Reading the Archive of Earthâ€™s Oxygenation: Volume 3: Global Events and the Fennoscandian Arctic Russia - Drilling Early Earth Project

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


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,


Pavel V. Medvedev Institute of Geology, Karelian Research Centre, Russian Academy of


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


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


4.3 The Onega Basin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387

V.A. Melezhik, P.V. Medvedev, and S.A. Svetov

xvii

.


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,


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,


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



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



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


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

mailto:[email protected]

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)


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


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)


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)


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)


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)


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


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.


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



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




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


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)


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


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


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)


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




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


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


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)


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)


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).


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


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.


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)


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)


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


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


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


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


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.


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


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




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


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


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))


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)


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


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.


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 (*)


Trondheim, Norway




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)


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 (*)


Trondheim, Norway


N-5007, Norway




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)


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))


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)


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.


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




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


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.


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 (*)


Trondheim, Norway


N-5007, Norway




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


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)


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


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


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





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)


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



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


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.


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-


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,



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,



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


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



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


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.


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


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.


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-


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,


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


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,





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.


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 (*)


Trondheim, Norway



3 7.3 The Palaeoproterozoic Perturbation of the Global Carbon Cycle 1113



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.


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




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.


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 (*)


Trondheim, Norway






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


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:


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)


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,


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


(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.


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 (*)


Trondheim, Norway





1124


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)


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


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)


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

––

–


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)


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.(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).


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


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))


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


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


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


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)


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)


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


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


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))


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)


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


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)


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)


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)


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


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.


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






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


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.


Hanski EJ, Huhma H, Smol’kin VF, Vaasjoki M (1990) The age of

ferropicritic volcanites and comagmatic Ni-bearing intrusions at


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 (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


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.


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.


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


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)


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




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.


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


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


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


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.


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)


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)


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


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)


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)


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


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)


References

Albarede F, van der Hilst RD (1999) New mantle convection model

may reconcile conflicting evidence. EOS Trans Am Geophys Union

80:535–539





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


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


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


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


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


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





precision ID-TIMS, AGU, San Francisco, 13–17 Dec 2010




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


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


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 (*)


Universit€at, Corrensstr. 24, 48149 M€unster, Germany




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 (*)





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.


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 (*)





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


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.


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 (*)





1176


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 (*)


Trondheim, Norway


N-5007, Norway



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.


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 (*)





1179


Fig. 7.51 The sulphur isotopic composition of Precambrian sedimentary sulphate and sulphide (After Thomazo et al. 2009)


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)


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


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)


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


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


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)


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)


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)


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)


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


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)


References



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-



Res 120:279–325




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,


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 (2011) On Proterozoic


associated with banded iron formations. In: Neves L et al (eds)

Modelacao de Sistemas. Geologicos, Universidade de Coimbra,

Portugal, pp 57–71



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)




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


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


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


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


and geological setting of the Palaeoproterozoic Magondi Super-



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,



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



442:908–911




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.


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


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)


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


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


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


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 (*)


Universit€at M€unster, Corrensstr. 24, 48149 M€unster, Germany




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 (*)


Trondheim, Norway



1196


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 (*)





1198


(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


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


(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.


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 (*)


Trondheim, Norway


N-5007, Norway



1202


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:


• 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.


http://www.britannica.com

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


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).


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


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


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


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


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


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.


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


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.


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 (*)


Trondheim, Norway


N-5007, Norway



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.


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)


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


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.


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


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)


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)


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)


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)


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))


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


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


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


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


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)


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)


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


http://far-deep.icdp-online.org

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)


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


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


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


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


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


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


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)


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)


http://oilandgasgeology.com

http://oilandgasgeology.com

http://www.CliffsNotes.com

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


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


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


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


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)


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


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


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)


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)


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)


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


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


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)


Fig. 7.75 Geological map of the Maksovo deposit, with cross- and longitudinal profiles through the maksovite lens (Modified after Kupryakov

1994)


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


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




Fig.7.79

Drillhole-based

sectionsshowinglateralandverticallithological

variationoftheZaonegaForm

ationandthree-dim

ensional

view

ofthemaksovitebodyintersectedbyFAR-D

EEP

Hole

12A

and12B(Compiled

byY.Deines

andV.Melezhik)


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


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


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


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


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


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)


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)


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)


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)


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


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


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))


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-



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


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


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




Springer Science+Business Media, New York, pp 169–188

Fallick AE, Melezhik VA, Simonson B (2011) On Proterozoic


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


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



Russia. Bull Geol Surv Finl 367:1–192

Hanski E, Huhma H (2005) Central Lapland greenstone belt. In:


Finland – key to the evolution of the Fennoscandian Shield.


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


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


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



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


http://www.CliffsNotes.com

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)


and geological setting of the Palaeoproterozoic Magondi Super-



McKirdy DM (1974) Organic geochemistry in Precambrian research.


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


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-



Melezhik VA, Fallick AE, Medvedev PV, Makarikhin VV (1999b)


dolomite-“ red beds” association in a global context: a case for the


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,




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:


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



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,


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,


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.


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.



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 (*)


Trondheim, Norway




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)


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


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


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.


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),



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)


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)


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)


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


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)


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


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)


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


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)


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)


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,


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-



Res 120:279–325




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


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.


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,


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




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


VA (2006) Precise 2004 � 9 Ma Re Os age for Pechenga black





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:


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





precision ID-TIMS, AGU meeting, 13–17 Dec 2010

McArthur JM (1985) Francolite geochemistry; compositional controls

during formation, diagenesis, metamorphism and weathering.



basins of the Baltic Shield. Nauka (Science), Leningrad, p 256 (in

Russian)




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)


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-



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, 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


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)


Trondheim, Norway




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,


1298 N. McLoughlin et al.



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


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;


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.


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


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.


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


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


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.


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.


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,


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


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


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.


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)


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



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)


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)


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)


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)


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))


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



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)


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))


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


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


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


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)


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)


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


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


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)


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


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)


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


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)


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


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


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)


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


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



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)


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


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)


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)


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


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)


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


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:


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.


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


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.


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.


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


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),


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.


Woodhead JD, Hergt JM (1997) Application of ‘double

spike’ technique to Pb-isotope geochronology. Chem Geol

138:311–321


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.



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


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


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


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).


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)


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


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


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)


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


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)


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


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)


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


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


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.


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)


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.


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


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.


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.


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


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


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,





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


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).


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


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


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


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.


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.


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)


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))

◂


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)

◂


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


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))


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



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


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)


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


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


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


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


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?


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.


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.


Lepot K, Philippot P, Benzerara K, Wang Y (2009)

Garnet-filled trails associated with carbonaceous matter

mimicking microbial filaments in Archean basalt.


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


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?


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


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


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


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




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


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


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).


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)


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)


References


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


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


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.


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


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.


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.


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


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


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.


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.


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


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




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



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


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))


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)


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


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


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,


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)


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


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.


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.


Tosca NJ, Johnson DT, Mushegian A, Rothman DH,

Summons RE, Knoll AH (2010) Clay mineralogy, organic

carbon burial, and redox evolution in Proterozoic oceans.


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


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


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





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


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


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


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)


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).


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).


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)


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))


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.


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



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)


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)


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


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-


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.



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


exposure surfaces in a Palaeoproterozoic 13C-rich dolostone




133:75–103


AB (2005) Palaeoproterozoic evaporites in Fennoscandia:



Nova 17:141–148



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


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





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


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


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


(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


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


(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


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


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)


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)


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)


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)


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.


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.


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.


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.


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)


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)


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



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,


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.


Garnett ER, Andrews JE, Preece RC, Dennis PF (2004)

Climatic change recorded by stable isotopes and trace


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.


Guo L, Riding R (1998) Hot-spring travertine facies and

sequences, Late Pleistocene, Rapolano Terme, Italy. Sedi-


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


exposure surfaces in a Palaeoproterozoic 13C-rich dolostone




133:75–103




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.



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-


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


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


7.10 Chemical Characteristics of Sedimentsand Seawater


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




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





Bekker A, Holland HD, Wang PL, Rumble D III, Stein



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


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


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


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



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



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)


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


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


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.


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.


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


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.


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


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



NN, Fallick AE (2003) Sr isotope composition in Paleopro-

terozoic carbonates extremely enriched in 13C: Kaniapiskau

Supergroup, the Labrador trough of the Canadian shield.


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


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




implications for seawater sulphate, d13C excursions and the

rise of atmospheric oxygen. Terra Nova 17:141–148




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


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


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.


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


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


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


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


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.



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


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


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.


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)


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)


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


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


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)


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)


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)


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


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



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.


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.


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:


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


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


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





Nova 17:141–148


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


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


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


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




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


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


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))


References

Anbar A (2003) Iron stable isotopes: beyond biosignatures.


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





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.


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


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


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



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, 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, 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)


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


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


Johnson CM, Beard BL, Klein C, Beukes NJ, Roden EE

(2008a) Iron isotopes constrain biologic and abiologic

processes in banded iron formation genesis. Geochim


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



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


perspective and Proterozoic case study. Annu Rev Earth


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


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.


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



Tomkins AG (2010) Windows of metamorphic sulfur

liberation in the crust: implications for gold deposit genesis.


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


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




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)


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)


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


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





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


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



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


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


Archean-Proterozoic transition. Geology 37:399–402


HD (2004) Primitive Os and 2316 Ma age for marine



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



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


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





Nova 17:141–148




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.


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,


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


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


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.



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))


(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))


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))


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,


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





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


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, 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


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.


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


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



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.


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


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.


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


Stirling CH, Andersen MB, Potter EK, Halliday AN

(2007) Low-temperature isotopic fractionation of uranium.


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


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.


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


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



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


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)


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


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






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



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


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



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




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



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.


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


Ravizza G, Turekian KK (1989) Application of the187Re-187Os system to black shale geochronometry.


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


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


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.


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


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




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


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


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


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.


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)


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))


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)


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)


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


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))


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)


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 (1989) Evidence from Lewisian limestones for

isotopically heavy carbon in two thousand million year old sea

water. Nature 337:352–354



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






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.


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




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



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-


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



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


VA (2006) Precise 2004 � 9 Ma Re‐Os age for Pechenga black





Russia. Geol Surv Finland Bull 367:1–192


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


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


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



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,


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


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


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)


dolomite-‘red beds’ association in a global context: a case for the


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




Morozov AF, Hahaev BN, Petrov OV, Gorbachev VI, Tarkhanov GB,




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



442:908–911




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



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)


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



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-


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


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 (*)


Trondheim, Norway


N-5007, Norway




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

http://www.icdp-online.org

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


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.


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


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


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


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


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


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


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


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


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


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)


References




Geology 37:399–402


VA (2006) Precise 2004 � 9 Ma Re-Os age for Pechenga black


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


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




Reuschel M, Melezhik VA, Lepland A, Fallick AE, Strauss H (2012)

Isotopic evidence for a sizeable seawater sulfate reservoir at 2.1 Ga.



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


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


Documents

Reading the Archive of Earthâ€™s Oxygenation: Volume 3: Global Events and the Fennoscandian Arctic Russia - Drilling Early Earth Project