration
gradients of hydrocarbons in Archean shales: live-oil escape
or fingerprint of contamination? Geochim Cosmochim Acta
75:3196–3213
Brocks JJ, Schaeffer P (2008) Okenane, a biomarker for
purple sulfur bacteria (Chromatiaceae), and other new carot-
enoid derivatives from the 1640 Ma Barney Creek Forma-
tion. Geochim Cosmochim Acta 72:1396–1414
Brocks JJ, Summons RE (2003) Sedimentary
hydrocarbons, biomarkers for early life. In: Holland HD,
Turekian K (eds) Treatise in geochemistry 8.03. Elsevier
Academic Press, San Diego, pp 65–115
Brocks JJ, Logan GA, Buick R, Summons RE (1999)
Archean molecular fossils and the early rise of eukaryotes.
Science 285:1033–1036
Brocks JJ, Love GD, Snape CE, Logan GA, Summons RE,
Buick R (2003) Release of bound aromatic hydrocarbons
from late Archean and Mesoproterozoic kerogens via
hydropyrolysis. Geochim Cosmochim Acta 67:1521–1530
Brocks JJ, Love GD, Summons RE, Knoll AH, Logan
GA, Bowden SA (2005) Biomarker evidence for green and
purple sulphur bacteria in a stratified Palaeoproterozoic sea.
Nature 437:866
Brocks JJ, Grosjean E, Logan GA (2008) Assessing bio-
marker syngeneity using branched alkanes with quaternary
carbon (BAQCs) and other plastic contaminants. Geochim
Cosmochim Acta 72:871–888
Canfield DE (2001) Biogeochemistry of sulphur isotopes.
In: Valley JW, Cole DR (eds) Stable isotope geochemistry,
vol 43, Reviews in mineralogy and geochemistry. Geologi-
cal Society of America, Washington, pp 607–633
Canfield DE, Thamdrup B, Kristensen E (2005) Aquatic
Geomicrobiology, vol 48, Advances in Marine Biology.
Elsevier Academic Press, San Diego, p 640
Casciotti KL (2009) Inverse kinetic isotope fractionation
during bacterial nitrite oxidation. Geochim Cosmochim
Acta 73:2061–2076
Clark I, Fritz P (1997) Environmental isotopes in hydro-
geology. CRC Press, Boca Raton/Florida, p 328
Dahl J, Hallberg R, Kaplan IR (1988) The effects of
radioactive decay of uranium on elemental and isotopic
ratios of Alum shale kerogen. Appl Geochem 3:583–589
Damste JSS, Muyzer G, Abbas B, Rampen SW, Masse G,
Allard WG, Belt ST, Robert J-M, Rowland SJ, Moldowan
JM, Barbanti SM, Fago FJ, Denisevich P, Dahl J, Trindade
LAF, Schouten S (2004) The rise of the rhizosolenid
diatoms. Science 304:584–587
Des Marais DJ (1997) Isotopic evolution of the biogeo-
chemical carbon cycle during the proterozoic eon. Org
Geochem 27:185–193
Des Marais DJ (2001) Isotopic evolution of the biogeo-
chemical carbon cycle during the Precambrian. In: Valley
JW, Cole DR (eds) Stable isotope geochemistry, vol 43,
Reviews in mineralogy and geochemistry. Mineralogical
Society of America, Washington, pp 555–578
Dutkiewicz A, Rasmussen B, Buick R (1998) Oil pre-
served in fluid inclusions in Archaean sandstones. Nature
395:885–888
Dutkiewicz A, Volk H, George SC, Ridley J, Buick R
(2006) Biomarkers from Huronian oil-bearing fluid
inclusions: an uncontaminated record of life before the
Great Oxidation Event. Geology 34:437–440
Eigenbrode JL, Freeman KH (2006) Late Archean rise of
aerobic microbial ecosystems. Proc Natl Acad Sci USA
103:15759–15764
Eigenbrode JL, Freeman KH, Summons RE (2008)
Methylhopane biomarker hydrocarbons in Hamersley prov-
ince sediments provide evidence for Neoarchean aerobiosis.
Earth Planet Sci Lett 273:323–331
1402
Fallick AE, Melezhik VA, Simonson B (2008) The
ancient anoxic biosphere was not as we know it. In:
Dobretsov N, Kolchanov N, Rozanov A, Zavarzin G (eds)
Biosphere origin and evolution. Springer Science + Busi-
ness Media, New York, pp 169–188
Garvin J, Buick R, Anbar AD, Arnold GL, Kaufman AJ
(2009) Isotopic evidence for an aerobic nitrogen cycle in the
latest Archean. Science 323:1045–1048
George SC, Volk H, Dutkiewicz A, Ridley J, Buick R
(2008) Preservation of hydrocarbons and biomarkers in oil
trapped inside fluid inclusions for>2 billion years. Geochim
Cosmochim Acta 72:844–870
Gerard E, Moreira D, Philippot P, Van Kranendonk MJ,
Lopez-Garcıa P (2009) Modern subsurface bacteria in pris-
tine 2.7 Ga-old fossil stromatolite drillcore samples from the
Fortesque Group, Western Australia. PLoS One 4:1–9
Godfrey LV, Falkowski PG (2010) The cycling and redox
state of nitrogen in the Archean ocean. Nat Geosci 2:725–729
Grassineau NV, Nisbet EG, Bickle MJ, Fowler CMR,
Lowry D, Mattey DP, Abell P, Martin A (2001) Antiquity
of the biological sulphur cycle: evidence from sulphur
and carbon isotopes in 2700 million year old rocks of
the Belingwe belt, Zimbabwe. Proc R Soc Lond B268:
113–119
Hallmann C, Kelly AE, Gupta SN, Summons RE (2011)
Reconstructing deep-time biology with molecular fossils. In:
Laflamme M, Schiffbauer JD, Dornbos SQ (eds)
Quantifying the Quantifying the evolution of early life,
vol 36, Topics in geobiology. Springer, Dordrecht/New
York, pp 355–401
Hayes JM, Kaplan IR, Wedeking W (1983) Precambrian
organic geochemistry, preservation of the record. In: Schopf
JW (ed) Earth’s earliest biosphere, its origin and evolution.
Princeton University Press, Princeton, pp 93–134
Hoefs J (2009) Stable isotope geochemistry. Springer,
Berlin, p 288
Hoering TC (1965) The extractable organic matter in
Precambrian rocks and the problem of contamination.
Carnegie Inst Wash Yearbook 64:215–218
Hoering TC, Navale V (1987) A search for molecular
fossils in the kerogen of Precambrian sedimentary rocks.
Precambrian Res 34:247–267
Holba AG, Dzou LIP, Masterson WD, Hughes WB,
Huizinga BJ, Singletary MS, Moldowan JM, Mello MR,
Tegelaar E (1998) Application of 24-norcholestanes for
constraining source age of petroleum. Org Geochem
29:1269–1283
Illing CJ, Ostertag-Henning C (2010) Deciphering the
transformation of nitrogen in sedimentary rocks during nat-
ural and artificial maturation. Geochim Cosmochim Acta
2010:A444
Jackson MJ, Sweet IP, Powell TG (1988) Studies on
petroleum geology and geochemistry, Middle Proterozoic,
McArthur Basin Northern Australia I: petroleum potential.
Aust Petrol Explor Assoc J 28:283–302
Jia Y (2006) Nitrogen isotope fractionations during pro-
gressive metamorphism: a case study from the Paleozoic
Cooma metasedimentary complex, south eastern Australia.
Geochim Cosmochim Acta 70:5201–5214
Johnston DT, Wing BA, Farquhar J, Kaufman AJ, Strauss
H, Lyons TW, Kah LC, Canfield DE (2005) Active micro-
bial sulphur disproportionation in the Mesoproterozoic. Sci-
ence 310:1477–1479
Karhu JA, Holland HD (1996) Carbon isotopes and the
rise of atmospheric oxygen. Geology 24:867–870
Kitchen NE, Valley JW (1995) Carbon isotope thermom-
etry in marbles of the Adirondack Mountains, New York.
J Metamorph Geol 13:577–594
Knoll AH, Summons RE, Waldbauer JR, Zumberge
J (2007) The geological succession of primary producers in
the oceans. In: Falkowski PG, Knoll AH (eds) The evolution
of primary producers in the sea. Elsevier Academic Press,
Amsterdam/Boston, pp 133–163
Leventhal JS, Suess SE, Cloud PE (1975) Nonprevalence
of biochemical fossils in kerogen from pre-Phanerozoic
sediments. Proc Natl Acad Sci USA 72:4706–4710
Love GD, Snape CE, Carr AD, Houghton RC (1995)
Release of covalently-bound alkane biomarkers in high
yields from kerogen via catalytic hydropyrolysis. Org
Geochem 23:981–986
Love GD, Grosjean E, Stalvies C, Fike DA,
Grotzinger JP, Bradley AS, Kelly AE, Bhatia M, Meredith
W, Snape CE, Bowring SA, Condon DJ, Summons RE
(2009) Fossil steroids record the appearance of Demo-
spongiae during the Cryogenian period. Nature 457:718–721
Lyons TW, Gill BC (2010) Ancient sulfur cycling and
oxygenation of the early biosphere. Elements 6:93–99
McCollom TM, Ritter G, Simoneit BRT (1999) Lipid
synthesis under hydrothermal conditions by Fischer-
Tropsch-Type reactions. Origins Life Evol B 29:153–166
Melezhik VA, Roberts D, Fallick AE, Gorokhov IM,
Kusnetzov AB (2005) Geochemical preservation potential
of high-grade calcite marble versus dolomite marble: impli-
cation for isotope chemostratigraphy. Chem Geol 216:
203–224
Melezhik VA, Huhma H, Condon DJ, Fallick AE,
Whitehouse MJ (2007) Temporal constraints on the
Paleoproterozoic Lomagundi-Jatuli carbon isotopic event.
Geology 35:655–658
Moldowan JM, Talyzina NM (1998) Biogeochemical
evidence for dinoflagellate ancestors in the early Cambrian.
Science 281:1168–1170
Nier AO (1950) A redetermination of the relative
abundances of the isotopes of carbon, nitrogen, oxygen,
argon and potassium. Phys Rev 77:789–793
1403
Nisbet EG, Grassineau NV, Howe CJ, Abell PI, Regelous
M, Nisbet RER (2007) The age of Rubisco: the evolution of
oxygenic photosynthesis. Geobiology 5:311–335
Philippot P, van Zuilen M, Lepot K, Thomazo C,
Farquhar J, van Kranendonk M (2007) Early Archean
microorganisms preferred elemental sulfur not sulfate. Sci-
ence 317:1534–1537
Raab M, Spiro B (1991) Sulfur isotopic variations during
seawater evaporation with fractional crystallization. Chem
Geol 86:323–333
Rashby SE, Sessions AL, Summons RE, Newman DK
(2007) Biosynthesis of 2-methylbacteriohopanepolyols by
an anoxygenic phototroph. Proc Natl Acad Sci 104:
15099–15104
Rasmussen B, Fletcher IR, Brocks JJ, Kilburn MR (2008)
Reassessing the first appearance of eukaryotes and
cyanobacteria. Nature 455:1101–1104
Robert F (1988) Carbon and oxygen isotope variations in
Precambrian cherts. GeochimCosmochimActa 52:1473–1478
Rosman JR, Taylor PD (1998) Isotopic compositions of
the elements (technical report): commission on atomic
weights and isotopic abundances. Pure Appl Chem 70:
217–235
Schidlowski M (2001) Carbon isotopes as biogeochemi-
cal recorders of life over 3.8 Ga of earth history: evolution of
a concept. Precambrian Res 106:117–134
Schidlowski M, Appel PWU, Eichmann R, Junge CE
(1979) Carbon isotope geochemistry of the 3.7 � 10�9 yr-
old Isua sediments, West Greenland: implications for the
Archaean carbon and oxygen cycles. Geochim Cosmochim
Acta 43:189–199
Seifert WK (1978) Steranes and terpanes in kerogen
pyrolysis for correlation of oils and source rocks. Geochim
Cosmochim Acta 42:473–484
Sessions AL, Doughty DM, Welander PV, Summons RE,
Newman DK (2009) The continuing puzzle of the great
oxidation event. Curr Biol 19:R567–R574
Shen Y, Pinti DL, Hashizume K (2006) Biogeochemical
cycles of sulphur and nitrogen in the Archean ocean and
atmosphere. In: Benn K, Mareschal JC, Condie KC (eds)
Archean geodynamics and environments, vol 164, Geophys-
ical monograph series. American Geophysical Union,
Washington, pp 305–320
Shen YN, Farquhar J, Masterson A, Kaufman AJ, Buick
R (2009) Evaluating the role of microbial sulfate reduction
in the early Archean using quadruple isotope systematics.
Earth Planet Sci Lett 279:383–391
Sherman LS, Waldbauer JR, Summons RE (2007)
Improved methods for isolating and validating indigenous
biomarkers in Precambrian rocks. Organ Geochem
38:1987–2000
Sherwood Lollar B, Westgate TD, Ward JA, Slater GF,
Lacrampe-Couloume G (2002) Abiogenic formation of
alkanes in the Earth’s crust as a minor source for global
hydrocarbon reservoirs. Nature 416:522–524
Sigman DM, Casciotti KL (2001) Nitrogen isotopes in
the ocean. In: Steele JH (ed) Encyclopedia of ocean
sciences. Academic, San Diego, pp 1884–1894
Strauss H (2002) The sulfur isotopic composition of
Precambrian sedimentary sulfides: seawater chemistry and
biological evolution. In: Altermann W, Corcoran P (eds)
Precambrian sedimentary environments: a modern approach
to ancient depositional systems, vol 33, Special publications
of the International Association of Sedimentologists.
Blackwell Science, Oxford/Malden, pp 67–105
Strauss H (2004) 4 Ga of seawater evolution: evidence
from the sulfur isotopic composition of sulfate. In: Amend
JP, Edwards KJ, Lyons TW (eds) Sulfur biogeochemistry:
past and present, vol 379, Geological society of America
special paper. Geological Society of America, Boulder,
pp 195–205
Strauss H, Des Marais DJ, Summons RE, Hayes JM
(1992) The carbon-isotopic record. In: Schopf JW, Klein C
(eds) The Proterozoic biosphere: a multidisciplinary study.
Cambridge University Press, Cambridge, pp 117–127
Summons RE, Jahnke LL, Hope JM, Logan GA (1999)
2-Methylhopanoids as biomarkers for cyanobacterial oxy-
genic photosynthesis. Nature 400:554–557
Summons RE, Bradley AS, Jahnke LL, Waldbauer JR
(2006) Steroids, triterpenoids and molecular oxygen. Philos
Trans R Soc B361:951–968
Page RW, Sweet IP (1998) Geochronology of basin
phases in the western Mt Isa Inlier, and correlation with
the McArthur Basin. Aust J Earth Sci 45:219–232
Thomazo C, Pinti DL, Busigny V, Ader M, Hashizume K,
Philippot P (2009) Biological activity and the Earth’s sur-
face evolution: insights from carbon, sulfur, nitrogen and
iron stable isotopes in the rock record. C R Palevol
8:665–678
Thomazo C, Ader M, Philippot P (2011) Extreme15N-enrichments in 2.72-Gyr-old sediments: evidence for
a turning point in the nitrogen cycle. Geobiology
9:107–120
Trudinger CM, Enting IG, Francey RJ, Etheridge DM,
Rayner PJ (1999) Long-term variability in the global carbon
cycle inferred from a high-precision CO2 and d13C ice-core
record. Tellus 51B:233–248
Vandenbroucke M, Largeau C (2007) Kerogen origin,
evolution and structure. Org Geochem 38:719–833
1404
Watanabe Y, Naraoka H, Wronkiewicz DJ, Condie KC,
Ohmoto H (1997) Carbon, nitrogen, and sulfur geochemis-
try of Archean and Proterozoic shales from the Kaapvaal
Craton, South Africa. Geochim Cosmochim Acta
61:3441–3459
Waldbauer JR, Sherman LS, Sumner DY, Summons RE
(2009) Late Archean molecular fossils from the Transvaal
Supergroup record the antiquity of microbial diversity and
aerobiosis. Precambrian Res 169:28–47
Waldbauer JR, Newman DK, Summons RE (2011)
Microaerobic steroid biosynthesis and the molecular fossil
record of Archean life. Proc Natl Acad Sci USA
108:13409–13414
Welander PV, Coleman M, Sessions AL, Summons RE,
Newman DK (2010) Identification of a methylase required
for 2-methylhopanoid production and implications for the
interpretation of sedimentary hopanes. Proc Natl Acad Sci
USA 107:8537–8542
Williams LB, Ferell RE, Chinn EW, Sassen R (1989)
Fixed-ammonium in clays associated with crude oils. Appl
Geochem 4:605–616
Zerkle AL, House CH, Brantley SL (2005) Bio-
geochemical signatures through time as inferred from
whole microbial genomes. Am J Sci 305:467–502
1405
7.9 Terrestrial Environments
7.9.1 Introductory Remarks
Lee R. Kump (Editor)
In contrast to a rather extensive marine sedimentary record
of the Palaeoproterozoic, the terrestrial record, preserved as
palaeosols (ancient soils; Retallack 2001) and caliches or
calcretes (carbonate layers in palaeosols; Wright and
Tucker 1991), is sparse. Nevertheless, these deposits are
important, because they have the potential to provide the
least ambiguous information on climatic and atmospheric
compositional change during this critical interval of Earth
history. In the best of circumstances, marine sediments
record oceanic conditions, which can differ markedly
from atmospheric conditions, especially in terms of redox.
In the modern world, soil environments themselves poorly
reflect those at the surface, especially in terms of CO2 and
O2 partial pressure, because of respiration by roots and soil
microbes. But in the Palaeoproterozoic, with its presum-
ably poorly developed terrestrial biota, soil conditions
likely track those of the atmosphere much more closely.
(We use “presumably” here, because terrestrial ecosystems
may have been extensive (Horodyski and Knauth 1994;
Watanabe et al. 2000), and the land surface may have
been the incubator for early evolutionary innovation,
including the origin of cyanobacteria; e.g. Battistuzzi
et al. (2004).
The chapters that follow review what is known about
Palaeoproterozoic palaeosols and caliches and what those
deposits can reveal about the surface environment at the
time of their formation. There are few examples of caliche
in the Palaeoproterozoic rock record, but when preserved
indicate an arid climate with perhaps a source of windborn
dust (providing the additional calcium) and long-term
stability of the soil surface. The palaeosol record is more
mature, thanks in large part to the pioneering and peren-
nial work of Heinrich D. Holland and his colleagues who
viewed palaeosols as the clearest window into the stages
of oxygenation of the surface environment during the
Great Oxidation Event (e.g. Holland and Rye 1997). The
work has been challenging, in part because the complete
weathering profile that existed when these rocks were soil
has rarely, if ever been preserved. Most profiles have
suffered erosion of their uppermost surface, the part that
was in direct contact with the atmosphere and whatever
biota existed at the time. What has been preserved is likely
the lower part of the regolith (the mantle of altered mate-
rial overlaying bedrock), i.e., the saprolite that exhibits the
base of chemical alteration and vestiges of original mineral-
ogy in their textures. Moreover, palaeosols and palaeosa-
prolites often had enhanced permeability, and thus have
been subject to enhanced postdepositional fluid flow and
hydrothermal alteration. Only with considerable care and
ingenious interpretative approaches can the original envi-
ronmental conditions be deduced from the palaesol and
caliche record. The FAR-DEEP core archive provides an
ideal opportunity to apply those techniques in an effort to
elucidate environmental change during the Great Oxidation
Event.
References
Battistuzzi FU, Feijao A, Hedges BA (2004) A genomic
timescale of prokaryote evolution: insights into the origin
of methanogenesis, phototrophy, and the colonization of
land. BMC Evol Biol 4:44
Holland HD, Rye R (1997) Evidence in pre-2.2 Ga
paleosols for the early evolution of atmospheric oxygen
and terrestrial biota: comment and reply. Geology
25:857–858
Horodyski RJ, Knauth LP (1994) Life on land in the
Precambrian. Science 263:494–498
L.R. Kump (Editor)
Department of Geosciences, Pennsylvanian State University,
503 Deike Building, University Park, PA 16870, USA
e-mail: [email protected]
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_9, # Springer-Verlag Berlin Heidelberg 2013
1407
Retallack GJ (2001) Soils of the past: an introduction to
paleopedology. Blackwell Science, Oxford, p 404
Watanabe Y, Martini JEJ, Ohmoto H (2000) Geochemi-
cal evidence for terrestrial ecosystems 2.6 billion years ago.
Nature 408:574–578
Wright VP, Tucker ME (1991) Calcretes: an introduction.
In: Wright VP, Tucker ME (eds) Calcretes. Blackwell Sci-
entific Publications, Oxford, pp 1–22
1408 L.R. Kump
7.9.2 Palaeoproterozoic Weathered Surfaces
Kalle Kirsim€ae and Victor A. Melezhik
Introduction
Precambrian weathering crusts and palaeosol profiles pro-
vide important and direct evidence for early weathering
conditions reflecting the past climate (temperature, precipi-
tation), atmospheric composition (pCO2, pO2) and (micro-)
biota (e.g. Retallack 2001). Palaeoweathering indicators
provide, in the absence of biological effects as is presumably
the case in Archaean-Proterozoic times, a semi-quantitative
reconstruction of atmospheric oxygen levels (Rye and Hol-
land 1998; Holland 2006). While photosynthesis is the pri-
mary source of O2, there are multiple environmental sinks,
most significantly reduced carbon, iron and sulphur.
Today, at current fluxes of reduced Fe and S from
volcanoes and mid-ocean ridges, the accumulation of free
oxygen is controlled by organic matter burial in sediments,
which is largely controlled by the sheltering/preservational
effects of detrital clay minerals in marine continental margin
depocentres (Kennedy et al. 2006). Kennedy et al. (2006)
show mineralogical and geochemical evidence for an
increase in clay mineral deposition in the late Neoproterozoic
that immediately predated the first metazoans at the last step
of atmospheric oxidation (Catling and Claire 2005). This was
interpreted as a result of initial expansion of a primitive land
biota and enhanced production of pedogenic clay minerals –
the “Inception of the Clay Mineral Factory” – leading to
increasedmarine burial of organic carbon via mineral surface
preservation. Indeed, terrestrial vegetation promotes deeper
weathering by creating corrosive soil-rock waters, by physi-
cally widening rock fractures with root propagation, and by
binding soils and rock particles for more prolonged chemical
weathering and soil development. As a consequence, the
appearance of land plants in the early to middle Palaeozoic
likely accelerated both chemical and physical weathering
(e.g. Berner 1992) relative to the Neoproterozoic.
Recently, however, Tosca et al. (2010) pointed to evi-
dence for intense leaching at moderate to intense weathering
conditions (chemical index of alteration, CIA ¼ 70 to 90
plus) in late Archaean to Mesoproterozoic palaeosols, and
perhaps for formation of smectite-kaolinite type mineral
phases. Moreover, Hassler and Lowe (2006) suggested an
aggressive weathering environment and almost entire
decomposition of labile materials (komatiite, basalt, and
coarse plagioclase grains) into clays in 3200 Ma Moodies
Group in the Barberton Greenstone Belt. Increased rainfall,
higher temperatures, and/or higher atmospheric pCO2
worked to offset the less effective weathering effects of a
plant-free environment, creating an aggressive weathering
environment in the Archaean-Proterozoic times. However,
the same degree of weathering could have been achieved
with longer exposure times and moderate rainfall amounts or
temperatures.
Weathering crusts and palaeosols formed in critical time
intervals are, therefore, of paramount importance for (a)
timing, (b) quantification and (c) understanding of GOE
processes-mechanisms as well as (d) palaeoclimatological-
palaeogeographical reconstruction of Archaean-Proterozoic
environments.
Reliable analysis of past environments is significantly
hampered by the limited number ofwell-preserved palaeosols,
leading to potentially biased and radically different
interpretations (see Rye and Holland 1998 for critical review).
Rye and Holland (1998) list only 14 Archaean-Proterozoic
sections world-wide that meet textural (incl. soft sediment
features), mineralogical and geochemical criteria for a
palaeosol, and some 13 likely palaeosols. The majority of
these definite palaeosol occurrences are limited to the
Transvaal basin in South Africa, the Huronian basin in
Canada, the Fennoscandian Shield, and Western Australia. A
number of these sections cover the critical period of timewhen
the GOE took place. Mt Roe palaeosols in Western Australia
are probably the oldest definite palaeosols and are indicative
of reducing atmospheric environment (Rye and Holland
1998).Most of the palaeosols in the Elliot Lake area, Huronian
basin, Canada are also believed to represent oxygen-free
atmospheric conditions, except the Ville Marie palaeosol,
which together with the Hokkalampi palaeosol in
Fennoscandia, are considered the oldest oxygenic weathering
profiles. However, the Hekpoort palaeosol in South Africa,
whose age overlaps with the Hokkalampi and Ville Marie
sections, has long been interpreted as the earliest (e.g. Rye
and Holland 1998) low-oxygen weathering profile. This inter-
pretation, however, has been challenged in recent years (see
discussion bellow).
The Fennoscandian (Baltic) Shield contains several
preserved weathering profiles (regoliths and saprolites).
These are formed on Late-Archaean granitoid basement
(including erosional greenstone belts) and early Proterozoic
metasedimentary-metavolcanic complexes of the Karelian
formations, 3100–2600 Ma and ~2450 to ~1900 Ma in age,
respectively (Ojakangas et al. 2001), and indicate variable
weathering intensity (e.g. Metzger 1924; Koryiakin 1971,
1975; Golovenok 1975; Negrutsa 1971, 1979, 1984; Sokolov
and Heiskanen 1984; Sokolov 1987; Bobrov and Shipakina
K. Kirsim€ae (*)
Department of Geology, Tartu University, Ravila 14A, 50411 Tartu,
Estonia
9 7.9 Terrestrial Environments 1409
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_9, # Springer-Verlag Berlin Heidelberg 2013
1409
1991; Marmo 1992; Melezhik and Sturt 1994; Sturt et al.
1994; Matrenichev et al. 2005; Alfimova and Matrenichev
2006). The most complete (and mature) palaeo-weathering
profiles in this area are located at the base of Sumian
(2504–3430 Ma), Sariolian (2430–2200 Ma) and Jatulian
(2200–2060 Ma) successions (Fig. 7.138).
Pre-Sariolian Palaeoweathering Crusts
The pre-Sariolian weathering crusts in Fennoscandia formed
in contact with Archaean basement, or in places, on Sumian
rocks. These widespread regolith-type weathering crusts can
be found in the Pechenga and Imandra/Varzuga Greenstone
belts, most complete sections in the Pasvik-Pechenga area
northern Norway and Kola Peninsula (Sturt et al. 1994), and
in eastern Finland and Karelia including the Onega Basin
(Laajoki 2005; Negrutsa 1979, 1984). The age of regolith
formation is estimated between 2453 and 2330 Ma in
Pechenga, which agrees with the 2493–2423 Ma range in
Imandra-Varzuga and 2440–2200 Ma in eastern Finland
(Sturt et al. 1994). This indicates a major unconformity
and regolith formation on the Fennoscandian Shield at the
base of Sariolian succession. The pre-Sariolian regolith is
typically composed of a-few-meter-thick fractured granitoid
residual breccia grading into Sariolian basal conglomerates
(Fig. 7.139). The chemical alteration of the parent rocks is
low (rarely moderate) with subdued sericitisation, carbon-
ation and chloritisation. The absence of deeply weathered
material has been interpreted as a primary feature; the pro-
duction of the weathering crust is ascribed to physical in situ
weathering (Laajoki et al. 1989) under an arid to semi-arid
climate with little chemical weathering of silicates and
sulphides (Sturt et al. 1994). Alternatively, Kohonen and
Marmo (1992) suggest for the <30-m-thick Ilvesvaara rego-
lith (eastern Finland) either mechanical disintegration in
fracture zones or disintegration (“ice-shattering”) under gla-
cial conditions (stresses). This would explain the depressed
chemical weathering of Pre-Sarioli crusts that were formed
prior to the onset of Huronian Glaciation that is represented
in the Sariolian sequences of diamictite, varves and shales
with dropstone clasts (for details see Chapter 7.2). On the
other hand, Pekkarinen (1979) infers incomplete preserva-
tion of pre-Sariolian weathering profiles to be due to erosion
during early stages of Sariolian sedimentation.
More importantly, the fluviatile polymict conglomerates
succeeding the regolith complex in Pasvik preserve pebbles
composed of magnetite (banded iron quartzites), pyrite ore
and magnetite-pyrite-rich skarn rocks indicating oxygen
deficiency in an arid to semi-arid climate during formation
of this regolith (Sturt et al. 1994). This conclusion is further
strengthened by relatively high abundance of carbonate,
particularly in the uppermost zone of the weathering crust
and in conglomerate units (Koryiakin 1971; Fedotov et al.
1975; Pekkarinen 1979; Sokolov and Heiskanen 1984)
suggesting caliche type soil (Aridisol) evolution in arid or
semi-arid climate, where evaporation causes alkali supersat-
uration in soil solution (Fig. 7.140). Usually calcium carbon-
ate precipitates, forming more or less temporary crusts at
the upper limit of the capillary (groundwater) rise (Meunier
2005). Carbonate accumulation implies, at least locally,
pH levels greater than 9 and possibly precipitation of
palygorskite (sepiolite) clay phases (Righi and Meunier
1995) that were converted into talc and chlorite in sub-
sequent metamorphic processes.
Conglomerate units topping the Fennoscandian pre-
Sariolian regolith resemble the uranium-bearing quartz peb-
ble conglomerates at the base of Huronian succession in the
Huronian Basin, Ontario, Canada where uraniferous and
pyritic placer conglomeratic deposits occur at the top of
the Livingstone Creek Formation and most abundantly, in
the Matienda Formation (e.g. Young et al. 2001). Chemi-
cally reduced detrital minerals such as pyrite and uraninite in
these mature sandstones and quartz conglomerates indicate,
similarly to the lowermost Sariolian conglomerates, low
atmospheric pO2.
Uraniferous conglomerates containing rip-up clasts of
weathered material from the Huronian Supergroup can be
associated with several occurrences of weathering crusts
underlying the Matienda Formation in Elliot Lake area as
at Denison/Stanleigh, Quirke II and Pronto saproliths/
palaeosols, probably formed between 2457 and 2440 Ma
(Rye and Holland 1998). In contrast to the Fennoscandian
pre-Sariolian regoliths, the Elliot Lake weathered profiles
are characterised by deep chemical weathering indicated by
high concentrations of Al2O3 in the upper parts of the
profiles, suggesting that primary phases had been weathered
into high-Al clay mineral phases such as kaolinite, smectite,
and/or mixed layer smectitic minerals (Gay and Grandstaff
1980; Rainbird et al. 1990; Sutton and Maynard 1993;
Young et al. 2001). Up-profile depletion of Fetotal and Fe3+
in the whole rock and in chlorite/mica phases, up-profile
depletion of Fe/Mg ratios in chlorites and fine-grained mus-
covite, and the presence of ferrous minerals (e.g. pyrite,
pyrrhotite, ilmenite) in these sediment profiles suggest
reducing atmospheric conditions and strong acidic leaching
due to elevated pCO2 (Prasad and Roscoe 1996; Rye and
Holland 1998; Sheldon 2006).
Pre-Jatulian Palaeoweathering Crusts
The base of Jatulian sedimentary (arenitic silicilastic, stro-
matolitic carbonate, redbed) and alkaline to tholeitic volca-
nic sequences in the Fennoscandian Shield is characterised
by palaeosols on Archaean, Sumian and Sariolian rocks
1410 K. Kirsim€ae and V.A. Melezhik
underlying the Jatulian-age rocks in a large number of
localities (Fig. 7.138; Kohonen and Marmo 1992; Negrutza
1984; Sokolov 1987; Marmo 1992; Laajoki 2005). The well-
developed nature of these palaeosols indicates intense chem-
ical weathering under a warm and humid climate (Marmo
1992). Though the correlation between different sections is
not clear (e.g. Kohonen and Marmo 1992), the lateral exten-
sion of this presumably contemporaneous unconformity
covers the eastern-northern part of the Fennoscandian
Shield.
The thickest and most widespread palaeosol on the pre-
Jatulia unconformity is the greenschist-facies Hokkalampi
palaeosol (Fig. 7.141) developed on late Archaean granites
and Sariolian glacigenic deposits in the northern North
Karelia Schist Belt (Marmo 1992). The palaeo-weathered
section consists of quartz-sericite-kyanite schists with a
maximum preserved thickness of up to 80 m, whereas
the proportions of kyanite, andalusite, and occasionally
chloritoid increase toward the top (Kohonen and Marmo
1992; Marmo 1992). The palaeosol is subdivided into three
gradually changing zones with upward increasing alteration.
The basal zone is characterised by vertical alteration from
unaltered rock to quartz-sericite schist, reflecting disintegra-
tion/replacement of feldspars and micas with sericite, car-
bonate, epidote and chlorite and a low chemical index of
alteration (CIA; Nesbitt and Young 1982) varying between
60 and 65. The intermediate zone consists entirely of quartz
and sericite while feldspar is absent. The rock lacks textures
inherited from the protoliths and the CIA values reach
70–80. In the uppermost zone sericite is progressively
replaced by kyanite and/or andalusite (most probably kao-
linite in original material), which may represent up to 25 %
of the rock; CIA values are 90–96 (Kohonen and Marmo
1992; Marmo 1992).
The Hokkalampi palaeosol and associated arenitic
metasedimentsis are interpreted as a result of strong chemi-
cal weathering. Palaeomagnetic data imply that during
2400–2300 Ma, Baltica (the Fennoscandian Shield) was
positioned at low palaeolatitudes of ~30� (Mertanen and
Pesonen 2005). Depletion of the upper palaeosol zones in
ferrous iron, sodium, calcium and magnesium, and also
potassium and ferric iron in the uppermost portions of the
crust (Fig. 7.141) indicates intense kaolinitisation of the
parent rock above the groundwater table, forming a Vertisol
type profile. Rye and Holland (1998), however, suggest that
the uppermost, presumably most weathered and perhaps
enriched in ferric iron, zone of the palaeosol at Hokkalampi
is missing and was eroded prior to the deposition of Jatulian
sediments. In contrast to the thick and nearly complete
Hokkalampi palaeosol section in the northern North Karelia
Schist Belt, the correlative palaeoweathered surfaces, as
well as weathered unconformities within the Jatulian
successions in the rest of the Shield, are relatively thin
(up to few m) alteration zones where typically the upper-
most kyanite-andalusite (kaolinite) zone is missing (e.g.
Sokolov and Heiskanen 1984). The age of the weathering
has been estimated at 2440–2200 Ma (2350 � 190 Ma
according to Sm-Nd isotope data, Stafford 2005), which
places the section within, or to just after, the GOE (Bekker
et al. 2004).
The environmental importance of the pre-Jatulian
Hokkalampi palaeosol, especially any inference from it
concerning atmospheric oxygen levels, rests on interpreta-
tion of the completeness of the palaeosol. Ohmoto (1996)
classifies the Hokkalampi palaeosol as a typical mixed
weathering profile (type–M sensu Ohmoto 1996), which
shows increased Fe3+/Ti ratio, an Fetotal-depleted sericite
(kaolinite) upper part, and Fetotal-enriched (chlorite rich)
composition in the middle part of the profile; all these are
characteristics of oxidised palaeosols. Strafford (2005)
shows that the upper portion of the preserved palaeosol has
lost more than 40 % of the iron relative to the parent rock,
while retaining ferric-ferrous iron ratios greater than one.
Based on that, Rye and Holland (1998) calculated an upper
limit on pO2 of about only 2·10�4 atm presuming that the
missing material at the top was not ferric-iron enriched. This
Fe loss from the upper portion of the Hokkalampi palaeosol
could be explained, in addition to an erosion model, by a
two-step leaching model, which includes dissolution of fer-
ric oxyhydroxides accumulated in the uppermost zone of the
profile under either H2-rich hydrothermal fluids or by
organic acids (Ohmoto 1996). Indeed, Driese (2004) shows
from comparison with modern and Palaeozoic oxygenic
Vertisols that significant translocation of Fe in the upper
parts of the early Proterozoic palaeosols could be accounted
for by the presence of a terrestrial biomass serving as a
source of organic ligands and not necessarily due to reducing
conditions.
The Hokkalampi palaeosol is considered to be one of the
oldest, though incomplete, weathering profiles indicative of
an oxygenic atmosphere. It is roughly coeval with the Ville
Marie profile of the Huronian Supergroup, Canada (Rainbird
et al. 1990), developed on Archaean granitic saprolite. The
Ville Marie palaeosol shows abundant sericite (i.e.,
kaolinitisation) as well as enrichment of Fe3+ upward in
profile that can be attributed to weathering under an
oxidising environment. However, the age of the Ville
Marie section is poorly constrained, and in particular, the
lower boundary of its age interval is somewhat unclear
(Holland and Rye 1997; Rye and Holland 1998).
If the Hokkalampi palaeosol developed just after the
GOE, then the Hekpoort palaeoweathering horizon of the
9 7.9 Terrestrial Environments 1411
Fig. 7.138 Location of main known pre–Sariolian (red circles) and pre–Jatulian (blue circles) weathering crust occurrences in the FennoscandianShield (The geological map is modified by Aivo Lepland from Koistinen et al. (2001))
1412 K. Kirsim€ae and V.A. Melezhik
Transvaal Supergroup in southern Africa, formed between
2240 and 2060 Ma (Yang and Holland 2003) on Hekpoort
basalt lavas, has been considered to be one of the youngest
Fe-depleted palaeosols (e.g. Rye and Holland 2000)
indicating a pO2 level of �8·10�4 atm. Interpretation of
low atmospheric oxygen levels during formation of the
Hekpoort palaeoweathering crust relies on gradual upward
increasing loss of Fe from the parent Hekpoort Basalt to
the sericite (pallid) zone. However, Beukes et al. (2002)
discovered a nearly complete lateritic palaeosol profile
near Gaborone, Botswana and Potchefstroom, western
Transvaal with preserved upper, haematite-rich laterite and
mottled zones. Yang and Holland (2003) revised earlier
interpretation by a mass balance calculation of Strata 1
profile of this laterite that suggests a value of pO2 between
2.5·10�4 and 9·10�3 atm. Nevertheless, Yamaguchi et al.
(2007) show, based on isotopic composition and concentra-
tion of Fe in the Hekpoort palaeosol, that it exhibited open-
system (governed by groundwater transport) behaviour of Fe
during weathering. As the result, they suggest that the
estimates by Yang and Holland (2003) are too low by at
least an order of magnitude.
A similar model (i.e., erosion of upper lateritic (goethite/
haematite) zones and significant Fe-mobilisation due to
interplay of groundwater transport and leaching by organic
acids) can be invoked for the Hokkalampi palaeoweathering
profile, which means that the Hokkalampi and Hekpoort,
as well as the Ville Marie and probably Drakenstein
(South Africa) palaeosols belong to the same group of oxic
weathering crusts that were formed after the GOE. It seems
that the Canadian Elliot Lake palaeoweathering sections in
Denison/Stanleigh, Quirke II and Pronto and the
Fennoscandian pre-Sariolian crusts, which agree in age
with the youngest rocks with large mass-independent sul-
phur isotope fractionation ~2450 Ma (Bekker et al. 2004),
are also probably the youngest pre-oxygenation palaeosols.
However, Ohmoto (1996) and Nedachi et al. (2005) have
disputed the interpretation of Fe distribution in Elliot Lake
palaeosol profiles and, thereby, suggest that the Denison and
Pronto palaeosols were formed under an oxic atmosphere. If
these palaeosols indeed require an oxygenated environment,
then it remains open whether any of these weathering
profiles indicate that the oxygen level rose from the levels
consistent with mass-independent fractionation of sulphur
Fig. 7.139 (a) Section of the pre-Sariolan regolith in Pasvik, NE
Norway at the contact of Archaean gneiss, pegmatite complex and
early Proterozoic fluvial sediments (Redrawn from Sturt et al. 1994).
A – fluvial deposits, B – disturbed regolith, C – regolith, D – jointed
pegmatite, E – jointed gneiss, F – gneiss. (b) Jointed pegmatite from
zone C–D contact (Photograph (b) by Victor Melezhik)
9 7.9 Terrestrial Environments 1413
isotopes (<0.001 % of present atmospheric level – PAL) to
~1 % of PAL required for iron immobilisation in weathering
profiles (e.g. Kump 2008).
Summary and Implications for FAR-DEEP Cores
Palaeosols, unlike marine climatic proxy records, are formed
in direct contact with the atmosphere, and thus are subject to
the compositional (e.g. O2 and CO2 concentrations) and
climate effects (temperature, rainfall) that prevailed at the
time of their formation (Sheldon and Tabor 2009). The
interpretational power of palaeosols, especially with respect
to the evolution of early Earth, relies on the quantitative
interpretation of different whole-rock and isotopic signatures
as proxies for reconstruction of palaeoenvironmental and
palaeoclimatic conditions. Palaeosols formed at the
Archaean–Proterozoic transition are of special interest for
the evolution of atmospheric oxygen, a phenomenon that
can be constrained using distribution and mobility of Fe (or
any other elements sensitive to redox states) in palaeosol
profiles. Critical re-evaluation of key palaeosol profiles
around the time of GOE (e.g. Rye and Holland 1998, Beukes
et al. 2002; Yang and Holland 2003; Nedachi et al. 2005;
Fig. 7.140 Chemical composition of the pre–Sariolian weathering
crust (S€arkilamppi, Kiihtelysvaara, Finland), after Pekkarinen (1979).
Note the increase in CaO content in upper portion of the crust, which is
due to presence of carbonate mineral phases. A – breccia conglomerate;
B – graded unconformity between conglomerate and broken basement;
C – heavily disintegrated Archaean basement rock, plagioclase and
biotite replaced with sericite, chlorite and carbonate; D – fractured
and disintegrated rock, sericite/chlorite in margins and cleavages of
grains, minor carbonate between grains; E – fractured rock with only
slight alteration
1414 K. Kirsim€ae and V.A. Melezhik
Yamaguchi et al. 2007) have significantly advanced the
understanding of the oxygenation event. Though the
palaeoweathering systems are not as sensitive indicators of
oxygen rise as mass-independent fractionation of sulphur
isotopes, they still provide a basis for quantification of tem-
perature, precipitation and more importantly, the pCO2 and
pO2 using different equilibrium/mass-balance models (e.g.
Sheldon 2006). Future developments in palaeosol studies are
focused (1) on applications of non-traditional isotope systems
using transition metals (e.g. Fe, Mo, V) to study the
palaeoredox conditions (Severmann and Anbar 2009); (2)
finding a reliable proxy for palaeo-pH in palaeosol profiles
(e.g. B-isotopes; Sheldon and Tabor 2009); (3) understanding
the biological (microbial) influences on Precambrian
weathering (e.g. Amundson et al. 2007) and (4) developing
new weathering indices and mass- and energy-balance
models for quantification of atmospheric composition and
climate (Sheldon and Tabor 2009; Nordt and Driese 2010).
As for the FAR-DEEP cores, the palaeoweathering stud-
ies could potentially be concentrated on several profiles
intersecting intervals that contain pre-Sariolian brecciated-
conglomeratic units (Holes 1A and 3A), and probably
pre-Jatulian weathering crusts (Hole 5A). A probable pre-
Ludicovian, haematite-rich weathering crust is present in the
Pechenga Greenstone Belt on top of the Kuetsj€arvi VolcanicFormation (Fig. 7.142a). This weathered surface was
intersected by Holes 7A and 8B at various depths with
respect to the present-day erosional surface (Fig. 7.142b,
c), allowing to address geochemical processes involved in
alteration of Kuetsj€arvi lavas in the aftermath of the
Lomagudi-Jatuli positive isotopic excursion of carbonate
carbon and prior to the onset of the Shunga Event (enhanced
accumulation of organic-rich rocks worldwide). In addition,
all FAR-DEEP drillholes intersected detrital shaley sedi-
mentary rocks from various time-intervals spanning
2500–2000 Ma. These can be potentially be used for miner-
alogical and geochemical investigations of the weathering
intensity and redox conditions in the sediment source areas.
Fig. 7.141 Hokkalampi palaeosol, Finland, after Kohonen and
Marmo (1992). (a) Composite diagram (vertical scale is arbitrary)
showing variation of main oxides and chemical index of alteration
(Nesbitt and Young 1982) A – quartz kyanite schist, zone I; B –
quartz–sericite schist, zone II; C – slightly altered granodiorite, zone
III; D – parent rock, Archaean granodiorite, Nuutilanvaara. (b) Quartz-
kyanite schist from zone I; photograph courtesy of Eero Hanski
9 7.9 Terrestrial Environments 1415
References
Alfimova NA, Matrenichev VA (2006) Continental
weathering in the Early Precambrian: specific features of
mineral transformations and composition of supergene
solutions. Lithol Mineral Resour 41:518–529 (in Russian)
Amundson R, Richter DD, Humphreys GS, Jobbagy EG,
Gaillardet J (2007) Coupling between biota and Earth
materials in the critical zone. Elements 3:327–332
Bekker A, Holland HD, Wang P-L, Rumble D III, Stein
HJ, Hannah JL, Coetzee LL, Beukes NJ (2004) Dating the
rise of atmospheric oxygen. Nature 427:117–120
Berner RA (1992) Weathering, plants, and the long–term
carbon cycle. Geochim Cosmochim Acta 56:3225–3231
Beukes NJ, Dorland H, Gutzmer J, Nedachi M, Ohmoto
H (2002) Tropical laterites, life on land, and the history of
atmospheric oxygen in the Paleoproterozoic. Geology
30:491–494
Bobrov ET, Shipakina IG (1991) Reconstruction of the
composition of metamorphosed weathering crusts. Nauka,
Moscow, p 164 (in Russian)
Catling DC, Claire M (2005) How Earth’s atmosphere
evolved to an oxic state: a status report. Earth Planet Sci Lett
237:1–20
Driese SD (2004) Pedogenic translocation of Fe in
modern and ancient Vertisols and implications for inter-
pretations of the Hekpoort paleosol (2.25 Ga). J Geol 112:
543–560
Fig. 7.142 Probable haematite-rich weathering crust formed on the c.
the 2060 Ma Kuetsj€arvi Volcanic Formation of the Pechenga Green-
stone Belt. (a) Polished slab parallel to the upper surface of the
Kuetsj€arvi Volcanic Formation beneath fluvial sedimentary rocks; the
image exhibits fragments of dacitic lava embedded into black, haema-
tite-magnetite-sericite matrix. (b) FAR-DEEP Core 8B exhibiting pink
fragment of alkaline dacite in haematite-rich lava with fluidal structure;
situated c. 0.5 m below the upper surface of the Kuetsj€arvi VolcanicFormation. (c) FAR-DEEP Core 8B showing dacitic lava breccia c.
12 m below the upper surface of the Kuetsj€arvi Volcanic Formation.
Core diameter in (b) and (c) is 5 cm (Photographs by Victor Melezhik)
1416 K. Kirsim€ae and V.A. Melezhik
Fedotov ZA, Melezhik VA, Predovsky AA (1975) Two
levels of the metamorphosed Proterozoic weathering crust in
the Montchegorsk region. In: Precambrian weathering
crusts. Nauka, Moscow, pp 36–99 (in Russian)
Gay AL, Grandstaff DE (1980) Chemistry and mineral-
ogy of Precambrian paleosols at Elliot Lake, Ontario,
Canada. Precambrian Res 12:349–373
Golovenok VK (1975) Precambrian weathering crusts,
their specific features, and lithogeochemical analysis
techniques. In: Precambrian weathering crusts. Nauka,
Moscow, pp 16–27 (in Russian)
Hassler AM, Lowe DR (2006) Weathering and sediment
generation in the Archean: an integrated study of the evolu-
tion of siliciclastic sedimentary rocks of the 3.2 Ga Moodies
Group, Barberton greenstone belt, South Africa. Precam-
brian Res 151:185–210
Holland HD (2006) The oxygenation of the atmosphere
and oceans. Philos Trans R Soc Biol Sci 361:903–915
Holland HD, Rye R (1997) Evidence in pre-2.2 Ga
paleosols for the early evolution of atmospheric oxygen
and terrestrial biota: comment and reply. Geology
25:857–858
Kennedy M, Droser M, Mayer LM, Peaver D, Mrofka D
(2006) Late Precambrian oxygenation; inception of the clay
mineral factory. Science 311:1446–1449
Kohonen J, Marmo J (1992) Proterozoic lithostratigraphy
and sedimentation of Sariola and Jatuli-type rocks in
the Nunnanlahti-Koli-Kaltimo area, eastern Finland;
implications for regional basin evolution models. Geol
Surv Finl Bull 364:67
Koistinen T, Stephens MB, Bogatchev V, Nordgulen Ø,
Wennestr€om M, Korhonen J (Comps.) (2001) Geological
map of the Fennoscandian Shield, Scale 1:2 000 000.
Espoo/Trondheim/Upsala/Moscow
Koryiakin AS (1971) Result of study of Proterozoic
weathering crusts in Karelia. Int Geol Rev 13:973–980
Koryiakin AS (1975) Diagnostic criteria of the
metamorphosed weathering crust. In: Precambrian
weathering crusts. Nauka, Moscow, pp 28–34
Kump LR (2008) The rise of atmospheric oxygen. Nature
451:277–278
Laajoki K, Strand K, H€arm€a P (1989) Lithostratigraphy
of the early Proterozoic Kainuu Schist belt in the Kurkikyl€a-
Siikavaara area, northern Finland, with emphasis on the
genetic approach. Geol Soc Finl Bull 61:65–93
Laajoki K (2005) Karelian supracrustal rocks. In:
Lehtinen M, Nurmi PA, R€am€o OT (eds) Precambrian geol-
ogy of Finland – key to the evolution of the Fennoscandian
Shield, vol 14, Developments in Precambrian Geology.
Elsevier B.V., Amsterdam, pp 279–342
Marmo JS (1992) The lower Proterozoic Hokkalampi
paleosol in North Karelia, eastern Finland. In: Schidlowski
M, Golubic S, Kimberley MM, Trudinger PA (eds) Early
organic evolution: implications for energy and mineral
resources. Springer–Verlag, Berlin/Heidelberg, pp 41–66
Matrenichev VA, Pinkova LO, Levchenkov OA, Makeev
AF, Yakovleva SZ, Alfimova NA (2005) The oldest conti-
nental weathering profile formed after granites on the Baltic
Shield. Dokl Earth Sci 400:62–66 (in Russian)
Melezhik VA, Sturt BA (1994) General geology and
evolutionary history of the early Proterozoic Polmak–Pas-
vik–Pechenga–Imandra/Varzuga–Ust’Ponoy greenstone
belt in the northeastern Baltic Shield. Earth Sci Rev
36:205–241
Mertanen S, Pesonen LJ (2005) Drift history of the
Shield. In: Lehtinen M, Nurmi PA, R€am€o OT (eds) Precam-
brian geology of Finland – key to the evolution of the
Fennoscandian Shield, vol 14, Developments in Precam-
brian Geology. Elsevier B.V., Amsterdam, pp 647–668
Metzger AA (1924) Die jatulishen Bildungen von
Suojarvi in Ostfinland. Bulletin de la Commission
Geologique de Finland 64:1–86
Meunier A (2005) Clays. Springer, Berlin/Heidelberg/
New York, p 472
Nedachi Y, Nedachi M, Bennett G, Ohmoto H (2005)
Geochemistry and mineralogy of the 2.45 Ga Pronto
paleosols, Ontario, Canada. Chem Geol 214:21–44
Negrutza TF (1971) Precambrian weathering crusts in
Karelian complexes of northeastern Karelia. Proc Leningrad
Univ 24:57–64 (in Russian)
Negrutza VZ (1979) Metamorphosed weathering crusts
in eastern part of the Baltic Shield. In: Weathering crusts in
eastern part of the Baltic Shield and their mineralization.
Kola Science Centre of the USSR Academy of Sciences,
Apatity, p 11–28 (in Russian)
Negrutza VZ (1984) Early Proterozoic stages of evolu-
tion of the Eastern part of the Baltic Shield. Nedra,
Leningrad, p 270 (in Russian)
Nesbitt HW, Young GM (1982) Early Proterozoic cli-
mate and plate motions inferred from major element chem-
istry of lutites. Nature 299:715–717
Nordt LC, Driese SD (2010) New weathering index
improves paleorainfall estimates from Vertisols. Geology
38:407–410
Ohmoto H (1996) Evidence in pre–2.2 Ga paleosols for
the early evolution of atmospheric oxygen and terrestrial
biota. Geology 24:1135–1138
Ojakangas RW, Marmo JS, Heiskanen KI (2001) Basin
evolution of the Paleoproterozoic Karelian Supergroup of
the Fennoscandian (Baltic) Shield. Sediment Geol
141–142:255–285
9 7.9 Terrestrial Environments 1417
Pekkarinen LJ (1979) The Karelian formations and their
depositional basement in the Kiihtelysvaara – V€artsil€a area,
East Finland. Geol Surv Finl Bull 301:1–141
Prasad N, Roscoe SM (1996) Evidence of anoxic to
oxic atmospheric change during 2.45–2.22 Ga from lower
and upper sub–Huronian paleosols, Canada. Catena 27:
105–121
Righi D, Meunier A (1995) Origin of clays by rock
weathering and soil formation. In: Velde B (ed) Origin and
mineralogy of clays: clays and the environment.
Springer–Verlag, Berlin/Heidelberg/New–York, pp 43–161
Rainbird RH, Nesbitt HW, Donaldson JA (1990) Forma-
tion and diagenesis of a sub–Huronian saprolith: comparison
with a modern weathering profile. J Geol 98:801–822
Retallack GJ (2001) Soils of the past: an introduction to
paleopedology. Blackwell Science, Oxford, p 404
Rye R, Holland HD (1998) Paleosols and the evolution of
atmospheric oxygen: a critical review. Am J Sci 298:
621–672
Rye R, Holland HD (2000) Geology and geochemistry of
paleosols developed on the Hekpoort basalt, Pretoria Group,
South Africa. Am J Sci 300:85–141
Severmann S, Anbar AD (2009) Reconstruction
paleoredox conditions through a multitracer approach: the
key to the past is the present. Elements 5:359–364
Sheldon ND (2006) Precambrian paleosols and atmo-
spheric CO2 levels. Precambrian Res 147:148–155
Sheldon ND, Tabor NJ (2009) Quantitative paleoenvir-
onmental and paleoclimatic reconstruction using paleosols.
Earth Sci Rev 95:1–52
Sokolov VA, Heiskanen KI (1984) Stages of formation
of chemical weathering crusts in the Precambrian. In:
Proceedings of the 27th international geological congress,
Invited Papers, Vol 5 Book, Geology of the Precambrian,
VNU Science Press, Moscow, pp 35–43
Sokolov VA (1987) Jatuli superhorizon. In: Sokolov VA
(ed) Geology of Karelia. Nauka (Science), Leningrad, pp
51–59
Stafford SL (2005) Precambrian paleosols as indicators of
paleoenvironments on the early Earth. Ph.D. thesis, Univer-
sity of Pittsburgh
Sturt BA, Melezhik VA, Ramsay DM (1994) Early Pro-
terozoic regolith at Pasvik, NE Norway: palaeotectonical
implications for the Baltic Shield. Terra Nova 6:618–633
Sutton SJ, Maynard JB (1993) Sediment– and
basalt–hosted regoliths in the Huronian supergroup: role of
parent lithology in middle Precambrian weathering profiles.
Can J Earth Sci 30:60–76
Tosca NJ, Johnson DT, Mushegian A, Rothman DH,
Summons RE, Knoll AH (2010) Clay mineralogy, organic
carbon burial, and redox evolution in Proterozoic oceans.
Geochim Cosmochim Acta 74:1579–1592
Yamaguchi KE, Johnson CM, Beard BL, Beukes NJ,
Gutzmer J, Ohmoto H (2007) Isotopic evidence for iron
mobilization during Paleoproterozoic lateritization of the
Hekpoort paleosol profile from Gaborone, Botswana. Earth
Planet Sci Lett 256:577–587
Yang W, Holland HD (2003) The Hekpoort paleosol
profile in Strata 1 at Gaborone, Botswana: soil formation
during the great oxidation event. Am J Sci 303:187–220
Young GM, Long DGF, Fedo CM, Nesbitt HW (2001)
Paleoproterozoic Huronian basin: Product of a Wilson cycle
punctuated by glaciations and a meteorite impact. Sediment
Geol 141–142:233–254
1418 K. Kirsim€ae and V.A. Melezhik
7.9.3 Caliche
Alexander T. Brasier, Victor A. Melezhik, andAnthony E. Fallick
Introduction: What Is Caliche?
Caliche, when referring to calcium carbonate accumulations
within soils, is synonymous with calcrete. The word
‘dolocrete’ has sometimes been used where the mineral is
dolomite rather than calcite (e.g. Melezhik et al. 2004). Note
that caliche is not a soil type itself. Broader use of the term
caliche encompassing other types of terrestrial carbonate
(e.g. Aspler and Donaldson 1986) can be misleading, and
the term caliche is not restricted in this article to certain parts
of a caliche profile as advocated by Bertrand-Sarfati and
Moussine-Pouchkine (1983). Palaeoproterozoic deposits
are rarely reported but are known to exist, and these are
discussed below. General reviews of the abundant literature
on Phanerozoic and Recent caliche are provided by Esteban
and Klappa (1983), Wright and Tucker (1991), Milnes
(1992), Wright (2007), and Alonso-Zarza and Wright
(2010). Brasier (2011) discusses calcite precipitation
mechanisms and the physical appearance of caliche before
the arrival of vascular plants in the Palaeozoic.
Because caliche is often taken as an indicator of subaerial
exposure and specific palaeoenvironmental conditions, it is
desirable to make a distinction between true soil-related ped-
ogenic caliches (e.g. Gile et al. 1981) and varieties produced
by precipitation from groundwaters within the shallow sub-
surface, commonly called groundwater calcrete (e.g. Arakel
1986; Nash and Smith 2003). Caliche/calcrete is often used
in palaeoenvironmental reconstructions, yet distinguishing
between caliche and groundwater calcrete formed in these
two separate circumstances using petrographic and geochem-
ical techniques is difficult. Perhaps this is because most
exposed caliche examples are presently within the soil zone
and thus groundwater calcretes can exhibit a post-depositional
soil-zone overprint. Nevertheless, there are features peculiar
to pedogenic caliche, which may be readily recognised in the
field or in drillcores. However, many of these features such as
rhizoliths and alveolar septal fabric, are believed to be caused
by the roots of vascular plants and their symbiotic fungi (see,
for example, Table 7.10 and Brasier 2011) and so would not
be expected in the Palaeoproterozoic. Although terrestrial
microbesmay have existed in the Proterozoic (e.g. Horodyskiand Knauth 1994), fossil evidence for their existence is cur-
rently limited to some examples of lacustrine stromatolites
(see Chaps. 7.8.2 and 7.9.4).
Pedogenic caliche of the Pleistocene of Spain (Esteban
and Klappa 1983; Alonso-Zarza 1999) and New Mexico
(Gile et al. 1981; Machette 1985) developed within spe-
cific soil horizons (often in aridisols and vertisols), becom-
ing increasingly well developed (maturing) with age.
Caliches can thus be classified according to their stage of
development (Gile et al. 1966; Machette 1985), and it has
been suggested that these models are universally applica-
ble to pedogenic caliches. Some examples of modern ped-
ogenic caliche profiles are given in Esteban and Klappa
(1983).
Caliche formation begins when carbonate clasts in the top
soil horizons, often sourced from wind-blown dust (e.g.
Wright and Tucker 1991; Capo and Chadwick 1999), are
leached by downward percolating waters. If there is an annual
soil moisture deficit, this calcium carbonate is not carried
away in solution but re-precipitated lower down in the soil
profile by abiological (degassing and evaporation) and
biological (microbial or root-related) processes. An alterna-
tive model (Goudie 1983) involves precipitation of caliche
from waters rising from below as a result of evaporation and
capillary action. Initial precipitates are often thin coatings on
clasts or on soil peds with a few small nodules (stage I of
Machette 1985). In Stage II (Machette 1985), the nodules
become larger and more common, and rhizocretions (from
the late Silurian onwards; Brasier 2011) also increase in size
and number. The nodules begin to coalesce in Stage III
(Machette 1985) and a ‘honeycomb’ texture may develop,
with zones of uncemented soil surrounded by a cemented
framework. This leads to development of an impermeable
plugged ‘K’ horizon (Stage IV of Machette 1985). A plugged
horizon is shown in Fig. 7.143a. Downward percolating
waters, which pond in topographic depressions on the
surfaces of plugged horizons, can produce centimetre-scale
laminar crusts like those of Fig. 7.143b (commonly but not
exclusively seen in StageV ofMachette 1985), although these
can also form in other ways, such as by calcification of plant
root mats (Wright et al. 1988). Above the laminar crust, one
might expect to find a calcite-poor, leached argillitic ‘B hori-
zon’, but this has a much lower preservation potential than the
indurated caliche. Brecciation of the calcrete profile, often
with abiologically produced pisoliths, represents Stage VI of
Machette (1985). Biologically-induced brecciation, which
plays a significant role in the re-working of modern caliche
through rhizobrecciation, would have been less important
during the Palaeoproterozoic, but the shrinking and swelling
of clay minerals (smectites in particular) known as argillope-
doturbation would have been an important process.
A.T. Brasier (*)
Faculty of Earth and Life Sciences, VU University Amsterdam, De
Boelelaan 1085, 1081 HV Amsterdam, the Netherlands
e-mail: [email protected]
9 7.9 Terrestrial Environments 1419
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_9, # Springer-Verlag Berlin Heidelberg 2013
1419
Many features of pedogenic caliche are also found in
groundwater calcrete deposits, making them hard to distin-
guish. However, groundwater calcretes lack the vertical
profile organisation seen in pedogenic examples. Groundwa-
ter calcretes are often massive in appearance, including
‘plugged horizons’ cemented entirely by carbonate (e.g.
Nash and Smith 2003), similar to those found in Stage IV
pedogenic examples, but distinguished from them by the
lack of association with a vertical soil profile. Laminar
carbonate sheets of less than a millimetre to several
centimetres in thickness, which strongly resemble those
found in soil zones, are not rare in groundwater settings.
Nodules are similarly not exclusive to the soil zone, although
where plant roots are involved, pedogenic nodules may
exhibit vertical elongation. Such a criterion obviously can-
not be applied to the Palaeoproterozoic. Like many vadose
cements which (because of their different, often sparry
appearance) would not be considered caliche or groundwater
calcrete, laminar groundwater calcrete coatings on alluvial
fanglomerate clasts are often thicker on the clast undersides.
This may be attributed either to the effects of gravity in the
vadose zone or to evaporation of carbonate-precipitating
waters (Mack et al. 2000). Groundwater calcrete cements
often fill voids in permeable fluvial and alluvial channel
facies rather than muddy and more impermeable overbank
facies, leading to lens-shaped, carbonate-plugged bodies
(e.g. Nash and Smith 2003).
Palaeoenvironmental Significance of Caliche
It has long been assumed that pedogenic caliche is restricted
to semi-arid (but not necessarily warm) environments, where
evaporation exceeds precipitation; the latter being generally
between 100 and 750 mm per year (Cerling 1984). Season-
ality is of particular importance, as a wet period allowing
illuviation (downward washing of dissolved and particulate
calcium into the soil) must be followed by a period of
evaporation and precipitation. Further palaeoenvironmental
significance attached to caliche stems from the widely held
view (Wright and Tucker 1991; Capo and Chadwick 1999;
Alonso-Zarza andWright 2010) that the source of calcium in
the majority of modern cases is wind-blown dust and not the
underlying bedrock. Consequently, one might expect
enhanced caliche deposition during windy arid intervals
when supply of calcium is high (e.g. Brasier 2007) although
Candy and Black (2009) advocate an opposing hypothesis of
increased calcite precipitation during warmer, wetter
intervals in the Quaternary of Spain.
Significant build-ups of pedogenic caliche are estimated
(based on examples from New Mexico, USA, such as Gile
et al. 1966; Gile et al. 1981; Machette 1985) to take tens of
thousands to millions of years to mature. Consequently, it
has been interpolated that mature caliches require significant
intervals of tectonic and climatic stability to develop (e.g.
Leeder et al. 2008). Critically, it is still not clear how easily
these rates can be extrapolated between caliche formed in
different settings under different environmental conditions
and at different times. For example, Candy et al. (2004)
suggested that mature caliche profiles in the Mediterranean
can form in tens of thousands rather than hundreds of
thousands or millions of years, and Wright (1990) noted
that estimates of the time required for a mature stage IV
pedogenic calcrete (sensu Machette 1985) to form range
from three thousand years to more than a million years.
Caliche Petrography
Wright and Tucker (1991) helpfully divided hand-specimen
and thin-section scale pedogenic caliche fabrics into those of
biological affinity (beta fabrics) and those believed to be
abiological (alpha fabrics). Post-Ordovician soil-dwelling
biota have had a significant impact on the local concentrations
and cycling of elements (including carbon, oxygen, calcium;
see Hinsinger 1998; Berner et al. 2003; Brasier 2011), affect-
ing the local supersaturation of fluids with respect to calcite
and other minerals leading to (for example) rhizolith forma-
tion. It is reasonable to assume that themajority of alpha fabric
features might be found in Palaeoproterozoic examples,
whereas the majority of beta fabric features will not apply
(see also Brasier 2011). Palaeoproterozoic caliche should
therefore be dominated by petrographic features typical of
abiogenic calcite precipitation, i.e. display an alpha fabric.
Such features include floating grains (e.g. Fig. 7.144a, f, g–i),
reflecting the displacive growth of calcite; spar fringes around
detrital grains, indicating calcite replacement of or cementa-
tion around sediment particles (e.g. Fig. 7.144h, i); and wavy
lamination (Fig. 7.144d). Laminar crusts (such as Fig. 7.144e)
generated by abiological processes should also be present in
Palaeoproterozoic examples, although root-mat related forms
(Wright et al. 1988) will clearly be absent. Argillopedo-
turbation (soil turbation caused by shrinking and swelling of
clays) may have caused abiological pisolith formation in the
Palaeoproterozoic (Figs. 7.143c, d and 7.144a). Dripstone
fabrics (Fig. 7.144b, c) are similarly expected in such ancient
rocks, and are an indicator of mineral precipitation in the
vadose zone.
Because biological processes are more conspicuous in the
vascular plant-influenced soil zone than in many (but not all)
modern groundwater calcrete-precipitating settings, it is
likely that Palaeoproterozoic pedogenic caliche will closely
resemble groundwater calcrete except where vertical soil
profiles can be recognised. Such caliche may form nodules,
millimetre to several centimetre thick laminar sheets and
densely cemented plugged ‘hardpan’ or K (Gile et al. 1966)
soil horizons. Recently it has been suggested that ground-
water calcrete precipitation may have increased in volume
1420 A.T. Brasier et al.
following oxidation of the Earth’s atmosphere in the Palaeo-
proterozoic (Brasier 2011). This is because dissolution of
calcium sulphates, which had just begun to accumulate in
shallow marine and lacustrine settings (e.g. the Tulomozero
Formation; see Chaps. 6.3.1 and 6.3.2, and Melezhik et al.
2005), will have supplied a concentrated, new and readily
accessible source of calcium to groundwaters. This extra
calcium from calcium sulphate dissolution can cause super-
saturation of groundwaters with respect to the less soluble
calcite via the ‘common-ion effect’ (see Brasier 2011, and
references therein). Consequently, differentiating Palaeopro-
terozoic pedogenic caliche from groundwater calcrete may
prove very difficult.
Caliche Geochemistry
Carbon isotopic values of modern pedogenic caliche are
strongly influenced by the photosynthetic system employed
by the dominant vegetation in the soil (C3, C4 or CAM; see
Cerling 1984). However, in the apparent absence of thick
soils, Palaeoproterozoic caliche/calcrete examples are likely
to have formed in similar ways to modern groundwater types
(Brasier 2011), and so will have obtained their d13C signa-
ture from bedrock (often marine carbonate). Consequently,
d13C values of most Palaeoproterozoic caliche examples are
predicted to be much more positive than found in present-
day pedogenic caliche. However, any bacterial influences
could result in more negative d13C values. One might alter-
natively consider that caliche formed in pre-Ordovician thin,
organic-poor soils could have d13C signatures dominated by
atmospheric CO2, as suggested by Cerling (1984, 1991,
1992). A possible Neoarchaean (2.6 Ga) calcrete, which
may have formed in equilibrium with atmospheric CO2,
has a carbonate d13C of around +2‰ VPDB (Watanabe
et al. 2000). But strict criteria must be met for accurate
assessments of palaeoatmospheric CO2 levels from
palaeosol carbonate d13C, including confident identification
of pedogenic caliche rather than groundwater calcrete (see
Cerling 1991). Laminar caliche of the Kuetsj€arvi Sedimen-
tary Formation has a d13C of around +8‰ VPDB (Melezhik
et al. 2004), which is similar to that of the surrounding
carbonate bedrock, and so does not seem to reflect a signifi-
cant contribution from palaeoatmospheric CO2.
Modern caliche mineralogy is predominantly low-
magnesian calcite, although some dolomite examples are
known. For example, authigenic dolomite fills fractures in
caliches of Olduvai, Tanzania (Hay and Reeder 1978), and
Alonso-Zarza et al. (1998a) reported very early
dolomitisation of ‘Microcodium b’ calcified plant root
structures, leaving the structures excellently preserved and
their primary isotopic signals intact. Because Mg/Ca ratios
of precipitating fluids are likely to influence caliche miner-
alogy (e.g. Wright and Tucker 1991), one can speculate that
weathering of dolomitic marine carbonate bedrocks and
komatiitic lavas in the Proterozoic would have supplied
high levels of magnesium to terrestrial carbonates and thus
favour dolomite as the usual mineral of Palaeoproterozoic
caliche.
Examples of Palaeoproterozoic Caliche
Few examples of Palaeoproterozoic caliche have been
documented. Those which are known are described here,
with their stratigraphic context shown in Fig. 7.145.
Chuniespoort Group, South Africa2.6 Ga pre-Chuniespoort Group silcretes and dolocretes of
grey colour form erosion-resistant escarpments in Eastern
Transvaal, South Africa (Martini 1994; Fig. 7.145e), and the
claim by Martini (1994) that these are the oldest reported
caliche examples still seems to be true (Brasier 2011).
Silcrete-dolocrete duricrust grades upwards into brecciated
silicrete in a matrix of silicified mudstone. The dolocrete is
found in ‘stratified lenses’ and inclusions up to several
decimetres across, although much of the dolostone has
been replaced by silica. It is intriguing that dolocrete is
found on dunite where magnesite might normally be
expected due to a lack of calcium. Martini (1994) suggests
a magnesite precursor may have been dolomitised by later
addition of calcium from a marine source. Alternatively,
externally sourced wind-blown calcium might be
implicated, similar to modern pedogenic caliche of the
south west USA (Capo and Chadwick 1999). Possible
2.6 Ga caliche from this area was further studied by
Watanabe et al. (2000; Fig. 7.145f). They suggested a pedo-
genic origin for the carbonate involving evaporatic concen-
tration of waters in a semi-arid environment. The latter
authors suggest that calcium may have been supplied from
weathering of nearby granitic rocks, and bicarbonate from
the high pCO2 atmosphere.
Kuetsj€arvi Sedimentary Formation,Fennoscandian RussiaProbable Palaeoproterozoic dolomitic caliche was described
by Melezhik et al. (2004) from the Kuetsj€arvi Sedimentary
Formation in the Pechenga Greenstone Belt (Fig. 7.145a).
This formation is overlain by volcanic rocks dated at
c. 2.06 Ga (Melezhik et al. 2007). The Kuetsj€arvi dolocretesare numerous 0.5–5-cm-thick crusts consisting of laminated
and non-laminated (zone 3 in Fig. 7.146a, b) dolomicrite and
partially silicified, orbicular rocks (zone 5 in Fig. 7.146b, d).
Silicification is interpreted as an early event on the basis of
petrography, quite possibly having occurred within the
vadose zone as reported in modern cases (e.g. Arakel et al.
1989). Individual laminar crusts cover up to a few square
metres and are developed on carbonate substrates where they
9 7.9 Terrestrial Environments 1421
follow irregularities and fill palaeo-topographic depressions
(Fig. 7.145a). Breccias (Fig. 7.146c; zone 4 on Fig. 7.146b)
interpreted to have formed through dissolution and collapse
(Melezhik et al. 2004) or perhaps from tension fracturing
during displacive growth of crystals in competing directions,
are found in smooth-walled microcavities. Laminae are
picked out by colour banding and changes in crystal size.
Discontinuities between, and truncations of, individual
laminae are not observed (Melezhik et al. 2004). As noted
by Melezhik et al. (2004), “Silicified nodular zones” of
laminar crusts illustrated here in Fig. 7.146d seem similar
to the Neoproterozoic “orbicular crusts” of Bertrand-Sarfati
and Moussine-Pouchkine (1983). Features described by
Melezhik et al. (2004), which suggest that these crusts are
laminar caliches, include the dissolution or fracture
microcavities (Fig. 7.146c), glaebules (nodules;
Fig. 7.146d) and carbonate-coated grains. Possible gypsum
pseudomorphs are also found in these dolocretes.
Gamagara Formation, South AfricaPalaeokarst-hosted pisolitic, haematitic laterites are found in
the Palaeoproterozoic (~2.0–2.2 Ga) Gamagara Formation,
South Africa (Gutzmer and Beukes 1998; Figs. 7.145d
and 7.147). The sediments hosting the laterites are
interpreted as fluvial overbank deposits, which overlie chan-
nel conglomerates. The laterite profiles are described as
consisting of “an iron-depleted white bleached topsoil zone
and a well-defined ferric duricrust underlain by a pisolitic to
mottled saprolitic zone” (Gutzmer and Beukes 1998). The
haematite duricrust is undulating, 2–5 mm thick and very
finely laminated. A strong similarity to modern terrestrial,
vegetation-influenced laterite profiles led Gutzmer and
Beukes (1998) to suggest that these Gamagara Formation
laterites are evidence of a microbial surface cover in Pre-
cambrian terrestrial settings at ~2.0–2.2 Ga. Equivalent cali-
che examples are known (J. Gutzmer, pers. comm.) but have
apparently never been studied.
Nonacho Basin, NW CanadaNodules in Palaeoproterozoic sandstones (Fig. 7.148a–c) of
the Nonacho Basin, NW Canada, were interpreted as non-
pedogenic, groundwater-related cements by Aspler and
Donaldson (1986). Unpublished dates from zircons confirm
that the Nonacho Basin is younger than 1.91 Ga, and cross-
cutting dykes give a minimum age of 1.83 Ga (Aspler 2010
pers. comm.). The possibility of a pedogenic origin for the
nodules is rejected by Aspler and Donaldson (1986) on the
basis that nodules are absent in fine-grained overbank
sediments. They further suggest that these have not formed
by pedogenic processes but are ‘merely zones of early
cementation’. Outcropping beds of ‘massive carbonate’ are
generally <1 m thick and separated by distances of 0.5–5 m
from each other. This calcite cement (Fig. 7.148d) ‘serves to
highlight bedding, channels and clast fabrics in otherwise
massive-appearing conglomerates’. Such carbonates may be
compared with modern groundwater calcretes, such as those
described by Nash and Smith (2003), and are unlikely to be
true pedogenic caliche.
Kanuyak Formation, NW CanadaThe Mesoproterozoic Kanuyak Formation (see Pelechaty
et al. 1991; Pelechaty and James 1991) of Northwest Canada
contains sediments interpreted as dolomitised caliche
horizons up to 5 m thick. These are dominantly found in
palaeokarst valleys. Ooliths and pisoliths up to 5 cm in
diameter are found in 1-mm- to 50-cm-wide, sediment-filled
cracks, as well as forming reversely graded beds (exhibiting
fewer and smaller ooliths and pisoliths with increasing
depth) up to 1 m thick near profile tops (Fig. 7.145c). Ooliths
and pisoliths are also found dispersed at other levels in
profiles. Some of these ooliths and pisoliths are concentri-
cally laminated coatings on clastic grains, and others are
described as non-laminated particles of massive, micritic
dolomite and so might be better termed peloids. Rare tepee
structures and brecciation attributed to expansion-related
processes are also compatible with a caliche origin.
Pelechaty and James (1991) suggest that speleothem fabrics
(see Chap. 7.9.4) were originally of calcitic composition on
the basis of crystallographic similarity with calcareous
cements. The original composition of the dolocretes is
uncertain, but Pelechaty and James (1991) suggest calcite
is likely. Carbonate carbon isotope values range from
�1.22 ‰ to +0.84 ‰ PDB, consistent with a lack of terres-
trial vegetation during formation of these pedogenic fabrics.
Oronto Group, Michigan, USACalcareous horizons in fluvial sediments of 1.1 Ga age
(Fig. 7.145b) are interpreted as Stage III maturity (Gile
et al. 1966) caliche by Kalliokoski (1986). Fabrics include
microscopic calcite veinlets, calcite-coated basalt clasts, and
3-mm-diameter ‘nodules’. Calcite rinds around pebbles are
sometimes thicker on the base of the encased clast, as often
seen in modern caliche formed in alluvial fan conglomerates
(e.g. Alonso-Zarza et al. 1998b; Mack et al. 2000; Brasier
2007). Other localities described by Kalliokoski (1986)
exhibit calcite-coated pebbles and void-filling cements in
conglomerates but there is a lack of any vertical profile.
These could be an example of Proterozoic gully-bed cements
in an alluvial fan setting.
Synthesis of Palaeoproterozoic Caliches
Documented examples of Palaeoproterozoic caliche are not
common, but this must partly reflect their lack of preserva-
tion, and not solely their non-deposition. From the few
examples uncovered to date, it is clear that many abiological
features seen in caliche of the modern world were also found
1422 A.T. Brasier et al.
in the Palaeoproterozoic. Easily recognised features such as
nodules (Melezhik et al. 2004) and pisoliths (Gutzmer and
Beukes 1998; Pelechaty and James 1991) might testify to the
former presence of soils, andmight even be taken as evidence
of a primitive microbial land cover. Indeed, the relatively
greater preservation potential of these duricrusts compared to
the more easily eroded argillitic horizons makes them prime
targets to look for in the search for early terrestrial
environments. However, without the key biological fabrics
of modern pedogenic caliche to aide identification (see also
Brasier 2011), many Palaeoproterozoic examples could also
be interpreted as ‘groundwater calcretes’. Fabrics of ground-
water calcretes which could (or have been) identified in the
Palaeoproterozoic include some nodules (e.g. Aspler and
Donaldson 1986), laminar crusts (e.g. Melezhik et al.
2004), wavy lamination and tepee structures (e.g. Pelechaty
and James 1991), floating grain fabric, circum-granular
cracks, spar fringes around grains (e.g. Kalliokoski 1986)
and crystallaria (Melezhik et al. 2004). The distinction
between pedogenic caliche and groundwater calcrete is
important. It has implications for the processes involved in
genesis of the rock, and also for the interpretation of any
geochemical data the rock holds. As an example, carbon
stable isotopes of pedogenic caliche can sometimes be used
to reconstruct palaeoatmospheric pCO2 levels. The same is
not true of those groundwater calcretes which do not precipi-
tate in equilibrium with atmospheric CO2, and which often
derive their carbon from bedrock rather than the atmosphere
(e.g. Cerling 1991). This means that any attempt to derive
Palaeoproterozoic atmospheric pCO2 levels from caliche-
like rocks will require a detailed study of their stratigraphic
context, and not just examination of individual hand-
specimens. The same is true when considering the possibility
that caliche might be evidence of climatic and tectonic sta-
bility over hundreds of thousands or millions of years: this is
not necessarily true of groundwater calcretes.
Table 7.10 Some common caliche fabrics and their apparent maximum ages (Modified and adapted from Brasier (2011))
Fabric Description Example reference(s)
Apparent
maximum age
Laminar crusts Laminar accretions of carbonate, commonly but not necessarily found
ponded on impermeable surfaces
Wright and Tucker
(1991), Wright et al.
(1988)
Archaean
Nodules/
glaebules
Indurated concentrations of carbonate which may be spherical to irregular in
shape, contained within a matrix.
Wieder and Yaalon
(1974)
Archaean
Circum-
granular
cracking
Non-tectonic fractures around grains produced by shrinking and swelling of
clay minerals during desiccation and hydration
Esteban and Klappa
(1983)
Archaean
Floating/
exploded grain
fabric
Silt and sand sized grains floating in a micritic matrix where the matrix
appears to be supporting the grains
Esteban and Klappa
(1983), Tandon and
Friend (1989)
Archaean
Pisoliths Concentrically laminated sphaeroids > 2 mm diameter Esteban and Klappa
(1983), Wright and
Tucker (1991)
Archaean
Tepees Antiformal structures produced by buckling during expansion Assereto and Kendall
(1977)
Archaean
Crystallaria Calcite-filled cracks, mostly formed through desiccation and expansion Wright and Tucker
(1991)
Archaean
Spar fringes Calcite spar coated grains, possibly formed as a result of the spar replacing
the grain
Wright and Tucker
(1991)
Archaean
Fungal needle
fibre calcite
Acicular crystals of calcite James (1972),
Verrecchia and
Verrecchia (1994)
Mesoproterozoic?
Faecal peloids Rounded micrite pellets resulting from invertebrate defaecation in the soil Esteban and Klappa
(1983)
Silurian
Rhizoliths Organosedimentary structures produced by roots Klappa (1980) Late Silurian?
Alveolar septal
fabric
‘Cylindrical to irregular pores, which may or may not be filled with calcite
cement, separated by a network of anastomosing micrite walls’ (Esteban and
Klappa 1983)
Esteban and Klappa
(1983)
Late Silurian?
Laminar root
mat crusts
Laminar caliche crusts resulting from calcification of root mats, exhibiting
‘tubular fenestrae’
Wright et al. (1988),
Alonso-Zarza (1999)
Late Silurian?
Microcodium b Calcified plant root cells Alonso-Zarza et al.
(1998a)
Carboniferous
Microcodium Calcite crystals resembling biological ‘cells in palisades around small
nucleii’
Klappa (1978) Cretaceous
9 7.9 Terrestrial Environments 1423
Fig. 7.143 Outcrop images of Holocene caliches from Kimberley
area, South Africa. (a) Thick, mature caliche profile developed in
overbank sediments. (b) Plan-view of a caliche cap completely coating
rock clasts. (c) Plan-view of partially eroded caliche crust exhibiting
angular rock fragments floating in a micritic calcite matrix. (d) Large
pisoids where rock fragments appear as nuclei surrounded by concen-
tric micritic and microspar laminae. Field book for scale is 17 cm long.
Divisions in the scale-bar are 1 cm (Photographs by Victor Melezhik)
1424 A.T. Brasier et al.
Fig. 7.144 Holocene caliche from Kimberley, South Africa. (a) Pho-
tomicrograph of pisoid in micritised groundmass with detrital quartz
grains. (b) Scanned slab of caliche cap showing calichified sediment
with micritic vadose cement (arrowed) above micritised groundmass
(M) (c) Photomicrograph of stalactite-like pendant cement shown in (b).
9 7.9 Terrestrial Environments 1425
Fig. 7.144 (continued) (d) Thin-section photomicrograph of a cali-
che cap with pisolitic layer. The overlying layer exhibits wavy lamina-
tion with smaller pisoids and coated grains. (e) Scanned polished slab
of laminated caliche crust developed on top of fluvial sand. (f) Thin-
section photomicrograph of cross-section through laminated caliche
composed of peloidal groundmass with scattered quartz grains (white)and a dissolution cavity filled with microsparitic calcite laminae (top);the latter is enlarged in (g) (Photographs by Victor Melezhik).
1426 A.T. Brasier et al.
Fig. 7.144 (continued) (h) Thin-section photomicrograph of a Pleis-
tocene caliche nodule from Corinth, central Greece. Clasts are
surrounded by sparry calcite cements (stained pink with Alizarin Red
S) and calcite-cemented, circum-granular cracks. The different crystal
sizes of the matrix, in which the clasts are ‘floating’, also give a
‘mosaic’ fabric. (i) Thin-section photomicrograph of a Pleistocene
caliche nodule from Corinth, central Greece. A silicate clast is being
dissolved through pressure solution, with calcite microspar around the
grain (stained pink) growing into the resulting void (Photomicrographs
by Alex Brasier)
9 7.9 Terrestrial Environments 1427
Fig. 7.145 Stratigraphic context of Proterozoic caliche and laterite
examples. (a) Kuetsj€arvi Sedimentary Formation, Fennoscandia, based
on Drillhole X and adapted from Melezhik et al. (2004). A more
extensive key to lithology is included in Melezhik et al. (2004).
(b) Calumet and Hecla Conglomerate, Oronto Group, Michigan, USA
(Adapted from Kalliokoski (1986)). (c) Kanuyak Formation, NW
Canada (Adapted from Pelechaty and James (1991)). (d) Gamagara
Formation haematitic laterites, South Africa (Adapted from Gutzmer
and Beukes (1998)). (e) pre-Chuniespoort Group Archaean basement,
Eastern Transvaal (Adapted from Martini (1994)). (f) Archaean base-
ment, Eastern Transvaal (Adapted from Watanabe et al. (2000))
1428 A.T. Brasier et al.
Fig. 7.146 Probable caliche of the c. 2060Ma Kuetsj€arvi Sedimentary
Formation, Pechenga Greenstone Belt. (a) Polished slab showing a
cross section through dolocrete crust developed above travertine (2)
veneering sandy dolostone (1). The crust consists of red, iron-stained,non-laminated dolomicrite (3) with a silicified, nodular, vuggy zone (4)
followed by white “chalky” dolostone (5). (b) Photomicrograph of a
section through non-laminated, iron-stained dolocrete (3) developed on
uneven, dissolved travertine surfaces, and capped by dissolution brec-
cia (4) and silicified nodular zone (5). The silicified nodular zone is
overlain by sandy dolomicrite (6) showing no silicifcation. Solution-
enlarged fractures developed in the dolocrete (3) are filled with
microdolospar and microcrystalline silica. Fabrics are crosscut by
later metamorphic veinlets filled with dolomite and quartz. (c) Photo-
micrograph showing details of dissolution or a fracture breccia zone
(4 in b). Solution cavities (or fractures) have smooth walls, lined by
earlier isopachous dolomite cement. Grey dolomicrite clast (DM) is
coated with earlier microdolospar and multiple generations of pendant
cement. Below the pendant cement are several partially dissolved
quartz grains suspended in dolomicrite. A larger vug beneath them is
filled with late equant dolospar. All earlier fabrics are crosscut by a
metamorphic quartz-dolomite veinlet.
9 7.9 Terrestrial Environments 1429
The link between modern pedogenic caliche and semi-
arid climates has been extrapolated through time to the
Palaeoproterozoic (e.g. Watanabe et al. 2000). However,
this only holds true if evaporation was an important process
in carbonate precipitation. Groundwater calcretes commonly
precipitate via the ‘common-ion’ effect (e.g. Arakel 1986).
The suggestion that the ‘common-ion’ effect might have
caused an increase in the number of groundwater calcrete
deposits following atmospheric oxygenation and calcium
sulphate deposition (Brasier 2011) is hard to judge on the
limited evidence currently available. This demands a search
for new examples, and further investigation of those which
are now known. The FAR-DEEP cores which penetrate the
Kuetsj€arvi Sedimentary Formation provide an excellent
opportunity for this.
Conclusions and Implications for the FAR-DEEPCores
Whilst many Phanerozoic biological pedogenic fabrics are
not applicable to the Palaeoproterozoic, a number of
abiological features are common to caliche of all ages.
Nodules, laminar crusts and coated grains are commonly
reported from Palaeoproterozoic examples, and their further
study may reveal evidence of palaeoenvironmental pro-
cesses and conditions during their formation. It is interesting
to note that most of these fabrics are recognised in modern
‘groundwater calcrete’ settings, and thus are not necessarily
indicative of pedogenic caliche in the Palaeoproterozoic, for
which evidence is more limited.
Understanding how and where different caliche fabrics
develop assists interpretations of depositional environments,
stratigraphic frameworks and post-depositional diagenesis.
The recognition and analysis of Palaeoproterozoic caliche
within the FAR-DEEP cores may thus help to illuminate
questions on the condition of terrestrial surfaces and their
weathering at a critical time in Earth history. Additionally,
study of these horizons will help to answer more general
questions on the nature of non-marine environments and
sedimentation in the absence of terrestrial vegetation:
questions which are as applicable to the majority of Earth
history as they are to terrestrial planets such as Venus and
Mars.
Fig. 7.146 (continued) (d) Photomicrograph showing details of a
section through the silicified nodular zone (4 in a) developed on
dissolved surface of the fractured iron-stained, non-laminated
dolomicrite (3 in a). Nodular dolomicrite (grey) is suspended in inten-
sively silicified dolomicritic and microdolosparitic cement. Silica
replaces earlier dolomicritic cement, some grey dolomicrite nodules
and is also found as white micronodules and silica crust (white).Multiple coatings of dolomicrite and silica indicate a complex diage-
netic history (Images are reproduced from Melezhik et al. (2004) with
permission of Elsevier)
1430 A.T. Brasier et al.
Fig. 7.147 Haematitic pisolitic laterite from the ~2.0–2.2 Ga
Gamagara Formation, South Africa. Similar textures should be found
in caliche examples. Equivalent caliches of this age are known from the
same area but have not been studied. Scale bar is 1 cm (Photograph
courtesy of Jens Gutzmer)
9 7.9 Terrestrial Environments 1431
Fig. 7.148 Palaeoproterozoic nodules and carbonate cements in
coarse-grained non-marine sediments of the Nonacho Basin, NW
Canada. (a), (b) and (c) Concretions (arrowed) representing early
partial carbonate cementation elongated parallel to bedding; with
deformation of layers due to differential compaction (fluvial
sandstones). Scale on notebook in (c) is in inches. (d) Calcite (arrowed)cementing, encrusting, and partially replacing pebbles in fluvial con-
glomerate (Photographs courtesy of Lawrence Aspler)
1432 A.T. Brasier et al.
References
Alonso-Zarza AM (1999) Initial stages of laminar calcrete
formation by roots: examples from the Neogene of central
Spain. Sediment Geol 126:177–191
Alonso-Zarza AM, Wright VP (2010) Calcretes. In:
Alonso-Zarza AM, Tanner LH (eds) Carbonates in continen-
tal settings: facies, environments, and processes, vol 61,
Developments in sedimentology. Elsevier, Amsterdam, pp
225–267
Alonso-Zarza AM, Sanz ME, Calvo JP, Estevez P
(1998a) Calcified root cells in Miocene pedogenic
carbonates of the Madrid Basin: evidence for the origin of
Microcodium b. Sediment Geol 116:81–97
Alonso-Zarza AM, Silva PG, Goy JL, Zazo C (1998b)
Fan-surface dynamics and biogenic calcrete development:
interactions during ultimate phases of fan evolution in the
semiarid SE Spain (Murcia). Geomorphology 24:147–167
Arakel AV (1986) Evolution of calcrete in
palaeodrainages of the Lake Napperby area, Central
Australia. Palaeogeogr Palaeocl Palaeoecol 54:283–303
Arakel AV, Jacobson G, Salehi M, Hill CM (1989) Silic-
ification of calcrete in palaeodrainage basins of the Austra-
lian arid zone. Aust J Earth Sci 36:73–89
Aspler LB, Donaldson JA (1986) Paleoclimatology of
Nonacho Basin (Early Proterozoic), Northwest territories,
Canada. Palaeogeogr Palaeocl Palaeoecol 56:17–34
Assereto RLAM, Kendall CG (1977) Nature, origin and
classification of peritidal tepee structures and related
breccias. Sedimentology 24:153–210
Berner EK, Berner RA, Moulton KL (2003) Plants and
mineral weathering: present and past. In: Holland HD,
Turekian KK (eds) Treatise on geochemistry, vol 5.
Pergamon, Oxford, pp 169–188
Bertrand-Sarfati J, Moussine-Pouchkine A (1983) Pedo-
genetic and diagenetic fabrics in the Upper Proterozoic
Sarnyere formation (Gourma, Mali). Precambrian Res
20:225–242
Brasier AT (2007) Palaeoenvironmental reconstructions
from Quaternary Tufas and Calcretes, Central Greece.
Unpublished Ph.D. thesis, University of East Anglia,
Norwich,p 293
Brasier AT (2011) Searching for travertines, calcretes and
speleothems in deep time: processes, appearances,
predictions and the impact of plants. Earth Sci Rev
104:213–239
Candy I, Black S (2009) The timing of Quaternary
calcrete development in semi-arid southeast Spain:
investigating the role of climate on calcrete genesis. Sedi-
ment Geol 218:6–15
Candy I, Black S, Sellwood BW (2004) Quantifying time
scales of pedogenic calcrete formation using U-series
disequilibria. Sediment Geol 170:177–187
Capo RC, Chadwick OA (1999) Sources of strontium and
calcium in desert soil and calcrete. Earth Planet Sci Lett
170:61–72
Cerling TE (1984) The stable isotopic composition of
modern soil carbonate and its relationship to climate. Earth
Planet Sci Lett 71:229–240
Cerling TE (1991) Carbon dioxide in the atmosphere:
evidence from Cenozoic and Mesozoic paleosols. Am J Sci
291:377–400
Cerling TE (1992) Further comments on using carbon
isotopes in palaeosols to estimate the CO2 content of the
palaeo-atmosphere. J Geol Soc Lond 149:673–676
Esteban M, Klappa CF (1983) Subaerial exposure envi-
ronment. In: Scholle PA, Bebout DG, Moore CH (eds)
Carbonate depositional environments. American Associa-
tion of Petroleum Geologists, Tulsa, pp 1–54
Gile LH, Peterson FF, Grossman RB (1966) Morphologi-
cal and genetic sequences of carbonate accumulations in
desert soils. Soil Sci 101:357–360
Gile LH, Hawley JW, Grossman RB (1981) Soils and
geomorphology in the Basin and Range area of southern
New Mexico – Guidebook to the desert project, Memoir
39, New Mexico Bureau of Mines and Mineral Resources,
Socorro, p 222
Goudie AS (1983) Calcrete. In: Goudie AS, Pye K (eds)
Chemical sediments and geomorphology. Academic,
London, pp 93–131
Gutzmer J, Beukes NJ (1998) Earliest laterites and possi-
ble evidence for terrestrial vegetation in the early Protero-
zoic. Geology 26:263–266
Hay RL, Reeder RJ (1978) Calcretes of Olduvai Gorge
and the Ndolanya Beds of northern Tanzania. Sedimentol-
ogy 25:649–673
Hinsinger P (1998) How do plant roots acquire mineral
nutrients? Chemical processes involved in the rhizosphere.
Adv Agron 64:225–265
Horodyski RJ, Knauth LP (1994) Life on land in the
Precambrian. Science 263:494–498
James NP (1972) Holocene and Pleistocene calcareous
crust (caliche) profiles, criteria for subaerial exposure. J
Sediment Res 42:817–836
Kalliokoski J (1986) Calcium carbonate cement (caliche)
in Keweenawan sedimentary rocks (1.1 Ga), upper Penin-
sula of Michigan. Precambrian Res 32:243–259
Klappa CF (1978) Biolithogenesis of Microcodium: elu-
cidation. Sedimentology 25:489–522
Klappa CF (1980) Rhizoliths in terrestrial carbonates:
classification, recognition, genesis and significance. Sedi-
mentology 27:613–629
Leeder MR, Mack GH, Brasier AT, Parrish RR, McIntosh
WC, Andrews JE, Duermeijer CE (2008) Late-Pliocene
timing of Corinth (Greece) rift-margin fault migration.
Earth Planet Sci Lett 274:132–141
9 7.9 Terrestrial Environments 1433
Machette MN (1985) Calcic soils of the southwestern
United States. Geol Soc Am Spec Pap 203:1–21
Mack GH, Cole DR, Trevino L (2000) The distribution
and discrimination of shallow, authigenic carbonate in the
Pliocene-Pleistocene Palomas Basin, southern Rio Grande
rift. Geol Soc Am Bull 112:643–656
Martini JEJ (1994) A late Archaean-Palaeoproterozoic
(2.6 Ga) palaeosol on ultramafics in the Eastern Transvaal,
South Africa. Precambrian Res 67:159–180
Melezhik VA, Fallick AE, Grillo SM (2004) Subaerial
exposure surfaces in a Palaeoproterozoic 13C-rich dolostone
sequence from the Pechenga Greenstone Belt: palaeoenvir-
onmental and isotopic implications for the 2330–2060 Ma
global isotope excursion of 13C/12C. Precambrian Res
133:75–103
Melezhik VA, Fallick AE, Rychanchik DV, Kuznetsov
AB (2005) Palaeoproterozoic evaporites in Fennoscandia:
implications for seawater sulphate, the rise of atmospheric
oxygen and local amplification of the d13C excursion. Terra
Nova 17:141–148
Melezhik VA, Huhma H, Condon DJ, Fallick AE,
Whitehouse MJ (2007) Temporal constraints on the
Palaeoproterozoic Lomagundi-Jatuli carbon isotopic event.
Geology 35:655–658
Milnes AR (1992) Calcrete. In: Martini IP, Chesworth W
(eds) Weathering, soils and Palaeosoils. Elsevier, Oxford, pp
309–347
Nash DJ, Smith RF (2003) Properties and development of
channel calcretes in a mountain catchment, Tabernas Basin,
southeast Spain. Geomorphology 50:227–250
Pelechaty SM, James NP (1991) Dolomitized middle
Proterozoic calcretes, Bathurst Inlet, Northwest Territories,
Canada. J Sediment Res 61:988–1001
Pelechaty SM, James NP, Kerans C, Grotzinger JP (1991)
A middle Proterozoic palaeokarst unconformity and
associated sedimentary rocks, Elu Basin, northwest Canada.
Sedimentology 38:775–797
Tandon SK, Friend PF (1989) Near-surface shrinkage and
carbonate replacement processes, Arran Cornstone Forma-
tion, Scotland. Sedimentology 36:1113–1126
Verrecchia EP, Verrecchia KE (1994) Needle-fiber cal-
cite; a critical review and a proposed classification. J Sedi-
ment Res 64:650–664
Watanabe Y, Martini JEJ, Ohmoto H (2000) Geochemi-
cal evidence for terrestrial ecosystems 2.6 billion years ago.
Nature 408:574–578
WiederM, Yaalon DH (1974) Effect of matrix composition
on carbonate nodule crystallization. Geoderma 11:95–121
Wright VP (1990) Estimating rates of calcrete formation
and sediment accretion in ancient alluvial deposits. Geol
Mag 127:273–276
Wright VP (2007) Calcrete. In: Nash DJ, McLaren SJ
(eds) Geochemical sediments and landscapes. Wiley-
Blackwell, Oxford, pp 10–45
Wright VP, Tucker ME (1991) Calcretes: an introduction.
In: Wright VP, Tucker ME (eds) Calcretes. Blackwell Sci-
entific Publications, Oxford, pp 1–22
Wright VP, Platt NH, Wimbledon WA (1988) Biogenic
laminar calcretes: evidence of calcified root-mat horizons in
paleosols. Sedimentology 35:603–620
1434 A.T. Brasier et al.
7.9.4 Earth’s Earliest Travertines
Alexander T. Brasier, Paula E. Salminen, Victor A.Melezhik, and Anthony E. Fallick
Introduction
Carbonates precipitated in terrestrial spring, stream, lacus-
trine and hot spring environments (tufas and travertines) are
mostly studied from Cenozoic and Recent strata, though
much older deposits include Proterozoic travertines of the
Pechenga Basin (Melezhik and Fallick 2001; Melezhik et al.
2004; Figs. 7.149 and 7.150). Despite the great age differ-
ence, aspects of many models relevant to the Pleistocene can
be considered relevant to the Proterozoic.
For example, the precipitation of such carbonates in non-
marine settings occurs when fluids become supersaturated
with respect to calcite. The calcium source is often bedrock
(commonly but not necessarily limestone or dolostone)
dissolved by CO2-rich waters. Degassing or photosynthetic
uptake of CO2 drives the reaction below to the right, causing
precipitation of calcium carbonate:
Ca2þ þ 2HCO3� $ CaCO3 þ CO2 þ H2O
This reaction has probably been driving calcite precipita-
tion in terrestrial environments since the Archaean, perhaps
with some assistance by bacteria and cyanobacteria.
The aim of this review is to provide a framework for
analysis of relatively rare Palaeoproterozoic travertines
based upon knowledge gleaned from younger rocks. Before
examining possible Precambrian examples, it is first neces-
sary to define terms and review depositional models relevant
to a world without land plants or CO2-rich soils.
Travertine or Tufa?
The words travertine and tufa have been used since Roman
times and have been continually re-defined (e.g. Irion and
Muller 1968; Pedley 1990; Koban and Schweigert 1993;
Pentecost and Viles 1994; Rainey and Jones 2009), so it is
now always necessary to give a definition when using
these terms. Here we follow common European usage as
advocated by Pedley (1990), Riding (1991) and Ford and
Pedley (1996) who suggested ‘travertine’ be reserved for
thermal and hydrothermal calcium carbonate deposits
lacking macrophyte remains. Tufa is used only in reference
to cool or ‘near ambient’ temperature deposits, which
because of their environment commonly contain moulds of
macrophytes in build-ups of Recent age (Pedley 1990; Ford
and Pedley 1996). Clearly pre-Ordovician deposits will lack
macrophytes (see Brasier, 2011), so the division here is
between hot and cool water deposits: a distinction which is
not easy to make with confidence in the ancient rock record.
Furthermore, Recent barite travertine is forming in Alberta
(e.g. Bonny and Jones 2008), so the composition does not
have to be calcium carbonate for the terms to be applicable.
Pentecost (1993) uses the term travertine in an all-
encompassing way for deposits precipitated predominantly
by de-gassing of carbon dioxide in environments below
springs. Pentecost and Viles (1994) and Pentecost (1995)
suggested meteogene travertine for deposits involving CO2
from soil atmospheres, and thermogene travertine for
deposits “whose carrier gas comes from thermal activity
involving oxidation, decarbonation and other deep
outgassing processes in tectonically active areas”.
Many authors have used the term travertine for ambient
temperature freshwater spring, stream and lacustrine
precipitates, including Love and Chafetz (1988); Pentecost
and Viles (1994); Soligo et al. (2002); O’Brien et al. (2006);
Anzalone et al. (2007); and Hammer et al. (2007) amongst
others. Such usage is common in North America.
Brasier (2011) suggested avoiding the terms ‘tufa’ and
‘travertine’ where there is no intention to imply depositional
conditions which cannot be measured or inferred. When
there is an intention to imply specific conditions of deposi-
tion, the term tufa can usefully be reserved for cold and
ambient temperature, and travertine for hydrothermal
precipitates (e.g. Pedley 1990). Because there is some con-
fusion in terminology, this chapter includes some discussion
of ambient temperature deposits (tufas) as well as those we
would consider true hydrothermal travertines.
To reiterate, we follow Pedley (1990) and use the term
travertine in a genetic way for non-marine authigenic spring,
stream and lacustrine precipitates where above ambient
(‘hydrothermal’) water temperatures are suspected; and tufawhere deposition is believed to have taken place at ambient
temperatures. When using the terms tufa and travertine, we
make no judgement on whether included (micro)organisms
caused mineral precipitation: a topic which itself is the sub-
ject of heated debate (see references spanning from Weed
1888, cited in Chafetz and Folk 1984, to Kawai et al. 2009).
An unfortunate consequence of using definitions based on
environmental conditions is a grey area where the more
descriptive term ‘stromatolite’ (which applies to both marine
and non-marine settings) may equally be applied to some
A.T. Brasier (*)
Faculty of Earth and Life Sciences, VU University Amsterdam,
De Boelelaan 1085, 1081 HV Amsterdam, The Netherlands
e-mail: [email protected]
9 7.9 Terrestrial Environments 1435
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_9, # Springer-Verlag Berlin Heidelberg 2013
1435
layered tufas (e.g. Andrews and Brasier 2005) and
travertines. Stromatolites are covered in more detail in
Chap. 7.8.2, with a few non-marine examples also included
here. Stromatolites formed in marine settings or by aggluti-
nation of clastic sediment involving cyanobacteria are never
considered tufas or travertines.
Travertine or Speleothem?
A useful distinction may be drawn between spring and
stream carbonates (be they tufa or travertine) and cave
deposits (speleothem) on the basis of absence vs. presence
of fossil photosynthesising organisms: speleothem preci-
pitates in the dark (see for example Thrailkill 1976) from
meteoric waters. This is not always an easy distinction to
make since tufa and travertine frequently contain dripstone
cements, stalactites and stalagmites.
There is a clear bias in the literature toward studies of
Pleistocene and younger speleothem materials. In part this
reflects their use in palaeoclimate reconstructions but a
lack of reports of Precambrian speleothem is conspicuous.
Reasons may include erosion of karst features at uncon-
formities and the lack of a high pCO2 soil in the Proterozoic
(see Brasier, 2011). In modern karst settings, groundwaters
which have filtered through biogenic CO2-rich soils dissolve
carbonate bedrock, which they re-precipitate as speleothem
on degassing in lower pCO2 cave atmospheres (e.g. Frisia
and Borsato 2010). Many modern speleothem cements are
thus found in caves below thick soils. However, there are
rare reported cases of Proterozoic karst dissolution and
speleothem precipitation, although reported karst surfaces
by far outnumber reported speleothem deposits (Brasier,
2011). For example, Skotnicki and Knauth (2007) reported
silicified flowstone and stalactite fabrics in palaeokarst
associated with the middle Proterozoic Mescal Limestone
of Arizona, USA, believed to be >1.1 Ga in age.
Tufa Facies Models
Pedley (1990), Pedley et al. (2003) and Pentecost and Viles
(1994) classified Quaternary tufas into groups according to
their environmental setting. One key feature of these
environments is the presence and influence of macrophytes:
a feature which would be conspicuously absent in the
Palaeoproterozoic and probably allows these ambient tem-
perature models to be narrowed down from five (Pedley
1990) to three settings. However, the physico-chemical pro-
cesses involved in carbonate precipitation together with the
common presence of cyanobacteria mean that some features
of Quaternary models are relevant to the Palaeoproterozoic.
It should be noted that many features are common to several
facies models, meaning that environmental interpretations
can rarely be made from single hand specimens without
stratigraphic context.
Perched Springline and Cascade SettingsPerched springline and cascade (waterfall) deposits both form
in hillslope settings where they have a low preservation
potential except as clasts in alluvial fan and fluvial
conglomerates (e.g. Pedley 1990). Because of this and the
probability that their constituents and modes of formation
were quite similar in the Proterozoic, they are treated together
here (Fig. 7.151a). They are known but rare in hydrothermal
systems (Guo and Riding 1998). In modern settings, low-
magnesian calcite precipitates around spring orifices on
slopes where high pCO2 calcite supersaturated groundwaters
emerge and degas (Andrews et al. 1997; Pedley et al. 2003).
Degassing of CO2 from groundwaters (and thus calcite pre-
cipitation) around springsmight have been less voluminous in
a Palaeoproterozoic world with higher atmospheric CO2
levels (see Brasier, 2011). Although high atmospheric CO2
levels might have allowed dissolution of carbonate bedrock
and carbonic acid weathering of silicates, a lower pCO2
atmosphere than that of the solution is required for such
degassing to occur. So in the presumed absence of organic
carbon-rich soils (Brasier, 2011) there may not have been a
sufficient gradient in CO2 concentration between groundwa-
ter and atmosphere for this carbonate precipitation mecha-
nism to work as effectively as it has since the Devonian.
Perched springline deposits have flat swampy tops and
steep proximal cascade fronts. In many modern cases,
mosses play a critical role as a substrate for calcification
(see Pedley 1990), and in their absence, one might expect
more ‘speleothem-like’ flowstone deposits over waterfalls.
Cyanobacteria are known to form tongue shaped mounds
of a few metres height and width in front of cascades, as seen
at Goredale Scar in Yorkshire, UK (Andrews and Brasier
2005). Such ‘tufa stromatolites’ (Riding 1991) could also
have been an important feature of the Palaeoproterozoic
terrestrial world (Fig. 7.151a). Fluvial incision in these
settings causes progressive karstification, leading to devel-
opment of speleothem-like cements in vugs and cavities of
older deposits not penetrated by light (Pedley 1990). Other
diagenetic processes in these settings might include post-
depositional crystal growth (e.g. Love and Chafetz 1988).
Fluviatile SettingsAspects of the braided fluviatile model of Pedley (1990) are
applicable to the Proterozoic (Fig. 7.151b). The barrage tufa
system of Pedley (1990) is more applicable to a world
1436 A.T. Brasier et al.
colonised by macrophytes, with meandering rather than
braided fluvial systems (e.g. Cotter 1978; Davies and
Gibling 2010). Further, the obstacles which cause barrage
formation are often logs or larger plants. Braided fluviatile
tufas are recognised by the presence of oncoids in fluvial
conglomerates. The channel deposits, in which many Pleis-
tocene oncoids are found, usually span metres in width and
only centimetres in depth. Oncoids are sub-spherical and
1–5 cm in diameter (Pedley 1990). Tufa intraclasts are
found as a matrix where oncoid conglomerates are clast
supported. Stromatolites develop on stable substrates in
channels and channel margins, aligned with long axes paral-
lel to the direction of flow (Fig. 7.151b).
Lacustrine and Paludal SettingsModern lacustrine environments (and presumably their
Palaeoproterozoic equivalents) exhibit stromatolite and
oncoid build-ups initiated on hard, stable substrates in
shallow-water marginal areas (Fig. 7.151c; see also Pedley
1990). Precipitation of carbonate is due to physico-chemical
degassing due to turbulence combined with photosynthetic
uptake of CO2 by cyanobacteria (e.g. Merz 1992). It has
been suggested (Riding 1979; Eggleston and Dean 1976)
that ‘phytoherm’ height corresponds with water depth.
Overhangs may develop where stromatolites reach the air-
water contact (Fig. 7.151c), leading to the possibility of later
dripstone fabrics. Recent ooid sand shoals of Lake
Tanganyika are most commonly found on shallow water
platforms in water depths of 0–4 m (Cohen and Thouin
1987), although Soreghan and Cohen (1996) describe the
pervasive presence of ooids with both tangential and radial
fabrics from shoreface depths down to 38 m below lake
level. As noted by Cohen and Thouin (1987), this depth
corresponds to the depth of waters agitated by waves on a
daily basis. Holocene tufa towers (Fig. 7.152) in Mono Lake,
California, are not restricted to lake margins but found where
fresh meteoric waters mixed with saline lake waters (Ford
and Pedley 1996). Such towers may be of bacterial or
cyanobacterial construction. Paludal (swampy) deposits are
not easily distinguishable from marginal lacustrine settings
in the rock record. Further, a lack of fauna and flora in the
Proterozoic hinders the recognition of paludal settings.
Hydrothermal Travertine Facies Models
Whilst modern ambient temperature deposits usually consist
of low-magnesian calcite, ‘hydrothermal’ precipitates also
include aragonite (Guo and Riding 1994; Pentecost 1990;
Fouke et al. 2000), barite (e.g. Bonny and Jones 2008) and
silica (e.g. Guidry and Chafetz 2003). Chafetz and Folk
(1984) suggest that ‘biologically harsh’ stagnant bodies of
water fed by sulphur-rich springs produce the greatest
amount of bacterially-mediated (or abiological?) travertine,
whereas situations which allow agitation and degassing of
H2S are more dominated by the presence of higher plants. In
a Proterozoic world, such contrasts between proximal and
distal fabrics might have been less pronounced.
Cooling and consequent degassing of hot waters on their
emergence from the ground causes minerals to precipitate in
hydrothermal settings (Fouke et al. 2000) and unlike ambi-
ent temperature springs, the rapid degassing and precipita-
tion causes spring vents to ‘self seal’ and migrate through
time (e.g. Fouke et al. 2000). Bacteria are able to tolerate
higher temperatures than cyanobacteria and may play a
significant role in precipitation (Chafetz and Folk 1984;
Guo et al. 1996; Fouke et al. 2000).
Terraced Mound SettingsProterozoic terraced mounds would probably have been
quite similar to those of today, which can reach substantial
sizes: Mammoth Hot Spring in Yellowstone National Park
measures 1.4 � 4 km (Pentecost 1990). The modern
environments of Angel Terrace Hot Spring in Yellowstone
National Park were described by Fouke et al. (2000) and
comply with the ‘Terraced mound’ model of Chafetz and
Folk (1984; see Fig. 7.151e). Hot springwater emerges from
irregularly shaped, 5-cm-diameter ‘vents’, where it rapidly
degasses both CO2 and H2S. Waters then flow rapidly across
gently sloped apron and channel facies before decelerating
and accumulating in ‘ponds’. These ‘ponds’ (Fouke et al.
2000) or ‘miniature lakes’ (Chafetz and Folk 1984) have
raised rims up to 10 cm high due to mineral precipitation.
Beyond these ponds, the Angel Terrace deposit has a steep
‘proximal slope’ covered in ‘microteracettes’, over which
waters cascade before reaching a ‘distal slope’ (Fouke et al.
2000). ‘Sinter terraces’ of Pastos Grandes, Bolivia (Risacher
and Eugster 1979) also fit this model, being rimmed pools
(often containing pisoliths) of several centimetres to metres
diameter, separated by elevation differences of a few
centimetres. Deposits of Rapolano Serre are described by
Chafetz and Folk (1984) and Guo et al. (1996). Here, ter-
raced mounds form on the flat valley floor, adjacent to
sloping fan complexes.
Fissure Ridge SettingsFissure ridges (Fig. 7.151f) are described as being 75–100 m
long, 10–15 m high and 10–20 m wide at their bases
(Chafetz and Folk 1984). A central fissure extends along
the length of the ridge, through which waters escape and
degas, precipitating minerals. Although fissure ridges have
been discriminated from other environmental settings on the
9 7.9 Terrestrial Environments 1437
basis of their morphology (Chafetz and Folk 1984; Pente-
cost and Viles 1994) and not water temperature, the pres-
ence of a fault conduit for water flow increases the
likelihood of hydrothermal fluid involvement. Two- to
five-centimetre-thick laminar crusts make up the ridges.
Microterraces develop on the steep outer surfaces of these
ridges. At Terme San Giovanni, Italy, Guo et al. (1996)
reported a Holocene calcite deposit with a fissure ridge
from which hot waters emanated before flowing over
200 m of shallow sloped drainage channel facies and a
further 150 m of steeper sloped cascades before reaching
the valley floor.
Hydrothermal Lake SettingsHydrothermal travertines are also found in lakes
(Fig. 7.151d). H2S-rich waters bubble up through
subaqeuous vents precipitating hydrothermal travertines at
Bagni di Tivoli in Italy. Active deposits have not been
extensively studied because of the prevalence of noxious
gases (Chafetz and Folk 1984). Nevertheless, Chafetz and
Folk (1984) describe purple bacteria amongst reeds in the
margins of the lake, with extensive paludal tufa in the
surrounding area. Quarries of fossil hydrothermal lacustrine
travertines here are hundreds of metres in length and tens of
metres in height. Ten-metre-thick packages consist of thin
layers of crudely laminated to massive carbonate mud, plus
layers of upward-branching “forest like” crystal shrubs,
which can be traced for tens of metres. Less continuous
regions of radiating calcite ray crystals are intercalated
with the muds and shrubs. Stratigraphic relationships and
rare desiccation cracks lead Chafetz and Folk (1984) to
suggest water depths of <1 m. Packages are separated by
gently sloping erosional surfaces exhibiting thin palaeosols
and evidence of karstification (Chafetz and Folk 1984).
Abundant geopetal sediment infills are found in these
travertines, and some clastic layers are interpreted to be a
result of storm floods (Chafetz and Folk 1984).
Petrography
Abiological fabrics and those produced by cyanobacteria
and bacteria in the Pleistocene may resemble those of the
Palaeoproterozoic (see Brasier, 2011). In many cases
cyanobacteria provide a framework for Pleistocene tufas,
and it has been demonstrated that the extracellular polymeric
substances (EPS) the cyanobacteria produce assist calcite
crystal nucleation (for example Rogerson et al. 2008).
Freytet and Verrecchia (1998) provide a useful compilation
of modern cyanobacteria and their associated crystal
morphologies.
Pervasive columnar calcite spar in Pleistocene tufas is
widely believed to be of early diagenetic origin, having
replaced primary micrite through Ostwald ripening (for
example Love and Chafetz 1988; Janssen et al. 1999).
Such spar is hard to distinguish from abiogenic speleothem.
In some instances columnar spar is clearly of primary and
possibly biogenic origin, having formed synchronously with
adjacent micrite (Brasier et al. 2011). Observations of
Freytet and Verrecchia (1998) support the contention that
spar can be a primary precipitate, though it is possible
(particularly given any biological involvement) that the
amount of primary versus secondary spar has changed
through time. Riding (2008) grouped Precambrian
stromatolites into “Fine-grained Crust”, believed to repre-
sent lithified microbial mat, and “Sparry Crust”, suggested to
be essentially abiogenic. In addition to these “Fine-grained
Crust” and “Sparry Crust” end-members, Riding (2008)
recognised “Hybrid Crusts” formed by a combination of
microbial growth and abiogenic precipitation (see also
Chap. 7.8.2). Riding (2008) envisaged a transition from a
world dominated by abiogenic “Sparry Crust” stromatolites
in the Archaean to one colonised by “Fine-grained Crusts” in
the Neoproterozoic, via a significant interval of “Hybrid
Crust” deposition during the Palaeo- and Mesoproterozoic.
Bacteria may be responsible for crystal shrubs (distinct
from calcified cyanobacterial shrubs) in hydrothermal
travertines as advocated by Chafetz and Folk (1984), Guo
and Riding (1994) and Guo et al. (1996) although they are
interpreted as abiological by Pentecost (1990). Shrubs are
found in harsh biological conditions in shallow pools such as
lakes fed by H2S-rich springs (Chafetz and Folk 1984) or
shallow pools in local pockets in slope systems (Guo et al.
1996). Aragonite shrubs may also form the dams around
terracette pools (Fig. 7.151e; see also Pentecost 1990) but
they are thickest in depressions and on flat surfaces (Guo and
Riding 1998).
Hydrothermal deposits also include fibrous crystalline
crusts composed of coarse, elongate “ray crystals” (Chafetz
and Folk 1984) or “feather crystals” (Guo and Riding 1992,
1998) oriented perpendicular to the depositional surface.
Bunches of these crystals may form conical radiating
patterns, and individual crystals often show internal lamina-
tion perpendicular to the direction of growth (Chafetz and
Folk 1984).
Spherical to irregularly shaped pisoids are common in
hydrothermal settings (Guo and Riding 1998). Concentri-
cally laminated pisoids form in splashing and turbulent
water, whereas radial shrub pisoids have similar fabrics
and form in similar settings to the crystal shrubs described
above (Guo and Riding 1998). Pisoliths of the Pastos
Grandes, Bolivia, range from a few mm to 20 cm in size
1438 A.T. Brasier et al.
(Risacher and Eugster 1979), with size showing some corre-
lation with water depth. Pisoliths may become cemented
together to form ‘cauliflower’ textures.
Paper-thin rafts (Guo and Riding 1998) are composed of
crystals precipitated rapidly (Chafetz et al. 1991) at the
surface of the water. They sometimes accumulate on the
floors of stagnant hot water pools (Guo and Riding 1998)
and may thicken through crystal growth. Rafts are often flat
on the upper surface since crystals grow downward into the
water (Chafetz et al. 1991). The undersides of rafts may
include mm- to cm-size calcitic and aragonitic coated
bubbles, preserved through rapid mineral precipitation,
which occurs “within minutes” (Chafetz et al. 1991; Guo
and Riding 1998). The bubbles are believed to result from
photosynthesis by microorganisms (Chafetz et al. 1991).
Geochemistry
Tufa stable isotope geochemistry (reviewed in Andrews and
Brasier 2005 and Andrews 2006) is increasingly used in
Quaternary palaeoenvironmental studies, and has also been
successfully employed in Permian (Szulc and Cwizewicz
1989) and Cretaceous (Nehza et al. 2009) cases. Such an
approach could and should be applied to Proterozoic
deposits. However, studies of the geochemistry of modern
systems are generally restricted to calcite and aragonite
deposits, whereas Melezhik and Fallick (2001) contend
that the Palaeoproterozoic Kuetsj€arvi travertines may be
primary dolomite. A possible analogue could be modern
microbial dolomite reported from ephemeral lakes of South
Australia (see references in Wacey et al. 2007), said to be
precipitated with the help of sulphate-reducing bacteria. The
geochemistry of Proterozoic non-marine carbonates thus
demands investigation.
Andrews et al. (1997) demonstrated that modern fresh-
water ambient temperature microbial carbonates often
accurately record the oxygen isotopic composition of their
depositing waters, modified by temperature and evapora-
tion. High-resolution studies of seasonally banded tufa
oxygen isotopes can provide information on groundwater
composition and stability as well as potentially revealing
seasonal changes in temperatures of carbonate precipitation
(for example Chafetz et al. 1991; Matsuoka et al. 2001;
Ihlenfeld et al. 2003; O’Brien et al. 2006; Brasier et al.
2010). Such a high-resolution petrographic and geochemi-
cal approach has not been fully applied to Proterozoic
deposits, although Awramik and Buchheim (2009)
attempted a high-resolution stable isotope transect through
Neoarchaean lacustrine stromatolite laminae, and Melezhik
and Fallick (2001) published a lower resolution stable
isotope transect (up to 5 mm spacing between samples)
of a specimen of the Palaeoproterozoic Kuetsj€arvi Sedi-
mentary Formation travertine. Unfortunately, carbonate
oxygen isotopic compositions are readily re-equilibrated
with meteoric waters, diagenetic and metamorphic fluids,
and records of Proterozoic age will always be questionable.
Nevertheless, original d18O values are sometimes preserved
and it has been suggested that this is the case with the
Kuetsj€arvi Sedimentary Formation travertines (Melezhik
and Fallick 2001). Mg/Ca molar ratios of calcitic tufas
can also relate to stream temperature (Ihlenfeld et al.
2003) although groundwater residence times (particularly
in regions with dolomite bedrock) have a significant input
(Fairchild et al. 2000; Garnett et al. 2004; Brasier et al.
2010). No palaeoenvironmental reconstructions based on
trace element analyses of dolomitic tufa or travertine
have been reported.
Tufas may also record the carbon isotopic composition of
dissolved inorganic carbon (see Andrews 2006 and
references therein). In modern stream systems, tufa carbon-
ate d13C reflects relative contributions of 12C-rich soil
organic matter versus more 13C rich marine carbonate
(Andrews et al. 1993; Andrews 2006). Proterozoic tufas
are thus likely to have heavier d13C than those of today.
Modern hydrothermal waters generally have a deep
source and so long residence times in marine carbonate
bedrock, leading to more 13C-rich compositions than
groundwaters derived directly from rainwater (Pentecost
and Viles 1994; Andrews 2006; Arenas et al. 2000). How-
ever, limestones and dolostones are probably more common
as bedrock today than they would have been in the
Palaeoproterozoic (Grotzinger 1989), and hence volcanic
sources of Ca ions may have been relatively more important.
Melezhik and Fallick (2001) speculated that the Palaeopro-
terozoic Kuetsj€arvi travertines with isotopically negative
d13C could have a volcanic source of CO2, and thus be
precipitated from hydrothermal waters.
Other possibilities for d13C depletion in Proterozoic
travertines and tufas could include incorporation of 12C
from ambient temperature groundwaters flowing through
volcanic or organic-carbon-rich rocks at the time of deposi-
tion; involvement of methane-rich groundwaters during car-
bonate precipitation or recrystallization; incorporation of
biotically respired CO2 from microbial aerobes in
cryptobiotic soils; and post-Ordovician recrystallization
involving 12CO2 from the soil zone.
Conversely, preferential degassing of 12CO2 from waters
through ‘prior calcite precipitation’ causes progressive
downstream enrichment in carbonate 13C in both hydrother-
mal and ambient temperature systems. For example, Pente-
cost and Spiro (1990) measured a 2.6 ‰ downstream
9 7.9 Terrestrial Environments 1439
increase in the d13C of dissolved inorganic carbon over a
distance of 531 m in a British tufa depositing stream. Similar
effects have been noted by Chafetz and Lawrence (1994),
Guo et al. (1996) and Fouke et al. (2000) amongst many
others. An important process in isolated stagnant pools is
local photosynthetic uptake of 12CO2 by cyanobacteria,
which has been reported to cause d13C enrichment by as
much as 8 ‰ in a modern hydrothermal fissure ridge setting
(Guo et al. 1996), but is relatively insignificant away from
cyanobacterial colonies and in areas of rapid water flow (see,
for example, Pentecost and Spiro 1990; Guo et al. 1996;
Andrews and Brasier 2005; Andrews 2006).
Examples of Precambrian Tufas and Travertines
Tumbiana Formation, AustraliaStromatolitic carbonates belonging to the Fortescue Group,
Pilbara Craton, Australia, are dated between 2.78 and
2.63 Ga and are of possible lacustrine origin (Bolhar and
van Kranendonk 2007; Awramik and Buchheim 2009). The
presence of oolites and sedimentary structures including
symmetrical ripples (Awramik and Buchheim 2009) favour
a lacustrine or shallow marine origin over a paludal setting.
These laminated, potentially microbial (see Chap. 7.8.2)
build-ups could be classified as lacustrine tufa (or travertine)
stromatolites (e.g. Freytet and Plet 1996; Andrews and
Brasier 2005; Takashima and Kano 2008; see Fig. 7.151c),
if the temperature of deposition could be deduced. Use of the
term ‘tufa stromatolite’ also assumes formation by autoch-
thonous carbonate precipitation on the outsides of
cyanobacterial sheaths (see Riding 1991, 2000), rather than
agglutination of clastic particles. Awramik and Buchheim
(2009) describe couplets of poorly preserved light and dark
coloured laminae, which vary in thickness along their length.
The thicker laminae, are light coloured and composed of
spar, whereas the dark laminae are composed of “microspar
with silt to fine sand-size volcaniclastic grains in calcareous
cement”. Stromatolite carbonate d13C values are in the
region of 0 ‰ PDB.
Kuetsj€arvi Sedimentary Formation, theFennoscandian Shield, RussiaThe Kuetsj€arvi Sedimentary Formation lies on subaerially
erupted amygdaloidal basaltic andesites of the Ahmalahti
Volcanic Formation and is overlain by Kuetsj€arvi Volcanites
dated at c. 2060 Ma (Melezhik et al. 2007). Probable hydro-
thermal travertines are known from an aggregate quarry and
drillholes. The hydrothermal interpretation of these
travertines seemingly hangs on petrographic arguments and
that negative d13C values (�6 ‰ VPDB) in some samples
could be a result of a volcanic CO2 source (Melezhik and
Fallick 2001). The reported range in d13C for these
travertines is from �6.1 ‰ to +7.7 ‰ VPDB.
Two types of travertine are distinguished by Melezhik
and Fallick (2001). The first (found lower in the stratigra-
phy) are laminated crusts developed on pure carbonate
substrates capped by stromatolitic dolostones
(Fig. 7.153a–c). These form laminar sheets 1–15 cm thick
(Fig. 7.153d) and up to 5 m across with botryoidal upper
surfaces (Fig. 7.153e, f). Separate stages of carbonate pre-
cipitation seen in hand specimens of these laminar crusts can
be distinguished by colour. The lowest (?oldest) fabrics are
‘yellowish micritic dolomite’, overlain by white radiating
dolomite crystals. Couplets of grey and dark grey dolomitic
ray crystal laminae (e.g. Fig. 7.153b) are found above the
white layer, and finally capped by a fibrous dolomite ‘sinter’.
Thin siliceous layers are interpreted by Melezhik and Fallick
(2001) as sinters. Occasional pisolites up to 5 cm in diameter
are also reported.
These sheets are found in association with flat-laminated
dolomitic stromatolites interbedded with sandy allochemical
dolostones, which contain abundant fenestrae, intraclasts of
algal dolostones, and exhibit tepee-related brecciation
(Melezhik and Fallick 2001). The interpreted depositional
environment of the host rocks is a lacustrine coastal plain
with strong evaporative pumping to produce the tepee
structures (Melezhik and Fallick 2001). Because no barrage
or slope morphology was observed, Melezhik and Fallick
(2001) suggested that the carbonate crusts formed on a
horizontal surface, and attributed them to spring, fluvial or
lacustrine origin following Pentecost and Viles (1994).
However, Melezhik et al. (2004) illustrated crusts over a
shallow metre-scale palaeo-slope (reproduced here as
Fig. 7.153d). The described crusts are laminar and not
oncoidal nor found within fluvial sediments, so a fluvial
environment can probably be ruled out.
The second group of ‘travertines’ are small-scale mounds
(heights of 1–10 cm) often clustered together and found
beneath red, laminated siltstones and sandstones
(Fig. 7.151g–j). Kuetsj€arvi Sedimentary Formation mounds
are associated with a 5-m-thick unit of interbedded,
allochemical, micritic dolostones and flat, laminated dolo-
mitic stromatolites with evidence of evaporitic minerals and
desiccation cracks. Mounds seem always oriented perpen-
dicular to bedding in a stalagmitic orientation (Fig. 7.153g),
and are found buried under red sandstone (Fig. 7.151g–i).
Some mounds are truncated and capped by laminar dolo-
mitic cement, which drapes the underlying topography
(Fig. 7.153g). The interior of each mound is composed of
buff yellow, micritic dolomite (Fig. 7.151i, k, l), coated by
layers of white dolomite with a few layers of grey and black
1440 A.T. Brasier et al.
(Fig. 7.151h, i, k–m). A thin veneer of ‘silica sinter’ coats
(and sometimes has replaced original carbonate around) the
outside of the mounds (Fig. 7.151l–p). The buff yellow,
micritic dolomite sometimes forms star shapes
(Fig. 7.153i) and could be interpreted as an early (syn-
depositional) channel-filling cement. The origin of the col-
our lamination is unclear, since black layers have a low
organic carbon content (<0.1 %). Melezhik and Fallick
(2001) suggested that the mounds were associated with the
orifices of hot springs in an arid playa lake setting. They
cited hydrothermal deposits of the Bolivian Altiplano
reported by Risacher and Eugster (1979) as a modern ana-
logue. Interpreted travertine veins (Fig. 7.153q) and possible
‘feeder channels’ (Fig. 7.153r) are compatible with such a
hydrothermal spring orifice origin. Laminar crusts coating
erosion surfaces (Fig. 7.153s) and clasts (Fig. 7.153t) is seen
as evidence of formation in a subaerial or near-surface
environment (Melezhik et al. 2004). However, the Bolivian
hot spring example exhibits rimmed pools compatible with
the terraced mound model of Chafetz and Folk (1984), a
feature that has not been reported in the Fennoscandian case.
Rocknest Formation, Wopmay Orogen, NorthwestCanadaCements of the 1.86 Ga Rocknest Formation exposed in the
Wopmay Orogen were originally reported as “tufa” by
Grotzinger (1986). The environmental interpretation of
Grotzinger (1986) suggests that the cements are of a mar-
ginal marine origin, albeit from the topographically highest
part of the platform. These ‘tufas’ are found above upper
intertidal to supratidal cryptalgalaminites, which contain
fenestrae, desiccation cracks and evaporite pseudomorphs.
The ‘tufas’ themselves are “cement laminae that are contin-
uous, smooth to undulatory or colloform”, often exhibiting
microdigitate stromatolites 1–10 mm wide and 0.1–5 mm
high (see Riding 2008). Layers are sometimes deformed into
tepee structures, and vadose cements include stalactitic bot-
ryoidal aragonite. These ‘tufas’ are said to have formed
when marine waters of a low siliciclastic content were
blown or washed over tidal flats, leading to precipitation of
aragonite by evaporation with some possible microbial
involvement. Such carbonate cements are now better
referred to as “seafloor cements” (Grotzinger 2010 pers.
comm.) or “seafloor crusts” (Grotzinger and Knoll 1995)
given that waters involved were marine, not meteoric.
Murky Formation, Athapuscow Aulacogen,Northwest Canada1.3–1.87 Ga ephemeral lake deposits associated with the
distal end of an alluvial fan of the Murky Formation contain
calcareous stromatolites within siliciclastic sediments
(Hoffman 1976). If some (perhaps unverifiable) assumptions
are made about conditions during deposition, these
stromatolites could be called tufa stromatolites (e.g. Riding
1991, 2000; Andrews and Brasier 2005). The stromatolites
form “small isolated colonies coalesced to form an extensive
basal sheet”. Individual stromatolite heads achieve heights
of up to 2 m above this basal sheet (Hoffman 1976).
Kunwak Formation, Northern CanadaDiscrete interbeds of encrusting and void-filling calcite
cements are briefly described and interpreted as hydrother-
mal travertine by Rainbird et al. (2006). The travertine was
Pb-Pb dated at 1.79 Ga by the same authors. The
travertines are associated with volcanic rocks and comprise
three types of cement: Type A is “laminar to colloform
calcite with Fe-Mn banding”, and this appears to be the
youngest generation, draping underlying fractured laminae
with voids filled by ‘Type B’ blocky void-filling spar.
Fabric ‘Type C’ of Rainbird et al. (2006) exhibits swallow-
tail shapes, suggesting calcite replacement after gypsum.
The prior presence of sulphate crystals is at least consistent
with a hydrothermal setting, but the temperature at which
these deposits were formed has not yet been conclusively
proven.
Mescal Limestone, Arizona, USASilicified flowstones of the Mescal Limestone, Arizona,
USA, are described by Skotnicki and Knauth (2007). These
are believed to be older than 1.1 Ga intrusions, which
“locally overprint the rocks with a coarse, recrystallized
texture”. In addition to outcrop exposures, clasts of
‘speleothem’ are found in quartzites overlying a palaeokarst
surface. Chert “fingers” perpendicular to bedding and paral-
lel to each other are 5–10 mm wide and up to 10 cm long.
The silica is interpreted to be secondary after carbonate on
the basis of rhomb-shaped cavities. Tubular cavities between
2 and 8 mm diameter and cones of a few cm width and
<10 cm length with fluted outer surfaces are also described
by Skotnicki and Knauth (2007), who interpret botryoidal
chert as secondary replacement of sulphates. A layer of
silicified coniform stromatolites with red and grey laminae
is found near the top of the palaeokarst. In places this layer is
brecciated.
Copper Harbor Conglomerate, Michigan, USAElmore (1983) published a study of non-marine stromatolites
of the Copper Harbor Conglomerate (dated at 1.087 Ga by
Davis and Paces 1990) that seem to be a good example of the
braided fluviatile facies model (Fig. 7.153b). Carbonates
9 7.9 Terrestrial Environments 1441
include ‘laminated cryptalgal’ stromatolites in coarse-
grained alluvial fan facies plus oolite-oncolite beds in
braided stream deposits and intraclasts of stromatolite and
oolite fragments (Elmore 1983). Stromatolites are ‘laterally
linked hemispheroids draped over cobbles’, fitting the
description in Pedley (1990) of stromatolites developed on
stable substrates in channel and channel margin facies.
Kanuyak Formation, Elu Basin, Northwest CanadaThe Mesoproterozoic Kanuyak Formation was deposited in
karst topography created by erosion of the underlying Parry
Bay Formation (Pelechaty et al. 1991). Possible lacustrine
tufas include ‘lithofacies 1’ dolostones, described as 60- to
220-cm-thick, white-pink coloured beds with sharp or
gradational bases and sharp upper contacts with overlying
sediments. The dolostones are composed of microcrystal-
line dolomite, minor quartz sand and chert pebbles, and
contain horizontally elongate fenestrae (Pelechaty and
James 1991) filled with ‘late-stage’ ferroan dolospar
cement. Pelechaty et al. (1991) suggest that the dolostones
formed by carbonate precipitation in a lacustrine environ-
ment based on their ‘isolated, random distribution’. Car-
bonate carbon isotope values range from �1.22 ‰ to
+0.84 ‰ PDB. Speleothem cements are mentioned by
Pelechaty and James (1991) but not described in great
detail. Similarly, the Kanuyak Formation contains braided
fluvial deposits (Pelechaty et al. 1991), but no associated
carbonates are mentioned.
Synthesis of Palaeoproterozoic Tufas andTravertines: What Is Known and What toLook For
In summary, the terms tufa and travertine can imply ambient
and hydrothermal temperatures of deposition, respectively.
Brasier (2011) suggested using terminology based on envi-
ronmental setting (deduced from stratigraphy), such as
‘spring carbonate’ or ‘lacustrine carbonate’, but there are
times when it may be desirable to use the term tufa or
travertine. So how can tufa be distinguished from travertine
in the ancient rock record? One guide may be the deposi-
tional setting in which the non-marine carbonate is found:
carbonates in braided fluvial systems are arguably likely to
be tufas, whereas carbonates from fissure ridges associated
with faults might be hydrothermal. However, application of
such criteria is necessarily uniformitarian, and does not
account for periods in Earth history when fluviatile
carbonates may actually have been deposited from waters
of high temperatures, such as might be expected on a hot,
young Earth (e.g. Walter 1996; Brasier, 2011). Relatively
few Palaeoproterozoic tufas and travertines have been
reported (see also Brasier, 2011). Along with the Kuetsj€arviSedimentary Formation travertines (Melezhik and Fallick
2001) are the younger Kunwak Formation travertines (Rain-
bird et al. 2006). The latter deposit has only been briefly
described, yet it is notable that both the Fennoscandian and
North American examples are closely associated with volca-
nic units, which supports but does not prove the hypotheses
that hot waters were involved in their formation.
Petrographic features might be diagnostic of depositional
temperature. The Kuetsj€arvi Sedimentary Formation traver-
tine includes ray crystals and pisoliths, which, although not
exclusively of hydrothermal origin, might be taken as
supporting evidence of the ‘travertine’ interpretation. More
diagnostic of travertines might be coated bubbles, paper-thin
rafts and crystal shrubs, none of which have yet been reported
from the Palaeoproterozoic. Coated bubbles, for example, are
known from the ancient rock record, including examples
from Jurassic siliceous hot springs (Guido and Campbell
2009). However, early diagenetic processes like ‘aggrading
neomorphism’ mean that not all petrographic features from
the time of deposition are necessarily preserved.
The biology of the system could help to distinguish ambi-
ent temperature deposits from hydrothermal ones. For exam-
ple, stromatolites built by cyanobacteria are arguably likely
to have formed at low temperatures, although it is notable
that some modern bacteria thrive at elevated temperatures.
Such bacteria may even be partly responsible for some
hydrothermal travertine build-ups. Geochemical data may
help to distinguish travertine from tufa, such as the sugges-
tion of Melezhik and Fallick (2001) that relatively negative
d13C values in the Palaeoproterozoic may indicate volcanic
CO2 presence and thus a hydrothermal (travertine) origin of
the carbonate. If correct, the implication is that unaltered
Palaeoproterozoic tufas derived from atmospheric CO2
and marine carbonate bedrock should normally have d13Cvalues closer to 0 ‰ (VPDB), whereas travertines will have
d13C values around �7 ‰ (VPDB). However, negative
Palaeoproterozoic non-marine carbonate d13C might also
reflect syn-depositional (or later post-depositional) equili-
bration with 12C-rich fluids which have passed through vol-
canic rocks or organic-carbon rich sediments. It is well
known that tufa/travertine carbonate d18O can record the
temperature of the depositing waters (e.g. Brasier et al.
2010), but oxygen isotopes are readily altered during
diagenesis, and hence in ancient cases their interpretation
is rarely simple or incontrovertible. One might also look for
indicators of the presence of high temperature minerals like
sulphides associated with hydrothermal systems. Thus a
1442 A.T. Brasier et al.
Fig. 7.149 Vertical section through dolomite travertine crust. The
crust starts with dismembered and displaced pink travertine bands;
voids filled with white dolospar. This is followed by alternating palepink and pink travertine bands, which, in turn, pass into pale beige and
pale pink travertine with a clotted microfabric. This part of the crust is
intersected by travertine veins composed of white, fibrous dolomite.
The partially dissolved, uneven upper surface is veneered with grey
travertine bands. The c. 2060 Ma Kuetsj€arvi Sedimentary Formation
from the Pechenga Belt. Width of image is 8 cm (Photograph by Victor
Melezhik)
9 7.9 Terrestrial Environments 1443
Fig. 7.150 Section cutting travertine mounds’ heads parallel to the
bedding surface, exhibiting a series of concentric white, pale grey anddark grey bands of fibrous dolomite with yellow dolomicrite having a
clotted microfabric in the centre. The outer dolomite rim is veneered by
silica sinter, which exhibits a replacive relationship to the underlying
travertine layer. The silica-veneered travertine mound is emplaced in
red sandstone. From the c. 2060 Ma Kuetsj€arvi Sedimentary Formation
from the Pechenga Belt. Width of image is 3 cm (Photograph by Victor
Melezhik)
1444 A.T. Brasier et al.
Fig. 7.151 Tufa and travertine facies models which may apply to
Palaeoproterozoic deposits. (a) Perched springline and cascade facies
(Adapted from Pedley et al. (2003) and modified from Brasier (2011)).
(b) Braided fluvial model adapted from Pedley (1990). (c) Lacustrine
model adapted from Pedley (1990). (d) Lacustrine hydrothermal model
based on information in Chafetz and Folk (1984). (e) Hydrothermal
spring mound model based on Mammoth Hot Springs, USA (left) andin cross-section modified from Fouke et al. (2000) shown on right. (f)Hydrothermal fissure ridge model adapted from Guo et al. (1996)
9 7.9 Terrestrial Environments 1445
hydrothermal lake setting seems at least consistent with the
data presented for the Palaeoproterozoic Kuetsj€arvi Sedi-mentary Formation.
Regarding the facies models presented here (Fig. 7.151),
there are presently no known Proterozoic hydrothermal ter-
raced mounds or fissure ridges. This might reflect their
preservation potential (Brasier, 2011) since marine hydro-
thermal deposits of similar age or older are well known
(e.g. Lindsay et al. 2005), and Walter (1996) suggests that
higher heat flow on the young Earth should have produced
more hydrothermal precipitates than found in the present
day. Lacustrine tufas might extend as far back in time
as the Neoarchaean Tumbiana Formation and might also
include stromatolites of the Murky Formation. Distin-
guishing marine from lacustrine and proving that stromato-
lite accumulation was accomplished by authigenic
precipitation are difficult. Nevertheless, the observation of
stromatolites and ooid shoals in the Tumbiana Formation
(Awramik and Buchheim 2009) fits the Proterozoic lacus-
trine tufa model outlined here (Fig. 7.151). Similarly,
Mesoproterozoic braided fluvial sediments with oncoids
and stromatolites of the Copper Harbor Conglomerate
(Elmore 1983) seem a good fit with the braided fluviatile
model as adapted here from Pedley (1990). It seems hopeful
that older, Palaeoproterozoic, lacustrine and fluviatile tufa
examples will be found. Perched springline and cascade
tufas form today by CO2 degassing in environments of low
preservation potential, and their remains might only be
found as clasts of ‘tufa stromatolite’ in conglomerates.
High atmospheric CO2 levels and lack of thick organic-rich
soil cover in the Palaeoproterozoic would not favour CO2
degassing (see Brasier, 2011). Thus it is perhaps unsurpris-
ing that perched springline and cascade deposits have not yet
been reported from these ancient times.
Potential for Future Study and Implicatons ofthe FAR-DEEP Cores
The relative lack of study of Proterozoic travertines leaves
many questions to be answered. The first is whether the
apparent lack of deposits (and speleothem in particular) of
Fig. 7.152 Holocene lacustrine tufa mounds of Mono Lake, California, found where fresh and saline waters mix. Towers range from a few
centimetres to a few metres in height (Photograph by Victor Melezhik)
1446 A.T. Brasier et al.
Fig. 7.153 Pechenga travertines from the Palaeoproterozoic c.
2.06 Ga Kuetsj€arvi Sedimentary Formation, the Pechenga Belt. (a)
Stack of thin travertine crusts beneath pale pink, bedded dolostone,
which is, in turn covered with white and pink travertine. (b) Greylaminated travertine overgrowing a pale beige clast in the form of a
pisolith above unevenly eroded/dissolved travertine crust. (c) Vertical
section through dolomite travertine crust composed of pale beige and
pale pink travertine with a clotted microfabric, and travertine veins of
white fibrous dolomite.
9 7.9 Terrestrial Environments 1447
Fig. 7.153 (continued) (d) Cross-section through eroded dolostone
with an erosion surface extending from left side downwards to lower
right corner, which is covered with pale grey, fibrous dolomite traver-
tine crust followed by thicker crust of beige and pale pink, finely
laminated, travertine. The remaining part of uneven palaeorelief, to
the right of the “travertine slope”, is filled with dolomite-cemented
sandstone, which are all, in turn, sealed by a sub-horizontally lying
dolarenite bed with a wavy upper surface veneered by travertine crust.
Pencil for scale is 15 cm long. (e) Bedding surface of dolomite traver-
tine crust with concentric banding and irregularly swirled layers. (f)
Clusters of small spheroidal structures (botryoids) on the surface of
dolomite travertine crust.
1448 A.T. Brasier et al.
Fig. 7.153 (continued) (g) Cross-section of dolomite travertine
mounds buried under bedded, red, clayey sandstone. The rippled sand-
stone surface is covered conformably by travertine crust. The mound
interior (yellowish micritic dolomite) is covered by a series of white
bands of radiating dolomite crystals. The white dolomite is conform-
ably overlain by a sequence of grey and dark grey bands of ‘ray
crystals’. The uppermost dark grey dolomite band is veneered by pale
grey, finely-crystalline silica sinter. (h) Cross-section of dolomite trav-
ertine crust and small mounds (some detached and displaced) buried
conformably under bedded, red, clayey sandstone. (i) Bedding-parallel
section through travertine mound heads emplaced into red clayey
sandstone. The yellowish, micritic dolomite with a clotted fabric
(shrub-like travertine) appears as starfish-like “channels”, which are
conformably overgrown by concentric alternating, grey and dark greybands of radiating dolomite crystals. The outer dolomite rim is
veneered by thin silica sinter. (j) Thin section photomicrograph of 4-
cm-long cross-section through travertine mound: (1) micritic dolomite
with a clotted fabric and voids filled with dolomite spar, (2) dark layer
of clumps and clots of radiating dolomite crystals, (3) sequence of
alternating white and dark grey bands of ‘ray crystals’, (4) dissolution
surface truncating dark and light couplets, (5) composite fibrous dolo-
mite crystals with compromise boundaries, (6) the uppermost layer of
micritic dolomite, (7) silica sinter, (8) bedded sandstone.
9 7.9 Terrestrial Environments 1449
Fig. 7.153 (continued) (k) Section through travertine crust, starting
with beige dolomite with a clotted microstructure, which is covered by
a white layer of radiating dolomite crystals followed by a sequence of
pale grey and grey couplets with ‘ray crystals’ and fibrous dolomite
composite crystals. (l) Section of the travertine crust parallel to the
bedding surface, exhibiting concentric banding of alternating grey and
dark grey couplets of radiating dolomite crystals; the underlying layers
of pale yellow micritic dolomite with a clotted microfabric appear as
starfish-like “channels”; the uppermost dark grey dolomite layer is
veneered by silica sinter. (m) Concentric banding of alternating grey
and dark grey couplets of radiating dolomite crystals on the section of
the travertine crust parallel to the bedding surface; the uppermost dark
grey dolomite layer of composite fibrous dolomite crystals is veneered
and partially replaced by a white silica sinter. (n, o) Vertical sections
through the upper part of the travertine crust with trough-like, silica-
veneered dissolution cavities of surface origin. The cavities cut through
the laminated travertine down into crust (pale yellow dolomicrite) with
a clotted microfabric. Three small, silica-filled dissolution pipes occur
to the left of the trough in “o”. (p) Photomicrograph in transmitted non-
polarised light showing cross-section through travertine (fibrous dolo-
mite) with three silica sinters (white), which are overlain by haematite-
rich mudstone (black) followed by sandstone. The lower sinter veneersreplacively the underlying travertine layer, and the upper sinters shows
black, mud-filled desiccation cracks, whereas the middle sinter is
dismembered and emplaced in black, haematite-rich mudstone.
1450 A.T. Brasier et al.
such great age is a result of lack of preservation, lack
of deposition or lack of recognition (Brasier, 2011).
Secondly, there are many obvious differences between
the modern terrestrial world and that of the Proterozoic.
The opportunity to study travertines formed in the absence
of CO2-rich soils and higher plants is a chance to examine
Fig. 7.153 (continued) (q) Cross section through pale beige and pale
yellow travertine crust intersected by travertine veins composed of
bluish, fibrous dolomite. (r) Photomicrograph in transmitted non-
polarised light of cross-section showing two banded travertine crusts
separated by bedded dolomicrite, which is cross-cut by travertine veins
resembling a feeder channel. (s) Photomicrograph in transmitted non-
polarised light of cross-section showing an eroded surface on laminated
dolomicrite and dolostone clast veneered by fibrous dolomite traver-
tine. (t) Photomicrograph in transmitted non-polarised light of plan
view showing dolomicrite clasts overgrown by fibrous dolomite traver-
tine (Photographs (b), (c), (e), (g), (h), (i), (q), (r) and (t) by Victor
Melezhik. Images (a), (d), (n), (o) and (p) are reproduced from
Melezhik et al. (2004) with permission of Elsevier. Images (f), (j), (k)
and (m) are reproduced from Melezhik and Fallick (2001) with permis-
sion of Elsevier. Images (b), (c), (e), (g), (h), (i), (q) are from the
aggregate quarry located in vicinity of FAR-DEEP Hole 5A
(Figs. 4.15 and 6.32). Images (r) and (t) are from drillhole X
(Figs. 4.15 and 6.32), depth 309 m)
9 7.9 Terrestrial Environments 1451
current understanding of models of travertine precipitation
and generation of petrographic fabrics. Conversely, traver-
tines potentially provide a unique window into the Protero-
zoic terrestrial biosphere: how widespread were bacteria
and cyanobacteria in non-marine environments? Were
Palaeoproterozoic land surfaces colonised? Was dolomite a
common primary mineral for travertines, and was this per-
haps related to the rising presence of sulphates and sulphate-
reducing bacteria?
The stable isotope geochemistry of non-marine carbo-
nates can be an excellent tool for palaeoenvironmental
reconstruction, and high-resolution carbon isotope studies
of the laminar travertines preserved in the Pechenga basin
and FAR-DEEP Hole 5A can potentially illuminate other-
wise dark corners of the debate on local versus global signals
during the Lomagundi-Jatuli excursion. These rocks have
not been metamorphosed at high metamorphic grade nor
extensively weathered during (recent) subaerial exposure.
The FAR-DEEP core allows both petrographic and geo-
chemical examination of travertines of the Kuetsj€arvi Sedi-mentary Formation. On a local scale, this allows testing of
depositional models developed in the field (Melezhik and
Fallick 2001; Melezhik et al. 2004). Unlike outcrops, these
accessible cores are from below the present regolith and
therefore invaluable in allowing combined petrographic
and geochemical examination of some of the earth’s earliest
travertines, presumed to be free from modern overprints by
fungi and bacteria (Figs. 7.154 and 7.155).
Fig. 7.154 Travertine crust cementing dolostone debris (pale yellow) in the Kuetsj€arvi Sedimentary Formation. Polished slab parallel to the
bedding surface. Width of the photograph is 20 cm (Photograph by Victor Melezhik)
1452 A.T. Brasier et al.
Fig. 7.155 FAR-DEEP drillcores illustrating abundant travertine
interacting with different sediments. (a) Pink and beige, cracked
cryptalgal laminites overlain by dolarenite, which is, in turn, capped
by beige travertine crust with quartz-filled cavity. (b) White, paleyellow and buff travertine crust with a quartz-filled cavity overlain by
dolomite-cemented sandstone with brown clay material at the base. (c,
d) White travertine cementing and corrosively replacing fragments of
brown clayey siltstone and buff dolarenite. Core diameter is 5 cm
(Photographs by Victor Melezhik)
9 7.9 Terrestrial Environments 1453
References
Andrews JE (2006) Palaeoclimatic records from stable
isotopes in riverine tufas: synthesis and review. Earth Sci
Rev 75:85–104
Andrews JE, Brasier AT (2005) Seasonal records of cli-
matic change in annually laminated tufas: short review and
future prospects. J Quaternary Sci 20:411–421
Andrews JE, Riding R, Dennis PF (1993) Stable isotopic
compositions of recent freshwater cyanobacterial carbonates
from the British Isles: local and regional environmental
controls. Sedimentology 40:303–314
Andrews JE, Riding R, Dennis PF (1997) The stable
isotope record of environmental and climatic signals in
modern terrestrial microbial carbonates from Europe.
Palaeogeogr Palaeocl Palaeoecol 129:171–189
Anzalone E, Ferreri V, Sprovieri M, D’Argenio B (2007)
Travertines as hydrologic archives: the case of the
Pontecagnano deposits (southern Italy). Adv Water Resour
30:2159–2175
Arenas C, Gutierrez F, Osacar C, Sancho C (2000) Sedi-
mentology and geochemistry of fluvio-lacustrine tufa
deposits controlled by evaporite solution subsidence in the
central Ebro Depression, NE Spain. Sedimentology 47:
883–909
Awramik SM, Buchheim HP (2009) A giant, Late
Archean lake system: the Meentheena member (Tumbiana
Formation; Fortescue Group), Western Australia. Precam-
brian Res 174:215–240
Bolhar R, Van Kranendonk MJ (2007) A non-marine
depositional setting for the northern Fortescue Group,
Pilbara Craton, inferred from trace element geochemistry
of stromatolitic carbonates. Precambrian Res 155:229–250
Bonny SM, Jones B (2008) Petrography and textural
development of inorganic and biogenic lithotypes in a relict
barite tufa deposit at Flybye Springs, NT, Canada. Sedimen-
tology 55:275–303
Brasier AT (2011) Searching for travertines, calcretes and
speleothems in deep time: processes, appearances, predictions
and the impact of plants. Earth Sci Rev. 104(4):213–239.
doi: 10.1016/j.earscirev.2010.10.007
Brasier AT, Andrews JE, Kendall AC (2011) Diagenesis
or dire genesis? The origin of columnar spar in tufa
stromatolites of central Greece and the role of Chironomid
larvae. Sedimentology. doi: 10.1111/J.1365-3091.2010.
01208.x
Brasier AT, Andrews JE, Marca-Bell AD, Dennis PF
(2010) Depositional continuity of seasonally laminated
tufas: implications for d18O based palaeotemperatures.
Global Planet Change 71:160–167
Chafetz HS, Folk RL (1984) Travertines: depositional
morphology and the bacterially constructed constituents.
J Sediment Petrol 54:289–316
Chafetz HS, Lawrence JR (1994) Stable isotopic
variability within modern travertines. Geogr Phys Quatern
48:257–273
Chafetz HS, Utech NM, Fitzmaurice SP (1991)
Differences in the delta 18 O and delta 13C signatures of
seasonal laminae comprising travertine stromatolites. J Sed-
iment Res 61:1015–1028
Cohen AS, Thouin C (1987) Nearshore carbonate
deposits in Lake Tanganyika. Geology 15:414–418
Cotter E (1978) The evolution of fluvial style, with spe-
cial reference to the central Appalachian Paleozoic. In: Miall
AD (ed) Fluvial sedimentology, vol 5, Canadian Society of
Petroleum Geologists Memoir. Canadian Society of Petro-
leum Geologists, Calgary, pp 361–383
Davies NS, Gibling MR (2010) Cambrian to Devonian
evolution of alluvial systems: the sedimentological impact
of the earliest land plants. Earth Sci Rev 98:171–200
Davis DW, Paces JB (1990) Time resolution of geologic
events on the Keweenaw Peninsula and implications for
development of the Midcontinent Rift system. Earth Planet
Sci Lett 97:54–64
Eggleston JR, Dean WE (1976) Chapter 8.7 Freshwater
stromatolitic bioherms in Green Lake, New York. In: Walter
M (ed) Developments in sedimentology, vol 20. Elsevier,
Amsterdam, pp 479–488
Elmore RD (1983) Precambrian non-marine stromatolites
in alluvial fan deposits, the Copper Harbor Conglomerate,
upper Michigan. Sedimentology 30:829–842
Fairchild IJ, Borsato A, Tooth AF, Frisia S, Hawkesworth
CJ, Huang Y, McDermott F, Spiro B (2000) Controls on
trace element (Sr-Mg) compositions of carbonate cave
waters: implications for speleothem climatic records. Chem
Geol 166:255–269
Ford TD, Pedley HM (1996) A review of tufa and traver-
tine deposits of the world. Earth Sci Rev 41:117–175
Fouke BW, Farmer JD, Des Marais DJ, Pratt L, Sturchio
NC, Burns PC, DiscipuloMK (2000) Depositional facies and
aqueous-solid geochemistry of travertine-depositing hot
springs (Angel Terrace, Mammoth Hot Springs, Yellowstone
National Park, U.S.A.). J Sediment Res 70:565–585
Freytet P, Plet A (1996) Modern freshwater microbial
carbonates: the Phormidium stromatolites (tufa-travertine)
of southeastern Burgundy (Paris Basin, France). Facies
34:219–237
Freytet P, Verrecchia EP (1998) Freshwater organisms
that build stromatolites: a synopsis of biocrystallization by
prokaryotic and eukaryotic algae. Sedimentology 45:
535–563
Frisia S, Borsato A (2010) Karst. In: Alonso-Zarza AM,
Tanner LH (eds) Developments in sedimentology, vol 61.
Elsevier, Amsterdam, pp 269–318
Garnett ER, Andrews JE, Preece RC, Dennis PF (2004)
Climatic change recorded by stable isotopes and trace
1454 A.T. Brasier et al.
elements in a British Holocene tufa. J Quaternary Sci
19:251–262
Grotzinger JP (1986) Cyclicity and paleoenvironmental
dynamics, Rocknest platform, northwest Canada. Geol Soc
Am Bull 97:1208–1231
Grotzinger JP (1989) Facies and evolution of Precam-
brian carbonate depositional systems: emergence of the
modern platform archetype. In: Crevello PD, Wilson JW,
Sarg JF, Read JF (eds) Controls on carbonate platform and
basin development. SEPM, Tulsa, pp 79–106
Grotzinger JP, Knoll AH (1995) Anomalous carbonate
precipitates: is the Precambrian the key to the Permian?
Palaios 10:578–596
Guido DM, Campbell KA (2009) Jurassic hot-spring
activity in a fluvial setting at La Marciana, Patagonia,
Argentina. Geol Mag 146:617–622
Guidry SA, Chafetz HS (2003) Anatomy of siliceous hot
springs: examples from Yellowstone National Park,
Wyoming, USA. Sediment Geol 157:71–106
Guo L, Riding R (1992) Aragonite laminae in hot water
travertine crusts, Rapolano Terme, Italy. Sedimentology
39:1067–1079
Guo L, Riding R (1994) Origin and diagenesis of Quater-
nary travertine shrub fabrics, Rapolano Terme, central Italy.
Sedimentology 41:499–520
Guo L, Riding R (1998) Hot-spring travertine facies and
sequences, Late Pleistocene, Rapolano Terme, Italy. Sedi-
mentology 45:163–180
Guo L, Andrews J, Riding R, Dennis P, Dresser Q (1996)
Possible microbial effects of stable carbon isotopes in hot-
spring travertines. J Sediment Res 66:468–473
Hammer Ø, Dysthe DK, Jamtveit B (2007) The dynamics
of travertine dams. Earth Planet Sci Lett 256:258–263
Hoffman P (1976) Environmental diversity of Middle
Precambrian stromatolites. In: Walter M (ed) Stromatolites,
Developments in sedimentology. Elsevier, Amsterdam,
pp 599–611
Ihlenfeld C, Norman MD, Gagan MK, Drysdale RN,
Maas R, Webb J (2003) Climatic significance of seasonal
trace element and stable isotope variations in a modern
freshwater tufa. Geochim Cosmochim Acta 67:2341–2357
Irion G, Muller G (1968) Mineralogy, petrology and
chemical composition of some calcareous tufa from the
Schwabische Alb, Germany. In: Muller G, Friedman GM
(eds) Recent developments in carbonate sedimentology in
Central Europe. Springer, Berlin, pp 157–171
Janssen A, Swennen R, Podoor N, Keppens E (1999)
Biological and diagenetic influence in Recent and fossil
tufa deposits from Belgium. Sediment Geol 126:75–95
Kawai T, Kano A, Hori M (2009) Geochemical and
hydrological controls on biannual lamination of tufa
deposits. Sediment Geol 213:41–50
Koban C, Schweigert G (1993) Microbial origin of trav-
ertine fabrics – two examples from Southern Germany
(Pleistocene stuttgart travertines and miocene ried€oschingenTravertine). Facies 29:251–263
Lindsay JF, BrasierMD,McLoughlin N, Green OR, Fogel
M, Steele A, Mertzman SA (2005) The problem of deep
carbon – an Archean paradox. Precambrian Res 143:1–22
Love KM, Chafetz HS (1988) Diagenesis of laminated
travertine crusts, Arbuckle Mountains, Oklahoma. J Sedi-
ment Res 58:441–445
Matsuoka J, Kano A, Oba T, Watanabe T, Sakai S, Seto K
(2001) Seasonal variation of stable isotopic compositions
recorded in a laminated tufa, SW Japan. Earth Planet Sci
Lett 192:31–44
Melezhik VA, Fallick AE (2001) Palaeoproterozoic
travertines of volcanic affiliation from a 13C-rich rift lake
environment. Chem Geol 173:293–312
Melezhik VA, Fallick AE, Grillo SM (2004) Subaerial
exposure surfaces in a Palaeoproterozoic 13C-rich dolostone
sequence from the Pechenga Greenstone Belt: palaeoenvir-
onmental and isotopic implications for the 2330–2060 Ma
global isotope excursion of 13C/12C. Precambrian Res
133:75–103
Melezhik VA, Huhma H, Condon DJ, Fallick AE,
Whitehouse MJ (2007) Temporal constraints on the
Paleoproterozoic Lomagundi-Jatuli carbon isotopic event.
Geology 35:655–658
Merz MUE (1992) The biology of carbonate precipitation
by Cyanobacteria. Facies 26:81–102
Nehza O, Woo KS, Lee KC (2009) Combined textural
and stable isotopic data as proxies for the mid-Cretaceous
paleoclimate: a case study of lacustrine stromatolites in the
Gyeongsang Basin, SE Korea. Sediment Geol 214:85–99
O’Brien GR, Kaufman DS, Sharp WD, Atudorei V,
Parnell RA, Crossey LJ (2006) Oxygen isotope composition
of annually banded modern and mid-Holocene travertine and
evidence of paleomonsoon floods, Grand Canyon, Arizona,
USA. Quaternary Res 65:366–379
Pedley HM (1990) Classification and environmental
models of cool freshwater tufas. Sediment Geol 68:143–154
Pedley M, Gonzalez Martın JA, Ordonez S, Garcıa Del
Cura MA (2003) Sedimentology of Quaternary perched
springline and paludal tufas: criteria for recognition, with
examples from Guadalajara Province, Spain. Sedimentology
50:23–44
Pelechaty SM, James NP (1991) Dolomitized middle
Proterozoic calcretes, Bathurst Inlet, Northwest Territories,
Canada. J Sediment Res 61:988–1001
Pelechaty SM, James NP, Kerans C, Grotzinger JP (1991)
A middle Proterozoic palaeokarst unconformity and
associated sedimentary rocks, Elu Basin, northwest Canada.
Sedimentology 38:775–797
9 7.9 Terrestrial Environments 1455
Pentecost A (1990) The formation of travertine shrubs;
Mammoth Hot Springs, Wyoming. Geol Mag 127:159–168
Pentecost A (1993) British travertines: a review. Proc
Geologist Assoc 104:23–39
Pentecost A (1995) The quaternary travertine deposits of
Europe and Asia Minor. Quaternary Sci Rev 14:1005–1028
Pentecost A, Spiro B (1990) Stable carbon and oxygen
isotope composition of calcites associated with modern
freshwater cyanobacteria and algae. Geomicrobiol J 8:17–26
Pentecost A, Viles H (1994) A review and reassessment of
Travertine classification. Geogr Phys Quatern 48:305–314
Rainbird RH, Davis WJ, Stern RA, Peterson TD, Smith
SR, Parrish RR, Hadlari T (2006) Ar-Ar and U-Pb Geochro-
nology of a Late Paleoproterozoic Rift Basin: support for a
genetic link with Hudsonian orogenesis, Western Churchill
Province, Nunavut, Canada. J Geol 114:1–17
Rainey DK, Jones B (2009) Abiotic versus biotic controls
on the development of the Fairmont Hot Springs carbonate
deposit, British Columbia, Canada. Sedimentology
56:1832–1857
Riding R (1979) Origin and diagenesis of lacustrine algal
bioherms at the margin of the Ries crater, Upper Miocene,
southern Germany. Sedimentology 26:645–680
Riding R (1991) Classification of microbial carbonates.
In: Riding R (ed) Calcareous algae and stromatolites.
Springer, Berlin, pp 21–51
Riding R (2000) Microbial carbonates: the geological
record of calcified bacterial-algal mats and biofilms. Sedi-
mentology 47:179–214
Riding R (2008) Abiogenic, microbial and hybrid
authigenic crusts: components of Precambrian stromatolites.
Geologia Croatica 61:73–103
Risacher F, Eugster HP (1979) Holocene pisoliths and
encrustations associated with spring-fed surface pools,
Pastos Grandes, Bolivia. Sedimentology 26:253–270
Rogerson M, Pedley HM, Wadhawan JD, Middleton R
(2008) New insights into biological influence on the geo-
chemistry of freshwater carbonate deposits. Geochim
Cosmochim Acta 72:4976–4987
Skotnicki SJ, Knauth LP (2007) The Middle Proterozoic
Mescal Paleokarst, Central Arizona, U.S.A.: karst develop-
ment, silicification, and cave deposits. J Sediment Res
77:1046–1062
Soligo M, Tuccimei P, Barberi R, Delitala MC, Miccadei
E, Taddeucci A (2002) U/Th dating of freshwater travertine
fromMiddle Velino Valley (Central Italy): paleoclimatic and
geological implications. Palaeogeogr Palaeocl Palaeoecol
184:147–161
Soreghan MJ, Cohen AS (1996) Textural and composi-
tional variability across littoral segments of Lake
Tanganyika: the effect of asymmetric basin structure on
sedimentation in Large Rift Lakes. Am Assoc Petrol Geolo-
gist Bull 80:382–409
Szulc J, Cwizewicz M (1989) The lower permian fresh-
water carbonates of the Slawkow graben, Southern Poland:
sedimentary facies context and stable isotope study.
Palaeogeogr Palaeocl Palaeoecol 70:107–120
Takashima C, Kano A (2008) Microbial processes
forming daily lamination in a stromatolitic travertine. Sedi-
ment Geol 208:114–119
Thrailkill J (1976) Speleothems. In: Walter MR (ed)
Stromatolites. Elsevier, Amsterdam, pp 73–86
Wacey D, Wright DT, Boyce AJ (2007) A stable isotope
study of microbial dolomite formation in the Coorong
Region, South Australia. Chem Geol 244:155–174
Walter MR (1996) Ancient hydrothermal ecosystems
on Earth: a new palaeobiological frontier. In: Evolution
of hydrothermal ecosystems on Earth (and Mars?),
vol 202, Ciba foundation symposium. Wiley, Chichester,
pp 112–127
1456 A.T. Brasier et al.
7.10 Chemical Characteristics of Sedimentsand Seawater
7.10.1 Introductory Remarks
Lee R. Kump (Editor)
The transition from an anoxic to oxygenated atmosphere
was arguably the most dramatic change in the history of the
Earth. This “Great Oxidation Event” (Holland 2006)
transformed the biogeochemical cycles of the elements by
imposing an oxidative step in the cycles, creating strong
redox gradients in the terrestrial and marine realms that
energised microbial metabolism. Although much past
research was focused on establishing when the rise of
atmospheric oxygen took place, recognition that substantial
mass-independent fraction (MIF) of the sulphur isotopes is
restricted to the time interval before 2.45 Ga and requires
an anoxic atmosphere (Farquhar et al. 2000, 2007; Mojzsis
et al. 2003; Ono et al. 2003; Bekker et al. 2004) argues the
atmosphere became permanently oxygenated at this time
(Pavlov and Kasting 2002). A false-start to the modern
aerobic biosphere and a “whiff” of atmospheric oxygen
(Anbar et al. 2007) may have occurred in the latest
Archaean, as reflected in a transient enrichment in the
redox-sensitive element molybdenum in marine shales
and a reduction in the extent of MIF precisely coincident
with the peak in Mo and FeS2 enrichment (Kaufman et al.
2007). Geochemical proxies are imperfect, and an earlier
(c. 3 Ga) appearance of atmospheric oxygen is possible
(Ohmoto et al. 2006) but disputed (Farquhar et al. 2007;
Buick 2008).
The interval between 2.45 and 2.0 Ga (the early
Palaeoproterozoic) was a period of transition, with evidence
for widespread glaciation (Evans et al. 1997; Kirschvink
et al. 2000), appearance of red beds, and disappearance of
detrital grains of pyrite and uraninite, unstable in an oxida-
tive weathering environment (Holland 1994). The banded
iron formations, so characteristic of the Archaean, largely
disappeared during the early Palaeoproterozoic, despite evi-
dence for the type, if not magnitude of volcanism that was
associated with BIF in the Archaean and earliest Palaeopro-
terozoic (Condie et al. 2001). Instead, large Mn deposits and
only minor BIFs that have Phanerozoic affinities (e.g. oolitic
ironstones) are present (Kirschvink et al. 2000; Bekker et al.
2004). Pb-Pb isotope studies indicate that the geochemical
cycles of U and Th became decoupled due to the redox
cycling of U (Pollack et al. 2007), but chondritic Os initial
ratios in 2.32 and 2.0 Ga shales (Hannah et al. 2004, 2006)
suggest that the ocean was dominated by hydrothermal Os
flux while Os flux from oxidative continental weathering
was either small and/or scavenged in epicontinental anoxic
and euxinic settings. A significant increase in mass-depen-
dent S isotope fractionation in sedimentary pyrite after
2.32 Ga (Canfield 2005) and the first appearance of sedi-
mentary sulphate evaporites (Melezhik et al. 2005b) likely
reflects an increase of seawater sulphate concentrations by
that time. The phenomenal Lomagundi-Jatuli carbon isotope
excursion (2.22–2.06 Ga) was followed, paradoxically, by
an interval of remarkable organic carbon accumulation and
the first sedimentary phosphorite accumulation (Melezhik
et al. 2005a).
A goal of FAR-DEEP is to document the sequence of
environmental changes associated with and pursuant to the
Great Oxidation Event by analysis of newly available drill
core from Fennoscandia and numerical modeling. Doing so
allows one to evaluate hypotheses including:
1. The disappearance of banded iron formation during the
interval of 2.45–1.84 Ga reflects an initial oxidation of the
deep ocean followed by the onset of euxinia.
2. The build-up of a sulphate-enriched ocean progressed
gradually through the Palaeoproterozoic to levels that
only by the time of the Shunga event (c. 2000 Ma)
could support a euxinic ocean.
3. Alternatively, euxinia was initiated earlier, during the
Lomagundi–Jatuli carbon isotope event (2.3–2.1 Ga),
whose high fractional organic-carbon burial rates resulted
L.R. Kump (Editor)
Department of Geosciences, Pennsylvanian State University, 503
Deike Building, University Park, PA 16870, USA
e-mail: [email protected]
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_10, # Springer-Verlag Berlin Heidelberg 2013
1457
from a decoupling of phosphate and organic-carbon
burial during widespread oceanic euxinia (Van Cappellen
and Ingall 1996; Aharon 2005).
4. Oxidative weathering was unimportant until ~2.3 Ga, as
reflected in the persistence of detrital uraninite and pyrite
and minor MIF (Papineau et al. 2007) in the older
sequences (Elliott Lake and Hough Groups) of Huronian
glacial sediments.
In this chapter we discuss a variety of environmental
(especially redox) proxies, including Sr isotopic composi-
tion of carbonates, Fe speciation, Mo abundance, Ca, Mg,
Fe, U, Cr and Mo isotopes, initial Os isotopic compositions,
and Fe mineral speciation that can be used to evaluate these
hypotheses. Obtaining precise radiometric dates using
Re-Os isotopes is critical to this task, allowing one to corre-
late among basins and thus establish the chronology and
pace of events, so the prospects for FAR-DEEP cores to
provide better geochronological constraints is discussed.
References
Aharon P (2005) Redox stratification and anoxia of the
early Precambrian oceans: implications for carbon isotope
excursions and oxidation events. Precambrian Res
137:207–222
Anbar AD, Duan Y, Lyons TW, Arnold GL, Kendall B,
Creaser RA, Kaufman AJ, Gordon GW, Scott C, Garvin J,
Buick R (2007) A whiff of oxygen before the great oxidation
event? Science 317:1903–1906
Bekker A, Holland HD, Wang PL, Rumble D III, Stein
HJ, Hannah JL, Coetzee LL, Beukes NJ (2004) Dating the
rise of atmospheric oxygen. Nature 427:117–120
Buick R (2008) When did oxygenic photosynthesis
evolve? Philos Trans Roy Soc B 363:2731–2743
Canfield DE (2005) The early history of atmospheric
oxygen: homage to Robert M. Garrels. Annu Rev Earth
Planet Sci 33:1–36
Condie KC, Des Marais DJ, Abbott D (2001) Precam-
brian superplumes and supercontinents: a record in black
shales, carbon isotopes, and paleoclimates? Precambrian
Res 106:239–260
Evans DA, Beukes MJ, Kirschvink JL (1997) Low-
latitude glaciation in the Palaeoproterozoic era. Nature
386:262–266
Farquhar J, Bao H, Thiemens M (2000) Atmospheric
influence of Earth’s earliest sulfur cycle. Science 289:
756–758
Farquhar J, Peters M, Johnston DT, Strauss H, Masterson
A, Wiechert U, Kaufman AJ (2007) Isotopic evidence for
Mesoarchaean anoxia and changing atmospheric sulphur
chemistry. Nature 449:706–710
Hannah JL, Bekker A, Stein HJ, Markey RJ, Holland
HD (2004) Primitive Os and 2316 Ma age for marine
shale: implications for Paleoproterozoic glacial events
and the rise of atmospheric oxygen. Earth Planet Sci Lett
225:43–52
Hannah JL, Stein HJ, Zimmerman A, Yang G, Markey
RJ, Melezhik VA (2006) Precise 2004 � 9 Ma Re-Os age
for Pechenga black shale: Comparison of sulfides and
organic material. Geochim Cosmochim Acta 70:A228
Holland HD (1994) Early Proterozoic atmospheric
change. In: Bengston S (ed) Early life on earth. Columbia
University Press, New York, pp 237–244
Holland HD (2006) The oxygenation of the atmosphere
and oceans. Philos Trans Roy Soc B 361:903–915
Kaufman AJ, Johnston DT, Farquhar J, Masterson AL,
Lyons TW, Bates S, Anbar AD, Arnold GL, Garvin J,
Buick R (2007) Late Archean biospheric oxygenation and
atmospheric evolution. Science 317:1900–1903
Kirschvink JL et al (2000) Paleoproterozoic snowball
Earth: extreme climatic and geochemical global change
and its biological consequences. Proc Natl Acad Sci 97:
1400–1405
Melezhik VA, Fallick AE, Hanski E, Kump L, Lepland
A, Prave A, Strauss H (2005a) Emergence of the modern
Earth system during the Archean-Proterozoic transition.
GSA Today 15:4–11
Melezhik VA, Fallick AE, Rychanchik DV, Kuznetsov
AB (2005b) Palaeoproterozoic evaporites in Fennoscandia:
implications for seawater sulphate, d13C excursions and the
rise of atmospheric oxygen. Terra Nova 17:141–148
Mojzsis SJ, Coath CD, Greenwood JP, McKeegan KD,
Harrison TM (2003) Mass-independent isotope effects in
Archean (2.5 to 3.8 Ga) sedimentary sulfides determined
by ion microprobe analysis. Geochim Cosmochim Acta
67:1635–1658
Ohmoto H, Watanabe Y, Ikemi H, Poulson SR, Taylor
BE (2006) Sulphur isotope evidence for an oxic Archaean
atmosphere. Nature 442:908–911
Ono S, Eigenbrode JL, Pavlov AA, Kharecha P, Rumble
III D, Kasting JF, Freeman KH (2003) New insights into
Archean sulfur cycle from mass-independent sulfur isotope
records from the Hamersley Basin, Australia. Earth Planet
Sci Lett 213:15–30
Papineau D, Mojzsis SJ, Schmitt AK (2007) Multiple
sulfur isotopes from Paleoproterozoic Huronian interglacial
sediments and the rise of atmospheric oxygen. Earth Planet
Sci Lett 255:188–212
Pollack GD, Krogstad EJ, Bekker A (2007) Evaluating
the nature of the rise of atmospheric oxygen using lead
isotope compositions from Paleoproterozoic shale suites,
2007 GSA Annual Meeting, vol 39, p 348
Van Cappellen P, Ingall ED (1996) Redox stabilization of
the atmosphere and oceans by phosphorus-limited marine
productivity. Science 271:493–496
1458 L.R. Kump
7.10.2 Sr Isotopes in Sedimentary Carbonates
Anton B. Kuznetsov, Igor M. Gorokhov, andVictor A. Melezhik
Study of the 87Sr/86Sr ratio variations in ancient oceans is an
important tool for the reconstruction of geodynamic settings
in the past and an assessment of the Earth’s crustal compo-
sition and its erosion intensity at different stages of its
evolution. The efficacy of this tool results from a uniformity
of the 87Sr/86Sr ratio in the global ocean at each instant of
the geologic history, inasmuch as the Sr residence time in
seawater is three orders of magnitude longer than that of
ocean mixing (Goldberg 1963; Faure 1986; Hodell et al.
1989). Strontium isotope variations in the ocean derive
from changes in the relationships among three global
variables: (1) the mantle Sr flux with low 87Sr/86Sr value,
(2) the continental discharge, and (3) the 87Sr/86Sr value in
this discharge (Faure et al. 1965; Veizer and Compston
1974; Brass 1976; Spooner 1976; Goldstein and Jacobsen
1987). The mantle flux originates from the hydrothermal
processing of basalts in mid-oceanic ridges and from
weathering of oceanic islands. The continental runoff
forms during the processes of continental crust weathering
resulting from the action of rain, soil, and groundwater on
Sr-bearing minerals. Thus, Sr isotope variations in the ocean
reflect a balance of the fluxes representative of the mantle
and crustal reservoirs differing in 87Sr/86Sr ratio.
Strontium concentrations in seawater are low enough for
it to be undersaturated with respect to Sr minerals, but this
does not preclude the isomorphic co-precipitation of Sr with
Ca in carbonates, sulphates and phosphates. Naturally occur-
ring fractionation of Sr isotopes among oceanic water and
carbonate sediments is insignificant due to rather high mass
numbers of these isotopes, and researchers dealing with the
Sr isotope variations in terrestrial materials ignored this
effect over many years and normalised measured isotope
ratios to 88Sr/86Sr ¼ 8.375209 (Nier 1938). Yet with the
advent of modern mass spectrometers and increasing preci-
sion of isotope measurements, the ability to detect and quan-
titatively evaluate natural fractionation of the Sr isotopes is
now achieved (Fietzke et al. 2008; Halicz et al. 2008;
Ruggeberg et al. 2008; Ohno et al. 2008). Krabbenhoft
et al. (2010) demonstrated that the average 88Sr/86Sr ratio
of modern marine carbonates and corals is about 0.22 ‰lower than that in seawater. This is caused by the preferential
uptake of lighter isotopes during carbonate precipitation.
The products of chemical weathering of terrestrial rocks
(the terra rossa soil and speleothem calcite) yield 0.45 ‰lower 88Sr/86Sr value than seawater (Halicz et al. 2008).
Temperature dependence of the natural fractionation
results in the distinction of the 88Sr/86Sr ratio between cold-
water and tropical corals within 0.33 ‰ (Fietzke et al. 2008;
Halicz et al. 2008; Ruggeberg et al. 2008). So, according to
available data, the stable Sr isotope fractionation in modern
biogenic carbonates amounts to a difference in 87Sr/86Sr value
of about 0.00015. In actual practice this fractionation effect is
corrected by normalisation of themeasured 87Sr/86Sr values to
the 88Sr/86Sr ¼ 8.375209, yet the uncertainty in the 87Sr/86Sr
of ancient carbonates is of the same order as associated with
the diagenetic overprint. Variations of 87Sr/86Sr ratio in
samples from the same stratigraphic level could be higher
than 0.00005 for Palaeozoic carbonates (Denison et al. 1997,
1998; Veizer et al. 1999; McArthur et al. 2001), and they
increase to 0.00020–0.00030 in Neo- and Mesoproterozoic
carbonates (Derry et al. 1992; Kaufman et al. 1993; Gorokhov
et al. 1995; Semikhatov et al. 2002; Ray et al. 2003; Halverson
et al. 2007; Kuznetsov et al. 2008). As a consequence, with
regard to scatter due to diagenetic overprint, the 87Sr/86Sr ratio
in minerals crystallised from Precambrian seawater can be
considered representative of the parent solution, and the Sr
isotopic composition of ancient seawater can be determined
by the analysis of marine authigenic minerals. In contrast to
carbonates, sulphates and phosphates are of limited geological
abundance. Therefore, with the exception of the Cretaceous to
recent, where marine barites provide an alternative (Paytan
et al. 1993), most of the Sr isotopic history in seawater can be
reconstructed by study of carbonate rocks (Peterman et al.
1970; Tremba et al. 1975; Koepnick et al. 1985; Derry et al.
1992, and others).
Present-day view of the Sr isotope composition in
Palaeoproterozoic seawater (Veizer and Compston 1976;
Veizer et al. 1992a, b; Mirota and Veizer 1994; Bekker
et al. 2003b; Frauenstein et al. 2009; Kuznetsov et al. 2010)
is based on analytical results for a moderate amount (about
200) of carbonate samples (Fig. 7.156) compared to that used
for the Neoproterozoic and Phanerozoic. Moreover, many of
these samples were taken from sedimentary successions not
always related with confidence to a chronostratigraphic scale,
and not all of those are actually feasible for the isotope
characterisation of marine sediments. A major portion of
the carbonate rocks were subjected to secondary recrystal-
lisation resulting in the disruption of the initial Rb-Sr systems
in the samples. Some of the samples were taken from deposits
that accumulated in intracontinental palaeobasins. As a
result, evolution of the Sr isotopic composition in Palaeopro-
terozoic ocean is only roughly known (Veizer and Compston
1976; Mirota and Veizer 1994).
The most representative set of the Sr isotopic data has
been obtained for Palaeoproterozoic carbonate samples from
the Transvaal Supergroup exposed in the Transvaal and
A.B. Kuznetsov (*)
Institute of Precambrian Geology and Geochronology, Russian
Academy of Sciences, Makarova 2, 199034 St. Petersburg, Russia
10 7.10 Chemical Characteristics of Sediments and Seawater 1459
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_10, # Springer-Verlag Berlin Heidelberg 2013
1459
Griqualand West areas of South Africa. This set involves
carbonates of the Malmani Subgroup (Veizer et al. 1992a)
and of the Duitschland (Bekker et al. 2001; Frauenstein et al.
2009), Silverton and Lucknow (Frauenstein et al. 2009)
formations in the Transvaal area, and those of the Kogelbeen
and Gamohaan (Kamber and Webb 2001), Rooinekke
(Frauenstein et al. 2009), Hotazel and Moodrai
(Schneiderhan et al. 2006) formations in Griqualand West
Area. The Sr isotope composition was studied also from the
Gandarela and Fecho-do-Funil carbonate formations, which
are parts of the Minas Supergroup in Sao Francisco Craton
of South America (Bekker et al. 2003b). Early Palaeopro-
terozoic sedimentary successions of North America are
presented by Dunphy, Portage, Alder and Uve formations
from the Kaniapiskau Supergroup in Labrador Trough
(Kuznetsov et al. 2003), and by the Nash Fork Formation
from Great Lakes area (Bekker et al. 2003a), the Kona
Dolomite Formation from Wyoming area (Bekker et al.
2006) and the Recluse Subgroup (including the Cowles
Lake, Odjick and Rocknest Formations) from northern
Canada (Veizer et al. 1992b). An appreciable contribution
to the Sr Palaeoproterozoic database is made by carbonate
successions of the Fennoscandian Shield: the Tulomozero
Formation in the Onega Lake area (Gorokhov et al. 1998;
Kuznetsov et al. 2010) and the Kuetsj€arvi Formation in the
Pechenga Greenstone Belt (Melezhik et al. 2005a). Except
for the metalliferous Hotazel Formation, the above-mentioned
stratigraphic units contain sedimentary carbonates (lime-
stones and/or dolostones) and were deposited in marine plat-
form or marine-influenced rift environments.
Geochemical criteria for evaluation of the Rb-Sr system
alteration in Phanerozoic and Neoproterozoic sedimentary
carbonate rocks are based on empirical data about re-
distribution Mn, Fe and Sr and modification of d18O during
low-temperature freshwater diagenesis (Banner and Hanson
1990). The above-listed criteria, however, are not univer-
sally suitable for Palaeoproterozoic carbonate rocks. These
rocks are generally metamorphosed in conditions from
prehnite-pumpellyite subfacies to amphibolite facies. Fur-
thermore, due to deficiency of dissolved oxygen, the marine
environment could be chemically reduced with high Mn and
Fe concentrations and thus high contents of Mn and Fe in
carbonate minerals could be expected. Nevertheless,
Palaeoproterozoic dolostones and limestones with elevated
Sr content, high d18O value and low Mn/Sr and Fe/Sr ratios
in many instances are marked by 87Sr/86Sr values close to
that in the seawater (Melezhik et al. 2005b). At the same
time, the 87Sr/86Sr ratio in ankerites is usually higher than in
limestones and dolostones of the same stratigraphic unit
(Veizer et al. 1992b; Frauenstein et al. 2009).
Among the early Palaeoproterozoic carbonate forma-
tions studied to date, only ten contain relatively unaltered
samples feasible for assessment of the 87Sr/86Sr ratio in the
2.5–1.9 Ga ocean. These samples were selected, however, on
the basis of dissimilar geochemical criteria and different
numerical values of these criteria. By way of illustration,
the least altered samples of marine limestones of the
Duitschland Formation (Pretoria Group, Frauenstein et al.
2009), Kona Dolomite Formation (Chocolay Group, Bekker
et al. 2006), Kuetsj€arvi Formation, (North Pechenga Group,
Melezhik et al. 2005a) and Cowles Lake Formation (Coro-
nation Supergroup, Veizer et al. 1992b) were selected on the
basis of low and variable Mn/Sr ratio, respectively <1.7,
<0.8, <0.5, and again <0.5. The Mn/Sr values in the least
altered marine dolostones of the Kaniapiskau Supergroup
(Kuznetsov et al. 2003) and the Tulomozero Formation
(Gorokhov et al. 1998; Kuznetsov et al. 2010) are less than
2.7 and 2.0, respectively. The lowest 87Sr/86Sr ratios in
limestone of the Gandarela Formation (Minas Supergroup,
Bekker et al. 2003b) and dolostone of the Nash Fork For-
mation (Libbly Creek Group, Bekker et al. 2003a) are
accompanied by the highest (�8.6 ‰ and �8.8 ‰ V-PDB,
respectively) d18O values. For the Gamohaan Formation
(Campbellrand Subgroup, Kamber and Webb 2001) and
the Fecho-do-Funil Formation (Minas Group, Bekker et al.
2003b), selection of the least altered samples feasible for
assessment of the Sr isotopic composition in seawater was
based solely on their minimal 87Sr/86Sr ratios.
Of the ten above-mentioned samples, only six are directly
dated and have reliable age constraints (Table 7.11 and
Fig. 7.157). The direct dates of the formations are obtained
by the U-Pb method for minerals (usually zircon) from
volcanic intercalations (Bowring and Grotzinger 1992;
Rohon et al. 1993; Sumner and Bowring 1996; Melezhik
et al. 2007) or by the Pb-Pb method for least-altered
carbonates previously enriched in initial carbonate material
by stepwise dissolution (Babinski et al. 1995; Ovchinnikova
et al. 2007). Great age uncertainties of the other formations
hinder their use for reconstruction of the 87Sr/86Sr ratio
variations in Palaeoproterozoic ocean.
As a result, the Palaeoproterozoic seawater 87Sr/86Sr
record remains in its infancy. The deficiency of knowledge
arises from several significant obstacles: (1) small quantity
of continuous carbonate successions, (2) the dearth of robust
radiometric ages of the formations, and (3) diagenetic alter-
ation of carbonates. Because of diagenetic overprint,
variations of 87Sr/86Sr in even the least-altered samples of
individual formations vary from 0.0002 to 0.0006
(Table 7.11) and are much more than those in Meso- and
Neoproterozoic carbonates. The positions of the most plau-
sible points on the age axis are separated by intervals
between 30 and 300 m.y. (Fig. 7.157). All this leads to the
conclusion that the 87Sr/86Sr ratio in 2.5–1.9 Ma ocean
varied within narrow limits of 0.7034–0.7048. These values
1460 Kuznetsov et al.
are higher than those in Archaean seawater (0.7015–0.7025)
(Veizer et al. 1989; Kamber and Webb 2001) and in the
Palaeoproterozoic mantle reservoir (0.7015–0.7020) (Faure
1986). This clearly demonstrates a Palaeoproterozoic
increase in the contribution from continental runoff of the
products of weathering of the growing Rb-rich continental
crust (Veizer and Compston 1976; Shields 2007). A lower-
ing of 87Sr/86Sr ratio in c. 2.1 Ga seawater could be due to an
Fig. 7.156 Sr isotope data for carbonate rocks from Palaeproterozoic
successions. Age constraints as inferred in publications. The data are
from: (1) Kogelbeen Formation, Campbellrand Subgroup (Kamber
and Webb 2001); (2) Gamohaan Formation, Campbellrand Subgroup
(Kamber and Webb 2001); (3) Malmani Subgroup (Veizer
et al. 1992a); (4) Gandarela Formation, Minas Supergroup (Bekker
et al. 2003b); (5) Duitschland Formation, Pretoria Group (Frauenstein
et al. 2009), (6) Duitschland Formation, Pretoria Group (Bekker
et al. 2001); (7) Rooinekke Formation, Koegas Subgroup (Frauenstein
et al. 2009), (8) Hotazel BIF Formation, Voelwater Subgroup
(Schneiderhan et al. 2006); (9) Mooidrai Formation, Voelwater Sub-
group (Schneiderhan et al. 2006), (10) Kona Dolomite Formation,
Chocolay Group (Bekker et al. 2006); (11) Dunphy and Portage
Formations, Seward Subgroup (Kuznetsov et al. 2003); (12) Alder
and Uve Formations, Pistolet Subgroup (Kuznetsov et al. 2003);
(13) Nash Fork Formation, Libbly Creek Group (Bekker et al.
2003a); (14) Fecho-do-Funil Formation, Minas Group (Bekker et al.
2003b); (15) Lucknow Formation, Postmasburg Group (Bekker et al.
2001; Frauenstein et al. 2009); (16) Tulomozero Formation, Jatuli
(Gorokhov et al. 1998; Kuznetsov et al. 2010); (17) Kuetsj€arvi Forma-
tion, North Pechenga Group (Melezhik et al. 2005a); (18) Cowles LakeFormation, Coronation Supergroup (Veizer et al. 1992b). Light-greyfield marks the pioneering temporal 87Sr/86Sr trend in early Palaeopro-
terozoic seawater set up by Veizer and Compston (1976)
10 7.10 Chemical Characteristics of Sediments and Seawater 1461
increase in the hydrothermal flux to ocean and erosion of
continental basalts (Melezhik et al. 2005a).
FAR-DEEP has been designed to advance our under-
standing of a series of global palaeoenvironmental events
(Fig. 1.1 in Chap. 1.1); the evolution of Sr-isotopic compo-
sition of seawater through the early Palaeoproterozoic is
closely linked to this objective. FAR-DEEP targeted several
formations containing continuous limestone and dolostone
successions ranging in age from 2430 to c. 2000 Ma, hence
associated with a series of global-scale perturbations
occurring on Earth surface, including the irreversible oxida-
tion of the terrestrial atmosphere. Consequently, the FAR-
DEEP core material allows studying Sr-isotope system in
continuous sections and calibrating seawater Sr-isotopic
composition against various global-scale processes. Deposi-
tion of 2430–2000 Ma carbonate formations has been
associated with intensive contemporaneous felsic to inter-
mediate volcanism, which, in principle, enables precise
radiometric dating. If proven positive, this should provide a
significant contribution to more robust construction of the
Table 7.11 Sr isotope data in least-altered carbonate samples from precisely dated early Palaeoproterozoic formations (2550–1850 Ma)
Formation Geochronological data Sr isotope data
Material dated Method Age, Ma Reference Rock 87Sr/86Sr Reference
Campbellrand Subgroup, Transvaal Supergroup, South Africa
Gamohaan Zircon from interbedded tuff U-Pb 2521 � 3 Sumner and
Bowring (1996)
Da 0.70230 Kamber and
Webb (2001)D 0.70238
D 0.70244
D 0.70250
Itabria Group, Minas Supergroup, Sao Francisco Craton, South America
Gandarela Dolostone Pb-Pb 2420 � 20 Babinski et al.
(1995)
La 0.70416 Bekker et al.
(2003b)L 0.70339
Pistolet Subgroup, Kaniapiskau Supergroup, Labrador Trough, North America
Alder Zircons from underlying and
overlying mafic sills
U-Pb >2142 � 4 Rohon et al.
(1993)
D 0.70479 Kuznetsov
et al. (2003)U-Pb <2169 � 2
Jatulian Superhorizon, Fennoscandian Shield, Onega Lake area, North Europe
Tulomozero, marine
facies (Member B)
Dolostone Pb-Pb 2090 � 70 Ovchinnikova
et al. (2007)
D 0.70418 Gorokhov
et al. (1998)
D 0.70430 Kuznetsov
et al. (2010)D 0.70427
D 0.70435
D 0.70442
D 0.70434
Tulomozero, marine
facies (Member G)
Dolostone Pb-Pb 2090 � 70 Ovchinnikova
et al. (2007)
D 0.70343 Gorokhov
et al. (1998)D 0.70409
D 0.70345 Kuznetsov
et al. (2010)D 0.70384
D 0.70367
D 0.70409
D 0.70403
D 0.70377
North Pechenga Group, Fennoscandian Shield, Kola Peninsula, North Europe
Kuetsjarvi, marine
facies
Zircon from overlying lava U-Pb 2058 � 6 Melezhik et al.
(2007)
L 0.70431 Melezhik et al.
(2005a)L 0.70407
L 0.70406
L 0.70410
Recluse Group, Coronation Supergroup, Canadian Shield, North America
Cowles Lake Zircon from interlayered ash
bed
U-Pb 1882 � 4 Bowring and
Grotzinger (1992)
L 0.70474 Veizer et al.
(1992b)aD dolostone, L limestone
1462 Kuznetsov et al.
87Sr/86Sr reference curve. Overall, new data will help to
detail 87Sr/86Sr variations in Palaeoproterozoic ocean.
The oldest drilled carbonate unit, the c. 2430 Ma
Seidorechka Sedimentary Formations (Fig. 7.158a), also
exposed in a nearby quarry, is associated with environments
transitional from chemically reduced to oxic atmosphere,
whereas the successive marine limestones of the Polisarka
Sedimentary Formation (Fig. 7.158b) are interbedded with
glaciomarine rocks, thus were formed during global-scale
Huronian glaciation. Sr-isotope compositions of these pre-
glacial and glacial marine carbonates have a great potential
for tracking, through 87Sr/86Sr of seawater, the erosional rate
of the pre-glacial flood-basalt provinces whose enhanced
weathering has been considered as one of many possible
factors causing the onset of the Huronian glacial conditions
(e.g. Melezhik 2006).
Several supposedly contemporaneously formed 13C-rich
carbonate formations have been drilled in basins whose
depositional settings range from rift-bound lacustrine and
open marine (Fig. 7.158c, d) to platformal carbonates
accumulated within restricted (evaporitic) through semi-
restricted to open marine environments (Fig. 7.158e–g).
The large array of carbonate lithologies is an ideal target
for studying, and perhaps quantifying, the influence of con-
tinental input on 87Sr/86Sr of carbonates precipitated in vari-
ous depositional and tectonic settings. The 13C-rich
carbonates record the Lomagundi-Jatuli positive d13Ccarb
excursion whose internal structure and second-order
variations remain poorly constrained (see Chap. 7.3).
A refined 87Sr/86Sr evolutionary trend along with the
d13Ccarb trend based on densely-sampled sections may assist
in interbasinal correlations and identification of small-scale
variations of the carbon isotopes, and their nature.
Some platformal carbonate successions contain abundant
Ca-sulphate. Although many sulphates are partially replaced
by dolomite and quartz (Fig. 7.158g), sulphate remnants are
large enough (e.g.Melezhik et al. 2005b) to obtain Sr-isotopic
composition by employing in situ analysis. Strontium isotopic
composition of early Palaeoproterozoic sulphates is in its
earliest infancy (Cameron 1983; Deb et al. 1991); thus the
available FAR-DEEP core material represents an attractive
target for exploring new area of research.
Fig. 7.157 Compilation of Sr isotope data in least-altered carbonate
samples from directly dated early Palaeoproterozoic units
(2.55–1.85 Ga) (see Table 7.11). Horizontal bars indicate uncertainty
in age. The data are from: (1) Gamohaan Formation, Campbellrand
Subgroup (Kamber and Webb 2001); (2) Gandarela Formation, Minas
Supergroup (Bekker et al. 2003b); (3) Alder Formation, Pistolet Sub-
group (Kuznetsov et al. 2003); (4a) marine facies of Member B of the
Tulomozero Formation, Jatuli (Gorokhov et al. 1998; Kuznetsov et al.
2010); (4b) Member G of the Tulomozero Formation, Jatuli (Gorokhov
et al. 1998; Kuznetsov et al. 2010); (5) marine facies of the Kuetsj€arviFormation, North Pechenga Group (Melezhik et al. 2005a); (6) CowlesLake Formation, Coronation Supergroup (Veizer et al. 1992b). Light-grey field marks the temporal 87Sr/86Sr trend in Palaeoproterozoic
seawater set up by Veizer and Compston (1976). Dark-grey field
marks presently accepted 87Sr/86Sr variations in early Palaeopro-
terozoic seawater (this work). Question-marks symbolise the least
altered samples from the formations without reliable age constraints
10 7.10 Chemical Characteristics of Sediments and Seawater 1463
Fig. 7.158 FAR-DEEP cores exemplifying carbonate rocks which
have accumulated in different depositional and tectonic settings, and
have a potential to address several questions concerning the Sr-isotopic
composition in early Palaeoproterozoic seawater; core width is 5 cm.
1464 Kuznetsov et al.
References
Babinski M, Chemale F Jr, Van Schmus WR (1995)
The Pb/Pb age of the Minas Supergroup carbonate
rocks, Quadrilatero Ferrıfero, Brazil. Precambrian Res
72:235–245
Banner JL, Hanson GN (1990) Calculation of simulta-
neous isotopic and trace element variations during water-
rock interaction with applications of carbonate diagenesis.
Geochim Cosmochim Acta 54:3123–3137
Bekker A, Kaufman AJ, Karhu JA, Beukes NJ, Swart QD,
Coetzee LL, Eriksson KA (2001) Chemostratigraphy of the
Paleoproterozoic Duitschland Formation, South Africa:
implication for coupled climate and carbon cycling. Am J
Sci 301:261–285
Bekker A, Karhu JA, Eriksson KA, Kaufman AJ (2003a)
Chemostratigraphy of the Paleoproterozoic carbonate
successions of the Wyoming Craton: tectonic forcing of
biogeochemical change? Precambrian Res 120:279–325
Bekker A, Sial AN, Karhu JA (2003b) Chemostra-
tigraphy of carbonates fromMinas Supergroup, Quadrilatero
Ferrifero (Iron Quadrangle), Brazil: a stratigraphic record of
Early Proterozoic atmospheric, biogeochemical and climatic
change. Am J Sci 330:865–904
Bekker A, Karhu JA, Kaufman AJ (2006) Carbon isotope
record for onset of the Lomagundi carbon isotope excursion
in Great Lakes area, North America. Precambrian Res
148:145–180
Bowring SA, Grotzinger JP (1992) Implications of new
chronostratigraphy for tectonic evolution of Wopmay
orogen, northwest Canadien Shield. Am J Sci 292:1–20
Brass GW (1976) The variation of the marine 87Sr/86Sr
ratio during Phanerozoic time: interpretation using a flux
model. Geochim Cosmochim Acta 40:721–730
Cameron EM (1983) Evidence from early Proterozoic
anhydrite for sulphur isotope partitioning in Precambrian
ocean. Nature 304:54–56
Deb M, Hoefs J, Baumann A (1991) Isotopic composition
of two Precambrian stratiform barite deposits from the
Indian shield. Geochim Cosmochim Acta 55:303–308
Denison RE, Koepnick RB, BurkeWH, Hetherington EA,
Fletcher A (1997) Construction of the Silurian and Devonian
seawater 87Sr/86Sr curve. Chem Geol 140:109–121
Denison RE, Koepnick RB, Burke WH, Hetherington EA
(1998) Construction of the Cambrian and Ordovician seawa-
ter 87Sr/86Sr curve. Chem Geol 152:325–340
Derry LA, Kaufman AJ, Jacobsen SB (1992) Sedimen-
tary cycling and environmental change in the Late Protero-
zoic: evidence from stable and radiogenic isotopes. Geochim
Cosmochim Acta 56:1317–1329
Faure G (1986) Principles of isotope geology, 2nd edn.
Willey, New York, p 589
Faure G, Hurley PM, Powell JK (1965) The isotopic
composition of strontium in surface water from the north
Atlantic Ocean. Geochim Cosmochim Acta 29:209–220
Fietzke J, Liebetrau V, Gunter D, Gurs K, Hametner K,
Zumholz K, Hansteen TH, Eisenhauer A (2008)
An alternative data acquisition and evaluation strategy for
improved isotope ratio precision using LA-MC-ICP-MS
applied to stable and radiogenic strontium isotopes in
carbonates. J Anal Atom Spectrom 23:955–961
Frauenstein F, Veizer J, Beukes N, van Niekerk HS,
Coetzee LL (2009) Transvaal Supergroup carbonates:
implications for Paleoproterozoic d18O and d13C records.
Precambrian Res 175:149–160
Goldberg ED (1963) The oceans as a chemical system. In:
Hill MN (ed) The sea, vol 2. Wiley, New York, pp 3–25
Goldstein SJ, Jacobsen SB (1987) The Nd and Sr isotopic
systematics of river water-dissolved material: implications
for the sources of Nd and Sr in seawater. Chem Geol
(Isot Geosci Sect) 66:245–272
Gorokhov IM, Semikhatov MA, Baskakov AV,
Kutyavin EP, Mel’nikov NN, Sochava AV, Turchenko TL
(1995) Sr isotopic composition in Riphean, Vendian, and
�
Fig. 7.158 (continued) (a) Massive dolostone bed with partially
eroded top overlain by dolarenite-matrix-supported breccia containing
dolostone clasts derived from the underlying bed; an incipient carbon-
ate platform recorded in the middle part of the Seidorechka Sedimen-
tary Formation in the western flank of the Imandra/Varzuga Greenstone
Belt; coin is 1.2 cm. (b) FAR-DEEP Core 3A exhibiting faintly-
bedded, Sr-rich dolostones accumulated in a deep-water marine setting
influenced by glacial conditions in a volcanically active continental rift
setting; the lower part of the Polisarka Sedimentary Formation in the
western Imandra/Varzuga Greenstone Belt. (c) FAR-DEEP Core 6A
exhibiting a pale beige dolarenite with a series of thin travertine crusts
composed of white and pale pink fibrous dolomite; a shallow water,
lacustrine environment in an intracontinental rift setting recorded in the
middle part of the Kuetsj€arvi Sedimentary Formation in the Pechenga
Greenstone Belt. (d) FAR-DEEP Core 4A exhibiting resedimented,
dolostones with lenses and layers of fine-grained ultramafic material
(dark grey), and faint bedding expressed by variation of colour from
white through beige to pale grey; a deep-water marine environment in a
volcanically active continental rift setting recorded in the middle part of
the Umba Sedimentary Formation in the western Imandra/Varzuga
Greenstone Belt. (e) FAR-DEEP Core 11A illustrating brick-coloured
stromatolitic dolostone formed in a subtidal environment in a platform
setting; the upper part of the Tulomozero Formation in the Onega
Basin. (f) FAR-DEEP Core 11A showing variegated dolostone with
disrupted and brecciated beds due to diagenetic mineral growth and
dissolution, primarily sulphates; a supratidal, platformal setting
recorded in the middle part of the Tulomozero Formation in the
Onega Basin. (g) FAR-DEEP Core 10A showing pink, clayey
dolostone with ubiquitous Ca-sulphate nodules partially replaced by
white dolomite and quartz; an evaporitic platformal setting recorded in
the middle part of the Tulomozero Formation in the Onega Basin
10 7.10 Chemical Characteristics of Sediments and Seawater 1465
Lower Cambrian carbonates from Siberia. Stratigr Geol
Correl 3:1–28
Gorokhov IM, Kuznetsov AB, Melezhik VA,
Konstantinova GV, Melnikov NN (1998) Sr isotopic com-
position in the Upper Jatulian (Early Paleoproterozoic)
dolomites of the Tulomozero Formation, southeastern
Karelia. Dokl Earth Sci 360:609–612
Halicz L, Segal I, Fruchter N, Stein M, Lazar B
(2008) Strontium stable isotopes fractionate in the soil
environments? Earth Planet Sci Lett 272:406–411
Halverson GP, Dudas FO, Maloof AS, Bowring SA
(2007) Evolution of 87Sr/86Sr composition of Neoproterozoic
seawater. Palaeogeogr Palaeocl Palaeoecol 256:103–129
Hodell DA, Mueller PA, McKenzie JA, Mead GA (1989)
Strontium isotope stratigraphy and geochemistry of the late
Neogene ocean. Earth Planet Sci Lett 92:165–178
Kamber BS, Webb G (2001) The geochemistry of late
Archean microbial carbonate: implications for ocean chem-
istry and continental erosion history. Geochim Cosmochim
Acta 65:2509–2525
Kaufman AJ, Jacobsen SB, Knoll AH (1993) The
Vendian record of Sr and C isotopic variations in seawater:
implications for tectonics and paleoclimate. Earth Planet Sci
Lett 120:409–430
Koepnick RB, Burke WH, Denison RE, Hetherington
EA, Nelson HF, Otto JB, Waite LE (1985) Construction of
the seawater 87Sr/86Sr curve for the Cenozoic and Creta-
ceous: supporting data. Chem Geol (Isot Geosci Sect)
58:55–81
Krabbenhoft A, Eisenhauer A, Bohm F, Vollstaedt H,
Fietzke J, Liebetrau V, Augustin N, Peucker-Ehrenbrink B,
Muller MN, Horn C, Hansen BT, Notle N, Wallmann K
(2010) Constraining the marine strontium budget with natu-
ral isotope fractionations (87Sr/86Sr*, d88/86Sr) of carbonates,hydrothermal solutions and river waters. Geochim
Cosmochim Acta 74:1097–4109
Kuznetsov AB, Melezhik VA, Gorokhov IM, Melnikov
NN, Fallick AE (2003) Sr isotope composition in Paleopro-
terozoic carbonates extremely enriched in 13C: Kaniapiskau
Supergroup, the Labrador trough of the Canadian shield.
Stratigr Geol Correl 11:209–219
Kuznetsov AB, Ovchinnikova GV, Semikhatov MA,
Gorokhov IM, Kaurova OK, Krupenin MT, Vasil’eva IM,
Gorokhovskii BM, Maslov AV (2008) The Sr isotopic char-
acterization and Pb–Pb age of carbonate rocks from the
Satka formation, the Lower Riphean Burzyan Group of the
southern Urals. Stratigr Geol Correl 16:120–137
Kuznetsov AB, Melezhik VA, Gorokhov IM, Melnikov
NN, Konstantinova GV, Kutyavin EP, Turchenko TL (2010)
Sr isotopic composition of Paleoproterozoic 13C-rich car-
bonate rocks: the Tulomozero Formation, SE Fennoscandian
Shield. Precambrian Res 182:300–312
McArthur JM, Howarth RJ, Bailey TR (2001) Strontium
isotope stratigraphy: LOWESS. Version 3. Best-fit line
to the marine Sr-isotope curve for 0 to 509 Ma and
accompanying look-up table for deriving numerical age. J
Geol 109:155–170
Melezhik VA (2006) Multiple causes of Earth’s earliest
global glaciation. Terra Nova 18:130–137
Melezhik VA, Fallick AE, Kuznetsov AB (2005a)
Palaeoproterozoic, rift-related, 13C-rich, lacustrine
carbonates, NW Russia. Part II: Global isotopic signal
recorded in the lacustrine dolostone. Trans Roy Soc Edinb
Earth Sci 95:423–444
Melezhik VA, Fallick AE, Rychanchik DV, Kuznetsov
AB (2005b) Palaeoproterozoic evaporites in Fennoscandia:
implications for seawater sulphate, d13C excursions and the
rise of atmospheric oxygen. Terra Nova 17:141–148
Melezhik VA, Huhma H, Condon DJ, Fallick AE,
Whitehouse MJ (2007) Temporal constraints on the
Paleoproterozoic Lomagundi-Jatuli carbon isotopic event.
Geology 35:655–658
Mirota MD, Veizer J (1994) Geochemistry of Precam-
brian carbonates: VI. Aphebian Albanel Formations, Que-
bec, Canada. Geochim Cosmochim Acta 58:1735–1745
Nier AO (1938) The isotopic constitution of strontium,
barium, bismuth, thallium andmercury. Phys Rev 5:275–279
Paytan A, Kastner M, Martin EE, Macdougall JD,
Herbert T (1993) Marine barite as a monitor of seawater
strontium isotope composition. Nature 366:445–449
Peterman ZL, Hedge CE, Tourtelot HA (1970) Isotopic
composition of strontium in sea water throughout Phanero-
zoic time. Geochim Cosmochim Acta 34:105–120
Ohno T, Komiya T, Ueno Yu, Hirata T, Maruyama S
(2008) Determination of 88Sr/86Sr mass-dependent isotopic
fractionation and radiogenic isotope variation of 87Sr/86Sr in
the Neoproterozoic Doushantuo Formation. Gondwana Res
14:126–133
Ovchinnikova GV, Kuznetsov AB, Melezhik VA,
Gorokhov IM, Vasil’eva IM, Gorokhovskii BM (2007) Pb-
Pb age of Jatulian carbonate rocks: the Tulomozero Forma-
tion of southeast Karelia. Stratigr Geol Correl 15:359–372
Ray JS, Veizer J, Davis WJ (2003) C, O, Sr and Pb isotope
systematics of carbonate sequences of the Vindhyan Super-
group, India: age, diagenesis, correlations and implications for
global events. Precambrian Res 121:103–140
Rohon M-L, Vialette Y, Clark T, Roger G, Ohnenstetter
D, Vidal Ph (1993) Aphebian mafic-ultramafic magmatism
in the Labrador trough (New Quebec): its age and the nature
of its mantle source. Can J Earth Sci 30:1582–1593
Rueggeberg A, Fietzke J, Liebetrau V, Eisenhauer A,
Dullo WC, Freiwald A (2008) Stable strontium isotope
(d88/86Sr) in cold-water corals – a new proxy for reconstruc-
tion of intermediate ocean water temperatures. Earth Planet
Sci Lett 269:569–574
Schneiderhan EA, Gutzmer J, Strauss H, Mezger K,
Beukes NJ (2006) The chemostratigraphy of a Paleopro-
terozoic MnF- and BIF succession – the Voelwater
1466 Kuznetsov et al.
Subgroup of the Transvaal Supergroup in Griqualand West,
South Africa. S Afr J Geol 109:63–80
Semikhatov MA, Kuznetsov AB, Gorokhov IM,
Konstantinova GV, Melnikov NN, Podkovyrov VN,
Kutyavin EP (2002) Low 87Sr/86Sr ratios in seawater of the
Grenville and post-Grenville time. Determining factors.
Stratigr Geol Correl 10:1–41
Shields GA (2007) A normalized seawater strontium
isotope curve: possible implications for Neoproterozoic-
Cambrian weathering rates and further oxygenation of the
Earth. eEarth 2:35–42
Spooner ETC (1976) The strontium isotopic composition
of seawater, and seawater-oceanic crust interaction. Earth
Planet Sci Lett 31:167–174
Sumner DY, Bowring SA (1996) U-Pb geochronologic
constraints on deposition of the Campbellrand Subgroup,
Transvaal Supergroup, South Africa. Precambrian Res
79:25–35
Tremba EL, Faure G, Katiskatos GC, Sumerson CH
(1975) Strontium-isotopic composition in the Thetys Sea,
Euboea, Greece. Chem Geol 16:109–120
Veizer J, Compston W (1974) 87Sr/86Sr composition of
seawater during the Phanerozoic. Geochim Cosmochim
Acta 38:1461–1484
Veizer J, Compston W (1976) 87Sr/86Sr in Precambrian
carbonates as an index of crustal evolution. Geochim
Cosmochim Acta 40:905–914
Veizer J, Hoefs J, Lowe DR, Thurston PC (1989) Geo-
chemistry of Precambrian carbonates: II. Archean green-
stone belts and Archean seawater. Geochim Cosmochim
Acta 53:859–871
Veizer J, Clayton RN, Hinton RW (1992a) Geochemistry of
Precambrian carbonates: IV. Early Paleoproterozoic (2.25
� 0.25) seawater. Geochim Cosmochim Acta 56:875–885
Veizer J, Plumb KA, Clayton RN, Hinton RW,
Grotzinger JP (1992b) Geochemistry of Precambrian
carbonates: V. Late Paleoproterozoic seawater. Geochim
Cosmochim Acta 56:2487–2501
Veizer J, Ala D, Azmy K, Bruckschen P, Buhl D, Bruhn F,
Carden G, Diener A, Ebneth S, Godderis Y, Jasper T, Korte C,
Pawellek F, PodlahaO, StraussH (1999) 87Sr/86Sr, 18O and 13C
evolution of Phanerozoic seawater. Chem Geol 161:59–88
10 7.10 Chemical Characteristics of Sediments and Seawater 1467
7.10.3 Ca and Mg Isotopes in SedimentaryCarbonates
Juraj Farkas, Ramananda Chakrabarti, Stein B.Jacobsen, Lee R. Kump, and Victor A. Melezhik
Introduction
Calcium (Ca) is the fifth most abundant element in the Earth’s
crust. Due to its numerous isotopes (40Ca, 42Ca, 43Ca, 44Ca, 46Ca
and 48Ca), it represents a potential tracer of geological and
biological processes. Magnesium (Mg), geochemically similar
to Ca, has three naturally occurring isotopes (24Mg, 25Mg
and 26Mg) and is the seventh most abundant element in the
Earth’s crust. The global geochemical cycles of Ca and Mg
are closely linked through processes of continental weathering,
marine carbonate cycle, carbonate burial and hydrothermal
exchange atmid-ocean ridges (Fig. 7.159). As these phenomena
also affect the functioning of the global carbon cycle, the isotope
studies ofCa andMghave implications for studies of theEarth’s
climate and its evolution through time.
Ca and Mg are delivered to the oceans by rivers via
weathering and dissolution of continental crust containing
carbonate and silicate rocks, expressed by an empirical
formula as CaMgCO3 and CaMgSiO3, respectively.
Weathering of the former supplies a dominant portion of
Ca and Mg to the oceans, but it is the dissolution of silicates
that consumes carbon dioxide (CO2) from the atmosphere,
which in turn regulates the Earth’s climate over geological
time (Berner and Berner 1997). The whole process of silicate
weathering and subsequent CO2 sequestration can be
approximated by a simple reaction below:
CaMgSiO3 silicateð Þ þ CO2 atmosphereð Þ! CaMgCO3 calcite=aragonite=dolomiteð Þ þ SiO2
Complementary to the weathering of the continental crust
by rivers is the hydrothermal alteration of oceanic crust by
seawater, which effectively exchanges dissolved Mg in sea-
water for igneous Ca derived mostly from the oceanic
basalts (Kump 2008):
Ca� silicate basaltð Þ þMg2þ seawaterð Þ! Mg� silicate serpentineð Þ þ Ca2þ seawaterð Þ
The net effect of this exchange reaction is a loss of Mg
from seawater that is being balanced by a hydrothermal
flux of Ca to the ocean. Dolomitisation, which releases Ca
from marine carbonates due to diffusional exchange with
Mg, represents yet another important process that couples
the geochemical cycles of Ca and Mg, according to the
reaction:
2CaCO3 calcite=aragoniteð Þ þMg2þ fluidð Þ! CaMg CO3ð Þ2 dolomiteð Þ þ Ca2þ fluidð Þ
Mutual interactions among these geological processes,
and variations in their intensities, control the inventory and
isotope composition of Ca and Mg in the ocean. Accord-
ingly, the oceanic Ca budget can be described quantitatively
by the following mass balance equation:
dMswCa
dt¼ Friv
Ca þ FhydrCa � Fcarb
Ca
where M is given in moles and F in moles per year. The
isotope mass balance of the marine Ca cycle is given by
(Holmden 2009; DePaolo 2004):
dðMswCa d
swCaÞ
dt¼ Friv
CadrivCa þ Fhydr
Ca dhydrCa � FcarbCa ðdswCa þ Dsw�carb
Ca Þ
whereMswCa is the mass of total dissolved Ca in the ocean. Friv
Ca
and FhydrCa are the input fluxes of Ca to the ocean from conti-
nental river runoff and submarine hydrothermal systems,
respectively. FcarbCa is the output flux of Ca from the ocean
due to deposition of marine carbonates. The 44Ca/40Ca iso-
tope ratios (i.e. d44/40Ca values) of the fluxes FrivCa, F
hydrCa , Fcarb
Ca
are, respectively, drivCa, dhydrCa and ðdswCa þ Dsw�carb
Ca Þ. Where dswCais the d44/40Ca value of contemporary seawater and Dsw�carb
Ca
is the average fractionation factor associated with the
removal of Ca from the ocean via carbonate deposition.
Accordingly, the rate of change of the Ca isotope composi-
tion of the ocean (dswCa) is given by:
MswCa
dðdswCaÞdt
¼ FrivCaðdrivCa � dswCaÞ þ Fhydr
Ca ðdhydrCa � dswCaÞ
� FcarbCa Dsw�carb
Ca
In a similar way, the mass and isotope balance of Mg in
the ocean can be approximated by the following equations
(Tipper et al. 2006a, b; Kump 2008):
dMswMg
dt¼ Friv
Mg � FhydrMg � Fdolo
Mg ;
J. Farkas (*)
Department of Geochemistry, Czech Geological Survey, Geologicka 6,
152 00 Prague 5, Prague, Czech Republic
Faculty of Environmental Sciences, Czech University of Life Sciences,
Kamycka 129, Prague 6, 165 21 Suchdol, Czech Republic
1468 J. Farkas et al.
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_10, # Springer-Verlag Berlin Heidelberg 2013
1468
and
dðMswMg d
swMgÞ
dt¼ Friv
MgdrivMg � Fhydr
Mg ðdswMg þ Dsw�basaltMg Þ � Fdolo
Mg
�ðdswMg þ Dsw�doloMg Þ
where MswMg is the mass of dissolved Mg in the ocean, and
FrivMg is the riverine input flux from weathering of Mg-bearing
silicates and carbonates (mostly dolomites). FhydrMg is the
hydrothermal uptake of Mg from the ocean via thermally-
driven circulation of seawater through the newly formed
oceanic crust. FdoloMg represents the output flux of Mg from
the ocean due to the deposition of dolomite in the marine
environment. The 26Mg/24Mg isotope ratios (i.e. d26Mg) of
the fluxes FrivMg, F
hydrMg , Fdolo
Mg are, respectively, drivMg, ðdswMg þDsw�basaltMg Þ and ðdswMg þ Dsw�dolo
Mg Þ. The dswMg is the average
d26Mg composition of seawater, and the terms Dsw�basaltMg
and Dsw�doloMg are the isotope fractionation factors associated
with the removal of Mg from the ocean via hydrothermal
uptake and dolomite formation, respectively. Accordingly,
the dependence of d26Mg in seawater to the oceanic Mg
fluxes is given by:
MswMg
dðdswMgÞdt
¼ FrivMgðdrivMg � dswMgÞ � Fhydr
Mg Dsw�basaltMg
� FdoloMg Dsw�dolo
Mg
Changing magnitudes of the above input and output fluxes
of Ca and Mg to the ocean can thus alter the chemical and
isotope composition of seawater over geological time. For
instance, an increase in carbonate deposition shifts d26Mg
and d44/40Ca of the ocean towards isotopically heavier values.This is because precipitation of marine carbonates (limestones
and dolostones) preferentially takes up lighter Ca and Mg
isotopes from seawater (Gussone et al. 2003; Tipper et al.
2006a), as both terms, Dsw�carbCa and Dsw�dolo
Mg , are negative. In
other words, calcite precipitated from modern seawater has
about 1 ‰ lighter d44/40Ca than the ambient ocean water
(1.88 � 0.1 ‰), (Gussone et al. 2003, 2005; DePaolo 2004;
Farkas et al. 2007). For Mg, this isotope discrimination is
even larger as the average d26Mg of modern and Holocene
marine dolomites is about 2 ‰ lighter compared to the
d26Mg of present-day seawater (Tipper et al. 2006a).
The isotope fractionation associated with the hydrothermal
uptake of Mg from seawater has yet to be determined (Tipper
et al. 2006a), but it is expected that during the high-
temperature regime at mid-ocean ridges, the isotope fraction-
ation between the fluid and solid phases should be negligible,
i.e. Dsw�basaltMg is close to zero. The Ca isotope signature of
modern hydrothermal fluids overlaps with that of the oceanic
basalts (Amini et al. 2008), and both sources have about 1 ‰lighter d44/40Ca than the present-day seawater. Consequently,an increase in submarine hydrothermal activity has a tendency
to lower d44/40Ca and d26Mg of the ocean (Tipper et al.
2006a), essentially driving its composition towards the aver-
age Ca and Mg isotope composition of the silicate crust.
The d44/40Ca of global river runoff is also about 1 ‰lighter relative to modern seawater (Schmitt et al. 2003;
Tipper et al. 2006b). Therefore, an increase in continental
riverine flux (FrivCa) is expected to lower d
44/40Ca of the ocean,
but this shift in seawater Ca isotope composition would be
relatively short-lived. This is because rivers commonly dis-
charge Ca2+ and CO2�3 ions to the ocean in approximately
equal proportions (Ca2þ/CO2�3 � 0.45; Gaillardet et al.
1999), and since the ocean remains near saturation with
respect to CaCO3, an increase in continental riverine flux
(FrivCa) happens relatively quickly, i.e. within the residence
time of carbon in the ocean (~100,000 years),
counterbalanced by an enhanced precipitation and burial of
marine CaCO3. The latter flux (FcarbCa ) thus efficiently
removes an excess of the river-derived Ca2+ and CO2�3 ions
in the ocean (Kump 2008). Unlike Ca isotopes, which show
a large (~1 ‰) difference between d44/40Ca of rivers and theocean, the d26Mg signature of global river runoff is only
about 0.2 ‰ lighter compared to present-day seawater
(Tipper et al. 2006a). Consequently, the marine Mg isotope
proxy is relatively insensitive to changes in the intensity of
continental river runoff or to the mixing of ocean water with
a typical freshwater. However, during geologic times of
enhanced dolomite production and deposition in continental
margin settings, the d26Mg signature of global river runoff
might have been significantly lighter compared to its modern
value, because larger areas of dolomite rocks with lower
d26Mg (Tipper et al. 2006a) were exposed to erosion and
chemical weathering.
Concentrations and isotope abundances of Ca and Mg in
the modern ocean are globally homogeneous (Hippler et al.
2003), as expected from the much shorter mixing time of the
ocean (~1,000 years) relative to the average residence times
of Ca and Mg in seawater, which are estimated at ~1 and ~14
million years (Ma), respectively (Broecker and Peng 1982;
de Villiers et al. 2005). The circulation and mixing of the
ocean water could, however, be restricted in shallow mar-
ginal seas or coastal evaporite basins, and thus seawater
in these depositional environments might develop unique
d44/40Ca and d26Mg signatures that differ from those of the
open ocean (Holmden 2009). The residence times of Ca and
Mg in these semi-restricted marine environments are
expected to be short, e.g. on the order of ~1,000 years
(Holmden 2009), since the total mass of these elements in
10 7.10 Chemical Characteristics of Sediments and Seawater 1469
a semi-closed basin or sea is considerably smaller compared
to their mass in the global ocean. There are at least two main
processes that could considerably modify the seawater Ca
and Mg isotope composition at a basinal scale: (1) locally
enhanced depositional fluxes of Ca-Mg carbonates and
evaporites, and (2) the input of continentally derived Ca
and Mg from groundwater discharge or river runoff
(Holmden 2009; and results of this study). The fact that the
marine d44/40Ca and d26Mg proxies are sensitive to local
depositional environment and elemental cycling at basinal
levels has implications for the Ca and Mg isotope record of
shallow-water carbonates from the FAR-DEEP cores (e.g.
Holes 10A, 10B and 11A), as these were formed in various
depositional environments, including the semi-restricted
evaporitic settings of coastal sabkha and playa (Melezhik
et al. 1999, 2005). The evidence of extensive and thick
development of sabkha deposits in the Onega Basin is over-
whelming (see Fig. 7.160). Several tens of metres of evapo-
rite and tidal deposits have been intersected by FAR-DEEP
drillholes (see Chaps. 6.31 and 6.3.2). The tidal deposits
comprise variegated, laminated and massive siltstones and
mudstones, flat-laminated stromatolites, rare flat-pebble
conglomerates (Fig. 7.160a, b); all reworked by pervasive
desiccation, growth of Ca-sulphates, and diagenetic
modifications such as disrupted bedding, dissolution and
replacement phenomena (Fig. 7.160a–d).
The evaporite deposits include bedded and nodular anhy-
drite partially pseudomorphed by dolomite and silica; mag-
nesite, dolomite and calcite (Fig. 7.160b–d). Importantly,
detailed mineralogical and geochemical studies of these
evaporite-replacement nodules revealed anhydrite relics in
pseudomorphed sulphate crystals (see the elemental X-ray
maps shown in Fig. 4, page 146, in Melezhik et al. 2005).
Overall, the highly disrupted tidal and evaporite deposits
exhibit many, if not all, features of modern sabkha deposits
reported in the literature (e.g. Demicco and Hardie 1994).
Playa deposits, which are interspersed with sabkha
evaporites, are less abundant in the Onega Basin, and occur
as a-few-metre-thick units of dark grey, brown and red
massive mudstones with abundant casts of halite and proba-
bly other soluble salts (Fig. 7.160e–g).
Ca and Mg Isotope Composition ofPalaeoproterozoic 13C-Rich Carbonates fromthe Tulomozero Formation: Implications forthe Depositional Environment and the Originof Dolomites
Here we present Ca and Mg isotopic compositions of bulk
dolomites from the Tulomozero Formation (TF), which was
deposited ~2100 Ma ago in the Onega Basin along the
southeast Fennoscandian Shield (Melezhik et al. 1999;
Ovchinnikova et al. 2007). The TF rock record obtained
from drillholes 5177 and 4699 can be directly correlated
with sedimentary carbonates recovered by the FAR-DEEP
Holes 10A, 10B and 11A. The TF carbonates preserve the
first major positive excursion in the marine carbon isotope
(d13C) record, the Lomagundi-Jatuli Isotope Event (Baker
and Fallick 1989; Karhu 1993; Melezhik et al. 2005).
The Ca and Mg isotope ratios were measured in near-
stoichiometric dolomites that yielded an average molar Mg/
Ca ratio of ~1.00 � 0.11 (2sd), (Fig. 7.161), and also rela-
tively low Mn/Sr ranging from 0.5 to 6 (Melezhik et al.
1999). Based on these elemental compositions and the sedi-
mentological features shown in Fig. 7.160, the TF
dolostones are considered to be primary or synsedimentary
precipitates (Melezhik et al. 2005).
The studied dolomites cover a wide range of palaeodepo-
sitional environments, ranging from marine intertidal to
supratidal sabkha and playa (Melezhik et al. 2000). The
d44/40Ca of the TF dolostones covaries with their d13C(Fig. 7.162), and these isotope proxies define a trend line
with a negative slope and moderate, yet statistically signifi-
cant, correlation (R2 ¼ 0.6, r ¼ �0.78, p < 0.001). Results
indicate that both d44/40Ca and d13C isotope tracers are
strongly dependent on a local depositional environment,
which in turn changes with stratigraphic depth (Figs. 7.162
and 7.163). The d44/40Ca (NIST) values of the TF dolostones
range from ~1.1 ‰ to 1.7 ‰, and the data show no radio-
genic 40Ca excesses larger than the analytical uncertainty of
� 0.12 ‰. Consequently, the observed d44/40Ca variations
(Figs. 7.162 and 7.163) are attributable to mass-dependent
isotope fractionation and not to the radiogenic 40Ca effects.
There are several plausible explanations for the observed
variation in d44/40Ca and d26Mg of the TF dolostones, and
these alternative scenarios and models of dolomitisation are
discussed below. Note, however, that due to the limited
nature of Ca and Mg data sets, the models presented are
based on qualitative assumptions rather than quantitative
data. In addition, due to a better understanding of the marine
Ca isotope system compared to the Mg system, which is still
in its infancy (Eisenhauer et al. 2009), the emphasis is given
to the interpretation of Ca isotope data and the Mg results are
discussed only marginally.
Seawater Evaporation and Locally EnhancedDeposition of Calcium Sulphates andCarbonates
This scenario assumes that the evaporation of seawater and
contemporaneous precipitation of marine carbonates and
calcium sulphates (gypsum, anhydrite) are the primary
1470 J. Farkas et al.
drivers of the heavy Ca and Mg isotope enrichments found in
the TF dolomites from the intertidal zone. Based on this
facies-dependent model, the lighter Ca isotope signatures
found in the supratidal zone dolomites are interpreted to be
influenced by an input of continentally derived Ca sources
from groundwater discharge and/or river runoff that both
tend to be isotopically lighter than seawater (DePaolo
2004; Heuser et al. 2005; Jacobson and Holmden 2008).
All intertidal dolomites with abundant evaporite relics,
which occur as dolomite-pseudomorphed nodules or crystals
after the replacement of calcium sulphates, yielded heavy
d44/40Ca signatures of 1.4–1.7 ‰ (Figs. 7.162 and 7.163, see
members D to H). This enrichment in the heavier 44Ca could
be explained by Rayleigh fractionation of Ca isotopes from
progressively evaporated seawater, in a closed system, due
to an ongoing precipitation of calcium sulphates and
carbonates. These Ca-bearing minerals preferentially take
up light 40Ca isotopes (Gussone et al. 2003, 2005; DePaolo
2004; Hensley 2006). Therefore, in a semi-restricted setting
of the Onega Basin the formation of CaCO3 and CaSO4
(alternatively CaSO4.2H2O) would drive the d44/40Ca of
the seawater or brine, and consequentially the dolomite
that formed from it, to isotopically heavier values. More-
over, the formation of dolomite is further enhanced by gyp-
sum or anhydrite precipitation as this process lowers the
concentration of dissolved sulphate, which is considered as
an inhibitor of dolomitisation especially in relatively low-
sulphate aqueous solution (Machel 2004), such as the
Palaeoproterozoic ocean water (Grotzinger 1989; but see
also discussion in Melezhik et al. 2005).
In contrast, the sabkha/playa and supratidal zone
dolomites (Figs. 7.162 and 7.163, members A and B) show
much lighter d44/40Ca values ranging from 1.1 ‰ to 1.3 ‰,
possibly reflecting a contribution of isotopically lighter Ca
from continental river runoff and/or groundwater, discharge
of which both have an estimated present-day d44/40Ca signa-ture close to 0.9 ‰ (Tipper et al. 2006b; Jacobson and
Holmden 2008). However, if one considers an order of
magnitude lower concentration of Ca in rivers compared to
ocean water (Holmden 2009), and also the arid conditions of
coastal sabkha and playa (Fig. 7.160), then any significant
portion of the continentally derived Ca must have been
supplied by a subsurface groundwater flux with a possible
but sporadic contribution from river runoff or aeolian depo-
sition. The influence of the continental weathering flux on
the Ca and Sr isotope composition of supratidal sabkha/
playa dolomites seems to be further supported by their
more radiogenic 87Sr/86Sr values, in excess of 0.7085
(Gorokhov et al. 1998); since Sr derived from the continental
sources is commonly enriched in radiogenic 87Sr. On the
other hand, all intertidal TF dolostones yielded much less
radiogenic 87Sr/86Sr, between ~0.7035 and 0.7060,
indicating a greater contribution from seawater-derived
sources (Gorokhov et al. 1998; Melezhik et al. 2005).
As to the Mg isotope variations, d26Mg of the TF
dolostones scatters around �0.8 � 0.3 ‰ and data show
no resolvable trend with stratigraphic depth or changing
depositional environment (Fig. 7.163). However, dolostones
collected in the vicinity of a major magnesite layer
(Fig. 7.163; member D in a depth of ~600 m) have distinctly
heavier d26Mg than the rest of the samples. More impor-
tantly, the Palaeoproterozoic TF dolostones have signifi-
cantly and systematically heavier d26Mg and d44/40Ca than
other published data from sedimentary dolomites ranging in
age form Neoproterozoic to Mesozoic (Fig. 7.164). Note
also that the average d26Mg of the TF dolostones overlaps
with that of present-day seawater (�0.8 � 0.1 ‰), and d44/40Ca of the intertidal dolostones (up to 1.7) is only slightly
lighter compared to modern ocean water (d44/40Ca ¼ 1.88
� 0.1 ‰), (Hippler et al. 2003).
Based on the presented facies-dependent model, the44Ca and 26Mg enrichments observed in the TF dolostones
could be explained by an enhanced local depositional flux
of gypsum, calcite and/or dolomite in the semi-restricted
evaporative settings of the Onega Basin, compared to the
fluxes that were operating in the contemporary global
ocean. An alternative interpretation could be that the
heavy d44/40Ca and d26Mg of the TF dolostones are of
global significance, which would indicate that the rates of
carbonate and dolomite accumulation in the global ocean
were significantly higher in the Palaeoproterozoic and
decreased gradually over time (Fig. 7.164). If correct, this
scenario would require global and synchronous distribution
of 44Ca- and 26Mg-rich carbonates in various basins of
Palaeoproterozoic age. However, our preliminary data
on Palaeoproterozoic limestones from the Transvaal
Supegroup in South Africa, i.e. the Duitschland Formation
dated at ~2350 Ma (Farkas et al. 2008; Frauenstein et al.
2009), yielded much lighter d44/40Ca values that are in the
range of Neoproterozoic and Palaeozoic data shown in
Fig. 7.164.
Taken as a whole, the evidence from Ca and Mg isotopes
and abundant relicts after evaporitic minerals (cf. Fig. 7.160)
suggest that the bulk chemical and isotopic composition of
the TF dolostones has been affected by local-scale processes,
such as seawater evaporation and precipitation of Ca-
carbonates and sulphates, which in turn may have signifi-
cantly overprinted the global seawater signal. Therefore,
future isotope studies on carbonates from the Onega Basin,
i.e. on samples recovered by the FAR-DEEP Holes 10A,
10B and 11A, should be performed with caution when
attempting to interpret the results solely in terms of a global
seawater signal or large-scale parameters of the contempo-
rary ocean-atmosphere system.
10 7.10 Chemical Characteristics of Sediments and Seawater 1471
Variable Rates of Dolomite Formation andOscillations in Mineralogy and PrecipitationRate of Marine Carbonates
An alternative explanation for the variation of d44/40Ca in theTF dolostones involves different rates of dolomite formation
in the Onega Basin, as dolomitisation of marine carbonates
tends to release light Ca isotopes to seawater through a
diffusional exchange with Mg (Heuser et al. 2005). Accord-
ingly, higher rates of dolomitisation would shift the d44/40Caof seawater or basinal water towards lighter values, whereas
lower rates would drive it towards relatively heavier values.
Hence the observed differences between d44/40Ca of the
supratidal and intertidal samples (Fig. 7.162) may also
reflect changing dolomitisation rates in the basin through
time; i.e. as a function of stratigraphic depth (Fig. 7.163).
One obvious drawback of this explanation, however, is
that the variations in d44/40Ca of the TF dolostones are not
accompanied by coeval changes in their d26Mg (Fig. 7.163),
which would be expected considering that dolomite forma-
tion is also a sink for light Mg isotopes (Tipper et al. 2006a),
and during dolomitisation Ca and Mg ions are exchanged
quantitatively.
Another plausible interpretation for the Ca isotope
variations in the TF dolostones considers the role of
changing precipitation rate during carbonate formation
(Tang et al. 2008). Accordingly, marine carbonates formed
at high precipitation rates should yield lower d44/40Ca sig-
nature than that of a precipitating fluid (i.e. seawater or
basinal water), due presumably to limited time for Ca
isotope equilibration between the fluid and carbonate min-
eral. In contrast, carbonates formed at low precipitation
rates have more time to re-equilibrate with the dissolved
Ca in seawater, and thus their d44/40Ca tends to be heavier,
i.e. closer to the Ca isotope composition of seawater
(Fantle and DePaolo 2007; Tang et al. 2008). Hence,
changes in the rate of carbonate precipitation in the
Onega Basin could, in theory, also explain the observed
variation in d44/40Ca of the TF dolostones. However, as
shown by Tang et al. (2008), changes in precipitation rates
affect not only the partitioning of Ca isotopes but also the
Sr content in the precipitating carbonate mineral. So if the
kinetics of carbonate precipitation indeed controlled the Ca
isotope composition of the TF dolostones, then one should
expect to observe a negative correlation between their d44/40Ca values and Sr concentrations (i.e. Sr/Ca ratios), as
shown in previous studies (Steuber and Buhl 2006; Tang
et al. 2008). Our data from the drillhole 5177 show no
apparent correlation between d44/40Ca and Sr/Ca values in
the TF dolostones (n ¼ 15), (Fig. 7.161, the right panel),
Fig. 7.159 The major geological processes controlling the chemical cycling of calcium (Ca) and magnesium (Mg) in terrestrial environments,
and its links to the global carbon (C) cycle (Modified after Elderfield 2010)
1472 J. Farkas et al.
Fig. 7.160 Sedimentological features of sabkha and playa deposits
exemplified by cores retrieved from FAR-DEEP Holes 10A and 10B.
Sabkha deposits: (a) Dark brown mudstone and pale pink, dolomite-
cemented sandstone overlain by variegated mudstone-siltstone tidal
couples disrupted by desiccation, and diagenetic growth of Ca-
sulphates partially replaced by white dolomite. (b) Pale pink,dolomite-cemented sandstones with lenses of flat-pebble conglomerate
overlain by dark grey and dark brown siltstone-mudstone tidal couples
capped by mudstone layer completely disrupted by diagenetic growth
of Ca-sulphates with nodular anhydrite at the base. (c) Dark grey
mudstone layer with abundant nodular anhydrite partially replaced by
red, haematite-stained dolomite and quartz. (d) Variegated tidal sand-
stone-siltstone-mudstone whose primary bedding was completely
obliterated by diagenetic growth of Ca-sulphates. Playa deposits: (e)Brick-red, (f) brown, and (g) dark grey mudstone with abundant halite
casts. Core diameter here and in Fig. 7.166 is 50 mm (Photographs by
Victor Melezhik)
10 7.10 Chemical Characteristics of Sediments and Seawater 1473
but samples (n ¼ 3) from the drillhole 4699 have system-
atically higher Sr/Ca and their d44/40Ca values are consis-
tently lower. Therefore, it is possible that a part of the Ca
isotope variation observed the TF dolostones could indeed
be explained via different precipitation rates at these two
sites (i.e. 5177 versus 4699). Alternatively, the higher
Sr/Ca and lower d44/40Ca of dolostones from the drillhole
4699 could be related to the mineralogy of their carbonate
precursor. That being said, the precursor of these
dolomites had to be considerably enriched in aragonite,
which is a polymorph that tends to have lower d44/40Caand higher Sr content relative to co-existing calcite and/or
dolomite phases (Gussone et al. 2005; Heuser et al. 2005).
The Role of Microbial Processes on Ca and CIsotopic Composition of the TF Dolostones
Taking into account the abundant evidence for microbial
activity in the sedimentary record of the TF (i.e. stromato-
litic structures shown in Fig. 7.166e, f), one should also
consider the role of micro-organisms like cyanobacteria or
sulphur-reducing bacteria in the formation of dolomites and
the associated fractionation of stable C and Ca isotopes
(Fig. 7.163). Previous studies (Melezhik et al. 2000, 2005)
suggested that the marine d13C record of the TF dolomites
has been amplified by local environmental factors such
as seawater evaporation and expansion of microbial
communities in partly restricted shallow-water settings of
the Onega Basin. Specifically, the anomalously high
13C-enrichment of the TF dolomites (d13C up to +16.8 ‰;
Fig. 7.163) has been attributed to “enhanced uptake of 12C
by cyanobacteria and penecontemporaneous oxidation of
organic material in cyanobacterial mats with the production
and consequent loss of CO2 in subaerial and shallow-water
environments” (p. 147, Melezhik et al. 2005). Although the
above biological processes could explain the 13C-enrichment
of the TF dolomites and their facies-dependent d13C trend
(Fig. 7.162), this mechanism is unlikely to explain the con-
temporaneous change in the dolomite’s Ca isotope record
or the correlation pattern between d44/40Ca and d13C data
(Figs. 7.162 and 7.163). This is because the biologically
driven flux of gaseous carbon species (CO2 and CH4),
which carries the ‘light’ isotope signatures from the basin
while leaving behind a ‘heavy’ carbonate reservoir, is not
expected to have a significant impact on the marine Ca
isotope record as Ca isotopes, unlike C isotopes, are not
cycled through a gaseous phase at near-surface conditions.
Thus, it appears unlikely that the observed covariance
between d44/40Ca and d13C records of the TF dolostones is
caused by a common underlying mechanism related to the
biological activity of stromatolitic communities. Hence, it
seems more plausible that the facies-dependent correlation
between d44/40Ca and d13C trends (Figs. 7.162 and 7.163) is
rather caused by temporal changes in physical parameters of
the depositional environment. Nevertheless, micro-
organisms might still have played an important role in the
formation of the TF dolomites by providing suitable nucle-
ation sites or by optimising the physico-chemical conditions
for dolomite precipitation (Warthmann et al. 2000).
Fig. 7.161 Left Panel: Cross-plot showing d44/40Ca versus molar
Mg/Ca ratios in the TF dolomites from drillholes 5177 (blue) and
4699 (orange rectangles). The vertical red line illustrates the ideal
stoichiometric Mg/Ca ratio of dolomite that is equal to unity. RightPanel: d44/40Ca values versus Sr/Ca ratios (mmol/mol) in the TF
dolomites
1474 J. Farkas et al.
New Isotopic Clues to an Old Standing ‘DolomiteProblem’ and Future Prospects for theFAR-DEEP Cores
The origin of sedimentary dolomites is a complex and long-
standing question that has puzzled earth scientists for more
than a century (Holland and Zimmermann 2000; Arvidson
and Mackenzie 1999; Von G€umbel 1857; Dolomieu 1791).
The ‘dolomite problem’ has numerous aspects but the major
issues are: (1) dolomite is rare in recent and Holocene
marine sediments but abundant in the older rock record;
(2) dolomite can be formed in many depositional and diage-
netic environments; (3) geochemical data frequently allow
more than one genetic interpretation; and (4) laboratory
Fig. 7.162 d44/40Ca and d13C of the TF dolostones from drillhole 5177
(Melezhik et al. 1999). Data are symbol- and colour-coded based on
their respective depositional environments. d44/40Ca represents per mil
deviations (‰) in the 44Ca/40Ca ratio of the sample relative to NIST
915a standard; where d44/40Ca ¼ (44Ca/40Casample/44Ca/40CaNIST �1)
*103. The external precision (2sd) of d44/40Ca values, based on the
long-term reproducibility of NIST and seawater (IAPSO) standards, is
estimated at �0.12 ‰. Note that selected samples of the TF dolomites
(n ¼ 10) were also analysed for their radiogenic 40Ca enrichments
(eCa) using the 42Ca/44Ca ratio of 0.31221 for the correction of instru-
mental fractionation (where eCa ¼ [40Ca/42Casample/40Ca/42CaNIST915a
�1]* 104). However, results showed no excess of the radiogenic 40Ca
larger than the analytical uncertainty of 1.5 e�unit
10 7.10 Chemical Characteristics of Sediments and Seawater 1475
experiments have not yet been successful to precipitate
inorganic stoichiometric dolomite at low temperatures
(Holland and Zimmermann 2000; Machel 2004).
Here we discuss the potential of Ca isotopes for tracing
the origin and source of dolomitising fluids, which in turn
might help us to better understand the underlying processes
responsible for the formation of massive dolomites in the
geological record. Specifically, we argue that Ca isotope
data, in combination with Mn/Sr ratios, can be used to
differentiate between various genetic models of dolomite
formation such as evaporative reflux, mixing zone, and/or
burial models (Machel 2004; Holmden 2009). The reasoning
behind the application of Ca isotopes to the ‘dolomite
problem’ is related to the kinetics of dolomite formation,
which typically requires low precipitation rates and large
water to rock ratios. Recent studies suggest that under
these conditions the Ca isotope fractionation between a
precipitating carbonate (i.e. calcite) and an ambient fluid
should be negligible (Fantle and DePaolo 2007; Jacobson
and Holmden 2008; Tang et al. 2008). Assuming that a
similar kinetics-dependent fractionation is also valid for
dolomite, then the Ca isotope composition of sedimentary
dolostones should closely reflect that of the dolomitising
fluid (Holmden 2009). As to the Mn/Sr ratio, this is com-
monly used as an index of diagenetic alteration of marine
carbonates because Mn and Sr are elements with greatly
different partitioning coefficients (Kd) between solid carbon-
ate phase and a fluid (e.g. Kd(Sr�Ca) ~ 30 and Kd
(Mn�Ca)
~ 0.05; Brand and Veizer 1980). Accordingly, during pro-
gressive alteration of marine carbonates by meteoric waters
the Mn concentration in limestones/dolostones increases and
the Sr concentration decreases, driving their Mn/Sr to higher
values (Veizer 1983). Consequently, the Mn/Sr ratio can
be used as a geochemical index to distinguish primary
Fig. 7.163 Carbon (d13C), calcium (d44/40Ca) and magnesium
(d26Mg) isotope profiles of dolostones from the Tulomozero Formation
sampled by drillholes 5177 and 4699 (Melezhik et al. 1999), which are
directly correlated with the FAR-DEEP Holes 10A, 10B and 11A. The
lithological profile and d13C data are fromMelezhik et al. (1999, 2005).
The d44/40Ca (NIST) values were determined by TIMS (GV IsoProbe-
T) using a 43Ca–48Ca double-spike (Farkas et al. 2007, Huang et al.
2010). The d26Mg values (reported with respect to DSM3) were
measured by MC-ICP-MS (GV IsoProbe-P) using a sample-standard
bracketing technique (Chakrabarti and Jacobsen 2010). Powdered
samples of dolostones were dissolved in 1N HCl at room temperature
over 24 h, and Ca and Mg fractions were separated from the carbonate
matrix using the conventional cation exchange chromatography with
BioRad AG50W-X12 resin (Huang et al. 2010; Chakrabarti and
Jacobsen 2010)
1476 J. Farkas et al.
syndepositional dolomites from those formed by secondary
post-depositional processes.
In Fig. 7.165, we plot Ca isotope compositions and Mn/Sr
ratios of sedimentary dolostones collected in the Onega
(Northwest Russia), Transvaal (South Africa) and Williston
(Mid-North America) Basins. These data indicate that
samples from the Onega Basin fall within a range of
syndepositional or early diagenetic dolomites that are
defined by Mn/Sr of less than 10 (Melezhik et al. 2000). In
general, low Mn/Sr and heavy d44/40Ca are indicative of
seawater-derived sources (Fig. 7.165) and point towards
the evaporative-brine models of dolomite formation (Machel
2004), involving either syndepositional or early diagenetic
processes. The former consider evaporative pumping of
seawater upward through a sedimentary column, causing
precipitation of calcium sulphate and carbonate
accompanied by an increase in Mg/Ca of remaining fluids,
which leads to precipitation of syndepositional (i.e. primary)
dolomite or even magnesite. On the other hand, early diage-
netic processes are associated with a dolomitisation model
driven by the seepage of dense Mg-rich brines, produced by
seawater evaporation, into underlying shallow-water marine
limestones, which are thus being progressively dolomitised
(i.e. a brine reflux model; Machel 2004). Both evaporative-
brine models, syndepositional and early diagenetic, are
applicable to the TF dolostones and could explain their
relatively heavy d44/40Ca and low Mn/Sr values
(Fig. 7.165). According to the discussed facies-dependent
model, the observed difference between the Ca isotope com-
position of intertidal and supratidal TF dolostones
(Fig. 7.165) could be explained by a two-component mixing
between (1) isotopically heavy seawater or evaporated brine
and (2) a much lighter Ca derived from continental sources
such as groundwater discharge or river runoff.
Low Mn/Sr ratios are also found in dolomites from the
Williston Basin (Fig. 7.165; Zenger 1996), but these
Palaeozoic dolostones were interpreted to be the product of
early diagenetic processes rather than being primary marine
precipitates (Holmden 2009). Such conclusions are based on
the fact that their light Ca isotope signatures match those of
the coeval Palaeozoic marine limestones (Fig. 7.165),
suggesting that a source of dolomitising fluid was derived
from the latter rather than from the contemporary ocean
water (Holmden 2009), which has a d44/40Ca value estimated
at ~1.2 � 0.2 ‰ (Farkas et al. 2007).
Unlike dolostones from the Onega and Williston Basins,
those from the Transvaal Basin have much higher Mn/Sr, in
excess of 20 (Fig. 7.165), suggesting involvement of later
postdepositional recrystallisation driven by interactions with
meteoric or basinal fluids (Frauenstein et al. 2009). Interest-
ingly, the d44/40Ca signature of these ‘late-diagenetic’
dolostones from the Transvaal Basin falls within a range of
Fig. 7.164 Left Panel: Histogram showing d26Mg variations in the
Palaeoproterozoic TF dolostones from the Onega Basin (this study),
along with other published data (all re-normalised to DSM3) from
sedimentary dolomites of Neoproterozoic-Cambrian (Galy et al.
2002), Palaeozoic (Chang et al. 2003) and Mesozoic ages (Galy et al.
2002). Right Panel: d44/40Ca variations in the TF dolostones compared
to published Ca isotope data (re-normalised to NIST915a) from
Neoproterozoic (Kasemann et al. 2005) and Palaeozoic marine
dolomites; the latter of Ordovician (Holmden 2009) and Carboniferous
ages (Steuber and Buhl 2006; Jacobson and Holmden 2008)
10 7.10 Chemical Characteristics of Sediments and Seawater 1477
present-day fresh water or groundwater (Fig. 7.165), which
provides additional support for their post-depositional, or
meteoric, origin.
Hence, a combined approach using Ca isotopes and Mn/Sr
ratios might be utilized to constrain a source of the
dolomitising fluid, and thus to differentiate between various
hydrological models of dolomite formation. Specifically, the
‘primary’ dolomites formed via syndepositional evaporative
pumping and/or brine refluxmechanism should yield lowMn/
Sr and heavy Ca isotope signatures, whereas those formed in
freshwater-seawater mixing zones are expected to have rela-
tively lighter Ca isotope compositions (Fig. 7.165). In con-
trast, ‘secondary’ dolomites formed through dolomitisation of
marine limestones in meteoric and/or burial realms should
have much higher Mn/Sr ratios (>10) and their Ca isotope
signature would be dependent on that of meteoric water or
local basinal fluids.
As to future research prospects for the FAR-DEEP
cores, a detailed study of Ca and Mg isotope geochemistry
in dolostones accumulated in different basinal and tectonic
settings (Fig. 7.166), as well as in various carbonate
components (primary matrix, later cements, pseudo-
morphs, and possibly fluid inclusions), will allow us to
unravel a complex history of dolomitisation and carbonate
diagenesis in the studied Palaeoproterozoic basins of the
Fennoscandian Shield. This information will be useful for
distinguishing the primary syndepositional marine
carbonates from the secondary replacement phases. Since
the former holds clues for the reconstruction of physical
and chemical properties of the contemporary ocean-
atmosphere system, further Ca and Mg isotope studies
are of interest and would be a valuable addition to the
FAR-DEEP project and its aspiration to understand the
causes and timing of the rise of atmospheric oxygen.
Finally, the d44/40Ca proxy can be used as a quantitative
palaeoenvironmental tracer because of its sensitivity to
seawater-freshwater mixing and the evaporation of ocean
water in restricted coastal systems.
Fig. 7.165 Cross-plot of d44/40Ca and Mn/Sr values in sedimentary
dolostones from the Onega Basin (this study), the Transvaal Basin
(Farkas et al. 2008) and the Williston Basin (Ca isotope data from
Holmden 2009; and Mn/Sr from Zenger 1996). The grey-lined rectan-
gle illustrates a range of d44/40Ca and Mn/Sr values found in dolostones
and dolomitised limestones, from the Williston Basin, whose Mg/Ca
ratios (mol/mol) vary from 0.4 to 0.9 (Holmden 2009; Zenger 1996).
The vertical dotted line indicates a threshold for Mn/Sr (~10) that
separates syndepositional or early diagenetic dolomites from those
formed by late-diagenetic processes (Melezhik et al. 2000). Horizontal
dashed lineswith error bars show the representative d44/40Ca signaturesof selected Ca sources (Tipper et al. 2006b, Jacobson and Holmden
2008; Farkas et al. 2007)
1478 J. Farkas et al.
Fig. 7.166 The FAR-DEEP cores exemplifying dolostones which
have accumulated in different depositional and tectonic settings; core
width is 5 cm. These carbonate rocks have a great potential to address
the “dolomite problem”, as well as other environmental aspects of
Palaeoproterozoic chemical sediments such as the first perturbation in
the global carbon cycle. (a) Pale blue and pale pink, Sr-rich limestones
with calcareous shale interlayers associated with the Huronian-age
glacial period; the Polisarka Sedimentary Formation, Imandra/Varzuga
Greenstone Belt, Hole 3A. (b) Intraplate rift-bound, lacustrine, 13C-rich
dolostone associated with the Lomagundi-Jatuli Isotpic Event; the
Kuetsj€arvi Sedimentary Formation, Pechenga Greenstone Belt, Hole
5A. (c) Resedimented deep-water ramp, 13C-rich dolostone associated
with the Lomagundi-Jatuli Isotopic Event; the Umba Sedimentary
Formation, Imandra/Varzuga Greenstone Belt, Hole 4A. (d)
Interbedded Corg-rich sale (black) and resedimented shale-draped,13C-rich dolostones (pale grey) associated with the Shunga Event;
volcanically-active continental rift on a continental margin; the
Zaonega Formation, Onega Basin, Hole 12B
10 7.10 Chemical Characteristics of Sediments and Seawater 1479
Acknowledgements
Financial support provided by the Canadian Institute for
Advanced Research (CIFAR-ESEP), the Earth System
Evolution Program, the GACR grant (P210/12/P631),
and a travel grant from Harvard University Department
of Earth and Planetary Sciences, are greatly acknowl-
edged. We also thank Melanie Mesli (Geological Survey
of Norway) for technical assistance with the core
sampling.
References
Amini M, Eisenhauer A, B€ohm F, Fietzke J, BachW, Garbe-
Sch€onberg D, Rosner M, Bock B, Lackschewitz KS, Hauff F
(2008) Calcium isotope (d44/40Ca) fractionation along
hydrothermal pathways, Logatchev field (Mid-Atlantic
Ridge, 14�45´N). Geochim Cosmochim Acta 72:4107–4122
Arvidson RS, MacKenzie FT (1999) The dolomite prob-
lem: control of precipitation kinetics by temperature and
saturation state. Am J Sci 299:257–288
Fig. 7.166 (continued) (e) Silicified stromatolitic dolostone
accumulated in a marine-influenced continental rift setting in the after-
math of the Lomagundi-Jatuli Isotopic Event; the Kolasjoki Sedimen-
tary Formation, Pechenga Greenstone Belt, Hole 8B. (f) Intertidal,
stromatolitic, 13C-rich dolostone accumulated in a carbonate platform;
the Tulomozero Formation, Onega Basin, Hole 11A. (g) Shale and
dolostone clasts filling cavity in karstified, 13C-rich dolostone; an
emerged carbonate platform; the Tulomozero Formation, Onega
Basin, Hole 10A. (h) Dissolution collapse breccia containing greyand pink shale fragments cemented by 13C-rich dolostone replacing
sulphates; the Tulomozero Formation, Onega Basin, Hole 10B
(Photographs by Victor Melezhik)
1480 J. Farkas et al.
Baker AJ, Fallick AE (1989) Evidence from Lewisian
limestone for isotopically heavy carbon in two-thousand-
million-year-old sea water. Nature 337:352–354
Berner RA, Berner EK (1997) Silicate weathering and
climate. In: Ruddiman WF (ed) Tectonic uplift and climate
change. Plenum Press, New York, pp 313–327
Brand U, Veizer J (1980) Chemical diagenesis of a mul-
ticomponent system: trace elements. J Sediment Petrol
50:1219–1236
Broecker WS, Peng TH (1982) Tracers in the sea. LDGO
Press, Palisades, p 690
Chakrabarti R, Jacobsen SB (2010) The isotopic compo-
sition of magnesium in the inner Solar System. Earth Planet
Sci Lett 293:349–358
Chang VTC, Makishima A, Belshaw NS, O’Nions RK
(2003) Purification of Mg from low-Mg biogenic carbonates
for isotope ratio determination using multiple collector ICP-
MS. J Anal Atom Spectrom 18:296–301
Demicco RV, Hardie LA (1994) Sedimentary structures
and early diagenetic features of shallow marine carbonate
deposits, vol 1, Atlas Series. Society for Sedimentary Geol-
ogy, Tulsa, p 265
de Dolomieu DG (1791) Sur un de pierres tres-peu
effervescentes avec les acides of phosphorescentes par la
collision. J Phys 39:3–10
de Villiers S, Dickson JAD, Ellam RM (2005) The com-
position of the continental riverine weathering flux deduced
from seawater Mg isotopes. Chem Geol 216:133–142
DePaolo DJ (2004) Calcium isotopic variations produced
by biological, kinetic, radiogenic and nucleosynthetic pro-
cesses. In: Johnson CM, Beard BL, Albarede F (eds) Geo-
chemistry of non-traditional stable isotopes, vol 55, Reviews
in Mineralogy and Geochemistry. Mineralogical Society of
America, Washington, DC, pp 255–288
Eisenhauer A, Kisak€urek B, B€ohm F (2009) Marine cal-
cification: an alkali earth metal isotope perspective.
Elements 5:365–368
Elderfield H (2010) Seawater chemistry and climate.
Science 327:1092–1093
Fantle MS, DePaolo DJ (2007) Ca isotopes in carbonate
sediments and pore fluid from ODP Site 807A: the Ca (aq)-
calcite equilibrium fractionation factor and calcite recrystal-
lization in Pleistocene sediments. Geochim Cosmochim
Acta 71:2524–2546
Farkas J, B€ohm F, Wallmann K, Blenkinsop J,
Eisenhauer A, van Geldern R, Munnecke A, Voigt S, Veizer
J (2007) Calcium isotope record of Phanerozoic oceans:
implications for chemical evolution of seawater and its
causative mechanisms. Geochim Cosmochim Acta
71:5117–5134
Farkas J, Jacobsen SB, Frauenstein F, Veizer J (2008)
Calcium isotope systematics of diagenetically altered
carbonates: example from the Proterozoic carbonates of
Transvaal Supergroup, South Africa. Eos Transactions, AGU
89 (53), Fall Meeting Supplement, Abstract B11C-0379
Frauenstein F, Veizer J, Beukes N, Van Niekerk HS,
Coetzee LL (2009) Transvaal Supergroup Carbonates:
implications for Paleoproterozoic d18O and d13C records.
Precambrian Res 175:149–160
Gaillardet J, Dupre B, Louvat P, Allegre CJ (1999)
Global silicate weathering and CO2 consumption rates
deduced from the chemistry of large rivers. Chem Geol
159:3–30
Galy A, Bar-Matthews M, Halicz L, O’Nions RK (2002)
Mg isotope composition of carbonate: insight from
speleothem formation. Earth Planet Sci Lett 201:105–115
Gorokhov IM, Kusnetsov AB, Melezhik VA,
Konstantinova GV, Melnikov NN (1998) Strontium isotope
ratios of the Paleoproterozoic, upper Jatulian, Tulomozero
dolostones from SE Karelia. Trans Russ Acad Sci
360:533–536
Grotzinger JP (1989) Facies and evolution of Precam-
brian carbonate depositional systems; emergence of the
modern platform archetype. In: Crevello PD, Wilson JJ,
Sarg JF, Read F (eds) Controls on carbonate platform and
basin development, vol 44, Special Publication Of the Soci-
ety of Economic Paleontologists and Mineralogists. Society
of Economic Paleontologists and Mineralogists, Tulsa, pp
79–106
Gussone N, Eisenhauer A, Heuser A, Dietzel M, Bock B,
B€ohm F, Spero HJ, Lea DW, Bijma J, N€agler TF (2003)
Model for kinetic effects on calcium isotope fractionation
(d44Ca) in inorganic aragonite and cultured planktonic fora-
minifera. Geochim Cosmochim Acta 67:1375–1382
Gussone N, B€ohm F, Eisenhauer A, Dietzel M, Heuser A,
Teichert BMA, Reitner J, W€orheide G, Dullo W-Ch (2005)
Calcium isotope fractionation in calcite and aragonite.
Geochim Cosmochim Acta 69:4485–4494
Hensley TM (2006) Calcium isotopic variation in marine
evaporites and carbonates: applications to late miocene med-
iterranean brine chemistry and late Cenozoic calcium
cycling in the oceans, Ph.D. thesis, University of California,
San Diego, p 139
Heuser A, Eisenhauer A, B€ohmF,WallmannK,GussoneN,
Pearson PN, N€agler TF, Dullo W-Ch (2005) Calcium
isotope (d44/40Ca) variations of Neogene planktonic fora-
minifera. Paleoceanography 20: PA2013, doi: 10.1029/
2004PA001048
Hippler D, Schmitt A-D, Gussone N, Heuser A, Stille P,
Eisenhauer A, N€agler TF (2003) Calcium isotope composi-
tion of various reference materials and seawater.
Geostandard Newslett 27:13–19
Holland HD, Zimmermann H (2000) The dolomite prob-
lem revisited. Int Geol Rev 42:481–490
Holmden Ch (2009) Ca isotope study of Ordovician dolo-
mite, limestone, and anhydrite in the Williston Basin:
10 7.10 Chemical Characteristics of Sediments and Seawater 1481
implications for subsurface dolomitization and local Ca
cycling. Chem Geol 268:180–188
Huang S, Farkas J, Jacobsen SB (2010) Calcium isotopic
fractionation between clinopyroxene and orthopyroxene
from mantle peridotites. Earth Planet Sci Lett 270:349–353
Jacobson AD, Holmden C (2008) d44Ca evolution in a
carbonate aquifer and its bearing on the equilibrium isotope
fractionation factor for calcite. Earth Planet Sci Lett
292:337–344
Karhu JA (1993) Paleoproterozoic evolution of the car-
bon isotope ratios of sedimentary carbonates in
Fennoscandian Shield. Geol Surv Finl Bull 371:1–87
Kasemann SA, Hawkesworth ChJ, Prave AR, Fallick AE,
Pearson PN (2005) Boron and calcium isotope composition
in Neoproterozoic carbonate rocks from Namibia: evidence
for extreme environmental change. Earth Planet Sci Lett
231:73–86
Kump LR (2008) The role of seafloor hydrothermal
systems in evolution of seawater composition during the Phan-
erozoic. In: Lowell RP, Seewald JS, Metaxas A, Perfit MR
(eds) Magma to microbe: modeling hydrothermal processes at
ocean spreading centers, vol 178, Geophysical Monograph
Series. American Geophysical Union, Washington, DC,
pp 275–284
Machel H (2004) Concepts and models of dolomitization:
a critical reappraisal. Geol Soc Lond 235:7–63
Melezhik VA, Fallick AE, Medvedev PV, Makarikhin
VV (1999) Extreme 13C enrichment in ca. 2.0 Ga
magnesite–stromatolite–dolomite–‘red beds’ association in
a global context: a case for the world-wide signal enhanced
by a local environment. Earth Sci Rev 48:71–120
Melezhik VA, Fallick AE, Medvedev PV, Makarikhin
VV (2000) Palaeoproterozoic magnesite–stromatolite–do-
lostone–‘red bed’ association, Russian Karelia: palaeoenvir-
onmental constraints on the 2.0 Ga-positive carbon isotope
shift. Norsk Geol Tidsskr 80:163–185
Melezhik VA, Fallick AE, Rychanchik DV, Kuznetsov
AB (2005) Palaeoproterozoic evaporites in Fennoscandia:
implications for seawater sulphate, the rise of atmospheric
oxygen and local amplification of the d13C excursion. Terra
Nova 17:141–148
Ovchinnikova GV, Kuznetsov AB, Melezhik VA,
Gorokhov IM, Vasileva IM, Gorokhovskii BM (2007)
Pb-Pb age of Jatulian carbonate rocks: the Tulomozero For-
mation of southeast Karelia. Stratigr Geol Correl 15:359–372
Schmitt A-D, Chabaux F, Stille P (2003) The calcium
riverine and hydrothermal isotopic fluxes and the oceanic
calcium mass balance. Earth Planet Sci Lett 213:503–518
Steuber T, Buhl D (2006) Calcium-isotope fractionation
in selected modern and ancient marine carbonates. Geochim
Cosmochim Acta 70:5507–5521
Tang JW, Dietzel M, B€ohm F, K€ohler SJ, Eisenhauer A
(2008) Sr2+/Ca2+ and 44Ca/40Ca fractionation during inor-
ganic calcite formation: II. Ca isotopes. Geochim
Cosmochim Acta 72:3733–3745
Tipper ET, Galy A, Gaillardet J, Bickle MJ, Elderfield H,
Carder EA (2006a) The magnesium isotope budget of the
modern ocean: constraints from riverine magnesium isotope
ratios. Earth Planet Sci Lett 250:241–253
Tipper ET, Galy A, Bickle MJ (2006b) Riverine evidence
for a fractionated reservoir of Ca and Mg on the continents:
implications for the oceanic Ca cycle. Earth Planet Sci Lett
247:267–279
Veizer J (1983) Chemical diagenesis of carbonates: the-
ory and application of the trace element technique. In: stable
Isotopes in Sedimentary Geology. Society of Economic
Paleontologists and Mineralogists, Short Course, vol 10.
p 3–1 to 3–100
Von G€umbel CW (1857) Untersuchungen in den
bayerischen Alpen zwischen Isar und Salzach. Jahrb Geol
Reichsanstalt 7:146–151
Warthmann R, van Lith Y, Vasconcelos C, McKenzie JA,
Karpoff A-M (2000) Bacterially induced dolomite precipita-
tion in anoxic culture experiments. Geology 28:1091–1094
Zenger DH (1996) Dolomitization of the “C” zone, Red
River Formation (Upper Ordovician) in a deep core,
Williston Basin, Richland County, eastern Montana. Contri-
bution to Geology, University of Wyoming, vol 31:57–75
1482 J. Farkas et al.
7.10.4 Iron Speciation and Isotope Perspectiveson Palaeoproterozoic Water Column Chemistry
Christopher T. Reinhard, Timothy W. Lyons,Olivier Rouxel, Dan Asael, Nicolas Dauphas,and Lee R. Kump
The Fe Palaeoredox Proxies
Ancient rocks record the redox conditions of the ocean-
atmosphere system through the distribution of iron (Fe)
between oxidised and reduced minerals, which can be
formulated into a suite of Fe palaeoredox proxies. The bal-
ance between Fe and S in a given system reflects the variance
in a range of high- and low-temperature sources and sinks.
Iron can be delivered by hydrothermal, diagenetic or clastic
fluxes and can be buried and removed as Fe-oxide phases, Fe-
bearing carbonates such as siderite or ankerite, relatively
unreactive silicate phases, which often pass through the sys-
tem in detrital form, or as a constituent of pyrite (FeS2) using
sulphide sourced by sulphate reduction. Sulphate is delivered
to the ocean primarily from continental weathering, which
requires that a surface oxidative cycle exists, and rates of
sulphate delivery and Fe removal as pyrite should thus
depend on ocean-atmosphere redox. Among other successes,
the iron proxies discussed here have proven their value in
studies of the 2.5 Ga Mt. McRae Formation and specifically
in delineating subtle increases in atmospheric oxygen prior to
the Great Oxidation Event, or ‘GOE’. (Anbar et al. 2007;
Kaufman et al. 2007; Reinhard et al. 2009). These Fe proxies
are our most effective inorganic proxy for ancient euxinia
(anoxic and H2S-rich conditions) on the local scale and are an
essential independent backdrop for meaningful application
of Mo isotopes to address extents of euxinia on ocean scales
(Arnold et al. 2004; Gordon et al. 2009). Thus, in addition to
being informative on their own, Fe-based palaeoredox
indicators are a crucial component of multi-proxy
approaches for distinguishing among oxic, anoxic and Fe
(II)-rich (ferruginous), and euxinic depositional conditions.
The quantity and speciation of highly reactive iron (FeHR) in
sediments and sedimentary rocks can provide crucial insight
into the redox state of the local depositional environment. The
total pool of FeHR consists of mineral phases that have the
potential to react with dissolved H2S when exposed on short
timescales (within the water column or during earliest diagen-
esis) plus Fe that has already reacted and is present as FeS2(Raiswell and Canfield 1998). Such minerals include ferrous
carbonates (siderite, FeCO3; ankerite, Ca(Fe,Mg,Mn)(CO3)2),
crystalline ferric oxides (haematite, Fe2O3; goethite, FeOOH),
and the mixed-valence Fe oxide magnetite (Fe3O4). These
phases are separated by means of a well-calibrated sequential
extraction scheme described in detail elsewhere (Poulton et al.
2004; Poulton and Canfield 2005; Reinhard et al. 2009).
Briefly, ~100 mg of sample powder is first treated with a
buffered sodium acetate solution for 48 h to mobilise ferrous
carbonate phases. A split of the extract is removed for
analysis, the sample is centrifuged, and the remaining super-
natant is discarded. The sample is then treated with a sodium
dithionite solution for 2 h to dissolve crystalline ferric oxides
and processed as before. Finally, the sample is treated with
an ammonium oxalate solution for 6 h to mobilise magnetite.
All extractions are performed at room temperature in 15 mL
centrifuge tubes under constant agitation. The sequential
extracts are analysed on an Agilent 7500ce ICP-MS after
100-fold dilution in trace-metal grade HNO3 (2 %). Pyrite
iron is calculated separately based on weight percent pyrite
sulphur extracted during a 2-h, hot chromous chloride distil-
lation followed by iodometric titration (Canfield et al. 1986),
assuming a stoichiometry of FeS2. For measurement of total
Fe (FeT), sample powders are ashed overnight at 450 �C (in
order to remove organic matter but preserve volatile metals,
such as rhenium) and digested using sequential HNO3-HF-
HCl acid treatments (see, for example, Kendall et al. 2009).
After digestion, samples are reconstituted in trace-metal
grade HNO3 (2 %), diluted, and analysed by ICP-MS
In modern oxic sediments deposited across a wide range
of environments, FeHR comprises 6–38 % of total sedimen-
tary Fe (i.e. FeHR/FeT ¼ 0.06–0.38), with an average value
for FeHR/FeT of 0.26 � 0.08 defining the modern
siliciclastic baseline (Raiswell and Canfield 1998).
Enrichments in FeHR that are in excess of this detrital back-
ground ratio indicate a source of reactive Fe that is
decoupled from the siliciclastic flux and thus reflect the
transport, scavenging and enrichment (see below) of Fe
within an anoxic basin (Canfield et al. 1996; Wijsman
et al. 2001). In this context, ratios of FeHR/FeT exceeding
the siliciclastic range point to anoxic deposition, and the
ratio FePY/FeHR can then be used to establish whether the
system was Fe(II)- or H2S-buffered. An anoxic system with
a relatively small amount of FeHR converted to pyrite
indicates a depositional environment in which reactive Fe
supply was greater than the titrating capacity of available
H2S produced microbially by sulphate reduction, and thus no
dissolved H2S was accumulating in pore fluids or the water
column. Importantly, this is true even if microbial sulphate
reduction and pyrite formation was occurring in the system
(Canfield 1989) because the preponderance of Fe precludes
the accumulation of free H2S. In contrast, if the vast majority
of FeHR is present as pyrite in an anoxic system, euxinic
depositional conditions are indicated – a consequence of the
C.T. Reinhard (*)
Department of Earth Sciences, University of California, Riverside, CA
92521, USA
10 7.10 Chemical Characteristics of Sediments and Seawater 1483
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_10, # Springer-Verlag Berlin Heidelberg 2013
1483
sparing solubility of Fe(II) in the presence of accumulating
dissolved H2S and the scavenging of the transported reactive
Fe as pyrite in the water column.
Recent work in the Black Sea, the world’s largest modern
euxinic basin, reveals that the additional reactive Fe, which is
deposited as pyrite in the deep basin, can derive from benthic
fluxes out of the oxic-suboxic sediments in the shallow mar-
gin. Mixing of porewater Fe(II) into the overlying water
column produces reactive Fe as nanoparticulate oxides in
modernwell-oxygenated systems, butmay also have liberated
dissolved Fe(II) during the Precambrian. These fluxes are
transported basinward and scavenged from the sulphidic
water column (Fig. 7.167; reviewed in Lyons et al. 2009).
Interpretation of this reactive Fe system can be complicated,
however, by variations in the ratio of the oxic-suboxic benthic
source area to the anoxic/euxinic sink (Canfield et al. 1996;
Lyons 1997; Raiswell and Canfield 1998; Wijsman et al.
2001; Anderson and Raiswell 2004; Raiswell and Anderson
2005), the export efficiency of Fe from shelf sediments
(Raiswell and Anderson 2005), sedimentation rates (Lyons
and Severmann 2006), and in mineralogical transformations
accompanying metamorphism (discussed in Lyons and
Severmann 2006; Reinhard et al. 2009). All three factors
will tend to obscure anoxic and/or euxinic depositional
conditions. For example, a relatively small ratio between the
areal extent of oxic-suboxic benthic source relative to anoxic/
euxinic sink will attenuate reactive Fe enrichments even in an
anoxic system. Similarly, relatively high sedimentation rates
will tend to dilute reactive Fe enrichments through an elevated
flux of material with siliciclastic FeHR/FeT and FeT/Al values
regardless of redox conditions in the overlying water column.
We note, however, that when this Fe cycling occurs on an
ocean scale, the reactive Fe pool can also be supplemented by
hydrothermal activity along mid-ocean ridge systems.
Taken in isolation, then, proxy reconstructions that rely
on the reactive Fe system can be somewhat ‘asymmetric’
because interpretations that indicate anoxic and/or euxinic
deposition based on clearly elevated FeHR/FeT and FeT/Al
(and FePY/FeHR for euxinia) are relatively straightforward
and robust, and false positives are probably rare (Lyons and
Severmann 2006). On the other hand, interpretations that
indicate oxic deposition are more problematic and are most
convincing when made in concert with other proxy data
(such as trace metal distributions). The last of these
concerns, metamorphism, is of particular importance and is
discussed in more detail below.
Mineralogical transformation accompanying metamor-
phism can complicate interpretations of the reactive Fe sys-
tem. For example, primary reactive mineral phases, such as
FeCO3 or Fe2O3, can be converted to poorly reactive Fe-
containing silicate mineralogies that are not fully removed in
the extraction methodology. This conversion will have the
duel effect of artificially lowering FeHR/FeT values while
increasing FePY/FeHR values (Fig. 7.168a, b). The formation
of authigenic Fe-containing silicate phases in Fe(II)-rich
environments would have a similar effect, particularly with
subsequent conversion to poorly reactive (and poorly-
extracted) Fe-containing silicates during burial diagenesis
and metamorphism.
Two approaches for dealing with the above concerns
involve (1) interpreting FeHR systematics within the context
of total Fe content (FeT/Al) and (2) using another, more
operationally defined Fe speciation parameter, Degree of
Pyritisation (DOP; Raiswell et al. 1988). DOP is defined as:
DOP ¼ FePY
FePY þ FeHCl
where FeHCl is extracted by boiling the sample for 1 min in
concentrated HCl and quantified using spectrophotometric
techniques (Stookey 1970). FePY is determined as described
above. DOP provides a conservative measure of the degree to
which reactive Fe has been converted to pyrite, because FeHClincludes some amount of poorly reactive silicate Fe that will
not react with dissolved H2S even on long timescales
(Canfield et al. 1992; Raiswell et al. 1994; Raiswell and
Canfield 1996; summarised in Lyons and Severmann 2006).
In addition to some amount of the detritally delivered Fe, this
fraction will include some authigenic Fe-containing silicate
phases formed in Fe(II)-rich environments and some Fe-
silicate phases where these formed secondarily at the expense
of reactive Fe pools during burial alteration. As a result, high
values for DOP are a convincing proxy for euxinic conditions.
In the same way that DOP can be viewed as complemen-
tary to FePY/FeHR in terms of delineating H2S-buffered,
euxinic systems, FeT/Al ratios can be viewed as an effective
partner with FeHR/FeT for recognising anoxic depositional
conditions. Because metamorphic repartitioning of reactive
Fe phases into poorly reactive silicate mineralogies will not
change the total Fe content, anoxic systems showing artifi-
cially low FeHR/FeT ratios as a result of metamorphism will
still display elevated FeT/Al values. In short, high values for
FeT/Al and DOP provide compelling evidence for anoxic
and sulphidic (euxinic) depositional conditions, even if sub-
stantial Fe mineral transformations have occurred attendant
to metamorphism.
Another issue that emerges when looking at more ancient
sedimentary rocks is the potential presence of significant
pyrrhotite. Two problems arise: one is analytical and con-
founded by uncertain Fe-S stoichiometries; the other is
interpretational. In the first case, because sulphur as pyrrho-
tite is extracted efficiently during the hot chromous chlo-
ride distillation used to quantify pyrite S contents, a
significant amount of Fe as pyrrhotite will result in inaccu-
rate estimates of reactive Fe inventories if we assume that
all chromium-reducible sulphur is present in the form FeS2(Fig. 7.168c, d). Pyrrhotite can be isolated from pyrite
through a boiling 6N HCl distillation, and although this
method is too aggressive for modern sediments (Chanton
and Martens 1985; Cornwell and Morse 1987), it may
preferable when analysing ancient sedimentary rocks
(Rice et al. 1993). However, the variable stoichiometry of
1484 C.T. Reinhard et al.
pyrrhotite (Fe1-xS, where 0 < x < 0.25) makes calculation
of the associated Fe content difficult.
The second issue is establishing the mechanism of pyr-
rhotite formation, and this is crucial for making meaningful
interpretation of the primary depositional environment. It
has been suggested that pyrrhotite can be present as a pri-
mary mineral phase, accumulating detritally or
authigenically in systems that are sulphur limited and thus
do not promote the conversion of metastable amorphous FeS
phases to pyrite (Roberts and Turner 1993; Horng and
Roberts 2006; Larrasoana et al. 2007). Separate quantifica-
tion of pyrrhotite allows one to roughly quantify the total
amount of ‘sulphidised’ Fe as follows (Fig. 7.168e, f):
FeS
FeHR¼ FePY þ FePO
FeHR
where FePO denotes pyrrhotite Fe. In this case, the reactive Fe
system can be interpreted, as before, by examining the frac-
tion of reactive Fe that is fixed via reaction with dissolved
H2S. We note, however, that the presence of primary pyrrho-
tite would tend to suggest a priori that the pyritisation process
has been ‘arrested’ by some factor, most plausibly limited
availability of sulphur intermediates (Sn2�, S2O3
2�, SnO62�)
(Schoonen and Barnes 1991a; Hurtgen et al. 1999).
On the other hand, there are reasons to suspect that the
formation of pyrrhotite as a low temperature authigenic or
diagenetic phase should be unlikely undermost circumstances
(Schoonen and Barnes 1991b; Lennie et al. 1995). A more
probable source for pyrrhotite in ancient sedimentary rocks is
metamorphic reactions involving pyrite and/or Fe-containing
silicate phases. In the simplest case, fluid produced through
metamorphic dehydration of silicates dissolves pyrite to yield
pyrrhotite and an S-bearing fluid (see Tomkins 2010, and
references therein). This desulphurization process can occur
at relatively low temperatures (i.e. sub-greenschist facies;
Lambert 1973; Ferry 1981), and the net result of producing
such a fluid is a loss of Fe-bound sulphur from the rock. The
effect of such a process on the reactive Fe system is depicted
schematically in Fig. 7.168c, d. Pyrrhotite can also be formed
through the later reaction of such a fluid with Fe-containing
silicate phases, but this process is only known to occur at
relatively high grades of metamorphism (Mallio and Gheith
1972; Guidotti et al. 1988; Tompkins 2010). It has also been
shown that reactions at relatively low temperature involving
gypsum can yield secondary pyrrhotite (Hall 1982), but these
are not likely to be a significant source of pyrrhotite in shales.
In short, care must be taken when interpreting Fe specia-
tion data for sedimentary rocks that have experienced even
low-grade metamorphism. Nevertheless, robust arguments
for primary depositional conditions can be constructed even
within the limitations imposed by potential metamorphic
overprinting. Such considerations will be particularly impor-
tant for the FAR-DEEP materials, as many sulphide-
containing siliciclastic units show elevated magnetic
susceptibility indicating that a substantial portion of the
sulphide mineral pool may be represented by pyrrhotite,
which is confirmed by visual examination of the core. It
will thus be crucial to consider the effects of metamorphism
on Fe mineral assemblages through both textural and micro-
scopic observations and wet chemical methods. Most impor-
tant of the latter will be a coupled approach that considers the
reactive Fe pools in terms of more conservative redox
indicators such as FeT/Al and DOP, in addition to the devel-
opment and use of a method for isolating pyrite from pyrrho-
tite that is tailored to ancient sedimentary rocks.
Such an approach has already yielded important informa-
tion about the structure and evolution of Earth’s ancient
oceans. For example, detailed studies of Archaean (Reinhard
et al. 2009; Scott et al. 2011), Palaeoproterozoic (Poulton et al.
2004, 2010), and Neoproterozoic (Canfield et al. 2008) shales
and iron formations have demonstrated that the history of
deep ocean chemistry on Earth has been variable on a number
of temporal and spatial scales, with complex responses to
changing ocean-atmosphere redox and concomitant effects
on the evolution of life. Given the rather poorly constrained
redox status of the deep ocean during the Palaeoproterozoic
(Canfield 2005), the FAR-DEEP drillcores will provide
important new constraints on the response of ocean chemistry
to the initial oxygenation of the ocean-atmosphere system.
Fe Isotope Approaches
Natural mass-dependent Fe isotope variations are defined by
comparing 56Fe/54Fe ratios between a given sample and a
standard reference material, and are expressed in units of
parts-per-thousand (‘per mil’ or ‰) using conventional
‘delta’ notation:
d56Fe ¼ 1,000 x [(56Fe/54FeÞsample=ð56Fe/54FeÞstandard � 1�:
Differences in the isotopic composition between two
phases X and Y (D56FeX�Y ¼ d56FeX � d56FeY) can be
related to the isotopic fractionation factor (a) through the
approximation D56FeX-Y ~ 103lnaX-Y. Values for d56Fe in a
variety of natural materials span a range of up to 5 ‰ (Anbar
2003; Beard et al. 1999; Dauphas and Rouxel 2006; Johnson
et al. 2004). Experimental and theoretical studies have shown
that Fe isotopes should fractionate strongly between ferric
and ferrous iron species in aqueous environments (Anbar
et al. 2005; Jarzecki et al. 2004; Johnson et al. 2002; Welch
et al. 2003), making the Fe isotope system responsive to
redox-dependent Fe cycling. Average isotope fractionations
between Fe(II)aq and Fe(III)aq species (D56FeFe(II)-Fe(III)) of
2.5–3.6 ‰ were measured by Johnson et al. (2002) and
Welch et al. (2003) in dilute HCl solutions at temperatures
of 0–22 �C, consistent with the theoretical calculations of
Anbar et al. (2005) and Jarzecki et al. (2004). Investigation of
isotope exchange kinetics between Fe(II)aq and Fe(III)aq by
10 7.10 Chemical Characteristics of Sediments and Seawater 1485
mixing solutions of dissolved ferrous and ferric iron have
shown that isotopic equilibrium is reached in approximately
150–300 seconds, meaning that isotopic equilibrium effects
between Fe(II) and Fe(III) compounds tend to dominate in
both biological and inorganic redox processes.
Microbiological experiments have also shown that signif-
icant Fe isotope fractionations up to 2–3 ‰ occur during
dissimilatory Fe(III) reduction (Beard et al. 2003; Icopini
et al. 2004; Johnson et al. 2005) and anaerobic photosyn-
thetic Fe(II) oxidation (Croal et al. 2004). Fe isotope
fractionations can also occur during abiotic Fe(II) oxidation
and precipitation of ferric hydroxides (Balci et al. 2006;
Bullen et al. 2001), and through sorption of aqueous Fe(II)
onto ferric hydroxides (Icopini et al. 2004).
Iron is readily soluble in oxygen-depleted environments
as Fe(II) species and typically precipitated as Fe(III) after
oxidation in solution. Therefore, the total reaction of Fe
(II)aq ! Fe(III)aq ! Fe(III)solid is a key process for under-
standing Fe isotopic variations in the geologic record, and
particularly in rocks associated with significant changes in
Earth surface redox such as the GOE at 2.3 Ga (e.g.
Farquhar et al. 2000; Hannah et al. 2004; Bekker et al.
2004; Guo et al. 2009). The hydrolysis and precipitation
of Fe(III) (the second step in the above reaction) may also
include net fractionation depending on the precipitated
phase, the kinetics of the Fe(III)aq ! Fe(III)solid reaction
(equilibrium vs. kinetic fractionation), and mass balance
relationships between reactant and product. Skulan et al.
(2002) measured equilibrium and kinetic fractionation
factors during haematite precipitation from Fe(III)aq of
103lnaFe(III)-haematite ¼ 0.1 � 0.2 ‰ and 103lnaFe(III)--haematite ¼ 1.3 ‰, respectively, and other precipitated
minerals may produce different values. For instance, Butler
et al. (2005) investigated Fe-isotope fractionation during
FeS (i.e. mackinawite) precipitation from Fe(II)aq solutions
and demonstrated kinetic Fe-isotope fractionation up to
�0.8 ‰ which may explain the generally negative d56Fevalues in pyrite found in sedimentary (Severmann et al.
2006) and hydrothermal environments (Rouxel et al.
2008a). These studies emphasise the importance of the
precipitation mechanism for the interpretation of Fe isoto-
pic compositions in the geological record. Clearly, mass
balance between the different Fe phases also plays an
important role. In cases where the oxidation and precipita-
tion of the aqueous iron is quantitative, fractionations will
cancel out and the isotopic composition of the sediment
will reflect directly that of the source solution (e.g. sea
water). If the process is not complete, the quantitative
relationship between the phases and the precipitation
model should be considered (i.e. Rayleigh distillation vs.
equilibrium fractionation).
Lithogenic sources of Fe on the modern oxygenated Earth
have similar Fe isotope composition to those of bulk silicate
Earth (Beard et al. 2003). In contrast, marked variations in Fe
isotope composition have been reported in organic-rich
sediments (Beard et al. 2003; Jenkyns et al. 2007; Matthews
et al. 2004; Rouxel et al. 2005; Yamaguchi et al. 2005),
banded iron formations (Dauphas et al. 2004; Johnson et al.
2003, 2008a; Steinhoefel et al. 2009), hydrothermal fluids and
precipitates (Rouxel et al. 2008a; Sharma et al. 2001), and
altered volcanic rocks (Rouxel et al. 2003). Initial study of the
Fe isotope composition of marine sediments and sedimentary
rocks over geological time has provided new insights into the
ancient Fe cycle (Rouxel et al. 2005) (Fig. 7.169). The general
pattern of this record divides Earth’s history into three stages
which are strikingly similar to the stages defined by the d34Sand D33S records as well as other indicators of the redox state
of the atmosphere and ocean such as the appearance of red
beds, oxidised palaeosols, haematitic o€olites and pisolites,
Mn-oxide deposits, and Ce anomalies in chemical sedimen-
tary deposits (Holland 1984; Cloud 1988; Baker and Fallick
1989; Bau and Moller 1993; Karhu and Holland 1996;
Canfield 1998; Gutzmer and Beukes 1998; Farquhar et al.
2000; Bekker et al. 2004). Highly variable and negative d56Fevalues of pyrite before 2.32Gamay reflect reservoir effects on
dissolved Fe in the ocean resulting from the removal of
isotopically heavy Fe during oxidative precipitation (Rouxel
et al. 2005). Similarly, the positive d56Fe values between 1.8
to 2.3 Ga might be related to the increased effect of sulphide
precipitation in a redox-stratified ocean. After 1.8 Ga, the
near-complete scavenging of dissolved Fe by the reaction
with dissolved oxygen or biogenic H2S to form Fe oxide and
sulphide minerals limits the extent of Fe isotope variability in
sediments (Anbar and Rouxel 2007; Rouxel et al. 2005).
Although both S- and Fe-isotope systematics, as well as Fe-
and S-speciation (Poulton et al. 2004) indicate a transition
from anoxic (Fe(II)-rich) to widespread euxinic (H2S-rich)
deep ocean conditions in the Palaeoproterozoic, there is
conflicting information in the distribution of rare earth
elements in late Palaeoproterozoic BIFs (Slack et al. 2007),
and the nature and timing of the relationship between oceanic
biogeochemical cycles of Fe and S during the rise of atmo-
spheric oxygen remains unclear.
Although several interpretations of the Fe-isotope record
in black shales were proposed (Archer and Vance 2006;
Rouxel et al. 2005; Severmann et al. 2006; Yamaguchi
et al. 2005), it is likely that the shift from high d56Fevariability in >2.3 Ga black shales to little variability
<1.8 Ga reflects redox-related changes in the global oceanic
Fe cycle (Fig. 7.169). It appears that Fe isotope variations in
sedimentary pyrite are particularly sensitive to the concen-
tration of dissolved Fe(II) (i.e. the size of aqueous Fe reser-
voir) and can be used to place important constraints on the
sources to and sinks from this Fe(II) reservoir in past oceans.
Subsequent studies of modern oxygen-deficient sedimen-
tary and oceanic systems (Rouxel et al. 2008b; Severmann
et al. 2006, 2008; Staubwasser et al. 2006) have helped to
better constrain the fractionation of Fe-isotopes during
early diagenesis and their palaeo-environmental implications,
thus providingmodern analogues to the Precambrian Fe cycle
1486 C.T. Reinhard et al.
in redox-stratified oceans. In particular, dissimilatory Fe
reduction (DIR) in sediment porewater may produce strongly
negative d56Fe signatures in pyrite in organic-rich sediments
similar to those observed in Precambrian black shales
(Fig. 7.167). By analogy with bacterial sulphate reduction, it
has been proposed that the secular variation of d56Fe values inblack shale pyrite is primarily controlled by the extent of Fe
(III) reduction during diagenesis, which is itself dependent on
the amount of reactive Fe(III) available for DIR. Since Fe(III)
oxides are insoluble (unlike sulphate), DIR is expected to
produce an isotopically heavy Fe(III)-rich reservoir which
should largely remain in the sediments, and diagenetic
remobilisation of reactive Fe (see above) will promote the
removal of light Fe as dissolved Fe(II) (see below). Although
d56Fe analyses of various coexisting Fe-pools in black shales,such as Fe-carbonates, Fe-oxides and disseminated pyrite and
silicates do not provide evidence for the complementary high
d56Fe components (Duan et al. 2010; Rouxel et al. 2006), an
important prospect of the FARDEEP research will further test
the effects of DIR on secular Fe-isotope variations during the
Archaean and Palaeoproterozoic era by coupling d56Feanalyses with Fe speciation analyses on the same sample.
It has also been demonstrated that suboxic porewaters on
modern continental shelves have characteristically light
(d56Fe < �2.0 ‰) iron isotope values (Rouxel et al.
2008b; Severmann et al. 2006). Consistent with a shelf-to-
basin iron transport inferred for modern euxinic basins
(Lyons and Severmann 2006), Palaeoproterozoic sediments
might have experienced a similar iron-shuttle that generated
an isotopically light benthic iron flux (Severmann et al.
2008). The Palaeoproterozoic sulphidic shales from the
Fennoscandian Shield available in FAR-DEEP drillcores
will allow the potential contribution of shelf-derived Fe to
be estimated through higher reactive Fe concentrations and
increased Fe/Al ratios in bulk shales. These contributions
should also correlate with negative d56Fe values, although
hydrothermal enrichment should also exert a strong influ-
ence on the Fe isotope mass balance in Archaean and
Palaeoproterozoic oceans.
In conclusion, by providing a context for interpreting
trace metal chemistry and transition metal isotope systemat-
ics, the Fe-speciation approach is critical to our use of the
FAR-DEEP cores as windows to the tempo and mode of
early Earth oxygenation.
Fig. 7.167 Schematic depiction of the benthic Fe shuttle as a mecha-
nism for enriching anoxic sediments in biogeochemically reactive Fe.
Reactive Fe is transported from shelf sediments, either as amorphous
Fe oxy(hydr)oxides or as dissolved Fe(II) subsequent to dissimilatory
Fe reduction (DIR) within shelf sediments, and is scavenged and buried
as syngenetic Fe phases in the deeper water column. This is manifested
in sediments by progressively increasing values for FeT/Al, FeHR/FeT,
and DOP. Shown here is a euxinic basin in which the ultimate reposi-
tory for reactive Fe is pyrite (FeS2) but in ferruginous systems this
syngenetic reactive Fe phase may be Fe carbonates or oxides (Modified
after Lyons and Severmann (2006))
10 7.10 Chemical Characteristics of Sediments and Seawater 1487
References
Anbar A (2003) Iron stable isotopes: beyond biosignatures.
Earth Planet Sci Lett 217:223–236
Anbar AD, Rouxel O (2007) Metal stable isotopes in
paleoceanography. Annu Rev Earth Planet Sci 35:717–746
Anbar AD, Jarzecki AA, Spiro TG (2005) Theoretical
investigation of iron isotope fractionation between
Fe(H2O)63+ and Fe(H2O)6
2+: implications for iron stable
isotope geochemistry. Geochim Cosmochim Acta 69:
825–827
Anbar AD, Duan Y, Lyons TW, Arnold GL, Kendall B,
Creaser RA, Kaufman AJ, Gordon GW, Scott C, Garvin J,
Buick R (2007) A whiff of oxygen before the great oxidation
event? Science 317:1903–1906
Anderson TF, Raiswell R (2004) Sources and
mechanisms for the enrichment of highly reactive iron in
euxinic Black Sea sediments. Am J Sci 304:203–233
Archer C, Vance D (2006) Coupled Fe and S isotope
evidence for Archean microbial Fe(III) and sulfate reduc-
tion. Geology 34:153–156
Arnold GL, Anbar AD, Barling J, Lyons TW (2004)
Molybdenum isotope evidence for widespread anoxia in
mid-Proterozoic oceans. Science 304:87–90
Canfield DE (1989) Reactive iron in marine sediments.
Geochim Cosmochim Acta 53:619–632
Baker AJ, Fallick AE (1989) Evidence from Lewisian
limestone for isotopically heavy carbon in two-thousand-
million-year-old sea water. Nature 337:352–354
Balci N, Bullen TD, Witte-Lien K, Shanks WC, Motelica
M, Mandernack KW (2006) Iron isotope fractionation dur-
ing microbially stimulated Fe(II) oxidation and Fe(III) pre-
cipitation. Geochim Cosmochim Acta 70:622–639
Bau M, Moller P (1993) Rare earth element systematics
of the chemically precipitated component in early Precam-
brian iron formations and the evolution of the terrestrial
atmopshere-hydrosphere-lithosphere system. Geochim
Cosmochim Acta 57:2239–2249
Beard BL, Johnson CM, Cox L, Sun H, Nealson KH,
Aguilar C (1999) Iron isotope biosignatures. Science
285:1889–1892
Fig. 7.168 Hypothesised shifts
in Fe speciation parameters
attendant to different
metamorphic overprints. The
y-axis represents values for eachgiven ratio. Open symbols
represent the primary sediment
values, while shaded symbols
represent values following
hypothetical metamorphic
overprinting. Panels (a) and (b)
show shifts expected under the
circumstance that some portion of
the originally reactive Fe pool is
incorporated into poorly reactive
Fe-containing silicates. Panels
(c–f) depict one possible series of
shifts accompanying
metamorphic desulphurisation of
pyrite and subsequent loss of S
from the rock in a reducing fluid.
It is assumed for illustration that
roughly two thirds of the sulphur
originally present as pyrite
(stoichiometry of FeS2) is
removed, and the remaining
Cr-reducible S is converted to
pyrrhotite (approximate
stoichiometry of FeS). The first
case (c, d) depicts a scenario
where all Cr-reducible S is
assumed to be present as pyrite.
The second case (e, f) depicts a
scenario utilising a refined
method for pyrrhotite separation
and correction of stoichiometry.
FeS is as defined in the text
1488 C.T. Reinhard et al.
Beard BL, Johnson CM, Skulan JL, Nealson KH, Cox L,
Sun H (2003) Application of Fe isotopes to tracing the geo-
chemical and biological cycling of Fe. ChemGeol 195:87–117
Bekker A, Holland HD, Wang PL, Rumble D III, Stein
HJ, Hannah JL, Coetzee LL, Beukes NJ (2004) Dating the
rise of atmospheric oxygen. Nature 427:117–120
Bekker A, Slack JF, Planavsky N, Krapez B, Hofmann A,
Konhauser KO, Rouxel OJ (2010) Iron formation: the sedi-
mentary product of a complex interplay among mantle,
tectonic, and biospheric processes. Econ Geol 105:467–508
Bullen TD, White AF, Childs CW, Vivit DV, Schulz MS
(2001) Demonstration of significant abiotic iron isotope
fractionation in nature. Geology 29:699–702
Butler IB, Archer C, Vance D, Oldroyd A, Rickard D
(2005) Fe isotope fractionation on FeS formation in ambient
aqueous solution. Earth Planet Sci Lett 236:430–442
Canfield DE (1998) A new model for Proterozoic ocean
chemistry. Nature 396:450–453
Canfield DE (2005) The early history of atmospheric
oxygen: homage to Robert M. Garrels. Annu Rev Earth
Planet Sci 33:1–36
Canfield DE, Raiswell R, Westrich JT, Reaves CM,
Berner RA (1986) The use of chromium reduction in the
analysis of reduced inorganic sulfur in sediments and shales.
Chem Geol 54:149–155
Canfield DE (2005) The early history of atmospheric
oxygen: homage to Robert M. Garrels. Annu Rev Earth
Planet Sci 33:1–36
Canfield DE, Raiswell R, Bottrell S (1992) The reactivity
of sedimentary iron minerals toward sulfide. Am J Sci
292:659–683
Canfield DE, Lyons TW, Raiswell R (1996) A model for
iron deposition to euxinic Black Sea sediments. Am J Sci
296:818–834
Canfield DE, Poulton SW, Knoll AH, Narbonne GM,
Ross G, Goldberg T, Strauss H (2008) Ferruginous
conditions dominated later Neoproterozoic deep-water
chemistry. Science 321:949–952
Chanton JP, Martens CS (1985) The effects of heat and
stannous chloride addition on the active distillation of acid
volatile sulfide from pyrite-rich marine sediment samples.
Biogeochemistry 1:275–383
Cloud P (1988) Oasis in space: earth history from the
beginning. WW Norton, New York, p 526
Cornwell JC, Morse JW (1987) The characterization of
iron sulfide minerals in anoxic marine sediments. Mar Chem
22:193–206
Croal LR, Johnson CM, Beard BL, Newman DK (2004)
Iron isotope fractionation by Fe(II)-oxidizing photoautotro-
phic bacteria. Geochim Cosmochim Acta 68:1227–1242
Dauphas N, Rouxel O (2006) Mass spectrometry and
natural variations of iron isotopes. Mass Spectrom Rev
25:515–550
Dauphas N, van Zuilen M, Wadhwa M, Davis AM, Marty
B, Janney PE (2004) Clues from Fe isotope variations on the
origin of early archean BIFs from Greenland. Science
306:2077–2080
Duan Y, Severmann S, Anbar AD, Lyons TW, Gordon
GW, Sageman B (2010) Isotopic evidence for Fe cycling and
repartitioning in ancient oxygen-deficient settings: examples
from black shales of the mid-to-late Devonian Appalachian
basin. Earth Planet Sci Lett 290:244–253
Fig. 7.169 Fe isotope
composition of sedimentary
pyrite in black shales and iron
formations over geological time.
Data include bulk mineral
analysis (magnetite, haematite,
Fe-rich carbonates: ankerite and
siderite, pyrite and metamorphic
Fe-silicates) and in situ analysis
using laser-ablation MC-ICPMS
and ion-microprobe (Data are
from recent compilation by Anbar
and Rouxel (2007); Bekker et al.
(2010); Johnson et al. (2008b);
Planavsky et al. (2012). GOE
refers to the Great Oxidation
Event (see text)
10 7.10 Chemical Characteristics of Sediments and Seawater 1489
Farquhar J, Bao H, Thiemens M (2000) Atmospheric
influence of Earth’s earliest sulfur cycle. Science
289:756–758
Ferry JM (1981) Petrology of graphitic sulfide-rich
schists from south-central Maine: an example of
desulfidation during prograde regional metamorphism. Am
Mineral 66: 908–930
Gordon GW, Lyons TW, Arnold GL, Roe J, Sageman
BB, Anbar AD (2009) When do black shales tell molybde-
num isotope tales? Geology 37:535–538
Guidotti CV, Cheney JT, Henry DJ (1988) Compositional
variation of biotite as a function of metamorphic reactions
and mineral assemblage in the peltitic schists of western
Maine. Am J Sci 288-A:270–292
Hall AJ (1982) Gypsum as a precursur to pyrrhotite in
metamorphic rocks. Miner Deposita 17:401–409
Hannah JL, Bekker A, Stein HJ, Markey RJ, Holland HD
(2004) Primitive Os and 2316 Ma age for marine shale:
implications for Paleoproterozoic glacial events and the
rise of atmospheric oxygen. Earth Planet Sci Lett 225:43–52
Holland HD (1984) The chemical evolution of the atmo-
sphere and oceans. Princeton University Press, New York,
p 598
Horng C-S, Roberts AP (2006) Authigenic or detrital
origin of pyrrhotite in sediments?: resolving a paleo-
magnetic conundrum. Earth Planet Sci Lett 241:750–762
Hurtgen MT, Lyons TW, Ingall ED, Cruse AM (1999)
Anomalous enrichments of iron monosulfide in euxinic
marine sediments and the role of H2S in iron sulfide
transformations: examples from Effingham Inlet, Orca
Basin, and the Black Sea. Am J Sci 299:556–588
Guo Q, Strauss H, Kaufman AJ, Schroeder S, Gutzmer J,
Wing B, Baker MA, Bekker A, Jin Q, Kim S-T, Farquhar J
(2009) Reconstructing Earth’s surface oxidation across the
Archean-Proterozoic transition. Geology 37(5):399–402
Gutzmer J, Beukes NJ (1998) Earliest laterites and possi-
ble evidence for terrestrial vegetation in the Paleopro-
terozoic. Geology 26:263–266
Icopini GA, Anbar AD, Ruebush SS, Tien M, Brantley
SL (2004) Iron isotope fractionation during microbial reduc-
tion of iron: the importance of adsorption. Geology
32:205–208
Jarzecki AA, Anbar AD, Spiro TG (2004) DFT analysis
of Fe(H2O)63+ and Fe(H2O)6
2+ structure and vibrations;
implications for isotope fractionation. J Phys Chem A
14:2726–2732
Jenkyns HC, Matthews A, Tsikos H, Erel Y (2007)
Nitrate reduction, sulfate reduction, and sedimentary iron
isotope evolution during the Cenomanian-Turonian oceanic
anoxic event. Paleoceanography:PA3208,doi:10.1029/
2006PA001355
Johnson CM, Skulan JL, Beard BL, Sun H, Nealson KH,
Braterman PS (2002) Isotopic fractionation between
Fe(III) and Fe(II) in aqueous solutions. Earth Planet Sci
Lett 195:141–153
Johnson CM, Beard BL, Beukes NJ, Klein C, O’Leary JM
(2003) Ancient geochemical cycling in the Earth as inferred
from Fe isotope studies of banded iron formations from the
Transvaal Craton. Contrib Miner Petrol 144:523–547
Johnson CM, Beard BL, Roden EE, Newman DK,
Nealson KH (2004) Isotopic constraints on biological
cycling of Fe. Rev Miner Geochem 55:359–408
Johnson CM, Roden EE, Welch SA, Beard BL (2005)
Experimental constraints on Fe isotope fractionation
during magnetite and Fe carbonate formation coupled to
dissimilatory hydrous ferric oxide reduction. Geochim
Cosmochim Acta 69:963–993
Johnson CM, Beard BL, Klein C, Beukes NJ, Roden EE
(2008a) Iron isotopes constrain biologic and abiologic
processes in banded iron formation genesis. Geochim
Cosmochim Acta 72:151–169
Johnson CM, Beard BL, Roden EE (2008b) The iron
isotope fingerprints of redox and biogeochemical cycling
in the modern and ancient Earth. Annu Rev Earth Planet
Sci 36:457–493
Karhu JA, Holland HD (1996) Carbon isotopes and the
rise of atmospheric oxygen. Geology 24:867–870
Kaufman AJ, Johnston DT, Farquhar J, Masterson AL,
Lyons TW, Bates S, Anbar AD, Arnold GL, Garvin J, Buick
R (2007) Late Archean biospheric oxygenation and atmo-
spheric evolution. Science 317:1900–1903
Kendall B, Creaser RA, Gordon GW, Anbar AD (2009)
Re-Os and Mo isotope systematics of black shales from the
Middle Proterozoic Velkerri and Wollogorang Formations,
McArthur Basin, northern Australia. Geochim Cosmochim
Acta 73:2534–2558
Lambert IB (1973) Post-depositional availabity of sul-
phur and metals and formation of secondary textures and
structures in stratiform sedimentary sulphide deposits.
J Geol Soc Aust 20:205–215
Larrasoana JC, Roberts AP, Musgrave RJ, Gracia E,
Pinero E, Vega M, Martinez-Ruiz F (2007) Diagenetic
formation of greigite and pyrrhotite in gas hydrate
marine sedimentary systems. Earth Planet Sci Lett
261:350–366
Lennie AR, England KER, Vaughan DJ (1995) Transfor-
mation of synthetic mackinawite to hexagonal pyrrhotite: a
kinetic study. Am Miner 80:960–967
Lyons TW (1997) Sulfur isotopic trends and pathways of
iron sulfide formation in upper Holocene sediments of the
anoxic Black Sea. Geochim Cosmochim Acta 61:3367–3382
Lyons TW, Severmann S (2006) A critical look at iron
paleoredox proxies: new insights from modern euxinic
marine basins. Geochim Cosmochim Acta 70:5698–5722
Lyons TW, Anbar AD, Severmann S, Scott C, Gill BC
(2009) Tracking euxinia in the ancient ocean: a multiproxy
1490 C.T. Reinhard et al.
perspective and Proterozoic case study. Annu Rev Earth
Planet Sci 37:507–534
Mallio WJ, Gheith MA (1972) Textural and chemical
evidence bearing on sulfide-silicate reactions in
metasediments. Miner Deposita 7:13–17
Matthews A, Morgans-Bell HS, Emmanuel S, Jenkyns
HC, Erel Y, Halicz L (2004) Controls on iron-isotope frac-
tionation in organic-rich sediments (Kimmeridge Clay,
Upper Jurassic, southern England). Geochim Cosmochim
Acta 68:3107–3123
Planavsky N, Rouxel OJ, Bekker A, Hofmann A, Little
CTS, Lyons TW (2012) Iron isotope composition of some
Archean and Proterozoic iron formations. Geochim
Cosmochim Acta 80:158–169
Poulton SW, Canfield DE (2005) Development of a
sequential extraction procedure for iron: implications for
iron partitioning in continentally derived particulates.
Chem Geol 214:209–221
Poulton SW, Fralick PW, Canfield DE (2004) The transi-
tion to a sulphidic ocean 1.84 billion years ago. Nature
431:173–177
Poulton SW, Fralick PW, Canfield DE (2010) Spatial
variability in oceanic redox structure 1.8 billion years ago.
Nat Geosci 3:486–490
Raiswell R, Anderson TF (2005) Reactive iron enrich-
ment in sediments deposited beneath euxinic bottom waters:
constraints on supply by shelf recycling. Geological Society
of London Special Publications 248:179–194
Raiswell R, Canfield DE (1996) Rates of reaction
between silicate iron and dissolved sulfide in Peru Margin
sediments. Geochim Cosmochim Acta 60:2777–2787
Raiswell R, Canfield DE (1998) Sources of iron for pyrite
formation in marine sediments. Am J Sci 298:219–245
Raiswell R, Buckley F, Berner RA, Anderson TF (1988)
Degree of pyritization of iron as a paleoenvironmental indi-
cator of bottom-water oxygenation. J Sediment Petrol
58:812–819
Raiswell R, Canfield DE, Berner RA (1994) A compari-
son of iron extraction methods for the determination of
degree of pyritisation and the recognition of iron-limited
pyrite formation. Chem Geol 111:101–110
Reinhard CT, Raiswell R, Scott C, Anbar AD, Lyons
TW (2009) A late Archean sulfidic sea stimulated by early
oxidative weathering of the continents. Science 326:713–716
Rice CA, Tuttle ML, Reynolds RL (1993) The analysis of
forms of sulfur in ancient sediments and sedimentary rocks:
comments and cautions. Chem Geol 107:83–95
Roberts AP, Turner GM (1993) Diagenetic formation of
ferromagnetic iron sulphide minerals in rapidly deposited
marine sediments, South Island, New Zealand. Earth Planet
Sci Lett 115:257–273
Rouxel O, Dobbek N, Ludden J, Fouquet Y (2003) Iron
isotope fractionation during oceanic crust alteration (Site
ODP 801). Chem Geol 202:155–182
Rouxel O, Bekker A, Edwards K (2005) Iron isotope
constraints on the Archean and Paleoproterozoic ocean
redox state. Science 307:1088–1091
Rouxel O, Bekker A, Edwards K (2006) Response to
Comment on “Iron isotope constraints on the Archean and
Paleo-Proterozoic Ocean Redox State” by Yamaguchi K,
Ohmoto H. Science 311:177b
Rouxel O, Shanks WC, Bach W, Edwards K (2008a)
Integrated Fe and S isotope study of seafloor hydrothermal
vents at East Pacific Rise 9-10N. Chem Geol 252:214–227
Rouxel O, Sholkovitz E, Charette M, Edwards K (2008b)
Iron isotope fractionation in subterranean estuaries.
Geochim Cosmochim Acta 72:3413–3430
Schoonen MAA, Barnes HL (1991a) Reactions
forming pyrite and marcasite from solution: II. Via FeS
precursors below 100 �C. Geochim Cosmochim Acta
55:1505–1514
Schoonen MAA, Barnes HL (1991b) Mechanisms of
pyrite and marcasite formation from solution. III. Hydrother-
mal processes. Geochim Cosmochim Acta 55:3491–3504
Scott C, Bekker A, Reinhard CT, Schnetger B, Krapez B,
Rumble D III, Lyons TW (2011) Late Archean euxinic
conditions before the rise of atmospheric oxygen. Geology
39:119–122
Severmann S, Johnson CM, Beard BL, McManus J
(2006) The effect of early diagenesis on the Fe isotope
compositions of porewaters and authigenic minerals in con-
tinental margin sediments. Geochim Cosmochim Acta
70:2006–2022
Severmann S, Lyons TW, Anbar A,McManus J, Gordon G
(2008) Modern iron isotope perspective on the benthic iron
shuttle and the redox evolution of ancient oceans. Geology
36:487–490
Sharma M, Polizzotto M, Anbar AD (2001) Iron isotopes
in hot springs along the Juan de Fuca Ridge. Earth Planet Sci
Lett 194:39–51
Skulan JL, Beard BL, Johnson CM (2002) Kinetic
and equilibrium Fe isotope fractionation between aqueous
Fe(III) and hematite. Geochim Cosmochim Acta 66:
2995–3015
Slack JF, Grenne T, Bekker A, Rouxel OJ, Lindberg PA
(2007) Suboxic deep seawater in the late Paleoproterozoic:
evidence from hematitic chert and iron formation related to
seafloor-hydrothermal sulfide deposits, central Arizona,
USA. Earth Planet Sci Lett 255:243–256
Staubwasser M, von Blanckenburg F, Schoenberg R
(2006) Iron isotopes in the early marine diagenetic iron
cycle. Geology 34:629–632
Stookey LL (1970) Ferrozine – a new spectrophotometric
reagent for iron. Anal Chem 42: 779–781
Steinhoefel G, Horn I, von Blanckenburg F (2009) Micro-
scale tracing of Fe and Si isotope signatures in banded iron
formation using femtosecond laser ablation. Geochim
Cosmochim Acta 73:5343–5360
10 7.10 Chemical Characteristics of Sediments and Seawater 1491
Tomkins AG (2010) Windows of metamorphic sulfur
liberation in the crust: implications for gold deposit genesis.
Geochim Cosmochim Acta 74:3246–3259
Welch SA, Beard BL, Johnson CM, Braterman PS (2003)
Kinetic and equilibrium Fe isotope fractionation between
aqueous Fe(II) and Fe(III). Geochim Cosmochim Acta
67:4231–4250
Wijsman JWM, Middelburg JJ, Heip CHR (2001) Reac-
tive iron in Black Sea sediments: implications for iron
cycling. Mar Geol 172:167–180
Yamaguchi KE, Johnson CM, Beard BL, Ohmoto H
(2005) Biogeochemical cycling of iron in the Archean–Pa-
leoproterozoic Earth: constraints from iron isotope
variations in sedimentary rocks from the Kaapvaal and
Pilbara Cratons. Chem Geol 218:135–169
1492 C.T. Reinhard et al.
7.10.5 Cr Isotopes in Near Surface ChemicalSediments
Mark van Zuilen and Ronny Schoenberg
Chromium in Earth’s Reservoirs
Chromium is the twenty-first most abundant element in the
Earth’s crust with an average concentration of 100 mg g�1,
while its concentration in Earth’s mantle, where it is the 8th
most abundant element, is as much as 2,600–4,000 mg g�1
(McDonough 2004; Walter 2004). Chromite (FeCr2O4) is
the economically most important Cr-bearing mineral and
occurs in mafic to ultramafic rocks. It is mainly mined
from metre-thick massive chromitite ore layers in layered
intrusions such as the Bushveld Igneous Complex in South
Africa, and the Stillwater Igneous Complex in the USA.
Chromium is a redox-sensitive element with the two stable
oxidation states +3 and +6. All chromium in igneous Earth
reservoirs is present in the +3 oxidation state, with the
exception of rare Cr(VI) mineral species, such as crocoite
PbCrO4, that crystallise from oxidised fluids near Earth’s
surface.
In Earth’s hydrosphere chromium is a prominent redox-
partner of hydrogen sulfide (H2S), iron and manganese
(Pettine et al. 1998a, b, 2006; Kim et al. 2002; Lee and
Hering 2003; Weaver and Hochella 2003; Wilkin et al.
2005). Both chromium oxidation states can therefore be
predominant in aquatic systems depending on the prevailing
Eh and pH conditions (Fig. 7.170a). This in turn renders the
speciation ratio of Cr(VI):Cr(III) a useful indicator for the
redox state of aqueous media (Sander and Koschinsky 2000;
Sirinawin et al. 2000). In aquatic systems at near neutral pH
conditions, Cr(III) mainly exists as the octahedrally coordi-
nated chromium hydroxides Cr(OH)2+ and Cr(OH)3
(Fig. 7.170b). Both these Cr(III) complexes are poorly solu-
ble (the equilibrium constant log K of Cr(III)-hydroxide in
water is less than �6.84; Rai et al. 1987), but highly particle
reactive, adsorbing efficiently onto inorganic and organic
matter. Cr(VI) on the other hand is highly soluble (the
solubility of K2CrO4 in water, for example, is 629 g L�1)
and occurs as the tetrahedrally coordinated chromate
(CrO42�), bichromate (Cr2O7
2�) or deprotonated chromic
acid (HCrO4�). The relative abundances of these
compounds in aquatic systems are mainly depending on the
ambient pH conditions and the overall Cr(VI) concentration
(Fig. 7.170c).
Cr Isotope Systematics
Chromium has four stable isotopes 50Cr, 52Cr, 53Cr and54Cr with the relative abundances of 0.043452 � 85,
0.837895 � 117, 0.095006 � 110 and 0.0236547 � 48
(Coplen et al. 2002). 53Cr is the decay product of the short-
lived radionuclide 53Mn (half-life ¼ 3.7 Ma), which already
became extinct in our solar system and thus in any terrestrial
material within the first few tens of million years after the
start of planetary accretion. Relative to 50Cr and 52Cr, small
radiogenic enrichments in the relative abundance of 53Cr
through the decay of 53Mn and small deficits in the abun-
dance of 54Cr through nucleosynthetic processes can be
found in various meteorites and in reservoirs of
differentiated planetesimals, such as the Howardite-
Eucrite-Diogenite (HED) parent body (Lugmair and
Shukolyukov 1998; Trinquier et al. 2007). However, no
such variations have been reported for any terrestrial mate-
rial so far, suggesting that these minor anomalies have been
homogenised by the dynamic nature of our planet.
Redox-related mass-dependent stable isotope fraction-
ation of several ‰ on the 53Cr/52Cr ratio was predicted
from empirical and ab initio force field models (Schauble
et al. 2004). Mass-dependent stable isotope fractionation can
be expressed through the fractionation factor a:
a ¼ Rproduct
Rreactant(1)
where Rproduct and Rreactant refer to the 53Cr/52Cr ratio of the
product and reactant of a specific chemical reaction. The
isotopic composition of a sample is reported relative to that
of the NIST SRM979 chromium isotope reference material
according to Eq. 2:
d53=52CrSRM 979 ¼53Cr=52Cr� �
sample
53Cr=52Crð ÞSRM 979
� 1 (2)
For reasons of simplification, the d values are expressed
in per mil by multiplication with a factor of 103. In case of
equilibrium isotope fractionation, the isotopic difference
between the product and the reactant is constant and can be
expressed as:
D53=52Crproduct�reactant ¼ d53=52Crproduct
� d53=52Crreactant (3)
which again is preferentially expressed in per mil by multi-
plication with a factor of 103. Consequently, for equilibrium
isotope fractionation reactions, the difference in isotopic
composition between product and reactant is a proxy for
the reaction’s isotopic fractionation through the relation:
M. van Zuilen (*)
Institut de Physique du Globe de Paris, Equipe Geobiosphere Actuelle
et Primitive, 1 rue Jussieu, 75238 cedex 5 Paris, France
10 7.10 Chemical Characteristics of Sediments and Seawater 1493
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_10, # Springer-Verlag Berlin Heidelberg 2013
1493
D53=52Crproduct�reactant � ln a (4)
While, in case of a unidirectional, kinetic Cr isotopic
fractionation process, the isotopic difference between the
instantaneously formed product and the reactant is still con-
stant, the difference between the accumulating product and
the reactant increases with ongoing reaction. Here, the reac-
tant and the accumulating product follow a Rayleigh type
fractionation model described in Eqs. 5 and 6:
d53=52Crreactant�ðtÞ
¼ d53=52Crreactant�initial þ 103� �
� fAa�1ð Þ
h i� 103
(5)
d53=52Craccum:product�ðtÞ
¼ d53=52Crreactant�initial � fA � d53=52Crreactant�ðtÞ� �� �
1� fA
(6)
where d53/52Crreactant•(t) and d53/52Craccum.product•(t) refer to
the isotopic composition of the unreacted pool (Rayl.A in
Fig. 7.171) and the accumulated product (Rayl.Baccum. in
Fig. 7.171) at time t, d53/52Crreactant•initial refers to the initial
isotopic composition of the reactant before the reaction
starts, and fA refers to the fraction of the unreacted pool.
The relation of the isotopic composition of product and
reactant in equilibrium and kinetic Cr isotope fractionation
is illustrated in Fig. 7.171.
Cr Isotope Cycling on Earth’s Surface
Manganese-oxides are known to be a major agent of Cr(III)
oxidation in soils and sediments (Kim et al. 2002). However,
little is known about the isotopic effects accompanying Cr
(III) oxidation. Preliminary experiments of Cr(III) oxidation
by the Mn-oxide birnessite (d-MnO2) revealed wide
variations in d53/52Cr values of the developing Cr(VI)
pools between �2.5 ‰ and +0.7 ‰ (Bain and Bullen
2005). This range in d53/52CrSRM979 values during Cr(III)
oxidation by MnO2 suggests the involvement of multiple
processes, such as the disproportionation of intermediate,
unstable Cr(IV) and Cr(V) complexes. Abiogenic and dis-
similatory Cr(VI) reduction experiments show large kinetic
isotopic fractionation with fractionation factors a between
0.9966 and 0.9950 (Ellis et al. 2002; Schoenberg et al. 2008;
Sikora et al. 2008; Zink et al. 2010), enriching the residual
unreacted Cr(VI) pool in heavy Cr isotopes. Enriched 50Cr
tracer experiments revealed no significant Cr isotope
exchange between soluble Cr(III) and Cr(VI) at a pH of
5.5–7, at least for reaction times of days to weeks. This
result is consistent with the lack of isotope exchange
between oxygen bound in dissolved chromate CrO42� and
that of the surrounding water (Bullen et al. 2009).
Although more experiments are needed to unravel the
full complexity of Cr isotope fractionation in natural
environments, a clear picture already emerges from the
knowledge obtained so far. A survey of the stable chromium
isotopic compositions of igneous Earth reservoirs reveals a
very narrow range in d53/52CrSRM979 of �0.124 � 0.101 ‰(Schoenberg et al. 2008). Since all chromium in igneous
terrestrial reservoirs is present in the trivalent oxidation
state, the release of Cr(III) from solid rocks through chemi-
cal weathering or by hydrothermal systems at mid-ocean
ridges is not expected to cause significant Cr isotope frac-
tionation. It is thus expected that soluble Cr(III) in aqueous
environments has a Cr isotopic composition equal or very
close to the igneous Earth value of �0.124 � 0.101 ‰ in
d53/52CrSRM979. Significant Cr isotope fractionation appears
to almost exclusively be restricted to redox changes, where
both Cr oxidation and reduction produce isotopically heavy
soluble Cr(VI) and isotopically light, preferentially adsorbed
Cr(III) pools, respectively (see Fig. 7.171). This statement is
corroborated by the observation that the Cr isotopic compo-
sition of soluble Cr(VI) in the anthropogenically uncontami-
nated groundwater of the western Mojave desert has
exclusively positive d53/52CrSRM979 values between +0.7 ‰and +5.1 ‰ (Ball and Izbicki 2004; Izbicki et al. 2008).
Cr Isotopes as a Tracer for Atmospheric Oxygen
Several lines of evidence point to a rapid oxygenation of the
Earth’s atmosphere to greater than 10�5 times the present
atmospheric level (PAL) at c. 2400–2300Ma (Farquhar et al.
2000; Hannah et al. 2004; Guo et al. 2009). This event is
commonly known as the Great Oxidation Event (GOE). The
onset of oxidative weathering, however, started already
before the GOE, as is evident from the enrichment of
redox-sensitive transition metals such as Mo and Re in
approximately 2500 Ma old shales (Anbar et al. 2007;
Wille et al. 2007). These ‘whiffs’ of oxygen are also
recorded by positively fractionated d53/52CrSRM979 values
of up to +0.29 ‰ in some, but not all, 2800–2500 Ma old
banded iron formations (BIFs) reported by Frei et al. (2009),
while even much higher d53/52CrSRM979 values of up to
+4.9 ‰ were found for Neoproterozoic BIFs representing
the rise of atmospheric oxygen to PAL (Fig. 7.172). These
authors argue that the positive d53/52CrSRM979 values of some
of the BIFs result from Cr isotope fractionation caused by
partial oxidation of Cr(III) to soluble Cr(VI) in erosion
products and soils on land through a catalytic reaction with
manganese oxides. The formation of manganese oxides in
soil, on the other hand, is depending on elevated oxygen
fugacity and thus free oxygen in the atmosphere. Riverine
transport flushed the heavy, dissolved Cr(VI) into the oceans,
where it was instantly reduced to Cr(III) by the presence of
soluble Fe(II) and effectively scavenged by the resulting
Fe(III)-oxyhydroxides, thereby incorporating the heavy Cr
isotope signature into the subsequent BIF deposits.
Intriguingly, the period between 2450 and 1900 Ma is
characterised by a decline in BIF deposition. Rare occurrences
1494 M. van Zuilen and R. Schoenberg
of BIFs from this period and 2500–2450 Ma old BIFs, which
directly predate the GOE, display little or no Cr isotope
anomalies, with d53/52CrSRM979 values of only up to
c. 0.01 ‰. Frei et al. (2009) interpreted this as a return to
more reduced atmospheric oxygen levels directly after the
GOE. The sulfur-MIF record, however, is not perturbed in
2450–1900 Ma old sediments, which according to Frei et al.
(2009) indicates that this decline in atmospheric oxygen did
not reach such low levels as in the early Archaean. As an
alternative to the decline in atmospheric oxygen levels, it is,
however, also conceivable that the oxygen level remained
high after the GOE and, accordingly strong Cr isotope
fractionations, were simply missing in the few BIF samples
analysed from this era. For instance, increased levels of
organic carbon burial – such as occurred between 2100 and
1900Ma during the Shunga event (Melezhik et al. 1999, 2004)
– should have caused an increase in atmospheric oxygen level.
Furthermore, the occurrence of strongly 13C-depleted dolo-
mite concretions in these organic-rich rocks (Melezhik et al.
1999) suggest that oxidative recycling of organic matter had
already developed at this stage (Fallick et al. 2008).
It is generally accepted that from 1840 Ma onwards there
was an increased flux of sulfate to the oceans due to
enhanced oxidative weathering of continental sulfide
minerals (Canfield 1998). Frei et al. (2009) studied the
stratigraphy of the 1880–1840 Ma Gunflint Iron Formation
(Ontario, Canada) and observed an increase towards positive
d53/52CrSRM979 values, directly in line with the model of
Canfield (1998).
Although the Cr isotope data of Frei et al. (2009) give
new interesting insights into the early evolution of atmo-
spheric oxygen, details of how exactly the atmosphere-ocean
system developed during the time interval between 2400 and
1800 Ma still escape our present knowledge. The collection
of drill cores of the ICDP FAR-DEEP project provide an
excellent sample set for studying this poorly understood
period. It can be expected that the presence of an oxidative
weathering cycle in the early Palaeoproterozoic would have
been recorded by Cr isotope variation in these rocks. It
therefore provides a direct proxy for the variation in the
atmospheric oxygen level, which can be placed in the con-
text of several important events, such as the Huronian
Fig. 7.170 (a) The stability of Cr(III) and Cr(VI) complexes
illustrated in an Eh-pH diagram. (b) The fraction of the dissociation
of Cr(III) species depending on the ambient pH. (c) The dependence of
the relative abundance of Cr(VI) species from the prevailing pH
conditions and a given overall Cr(VI) concentration (Figure is modified
after Zink et al. 2010)
10 7.10 Chemical Characteristics of Sediments and Seawater 1495
Glaciation, the first appearance of red beds, the Lomagundi-
Jatuli and the Shunga events (Fig. 7.173).
Possible Implications of the FAR-DEEP Core
A brief overview is given of sample types and specific
targets in selected drill holes for future Cr isotope studies.
In line with the study of Frei et al. (2009), positive Cr isotope
ratios are to be expected in rocks that were deposited during
the GOE at c. 2400 Ma. The Polisarka Sedimentary Forma-
tion (Hole 3A) in the Imandra-Varzuga Greenstone Belt
recorded this time interval, and includes remnants of the
first global glacial events of the Huronian age. The only
age constraint of this formation is provided by the underly-
ing 2442 Ma felsic metavolcanic rocks of the Seidorechka
Volcanic Formation (Amelin et al. 1995). The carbonate
deposits in the Limestone member and the Limestone-
Greywacke-Diamictite member documented in Hole 3A
would be excellent targets to investigate whether the onset
of the oxidative continental weathering cycle is also
portrayed by positive Cr isotope anomalies in near shore,
shallow marine chemical sediments.
The lowermost part of the Kuetsj€arvi Sedimentary Forma-
tion in the Pechenga Greenstone Belt (Hole 5A) represents a
redeposited palaeo-weathering crust and “red beds” which
apparently provide the first clear record of oxidative
weathering on Earth, and therefore form an important marker
in future Cr isotope studies. The rest of the formation consists
of siliciclastic rocks, hot spring travertines, and 13C-enriched
lacustrine and marine dolostones. The general shift towards
positive d13Ccarb values represents the Lomagundi-Jatuli
event (Melezhik et al. 2005). The processes that caused this
event are still being debated and include increased burial
rates of organic matter (Baker and Fallick 1989a, b; Karhu
and Holland 1996), and a shift in the dominant autotrophic
metabolism, such as the onset of intensive biologic methane
cycling (Hayes and Waldbauer 2006), or a redox-stratified
ocean (Aharon 2005; Bekker et al. 2008). Local
amplifications of the global signals in restricted basins have
also been reported (Melezhik et al. 2005). In order to shed
light on this important perturbation of the global carbon
cycle, it would be essential to better understand the effects
of the general oxygen-based weathering cycle in this time
period. It is therefore important that Cr isotope studies be
performed directly on carbonate deposits throughout Hole
Fig. 7.171 The development of the Cr isotope compositions of the
product (B) and its corresponding reactant (A) in case of equilibrium
(Equi) and kinetic (Rayl) isotope fractionation. Note that in the case of
a kinetic isotope fractionation, product and reactant do not exchange
isotopes. As such, the instantaneous product Rayl·Binst. of a kinetic
reaction at a certain fA has a constant Cr isotopic difference to the
unreacted pool A, while the accumulating product (Rayl·Baccum.)
develops an increasing Cr isotopic difference to the unreacted pool A
(i.e. D53/52CrRayl·Baccum.- Rayl·A)
1496 M. van Zuilen and R. Schoenberg
5A (dolostones as well as travertines), in which these positive
d13Ccarb anomalies are being found. If an increased burial rate
of organic matter occurred at this time, it is to be expected
that the heavy Cr isotope signature of Cr(IV)-rich marine
surface waters would be diminished. Organic particles in
a water columnwould cause reduction of weathering-derived
Cr(VI) and subsequently act as an adsorption surface for
Cr(III). This removal process of Cr ions from the upper
water column to deep marine sediments is therefore compa-
rable to that proposed by Frei et al. (2009), where Cr(VI) is
reduced to Cr(III) by upwelling hydrothermal Fe(II) and
subsequently adsorbed on ferrihydrite particles in oxidised
surface waters followed by accumulation into banded iron
formations. The organic-rich carbonates that were deposited
during this time interval thus bear the potential to record
a shift in the surface seawater Cr isotope signature.
The Kolosjoki Sedimentary Formation (Hole 8A)
deposited shortly after 2058 Ma in the Pechenga Green-
stone Belt contains a marine carbonate succession that
postdates the positive d13Ccarb excursion of the
Lomagundi-Jatuli event (Melezhik et al. 2007). The
Dolostone member in the upper part of the stratigraphy
contains stromatolite-dominated successions that would be
important for a direct Cr isotope record of post-GOE
marine redox conditions. In addition, the formation
contains Fe-oxide-rich lithofacies including near-shore jas-
per deposits. Given the fact that banded iron formations
are rare from the 2400–1900 Ma interval in Earth history,
and the fact that positive Cr-anomalies have so far not
been observed (Frei et al. 2009), it would be of crucial
importance to study the Cr isotope variation in Fe-oxide
lithofacies of Hole 8A. Such measurements could poten-
tially either confirm or reject the possibility of a post-GOE
chemically-reduced atmosphere.
The Zaonega Formation in the Onega Basin (Holes 12A,
12B and 13A), Karelia, contains a thick succession of
organic-rich, rhythmically bedded silt- and sandstones.
These rocks, which have been deposited at around
2000 Ma, represent one of the largest accumulations of
organic material in Earth history, known as the Shunga
event (Melezhik et al. 1999, 2004). The organic-rich silt-
and sandstones, usually described as shungites in the litera-
ture, form an important target for Cr isotope analysis, since
they may have incorporated chromium derived from conti-
nental runoff and surface waters. As was explained above, it
is to be expected that organic particles would carry a
strongly positive Cr isotope anomaly. This finding, com-
bined with a diminished positive Cr isotope anomaly in
marine carbonates, would shed light on the overall burial
rates of organic carbon as well as the oxidation state of the
atmosphere-hydrosphere system 2000 Ma ago. The Zaonega
Formation therefore provides an important alternative Cr
isotope record of surface water redox processes at a time
when banded iron formations are almost absent.
Fig. 7.172 d53/52CrSRM979 values of BIFs through the geological
record as reported by Frei et al. (2009). Frei et al. (2009) identified
six stages in the BIF Cr isotope evolution (separated by the dashedlines), which these authors interpreted by an anoxygenic atmosphere in
stage 1; the first ‘whiffs’ of atmospheric oxygen recorded by Cr
isotopes but not by MIF-S and the significant increase in oxygen during
the GOE in stage 2; stage 3 may indicate a transition from Fe-enriched
to Fe-depleted oceans with a decline in atmospheric oxygen after the
GOE; the H2S dominated Proterozoic oceans of stage 5 (Canfield 1998)
and the fully oxygenated oceans of stage 6
10 7.10 Chemical Characteristics of Sediments and Seawater 1497
References
Aharon P (2005) Redox stratification and anoxia of the early
Precambrian oceans: implications for carbon isotope
excursions and oxidation events. Precambrian Res
137:207–222
Amelin YV, Heaman LM, Semenov VS (1995) U-Pb
geochronology of layered mafic intrusions in the eastern
Baltic Shield – implications for the timing and duration of
Paleoproterozoic continental rifting. Precambrian Res
75:31–46
Anbar AD, Duan Y, Lyons TW, Arnold GL, Kendall B,
Creaser RA, Kaufman AJ, Gordon GW, Scott C, Garvin J,
Buick R (2007) A whiff of oxygen before the great oxidation
event? Science 317:1903–1906
Bain DJ, Bullen TD (2005) Chromium isotope fraction-
ation during oxidation of Cr(III) by manganese oxides.
Geochim Cosmochim Acta 69:A212–A212
Baker AJ, Fallick AE (1989a) Evidence from Lewisian
limestones for isotopically heavy carbon in 2000-million-
year-old seawater. Nature 337:352–354
Baker AJ, Fallick AE (1989b) Heavy carbon in 2-billion-
year-old marbles from Lofoten-Vesteralen, Norway –
implications for the Precambrian carbon-cycle. Geochim
Cosmochim Acta 53:1111–1115
Ball JW, Izbicki JA (2004) Occurrence of hexavalent
chromium in ground water in the western Mojave Desert,
California. Appl Geochem 19:1123–1135
Bekker A, Holmden C, Beukes NJ, Kenig F, Eglingtone
B, Patterson WP (2008) Fractionation between inorganic
and organic carbon during the Lomagundi (2.22–2.1 Ga)
carbon isotope excursion. Earth Planet Sci Lett 271:278–291
Bullen T, Widory D, Petelet-Giraud E (2009) Multi-
isotope approaches for identification of metal contamination
sources in environmental systems. Geochim Cosmochim
Acta 73:A173–A173
Canfield DE (1998) A new model for Proterozoic ocean
chemistry. Nature 396:450–453
Coplen TB, Bohlke JK, De Bievre P, Ding T, Holden NE,
Hopple JA, Krouse HR, Lamberty A, Peiser HS, Revesz K,
Rieder SE, Rosman KJR, Roth E, Taylor PDP, Vocke RD,
Xiao YK (2002) Isotope-abundance variations of selected
elements – (IUPAC Technical Report). Pure Appl Chem
74:1987–2017
Ellis AS, Johnson TM, Bullen TD (2002) Chromium
isotopes and the fate of hexavalent chromium in the envi-
ronment. Science 295:2060–2062
Fallick AE, Melezhik VA, Simonson B (2008) The
ancient anoxic biosphere was not as we know it. In:
Dobretsov N, Kolchanov N, Rozanov A, Zavarzin G (eds)
Biosphere origin and evolution, Springer Science +Business
Media, LLC., 233 Spring Street, New York, NY 10013,
USA, p 169–188
Farquhar J, Bao HM, Thiemens M (2000) Atmospheric
influence of Earth’s earliest sulfur cycle. Science
289:756–758
Fig. 7.173 Overview of the Archaean-Palaeoproterozoic transition
period. The bar at the bottom represents the shift from a hydrothermal
Fe2+-dominated ocean to a sulfide-dominated ocean. The Great Oxida-
tion Event (GOE) is shown in relation to several other important events.
The top panel shows the major early Palaeoproterozoic chronostra-
tigraphic units in Fennoscandia (for details see Chap. 3.2) and position
of the FAR-DEEP drillholes
1498 M. van Zuilen and R. Schoenberg
Frei R, Gaucher C, Poulton SW, Canfield DE (2009)
Fluctuations in Precambrian atmospheric oxygenation
recorded by chromium isotopes. Nature 461:250–253
Guo QJ, Strauss H, Kaufman AJ, Schroder S, Gutzmer J,
Wing B, Baker MA, Bekker A, Jin QS, Kim ST, Farquhar J
(2009) Reconstructing Earth’s surface oxidation across the
Archean-Proterozoic transition. Geology 37:399–402
Hannah JL, Bekker A, Stein HJ, Markey RJ, Holland
HD (2004) Primitive Os and 2316 Ma age for marine
shale: implications for Paleoproterozoic glacial events
and the rise of atmospheric oxygen. Earth Planet Sci Lett
225:43–52
Hayes JM, Waldbauer JR (2006) The carbon cycle and
associated redox processes through time. Philos Trans Roy
Soc B Biol Sci 361:931–950
Izbicki JA, Ball JW, Bullen TD, Sutley SJ (2008) Chro-
mium, chromium isotopes and selected trace elements, west-
ern Mojave Desert, USA. Appl Geochem 23:1325–1352
Karhu JA, Holland HD (1996) Carbon isotopes and the
rise of atmospheric oxygen. Geology 24:867–870
Kim JG, Dixon JB, Chusuei CC, Deng YJ (2002) Oxida-
tion of chromium(III) to (VI) by manganese oxides. Soil Sci
Soc Am J 66:306–315
Lee G, Hering JG (2003) Removal of chromium(VI) from
drinking water by redox-assisted coagulation with iron(II). J
Water Supply Res Technol Aqua 52:319–332
Lugmair GW, Shukolyukov A (1998) Early solar system
timescales according to 53Mn-53Cr systematics. Geochim
Cosmochim Acta 62:2863–2886
McDonough WF (2004) Compositional model for the
Earth’s core. In: Holland HD, Turrekian KK (eds) Treatise
on geochemistry, vol 2. Elsevier, Amsterdam, pp 547–568
Melezhik VA, Fallick AE, Filippov MM, Larsen O
(1999) Karelian shungite – an indication of 2.0-Ga-old
metamorphosed oil-shale and generation of petroleum: geol-
ogy, lithology and geochemistry. Earth Sci Rev 47:1–40
Melezhik VA, Filippov MM, Romashkin AE (2004) A
giant Palaeoproterozoic deposit of shungite in NW Russia:
genesis and practical applications. Ore Geol Rev
24:135–154
Melezhik VA, Fallick AE, Rychanchik DV, Kuznetsov
AB (2005) Palaeoproterozoic evaporites in Fennoscandia:
implications for seawater sulphate, the rise of atmospheric
oxygen and local amplification of the d13C excursion. Terra
Nova 17:141–148
Melezhik VA, Huhma H, Condon DJ, Fallick AE,
Whitehouse MJ (2007) Temporal constraints on the
Paleoproterozoic Lomagundi-Jatuli carbon isotopic event.
Geology 35:655–658
Pettine M, Barra I, Campanella L, Millero FJ (1998a)
Effect of metals on the reduction of chromium (VI) with
hydrogen sulfide. Water Res 32:2807–2813
Pettine M, D’Ottone L, Campanella L, Millero FJ,
Passino R (1998b) The reduction of chromium (VI) by iron
(II) in aqueous solutions. Geochim Cosmochim Acta
62:1509–1519
Pettine M, Tonnina D, Millero FJ (2006) Chromium(VI)
reduction by sulphur(IV) in aqueous solutions. Mar Chem
99:31–41
Rai D, Sass BM, Moore DA (1987) Chromium(III)
hydrolysis constants and solubility of chromium(III) hydrox-
ide. Inorg Chem 26:345–349
Sander S, Koschinsky A (2000) Onboard-ship redox spe-
ciation of chromium in diffuse hydrothermal fluids from the
North Fiji Basin. Mar Chem 71:83–102
Schauble E, Rossman GR, Taylor HP (2004) Theoretical
estimates of equilibrium chromium-isotope fractionations.
Chem Geol 205:99–114
Schoenberg R, Zink S, Staubwasser M, von
Blanckenburg F (2008) The stable Cr isotope inventory of
solid Earth reservoirs determined by double spike MC-ICP-
MS. Chem Geol 249:294–306
Sikora ER, Johnson TM, Bullen TD (2008) Microbial
mass-dependent fractionation of chromium isotopes.
Geochim Cosmochim Acta 72:3631–3641
Sirinawin W, Turner DR, Westerlund S (2000) Chro-
mium(VI) distributions in the Arctic and the Atlantic Oceans
and a reassessment of the oceanic Cr cycle. Mar Chem
71:265–282
Trinquier A, Birck JL, Allegre CJ (2007) Widespread54Cr heterogeneity in the inner solar system. Astrophys J
655:1179–1185
Walter MJ (2004) Melt extraction and compositional
variability in mantle lithosphere. In: Holland HD, Turrekian
KK (eds) Treatise on geochemistry, vol 2. Elsevier,
Amsterdam, pp 363–394
Weaver RM, Hochella MF (2003) The reactivity of seven
Mn-oxides with Cr-aq(3+): a comparative analysis of a
complex, aq environmentally important redox reaction. Am
Miner 88:2016–2027
Wilkin RT, Su CM, Ford RG, Paul CJ (2005) Chromium-
removal processes during groundwater remediation by a
zerovalent iron permeable reactive barrier. Environ Sci
Technol 39:4599–4605
Wille M, Kramers JD, Nagler TF, Beukes NJ, Schroder S,
Meisel T, Lacassie JP, Voegelin AR (2007) Evidence for a
gradual rise of oxygen between 2.6 and 2.5 Ga from Mo
isotopes and Re-PGE signatures in shales. Geochim
Cosmochim Acta 71:2417–2435
Zink S, Schoenberg R, Staubwasser M (2010) Isotopic
fractionation and reaction kinetics between Cr(III) and Cr
(VI) in aqueous media. Geochim Cosmochim Acta
74:5729–5745
10 7.10 Chemical Characteristics of Sediments and Seawater 1499
7.10.6. Mo and U Geochemistry and Isotopes
Francois L.H. Tissot, Nicolas Dauphas,Christopher T. Reinhard, Timothy W. Lyons,Dan Asael, and Olivier Rouxel
The molybdenum and uranium isotopic systems can poten-
tially be used to reconstruct the oxidation state of an ancient
the global ocean as recorded in the FAR-DEEP samples. As
previous studies have shown (e.g. Anbar et al. 2007; Kendall
et al. 2010), the use of multiple proxies can help lift
uncertainties and avoid the pitfalls associated with the use
of a single redox proxy. Molybdenum is the most abundant
transition metal in modern seawater, occurring dominantly
as the molybdate anion (MoO42�; Morford and Emerson
1999). Along with many other transition metals, Mo
becomes authigenically enriched in sulphidic environments,
where molybdate is converted to particle-reactive oxythio-
molybdates (MoOxS4-x2�) and sequestered in sediments
(Erickson and Helz 2000; Helz et al. 1996; Tribovillard
et al. 2004). Although well-oxygenated marine sediments
are characterised by approximately crustal Mo concen-
trations (~1–2 ppm), and reducing sediments overlain by
oxygenated water do not typically show Mo concentrations
in excess of ~10–20 ppm, most modern and Phanerozoic
euxinic sediments show high Mo concentrations (often in
excess of 100s of ppm) that correlate strongly with the total
organic carbon (TOC) content of the sediments. However,
scavenging can be efficient enough in some environments to
remove Mo from solution quantitatively (Erickson and Helz
2000), such as in the modern Black Sea, so that muted
euxinic enrichments can also reflect drawdown of the Mo
inventory on an oceanic scale during times of widespread
oxygen deficiency (Algeo and Lyons 2006; Emerson
and Huested 1991; Scott et al. 2008), as suggested for the
Proterozoic (Canfield 1998; Lyons et al. 2009). Moly-
bdenum depletion can have consequences for biological
pathways that are dependent on Mo, such as nitrogen fixa-
tion (Anbar and Knoll 2002; Zerkle et al. 2006) and inor-
ganic nitrogen assimilation (Milligan and Harrison 2000).
Recent work (Scott et al. 2008) showed a dramatic increase
in Mo concentration in euxinic shales at ~2.2–2.1 Ga, around
200 million years after the Great Oxidation Event (GOE).
While almost certainly tied to an increased weathering
flux of Mo to the ocean, the existing deep marine record is
sparse in this interval, and ocean redox conditions for the
early Palaeoproterozoic in general are not well known.
In sum, Mo concentration relationships provide an additional
constraint on local palaeoenvironments. When viewed at the
same time from the perspective of global mass balance,
magnitudes of Mo enrichment in black shales can speak to
ocean-scale palaeoredox.
The ocean has a homogeneous present-day d98Mo value
(expressed in conventional ‘delta’ notation, where d98Mo
¼ 1,000 � [(98Mo/95Mo)sample/(98Mo/95Mo)standard – 1]) of
about +2.4 ‰ relative to the Specpure®Mo plasma standard
(Barling et al. 2001; Siebert et al. 2003). Although
uncertainties still remain on the modern oceanic budget of
Mo isotopes (e.g., the Mo isotopic composition of rivers and
estuarine systems, the role of suboxic sediments with respect
to Mo removal from the ocean), Mo-isotopes are promising
tracers of ocean-scale palaeoredox conditions (e.g. Anbar
2004; Anbar and Rouxel 2007; Arnold et al. 2004; Siebert
et al. 2003). Also, new constraints are emerging, for exam-
ple, for the riverine contribution and Mo cycling and
associated isotope effects in suboxic settings (Archer and
Vance 2008; Poulson et al. 2006).
Mo-isotope fractionation is associated with the transition
between the conservative molybdate ion (MoO4�2) and the
more reactive thiomolybdate (MoS4�2) species through
intermediate oxythiomolybdate species (MoOxS(4-x)�2).
The total reaction can be written as follows:
MoO4�2 þ 4H2S aqð Þ $ MoS4
�2 þ 4H2O:
The equilibrium constant of this reaction series indicates
that oxythiomolybdate species coexist in solution only at a
very narrow range of [H2S], from about 2 � 10�5 to
5 � 10�5 molar (Helz et al. 1996; Fig. 7.174). In other
words, in most natural systems, Mo will be dominantly
present either as molybdate or thiomolybdate.
Molybdenum is typically transferred to the oceans as
molybdate and remains in solution as long as H2S concentra-
tion is low. If conditions become euxinic, Mo will be trans-
ferred to the reactive thiomolybdate form and can be
quantitatively removed from the solution to the sediments
(Algeo and Lyons 2006). In such cases, the Mo isotopic
composition of the sediments will represent that of the source
solution. For that reason, it is expected that sedimentation of
black shales under sulphidic conditions can faithfully record
the Mo-isotope composition of seawater (Arnold et al. 2004;
Barling et al. 2001; Siebert et al. 2003). Because adsorption
of Mo onto Mn oxyhydroxide phases in oxic settings imparts
a strong negative fractionation (Barling and Anbar 2004), the
isotopic composition of seawater loosely reflects the balance
between oxic and euxinic sequestration (and the varying role
of suboxia) and thus approximates the areal extent of each
environment in the global ocean, permitting us to place
unique constraints on the relative balance between anoxic
(i.e. euxinic) and oxic removal of Mo from past oceans.
F.L.H. Tissot (*)
Origins Lab, Department of the Geophysical Sciences and Enrico
Fermi Institute, The University of Chicago, 5734 South Ellis Avenue,
Chicago, IL 60637, USA
1500 F.L.H. Tissot et al.
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_10, # Springer-Verlag Berlin Heidelberg 2013
1500
Figure 7.175 summarises several studies of Mo isotopes
in black shales and similar sediments. Samples older than
2.8 Ga show the same Mo isotopic compositions as their
continental crust source rocks, indicating a purely detrital
input of Mo to the oceans occurring in a no free oxygen
environment (Siebert et al. 2005; Wille et al. 2007). Black
shale from 2.5 to 2.7 Ga show general increase in Mo
concentration and fractionated isotopic compositions com-
pared to the continental crust and older sediments, indicating
a gradual rise of oxygen during this period (Wille et al.
2007). The large Mo isotopic variations observed in these
samples could reflect local fluctuations in the redox
conditions (Siebert et al. 2005; Wille et al. 2007). The
euxinic sediments of 2.3–2.2 Ga show Mo isotopic compo-
sition similar to that of their source and apparently represent
quantitative or near quantitative scavenging of Mo at euxinic
bottom water. Younger euxinic sediments typically repre-
sent the Mo isotopic composition of seawater where signifi-
cant Mo isotopic variations are related to redox events such
as the Toarcian Oceanic Anoxic Event (OAE) (Arnold et al.
2004; Kendall et al. 2009; Pearce et al. 2008). Recent
euxinic sediments from the Black Sea and the Cariaco
Trench represent the seawater value where recent oxic
sediments show much lower d98Mo values (Arnold et al.
2004; Barling et al. 2001).
The combined use of Mo and U will allow the definition
of a time window for the duration of the oxygen rise during
the GOE, as U needs more oxidising conditions to be
mobilised. In the modern oxic ocean, uranium occurs mainly
as the uranyl carbonate ion UVIO2(CO3)34� (Djogic et al.
1986) and is conservative with a concentration of 13.9 � 0.9
nmol/L (Chen et al. 1986). Several estimates of the U budget
for the modern ocean have been reported (Morford and
Emerson 1999; Dunk et al. 2002; Henderson and Anderson
2003), and generally agree within uncertainties. According
to Dunk et al. (2002), three main sinks, which represent
roughly similar removal rates of U, are identified: biogenic
carbonates, anoxic sediments and suboxic sediments
(suboxic sediment is defined as O2 <10 mmol/L and H2S
<10 mmol/L, Murray et al. 2007; Berner 1981). The U
accumulation rate in suboxic sediments is one order of
magnitude smaller than in anoxic sediments but this is bal-
anced by the greater areal extent of suboxic sediments. It is
worth noting that the notion of suboxic sediments is poorly
defined, which has led to the introduction of a new classifi-
cation scheme (Canfield and Thamdrup 2009). Prior to the
GOE, anoxic sediments dominated the ocean and little oxi-
dative weathering of U took place as suggested by the low U
abundance recorded in late Archaean shales (Kendall et al.
2010). This is supported by recent calculations (Sverjensky
and Lee 2010), which showed that the enrichment in Mo and
Re observed in Archaean shales (Anbar et al. 2007) is
consistent with local pulses in the production of O2 by
photosynthesis, that mobilised Mo and Re but did not
oxidise the atmosphere enough for U to be mobilised.
Archaean sediments are further distinguished by the presence
of detrital uraninite (UIVO2) not found in later deposits,
providing evidence of low atmospheric oxygen at that time
(Ramdohr 1958; Rasmussen and Buick 1999; Frimmel 2005).
Uranium ultimately decays into lead and it is commonly
assumed in Pb-Pb dating that the 238U/235U ratio is constant.
Early isotopic studies estimated the value of the 238U/235U
ratio to be ~139 (Lounsbury 1956; Nier 1939; Senftle et al.
1957). In the late 1970s, U isotopic abundances were
measured more precisely in uranium ore deposits (Cowan
and Adler 1976) and meteorites (Arden 1977; Tatsumoto
and Shimamura 1980) using thermal ionisation mass spec-
trometry (TIMS). Excess 235U up to 200 % was found but
this was not confirmed by subsequent studies (Chen and
Wasserburg 1980, 1981; Shimamura and Lugmair 1984a,
b). The research triggered by these early investigations
stopped in the 1980s because of limited analytical precision,
and resumed in the middle 1990s when Multi-Collector
Inductively Coupled Plasma Mass Spectrometer (MC-
ICPMS) improved the precision from � 4 ‰ to � 0.2 ‰.
With such instrumentation Bopp et al. (2009), Stirling et al.
Fig. 7.174 The transition between molybdate (MoO4�2) and thiomolybdate (MoS4
�2) species and their isotopic composition as a function of
aH2S (Fractionation factor taken from Tossell (2005))
10 7.10 Chemical Characteristics of Sediments and Seawater 1501
(2007), and Weyer et al. (2008) reported 238U/235U
variations that seemed to be related to the redox state of
the depositional environment (Fig. 7.176).
Experimental results and theory (Abe et al. 2008;
Bigeleisen 1996; Fujii et al. 1989; Schauble 2006, 2007)
explain these variations as being due to a mass-independent,
volume-dependent fractionation mechanism called Nuclear
Field Shift (NFS – the difference in nuclear charge radius
associated with the number of neutrons shifts the atomic
energy levels of the isotopes). For U, the implication of the
NFS is a preferential incorporation of the heavy isotope
(238U) into the reduced species U(IV). However, redox reac-
tion is not the only process inducing d238U fractionation and
it has been shown (Brennecka et al. 2008; Wasylenki et al.
2010) that adsorption onto Mn oxyhydroxide fractionates U
towards lighter d238U in ferromanganese sediments due to a
difference in the U-O coordination shell between dissolved
and absorbed U.
From Fig. 7.176, it can be seen that the main sources
(continental crust, i.e. granites, basalts) and two of the main
sinks (suboxic sediments and biogenic carbonates, i.e. corals)
of U in the modern oceanic mass balance have isotopic
compositions close to that of modern seawater. Little U isoto-
pic fractionation is thought to occur during weathering and
transport of U from the land to the ocean. Modern
hydrogeneous Mn-crusts and 2.9–2.2 Ga BIFs have lower
isotopic compositions (down to ~ �0.9 ‰ relative to
CRM112) and black shales from both modern and ancient
ocean show heavier isotopic compositions (up to +0.43 ‰).
Although biogenic carbonate shows no fractionation relative
to seawater, Herrmann et al. (2010a) observed that non-bio-
genic carbonates show slightly heavier compositions relative
to seawater (~ � 0.26 ‰). Two recent studies used U
isotopes to investigate the intensity of oxidative weathering
in the late Archaean and the extent of marine anoxia in the
mid-Cretaceous Ceromanian-Turomian OAE (Kendall et al.
2010; Montoya-Pino et al. 2010). Kendall et al. (2010)
analysed late Archaean shales and found that d238U composi-
tion correlated with authigenic enrichments of Re andMo, but
notU. The enrichment patterns imply the presence ofO2 in the
atmosphere at a level sufficient for oxidative weathering of Re
andMo to take place, but too low to mobilise U. Additionally,
the preservation of a heavy d238U signature in the sediments
indicates the survival of enough dissolved U during shallow-
to-deep water transport prior to d238U fractionation in the
sediments, i.e. seawater was oxidised enough for U to be in
its soluble form, U(VI). Data point toward mild oxygenation
of the late Archaean surface ocean. Montoya-Pino et al.
(2010) proposed the first quantitative reconstruction of the U
mass balance of the ocean. They speculated that the ocean
prior to the OAE was slightly anoxic compared to today’s
ocean, and that the OAE corresponds to a global increase of
oceanic anoxia by at least a factor of three compared to the
present day.
The detailed study of sedimentologic-stratigraphic
variations of the redox state of the ocean and atmosphere as
it evolved during and after the GOE is made possible because
the Fennoscandian Shield sedimentary rocks are exceptionally
well-preserved and diverse. The work devoted to Mo and U
isotopes aims at establishing the pace of ocean/atmosphere
oxygenation. However, both systems suffer limitations. The
Mo-isotope record in black shales can be affected by local
redox conditions (Poulson et al. 2006), and local palaeocea-
nographic conditions need to be taken into account for U, as
Fig. 7.175 d98Mo vs. time diagram. Isotopic compositions are reported relative to the Johnson Matthey Specpure® Mo plasma standard (Data
taken from Arnold et al. (2004), Barling et al. (2001), Kendall et al. (2009), Pearce et al. (2008), Siebert et al. (2005) and Wille et al. (2007))
1502 F.L.H. Tissot et al.
Fig.7.176
Measurements
oftheisotopic
compositionofU
indifferentmedia
throughtime.
d238U
¼((238U/235U) sample/(238U/235U) reference�1
)�
1,000.Theintroductionofmulti-collector
ICP-M
Spermittedtheresolutionofvariationsin
Uisotopes
assm
allas
0.1
‰.Theblack
horizontalline(0
‰)correspondstotheisotopiccompositionoftheCRM112areference
material
(alsonam
edSRM960,N
BL112aandCRM145)with238U/235U
¼137.880137.837(Richteretal.
(2010).Thebluelineistheisotopiccompositionofseaw
ater
determined
byWeyer
etal.(2008).
Theinsetdiagram
showsazoom
ofthemeasurements
since
2007,showingtheenrichmentof
reduced/anoxic
sedim
ents
in238U
relative
toseaw
ater.Other
sources/sinkshave
isotopic
compositionssimilar
toorlower
than
seaw
ater
(Datafrom
Boppetal.(2009),Cowan
andAdler
(1976),Lounsbury
(1956),Montoya-Pinoetal.(2010),Nier(1939),Senftleetal.(1957),Stirling
etal.(2007)andWeyer
etal.(2008))
10 7.10 Chemical Characteristics of Sediments and Seawater 1503
correlated d238U-d15N variations observed in epicontinental
black shales indicate that local effectsmight obscure the global
signal (Herrmann et al. 2010b). In general, it is expected that
an increase in ocean oxygenation during the Palaeoproterozoic
(i.e. expansion of oxic Mo sink) will produce significant
increase in d98Mo and d238U values measured in black shales.
Comparing the d98Mo values with other geochemical tracers
and proxies for euxinia in porewater or seawater will allow us
to overcome the limitations of each system individually and
identify local effects, thus strengthening interpretation of the
isotopic record. Since Re is enriched in mildly reducing,
suboxic sediments, close to or slightly later in the redox
sequence than U, and before reduction of Mo in more reduc-
ing, sulphidic sediments (Crusius et al. 1996), we expect
various degrees of Mo-U-Re enrichment and Mo/U-isotope
fractionation due to the relative effects of oxic weathering vs.
expansion of oxic and/or suboxic conditions in continental
shelf environments during the rise of atmosphere oxygen.
References
Abe M, Suzuki T, Fujii Y, Hada M, Hirao K (2008) An ab
initio molecular orbital study of the nuclear volume effects
in uranium isotope fractionations. J Chem Phys 129:164309
Algeo TJ, Lyons TW (2006) Mo-total organic carbon
covariation in modern anoxic marine environments: implica-
tions for analysis of paleoredox and paleohydrographic
conditions. Paleoceanography 21:PA1016
Anbar AD (2004) Molybdenum stable isotopes:
observations, interpretations and directions. In: Johnson
CM, Beard BL, Albarede F (eds) Geochemistry of non-
traditional stable isotopes, vol 55, Reviews in Mineralogy
& Geochemistry. Mineralogical Society of America,
Washington, DC, pp 429–454
Anbar AD, Knoll AH (2002) Proterozoic ocean chemistry
and evolution: a bioinorganic bridge? Science 297:1137–1142
Anbar AD, Rouxel O (2007) Metal stable isotopes in
paleoceanography. Annu Rev Earth Planet Sci 35:717–746
Anbar AD, Duan Y, Lyons TW, Arnold GL, Kendall B,
Creaser RA, Kaufman AJ, Gordon GW, Scott C, Garvin J,
Buick R (2007) A whiff of oxygen before the great oxidation
event? Science 317:1903–1906
Archer C, Vance D (2008) The isotopic signature of the
global riverine molybdenum flux and anoxia in the ancient
oceans. Nat Geosci 1:597–600
Arden JW (1977) Isotopic composition of uranium in
chondritic meteorites. Nature 269:788–789
Arnold GL, Anbar AD, Barling J, Lyons TW (2004)
Molybdenum isotope evidence for widespread anoxia in
mid-Proterozoic oceans. Science 304:87–90
Barling J, Anbar AD (2004) Molybdenum isotope frac-
tionation during adsorption by manganese oxides. Earth
Planet Sci Lett 217:315–329
Barling J, Arnold GL, Anbar AD (2001) Natural mass-
dependent variations in the isotopic composition of molyb-
denum. Earth Planet Sci Lett 193:447–457
Berner RA (1981) A new geochemical classification of
sedimentary environments. J Sediment Res 51:359–365
Bigeleisen J (1996) Nuclear size and shape effects in
chemical reactions. Isotope chemistry of the heavy elements.
J Am Chem Soc 118:3676–3680
BoppCJ, LundstromCC, Johnson TM,Glessner JJG (2009)
Variations in U-238/U-235 in uranium ore deposits: isotopic
signatures of the U reduction process? Geology 37:611–614
Brennecka GA, Wasylenki LE, Weyer S, Anbar AD
(2008) Experiments demonstrate that uranium isotopes
fractionate during adsorption to Mn-oxides. Geochim
Cosmochim Acta 72:A114, Suppl
Canfield DE (1998) A new model for Proterozoic ocean
chemistry. Nature 396:450–453
Canfield DE, Thamdrup B (2009) Towards a consistent
classification scheme for geochemical environments, or,
why we wish the term ‘suboxic’ would go away. Geobiology
7:385–392
Chen JH, Wasserburg GJ (1980) A search for isotopic
anomalies in uranium. Geophys Res Lett 7:275–278
Chen JH, Wasserburg GJ (1981) Isotopic determination
of uranium in picomole and subpicomole quantities. Anal
Chem 53:2060–2067
Chen JH, Edwards RL, Wasserburg GJ (1986) 238U, 234U
and 232Th in seawater. Earth Planet Sci Lett 80:241–251
Cowan GA, Adler HH (1976) The variability of the
natural abundance of 235U. Geochim Cosmochim Acta
40:1487–1490
Crusius J, Calvert S, Pedersen T, Sage D (1996) Rhenium
and molybdenum enrichments in sediments as indicators of
oxic, suboxic and sulfidic conditions of deposition. Earth
Planet Sci Lett 145:65–78
Djogic R, Sipos L, Branica M (1986) Characterization of
uranium(VI) in seawater. Limnol Oceanogr 31:1122–1131
Dunk RM, Mills RA, Jenkins WJ (2002) A reevaluation
of the oceanic uranium budget for the Holocene. Chem Geol
190:45–67
Emerson SR, Huested SS (1991) Ocean anoxia and the
concentrations of molybdenum and vanadium in sea-water.
Mar Chem 34:177–196
Erickson BE, Helz GR (2000) Molybdenum(VI) specia-
tion in sulfidic waters: stability and lability of thiomoly-
bdates. Geochim Cosmochim Acta 64:1149–1158
Frimmel HE (2005) Archaean atmospheric evolution:
evidence from the Witwatersrand gold fields, South Africa.
Earth Sci Rev 70:1–46
Fujii Y, Nomura M, Okamoto M, Onitsuka H, Kawakami
F, Takeda K (1989) An anomalous isotope effect of 235U in
U(IV)-U(VI) chemical exchange. Z Naturforsch A
44:395–398
Helz GR, Miller CV, Charnock JM, Mosselmans JFW,
Pattrick RAD, Garner CD, Vaughan DJ (1996) Mechanism
1504 F.L.H. Tissot et al.
of molybdenum removal from the sea and its concentration
in black shales: EXAFS evidence. Geochim Cosmochim
Acta 60:3631–3642
HendersonGM,AndersonRF (2003) TheU-series toolbox
for paleoceanography. Rev Miner Geochem 52:493–531
Herrmann AD, Wasylenki LE, Anbar AD (2010a) Ura-
nium isotopic composition of carbonate sediments as a
potential redox proxy. Geologycial Society of America
Abstracts with programs 42:514
Herrmann AD, Algeo TJ, Anbar AD (2010b) Do uranium
isotopes of epicontinental black shales reflect global oceanic
conditions?. Geologycial Society of America Abstracts with
programs 42:467
Kendall B, Creaser RA, Gordon GW, Anbar AD (2009)
Re-Os and Mo isotope systematics of black shales from the
Middle Proterozoic Velkerri and Wollogorang Formations,
McArthur Basin, northern Australia. Geochim Cosmochim
Acta 73:2534–2558
Kendall B, Brennecka G, Weyer S, Anbar AD (2010)
Onset of oxidative uranium mobilization on the late
Archaean earth? Astrobiology Science Conference
Lounsbury M (1956) The natural abundances of the ura-
nium isotopes. Can J Chem 34:259–264
Lyons TW, Reinhard CT, Scott C (2009) Redox redux.
Geobiology 7:489–494
Milligan AJ, Harrison PJ (2000) Effects of non-steady-
state iron limitation on nitrogen assimilatory enzymes
in the marine diatom Thalassiosira weissflogii (Bacillar-
iophyceae). J Phycol 36:78–86
Montoya-Pino C, Weyer S, Anbar AD, Pross J,
Oschmann W, van de Schootbrugge B, Arz HW (2010)
Global enhancement of ocean anoxia during Oceanic Anoxic
Event 2: a quantitative approach using U isotopes. Geology
38:315–318
Morford JL, Emerson S (1999) The geochemistry of
redox sensitive trace metals in sediments. Geochim
Cosmochim Acta 63:1735–1750
Murray JW, Stewart K, Kassakian S, Krynytzky M,
DiJulio D (2007) Oxic, suboxic, and anoxic conditions in
the Black Sea. In: Yanko-Hombach V, Gilbert AS, Panin N,
Dolukhaov PM (eds) The Black Sea flood question: changes
in coastline, climate, and human settlement. Springer,
Dordrecht, p 1–21
Nier AO (1939) The isotopic constitution of uranium and
the half-lives of the uranium isotopes. I. Phys Rev
55:0150–0153
Pearce CR, Cohen AS, Coe AL, Burton KW (2008)
Molybdenum isotope evidence for global ocean anoxia cou-
pled with perturbations to the carbon cycle during the early
Jurassic. Geology 36:231–234
Poulson RL, Siebert C, McManus J, Berelson WM (2006)
Authigenic molybdenum isotope signatures in marine
sediments. Geology 34:617–620
Ramdohr P (1958) New observations on the ores of the
Witswatersrand in South Africa and their genetic signifi-
cance. Geol Soc S Afr Trans 61:1–50
Rasmussen B, Buick R (1999) Redox state of the Archean
atmosphere: evidence from detrital heavy minerals in ca.
3250–2750 Ma sandstones from the Pilbara Craton,
Australia. Geology 27:115–118
Richter S, Eykens R, Kuhn H, Aregbe Y, Verbruggen A,
Weyer S (2010) New average values for the n(238U)/n
(235U) isotope ratios of natural uranium standards. Int J
Mass Spectrom 295:94–97
Schauble EA (2006) Equilibrium uranium isotope frac-
tionation by nuclear volume and mass-dependent processes,
AGU Fall Meeting
Schauble EA (2007) Role of nuclear volume in driving
equilibrium stable isotope fractionation of mercury, thal-
lium, and other very heavy elements. Geochim Cosmochim
Acta 71:2170–2189
Scott C, LyonsTW,BekkerA, ShenY, Poulton SW,ChuX,
Anbar AD (2008) Tracing the stepwise oxygenation of the
Proterozoic ocean. Nature 452:456–459
Senftle FE, Stieff L, Cuttitta F, Kuroda PK (1957) Com-
parison of the isotopic abundance of U235 and U238 and the
radium activity ratios in Colorado Plateau uranium ores.
Geochim Cosmochim Acta 11:189–193
Shimamura T, Lugmair GW (1984a) U-isotopic
abundances. Proc Lunar Planet Sci Conf 12:976–978
Shimamura T, Lugmair GW (1984b) Uranium isotopic
abundance in Allende residue. Proc Lunar Planet Sci Conf
15:776–777
Siebert C, N€alger TF, von Blanckenburg F, Kramers JD
(2003) Molybdenum isotope recordes as a potential new
proxy for paleoceanography. Earth Planet Sci Lett
211:159–171
Siebert C, Kramers JD, Meisel T, Morel P, N€agler TF
(2005) PGE, Re-Os, and Mo isotope systematics in
Archaean and early Proterozoic sedimentary systems as
proxies for redox conditions of the early Earth. Geochim
Cosmochim Acta 69:1787–1801
Stirling CH, Andersen MB, Potter EK, Halliday AN
(2007) Low-temperature isotopic fractionation of uranium.
Earth Planet Sci Lett 264:208–225
Sverjensky DA, Lee N (2010) The great oxidation event
and mineral diversification. Elements 6:31–36
Tatsumoto M, Shimamura T (1980) Evidence for live247Cm in the early solar system. Nature 286:118–122
Tossell JA (2005) Calculating the partitioning of the
isotopes of Mo between oxidic and sulfidic species
in aqueous solution. Geochim Cosmochim Acta
69:2981–2993
Tribovillard N, Riboulleau A, Lyons T, Baudin FO
(2004) Enhanced trapping of molybdenum by sulfurized
10 7.10 Chemical Characteristics of Sediments and Seawater 1505
marine organic matter of marine origin in Mesozoic
limestones and shales. Chem Geol 213:385–401
Wasylenki LE, Brennecka GA, Bargar JR, Anbar AD
(2010) The mechanism of uranium isotope fractionation
during adsorption to Mn oxyhydroxide. Geochim
Cosmochim Acta 74:A1116, Suppl
Weyer S, Anbar AD, Gerdes A, Gordon GW, Algeo TJ,
Boyle EA (2008) Natural fractionation of U-238/U-235.
Geochim Cosmochim Acta 72:345–359
Wille M, Kramers JD, Nagler TF, Beukes NJ, Schroder S,
Meisel T, Lacassie JP, Voegelin AR (2007) Evidence for a
gradual rise of oxygen between 2.6 and 2.5 Ga from Mo
isotopes and Re-PGE signatures in shales. Geochim
Cosmochim Acta 71:2417–2435
Zerkle AL, House CH, Cox RP, Canfield DE (2006)
Metal limitation of cyanobacterial N-2 fixation and
implications for the Precambrian nitrogen cycle. Geobiology
4:285–297
7.10.7 Re-Os Isotope Geochemistry
Judith L. Hannah and Holly J. Stein
Introduction
Charting the growth of atmospheric oxygen and the comple-
mentary evolution of the biosphere requires both accurate
time pins in the stratigraphic record and proxies for the
changing geochemical cycles in Earth’s surface and near
surface environments. The redox-sensitive metals rhenium
(Re) and osmium (Os) work in partnership to meet both of
these requirements. Both elements are concentrated in
organic matter and sulphides; that is, they are notably
enriched in organic carbon-rich sedimentary rocks (ORS).
Both are mobilised under oxidising conditions and fixed by
reduction; that is, an oxygenated atmosphere is necessary to
release Re and Os by chemical weathering and transport
them from the continents to the oceans. Thus, changing
atmospheric conditions are reflected in changing Re and Os
cycling in the hydrosphere and lithosphere. Significantly,187Re decays to 187Os with a half-life of 41.6 billion years,
providing a geochronometer for dating ORS and related
hydrocarbons. Moreover, the steady decay of 187Re produces
increasing 187Os/188Os over time, recording the time-
integrated Re/Os ratio and providing a tracer for movement
of the metals between geochemical reservoirs. Re-Os iso-
tope geochemistry, therefore, gives us a timelines for
changes in redox conditions and weathering rates in surface
environments. Re-Os data, combined with other geochemi-
cal proxies described in this section, help define the history
of Earth’s environmental changes through time.
Geochemical Reservoirs for Re and Os
There are two naturally occurring isotopes of Re: stable185Re and radioactive 187Re. Mass fractionation of the two
Re isotopes is readily observed during the extreme
conditions imposed by mass spectrometry (e.g. Suzuki
et al. 2004; Zimmerman et al. 2007). The few reported
natural variations, however, are less than 1 ‰ (Miller et al.
2009), and thus introduce errors less than other sources of
uncertainty for Re-Os geochronology. As with other high
mass elements used in geochronology, we therefore assume
uniform present-day abundances of 185Re (37.40 %) and187Re (62.60 %; Gramlich et al. 1973) as an underpinning
assumption for radiometric dating.
There are seven naturally occurring isotopes of Os: 184Os,186Os, 187Os, 188Os, 189Os, 190Os, and 192Os. 187Os is the
product of beta-decay of 187Re, with a decay constant of
1.666 � 10�11 a�1 (Smoliar et al. 1996). 186Os is the prod-
uct of alpha-decay of 190Pt with a decay constant of
1.542 � 10�12 a�1 (Walker et al. 1997). The abundances
of the remaining isotopes may be considered constant for
present day Re-Os isotope geochemistry and geochronology.
Current convention is to normalise abundances of 187Os to188Os, reporting isotopic variations in terms of the ratio187Os/188Os.
In order to quantify Re-Os systematics in surface
environments, we must specify the concentrations and isoto-
pic compositions of Re and Os in Earth’s reservoirs and
define the interactions among those reservoirs (Fig. 7.177).
Both Re and Os are enriched in the mantle relative to the
crust. Os is highly compatible and Re mildly incompatible
during mantle melting. Hence, 187Re/188Os ratios are gener-
ally high in crustal materials. 187Re/188Os varies over more
than an order of magnitude in common reservoirs, being
about 0.4 in chondritic mantle (Meisel et al. 1996), about
50 in average currently eroding continental crust (Esser and
Turekian 1993), and 100–1,000 (and as high as 6,000) in
ORS (Georgiev et al. 2011). Thus, the mantle has a charac-
teristically low 187Os/188Os ratio (present day 0.113; Shirey
and Walker 1998), whereas crustal reservoirs have higher187Os/188Os because they accumulate radiogenic 187Os over
time from decay of 187Re in high Re/Os materials.
The largest crustal reservoirs of Re and Os are those rich in
sulphides and/or organic material. Sulphideminerals (exclud-
ing the exceptional case of molybdenite) commonly have Re
1506 J.L. Hannah and H.J Stein
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_10, # Springer-Verlag Berlin Heidelberg 2013
1506
and Os concentrations up to 5 ppm and 50 ppb, respectively
(summarised in Hannah and Stein 2002). ORS have Re
concentrations of about 0.5 to several hundred ppb and Os
concentrations of 50 ppt to several ppb (e.g. Kendall et al.
2009a, Georgiev et al. 2011). In both cases, 187Os/188Os ratios
may be extremely high, especially in ancient materials. Two
important special cases have unique Re-Os signatures.
Molybdenite, on its formation, has high Re concentrations
(10s to 1000s ppm) but vanishingly lowOs concentrations; its
Os as measured today is therefore almost 100 % radiogenic187Os (Stein et al. 2001, 2003). In contrast, ultramafic rocks
have sulphides with elevated Os concentrations (10s ppb or
even 10s ppm) and low 187Re/188Os (commonly less than 0.4);
their 187Os/188Os ratios therefore may fall below depleted
mantle ratios (c. 0.125 today; Shirey and Walker 1998).
Given the overwhelming predominance of high 187Os/188Os
materials in exposed continental crust, present-day riverine
input of Os to seawater is predictably radiogenic (e.g.187Os/188Os � 1.26, Esser and Turekian 1993; Peucker-
Ehrenbrink and Ravizza 2000; Chen et al. 2006).
There are two principal sources of non-radiogenic Os for
the Earth’s hydrosphere and surface environments – high-
temperature mantle-derived submarine hydrothermal fluids
and cosmic dust. Large bolide impacts produce an abrupt
plunge in seawater 187Os/188Os followed by a rapid recovery
to pre-impact values (Paquay et al. 2008, and references
therein). Local depressions have been detected in the Os
concentration in seawater and the 187Os/188Os ratio in
sediments proximal to hydrothermal plumes (Woodhouse
et al. 1999; Cave et al. 2003), but this chondritic Os influx
is quickly mixed with background seawater Os (Sharma
et al. 2007). Cosmic dust is a significant source of Os in
surface environments, delivering an annual flux perhaps
comparable to that of hydrothermal fluids.
Modern seawater has an Os concentration of about
0.01 ppt, with a 187Os/188Os ratio of about 1.06 and187Re/188Os of about 4,200 (Sharma et al. 1997; Levasseur
et al. 1998; Peucker-Ehrenbrink and Ravizza 2000). The Os
isotopic composition is a composite of inputs from continen-
tal, hydrothermal, and cosmic sources. Mass balance
suggests that, under present-day atmospheric conditions,
about 80 % of the Os in seawater is derived from the
continents, with the remaining non-radiogenic Os derived
from hydrothermal circulation at mid-ocean ridges and cos-
mic dust (Sharma et al. 1997). Under an anoxic Archaean
atmosphere, input from continental sources would be greatly
reduced; further, a significant lag time is expected between
the onset of oxidative weathering with the rise of atmo-
spheric oxygen and the increase in seawater 187Os/188Os
(Hannah et al. 2004).
ORS inherit their Re and Os from the water in which
they are deposited, and/or by exchange with pore waters
after deposition. Therefore, the principal control on Re-Os
geochemistry of marine ORS is the composition of seawa-
ter. The residence time of Os in modern oceans is most
likely about 25 ky, and certainly within the range of
10–60 ky (Peucker-Ehrenbrink and Ravizza 2000; Paquay
et al. 2008). Given its short residence time, Os can record
short-term variations in seawater Os isotopic composition
caused by, for example, massive eruption of mantle-
derived magmas in large igneous provinces or a large
bolide impact. Os is a largely conservative element, and
the residence time exceeds the time required for thorough
mixing among major ocean basins. Nevertheless, rapid
draw-down of redox-sensitive trace metals, including Os,
in restricted basins or anoxic zones on shelf margins (e.g.
Woodhouse et al. 1999; Poirier 2006; Xu et al. 2009), or
non-conservative mixing across salinity gradients (e.g.
Martin et al. 2001), may result in basin-scale spatial
variations in 187Os/188Os of marine waters (e.g. Poirier
2006; Xu et al. 2009). Importantly, the residence time
may have been significantly shorter under anoxic
conditions in the Archaean, or with stratified oceans in
the Palaeoproterozoic, resulting in regional variations in
seawater 187Os/188Os (Kendall et al. 2009b).
Developments in Re-Os Geochronology:Applications to Organic-Rich Shales
Re-Os geochronology has its historic roots in the recognition
of high Re with very low initial Os concentrations in
molybdenite, and hence, the mineral’s potential as
a geochronometer (Herr and Merz 1955; Hirt et al. 1963;
Herr et al. 1967; McCandless and Ruiz 1993). Early low-
precision measurements demonstrated proof-of-concept, but
meaningful application awaited analytical breakthroughs
(e.g. Du et al. 1993). More refined work began in the 1980s
with a better constrained decay constant (Lindner et al. 1986,
1989), improved Re and Os separation methods (Walker
1988; Morgan and Walker 1989), and studies of relatively
Os-rich materials (e.g. osmiridium alloys, meteorites, terres-
trial ultramafic rocks, marine Mn nodules; Allegre and Luck
1980; Turekian 1982; Luck and Allegre 1983; Morgan 1985;
Palmer and Turekian 1986; Hart and Kinloch 1989; Lambert
et al. 1989;Martin 1989;Walker andMorgan 1989). Themost
significant analytical advances took place after 1991 with the
demonstration of high ion yields using negative thermal
ionisation mass spectrometry (N-TIMS; Creaser et al. 1991;
V€olkening et al. 1991), and a dramatic improvement in preci-
sion for the 187Re decay constant by tying it to the well-
determine 235U and 238U decay constants (Smoliar et al.
1996). The way was now opened for meaningful chronology
using the Re-Os method.
Applications of Re-Os geochronology in more felsic
crustal systems soon followed, notably with the develop-
ment of the molybdenite geochronometer (Stein et al.
1997, 1998, 2001; Chesley and Ruiz 1999; Selby and
Creaser 2001). This single-mineral chronometer typically
assumes negligible “common”, or initial Os so that model
10 7.10 Chemical Characteristics of Sediments and Seawater 1507
ages can be calculated directly from concentrations of 187Re
and 187Os. By using a mixed double spike
(185Re-188Os-190Os), both instrumental mass fractionation
of Os and abundance of common Os can be measured,
increasing the precision of molybdenite ages, especially for
very young and/or low Re samples (Markey et al. 2003).
Other sulphides, primarily arsenopyrite or pyrite, may con-
tain sufficiently high Re/Os ratios to yield single-mineral
ages, despite low concentrations (Stein et al. 2000; Arne
et al. 2001; Morelli et al. 2005, 2010). These “low-level
highly radiogenic” (LLHR; Stein et al. 2000) minerals are
valuable for their age information, but, like molybdenite, do
not constrain the initial 187Os/188Os (Osi ) that might other-
wise help trace the source of the metals. Still other sulphides
yield both age and Osi by regression of data from multiple
cogenetic samples using the isochron method (Fig. 7.178,
detailed below). Osi determined for a sulphide ore system,
for example, records the time-integrated radiogenic Os
accumulated in source rocks – in some cases with surprising
results, such as apparent crustal source for magmatic
sulphides in anorthositic or ultramafic intrusions (Morgan
et al. 2000; Hannah and Stein 2002), or crustal sedimentary
rocks for the source of gold at Bendigo in the Lachlan fold
belt (Arne et al. 2001).
Given the success of sulphide isochrons for magmatic-
hydrothermal systems, it was a short leap to syndepositional
or diagenetic sulphides (Horan et al. 1994; Mao et al. 2002).
For example, syndepositional sulphide framboids and
concretions from the Pretoria Group yield a remarkably
precise isochron age of 2316 � 4 Ma (Fig. 7.178; Hannah
et al. 2004). In fact, the analytical uncertainty is less than the
uncertainty for the decay constant; propagating the decay
constant uncertainty generates a statistical uncertainty of
� 7 m.y. The Osi is tightly constrained to 0.112 � 0.001,
precisely the model ratio for a chondritic mantle at 2.32 Ga.
Two major conclusions derive from this result: (1) given the
lack of a mass-independent fractionation signal in the
sulphides, oxygen must have begun accumulating in Earth’s
atmosphere by 2.32 Ga (Bekker et al. 2004), and (2) oxida-
tive weathering was still insufficient at this time to deliver
significant radiogenic 187Os from exposed continental crust
to the oceans (Hannah et al. 2004).
Given their redox properties, Re and Os are predictably
concentrated in organic material as well as in sulphides. This
prediction was first confirmed by measurement of high
concentrations of both Re and Os in sulphide-free humic
particles in sandstones of the Late Jurassic Morrison Forma-
tion in New Mexico (Hannah et al. 2001). The porous
sandstones, however, are prone to transgression by oxidised
groundwaters. Post-depositional adsorption and exchange of
Re and Os by reducing sulphides and organic material in the
sandstone thus created complex isotopic mixtures. Such open
system behaviour, especially prevalent under oxidising
conditions, is an anathema to accurate isotope geochronology.
ORS afford a promising alternative to organic material or
diagenetic sulphides in sandstones. The clay-rich mineral-
ogy typically produces an aquiclude; that is, once dewatered,
there is limited opportunity for the ORS to exchange fluids
with surrounding units. Moreover, the strongly reducing
micro-environments within ORS units reduce the solubility,
and thus the mobility, of both Re and Os.
Three key papers demonstrated the potential utility of
whole-rock Re-Os dating of ORS (Ravizza and Turekian
1989; Cohen et al. 1999; Creaser et al. 2002). Since then,
two fundamental concepts have improved precision. First,
selective dissolution of organics and sulphides, but not
silicates or oxides, limits contributions of Re and Os from
detrital components (Selby and Creaser 2003; Kendall et al.
2004). Second, carefully designed sampling strategies, based
on local geologic conditions, minimise variability in either
depositional age or Osi while maximising the range of187Re/188Os (Hannah and Schersten 2001; Xu et al. 2009;
Yang et al. 2010). Although precision may vary because of
sedimentologic and diagenetic processes that are difficult to
predict in advance, basic procedures for dating ORS are now
well established. For older materials, the decay constant
uncertainty (0.3 %) is a relatively large contribution to the
total age uncertainty. Nevertheless, Re-Os is an important
new method for establishing ages in Precambrian sedimen-
tary sections, particularly since biostratigraphic control is
limited (e.g. Hannah et al. 2006; Kendall et al. 2006,
2009a, b; Yang et al. 2009).
Analytical Methods in Current Use
A radiometric age is determined from the standard age
equation, an expression of the change in isotopic composi-
tion of the daughter isotope (187Os) resulting from decay of
the parent isotope (187Re):
187Os188Os
�
measured
¼187Os188Os
�
0
þ187Re188Os
�
measured
ðelt � 1Þ
where l is the decay constant (1.666 � 10�11a�1; Smoliar
et al. 1996), t is the age in years, and the subscript 0 refers to
the initial ratio, or 187Os/188Os at the time the system was
isotopically closed. This is the equation for a straight line
(y ¼ mx + b) in a plot of 187Re/188Os vs. 187Os/188Os – an
isochron diagram – in which the slope, m, is (elt – 1) and the
y-intercept, b, is (187Os/188Os )0. The slope and intercept aredetermined by linear regression of measured 187Re/188Os
and 187Os/188Os ratios for multiple related samples, known
to have equilibrated isotopically at the same time and with
the same fluid. The program Isoplot (Ludwig 2003) is widely
used for regressing isotopic data.
Figure 7.178 illustrates the salient features of an isochron
diagram. To yield geologically valid results, all samples
plotted must have (1) formed at the same time, (2)
1508 J.L. Hannah and H.J Stein
assimilated Os with the same 187Os/188Os at the time of
formation (i.e. the same Osi), and (3) experienced no
subsequent gain, loss, or isotopic exchange of Re or Os.
Uncertainties in the isotopic ratios for each data point are
propagated from uncertainties at each stage of measurement,
including sample weights, spike calibration, mass ratio
measurements, and blank corrections. Uncertainties on the
age include the uncertainty for the 187Re decay constant. The
mean square weighted deviates (MSWD) are determined by
the offset of each data point from the regression line –
a measure of the goodness of fit – and should have a value
near 1 in the absence of geologic scatter. The MSWDmay be
artificially low, however, if uncertainties on the individual
data points are high. Most importantly, the MSWD will be
altered significantly if analytical uncertainties are over- or
under-estimated; that is, correct determination and propaga-
tion of analytical uncertainties is essential for correct inter-
pretation of isochron statistics. The uncertainty on the slope,
and hence the precision of the age, is strongly affected by the
spread of data points along the 187Re/188Os axis; a larger
spread generally results in a higher precision. Similarly, the
precision on the y-intercept, the Osi, may be poor for
samples that are very old or that have very high187Re/188Os ratios. This is readily visualised from the geom-
etry of the isochron diagram. A precise determination of the
intercept requires some data points close to that intercept. If
all 187Re/188Os are high, then measured 187Os/188Os will be
near the Osi only if the sample is very young; that is, the
slope is low, and the regression line points directly back to
the y-intercept on an expanded 187Os/188Os scale. If the
sample is old, then in-growth of 187Os produces high
measured 187Os/188Os, far removed from the Osi; that is,
the slope is high and must project at an acute angle toward
a greatly compressed 187Os/188Os scale.
Re-Os isotopic analyses are done in four fundamental
steps: (1) selection and separation of specific material to be
analysed in a well-constrained geologic context, (2) com-
plete dissolution of sample and full equilibration with isoto-
pic spikes, (3) chemical separation of Re and Os from other
elements and from each other, and (4) high precision mea-
surement of isotopic ratios by N-TIMS.
Preparation of ORS samples for dissolution and analyses
begins with the outcrop or drill core. Multiple samples are
needed to define an isochron. The stratigraphic interval sam-
pled must be limited to avoid mixing strata of differing age
and/or initial 187Os/188Os, and to avoid any depositional
hiatuses. Yet the samples must have sufficient spread in
Re/Os ratios to formulate a precise slope by linear regression
of 187Re/188Os vs. 187Os/188Os. Some labs recommend
pulverising and thoroughly mixing 20–100 mg of ORS for
each analysis in order to minimise variation in initial187Os/188Os (e.g. Kendall et al. 2009a; Selby et al. 2009).
We have found that small samples (at most a few mg)
extracted from carefully selected spots using a diamond-
tipped drill commonly yield good results (Yang et al.
2010). Laboratory analysis reveals negligible contamination
from the drill bits. Most importantly, there is no magic
number for the aliquant size that will yield the best data.
Redistribution of Re and Os among organic and sulphide
phases during diagenetic, metamorphic, or other metaso-
matic processes may require larger samples to capture any
potential decoupling of parent and daughter isotopes, but this
must be determined on a case-by-case basis. The ‘recipe’
varies with geologic conditions; sample size tests may be
required to determine the best approach for a given setting.
Most labs have now adopted Carius tube dissolution, in
which sample, spike, and reagents are sealed in a thick-
walled glass vessel, enclosed in a steel jacket, and digested
at ~240 �C for ~48 h (Shirey and Walker 1995). For most
applications, inverse aqua regia (2:1 ratio 16N HNO3 and
12N HCl) is the reagent of choice. For analysis of ORS,
however, selective dissolution of hydrogenous components
using CrO3-H2SO4 commonly yields more consistent
results, as detrital Re and Os contributions are not attacked
(Selby and Creaser 2003; Kendall et al. 2004; Xu et al.
submitted). Os may be extracted from the solution either
by distillation directly from the Carius tube (Markey et al.
2007) or by solvent extraction (Cohen and Waters 1996) and
further purified by microdistillation (Birck et al. 1997). Re is
then separated from the remaining solution by anion
exchange (Morgan et al. 1991; Markey et al. 1998). The
most precise isotope ratio measurements are achieved by
N-TIMS, but robust results may also be acquired for Os-
rich samples by multi-collector inductively coupled plasma
mass spectrometry (MC-ICP-MS; e.g. Nowell et al. 2007).
Achievements
To date, there are relatively few Re-Os geochronology stud-
ies of sedimentary rocks >2.0 Ga. The example of 2.3 Ga
ORS from the Transvaal Supergroup (Hannah et al. 2004,
described above) was succeeded by a review of Re/Os con-
centration ratios in Archaean and younger ORS by Siebert
et al. (2005). The latter study demonstrates increasing mobil-
ity of Re and Os with the development of oxygenated
environments, suggesting complex and episodic growth of
atmospheric oxygen from the Mesoarchaean through the
Palaeoproterozoic. Anbar et al. (2007) documented an
increased flux of redox-sensitive Mo and Re to ORS at
2.5 Ga, confirming the age of the ORS with Re-Os geochro-
nology. Most recently, Yang et al. (2009) used a precise
2.69 Ga age for slates from the Wawa subprovince of the
Superior Province, Minnesota to constrain the timing of
assembly of the southern Superior Province.
Work in progress suggests that Re-Os ages can be deter-
mined for ORS deposited near the close of the Lomagundi
C-isotope excursion (2.22–2.06 Ga; Karhu and Holland
1996) in the Pechenga Greenstone Belt and the Onega
basin (Hannah et al. 2006, 2010). Persistent chondritic initial
10 7.10 Chemical Characteristics of Sediments and Seawater 1509
187Os/188Os ratios for these units do not reflect the expected
increase in oxidative weathering of high 187Re/188Os (and
hence, high 187Os/188Os) continental rocks. Rather, the187Os/188Os ratios in these units may reflect restricted basins
dominated by hydrothermal Os input.
Unsolved Problems in Palaeoproterozoic Re-OsGeochemistry
Re-Os geochronology has the potential to contribute enor-
mously to reconstruction and interpretation of Palaeopro-
terozoic environmental changes. In order to fulfill that
potential, however, advances in our knowledge and under-
standing of the behaviour of Re and Os in sedimentary
systems are required.
How are the onset and acceleration of oxidative
weathering reflected by changes in Os cycling through the
Archaean/Palaeoproterozoic transition? To answer this
question, many more Re-Os data sets from Late Archaean
and Palaeoproterozoic sedimentary units are needed to chart
changes in Re-Os behaviour through time. Initial187Os/188Os ratios are commonly less precise in older rocks
because of the steep slope of the regression line on the
isochron diagram (Fig. 7.178). Therefore, highly precise
measurements on undisturbed materials are essential. Con-
centration data require normalisation in order to separate
variations in global cycling of Re and Os from local
variations in, for example, total organic carbon or sulphide
concentrations. Yet there are very few studies in which Re-
Os measurements are combined with other geochemical
data. Such corroborating data sets are also needed to tease
apart causes of variability in Re-Os systematics (Georgiev
et al. in press). We know that redox-sensitive elements
behave very differently in open marine systems compared
to restricted basins or broad shelf environments, but the
details remain poorly understood.
When and why did the first organic carbon-rich strata
generate hydrocarbons? Much remains to be learned about
the behaviour of Re and Os in hydrocarbon systems.
Improved knowledge of Re/Os fractionation and Os isotopic
exchange during hydrocarbon maturation is needed to link
migrated hydrocarbons to source rocks and decipher the
timing of source rock deposition and hydrocarbon expulsion.
Simultaneous study of well-constrained Phanerozoic hydro-
carbon systems and Earth’s earliest concentrations of organic
material is crucial, but details of distribution coefficients and
exchange rates will come only from younger systems. Above
all, careful sampling strategies are required to distinguish the
Os isotopic ratio of the source rock from isotopic changes or
overprints introduced by hydrocarbon migration.
7.13.5.7 Implications for Results from theFAR-DEEP Core
The FAR-DEEP core holds two key positive attributes for
maximising what we learn from Re-Os isotopic studies:
Fig. 7.177 Re and Os reservoirs with sources and sinks for seawater.
Riverine concentrations of Re and Os depend strongly on bedrock in
the drainage; flux to the oceans is a mixture of dissolved and suspended
load. Therefore, global averages are not available (Data are
summarised from a variety of sources, all cited in the main text.
Graphic by Lisa Løseth, Geological Survey of Norway)
1510 7.10 Chemical Characteristics of Sediments and Seawater
(1) samples are tightly controlled for stratigraphic position
and sedimentary environments, and (2) interdisciplinary
collaboration assures that Re-Os data can be correlated
with other chronometers and redox indicators on the
same (or adjacent) samples. In turn, Re-Os geochronology
adds time constraints to the core, and thus guides
interpretations of rates of change. And, the history of
changes in Re and Os cycling through Earth time
corroborates other redox indicators of changing surface
conditions in the deep past.
The FAR-DEEP materials also present challenges. ORS
horizons were intersected in only a few of the drill holes.
Geologically accurate interpretation of Re-Os data requires
persistent local anoxia during deposition and minimal
subsequent oxidation. Most of the FAR-DEEP drill core,
however, penetrated rocks originally deposited under oxic
conditions. In contrast, Holes 12A, 12B, and 13A intersected
many metres of carbon-rich rocks. These units present their
own challenges, however. The sulphide record alone in these
cores shows early laminated, nodular sulphides with
multiple overgrowths, and vein sulphides (pyrite and pyr-
rhotite). The organic material occurs as residual kerogen,
mobilised organosiliceous rocks, pyrobitumen-cemented
breccias and complex pyrobitumen veining. Multiple events
– early maturation and migration of hydrocarbons, thermal
metamorphism associated with magmatism, and subsequent
dynamic metamorphism – provided multiple opportunities
for chemical exchange among organic and sulphide phases.
The FAR-DEEP record is fraught with closely-spaced
Palaeoproterozoic events that redistributed both organic
material and sulphides. Thoughtful sampling, precise
analyses, and informed interpretation will be required to
tease apart the influences of each event.
References
Allegre CJ, Luck JM (1980) Osmium isotopes as petro-
genetic and geological tracers. Earth Planet Sci Lett
48:148–154
Fig. 7.178 Example of an isochron diagram (from Hannah et al.
2004). Isotope ratios for seven discrete samples are regressed using a
York fit (Ludwig 2003) to yield a straight line with slope (elt – 1). The
age, t, is then calculated from the slope (see text for details). The initial187Os/188Os, the ratio at the time the pyrites crystallised, is the y-
intercept. That is, if 187Re/188Os is zero, there will be no in-growth of187Os, and the measured 187Os/188Os will equal the initial ratio. All
analytical uncertainties for each ratio measurement are propagated
through the statistical analysis. “Model 1” assumes no geologic varia-
tion in the initial 187Os/188Os of the samples, and is an indicator of a
good statistical fit. “Model 3” is used when scatter of data points
suggests real variation in the initial 187Os/188Os, and generally results
in a less satisfactory fit of line to points. Assumptions of large
uncertainties for individual data points, however, may inappropriately
result in a Model 1 fit. MSWD is the mean square weighted deviates,
and is a measure of the deviation of the data points from the regression
line. The MSWD may be made to look artificially low if large
uncertainties are assigned to data points. The precision on the age
may be artificially high if the data cluster into two disparate regions,
as is the case here. Separate regression of the four data points near the
origin, however, shows that the statistics are not being enhanced by two
data clusters in this case. The uncertainty on the 187Re decay constant is
0.31 % when, as in this case, the Os spike is tied to the Os spike solution
used at the University of Maryland in 1996 (Smoliar et al. 1996, note
16); hence the true uncertainty on the age cannot be less than 7 Ma.
That is, the statistical precision of the data in this example is better than
the minimum true precision including the decay constant uncertainty.
The larger uncertainty of 7 Ma is applied for comparison with other
chronometers
10 7.10 Chemical Characteristics of Sediments and Seawater 1511
Anbar AD, Duan Y, Lyons TW, Arnold GL, Kendall B,
Creaser RA, Kaufman AJ, Gordon GW, Scott C, Garvin J,
Buick R (2007) A whiff of oxygen before the great oxidation
event? Science 317:1903–1906
Arne D, Bierlein F, Morgan JW, Stein HJ (2001) Re-Os
dating of sulfides associated with gold mineralization in
central Victoria, Australia. Econ Geol 96:1455–1459
Bekker A, Holland HD, Wang P-L, Rumble D III, Stein
HJ, Hannah JL, Coetzee LL, Beukes NJ (2004) Dating the
rise of atmospheric oxygen. Nature 427:117–120
Birck JL, Barman MR, Capmas F (1997) Re–Os isotopic
measurements at the femtomole level in natural samples.
Geostandard Newslett 20:19–27
Cave RR, Ravizza GE, German CR, Thomson J, Nesbitt
RW (2003) Deposition of osmium and other platinum-group
elements beneath the ultramafic-hosted Rainbow hydrother-
mal plume. Earth Planet Sci Lett 210:65–79
Chen C, Sharma M, Bostick BC (2006) Lithologic
controls on osmium isotopes in the Rio Orinoco. Earth
Planet Sci Lett 252:138–151
Chesley JT, Ruiz J (1999) Preliminary Re-Os dating
on molybdenite mineralization from the Bingham Canyon
porphyry copper deposit, Utah. In: John DA, Ballantyne H
(eds) Geology and ore deposits of the Oquirrh and
Wasatch Mountains, Utah. Society of Economic Geologists
Guidebook Series. 29:165–169
Cohen AS, Waters FG (1996) Separation of osmium from
geological materials by solvent extraction for analysis by
thermal ionization mass spectrometry. Anal Chim Acta
332:269–275
Cohen AS, Coe AL, Bartlett JM, Hawkesworth CJ (1999)
Precise Re–Os ages of organic-rich mudrocks and the Os
isotope composition of Jurassic seawater. Earth Planet Sci
Lett 167:159–173
Creaser RA, Papanastassiou DA, Wasserburg GJ
(1991) Negative thermal ion mass spectrometry of osmium,
rhenium, and iridium. Geochim Cosmochim Acta
55:397–401
Creaser RA, Sannigrahi P, Chacko T, Selby D (2002)
Further evaluation of the Re–Os geochronometer in
organic-rich sedimentary rocks: a test of hydrocarbon
maturation effects in the Exshaw Formation, Western
Canada Sedimentary Basin. Geochim Cosmochim Acta
66:3441–3452
Du AD, He HL, Yin NW, Zou XQ, Sun YL, Sun DZ,
Chen SZ, Qu WJ (1993) Direct dating of molybdenites
using the Re-Os geochronometer. Chinese Sci Bull 38:
1319–1320
Esser BK, Turekian KK (1993) The osmium isotopic
composition of the continental crust. Geochim Cosmochim
Acta 57:3093–3104
Georgiev S, Stein HJ, Hannah JL, Bingen B, Weiss HM,
Piasecki S (2011) Hot acidic Late Permian seas stifle life in
record time. Earth Planet Sci Lett 310:389–400
Georgiev S, Stein HJ, Hannah JL, Weiss H, Bingen B,
Xu G, Rein E, Hatlø V, Løseth H, Nali M, Piasecki H (2012)
Trace element geochemistry of black shales in the Arctic
region: effects of weathering on redox-sensitive elements
and Re-Os isotope systematics. Chem Geol (in press)
Gramlich JW, Murphy TJ, Garner EL, Shields WR (1973)
The atomic weight of rhenium. J Res Natl Bur Stand A
77:691–98
Hannah JL, Schersten A (2001) Open-system behavior of
Re and Os in uraninite, pyrite, and organic material in
continental sedimentary rocks, Grants uranium region,
New Mexico. Geological Society of America Abstracts
with Programs 33(5):A-57
Hannah JL, Stein HJ (2002) Re-Os model for the origin of
sulfide deposits in anorthosite-associated intrusive
complexes. Econ Geol 97:371–383
Hannah JL, Schersten A, Stein HJ (2001) Re-Os behavior
in mafic magmas and terrestrial diagenetic materials:
extracting information from open systems. Geological Soci-
ety of America Abstracts with Programs 33(6):A-238
Hannah JL, Bekker A, Stein HJ, Markey RJ,
Holland HD (2004) Primitive Os and 2316Ma age for marine
shale: implications for Paleoproterozoic glacial events
and the rise of atmospheric oxygen. Earth Planet Sci Lett
225:43–52
Hannah JL, Stein HJ Zimmerman A, Yang G, Markey RJ,
Melezhik VA (2006) Precise 2004 � 9 Ma Re-Os age for
Pechenga black shale: comparison of sulfides and organic
material. Geochim Cosmochim Acta 70(Supplement):A-228
Hannah J, Yang G, Stein H, Zimmerman A (2010) Birth
of a giant Paleoproterozoic oil field: Re-Os ages for source
rocks and maturation. Geochim Cosmochim Acta 74(Sup-
plement):A-379
Hart SR, Kinloch ED (1989) Osmium isotope systematics
in Witwatersrand and Bushveld ore deposits. Econ Geol
84:1651–1655
Herr W, Merz E (1955) Eine neue Methode zur
Alterbestimmung von Rhenium-Haltigen Mineralien Mittels
Neutronaktivierung. Z Naturforsch 10a:613–615
Herr W W€olfle EP, Kopp E (1967) Development and
recent applications of the Re/Os dating method. In: Radio-
active dating and methods of low-level Counting. Interna-
tional Atomic Energy Agency, Vienna, p 499–508
Hirt B, Herr W, Hofmeister W (1963) Age determinations
by the rhenium-osmium method. In: Radioactive dating.
International Atomic Energy Agency, Vienna, p 35–43
Horan MF, Morgan JW, Grauch RI, Coveney RM Jr,
Murowchick JB, Hulbert LJ (1994) Rhenium and
osmium isotopes in black shales and Ni–Mo–PGE-rich
sulfide layers, Yukon territory, Canada, and Hunan and
Guizhou provinces, China. Geochim Cosmochim Acta
58:257–265
Karhu JA, Holland HD (1996) Carbon isotopes and the
rise of atmospheric oxygen. Geology 24:867–870
1512 J.L. Hannah and H.J Stein
Kendall BS, Creaser RA, Ross GM, Selby D (2004)
Constraints on the timing of Marinoan Snowball Earth gla-
ciation by 187Re–187Os dating of a Neoproterozoic, post-
glacial black shale in Western Canada. Earth Planet Sci Lett
222:729–740
Kendall BS, Creaser RA, Selby D (2006) Re–Os geochro-
nology of postglacial black shales in Australia: constraints
on the timing of “Sturtian” glaciation. Geology 34:729–732
Kendall B, Creaser RA, Selby D (2009a) 187Re-187Os
geochronology of Precambrian organic-rich sedimentary
rocks. Geological Society of London Special Publication
326:85–107
Kendall B, Creaser RA, Gordon GW, Anbar AD (2009b)
Re–Os and Mo isotope systematics of black shales from the
Middle Proterozoic Velkerri and Wollogorang Formations,
McArthur Basin, northern Australia. Geochim Cosmochim
Acta 73:2534–2558
Lambert DD, Morgan JW,Walker RJ, Shirey SB, Carlson
RW, Zientek MI, Koski MS (1989) Rhenium-osmium and
samarium-neodymium isotopic systematics of the Stillwater
complex. Science 244:1169–1174
Levasseur S, Birk J-L, Allegre CJ (1998) Direct measure-
ment of femtomoles of osmium and the 187Os/186Os ratio in
seawater. Science 282:272–274
Lindner M, Leich DA, Borg RJ, Russ GP, Bazan JM,
Simons DS, Date AR (1986) Direct laboratory determination
of the 187Re half-life. Nature 320:246–248
Lindner M, Leich DA, Price R, Bazan JM, Borg RJ
(1989) Direct determination of the half-life of 187Re.
Geochim Cosmochim Acta 53:1597–1606
Ludwig KR (2003) User’s manual for Isoplot 3.0.
A Geochronological toolkit for Microsoft Excel, Berkeley
Geochronology Center Special Publication 4:70
Luck JM, Allegre CJ (1983) 187Re-187Os systematics in
meteorites and cosmochemical consequences. Nature
302:130–132
Mao J, Lehmann B, Du A, Zhang G, Ma D, Wang Y,
Zeng M, Kerrich R (2002) Re–Os dating of polymetallic Ni-
Mo-PGE-Au mineralization in Lower Cambrian black
shales of South China and its geologic significance. Econ
Geol 97:1051–1061
Markey R, Stein H, Morgan J (1998) Highly precise
Re–Os dating for molybdenite using alkaline fusion and
NTIMS. Talanta 45:935–946
Markey RJ, Hannah JL, Morgan JW, Stein HJ (2003)
A double spike for osmium analysis of highly radiogenic
samples. Chem Geol 200:395–406
Markey R, Stein HJ, Hannah JL, Zimmerman A, Selby D,
Creaser RA (2007) Standardizing Re-Os geochronology:
a new molybdenite Reference Material (Henderson, USA)
and the stoichiometry of Os salts. Chem Geol 244:74–87
Martin CE (1989) Re-Os isotopic investigation of the
Stillwater complex, Montana. Earth Planet Sci Lett
93:336–344
Martin CE, Peucker-Ehrenbrink B, Brunskill G,
Szymczak R (2001) Osmium isotope geochemistry of a
tropical estuary. Geochim Cosmochim Acta 65:3193–3200
McCandless TE, Ruiz J (1993) Rhenium-osmium evi-
dence for regional mineralization in southwestern North
America. Science 261:1262–1266
Meisel T, Walker RJ, Morgan JW (1996) The osmium
isotopic composition of the Earth’s primitive upper mantle.
Nature 383:517–520
Miller CA, Peucker-Ehrenbrink B, Ball L (2009) Precise
determination of rhenium isotope composition by multi-
collector inductively-coupled plasma mass spectrometry.
J Anal Atom Spectrom 24:1069–1078
Morelli RM, Creaser RA, Selby D, Kontak DJ, Horne RJ
(2005) Rhenium-osmium geochronology of arsenopyrite in
Meguma Group gold deposits, Meguma Terrane, Nova Sco-
tia, Canada: evidence for multiple gold-mineralizing events.
Econ Geol 100:1229–1242
Morelli RM, Bell CC, Creaser RA, Simonetti A (2010)
Constraints on the genesis of gold mineralization at the
Homestake gold deposit, Black Hills, South Dakota from
rhenium-osmium sulfide geochronology. Miner Deposita
45:461–480
Morgan JW (1985) Osmium isotope constraints on
Earth’s late accretionary history. Nature 317:703–705
Morgan JW, Walker RJ (1989) Isotopic determinations of
rhenium and osmium in meteorites by using fusion, distilla-
tion and ion-exchange separations. Anal Chim Acta
222:291–300
Morgan JW, Golightly DW, Dorzapf AF Jr (1991)
Methods for the separation of rhenium, osmium, and molyb-
denum applicable to isotope geochemistry. Talanta
38:259–265
Morgan JW, Stein HJ, Hannah JL, Markey RJ,
Wiszniewska J (2000) Re-Os study of ores from the Suwalki
anorthosite massif, northeast Poland. Miner Deposita
35:391–401
Nowell GM, Luguet A, Pearson DG, Horstwood MSA
(2007) Precise and accurate 186Os/188Os and 187Os/188Os
measurements by multi-collector plasma ionisation mass
spectrometry (MC-ICP-MS) part I: solution analyses.
Chem Geol 248:363–393
Palmer MR, Turekian KK (1986) 187Os/186Os in marine
manganese nodules and the constraints on the crustal
geochemistries of rhenium and osmium. Nature
319:216–220
Paquay FS, Ravizza GE, Dalai TK, Peucker-Ehrenbrink B
(2008) Determining chondritic impactor size from the marine
osmium isotope record. Science 320:214–218
Peucker-Ehrenbrink B, Ravizza G (2000) The marine
osmium isotope record. Terra Nova 12:205–219
Poirier A (2006) Re–Os and Pb isotope systematics in
reduced fjord sediments from Saanich Inlet (Western
Canada). Earth Planet Sci Lett 249:119–131
10 7.10 Chemical Characteristics of Sediments and Seawater 1513
Ravizza G, Turekian KK (1989) Application of the187Re-187Os system to black shale geochronometry.
Geochim Cosmochim Acta 53:3257–3262
Selby D, Creaser RA (2001) Re-Os geochronology and
systematics in molybdenite from the Endako porphyry
molybdenum deposit, British Columbia, Canada. Econ
Geol 96:197–204
Selby D, Creaser RA (2003) Re–Os geochronology of
organic rich sediments: an evaluation of organic matter
analysis methods. Chem Geol 200:225–240
Selby D, Mutterlose J, Condon DJ (2009) U-Pb and
Re-Os geochronology of the Aptian/Albian and
Cenomanian/Turonian stage boundaries: implications for
timescale calibration, osmium isotope seawater composition
and Re-Os systematics in organic-rich sediments. Chem
Geol 265:394–409
Sharma M, Papanastassiou DA, Wasserburg GJ (1997)
The concentration and isotopic composition of osmium in
the oceans. Geochim Cosmochim Acta 61:3287–3299
Sharma M, Rosenberg EJ, Butterfield DA (2007) Search
for the proverbial mantle osmium sources to the oceans:
hydrothermal alteration of mid-ocean ridge basalt. Geochim
Cosmochim Acta 71:4655–4667
Shirey SB, Walker RJ (1995) Carius tube digestion
for low-blank rhenium-osmium analysis. Anal Chem
67:2136–2141
Shirey SB, Walker RJ (1998) The Re–Os isotope system
in cosmochemistry and high-temperature geochemistry.
Annu Rev Earth Planet Sci 26:423–500
Siebert C, Kramers JD, Meisel T, Morel P, N€agler TF(2005) PGE, Re-Os, and Mo isotope systematics in Archean
and early Proterozoic sedimentary systems as proxies for
redox conditions of the early Earth. Geochim Cosmochim
Acta 69:1787–1801
Smoliar MI, Walker RJ, Morgan JW (1996) Re-Os ages
Pt-Os isotopes of group IIA, IIIA, IVA, and IVB iron
meteorites. Science 271:1099–1102
Stein HJ, Markey RJ, Morgan JW, Du A, Sun Y (1997)
Highly precise and accurate Re-Os ages for molybdenite
from the East Qinling molybdenum belt, Shaanxi Province,
China. Econ Geol 92:827–835
Stein HJ, Sundblad K, Markey RJ, Morgan JW, Motuza
G (1998) Re-Os ages for Archaean molybdenite and pyrite,
Kuittila-Kivisuo, Finland and Proterozoic molybdenite,
Kabeliai, Lithuania: testing the chronometer in a meta-
morphic and metasomatic setting. Miner Deposita 33:
329–345
Stein HJ, Morgan JW, Schersten A (2000) Re-Os dating
of low-level highly-radiogenic (LLHR) sulfides: the Harn€as
gold deposit, southwest Sweden records continental scale
tectonic events. Econ Geol 95:1657–1671
Stein HJ, Markey RJ, Morgan JW, Hannah JL, Schersten
A (2001) The remarkable Re-Os chronometer in molybde-
nite: how and why it works. Terra Nova 13:479–486
Stein H, Schersten A, Hannah J, Markey R (2003) Sub-
grain scale decoupling of Re and 187Os and assessment of
laser ablation ICP-MS spot dating in molybdenite. Geochim
Cosmochim Acta 67:3673–3686
Suzuki K, Miyata Y, Kanazawa N (2004) Precise Re
isotope ratio measurements by negative thermal ionization
mass spectrometry (NTI-MS) using total evaporation tech-
nique. Int J Mass Spectrom 235:97–101
Turekian KK (1982) Potential of 187Os/186Os as a cosmic
versus terrestrial indicator in high iridium layers of sedimen-
tary strata. In: Silver LT, Schultz PH (eds) Geological
implications of impacts of large asteroids and comets on
the Earth. Geol Soc Am Spec Pap 190:243–249
V€olkening J, Walczyk T, Heumann KG (1991) Osmium
isotope ratio determinations by negative thermal ionization
mass spectrometry. Int J Mass Spectrom Ion Process
105:147–159
Walker RJ (1988) Low-blank chemical separation of
rhenium and osmium from gram quantities of silicate rock
for measurement by resonance ionization mass spectrome-
try. Anal Chem 60:1231–1234
Walker RJ, Morgan JW (1989) Rhenium-osmium isotope
systematics of carbonaceous chondrites. Science
243:519–522
Walker RJ, Morgan JW, Beary ES, Smoliar MI,
Czamanske GK, Horan MF (1997) Applications of the190Pt-186Os isotope system to geochemistry and cosmo-
chemistry. Geochim Cosmochim Acta 61:4799–4807
Woodhouse OB, Ravizza G, Kenison Falkner K, Statham
PJ, Peucker-Ehrenbrink B (1999) Osmium in seawater: ver-
tical profiles of concentration and isotopic composition in
the eastern Pacific Ocean. Earth Planet Sci Lett 173:223–233
Xu G, Hannah JL, Stein HJ, Bingen B, Yang G,
Zimmerman A, Weitschat W, Mørk A, Weiss HM (2009)
Re–Os geochronology of Arctic black shales to evaluate the
Anisian–Ladinian boundary and global faunal correlations.
Earth Planet Sci Lett 288:581–587
Xu G, Hannah JL, Bingen B, Georgiev S, Stein HJ,
(2012) Digestion methods for trace element measurements
in shales: paleoredox proxies examined. Chem Geol, in
press
Yang G, Hannah JL, Zimmerman A, Stein HJ, Bekker A
(2009) Re-Os depositional age for Archean carbonaceous
slates from the southwestern Superior Province: challenges
and insights. Earth Planet Sci Lett 280:83–92
Yang G, Stein H, Hannah J, Zimmerman A (2010)
Targeted drilling for Re-Os geochronology to decipher com-
plex history of overmature source rocks and migrated
hydrocarbons. Geochim Cosmochim Acta 74(12), Supple-
ment 1:A1172
Zimmerman A, Stein HJ, Hannah JL, Tuttas B, Yang G,
Beitscher B (2007) Instrumental mass fractionation over-
come by total evaporation. Geochim Cosmochim Acta 71
(Supplement):A1176
1514 J.L. Hannah and H.J Stein
Part VIII
The Great Oxidation Event: State of the Art andMajor Unresolved Problems
8.1 The Great Oxidation Event
Lee R. Kump, Anthony E. Fallick, Victor A. Melezhik, Harald Strauss,and Aivo Lepland
The Fennoscandian Arctic Russia-Drilling Early Earth Proj-
ect (FAR-DEEP) was designed to capture the sequence of
environmental upheavals associated with the establishment
of an aerobic biosphere in the Palaeoproterozoic as
represented in several palaeobasins and greenstone belts in
the Fennoscandian Shield; the Pechenga Greenstone Belt is
one of them (Fig. 8.1). From weathering and deposition on
Archaean basement overlain by evidence for “Huronian
Glaciation,” through early evidence for atmospheric oxygen
in the form of redbeds and highly oxidised lavas, which
themselves are coincident with evidence for massive
perturbations of the global carbon cycle expressed through
substantial shifts in d13C, and ending with the deposition of
high-organic-C rocks and strong evidence for modern-style
early diagenetic environments with abundant sulphate
reduction and diagenetic concretions, the FAR-DEEP cores
preserve an invaluable archive of the “Great Oxidation
Event” or GOE (Holland 2002, 2006) and related environ-
mental consequences. This transition, known for decades to
roughly coincide with the Archaean – Proterozoic boundary,
is marked not only by the appearance of “red beds,” reddish
sedimentary rocks, typically sand and silt particles coated in
ferric oxides that were deposited in terrestrial environments,
but also with the retention of Fe in ancient soil profiles or
palaeosols and the end of uranium ore accumulation in
detrital rocks as uranium-bearing conglomerates (reviews
by Knoll and Holland 1995; Canfield 2005; Holland 2006).
Other changes, especially the abrupt cessation (but episodic
recurrence) of banded iron formation and an increase in the
Fe(III)/Fe(II) ratio in shales (Bekker et al. 2003), reflect on
the oxidation state of the oceans and diagenetic
environments, not the atmosphere per se, and thus require
additional consideration before they can be used as a proxy
for atmospheric oxygen. As discussed below, the discovery
of mass-independent fractionation of the sulphur isotopes
exclusively in Archaean sedimentary rocks (Farquhar et al.
2000) provided the direct proxy for atmospheric
oxygenation during the transition from the Archaean to the
Proterozoic.
In the 1990s, it appeared that the GOE coincided with the
large and prolonged “Lomagundi-Jatuli” positive carbon
isotope excursion (LJ-CIE; Baker and Fallick 1989; Karhu
1993; Melezhik and Fallick 1996; Karhu and Holland 1996).
This provided a mechanism for the rise of oxygen: the
implied net generation of considerable quantities of oxygen
through enhanced burial of organic matter produced by
oxygenic photosynthesis, a process that protects the organic
matter from “back reaction” with oxygen and preferentially
sequesters 12C, driving the residual C in the ocean/atmo-
sphere system toward higher values of d13C. However, evenat the time it was recognised that a burst of O2 release was
insufficient cause for the GOE, because the atmosphere has
sustained its oxygenated state ever since. A singular event
could perturb the system, but could not explain a permanent
shift in the long-term, steady-state oxygen content of the
atmosphere.
8.1.2 The Timing of the Great Oxidation Event
Another problem developed with the LJ-CIE explanation for
the rise of atmospheric oxygen. As geochronological age
constraints improved, the isotope excursion came to postdate
the GOE. This became particularly clear with the discovery
that large mass-independent fractionation of sulphur
isotopes (S-MIF) of atmospheric sulphur, preserved in
sulphur-bearing minerals, disappeared before the LJ-CIE
(Farquhar et al. 2000). Theory suggests that S-MIF will
only be formed when the atmosphere is anoxic and enriched
in methane (Pavlov and Kasting 2002; Zahnle et al. 2006).
Detailed work in South Africa on the Duitschland Formation
now shows clearly that the MIF signature disappeared just
L.R. Kump (*)
Department of Geosciences, Pennsylvanian State University,
503 Deike Building, University Park, PA 16870, USA
e-mail: [email protected]
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_11, # Springer-Verlag Berlin Heidelberg 2013
1517
before the onset of the LJ-CIE (see Chaps. 7.1 and 7.5; Guo
et al. 2009). Thus, the LJ-CIE seems to be a consequence or
even perhaps non-related event, rather than a cause, of the
GOE. In detail, the MIF signature vanishes between the first
and second diamictite in South Africa (Guo et al. 2009) and
Canada (Papineau et al. 2007) (Chap. 7.2). The ages of
individual diamictites are not known, but the Duitschland
and Huronian deposits rest on volcanic rocks with U/Pb ages
of 2480 (�6) and 2450 (+25/�10) Ma (respectively) and in
South Africa are overlain by sediments with a Re/Os age of
2317 � 7 Ma (summarised in Chap. 7.2). Thus, the GOE
appears to have commenced somewhere after 2450 Ma, and
atmospheric pO2 exceeded 10�5 of the present atmospheric
level (PAL) (Pavlov and Kasting 2002) and atmospheric
pCH4 fell below ~10� PAL (Zahnle et al. 2006) by
2317 Ma.
Kirshvink and Kopp (2008) argue that the GOE occurred
at the time of the Kalahari manganese deposit in the Hotazel
Formation, Transvaal Supergroup of South Africa; the ear-
lier loss of the MIF signature was a consequence of the
oxidation of reduced sulphur and homogenisation of the
isotopes in an ocean rich in hydrogen peroxide (H2O2)
derived from glacial melting. Whether or not this interpreta-
tion of the timing of the GOE is in conflict with the 2317 Ma
minimum age cited above, depends on whether one accepts
the relatively young whole-rock Pb-Pb age of the underlying
Ongeluk lavas (2220 Ma as a maximum age; Cornell et al.
1996) or the relatively old whole-rock Pb-Pb carbonate age
of the overlying Mooidraai Formation (2390 Ma as a mini-
mum age; Bau et al. 1999). As discussed in Tsikos et al.
(2010), the large uncertainty in the age of the Hotazel For-
mation prevents us from placing it confidently within the
sequence of events associated with the GOE.
8.1.3 The Cause of the Great Oxidation Event
So what did cause the GOE? Two types of viable
explanations have been proffered: either this was when
cyanobacteria invented oxygenic photosynthesis (Fig. 8.2a),
or they did so long before the GOE but the sinks for O2
exceeded the sources until the GOE (Fig. 8.2b, c).
Cyanobacterial Origin for the GOE
Kopp et al. (2005) invoke the origin of cyanobacterial oxy-
genic photosynthesis to explain the Makganyene glacial
deposits and overlying Hotazel Mn deposits. The onset of
oxygen production destroyed a strong methane greenhouse
climate control, driving the world into a “snowball” glacia-
tion. Subsequent oxidation of Mn led to the concentrated
accumulation of Mn ores. Although early analyses put the
origin of cyanobacteria at the GOE (Hedges et al. 2001),
molecular clock determinations now place the origin of
cyanobacteria at around ~2700 Ma (2920–2592 Ma; Hedges
and Kumar 2009), consistent with biomarker evidence for
cyanobacterial hopanes and steranes (Brocks et al. 1999,
2003) at 2700 Ma. The biomarker evidence has been
challenged in terms of its specificity (Kopp et al. 2005;
Rashby et al. 2007; Kirschvink and Kopp 2008; Welander
et al. 2010) and syngenicity (Rasmussen et al. 2008). Other
work, however, supports the original interpretation of the
biomarker evidence, i.e., that hopanes most abundantly pro-
duced by (some) cyanobacteria, and steranes indicative of
the presence of oxygen, were being preserved in sediments
by 2700 Ma (e.g. Waldbauer et al. 2009; Eigenbrode et al.
2008; Dutkiewicz et al. 2006) and in Huronian rocks just
before the GOE (Dutkiewicz et al. 2006).
Reduced Volcanic/Metamorphic Sinks Causedthe GOE
The alternative, then, is that sinks exceeded sources of
oxygen prior to the GOE. In the modern oxygen cycle,
oxidative weathering of fossil organic matter and sedimen-
tary pyrite is the largest sink for oxygen in its exogenic
cycle, although there are smaller sinks associated with oxi-
dation of volcanic fluids (e.g. Holland 1978). Presumably the
reaction with reduced gases (H2, CH4, H2S, CO) would have
readily occurred at much lower oxygen concentrations (e.g.
Claire et al. 2006; Goldblatt et al. 2006), suppressing the
buildup of O2 if the supply of reducing gases was larger in
the Archaean.
An enhanced sink for oxygen in the Archaean could be
the result of (1) more reduced volcanic fluids equilibrated
with a lower fO2 mantle (i.e., mantle redox evolution;
Kasting et al. 1993; Kump et al. 2001; Holland 2002)
(Fig. 8.2b); (2) more reduced metamorphic fluids emanating
from a more reduced crust (Catling et al. 2001; Claire et al.
2006) (Fig. 8.2b); or (3) a lower mean fO2 of volcanic fluids
in the Archaean because of the predominance of submarine
volcanism prior to the Archaean-Proterozoic boundary
(Kump and Barley 2007) (Fig. 8.2c). The presumed higher
rate of volcanism on the early Earth itself would not neces-
sarily provide a net enhanced sink for O2. As Holland (2002)
has shown, volcanic gases equilibrated with the presumed
nominal mean upper mantle mineral assemblage fayalite-
magnetite-quartz (FMQ), when “disproportionated” into
oxidised and reduced components, sets the redox balance
for the planet with the canonical 20 % of the C converting to
organic matter (with attendant oxygen production) and all
the reduced gases (including the sulphur gases) becoming
1518 L.R. Kump et al.
oxidised. Thus, increased volcanism at FMQ increases both
the sink for oxygen associated with the increased flux of
reduced gases but also the source for oxygen associated with
organic matter production and burial. Only if those fluids
were more reduced in the Archaean, would the balance be
shifted toward an anoxic atmosphere.
The mantle redox evolution arguments (Kasting et al.
1993; Kump et al. 2001; Holland 2002) arose out of the
recognition that at some point in Earth history, as the core
was segregating, the mantle likely was buffered at the Fe-FeO
boundary. From that point forward to today it has evolved
toward the FMQ buffer, presumably by hydrogen escape to
space (Kasting et al. 1993). If that change were gradual, then
one would expect that the fO2 of volcanic gases would have
been significantly lower in the early Archaean relative to the
late Archaean and Palaeoproterozoic. However, observations
(Canil 1999; Delano 2001; Li and Lee 2004) indicate that this
progression occurred before the Archaean, and that for much
of the Archaean, the mantle was buffered near FMQ. In detail,
the analyses of Li and Lee (2004) allow for a small (~0.5 log
unit) increase in mantle fO2 over the Archaean. This could be
enough (Fig. 8.2b), if the system were poised near the critical
threshold where the reducing power of volcanoes is just
sufficient to quench the oxidising power of the burial of
organic matter (Holland 2002).
The crustal redox evolution argument is similar, except
the H loss to space is presumed to drive an increase in the
oxidation state of the crust and the metamorphic fluids
derived from it. Metamorphic fluids contain CO2 and H2.
Released to an anoxic atmosphere, the H2 is subject to
escape to space, so, similar to the mechanism for the mantle,
the crust/hydrosphere/atmosphere becomes more oxidised.
The rock cycle (uplift, weathering, sedimentation, burial and
metamophism) mixes the crust, and gradually its oxidation
state increases (Catling et al. 2001; Claire et al. 2006).
A variant of the mantle redox evolution hypothesis is
based on the observation that evidence for subaerial
volcanism prior to the Archaean-Proterozoic boundary
is sparse, whereas from the earliest Palaeoproterozoic
onward the evidence is abundant and clear. The growth
in abundance of subaerial volcanism was linked to the
stabilisation of the cratons of the continents that allowed
them to grow, supporting more and larger volcanoes.
Kump and Barley (2007) proposed that this transition
was abrupt, occurring coincidently with the initiation of
the GOE, and that it increased the mean fO2 of volcanic
gases, reducing the sink for O2 associated with volcanic
gas oxidation. Support for the Kump and Barley (2007)
hypothesis comes from the detailed modeling of the
evolving Archaean MIF-S signature, which indicates a
substantial increase in the volcanic SO2:H2S ratio in the
Late Archaean (Halevy et al. 2010).
In actuality, these three mechanisms for a reduction in the
sink for oxygen are not mutually exclusive. The first two
share the common presumption that the crust/mantle/hydro-
sphere/atmosphere must become progressively more
oxidised as hydrogen escapes from an anoxic atmosphere.
The third can accommodate this secular change while
providing an explanation for why the “final straw” reduction
in O2 sink occurred when it did, at the Archaean/Proterozoic
boundary.
Archaean “Whiffs” of Oxygen and BistableStates
If oxygenic photosynthesis did evolve and become a glob-
ally important source of organic matter (and net oxygen
production) prior to the GOE, then one might expect that
the volcanic or metamorphic sink for oxygen might have
become overwhelmed at times. Moreover, oxygen levels
might have been intermittently high in productive regions
of the ocean or land surface.
Support for this inference is provided by geochemical and
isotopic indications of “whiffs” of oxygen during the
Neoarchaean (between 2800 and 2500 Ma; Anbar et al.
2007; Kaufman et al. 2007). In South Africa there is a
trend toward Mo concentrations and d98/95Mo values above
the crustal average in Ghaap Group sediments between 2640
and 2500 Ma that Wille et al. (2007) interpret to reflect a
gradual increase, with significant fluctuations, in atmo-
spheric oxygen levels in the latest Archaean. A similar
trend in Mo concentration has been documented for the
c. 2500 Ma old Mt. McRae Shale in Western Australia
(Anbar et al. 2007). The D33S of Mt. McRae pyrites was
non-zero but negative during the interval of Mo enrichment,
interpreted as indicative of an atmospheric source of
sulphate and accumulation at low concentration in the deep
ocean. Molybdenum is mobilised under even mildly oxic
weathering conditions (below the MIF-S threshold; Anbar
et al. 2007), can accumulate as molybdate ion in oxic sea-
water, and will be quantitatively removed from the water
column into the sediment under euxinic conditions. Thus, it
is unclear how high atmospheric O2 levels rose during the
“whiffs”. Cr isotopes in banded iron formations also indicate
episodic atmosphere/ocean oxidation in the Neoarchaean
(Frei et al. 2009) perhaps involving the activity of aerobic,
pyrite-oxidising, acid-generating bacteria (Konhauser et al.
2011).
Transient oxygen increases prior to the GOE are an
expected consequence of a system like the global O2 cycle
gradually approaching a threshold value in which stochastic
events briefly force the system into the new state “prema-
turely.” Goldblatt et al. (2006) envisioned a bistable region
11 8.1 The Great Oxidation Event 1519
of low and high O2 concentrations (Fig. 8.3) that existed just
before the GOE, as a consequence of steadily declining input
of reductant to the surface environment (e.g. from volcanic
emissions; see above). Strong positive feedback would link
increasing atmospheric oxygen concentrations to the devel-
opment of a stratospheric ozone layer that reduces methane
oxidation rates, thus further increasing atmospheric oxygen
levels (Goldblatt et al. 2006) until they were sufficiently high
to support aerobic respiration, which would then replace
anaerobic metabolism as the predominant pathway of
organic-matter remineralisation. O2 levels would fluctuate
between these states until the GOE, when the low O2 equi-
librium state was lost.
8.1.4 Major Unresolved Problems, and thePromise of FAR-DEEP
There has been a considerable amount of research performed
on the GOE interval of time, including decades of research
in Fennoscandia, but still important problems remain unre-
solved. Thus, there is great promise for future researchers to
utilise the FAR-DEEP core materials and provide answers to
the following vexing questions.
Were GOE Groundwaters Oxidising?
Particularly understudied is the expression of the GOE in
terrestrial environments, especially in terms of penetration
oxidative weathering processes into soils and regolith. The
Kuetsj€arvi-age volcanic rocks of Fennoscandia are highly
oxidised with haematite-containing amygdales (Fig. 8.4a).
A possible explanation for these features is alteration by
oxidising groundwaters (see Chap. 7.4). A comprehensive
study of palaeoweathering features in FAR-DEEP materials
(Fig. 8.1) from Hole 1A (Seidorechka weathering crusts)
through the Kuetsj€arvi volcanic rocks (Holes, 6A, 7A and
8B) to the volcaniclastic red beds of the Kolosjoki Sedimen-
tary Formation in Holes 8A and 8B and the palaeoweathered
surfaces at the top of the Tulomozero Formation (Hole 11A,
Onega Basin) should be performed, and include a careful
examination of various other volcanic flow surfaces for
evidence of surface alteration by weathering.
Did GOE Magmas Have Abnormally High fO2?
Alternatively, the highly oxidised lavas of the Kuetsj€arvi
Formation (Fig. 8.4b) may reflect a high fO2 mantle source
region. In the hypothesis of Kump et al. (2001), Archaean
subduction brought highly oxidised, oceanic slabs (perhaps
incorporating banded iron formations; Dobson and Brodholt
2005) to the core/mantle boundary where they accumulated.
Then, in the Neoarchaean, these materials became buoyantly
unstable and erupted at the surface as deep-sourced mantle
plumes. The reduction in oxygen demand associated with
eruption of these high fO2 lavas finally allowed for the
accumulation of oxygen in the atmosphere at the
Archaean-Proterozoic boundary. The high Fe(III)/Fe(II)
Kuetsj€arvi lavas, if not simply the result of surface oxida-
tion, may represent the preserved remnants of these high fO2
eruptions. Detailed petrographic analysis of the distribution
of haematite in the FAR-DEEP Kuetsj€arvi lavas, leading to aclear description of the origin of Fe(III) enrichment, together
with analysis of trace-metal (Cr, V) proxies of original
oxidation state, may reveal whether the present-day highly
oxidised nature of these rocks reflects original upper mantle
conditions, oxidative weathering penecontemporaneous
with eruption, or modern-day oxidation.
What Does the Appearance and Abundanceof Sulphate Minerals Represent?
In Chap. 7.5, we presented evidence for substantial evaporite
sulphate (gypsum) accumulation during the Palaeopro-
terozoic, indicating that ocean sulphate concentrations in
the millimolar (mM) range were achieved, presumably for
the first time in Earth history. The FAR-DEEP core materials
exhibit many of these features, including pseudomorphs
after gypsum and anhydrite. In addition, other available
core materials demonstrate that massive sulphate deposition
occurred during the Palaeoproterozoic (Morozov et al. 2010;
Krupenik et al. 2011; Fig. 8.5). Moreover, based on their
study of sulphate relicts in pseudomorphic evaporite
minerals as well as carbonate-associated sulphate from the
FAR-DEEP drill cores, Reuschel et al. (2012) found little
stratigraphic variability of d34S over several hundred meters
of section in the c. 2100 Ma Tulomozero Formation, arguing
for a minimal sulphate concentration of 2.5 mM at this time.
Thus, physical and chemical evidence all point to the devel-
opment of a sulphate-rich ocean in the early Palaeopro-
terozoic at the time of the Lomagundi-Jatulian carbon
isotope excursion. This conclusion is at odds with earlier
arguments for low oceanic sulphate concentrations in the
Palaeoproterozoic based on the presumed absence of signifi-
cant gypsum accumulation, the limited degree of sulphur
isotope fractionation reflected in pyrite d34S (Canfield 1998).
Most important, however, is the temporal and presumed
causal relationship between the appearance of massive
sulphate accumulations in the rock record in the early
Palaeoproterozoic and the time when atmospheric oxygen
concentration displayed a substantial rise. This rise in
1520 L.R. Kump et al.
atmospheric oxygen abundance is tied to the permanent
disappearance of the mass-independently fractionated sul-
phur isotope signature (MIF-S) from the sedimentary record
at c. 2.3 Ga (Bekker et al. 2004; Guo et al. 2009). Thus, this
point in time represents the termination of the delivery of
photochemically produced sulphate carrying a mass-
independently fractionated sulphur isotope signal from the
atmosphere to Earth surface environments. Instead, a rising
atmospheric oxygen concentration prevents this atmospheric
delivery, but fosters oxidative weathering of sulphides on the
continents and the subsequent delivery of sulphate to the
ocean. Thus, the appearance of massive sulphate
occurrences in the aftermath of the GOE marks an important
turning point in the evolution of the global sulphur cycle,
notably a change from a strong atmospheric contribution to a
modern-style sulphur cycle that is governed by oxidative
weathering, bacterial sulphate reduction and subsequent
pyrite burial. But, was this the first time in Earth history
that sulphate had accumulated in the ocean?
There are indeed sulphate mineral accumulations
that have been interpreted controversially to indicate a
much earlier, perhaps local oxygenation of ocean seawater
(Lambert et al. 1978). These are stratiform barite (barium
sulphate) deposits in Western Australia, South Africa, and
India, ranging in age from 3500 to 3200 Ma (Hickman 1973;
Heinrichs and Reimer 1977; Reimer 1980; van Kranendonk
et al. 2008; Huston and Logan 2004). Different modes of
formation have been discussed (most specifically for the
North Pole barite in Western Australia) ranging from an
evaporitic origin (Lambert et al. 1978) or the discharge of
barium-laden hydrothermal fluids into a sulphate-rich
marine basin (Buick and Dunlop 1990) to a hydrothermal
origin of the Australian barites (van Kranendonk et al. 2008).
Most importantly, however, is the fact that the barites carry a
mass-independently fractionated sulphur isotope signature
(e.g. Farquhar et al. 2000; Bao et al. 2007; Ueno et al.
2008), inconsistent with their formation in the presence of
abundant atmospheric oxygen. Independent of their mode of
precipitation, the presence of MIF-S in these barites
indicates a strong input signal of atmospherically derived
sulphate rather than sulphate derived from oxidative
weathering on land and delivery of sulphate to the ocean.
However, it does not rule out the possibility of deposition in
a locally oxygenated aquatic environment.
Little is known about the early Archaean oceanic sulphate
concentration. Some indication is provided by microscopic
pyrite co-existing with the barite from the North Pole area
(Dresser Formation) of Western Australia and their sulphur
isotopic composition. These pyrites also carry a MIF-S sig-
nature. In addition and most importantly, however, these
pyrites display mass-dependent fractionations of the sulfur
isotopes of up to 24 ‰ (d34Sbarite-d34Spyrite), potentially
indicating the activity of sulphate-reducing bacteria (Shen
et al. 2009; Ueno et al. 2008), although alternative
conclusions have been drawn (Philippot et al. 2007). Based
on experimental work, Habicht et al. (2002) proposed a
minimum sulphate concentration of 200 mM for the expres-
sion of high-magnitude mass-dependent sulphur isotopic
fractionation. Recently, Canfield et al. (2010) reported
60–70 ‰ fractionations during bacterial sulphate reduction
at ambient sulphate concentrations of 1.1–2 mM. Thus, the
expression of high-magnitude mass-dependent sulphur iso-
topic fractionations provides only a minimum level for oce-
anic sulphate abundance somewhere in the 1 mM range.
In conclusion, despite recent advances in our understand-
ing of atmospheric oxygenation, problems remain with the
simplified story of an anoxic Archaean and an oxygen-rich
post-Archaean (Ohmoto et al. 2006). The observation of a
strong temporal and likely causal link between a significant
rise in atmospheric oxygen and the appearance of massive
evaporitic sulphate occurrences in the early Palaeopro-
terozoic rock record remains. The signal of high-magnitude
mass-dependent sulphur isotopic fractionation as indicative
of bacterial sulphate reduction is most prominently and
consistently developed in the sedimentary rock record
starting at 2.3 Ga (e.g. Cameron 1982; Bekker et al. 2004),
but earlier occurrences have been reported from 2.7 Ga (e.g.
Grassineau et al. 2006) and, as discussed above, the 3.5 Ga
Dresser Formation. Oxygen may have first appeared in
microbial mats or surface water “oxygen oases” long before
it accumulated in the atmosphere (e.g. Kasting et al. 1992),
and in these environments sulphate may have accumulated.
Detailed analysis of FAR-DEEP core materials, together
with ongoing work on Archaean (and also younger Precam-
brian) rocks, will almost certainly reveal a far more complex
history of surface oxidation.
What Are the Implications of the GOE forOrganic-Matter Recycling in Sediments?
Today, aerobic decomposition in the water column and
surface sediments is the predominant and most energetic
mode of organic matter remineralisation. In sediments, a
predictable sequence of remineralisation reactions with
depth ensues: aerobic decomposition followed by denitrifi-
cation, Mn- and Fe-oxide reduction, sulphate reduction, and
finally fermentative methanogenesis (e.g. Berner 1980). In
an anoxic, low-sulphate world the series would be signifi-
cantly truncated, with methanogenesis as the principal
mechanism to recycle organic matter in the water column
and sediments (Hayes and Waldbauer 2006). Thus, the
oxygenation of the atmosphere–ocean system presumably
provided additional terminal electron acceptors such as
11 8.1 The Great Oxidation Event 1521
sulphate, which would result in a more diverse carbon
cycling in the sedimentary realm. Melezhik et al. (1999,
2005) and Fallick et al. (2008, 2011) presented data
supporting this hypothesis in the form of carbon isotopic
compositions of diagenetic carbonates of Archaean and Pro-
terozoic age. Interestingly, pre-Palaeoproterozoic diagenetic
carbonates, with the exception of those found in banded iron
formations, have d13C values close to 0 ‰ (� 3 ‰). Diage-
netic concretions are also exceedingly rare in the Archaean.
In contrast, Palaeoproterozoic diagenetic carbonates, nota-
bly those from Fennoscandia but from elsewhere as well, are
abundant (Fig. 8.6), and display negative d13C values as low
as �22 ‰, indicative of incorporation of C derived from
organic-matter remineralisation (Fig. 8.7).
Fallick, Melezhik and colleagues (op. cit.) hypothesised
that oxygenation of the water column was accompanied by
obligate anaerobes deserting this now uncongenial environ-
ment and colonising the oxygen-free zones of sediments.
Aerobic recycling of organic matter could now take place
in appropriate, oxygenated areas of the water column, and
perhaps upper sediment. In deeper zones of the sediment,
sulphate reduction (stimulated by the accumulation of oce-
anic dissolved sulphate from oxidative weathering) released13C-depleted carbon into the porewaters, producing diage-
netic carbonates with negative d13C. The much enhanced
frequency of nodular carbonate concretions is a consequence
of this change. Deeper (or at least elsewhere) in the sediment,
fermentative methanogenesis produced strongly 13C-depleted
methane, which was newly available for incorporation into
clathrate-hydrates and potential long(�ish)-term sequestration.
The relative paucity of observed carbonate cements and
nodules with positive d13C approaching +10 to +15 ‰suggests that other, likely geochemical, considerations
militated against direct precipitation of the oxidised carbon
released during fermentation reactions. The crucial point is
that following oxygenation of the Earth’s surface, novel
recycling of organic matter within sediments created new,
or enhanced, opportunities for sedimentary sequestration of13C-depleted carbon over timescales longer than the ocean
residence time of carbon.
Hayes and Waldbauer (2006) have argued that as
methanogens were forced deeper into the sediment during
the GOE, 13C-enriched dissolved inorganic carbon began to
accumulate in porewaters, leading to diagenetic carbonates
with elevated d13C; in other words, the LJ carbonates are
themselves of diagenetic origin, and their d13C is not repre-
sentative of the contemporaneous ocean.
Both the Fallick-Melezhik (F-M) and Hayes-Waldbauer
(H-W) models invoke a migration of obligate anaerobes, and
hence sulphate reduction and methanogenesis, from the
water-column and sediment-water-interface to a deeper dia-
genetic setting. They differ distinctly in their predictions in
terms of the d13C of the resulting diagenetic carbonates, with
F-M predicting low d13C carbonates and H-W predicting
high d13C carbonates. Key to resolving the two models
resides in the interpretation of H-W that the LJ high d13Ccarbonates are of diagenetic origin. Careful petrographic
work combined with the dedicated isotopic study involving
high spatial resolution analyses on the FAR-DEEP LJ
carbonates could go a long way toward reconciling the two
models.
Why Did the Lomagundi-Jatuli C IsotopeExcursion Postdate the Onset of the GOE?
The precise timing of the onset of the Lomagundi-Jatuli (LJ)
C isotope excursion is a key uncertainty that can be
addressed through analysis of the FAR-DEEP core
materials. Evidence from South Africa (Duitschland Forma-
tion) indicates a shift toward positive values just after the
loss of the MIF-S signature (Guo et al. 2009), but the
prevailing interpretation is that this is a pre-LJ excursion
rather than the onset of the LJ-CIE because of the occurrence
of “normal” (near-zero) d13C values of the (overlying?)
Mooidraai Formation dolomites (Bekker et al. 2001). If
instead the LJ-CIE was initiated in the Duitschland, then a
much longer excursion is indicated. In the Pechenga Green-
stone Belt the maximum age of the termination of the LJ-
CIE has been refined by Melezhik et al. (2007) as 2058 � 6
Ma by U-Pb zircon dating. Further refinement is possible
given the greater stratigraphic completeness of the FAR-
DEEP core materials and the potential overlap of isotopi-
cally heavy carbonate sequences with organic-C rich shales
in the Onega Basin that could extend the LJ-CIE into the
interval of “Shunga” deposition.
The concept of the GOE as an “event” may need
rethinking as well. Beginning with the “whiffs” of the
Neoarchaean indicating an initiation of oxidative weathering
at perhaps very low atmospheric pO2 (10�8–10�5 PAL;
Reinhard et al. 2009), through the loss of MIF-S (>10�5
PAL; Pavlov and Kasting 2002) in the early Palaeopro-
terozoic, to the indications of deep crustal oxidation
associated with uranium accumulation (Gauthier-Lafaye
and Weber 2003) and supergene iron-ore concentration
(M€uller et al. 2005) in the later Palaeoproterozoic, the oxi-
dation of the Earth’s surface spans a half-billion years,
crossing progressively higher redox thresholds for surface
oxidation processes along the way. In this context the LJ-
CIE may indeed represent a significant oxygen-producing
interval (if it was indeed driven by an interval of enhanced
burial of organic-carbon derived from oxygenic photosyn-
thesis), postdating the MIF-S threshold crossing but
representing the quantitatively most important oxygen rise
1522 L.R. Kump et al.
of the GOE. As it spans this important interval of time, the
FAR-DEEP core archive holds great promise for future
efforts to further refine the detailed history of surface
oxygenation through the Palaeoproterozoic.
The answers to these and a host of other scientific
questions are locked in the chemical, isotopic, chronostra-
tigraphic and petrographic characteristics of the FAR-DEEP
core materials. It is up to future researchers to find the keys
that unlock these mysteries. Successful researchers will
begin by obtaining the necessary background information
on the Palaeoproterozoic of Fennsocandia in Volume 1 of
this treatise and familiarising oneself with what is available
in the FAR-DEEP core repository by perusing Volume 2.
The principal investigators of FAR-DEEP are dedicated to
providing anyone the opportunity to access these valuable
materials so that we may one day have a much more com-
plete view of the establishment of an aerobic Earth system in
the Palaeoproterozoic.
11 8.1 The Great Oxidation Event 1523
Fig. 8.1 Simplified lithological column of the North Pechenga Group,
positions of FAR-DEEP and other relevant drillholes. Also shown is
how the evolution of the Pechenga Greenstone Belt is related to global
palaeoenvironmental events. Superscripts denote radiometric ages
from (1) Amelin et al. (1995), (2) Melezhik et al. (2007), (3) Hannah
et al. (2006), and (4) Hanski (1992)
1524 L.R. Kump et al.
Fig. 8.2 Explanations for the timing of
the Great Oxidation Event near the
Archaean-Proterozoic boundary. (a) This
is the time when cyanobacteria evolve
oxygenic photosynthesis. Prior to this
time, there is no source of oxygen. Then
oxygenic cyanobacteria evolve and
spread, creating a source for oxygen that
ultimately comes to match the (potential)
sink for oxygen associated with reactions
with volcanic gases and reduced crustal
rocks during weathering. (b) Oxygenic
photosynthesis evolves in the
Neoarchaean, but sinks associated with
reduced volcanic and metamorphic fluids
exceed the source of oxygen. These sinks
diminish with time because hydrogen
escape from the anoxic atmosphere leads
to net oxidation of the upper mantle
(Kasting et al. 1993; Kump et al. 2001)
and/or the crust (Catling et al. 2001). Near
the Archaean-Proterozoic boundary, the
sink becomes smaller than the oxygen
source, oxygen levels rise, and balance is
achieved as oxygen-dependent crustal
weathering comes to balance the deficit in
oxygen sink between oxygenic production
and consumption by reaction with
volcanic and metamorphic gases. (c)
Similar to case b, but here there is a
sudden decrease in the volcanic sink for
oxygen associated with a substantial
reduction in the proportion of submarine
(more reducing) volcanism (increase in
subaerial volcanism) associated with the
major episode of continental stabilisation
that defines the Archaean-Proterozoic
boundary (Kump and Barley 2007)
Fig. 8.3 Cartoon of the evolution of the
stability of atmospheric oxygen levels as
the supply of reductants to the Earth
surface diminishes (After Goldblatt et al.
2006). At high reductant supplies, only a
low oxygen level is stable. As reductant
supply diminishes, a region of bistability
emerges in which “whiffs” of oxygen are
possible, representing transient
perturbations into the metastable high
oxygen state. Eventually only the high
oxygen level is stable (After Scheffer and
Carpenter (2003))
11 8.1 The Great Oxidation Event 1525
Fig. 8.4 Igneous rocks from the Kuetsj€arvi Volcanic Formation in the
Pechenga Greenstone Belt. (b) Lava breccia of trachydacite occurring
at the contact between the Kuetsj€arvi Volcanic Formation and the
Kolosjoki Sedimentary Formation. Lava fragments are separated by
black, haematite-magnetite-rich “bands” that contain up to 25 wt.%
Fetot (mainly as haematite) and 7 wt.% K2O, apparently accumulated as
the result of upper mantle oxidation, or post-volcanic hydrothermal
alteration, or surface weathering, or combination of all. (a) Highly-
oxidised, trachytic lava containing vesicles filled with quartz and jasper
(dark red); core diameter is 5 cm (Photographs by Victor Melezhik)
1526 L.R. Kump et al.
Fig. 8.5 Sulphates and their pseudomorphes in the Tulomozero For-
mation from the Onega Basin (for details see Chaps. 4.3, 6.3.1, 6.32,
and 7.5). (a, b) Dark grey dolomarl and brown mudstone with silica
(pink)- and dolospar (white, yellow)-psedomorphed nodular calcium
sulphate retaining micro-relicts of anhydrite, barite and celestite;
FAR-DEEP Cores 10A and 10B. (c) Dark brown mudstone with
rosettes and individual crystals of gypsum replaced by white blocky
dolomite; FAR-DEEP Core 10A. (d) Unsawn cores of massive anhy-
drite from c. 3,500-m-deep Onega parametric hole; sample courtesy of
the Institute of Geology, Karelian Science Center, photograph courtesy
of Dmitry Rychanchik. Core diameter in (a–c) is 5 cm, and 10 cm in (d)
(Photographs (a–c) by Victor Melezhik)
11 8.1 The Great Oxidation Event 1527
Fig. 8.6 Diagenetic carbonates in 2000–1900 Ma sedimentary
formations from various regions. C. 2000 Ma Pulguj€arvi SedimentaryFormation, Pechenga Greenstone Belt, northwestern Fennoscandia:(a) Unsawn core with large, lensoidal, 13C-depleted (d13C ¼�10.5 ‰; V-PDB) calcite concretions in marine greywacke.
(b) Unsawn core with large, zoned, 13C-depleted (d13C ¼ �13.6 ‰;
V-PDB) calcite concretions in marine greywacke; pale grey core is
calcite-rich, whereas the outer rim is enriched in SiO2. (c) Sawn core
with 13C-depleted (d13C ¼ �11.1 ‰; V-PDB) calcite concretions
with irregular form in marine greywacke; concretions show recessive
relief because core slab was treated by hydrochloric acid. (d) A slab
showing marine, turbiditic greywacke-siltstone; grey-greenish
greywacke ripples are cemented with 13C-depleted (d13C ¼ �12.5 ‰;
V-PDB) calcite; note the calcite concretion with concentric zoning
occurring in the lower right corner. C. 1960–1920 Ma PovungnitukSlate (Lesher 1999), East-Central Cape Smith Belt, Canada: (e,f)
Pyrrhotite-rimmed, 13C-depleted (d13C ¼ �10.8 ‰ to �8.6 ‰;
V-PDB) calcite concretions in turbiditic shale. Core diameter in (a–c)
is 5 cm, in (e, f) is 3.5 cm
1528 L.R. Kump et al.
Fig. 8.6 (continued) C. 1900 Ma Omarolluk Formation, BelcherIslands, Canada: (g) Large, spherical calcite concretions in organic
carbon-rich argillites; hammer head is c. 14 cm. (h) Enlarged image of
a concretion (denoted by rectangle in “g”) showing zoning. (i) Large,
lensoidal carbonate concretions in organic carbon-rich argillites; the
length of the largest concretion is c. 2 m. C. 1950 Ma, Ladoga series,Ladoga lake area: (j) Former carbonate concretion in a high-grade
amphibolites-facies gneiss (originally marine turbiditic sandstone-
siltstone) affected by partial melting (bright patches). C. 1950 MaKondopoga Formation, Onega Basin: (k) Lacustrine, turbiditic
greywacke-siltstone with concretionary layers of 13C-depleted (d13C ¼
�10.2 ‰ to �16.5 ‰; V-PDB) ankerite (pale brown bands). (l)
Ankerite-cemented sandstone bed (d13C ¼ �12.3 ‰; V-PDB) in
lacustrine turbiditic succession; coin diameter is 2 cm. (m) Ankeritic,13C-depleted (d13C ¼ �17.3 ‰; V-PDB) concretion in lacustrine,
turbiditic greywacke siltstone; note considerable differential compac-
tion of layers outside the concretion, implying early diagenetic cemen-
tation. (n) Bedding surface of a sandstone bed with numerous sausage-
shaped, 13C-depleted (d13C ¼ �15.5 ‰; V-PDB) calcite concretions;
hammer head is 14 cm. Photographs (g–i) courtesy of Dominic
Papineau. Photographs (a–f, j–n) by Victor Melezhik (Carbon isotope
data by Melezhik and Fallick (unpublished))
11 8.1 The Great Oxidation Event 1529
Fig. 8.7 Carbon and oxygen isotope composition of diagenetic
carbonates from the c. 2000 Ma Pilguj€arvi Sedimentary Formation of
the Pechenga Greenstone Belt in Fennoscandia. d13C values are uni-
formly low, consistent with their formation in an early diagenetic
setting dominated by aerobic remineralisation and sulphate reduction,
pursuant to the establishment of an aerobic biosphere. The diagram is
based on drillcore 2900 (see Fig. 8.1 for stratigraphic location) and
unpublished analyses of A (Fallick and V. Melezhik. For analytical
protocol see Chap. 5)
1530 L.R. Kump et al.
References
Amelin YuV, Heaman LM, Semenov VS (1995) U-Pb geochronology
of layered mafic intrusions in the eastern Baltic Shield: implications
for the timing and duration of Palaeoproterozoic continental rifting.
Precambrian Res 75:31–46
Anbar AD, Duan Y, Lyons TW, Arnold GL, Kendall B, Creaser RA,
Kaufman AJ, Gordon GW, Scott C, Garvin J, Buick R (2007) A
whiff of oxygen before the great oxidation event? Science
317:1903–1906
Baker AJ, Fallick AE (1989) Evidence from Lewisian limestones for
isotopically heavy carbon in two thousand million year old sea
water. Nature 337:352–354
Bao H, Rumble D, Lowe DR (2007) The five stable isotope
compositions of Fig Tree barites: implications on sulfur cycle in
ca. 3.2 Ga oceans. Geochim Cosmochim Acta 71:4868–4879
Bau M, Romer R, Luders V, Beukes NJ (1999) Pb, O, and C isotopes in
silicified Mooidraai dolomite (Transvaal Supergroup, South
Africa): implications for the composition of Palaeoproterozoic sea-
water and ‘dating’ the increase of oxygen in the Precambrian
atmosphere. Earth Planet Sci Lett 174:43–57
Bekker A, Kaufman AJ, Karhu JA, Beukes NJ, Swart QD, Coetzee LL,
Eriksson KA (2001) Chemostratigraphy of the Paleoproterozoic
Duitschland Formation, South Africa: implications for coupled
climate change and carbon cycling. Am J Sci 301:261–285
Bekker A, Holland HD, Young GM, Nesbitt HW (2003) Fe2O3/FeO ratio
in average shale through time: a reflection of the stepwise oxidation of
the atmosphere. Abstr Program Annu Meeting Geol Soc Am 34:83
Bekker A, Holland HD, Wang P-L, Rumble D, Stein HJ, Hannah JL,
Coetzee LL, Beukes NJ (2004) Dating the rise of the atmospheric
oxygen. Nature 427: 117–120
Berner RA (1980) Early diagenesis, a theoretical approach. Princeton
University Press, New Jersey, p 256
Brocks JJ, Logan GA, Buick R, Summons R (1999) Archaean molecular
fossils and the early rise of eukaryotes. Science 285:1033–1036
Brocks JJ, Buick R, Summons RE, Logan GA (2003) A reconstruction
of Archaean biological diversity based on molecular fossils from
the 2.78 to 2.45 billion-year-old Mount Bruce Supergroup,
Hamersley Basin, Western Australia. Geochim Cosmochim Acta
67:4321–4335
Buick R, Dunlop JSR (1990) Evaporitic sediments of early Archaean
age from the Warrawoona Group, North Pole, Western Australia.
Sedimentology 37:247–277
Cameron EM (1982) Sulphate and sulphate reduction in early Precam-
brian oceans. Nature 296:145–148
Canfield DE (1998) A new model for Proterozoic ocean chemistry.
Nature 396:450–453
Canfield DE (2005) The early history of atmospheric oxygen: homage
to Robert M. Garrels. Annu Rev Earth Planet Sci 33:1–36
Canfield DC, Steward FJ, Thamdrup B, de Brabandere L, Dalsgaard T,
DeLong EF, Revsbech NP, Uloa O (2010) A cryptic sulfur cycle in
oxygen-minimum-zone waters off the Chilean coast. Science
330:1375–1378
Canil D (1999) Vanadium partitioning between orthopyroxene, spinel
and silicate melt and the redox states of mantle source regions for
primary magmas. Geochim Cosmochim Acta 63:557–572
Catling DC, Zahnle KJ, McKay CP (2001) Biogenic methane, hydro-
gen escape, and the irreversible oxidation of early Earth. Science
293:839–843
Claire MW, Catling DC, Zahnle KJ (2006) Biogeochemical modelling
of the rise in atmospheric oxygen. Geobiology 4:239–269
Cornell DH, Schutte SS, Eglington BL (1996) The Ongeluk basaltic
andesite formation in Griqualand West, South Africa: submarine alter-
ation in a 2222 Ma Proterozoic sea. Precambrian Res 79:101–123
Delano JW (2001) Redox history of the Earth’s interior: implications
for the origin of life. Orig Life Evol Biosph 31:311–341
Dobson DP, Brodholt JP (2005) Subducted banded iron formations as a
source of ultralow-velocity zones at the core-mantle boundary.
Nature 434:371–374
Dutkiewicz A, Volk H, George SC, Ridley J, Buick R (2006)
Biomarkers from Huronian oil-bearing fluid inclusions: an uncon-
taminated record of life before the Great Oxidation Event. Geology
34:437–440
Eigenbrode JL, Freeman KH, Summons RE (2008) Methylhopane
biomarker hydrocarbons in Hamersley Province sediments provide
evidence for NeoArchaean aerobiosis. Earth Planet Sci Lett
273:323–331
Fallick AE, Melezhik VA, Simonson BM (2008) The ancient anoxic
biosphere was not as we know it. In: Dobretsov N, Kolchanov N,
Rozanov A, Zavarzin G (eds) Biosphere origin and evolution.
Springer, New York, pp 169–188
Fallick A, Melezhik V, Simonson B (2011) On Proterozoic ecosystems
and the carbon isotopic composition of carbonates associated
with Banded Iron Formations. In: Neves L et al (eds) Modelacao
de Sistemas Geologicos. Univeridade de Coimbra, Portugal, pp
57–71
Farquhar J, Bao H, Thiemens M (2000) Atmospheric influence of
Earth’s earliest sulphur cycle. Science 289:756–758
Frei R, Gaucher C, Poulton SW, Canfield DE (2009) Fluctuations in
Precambrian atmospheric oxygenation recorded by chromium
isotopes. Nature 461:250–254
Gauthier-Lafaye F, Weber F (2003) Natural nuclear fission reactors:
time constraints for occurrence and their relation to uranium and
manganese deposits and to the evolution of the atmosphere. Pre-
cambrian Res 120:81–101
Goldblatt C, Lenton TM, Watson AJ (2006) Bistability of atmospheric
oxygen and the great oxidation. Nature 443:683–686
Grassineau NV, Abell P, Appel PWU, Lowry D, Nisbet EG (2006)
Early life signatures in sulfur and carbon isotopes, from Isua,
Barberton, Wabigoon (Steep Rock), and Belingwe greenstone
belts. Geol Soc Am Mem 198:33–52
Guo Q, Strauss H, Kaufman AJ, Schr€oder S, Gutzmer J, Wing B, Baker
MA, Bekker A, Jin Q, Kim S-T, Farquhar J (2009) Reconstructing
Earth’s surface oxidation across the Archaean-Proterozoic transi-
tion. Geology 37:299–402
Habicht KS, Gade M, Thamdrup B, Berg P, Canfield DE (2002)
Calibration of sulfate levels in the Archean ocean. Science 298:
2372–2374
Halevy I, Johnston DT, Schrag DP (2010) Explaining the structure of
the Archaean mass-independent sulphur isotope record. Science
329:204–207
Hannah JL, Stein HJ, Zimmerman A, Yang G, Markey RJ, Melezhik
VA (2006) Precise 2004 � 9 Ma Re‐Os age for Pechenga black
shale: comparison of sulfides and organic material. Geochim
Cosmochim Acta 70:A228
Hanski EJ (1992) Petrology of the Pechenga ferropicrites and
cogenetic, Ni-bearing gabbro-wehrlite intrusions, Kola Peninsula,
Russia. Geol Surv Finland Bull 367:1–192
Hayes JM,Waldbauer JR (2006) The carbon cycle and associated redox
processes through time. Philos Trans R Soc B 361:931–950
Hedges SB, Kumar S (2009) The timetree of life. United Kingdom,
Oxford, p 551
Hedges SB, Chen H, Kumar S, Wang DY-C, Thompson AS, Watanabe
H (2001) A genomic timescale for the origin of eukaryotes. BMC
Evol Biol 1:4
Heinrichs TK, Reimer TO (1977) A sedimentary barite deposit from the
Archean Fig Tree Group of the Barberton Mountain Land (South
Africa). Econ Geol 72: 1426–1441
Hickman AH (1973) The North Pole barite deposits, Pilbara Goldfield.
Ann Rept Geol Surv W Aust 1972:57–60
11 8.1 The Great Oxidation Event 1531
Holland HD (1978) The chemistry of the atmosphere and oceans.
Wiley, New York, p 351
Holland HD (2002) Volcanic gases, black smokers, and the great
oxidation event. Geochim Cosmochim Acta 66:3811–3826
Holland HD (2006) The oxygenation of the atmosphere and oceans.
Philos Trans R Soc B 361:903–915
Huston DL, Logan GA (2004) Barite, BIFs and bugs: evidence for the
evolution of the Earth’s early hydrosphere. Earth Planet Sci Lett
220:41–55
Karhu JA (1993) Palaeoproterozoic evolution of the carbon isotope
ratios of sedimentary carbonates in the Fennoscandian Shield.
Geol Surv Finland Bull 371:1–87
Karhu JA, Holland HD (1996) Carbon isotopes and rise of atmospheric
oxygen. Geology 24:867–879
Kasting JF, Holland HD, Kump LR (1992) Atmospheric evolution: the
rise of oxygen. In: Schopf JW, Klein C (eds) The Proterozoic
biosphere: a multidisciplinary study. Cambridge University Press,
Cambridge, pp 159–165
Kasting JF, Eggler DH, Raeburn SP (1993) Mantle redox evolution
and the case for a reduced Archaean atmosphere. J Geol
101:245–257
Kaufman AJ, Johnston DT, Farquhar J, Masterson AL, Lyons TW,
Bates S, Anbar AD, Arnold GL, Garvin J, Buick R (2007) Late
Archaean biospheric oxygenation and atmospheric evolution. Sci-
ence 317:1900–1903
Kirschvink JL, Kopp RE (2008) Palaeoproterozoic ice houses and the
evolution of oxygen-mediating enzymes: the case for a late origin of
photosystem II. Philos Trans R Soc B 363:2755–2765
Knoll AH, Holland HD (1995) Oxygen and Proterozoic evolution: an
update. In: Board on Earth Sciences and Resources (ed) Effects of
past global change on life. National Research Council, Washington,
DC, pp 21–33
Konhauser KO, Lalonde SV, Planavsky NJ, Pecoits E, Lyons TW,
Mojzsis SJ, Rouxel OJ, Barley ME, Rosıere C, Fralick PW,
Kump LR, Bekker A (2011) Aerobic bacterial pyrite oxidation
and acid rock drainage during the great oxidation event. Nature
478:369–373
Kopp RE, Kirschvink JL, Hilburn IA, Nash CZ (2005) The Paleopro-
terozoic snowball Earth: a climate disaster triggered by the evolu-
tion of oxygenic photosynthesis. Proc Natl Acad Sci USA
102:11131–11136
Krupenik VA, Akhmedov AM, Sveshnikova KYu (2011) Structure of
the section of the Onega structure based on the Onega parametric
drillhole. In: Glushanin LV, Sharov NV, Shchiptsov VV (eds) The
Onega Palaeoproterozoic structure (Geology, tectonics, deep struc-
ture and minerageny). Institute of Geology, Karelian Research
Centre of RAS, Petrozavodsk, pp 172–189 (in Russian)
Kump LR, Barley ME (2007) Continental tectonics, increased subaerial
volcanism and the rise of atmospheric oxygen. Nature 448:
1033–1036
Kump LR, Kasting JF, Barley ME (2001) Rise of atmospheric oxygen
and the “upside-down” Archaean mantle. Geochem Geophys
Geosyst 2: Paper number 2000GC000114
Lambert IB, Donnelly TH, Dunlop JSR, Groves DI (1978) Stable
isotopic compositions of early Archaean sulphate deposits of prob-
able evaporitic and volcanogenic origins. Nature 276:808–810
Lesher CM (1999) Regional stratigraphic, structural, and metamorphic
setting of the East-Central Cape Smith Belt. In: Lesher CM (ed)
Komatiitic Peridotite-Hosted Ni-Cu-(PGE) Deposits of the Raglan
Area, Cape Smith belt, New Quebec, vol 2, Guide Series. Mineral
Exploration Research Centre/Laurentian University, Sudbury,
pp 17–34
Li ZXA, Lee CTA (2004) The constancy of upper mantle fO2 through
time inferred from the V/Sc ratios in basalts. Earth Planet Sci Lett
228:483–493
Melezhik VA, Fallick AE (1996) A widespread positive d13Ccarb anom-
aly at around 2.33-2.06 Ga on the Fennoscandian shield: a paradox?
Terra Nova 8:141–157
Melezhik VA, Fallick AE, Medvedev PV, Makarikhin VV (1999)
Extreme 13Ccarb enrichment in ca. 2.0 Ga magnesite-stromatolite-
dolomite-‘red beds’ association in a global context: a case for the
world-wide signal enhanced by a local environment. Earth Sci Rev
48:71–120
Melezhik VA, Fallick AE, Hanski EJ, Kump LR, Lepland A, Prave AR,
Strauss H (2005) Emergence of the aerobic biosphere during the
Archaean-Proterozoic transition: challenges of future research.
GSA Today 15:4–11
Melezhik VA, Huhma H, Condon DJ, Fallick AE, Whitehouse MJ
(2007) Temporal constraints on the Paleoproterozoic Lomagundi-
Jatuli carbon isotopic event. Geology 35:655–658
Morozov AF, Hahaev BN, Petrov OV, Gorbachev VI, Tarkhanov GB,
Tsvetkov LD, Erinchek YuM, Akhmedov AM, Krupenik VA,
Sveshnikova KYu (2010) Rock-salts in Palaeoproterozoic strata of
the Onega depression of Karelia (based on data from the Onega
parametric drillhole). Commun Russ Acad Sci 435:230–233
(in Russian)
M€uller SG, Krapez B, Barley ME, Fletcher IR (2005) Giant iron-ore
deposits of the Hamersley province related to the breakup of
Paleoproterozoic Australia: new insights from in situ SHRIMP
dating of baddeleyite from mafic intrusions. Geology 33:577–580
Ohmoto H, Watanabe Y, Ikemi H, Poulson SR, Taylor BE (2006)
Sulphur isotope evidence for an oxic Archaean atmosphere. Nature
442:908–911
Papineau D, Mojzsis SJ, Schmitt AK (2007) Multiple sulphur isotopes
from Paleoproterozoic Huronian interglacial sediments and the rise
of atmospheric oxygen. Earth Planet Sci Lett 255:188–212
Pavlov AA, Kasting JF (2002) Mass-independent fractionation of sulfur
isotopes in Archaean sediments: strong evidence for an anoxic
Archaean atmosphere. Astrobiology 2:27–41
Philippot P, van Zuilen M, Lepot K, Thomazo C, Farquhar J, van
Kranendonk MJ (2007) Early Archaean mircroorganisms preferred
elemental sulfur, not sulfate. Science 317:1534–1537
Rashby SE, Sessions AL, Summons RE, Newman DK (2007) Biosyn-
thesis of 2-methylbacteriohopanepolyols by an anoxygenic
phototroph. Proc Natl Acad Sci USA 104:15099–15104
Rasmussen B, Fletcher IR, Brocks JJ, Kilburn MR (2008) Reassessing
the first appearance of eukaryotes and cyanobacteria. Nature
455:1101–1104
Reimer TO (1980) Archean sedimentary barite deposits of the
Swaziland supergroup barites and the evolution of the atmosphere
Reinhard CT, Raiswell R, Scott C, Anbar AD, Lyons TW (2009) A late
Archaean sulfidic sea stimulated by early oxidative weathering of
the continents. Science 326:713–716
Reuschel M, Melezhik VA, Whitehouse MJ, Lepland A, Fallick AE,
Strauss H (2012) Isotopic evidence for a sizeable seawater sulfate
reservoir at 2.1 Ga. Precambrian Geol 192–195:78–88
SchefferM,Carpenter SR (2003)Catastrophic regime shifts in ecosystems:
linking theory to observation. Trends Ecol Evol 18:648–656
Shen Y, Farquhar J, Masterson A, Kaufman AJ, Buick R (2009)
Evaluating the role of microbial sulfate reduction in the early
Archaean using quadruple isotope systematics. Earth Planet Sci
Lett 279:383–391
Tsikos H, Matthews A, Erel Y, Moore JM (2010) Iron isotopes con-
strain biogeochemical redox cycling of iron and manganese in a
Palaeoproterozoic stratified basin. Earth Planet Sci Lett
298:125–134
Ueno Y, Ono S, Rumble D, Maruyamas S (2008) Quadruple sulfur
isotope analysis of c. 3.5 Ga Dresser Formation: new evidence for
microbial sulfate reduction in the early Archaean. Geochim
Cosmochim Acta 72:5675–5691
1532 L.R. Kump et al.
Van Kranendonk MJ, Philippot P, Lepot K, Bodorkos S, Pirajno F
(2008) Geological setting of Earth’s oldest fossils in the ca.
3.5 Ga Dresser Formation, Pilbara Craton, Western Australia. Pre-
cambrian Res 167:93–124
Waldbauer JR, Sherman LS, Sumner DY, Summons RE (2009) Late
Archean molecular fossils from the Transvaal Supergroup record
the antiquity of microbial diversity and aerobiosis. Precambrian Res
169:28–47
Welander PA, Coleman ML, Sessions AL, Summons RE, Newman DK
(2010) Identification of a methylase required for 2-methylhopanoid
production and implications for the interpretation of sedimentary
hopanes. Proc Natl Acad Sci 107:8537–8542
Wille M, Kramers JD, Nagler TF, Beukes NJ, Schroder S, Meisel Th,
Lacassie JP, Voegelin AR (2007) Evidence for a gradual rise of
oxygen between 2.6 and 2.5 Ga from Mo isotopes and Re-PGE
signatures in shales. Geochim Comsochim Acta 71:2417–2435
Zahnle K, Claire M, Catling D (2006) The loss of mass-independent
fractionation in sulphur due to a Paleoproterozoic collapse of atmo-
spheric methane. Geobiology 4:271–283
11 8.1 The Great Oxidation Event 1533
Part IX
FAR-DEEP Core Archive: Future Opportunitiesfor Geoscience Research and Education
FAR-DEEP Core Archive: Further Opportunitiesfor Earth Science Research and Education
Victor A. Melezhik and A.R. Prave
9.1 Introduction
The three-volume treatise, Reading the Archive of Earth’s
Oxygenation, documents the scientific findings of the Inter-
national Continental Drilling Program’s (ICDP) FAR-DEEP
venture (Fennoscandian Arctic Russia – Drilling Early
Earth Project). The major outcome of the drilling project
was the successful recovery of a total of 3.5 km of pristine
cores that provide the international Earth science community
an exceptional opportunity to study Palaeoproterozoic Earth
history. The core archive contains a record of many of the
hallmark events of that time, such as the first worldwide
glaciation, the earliest and one of the largest positive isoto-
pic excursions of the global C cycle, and arguably the
world’s oldest supergiant oilfield, and offers the chance for
undertaking a wide variety of exciting research and educa-
tional activities.
The core is a window into the Palaeoproterozoic world as
seen through the geology of the Russian sector of the
Fennoscandian Shield. Combined, the treatise and core are
a one-stop-shop to either browse through or ponder deeply
this geology and its linkages to global events. In that many of
these events represent worldwide happenings, the core and
its archived data represent an efficient way to become famil-
iar with both that time period and a region in which an
outstanding Palaeoproterozoic rock record is preserved.
The core is readily accessible in its current stored location
with the Norwegian Geological Survey, Trondheim. Unique
rock samples can be obtained for use in testing ideas regard-
ing the nature of Palaeoproterozoic geological processes and
to calibrate isotope chemistry with rock composition. The
core is also an ideal learning aid for educators, researchers
and students interested in observing first-hand the rock types
that mark this period of Earth history.
The following pages show examples of the material and
information available for study. We also give a brief discus-
sion of the problems that yet remain in understanding how
Earth became an oxygen-rich planet. Future research will
determine whether or not Earth’s oxygenation was a unidi-
rectional or a stuttered, stepwise process and, ultimately,
enable constructing a self-consistent model of how this
happened. The FAR-DEEP core and database represent
valuable archives in this quest, ones that can be returned to
again and again over the years to investigate this profound
time period.
9.2 Educational Opportunities
The FAR-DEEP core and database offers an unprecedented
opportunity for geological training and education. The
3.5 km of cores represent palaeoenvironmental settings
ranging from lacustrine to deep marine, and from rifted-
margins to continental slopes. The cores contain exception-
ally well-preserved rocks including komatiitic lavas
(Fig. 9.1a), pillowed basalts (Fig. 9.1b), alkaline amygdaloi-
dal lavas (Fig. 9.1c), felsic lava breccias (Fig. 9.1d), Earth’s
first red beds (Fig. 9.1e–g), stromatolitic dolostones
(Fig. 9.1g), lacustrine travertines (Fig. 9.1h), abundant
sulphates (Fig. 9.1i–k), tidalites (Fig. 9.1l), glacigenic
rocks (Fig. 9.1m, n), organic-rich oil shales and
pyrobitumens of ancient oil seeps (Fig. 9.1o–v), and Earth’s
earliest phosphorites (Fig. 9.1w). All of these rocks can be
cross-referenced to detailed geochemical data (Appendices
1–43). Consequently, the FAR-DEEP core is a unique repos-
itory for students to observe and study first-hand such a
variety of rocks housed in a single, easily accessible loca-
tion. In addition, the core can be used as a means of teaching
students how to recognise and describe rock in core sections,
a valuable transferrable skill for many industry-related tasks
V.A. Melezhik (*)
Geological Survey of Norway, Postboks 6315 Sluppen, NO-7491
Trondheim, Norway
Centre for Geobiology, University of Bergen, Allegaten 41, Bergen
N-5007, Norway
e-mail: [email protected]
V.A. Melezhik et al. (eds.), Reading the Archive of Earth’s Oxygenation, Volume 3: Global Events and the Fennoscandian
Arctic Russia - Drilling Early Earth Project, DOI 10.1007/978-3-642-29670-3_12, # Springer-Verlag Berlin Heidelberg 2013
1537
involving the study of fresh rocks. The core provides
examples of rocks that have not been exposed to weathering
and degradation on Earth’s surface. This enables students to
compare and contrast surface outcrops to core and see how
they vary, both texturally and chemically, from one another.
Those educators and researchers who are unable to visit
the core can nevertheless also access this wealth of informa-
tion via the FAR-DEEP website. This web portal is the
gateway (http://www.icdp-online.org) to a vast collection
of core and outcrop photos and geochemical data, organised
by hole, core box and easy-to-follow spreadsheet arrays.
Further, it can be used as a virtual fieldtrip for educational
training. The high quality images, detailed core descriptions,
and the geochemical databases linked to individual cores and
rock types provide students with the next-best-thing to being
able to take an actual fieldtrip to Fennoscandia Russia.
Importantly, many of the rocks are iconic to the Palaeopro-
terozoic, rather than specific to a geographic locality. Thus,
examination of the core, either by actual visits or virtually,
enables students to learn how to recognise and identify
different rock types, as well as appreciate how rock
compositions, textures, and the data obtained from a variety
of analytical techniques, are used by Earth scientists to gain
insights into ancient geological processes and products.
The educational aspects associated with FAR-DEEP
cores could accompany research activities involving
educators, students and researchers. The well-preserved
core material can be successfully employed for targeting
several unresolved sedimentological, petrological and geo-
chemical problems associated with the emergence of the
aerobic Earth system. The core material enables defining
and undertaking research activities ranging from focused
master programmes and doctoral studies to large-scale mul-
tidisciplinary projects.
9.3 Further Opportunities for Earth ScienceResearch
The Palaeoproterozoic Earth system was marked by unprec-
edented global-scale tectonic and biogeochemical changes.
Earth’s surface environments underwent an irreversible
alteration in oxidation state, large continental landmasses
were established, and the biosphere likely became
dominated by photosystem-II autotrophs. As this treatise
reveals, much has become known about these events, yet
much remains to be understood. At the present time of this
writing, there are many unresolved problems that offer
opportunities for future research. Many of these have been
identified and presented in several topical chapters in the
current volume. We highlight a few of these below, includ-
ing those which have not been attended to, and stress that it
is the rocks that afford the ultimate test of models and
interpretations aimed at explaining how the modern aero-
bic Earth system evolved.
9.3.1 Geochronology
The adage, ‘no dates, no rates’, is apropos for any attempt to
define the timing, rates and durations of geological processes
and events. Consequently, the need to have precise and
accurate radiometric ages is a basic requirement for
constructing stratigraphic frameworks and correlations. Fur-
ther, assessing geochemical, geobiological and basin evolu-
tion models demand geochronological control, yet
frustratingly few robust ages exist to bracket the events of
the Palaeoproterozoic time slice.
The FAR-DEEP cores offer numerous radiometric dating
opportunities to obtain geochronological constraints and
help efforts to construct regional to global correlations,
assess the synchronicity or diachronicity of isotopic
excursions, the timing and duration of climatic events, the
timing and duration of volcanism, and the temporal (in)
completeness of the sedimentary record. Felsic to intermedi-
ate composition igneous bodies (Fig. 9.1c, d), from flows
and tuffs to intrusive rocks, as well as phosphate-bearing
intervals (Fig. 9.1w) are present in a number of the FAR-
DEEP cores and these could contain U-bearing minerals
(zircon, baddeleyite, phosphorites, xenotime) for U-Pb dat-
ing techniques. Organic- and sulphide-rich rocks in the
Zaonega Formation (Fig. 9.1o–w) and their equivalents
could be targeted for Re-Os methodologies (for details see
Chap. 7.10.7). The Zaonega Formation hosts a unique
petrified oil field (for details see Chap. 7.6) that includes
source rocks, reservoirs, vein-trapped migrated petroleum,
petrified oil, and subaqueous and surface oil seeps
(Fig. 9.1o–w). A long history of the oil field can be poten-
tially tracked by dating various organic-carbon-rich phases.
Trial projects on dating of organic matter and sedimentary/
diagenetic sulphides from the Pilguj€arvi and Zaonega
formations by the Re-Os technique proved successful, as
long as sampling was underpinned by detailed and robust
sedimentological, petrographic and geochemical work (e.g.
Hannah et al. 2006, 2008). The Re-Os dating is also essential
for providing time constraints for the enhanced global accu-
mulation of organic-rich rock during the Shunga Event, and
to test if this event was contemporaneous in different basins.
A (U-Th)/He series dating programme could be
attempted on haematite-bearing rocks (Fig. 9.1x) in order
to assess if the timing of the oxidising event was syn- or
post-depositional. The application of (U-Th)/He dating to
Fe-oxides is not new (e.g. Lippolt et al. 1995, 1998), but
long-term helium retention is a potential problem (Shuster
et al. 2005). The technique can be potentially tested on the
Kuetsj€arvi volcanic and Kolosjoki sedimentary rocks that
1538 V.A. Melezhik and A.R. Prave
represent a varied suite of igneous and volcaniclastic “red
beds” and haematite-rich lithofacies such as haematite-
cemented sandstones and jaspers.
9.3.2 Palaeogeography and Tectonics
The present-day longitudinal alignment of continents
permits the relatively free circulation of oceanic water
masses from tropical to polar regions thereby distributing
heat from low to high latitudes, which, in turn, influences
strongly climate. Further, it is readily apparent which conti-
nental margins are tectonically active or quiescent, and
which sedimentary basins are related to collisional, exten-
sional or strike-slip tectonics. In contrast, reconstructing
Palaeoproterozoic palaeogeography and tectonics is diffi-
cult: the size and number of continental plates is unknown,
the sizes of ocean basins are unknown, and the timing and
nature of the genesis of continental crust and subduction-
related volcanism is under lively debate. It is likely that we
will never be able to reconstruct completely, owing to the
incompleteness of the rock record and the poly-phase defor-
mation experienced by many ancient continental margins,
what the exact plate tectonic configurations were for the
Palaeoproterozoic Earth. However, the preserved remnants
of sedimentary basins afford one means of reconstructing, as
best we can, ancient plate tectonics.
The FAR-DEEP database and core can provide much of
the underpinning science, namely, the detailed sedimentol-
ogy, stratigraphy, igneous petrogenesis and palaeomagnetic
sampling, to construct tectonostratigraphic frameworks for
the Pechenga and Imandra/Varzuga belts and Onega basin.
Future work to obtain, for example, better geochronology
and more robust palaeomagnetic data, will refine deposi-
tional models and, hence, the evolution of the various sedi-
mentary basins through time. This will enable a better
understanding of the tectonic interactions between the Kola
and Karelian cratons, and their intervening oceans.
9.3.3 The Advent of the Progressive Oxidationof the Atmosphere
The Fennoscandian geological record documented in FAR-
DEEP core starts with c. 2442 Ma tidal sandstone-siltstone-
shale (Fig. 9.1l) and marine dolostones of the Seidorechka
Sedimentary Formation (see Chaps. 4.1 and 6.1.1). These
rocks accumulated during a transitional period in atmo-
sphere evolution, from largely anoxic to a state of its
incipient oxidation, as indicated by disappearance of mass-
independent fractionation of sulphur isotope (MIF-S)
reported from Canada and South Africa (Papineau et al.
2007; Guo et al. 2009). The obtained drillcore has great
potential to address the global sulphur and carbon cycles at
this dawn of progressive atmospheric oxidation and prior to
the first global Huronian-time glaciation(s) (see Chap. 7.1).
The continuous core through a c. 120-m-thick succession of
marine clastic and carbonate sedimentary rocks allows
obtaining high-resolution carbon and multiple sulphur iso-
tope measurements and a great opportunity to resolve the
internal structure of the termination of MIF-S (e.g. abrupt,
gradually, stuttered) and ultimately a quantitative under-
standing of the related environmental changes. This will
also contribute to more complete understanding of the global
carbon and sulphur cycles and seawater chemistry.
Comparison/contrasting the Fennoscandian isotopic
record with those reported from two other continents
(Canada and S. Africa) should reveal similarities or
discrepancies between them and thereby provide a more
robust understanding of the termination of MIF-S.
9.3.4 Palaeoclimate
Models that inform on the early evolution of atmospheric
composition and its role in regulating Earth’s surface tem-
perature require validation with the rock record. A good
case-in-point is the causes and consequences of the first
global glaciations, known collectively as the Huronian, and
the atmospheric-oceanic-lithospheric chemistry changes that
led to oxygen becoming freely available.
The Polisarka Sedimentary Formation core contains
uniquely a record of one of the Palaeoproterozoic glacial
episodes (Fig. 9.1n). Unlike anywhere else in the world, the
Scandinavian diamictic units intersected by FAR-DEEP
hole 3A (see Chap. 6.1.2) are encased in high-Sr, apparently
marine, bedded carbonate rocks and varves (Fig. 9.1m, y)
that can inform on oceanic chemistry and C-, Sr- and
S-isotopic composition. In turn these can be used as proxies
for atmospheric composition and a means to test and con-
strain models about Earth’s decent into the first global ice-
house. Ash beds associated with the glacial deposits have a
great potential for radiometric dating, thus providing time
constraints on the Fennoscandian glaciations.
9.3.5 Palaeobiology
An aspect that makes Earth unique is its oxygen-rich atmo-
sphere and determining why Earth was saved from sharing
the oxygen-starved fate of her sister planets is challenging.
The Palaeoproterozoic was the time when a tipping point
was reached on Earth such that the amount of oxygen pro-
duction (via oxygenic photosynthesising cyanobacteria)
outpaced the amount consumed (via oxygen sinks largely
in the form of reduced volcanic rocks and gases). Geologists,
12 FAR-DEEP Core Archive: Further Opportunities for Earth Science Research and Education 1539
biologists and geochemists are researching worldwide to
document the evidence of this event and ascertain what
type of microbial biota defined Earth’s earliest microbial
ecosystems.
The organic-rich rocks in the Zaonega Formation cores
(Fig. 9.1o, p) are prime candidates for microfossil and
molecular biomarker studies (see Chaps. 7.8.3 and 7.8.5).
The Kolosjoki Volcanic Formation affords the opportunity
to study the potential for microbiota existing in the rims of
pillow basalts (see Chap. 7.8.4). Further, there are numerous
stromatolitic occurrences (e.g. Fig. 9.1g) that require study
regarding their potential as stratigraphic tools (see Chap.
7.8.2). Hence, palaeobiological research into Palaeopro-
terozoic microbial ecologies and habitats await investigation
utilising the FAR-DEEP cores.
9.3.6 ‘Red Beds’
An oft-cited line-of-evidence for the timing of free oxygen is
the appearance in the geological record of the first ‘red
beds’. However, the environmental significance of red beds
varies significantly through Earth history. In Phanerozoic
successions that post-date the advent of land plants, ‘red
beds’ are commonly viewed as an indicator of aridity
because humid climates generate humus which can act as a
reductant, converting ferric to ferrous iron. In pre-
Phanerozoic successions, the absence of an extensive land
biota means that ‘red beds’ could potentially occur in any
climatic setting following the establishment of free oxygen,
hence they are non-climatic but still environmental
indicators reflecting emergence of an oxic atmosphere.
Another, non-climatic type of red bed is that formed in
association with post-orogenic, volcaniclastic molasse;
these are widespread throughout the Phanerozoic (e.g.
Turner 1980; Zharkov et al. 1998) and are also known in
Precambrian successions (Fig. 9.1z). Thus, assessing the
exact origin and hence geological significance of ‘red
beds’ is important in establishing the sinks and sources for
atmospheric oxygen. The ‘red beds’ of the Tulomozero,
Keutsj€arvi, and Kolosjoki formations represent sedimentary,
volcanic, and volcaniclastic units (Fig. 9.1e–g) and could
provide such information.
9.3.7 Palaeoproterozoic Seawater SulphateReservoir
Since the appearance of free oxygen in the atmosphere, the
sulphur cycle has been governed by oxidative weathering of
sulphides and microbial turnover of oceanic sulphate (for
details see Chap. 7.5). The currently available database on
d34S of Palaeoproterozoic sulphate remains limited and
shows a large range from +9.1 ‰ to +42.3 ‰ (VCTD)
(Schr€oder et al. 2008 and references therein; Guo et al.
2009; Krupenik et al. 2011; Reuschel et al. 2012). Such
large variation suggests strongly that the true isotopic com-
position of Palaeoproterozoic seawater sulphate is yet to be
established.
Two recently published reports on sulphate sulphur isoto-
pic composition from the Tulomozero successions also show
some discrepancy. Reuschel et al. (2012) provided an
account for the sulphur isotopic composition based on 9
ex situ analyses of carbonate-associated (CAS) and,
breccia-hosted sulphate (BHS), and 92 in situ analyses of
anhydrite and barite relicts in quartz-pseudomorphed Ca-
sulphate nodules. All show a rather narrow range in d34Sbetween +7.8 ‰ and +11.3 ‰ (one outlier is at +15.8 ‰)
over c. 500 m of stratigraphy. In contrast, Krupenik et al.
(2011) reported 14 analyses with lower values (+4.8 ‰ to
+5.9‰) that were obtained from a c. 400-m-thick succession
of interbedded massive anhydrite and magnesite (see Chap.
7.5). The succession occurs just beneath the units sampled
by Reuschel et al. (2012). Such d34S difference within the
same formation may suggest either an evolutionary trend in
seawater isotopic composition or involvement of late,
evolved, brines influencing the isotopic composition of the
CAS, BHS and anhydrite and barite relicts in quartz-
pseudomorphed sulphate concretions. Cores 10A, 10B
and 11A contain abundant relicts of sulphates preserved in
dolomite- and quartz-pseudomorphed chicken-wire and
enterolithic structures, concretions, crystals and rosettes
(Fig. 9.1i–k); these may hold a key to resolving the
intriguing and puzzling datasets.
9.3.8 Palaeoproterozoic Perturbation of theGlobal Carbon Cycle, the Lomagundi-JatuliEvent
The current understanding of the cause(s) for the c. 160-Ma-
long positive isotopic excursion of carbonate carbon (see for
details Chap. 7.3) is hampered by several shortcomings,
among them the following three are important: (1) the true
global d13C values reflecting isotopic composition of seawa-
ter; (2) local amplification factors; and (3) primary carbon
isotopic difference between coeval marine carbonates and
unaltered organic matter.
FAR-DEEP cores 5A, 10A, 10B and 11A together with
previously published data offer an opportunity to construct
two- or even three-dimensional isotopic models across the
Pechenga and Onega basins which may inform on basinal
d13C variations associated with local amplifying factors.
Providing that high-precision age constraints are obtainable,
the FAR-DEEP holes 4A, 5A, 10A, 10B and 11A have a
potential for comparison/contrast time-equivalent carbonate
1540 V.A. Melezhik and A.R. Prave
successions accumulated in lacustrine (Core 5A), open
marine (Core 4A) and restricted evaporitic (Cores 10A,
10B and 11A) environments (Fig. 9.1f–h, y, ac–af).
9.3.9 Enhanced Accumulation of OrganicCarbon, Petrified Supergiant Oil Field and theShunga Event
FAR-DEEP cores 12A, 12B and 13A hold information on
the enhanced accumulation of organic matter (the Shunga
event) and an associated petrified supergiant oil field
(Fig. 9.1o–v). Despite many years of research, many geolog-
ical, geochemical and petrological features of this unique
phenomenon remain understudied (see Chap 7.6). Coupling
carbon and sulphur isotopic systems, and involving paired
carbonate carbon – organic carbon isotopic studies on pri-
mary and diagenetic carbonate phases may reveal additional
and crucial information on evolution of primary producers of
organic matter and its recyclers. These FAR-DEEP cores
warrant a series of fascinating research projects.
9.3.10 Stable Isotope and Trace ElementGeochemistry
Stable isotopes such as C, O, S and Fe, concentrations of
redox-sensitive transition metals such as Re and Mo, and
variations in specific elemental ratios such as FeII and FeIII,
are yielding fascinating insights into palaeoenvironmental
conditions, biological evolution and ancient ecologies.
Excursions in their stratigraphic profiles inform on the
chemical composition of ancient oceans, atmospheres and
ecosystems. Such data also provide insight into the potential
for modification of original (i.e. depositional) isotopic and
geochemical signals due to local, basinal effects as well as
subsequent diagenetic and metamorphic overprinting.
FAR-DEEP cores contain abundant examples of carbon-
ate rocks, such as those of the Umba, Keutsj€arvi, Kolosjoki
and Tulomozero formations. These can be targeted to under-
take studies of Ca, Mg and B isotopes to assess the changing
chemistry of ocean waters, the inputs from, as well as the
intensity of, weathering, and the lithological makeup of
provenances. Study of the Ca and Mg isotope systematics
in Precambrian carbonate rocks is in its infancy (see Chap.
7.10.3). The carbonate formations intersected by FAR-
DEEP drillholes show a great variety of depositional settings
ranging from glacial, surface-hydrothermal, lacustrine,
sabkha to open marine (Figs. 9.1f–h, y, ac–af), and hence
offer an excellent opportunity to test the environmental
influence on various isotope systems.
FAR-DEEP cores obtained from the Keutsj€arvi Volcanic
and Kolosjoki Sedimentary formations contain several iron-
oxide phases such as magmatic and post-magmatic hydro-
thermal varieties (Fig. 9.1c, aa). In some instances these
were oxidised by surface and/or subsurface waters, then
eroded and transported into a non-marine basin in detrital
form (Fig. 9.1x, ag). In other cases, magmatic iron-oxides
were dissolved and transported by hydrothermal fluids from
which the iron was precipitated as haematite cement in
fluvial and coastal sandstones and as sea-floor jaspers
(Fig. 9.1ah, ai). Although the field of Fe-isotope geochemis-
try is relatively new (reviewed in Johnson et al. 2003; see
Chap. 7.10.4), experimental and theoretical studies show
that Fe isotopes fractionate in aqueous environments
between ferric and ferrous iron species due to both
biological and non-biological processes (e.g. Sharma et al.
2001). Hence, Fe-isotope studies could be employed to (1)
reveal isotopic differences in Fe-oxides, (2) investigate the
sources/origin of ferric iron, and (3) develop a model for the
mobilisation, precipitation, erosion and redeposition of iron
oxides in the Keutsj€arvi and Kolosjoki depositional systems.
Such data on the appropriate rock types may enable
testing and constraining models of atmospheric chemistry
versus mantle oxidation processes, as well as evaluate the
magnitude of oxygen sources and sinks during the course of
the progressive oxygenation of terrestrial environments dur-
ing the Palaeoproterozoic.
12 FAR-DEEP Core Archive: Further Opportunities for Earth Science Research and Education 1541
Fig. 9.1 Selected images of Palaeoproterozoic rocks documented in
some FAR-DEEP cores. Core diameter for scale is 5 cm unless other-
wise specified. (a) High-magnesium komatiitic basalt with a pyroxene-
spinifex texture; Polisarka Sedimentary Formation; Imandra/Varzuga
Belt, Core 3A. (b) High-magnesium tholeiitic basalt, Zaonega
Formation, Onega basin, Core 12B. (c) Microcrystalline, amygdaloidal,
trachydacitic lava with amygdales filled by quartz, albite and carbonate
(all white), chlorite (green) and jasper (red); Kuetsj€arvi Volcanic For-mation, Pechenga Belt, Core 6A. (d) Rhyodacitic lava breccia;
Kuetsj€arvi Volcanic Formation, Pechenga Belt, Core 6A
1542 V.A. Melezhik and A.R. Prave
Fig. 9.1 (continued) (e) A “red bed”: rhythmically interbedded pink
dolarenite and dark-coloured siltstone with parallel, and wavy, ripple-
cross lamination; Tulomozero Formation, Onega Basin, Core 10A. (f)
A “red bed”: rounded, pale pink, dolarenite clasts in a massive, pink
dolomarl matrix filling a large dissolution cavity; Tulomozero Forma-
tion, Onega Basin, Core 10A. (g) “Clumpy”, marine stromatolites with
divergent morphology and individual columns and beds are separated
by dark-coloured, haematite-rich, silt-sized material; Tulomozero For-
mation, Onega Basin, Core 10A. (h) Lacustrine dolorudite (at the base)
overlain by white massive and banded, dolomitic travertine crust
overlain by yellowish, dolomitic travertine with a clotted microfabric;
Kuetsj€arvi Sedimentary Formation, Pechenga Belt
12 FAR-DEEP Core Archive: Further Opportunities for Earth Science Research and Education 1543
Fig. 9.1 (continued) (i) Dark grey, indistinctly bedded, sabkha sand-
stone with large sulphate nodules partially replaced by pink quartz;
Tulomozero Formation, Onega Basin, Core 10B. (j) Grey-brown, non-
bedded dolomarl with sulphate crystals partially replaced by white
quartz and dolospar; Tulomozero Formation, Onega Basin, Core 10A.
(k) Dark brown, sabkha dolomarl with crystals of sulphates partially
replaced by white quartz and dolospar, Tulomozero Formation, Onega
Basin, Core 10A
1544 V.A. Melezhik and A.R. Prave
Fig. 9.1 (continued) (l) Flaser and wavy bedding in tidal siltstone-
shale of the Seidorechka Sedimentary Formation, Imandra/Varzuga
Belt, outcrop is nearby to Hole 1A. (m) Finely laminated (varved),
glacio-marine limestone (pale grey) and siltstone (dark grey) couplets;
Polisarka Sedimentary Formation, Imandra/Varzuga Belt, Core 3A. (n)
Diamictite composed of scattered andesite and dacite clasts set in a
massive clayey siltstone matrix (rock flower); Polisarka Sedimentary
Formation, Imandra/Varzuga Belt, Core 3A
12 FAR-DEEP Core Archive: Further Opportunities for Earth Science Research and Education 1545
Fig. 9.1 (continued) (o) Rhythmically bedded, sulphide- and Corg-rich
greywacke-shale; Zaonega Formation, Onega Basin, Core 13A. (p)
Dark-coloured, laminated, Corg-rich mudstone-shale with chert nodule;
Zaonega Formation, Onega Basin, Core 13A. (q) Back-scattered
electron image of pyrobitumen-rich (black) graded sandstone with
pyrobitumen accumulation (black, originally oil) in inter-layer space;
Zaonega Formation, Onega Basin, Core 12B
1546 V.A. Melezhik and A.R. Prave
Fig. 9.1 (continued) (r) Photomicrograph in reflected light showing
pyrobitumen-rich vein with wall-parallel banding in calcareous
greywacke; Zaonega Formation, Onega Basin, Core 12B. (s) Photomi-
crograph in reflected light showing fragment of quartz (pale grey)-pyrobitumen (bright) vein in calcareous greywacke; Zaonega Forma-
tion, Onega Basin, Core 12B. (t) Back-scattered electron image of
pyrobitumen “roses” (black) in calcite (bright) occurring in gabbro-
hosted veinlet; Zaonega Formation, Onega Basin, Core 12B. (u) Back-
scattered electron image of graphic intergrowth between pyrobitumen
(black) and calcite (bright) occurring in gabbro-hosted veinlet;
Zaonega Formation, Onega Basin, Core 12B
12 FAR-DEEP Core Archive: Further Opportunities for Earth Science Research and Education 1547
Fig. 9.1 (continued) (v) Soft-sediment deformed, laminated siltstone
and mudstone with two fragments of massive pyrobitumen (right side)
overlain by subaqueous oil seep represented by breccia that consists of
slumped and partially disintegrated massive pyrobitumen “clumps”
(grey) in black mudstone matrix; Zaonega Formation, Onega Basin,
Core 12B. (w) Back-scattered electron image of nodular phosphates in
Corg-rich mudstone; Zaonega Formation, Onega Basin, an outcrop at
the type locality at Shunga, in the vicinity of Hole 13A. (x) Thickly
bedded, volcaniclastic, coarse-grained sandstone with angular, platy
rip-up clasts of fine-grained haematite layers; Kolosjoki Sedimentary
Formation, Pechenga Belt, Core 8B
1548 V.A. Melezhik and A.R. Prave
Fig. 9.1 (continued) (y) Interbedded, finer- to thicker- bedded, glacio-
marine, limestone-shale couplets; Polisarka Sedimentary Formation,
Imandra/Varzuga Belt, Hole 3A. (z) Haematite-stained, fragment-
supported, volcaniclastic conglomerate with polymict clasts; Levi
Formation, Central Lapland Belt. (aa) Subaerialy erupted dacitic lava
breccia with fragments separated by black, haematite-magnetite-rich
(25 wt% Fetot) bands; Kuetsj€arvi Volcanic Formation, Pechenga Belt,
an outcrop in the vicinity of Hole 7A
12 FAR-DEEP Core Archive: Further Opportunities for Earth Science Research and Education 1549
Fig. 9.1 (continued) (ab) Fluvial, volcaniclastic conglomerate com-
posed of irregularly scattered fragments of dacite and andesite
supported by arkosic-gritstone matrix; Kolosjoki Sedimentary Forma-
tion, Pechenga Belt, Hole 7A. (ac) Pale pink, indistinctly bedded,
lacustrine dolostone with white, bedding-parallel, travertine veins and
abundant voids filled with travertine dolomite; Kuetsj€arvi Sedimentary
Formation, Pechenga Belt, Hole 5A. (ad) Soft-sediment deformed,
thin-bedded, variegated, shallow-marine, dolarenite overlying
dolostone with wavy, wrinkled and domed lamination (small cumulate
stromatolite); Tulomozero Formation, Onega Basin, Hole 10A. (ae)
White, sabkha dolarenite with hummocky bedding overlain by clayey
dolostone with red dolomite crystals, followed by dark-coloured marl
with crystal rosettes of sulphates partially replaced by white dolospar;
Tulomozero Formation, Onega Basin, Core 10A
1550 V.A. Melezhik and A.R. Prave
Fig. 9.1 (continued) (af) White, finely crystalline, parallel-bedded
deep-marine dolostone overlain by black shale; Umba Sedimentary
Formation, Imandra/Varzuga Belt, Hole 4A. (ai) Back-scattered elec-
tron image of fluvial-deltaic sandstone containing abundant haematite-
magnetite (right) clasts; Kolosjoki Sedimentary Formation, Pechenga
Belt, an outcrop in vicinity of Hole 8B. (ah) Back-scattered electron
image of fluvial-deltaic sandstone composed of rounded and sorted
clasts of quartz (dark coloured) and feldspar (grey) cemented by
haematite (bright); Kolosjoki Sedimentary Formation, Pechenga Belt,
Core 8B. (ai) Thin jasper layers rhythmically interbedded with dark-
coloured, marine siltstone; Kolosjoki Sedimentary Formation,
Pechenga Belt, Core 8B (Photographs (a–v, x, y, aa–ai) by Victor
Melezhik, photograph (w) courtesy of Aivo Lepland, sample (z) cour-
tesy of the Geological Museum of the Department of Geosciences,
University of Oulu, photograph by Eero Hanski)
12 FAR-DEEP Core Archive: Further Opportunities for Earth Science Research and Education 1551
References
Guo Q, Strauss H, Kaufman AJ, Schr€oder S, Gutzmer J, Wing B, Baker
MA, Bekker A, Jin Q, Kim S-T, Farquhar J (2009) Reconstructing
Earth’s surface oxidation across the Archean-Proterozoic transition.
Geology 37:399–402
Hannah JL, Stein HJ, Zimmerman A, Yang G, Markey RJ, Melezhik
VA (2006) Precise 2004 � 9 Ma Re-Os age for Pechenga black
shale: comparison of sulfides and organic material. Geochim
Cosmochim Acta 70(Supplement):A–228
Hannah JL, Stein HJ Zimmerman A, Yang G, Melezhik VA, Filippov
MM, Turgeon SC, Creaser RA (2008) Re-Os geochronology of a
2.05 Ga fossil oil field near Shunga, Karelia, NW Russia. In:
Abstracts of the 33rd international geological congress, Oslo,
6–14 Aug 2008
Johnson CM, Beard BL, Beukes NJ, Klein C, O’Leary JM (2003)
Ancient geochemical cycling in the Earth as inferred from Fe
isotope studies of banded iron formations from the Transvaal Cra-
ton. Contrib Mineral Petrol 144:523–547
Krupenik VA, Akhmedov AM, Sveshnikova KYu (2011) Carbon,
oxygen and sulphur isotopic composition of rocks from Ludicoviian
and Jatulian Super-Horizons. In: Glushanin LV, Sharov NV,
Shchiptsov VV (eds) The onega Palaeoproterozoic structure (Geol-
ogy, tectonics, deep structure and minerageny). Karelian Science
Centre, Petrozavodsk, pp 250–255 (in Russian)
Lippolt HJ, Wernicke RS, B€ahr R (1995) Paragenetic specularite and
adularia (Elba, Italy): concordant (U+Th)-He and K-Ar ages. Earth
Planet Sci Lett 132:43–51
Lippolt HJ, Brander T, Mankopf NR (1998) An attempt to determine
formation ages of goethites and limonites by (U+Th)-4He dating.
Neues Jahrb Miner Monatsh 11:505–528
Papineau D, Mojzsis SJ, Schmitt AK (2007) Multiple sulphur isotopes
from Paleoproterozoic Huronian interglacial sediments and the rise
of atmospheric oxygen. Earth Planet Sci Lett 255:188–212
Reuschel M, Melezhik VA, Lepland A, Fallick AE, Strauss H (2012)
Isotopic evidence for a sizeable seawater sulfate reservoir at 2.1 Ga.
Precambrian Res 192:78–88
Schr€oder S, Bekker A, Beukes NJ, Strauss H, van Niekerk HS (2008)
Rise in seawater sulphate concentration associated with the
Paleoproterozoic positive carbon isotope excursion: evidence from
sulphate evaporites in the 2.2–2.1 Gyr shallow-marine Lucknow
Formation, South Africa. Terra Nova 20:108–117
Sharma M, Polizzotto M, Anbar AD (2001) Iron isotopes in hot springs
along the Juan de Fuca Ridge. Earth Planet Sci Lett 194:39–51
Shuster DL, Vasconcelos PM, Heim JA, Farley KA (2005) Weathering
geochronology by (U-Th)/He dating of goethite. Geochim
Cosmochim Acta 69:659–673
Turner P (1980) Continental red beds, vol 29, Developments in Sedi-
mentology. Elsevier, Amsterdam, p 562
Zharkov MA, Murdmaa IO, Filatova NI (1998) Paleogeography of the
Berriasian-Barremian ages stratigraphy and geological correlation.
Stratigr Geo Correl 6:47–69
1552 V.A. Melezhik and A.R. Prave