Geophysical Monograph Series · Figures, tables and short excerpts may be reprinted in scientific books and journals if the source is properly cited. Authorization to photocopy items

Geophysical Monograph Series

IncludingIUGG Volumes

Maurice Ewing VolumesMineral Physics Volumes

157 Seismic Earth: Array Analysis of Broadband Seismograms Alan Levander and Guust Nolet (Eds.)

158 The Nordic Seas: An Integrated Perspective Helge Drange, Trond Dokken, Tore Furevik, Rüdiger Gerdes, and Wolfgang Berger (Eds.)

159 Inner Magnetosphere Interactions: New Perspectives From Imaging James Burch, Michael Schulz, and Harlan Spence (Eds.)

160 Earth’s Deep Mantle: Structure, Composition, and Evolution Robert D. van der Hilst, Jay D. Bass, Jan Matas, and Jeannot Trampert (Eds.)

161 Circulation in the Gulf of Mexico: Observations and Models Wilton Sturges and Alexis Lugo-Fernandez (Eds.)

162 Dynamics of Fluids and Transport Through Fractured Rock Boris Faybishenko, Paul A. Witherspoon, and John Gale (Eds.)

163 Remote Sensing of Northern Hydrology: Measuring Environmental Change Claude R. Duguay and Alain Pietroniro (Eds.)

164 Archean Geodynamics and Environments Keith Benn, Jean-Claude Mareschal, and Kent C. Condie (Eds.)

165 Solar Eruptions and Energetic Particles Natchimuthukonar Gopalswamy, Richard Mewaldt, and Jarmo Torsti (Eds.)

166 Back-Arc Spreading Systems: Geological, Biological, Chemical, and Physical Interactions David M. Christie, Charles Fisher, Sang-Mook Lee, and Sharon Givens (Eds.)

167 Recurrent Magnetic Storms: Corotating Solar Wind Streams Bruce Tsurutani, Robert McPherron, Walter Gonzalez, Gang Lu, José H. A. Sobral, and Natchimuthukonar Gopalswamy (Eds.)

168 Earth’s Deep Water Cycle Steven D. Jacobsen and Suzan van der Lee (Eds.)

169 Magnetospheric ULF Waves: Synthesis and New Directions Kazue Takahashi, Peter J. Chi, Richard E. Denton, and Robert L. Lysal (Eds.)

170 Earthquakes: Radiated Energy and the Physics of Faulting Rachel Abercrombie, Art McGarr, Hiroo Kanamori, and Giulio Di Toro (Eds.)

171 Subsurface Hydrology: Data Integration for Properties and Processes David W. Hyndman, Frederick D. Day-Lewis, and Kamini Singha (Eds.)

172 Volcanism and Subduction: The Kamchatka Region John Eichelberger, Evgenii Gordeev, Minoru Kasahara, Pavel Izbekov, and Johnathan Lees (Eds.)

173 Ocean Circulation: Mechanisms and Impacts—Past and Future Changes of Meridional Overturning Andreas Schmittner, John C. H. Chiang, and Sidney R. Hemming (Eds.)

174 Post-Perovskite: The Last Mantle Phase Transition Kei Hirose, John Brodholt, Thorne Lay, and David Yuen (Eds.)

Geophysical Monograph Series

175 A Continental Plate Boundary: Tectonics at South Island, New Zealand David Okaya, Tim Stem, and Fred Davey (Eds.)

176 Exploring Venus as a Terrestrial Planet Larry W. Esposito, Ellen R. Stofan, and Thomas E. Cravens (Eds.)

177 Ocean Modeling in an Eddying Regime Matthew Hecht and Hiroyasu Hasumi (Eds.)

178 Magma to Microbe: Modeling Hydrothermal Processes at Oceanic Spreading Centers Robert P. Lowell, Jeffrey S. Seewald, Anna Metaxas, and Michael R. Perfit (Eds.)

179 Active Tectonics and Seismic Potential of Alaska Jeffrey T. Freymueller, Peter J. Haeussler, Robert L. Wesson, and Göran Ekström (Eds.)

180 Arctic Sea Ice Decline: Observations, Projections, Mechanisms, and Implications Eric T. DeWeaver, Cecilia M. Bitz, and L.-Bruno Tremblay (Eds.)

181 Midlatitude Ionospheric Dynamics and Disturbances Paul M. Kintner, Jr., Anthea J. Coster, Tim Fuller-Rowell, Anthony J. Mannucci, Michael Mendillo, and Roderick Heelis (Eds.)

182 The Stromboli Volcano: An Integrated Study of the 2002–2003 Eruption Sonia Calvari, Salvatore Inguaggiato, Giuseppe Puglisi, Maurizio Ripepe, and Mauro Rosi (Eds.)

183 Carbon Sequestration and Its Role in the Global Carbon Cycle Brian J. McPherson and Eric T. Sundquist (Eds.)

184 Carbon Cycling in Northern Peatlands Andrew J. Baird, Lisa R. Belyea, Xavier Comas, A. S. Reeve, and Lee D. Slater (Eds.)

185 Indian Ocean Biogeochemical Processes and Ecological Variability Jerry D. Wiggert, Raleigh R. Hood, S. Wajih A. Naqvi, Kenneth H. Brink, and Sharon L. Smith (Eds.)

186 Amazonia and Global Change Michael Keller, Mercedes Bustamante, John Gash, and Pedro Silva Dias (Eds.)

187 Surface Ocean–Lower Atmosphere Processes Corinne Le Quèrè and Eric S. Saltzman (Eds.)

188 Diversity of Hydrothermal Systems on Slow Spreading Ocean Ridges Peter A. Rona, Colin W. Devey, Jérôme Dyment, and Bramley J. Murton (Eds.)

189 Climate Dynamics: Why Does Climate Vary? De-Zheng Sun and Frank Bryan (Eds.)

190 The Stratosphere: Dynamics, Transport, and Chemistry L. M. Polvani, A. H. Sobel, and D. W. Waugh (Eds.)

191 Rainfall: State of the Science Firat Y. Testik and Mekonnen Gebremichael (Eds.)

Geophysical Monograph 192

Antarctic SubglacialAquatic Environments

Martin J. SiegertMahlon C. Kennicutt IIRobert A. Bindschadler

Editors

American Geophysical UnionWashington, DC

Published under the aegis of the AGU Books Board

Kenneth R. Minschwaner, Chair; Gray E. Bebout, Kenneth H. Brink, Jiasong Fang, Ralf R. Haese, Yonggang Liu, W. Berry Lyons, Laurent Montési, Nancy N. Rabalais, Todd C. Rasmussen, A. Surjalal Sharma, David E. Siskind, Rigobert Tibi, and Peter E. van Keken, members.

Library of Congress Cataloging-in-Publication Data

Antarctic subglacial aquatic environments / Martin J. Siegert, Mahlon C. Kennicutt II, Robert A. Bindschadler, editors. p. cm. — (Geophysical monograph ; 192) Includes bibliographical references and index. ISBN 978-0-87590-482-5 (alk. paper) 1. Subglacial lakes—Antarctica—History. 2. Subglacial lakes—Antarctica—Discovery and exploration. 3. Subglacial lakes— Polar regions—History. 4. Subglacial lakes—Polar regions—Discovery and exploration. 5. Aquatic ecology—Antarctica. I. Siegert, Martin J. II. Kennicutt, Mahlon C. III. Bindschadler, R. A. (Robert A.) GC461.A596 2011 551.48'2—dc22 2011007605

ISBN: 978-0-87590-482-5 ISSN: 0065-8448

Cover Image: The location (red triangles) of 387 subglacial lakes superimposed on the BEDMAP database depiction of Antarctic sub-ice topography. (top) The ice sheet surface is illustrated, which is used along with basal topography to predict (bottom) hydrological pathways (blue lines). Image credit: Andrew Wright and Martin Siegert.

Copyright 2011 by the American Geophysical Union2000 Florida Avenue, N.W.Washington, DC 20009

Figures, tables and short excerpts may be reprinted in scientific books and journals if the source is properly cited.

Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by the American Geophysical Union for libraries and other users registered with the Copyright Clearance Center (CCC) Transactional Reporting Service, provided that the base fee of $1.50 per copy plus $0.35 per page is paid directly to CCC, 222 Rosewood Dr., Danvers, MA 01923. 0065-8448/11/$01.50+0.35.

This consent does not extend to other kinds of copying, such as copying for creating new collective works or for resale.The reproduction of multiple copies and the use of full articles or the use of extracts, including figures and tables, for commercial purposes requires permission from the American Geophysical Union. geopress is an imprint of the American Geophysical Union.

Printed in the United States of America.

CONTENTS

PrefaceMartin J. Siegert, Mahlon C. Kennicutt II, and Robert A. Bindschadler ..................................................................vii

Section I: History and Background

Subglacial Aquatic Environments: A Focus of 21st Century Antarctic ScienceMahlon C. Kennicutt II and Martin J. Siegert ...........................................................................................................1

The Identification and Physiographical Setting of Antarctic Subglacial Lakes: An Update Based on Recent DiscoveriesAndrew Wright and Martin J. Siegert .......................................................................................................................9

Antarctic Subglacial Lake DischargesFrank Pattyn .........................................................................................................................................................27

Section II: Vostok Subglacial Lake and Recognition of Subglacial Aquatic Environments

Vostok Subglacial Lake: A Review of Geophysical Data Regarding Its Discovery and Topographic SettingMartin J. Siegert, Sergey Popov, and Michael Studinger ........................................................................................45

Microbial Communities in Antarctic Subglacial Aquatic EnvironmentsMark Skidmore .....................................................................................................................................................61

Subglacial Lake Sediments and Sedimentary Processes: Potential Archives of Ice Sheet Evolution, Past Environmental Change, and the Presence of LifeM. J. Bentley, P. Christoffersen, D. A. Hodgson, A. M. Smith, S. Tulaczyk, and A. M. Le Brocq ............................83

The Geomorphic Signature of Massive Subglacial Floods in Victoria Land, AntarcticaDavid R. Marchant, Stewart S. R. Jamieson, and David E. Sugden ...................................................................... 111

Subglacial Environments and the Search for Life Beyond the EarthCharles S. Cockell, Elizabeth Bagshaw, Matt Balme, Peter Doran, Christopher P. McKay, Katarina Miljkovic, David Pearce, Martin J. Siegert, Martyn Tranter, Mary Voytek, and Jemma Wadham ..........................................129

Section III: Future Exploration Missions

Environmental Protection and Stewardship of Subglacial Aquatic EnvironmentsPeter T. Doran and Warwick F. Vincent ..............................................................................................................149

Probe Technology for the Direct Measurement and Sampling of Ellsworth Subglacial LakeMatthew C. Mowlem, Maria-Nefeli Tsaloglou, Edward M. Waugh, Cedric F. A. Floquet, Kevin Saw, Lee Fowler, Robin Brown, David Pearce, James B. Wyatt, Alexander D. Beaton, Mario P. Brito, Dominic A. Hodgson, Gwyn Griffiths, M. Bentley, D. Blake, L. Capper, R. Clarke, C. Cockell, H. Corr, W. Harris, C. Hill, R. Hindmarsh, E. King, H. Lamb, B. Maher, K. Makinson, J. Parnell, J. Priscu, A. Rivera, N. Ross, M. J. Siegert, A. Smith, A. Tait, M. Tranter, J. Wadham, B. Whalley, and J. Woodward ............................................................159

Vostok Subglacial Lake: Details of Russian Plans/Activities for Drilling and SamplingValery Lukin and Sergey Bulat .............................................................................................................................187

Siple Coast Subglacial Aquatic Environments: The Whillans Ice Stream Subglacial Access Research Drilling Project Helen Amanda Fricker, Ross Powell, John Priscu, Slawek Tulaczyk, Sridhar Anandakrishnan, Brent Christner, Andrew T. Fisher, David Holland, Huw Horgan, Robert Jacobel, Jill Mikucki, Andrew Mitchell, Reed Scherer, and Jeff Severinghaus .......................................................................................199

Ellsworth Subglacial Lake, West Antarctica: A Review of Its History and Recent Field CampaignsN. Ross, M. J. Siegert, A. Rivera, M. J. Bentley, D. Blake, L. Capper, R. Clarke, C. S. Cockell, H. F. J. Corr, W. Harris, C. Hill, R. C. A. Hindmarsh, D. A. Hodgson, E. C. King, H. Lamb, B. Maher, K. Makinson, M. Mowlem, J. Parnell, D. A. Pearce, J. Priscu, A. M. Smith, A. Tait, M. Tranter, J. L. Wadham, W. B. Whalley, and J. Woodward ..................................................................221

AGU Category Index ..........................................................................................................................................235

Index ..................................................................................................................................................................237

PREFACE

Between 15 and 17 March 2010, 83 scientists from 11nations gathered in Baltimore, Maryland, United States, foran AGU Chapman Conference to discuss the current status ofknowledge about, and future exploration plans for, Antarcticsubglacial aquatic environments. This was the fifth in a seriesof international conferences. In response to recent fundingof two new major exploration programs and the continuation ofwork at Vostok Subglacial Lake, this meeting focused at-tention on emerging scientific frontier and the attendantenvironmental stewardship issues. The chapters of this bookexpand on keynote presentations and are augmented byselected invited authors to produce the first comprehensivesummary of research on, and planning for, the explorationof subglacial aquatic environments. The chapters includesummaries of the most recent identification, location, andphysiography of 387 subglacial lakes; a detailed analysis ofthe results, from years of study, from Vostok SubglacialLake; the rationale for subglacial lakes as analogues for

extraterrestrial environments; protocols for the protection andstewardship of these unique environments; critiques of thetechnological issues facing future exploration programs;and, finally, summaries of the three projects that will enterand sample subglacial aquatic environments in the next 3to 5 years. This book serves as a benchmark in subglacialaquatic environmental research, marking the beginnings of themain phase of exploration for this new frontier in Antarcticscience.

Martin J. SiegertUniversity of Edinburgh

Mahlon C. Kennicutt IITexas A&M University

Robert A. BindschadlerNASA Goddard Space Flight Center

Antarctic Subglacial Aquatic EnvironmentsGeophysical Monograph Series 192Copyright 2011 by the American Geophysical Union10.1029/2010GM001005

vii

Subglacial Aquatic Environments: A Focus of 21st Century Antarctic Science

Mahlon C. Kennicutt II

Department of Oceanography, Texas A&M University, College Station, Texas, USA

Martin J. Siegert

School of GeoSciences, University of Edinburgh, Edinburgh, UK

In 1996, growing evidence suggested a massive lake of liquid water had pooledbeneath the East Antarctic Ice Sheet. This feature became known as “Lake Vostok.”Early on, two hypotheses were posed: the lake contained microbial life that hadevolved over millions of years in isolation beneath the ice and lake sedimentscontained records of past climate change obtainable nowhere else in Antarctica.Manysubglacial lakes, in a number of locales, have been identified, suggesting that studiesat multiple locations will be needed to fully understand the importance of subglacialaquatic environments. As of 2010, more than 300 lakes have been identified; this willincrease as surveys improve spatial coverage. Given the likely pristine nature of theseenvirons and the low levels of microbial life expected, exploration must be done in amanner that causes minimal impact or contamination. It has been shown that many ofthese lakes are part of an active, sub-ice hydrological system that experiences rapidwater flow events over time frames of months, weeks, and even days. Microbial life insubglacial environments has been inferred, and is expected, but it has yet to be directlyconfirmed by in situ sampling. Current understanding of subglacial environments isincomplete and will only be improved when these subglacial environments areentered and sampled, which is projected to occur in the next few years. This booksynthesizes current understanding of subglacial environments and the plans for theirexploration as a benchmark for future discoveries.

1. INTRODUCTION

The recent study and exploration of subglacial aquaticenvironments has transformed our understanding of processesoperating at the Antarctic ice sheet bed and the role they haveplayed in the evolution of the continental ice mass. In 1996,an article in Nature reported that a giant lake existed beneath~4 km of ice in East Antarctica [Kapitsa et al., 1996]. The

lake was large enough to be visible in satellite altimetry dataof the ice sheet surface [Ridley et al., 1993]. The Naturearticle marked the beginning of modern subglacial aquaticenvironment research. Even though the original data hadbeen collected in the 1960s and 1970s, these features hadgone largely unnoticed by the broader scientific communityfor more than two decades [e.g., Oswald and Robin, 1973;Robin et al., 1977]. While the existence of liquid bodies ofwater beneath ice sheets was in itself of interest to glaciolo-gists, attention swiftly turned to whether these environmentsharbored unusual life forms. Some also conjectured that ifsediments were preserved in the lake they would containotherwise unavailable, valuable records of ice and climatechange in the interior of Antarctica. Kapitsa et al. [1996] had


1

identified Vostok Subglacial Lake buried beneath ∼4 km ofice. Not only was the lake’s area an order of magnitudegreater than any other subglacial lake known at the time,Kapitsa et al. [1996] also discoverd that the lake contained awater column at least 510 m deep, with an estimated volumeof 1800 km3.An international community of scientists became con-

vinced that subglacial lakes represented a new frontier inAntarctic research. Within a decade, this community devel-oped the scientific rationale for the exploration and study ofthese environments through a series of international meetings.The first was convened in 1994 in expectation of the publica-tion of the Nature article (Scott Polar Research Institute,University of Cambridge, 1994). The dimensions and settingof Vostok Subglacial Lake were discussed, and a preliminaryinventory of Subglacial Lakes was presented. By the time ofthe publication of the Nature article, 77 subglacial lake fea-tures had been identified based on analysis of existing radioecho sounding records [Siegert et al., 1996]. Three moreworkshops were convened in the late 1990s in quick succes-sion: (1) “Lake Vostok Study: Scientific Objectives and Tech-nological Requirements” (St. Petersburg, March 1998); (2)“Lake Vostok: A Curiosity or a Focus for Scientific Re-search?” (Washington D. C., United States, November 1998[Bell andKarl, 1998]); and (3) “Subglacial Lake Exploration”(Scientific Committee on Antarctic Research (SCAR), Cam-bridge, September 1999). The general conclusion of thesegatherings was that to adequately explore subglacial environ-ments a major, sustained investment in time, resources, andscientific effort would be needed for at least a decade. Inrecognition of this emerging frontier, SCAR convened aforum for scientists and technologists to gather, exchangeideas, and plan for the future: the Subglacial Antarctic LakeEnvironments Group of Specialists (SALEGoS) (2000–2004). In due course, SALEGoS transformed into a majorSCAR Scientific Research Program entitled Subglacial Ant-arctic Lake Environments (SALE) (2004–2010).

2. EARLY SCIENTIFIC DEVELOPMENTS

In the early stages, understanding of subglacial environ-ments was refined by remote sensing studies and theoreticalmodeling [Ridley et al., 1993; Siegert and Ridley, 1998;Wüest and Carmack, 2000; Mayer and Siegert, 2000]. Theinterface between the ice sheet and the underlying bed wasshown to contain liquid water at many locations. Ongoingspeculation about life in these lakes caught the imagination ofnot only scientists but the public in general. At the time, theonly available samples were surrogates of lake water, the so-called “accreted ice” that forms as lake water adfreezes to theunderside of the ice sheet [Karl et al., 1999; Jouzel et al.,

1999; Priscu et al., 1999; Bell et al., 2002]. This “accretedice” was unexpectedly encountered and recovered duringdeep coring of the Vostok ice borehole. The ice was recog-nized as unique because of its unusually large crystal sizes(feet in length), lack of meteoric gasses, and the purity of thewater collected on melting [Jouzel et al., 1999], comparedwith the meteoric glacier ice above that contained a well-characterized record of climate change [Petit et al., 1999].As additional geophysical surveys were conducted and

integrated with previously collected data [Tabacco et al.,2002; Studinger et al., 2003; Wright and Siegert, this vol-ume], it was established that subglacial lakes were commonbeneath thick (>2 km) ice sheets. In early inventories [Siegertet al., 1996], the number and distribution of features waslimited by the coverage of surveys. However, it was expectedthat identification of additional features would continue tomount as unexplored areas of Antarctica were surveyed. Onthe basis of fundamental considerations, subglacial lakeswere expected to occur across the Antarctic continent wher-ever thick accumulations of ice occurred, a hydrologicalcollection basin was accessible, and a source of water wasavailable. Vostok Subglacial Lake dominated early discus-sions as it was the only lake whose shape and size wereknown well; it remains the largest known subglacial lakewith an area of about 14,000 km2 and water depths reaching>1000 m [Siegert et al., this volume].As the inventory of lakes expanded, it was apparent that

subglacial aquatic features are not randomly distributedacross Antarctica. Instead they are located in preferred set-tings suggesting that a spectrum of lakes exist that might wellhave differing histories, ages, origins, and possibly livingresidents [Dowdeswell and Siegert, 1999]. Clusters of lakeswere documented in regions that exhibit distinct ice sheetdynamics in settings defined by the underlying basementmorphology [Dowdeswell and Siegert, 2002]. In the vicinityof Dome C and Concordia Station, “lake districts” wereidentified where subglacial features clustered near ice dividesand also at the heads of ice streams [Siegert and Ridley,1998; Siegert and Bamber, 2000]. It was speculated thatsome lakes were hydrologically connected in a manner anal-ogous to subaerial lake, stream, and wetland systems [Dow-deswell and Siegert, 2002]. The existence of sub-icehydrological systems transformed ideas about the evolutionand functioning of subglacial environments and redefinedinterests in these settings to include a wide variety of sub-glacial aquatic environments.

3. MOMENTUM BUILDS

Planning for, and discussions of, subglacial aquatic envi-ronment exploration and study gained additional momentum

2 INTRODUCTION

with the formation and approval of the International PolarYear 2007–2008 program “Subglacial Antarctic Lake Envi-ronments Unified International Team for Exploration andDiscovery (SALE UNITED).” Together SCAR SALE andSALE UNITED served as forums to exchange informationamong those interested in the study of these environments. Incombination, the programs included scientists and technolo-gists from Belgium, Canada, China, France, Germany, Italy,Russia, the United Kingdom, and the United States. Meet-ings were convened in Austria (2005), France (2006), theUnited States (2007), Russia (2008), and Belgium (2009) todevelop and refine plans for exploration and to share thelatest geophysical, microbiological, and modeling informa-tion. An international workshop entitled “Subglacial Antarc-tic Lake Environment in the IPY 2007-2008: AdvancedScience and Technology Planning Workshop” was convenedin Grenoble, France, in 2006 bringing together 84 partici-pants from 11 countries [Kennicutt and Petit, 2006]. Duringthis period, understanding of subglacial environments tookan unexpected turn.Analyses of changes in ice sheet surface elevations in

central East Antarctica, using satellite remote sensing, dem-onstrated that a lake in the Adventure Subglacial Trenchdischarged approximately 1.8 km3 of water over a period of14 months [Wingham et al., 2006]. The water flowed alongthe axis of a trench and into at least two other lakes about200 km downstream. The flux of water, ~ 50 m3 s�1, wasequivalent to the flow of the River Thames in London. Thisdiscovery was particularly interesting as, up until then, thecentral East Antarctica Ice Sheet was considered to be un-dynamic compared with West Antarctica. If significant flowof water occurred at the center of East Antarctica, flows ofsubglacial water were thought to be commonplace in Ant-arctica. Subglacial aquatic features appeared to be linked bya network of hydrological channels that were defined bybasal topography and surface ice sheet slope. Siegert et al.[2007] suggested that groups of lakes were likely to bejoined in discrete clusters acting as a system. Wright et al.[2008] established that flow channels were sensitive to theice surface slope, concluding that small changes in surfaceslope could result in major alterations of basal water flow.Periods of ice sheet changes, such as after the Last GlacialMaximum, or even as a consequence of global warming,might affect the frequency, magnitude, and direction of theseflow events. Up topographic slope (uphill) flow could beexpected as discharges were predicted to follow the hydro-logic potential established by variations in overlying icethickness interacting with underlying basement elevations.Further satellite remote sensing analyses illustrated that

subglacial discharge and water flow were indeed common-place in Antarctica [Smith et al., 2009]. It was confirmed that

many newly identified lakes and discharge areas were pref-erentially located at the heads of ice streams [Siegert andBamber, 2000; Bell et al., 2007]. Smith et al. [2009] furthersuggested that lakes actively discharge water into ice streambeds in response to varying basal flows. Satellite investiga-tions of the Byrd Glacier established that subglacial lakedischarges coincided with variations in flow velocities ob-served at an outlet glacier that drained East Antarctica[Stearns et al., 2008]. This inferred subglacial dynamics bothinfluenced, and were influenced by, overlying ice sheetdynamics.As subglacial lakes represent unique habitats, environmen-

tal stewardship during their eventual exploration was seen asa critical issue, and, early on (i.e., Cambridge, 1999), guidingprinciples were developed and adopted by the community.These concerns included the cleanliness of access techni-ques, contamination by the experiments that might be per-formed, the introduction of alien chemicals and biota, how tocollect unadulterated samples for laboratory analysis (espe-cially microbiological samples), and how best to protectsubglacial aquatic environments as sites of scientific andpublic interest. The U.S. National Academies convened acommittee to review aspects of subglacial lake explorationfrom an environmental protection and conservation perspec-tive [Committee on Principles of Environmental Stewardshipfor the Exploration and Study of Subglacial Environments,National Research Council, 2007; Doran and Vincent, thisvolume]. The National Academy findings were introduced atthe Antarctic Treaty Consultative Meeting in 2008 in Kiev,Ukraine, and SCAR subsequently provided guidance onthese issues as a code of conduct for subglacial lake explo-ration [Doran and Vincent, this volume]. These deliberationsserve as the basis for promulgating standards and proceduresfor the responsible conduct of subglacial aquatic environ-ment study and exploration.

4. VOSTOK SUBGLACIAL LAKE

Vostok Subglacial Lake has been, and continues to be, amajor focus of subglacial lake research [Lukin and Bulat, thisvolume]. A consortium of Russian research institutions led bythe Arctic and Antarctic Research Institute of Roshydrometconducted extensive geophysical surveys of the Vostok Sub-glacial Lake area and its vicinity within the framework ofthe Polar Marine Geological Research Expedition and theRussian Antarctic Expedition (RAE) [Masolov et al., 2006;Popov et al., 2006, 2007;Popov andMasolov, 2007]. A seriesof 1:1,000,000 maps of Vostok Subglacial Lake’s extent,ice and water body thicknesses, and bedrock relief were pro-duced as well as maps of the spatial pattern of internal layersin the overlying ice sheet. From this work, the lake’s

KENNICUTT AND SIEGERT 3

dimensions were better defined, the inclination of the ice-water interface was confirmed, and it was recognized thatVostok Subglacial Lake lies in a deep trough. Studinger et al.[2003] collected more than 20,000 km of aerogeophysicaldata producing detailed assessments of the lake and its glaci-ological setting. The existence of two basins was confirmedby gravity modeling of lake bathymetry, and the southernbasin of the lake was determined to be more than 1 km deep[Studinger et al., 2004;Masolov et al., 2001, 2006; Siegert etal., this volume].Geophysical, geodetic, and glaciological traverses, under-

taken by RAE, measured ice flow lines starting at Ridge Band passing through the drilling site at Vostok Station. AnItalian/French/Russian partnership also conducted traversesfrom Talos Dome via Dome C, Vostok Station, Dome B, andDome A. Thermomechanical ice flow line models werefurther constrained by this new information [Richter et al.,2008; Salamatin et al., 2008] to yield accurate estimatesof the distribution of accreted ice thickness and freezingrates, refined ice depth ages and temperature profiles, andestimated basal melt rates in the northern part of VostokSubglacial Lake.Continued deepening of the borehole at Vostok Station

extended the ice core isotopic profiles revealing significantspatial and/or temporal variability in physical conditions dur-ing accreted ice formation [Ekaykin et al., 2010]. Analysis ofaccreted ice revealed a distribution of helium isotopes in thelake water that could be explained by hydrothermal activitycontributing to the lake water hydrochemistry [Jean-Baptistet al., 2001; Bulat et al., 2004; de Angelis et al., 2004;P. Jean-Baptist, personal communication, 2009]. Althoughthe lake is known to possess small tides [Dietrich et al.,2001], geodetic GPS observations in the southern part ofVostok Subglacial Lake demonstrated that, on a time scale of5 years, the lake and ice sheet in the vicinity of Vostok Stationwere in steady state in contrast to other subglacial lakes thatwere then known to be dynamic [Richter et al., 2008].

5. LIFE IN SUBGLACIAL AQUATIC ENVIRONMENTS

As understanding of the physical conditions in subglacialenvironments (temperature, pressure, salinity, etc.) was be-ing refined, the existence of life in the lakes remained a focusof great speculation [Skidmore, this volume]. A consensusgrew that extremely low nutrient levels were to be expected,suggesting these habitats could be challenging for possiblemicrobial inhabitants. Superoxic conditions caused by clath-rate decomposition and formation, especially at the water-sediment interface on the lake floor, were also speculated,and it was suggested that these conditions would be toxic toorganisms other than anaerobes [Siegert et al., 2003].

At this time, the only clues about possible life in VostokSubglacial Lake came from extrapolations based on theanalyses of the accreted ice [Karl et al., 1999; Priscu et al.,1999]. Contamination of accreted ice samples during dril-ling, recovery, transportation, and analysis called these re-sults into question as these samples were not originallyretrieved for microbiological analyses. The effects of parti-tioning of lake water constituents during ice formation,under subglacial lake conditions, were poorly understoodmaking inferences of lake water chemistry difficult. Theoutcome was conflicting evidence for life in the lake andambiguity about the biogeochemistry of lake water. Theseuncertainties led to differing opinions about whether hydro-thermal effluents contributed to Vostok Subglacial Lakewaters. A general consensus evolved that these environ-ments would most likely contain life and that organismsmore complex than microbes were highly unlikely. Therecognition of hydrological connections among these envir-onments meant that water in many subglacial lakes waslikely isolated for far fewer years than first speculated,decreasing the possibility of long-term (>1 Ma) isolation.Depending on the method, the turnover times for water inVostok Subglacial Lake have been calculated to be between50,000 and 100,000 years but certainly not millions of years[Siegert et al., 2001; Bell et al., 2002].More recently, additional accretion ice, and samples of

snow collected from layers deposited before the beginningof coring at Vostok Station, contributed further to the debateabout possible life within the lake [Bulat et al., 2004, 2007b;Alekhina et al., 2007]. These results suggest that extremelylow biomass of both atmospheric and lake water origins ispresent [Bulat et al., 2009]. Similar studies by United Statesand United Kingdom researchers confirmed the low cellnumbers and low microbial diversity in glacial and accretedice, though a range of cell numbers and greater diversity havebeen detected by some investigators [Christner et al., 2006].The few bacterial phylotypes recovered from accreted icewere isolated from ice layers that contain mineral inclusionsraising further questions about their origin [Bulat et al.,2009].Current knowledge of the lake conditions, inferred from

the chemistry of accretion ice studies and from modeling,suggests that the Vostok Subglacial Lake may be inhabitedby chemoautotrophic psychrophiles that can tolerate highpressures and possibly high oxygen concentrations, thoughno conclusive evidence of such microorganisms has yet beenfound because of a lack of direct sampling of lake water[Bulat et al., 2007a]. The presence of a thermophilic, che-moautotrophic bacterium, Hydrogenophilus thermoluteolus(previously identified in other areas influenced by hydrother-mal activity remote from Antarctica), has been reported

4 INTRODUCTION

[Bulat et al., 2004; Lavire et al., 2006]. It has been speculatedthat water in Vostok Subglacial Lake will be an extremelydilute biological solution suggesting that life, if present,may be primarily restricted to lake sediments and the basalwater interface [Bulat et al., 2009]. While studies of theVostok Subglacial Lake accretion ice have improved com-prehension of physical, chemical, and biological processes inthe lake, considerable debate continues as to the level andtype of life expected in these environments. The debate willnot be resolved until direct measurement and sampling ofthese environments has taken place.

6. OTHER SUBGLACIAL LAKES

Amajor set of subglacial lakes was recently identified at theonset of the Recovery Ice Stream a major East Antarctic iceflow unit [Bell et al., 2007]. Three or possibly four largesubglacial lakes (smaller than Vostok Subglacial Lake but,nonetheless, far larger than most) are thought to be coincidentwith the onset of rapid ice flow. The lakes exhibit distinctiveice surface morphologies including extensive, relatively flatfeatureless regions bounded by upstream troughs and down-stream ridges generated by changes in bottom topography.The Recovery subglacial lakes are hypothesized to containwater derived from basal melting routed to the lake from alarge upstream catchment area. To study the Recovery lakesregion a U.S.-Norway traverse conducted surface geophysi-cal surveys and installed GPS stations. Ice sheet motion wasquantified by collecting gravity magnetics, laser, and radardata over the two southernmost Recovery lakes [Block et al.,2009]. Once fully interpreted, these data will clarify thedynamics of the origins of subglacial water in the lakes andthe upstream catchment as well as evaluate the geologicsetting of these features. All four of the Recovery lakes werecrossed by the U.S.-Norwegian traverse in January 2009, andlow-frequency radar was used to map the morphology of thesubglacial lakes and image the ice sheet bed of the lakesidentified by Smith et al. [2009]. In the coming years, as thesedata sets are processed, the role that subglacial lakes play incontrolling the onset of fast ice flow will be better defined.

7. NEW FRONTIER

Significant progress in the study of subglacial aquaticenvironments is now at hand with the initiation of an impor-tant phase with three exploration programs likely to advanceunderstanding of these environments over the next 3 to 5 years.A United Kingdom–led international program has completeda full survey of Ellsworth Subglacial Lake located in WestAntarctica, and plans to undertake direct clean measurementsand sampling of the lake in 2012/2013 are in place [Ross

et al., this volume]. The United States has launched amajor program to survey, enter, instrument, and sample an“actively discharging” subglacial aquatic system beneathWhillans Ice Stream in West Antarctica at around the sametime [Fricker et al., this volume]. Russian researchers aredeveloping further strategies for penetration of Vostok Sub-glacial Lake, and lake entry is expected in the next few fieldseasons [Lukin and Bulat, this volume].In the past decade, our understanding of the importance

of subglacial aquatic systems as habitats for life, and oftheir influence on ice sheet dynamics, has been greatlyadvanced. Subglacial features that contain liquid water arenow known to be common beneath the ice sheets of Ant-arctica. A spectrum of subglacial environments exists asconnected subglacial hydrologic systems and water move-ment beneath ice sheets can and does occur over a range ofspatial and temporal scales. The location of subglacialaquatic accumulations and the onset of ice streams havebeen shown to be linked in some areas, suggesting that icesheet dynamics can be affected by hydrological systems atthe base of the ice sheet.The exploration and study of subglacial aquatic environ-

ments remains at its early stages and if the major advancesrealized to date are an indication of what is to come, evenmore fundamental discoveries will be realized in the yearsahead. In little more than a decade, findings regarding sub-glacial aquatic systems have transformed fundamental con-cepts about Antarctica and its ice sheets. Ice sheet bases arenow seen as being highly dynamic at their beds, involving acomplex interplay of hydrology, geology, glaciology, tecton-ics, and ecology now and in the past. Ongoing and plannedprojects to directly sample these environments will ulti-mately determine if subglacial waters house unique andspecially adapted microbiological assemblages and recordsof past climate change. The most remarkable advances to berealized from the study of this next frontier in Antarcticscience will probably be wholly unexpected as these recentlyrecognized environments are explored.This volume serves as a benchmark for knowledge about

subglacial aquatic environments and as an update on thelatest research developments, setting the stage for major newexploration efforts.

REFERENCES

Alekhina, I. A., D. Marie, J.-R. Petit, V. V. Lukin, V. M. Zubkov,and S. A. Bulat (2007), Molecular analysis of bacterial diversityin kerosene-based drilling fluid from the deep ice borehole atVostok, East Antarctica, FEMS Microbiol. Ecol., 59, 289–299.

Bell, R., and D. M. Karl (1998), Lake Vostok Workshop: “ACuriosity or a Focus for Interdisciplinary Study,” final report,Natl. Sci. Found., Washington, D. C.


Bell, R. E.,M. Studinger, A. Tikku, G. K. C. Clarke,M.M. Gutner, andC. Meertens (2002), Origin and fate of Lake Vostok water frozen tothe base of the East Antarctic ice sheet, Nature, 416, 307–310.

Bell, R. E., M. Studinger, C. A. Shuman, M. A. Fahnestock, and I.Joughin (2007), Large subglacial lakes in East Antarctica at theonset of fast-flowing ice, Nature, 445, 904–907.

Block, A. E., R. E. Bell, and M. Studinger (2009), Antarctic crustalthickness from satellite gravity: Implications for the Transantarc-tic and Gamburtsev subglacial mountains, Earth Planet. Sci.Lett., 288, 194–203.

Bulat, S. A., et al. (2004), DNA signature of thermophilic bacteriafrom the aged accretion ice of Lake Vostok, Antarctica: Implica-tions for searching life in extreme icy environments, Int. J.Astrobiol., 3(1), 1–12, doi:10.1017/S1473550404001879.

Bulat, S. A., I. A. Alekhina, V. Y. Lipenkov, V. V. Lukin, and J. R.Petit (2007a), Microbial life in extreme subglacial Antarctic lakeenvironments: Lake Vostok, in Bresler Memorial Lectures II,pp. 264–269, Petersburg Nucl. Phys. Inst., Russ. Acad. of Sci.,St. Petersburg, Russia.

Bulat, S. A., I. A. Alekhina, J.-R. Petit, V. Y. Lipenkov, and V. V.Lukin (2007b), An assessment of the biogeochemical potential ofsubglacial Lake Vostok, East Antarctica, in relation to microbiallife support (in Russian), Probl. Arktiki Antarktiki, 76, 106–112.

Bulat, S. A., I. A. Alekhina, V. Y. Lipenkov, V. V. Lukin, D. Marie,and J. R. Petit (2009), Cell concentration of microorganisms inglacial and lake ice of the Vostok ice core, East Antarctica (inRussian), Mikrobiologiya, 78(6), 850–852.

Christner, B., G. Royston-Bishop, C. M. Foreman, B. R. Arnold, M.Tranter, K. A. W. B. Welch, W. B. Lyons, A. I. Tsapin, M. Studin-ger, and J. C. Priscu (2006), Limnological conditions in subglacialLake Vostok, Antarctica, Limnol. Oceanogr., 51(6), 2485–2501.

Committee on Principles of Environmental Stewardship for theExploration and Study of Subglacial Environments, NationalResearch Council (2007), Exploration of Antarctic SubglacialAquatic Environments: Environmental and Scientific Steward-ship, 152 pp., Natl. Acad. Press, Washington, D. C.

de Angelis, M., J. R. Petit, J. Savarino, R. Souchez, and M. H.Thiemens (2004), Contributions of an ancient evaporitic-typereservoir to subglacial Lake Vostok chemistry, Earth Planet. Sci.Lett., 222, 751–765.

Dietrich, R., K. Shibuya, A. Pötzsch, and T. Ozawa (2001), Evi-dence for tides in the subglacial Lake Vostok, Antarctica, Geo-phys. Res. Lett., 28, 2971–2974, doi:10.1029/2001GL013230.

Doran, P. T., and W. F. Vincent (2011), Environmental protectionand stewardship of subglacial aquatic environments, in AntarcticSubglacial Aquatic Environments, Geophys. Monogr. Ser., doi:10.1029/2010GM000947, this volume.

Dowdeswell, J. A., and M. J. Siegert (1999), The dimensions andtopographic setting of Antarctic subglacial lakes and implicationsfor large-scale water storage beneath continental ice sheets, Geol.Soc. Am. Bull., 111, 254–263.

Dowdeswell, J. A., and M. J. Siegert (2002), The physiography ofmodern Antarctic subglacial lakes, Global Planet. Change, 35,221–236.

Ekaykin, A. A., V. Y. Lipenkov, J. R. Petit, S. Johnsen, J. Jouzel,and V. Masson-Delmotte (2010), Insights into hydrological re-gime of Lake Vostok from differential behavior of deuterium andoxygen-18 in accreted ice, J. Geophys. Res., 115, C05003,doi:10.1029/2009JC005329.

Fricker, H. A., et al. (2011), Siple Coast subglacial aquatic environ-ments: The Whillans Ice Stream Subglacial Access ResearchDrilling project, in Antarctic Subglacial Aquatic Environments,Geophys. Monogr. Ser., doi: 10.1029/2010GM000932, this volume.

Jean-Baptiste, P., J. R. Petit, V. Y. Lipenkov, D. Raynaud, and N. I.Barkov (2001), Constraints on hydrothermal processes and waterexchange in Lake Vostok from helium isotopes, Nature, 411,460–462.

Jouzel, J., J. R. Petit, R. Souchez, N. I. Barkov, V. Y. Lipenkov, D.Raynaud, M. Stievenard, N. I. Vassiliev, V. Verbeke, and F.Vimeux (1999), More than 200 meters of lake ice above subgla-cial Lake Vostok, Antarctica, Science, 286(5447), 2138–2141.

Kapitsa, A. P., J. K. Ridley, G. Q. Robin, M. J. Siegert, and I. A.Zotikov (1996), A large deep freshwater lake beneath the ice ofcentral East Antarctica, Nature, 381, 684–686.

Karl, D. M., D. F. Bird, K. Bjorkman, T. Houlihan, R. Shackelford,and T. Tupas (1999), Microorganisms in the accreted ice of LakeVostok, Antarctica, Science, 286(5447), 2144–2147.

Kennicutt, M. C., and J. R. Petit (2006), Subglacial Antarcticlake environments (SALE) in the International Polar Year:Advanced science and technology planning - Grenoble France,April, 2006, workshop report, 43 pp., Natl. Sci. Found., Wa-shington, D. C.

Lavire, C., P. Normand, I. Alekhina, S. Bulat, D. Prieur, J. L.Birrien, P. Fournier, C. Hänni, and J. R. Petit (2006), Presenceof Hydrogenophylus thermoluteolus DNA in accretion ice in thesubglacial Lake Vostok, Antarctica, assessed using rrs, cbb andhox, Environ. Microbiol., 8(12), 2106–2114.

Lukin, V., and S. Bulat (2011), Vostok Subglacial Lake: Details ofRussian plans/activities for drilling and sampling, in AntarcticSubglacial Aquatic Environments, Geophys. Monogr. Ser., doi:10.1029/2010GM000951, this volume.

Masolov, V. N., V. V. Lukin, A. N. Sheremetyev, and S. V. Popov(2001), Geophysical investigations of the subglacial Lake Vostokin Eastern Antarctica (in Russian), Dokl. Akad. Nauk, 379(5),680–685.

Masolov, V. N., S. V. Popov, V. V. Lukin, A. N. Sheremet’ev, andA. M. Popkov (2006), Russian geophysical studies of LakeVostok, central East Antarctica, in Antarctica: Contributions toGlobal Earth Sciences, edited by D. K. Fütterer et al., pp. 135–140, Springer, New York.

Mayer, C., and M. J. Siegert (2000), Numerical modelling of ice-sheet dynamics across the Vostok subglacial lake, central EastAntarctica, J. Glaciol., 46, 197–205.

Oswald, G. K. A., and G. d. Q. Robin (1973), Lakes beneath theAntarctic ice sheet, Nature, 245, 251–254.

Petit, J. R., et al. (1999), Climate and atmospheric history of the past420,000 years from the Vostok ice core, Antarctica, Nature, 399,429–436.

6 INTRODUCTION

Popov, S. V., and V. N. Masolov (2007), Forty-seven new subglaciallakes in the 0–110- sector of East Antarctica, J. Glaciol., 53,289–297.

Popov, S. V., A. N. Lastochkin, V. N. Masolov, and A. M.Popkov (2006), Morphology of the subglacial bed relief ofLake Vostok basin area (central East Antarctica) based on RESand seismic data, in Antarctica: Contributions to Global EarthSciences, edited by D. K. Fütterer et al., pp. 141–146, Springer,New York.

Popov, S. V., V. Y. Lipenkov, A. V. Enalieva, and A. V. Preobraz-henskaya (2007), Internal isochrone layers in the ice sheet aboveLake Vostok, East Antarctica (in Russian), Probl. Arktiki An-tarktiki, 76, 89–95.

Priscu, J. C., et al. (1999), Geomicrobiology of subglacial ice aboveLake Vostok, Antarctica, Science, 286(5447), 2141–2144.

Richter, A., S. V. Popov, R. Dietrich, V. V. Lukin, M. Fritsche, V. Y.Lipenkov, A. Y. Matveev, J. Wendt, A. V. Yuskevich, and V. N.Masolov (2008), Observational evidence on the stability of thehydro-glaciological regime of subglacial Lake Vostok, Geophys.Res. Lett., 35, L11502, doi:10.1029/2008GL033397.

Ridley, J. K., W. Cudlip, and S. W. Laxon (1993), Identification ofsubglacial lakes using ERS-1 radar altimeter, J. Glaciol., 39,625–634.

Robin, G. d. Q., D. J. Drewry, and D. T. Meldrum (1977), Interna-tional studies of ice sheet and bedrock, Philos. Trans. R. Soc.London, 279, 185–196.

Ross, N., et al. (2011), Ellsworth Subglacial Lake, West Antarctica:Its history, recent field campaigns, and plans for its exploration,in Antarctic Subglacial Aquatic Environments, Geophys.Monogr. Ser., doi: 10.1029/2010GM000936, this volume.

Salamatin, A. N., E. A. Tsyganova, S. V. Popov, and V. Y. Lipenkov(2008), Ice flow line modeling in ice core data interpretation:Vostok Station (East Antarctica), in Physics of Ice Core Records,vol. 2, edited by T. Hondoh, Hokkaido Univ. Press, Sapporo, Japan.

Siegert, M. J., and J. L. Bamber (2000), Subglacial water at theheads of Antarctic ice-stream tributaries, J. Glaciol., 46,702–703.

Siegert, M. J., and J. K. Ridley (1998), An analysis of the ice-sheetsurface and subsurface topography above the Vostok Stationsubglacial lake, central East Antarctica, J. Geophys. Res., 103,10,195–10,208.

Siegert, M. J., J. A. Dowdeswell, M. R. Gorman, and N. F. McIntyre(1996), An inventory of Antarctic sub-glacial lakes, Antarct. Sci.,8, 281–286.

Siegert, M. J., J. C. Ellis-Evans, M. Tranter, C. Mayer, J.-R. Petit, A.Salamatin, and J. C. Priscu (2001), Physical, chemical and bio-logical processes in Lake Vostok and other Antarctic subglaciallakes, Nature, 414, 603–609.

Siegert, M. J., M. Tranter, C. J. Ellis-Evans, J. C. Priscu, and W. B.Lyons (2003), The hydrochemistry of Lake Vostok and the po-tential for life in Antarctic subglacial lakes, Hydrol. Processes,17, 795–814.

Siegert, M. J., A. Le Brocq, and A. Payne (2007), Hydrologicalconnections between Antarctic subglacial lakes and the flow ofwater beneath the East Antarctic ice sheet, in Glacial Sedimen-tary Processes and Products, edited by M. J. Hambrey et al.,Spec. Publ. Int. Assoc. Sedimentol., 39, 3–10.

Siegert, M. J., S. Popov, and M. Studinger (2011), Vostok Subgla-cial Lake: A review of geophysical data regarding its discoveryand topographic setting, in Antarctic Subglacial Aquatic Envir-onments, Geophys. Monogr. Ser., doi: 10.1029/2010GM000934,this volume.

Skidmore, M. (2011), Microbial communities in Antarctic subgla-cial aquatic environments, in Antarctic Subglacial Aquatic Envir-onments, Geophys. Monogr. Ser., doi: 10.1029/2010GM000995,this volume.

Smith, B. E., H. A. Fricker, I. R. Joughin, and S. Tulaczyk (2009),An inventory of active subglacial lakes in Antarctica detected byICESat (2003–2008), J. Glaciol., 55, 573–595.

Stearns, L. A., B. E. Smith, and G. S. Hamilton (2008), Increasedflow speed on a large East Antarctic outlet glacier caused bysubglacial floods, Nat. Geosci., 1, 827–831.

Studinger, M., et al. (2003), Ice cover, landscape setting and geo-logical framework of Lake Vostok, East Antarctica, Earth Planet.Sci. Lett., 205, 195–210.

Studinger, M., R. E. Bell, and A. A. Tikku (2004), Estimating thedepth and shape of subglacial Lake Vostok’s water cavity fromaerogravity data, Geophys. Res. Lett., 31, L12401, doi:10.1029/2004GL019801.

Tabacco, I. E., C. Bianchi, A. Zirizzotti, E. Zuccheretti, A. Forieri,and A. Della Vedova (2002), Airborne radar survey above Vostokregion, East Antarctica: Ice thickness and Lake Vostok geometry,J. Glaciol., 48, 62–69.

Wingham, D. J., M. J. Siegert, A. Shepherd, and A. S. Muir (2006),Rapid discharge connects Antarctic subglacial lakes, Nature,440, 1033–1036.

Wright, A., and M. J. Siegert (2011), The identification and phys-iographical setting of Antarctic subglacial lakes: An update basedon recent discoveries, in Antarctic Subglacial Aquatic Environ-ments, Geophys. Monogr. Ser., doi: 10.1029/2010GM000933,this volume.

Wright, A. P., M. J. Siegert, A. Le Brocq, and D. Gore (2008), Highsensitivity of subglacial hydrological pathways in Antarctica tosmall ice sheet changes, Geophys. Res. Lett., 35, L17504,doi:10.1029/2008GL034937.

Wüest, A., and E. Carmack (2000), A priori estimates of mixing andcirculation in the hard-to-reach water body of Lake Vostok,Ocean Modell., 2(1), 29–43.

M. C. Kennicutt, Department of Oceanography, Texas A&MUniversity, College Station, TX 77843-3146, USA.M. J. Siegert, School of GeoSciences, University of Edinburgh,

Edinburgh EH9 3JW, UK. ([email protected])


The Identification and Physiographical Setting of Antarctic Subglacial Lakes:An Update Based on Recent Discoveries

Andrew Wright and Martin J. Siegert

Grant Institute, School of GeoSciences, University of Edinburgh, Edinburgh, UK

We investigate the glaciological and topographic setting of known Antarcticsubglacial lakes following a previous assessment by Dowdeswell and Siegert(2002) based on the first inventory of 77 lakes. Procedures used to detectsubglacial lakes are discussed, including radio echo sounding (RES) (which wasfirst used to demonstrate the presence of subglacial lakes), surface topography,topographical changes, gravity measurements, and seismic investigations. Recentdiscoveries of subglacial lakes using these techniques are detailed, from which arevised new inventory of subglacial lakes is established, bringing the totalnumber of known subglacial lakes to 387. Using this new inventory, we examinevarious controls on subglacial lakes, such as overlying ice thickness and positionwithin the ice sheet and formulate frequency distributions for the entire subglaciallake population based on these (variable) controls. We show how the utility ofRES in identifying subglacial lakes is spatially affected; lakes away from the icedivide are not easily detected by this technique, probably due to scattering at theice sheet base. We show that subglacial lakes are widespread in Antarctica, and itis likely that many are connected within well-defined subglacial hydrologicalsystems.

1. INTRODUCTION

A variety of methods have been used in the discoveryand characterization of subglacial lakes and the identifica-tion of subglacial water movement in Antarctica (Figure 1).The first inventory of subglacial lakes, recording 77 lakelocations, used the technique of radio echo sounding [Sie-gert et al., 1996]. This was later updated to 145 lakes bySiegert et al. [2005]. Several other techniques are availablefor the detection of subglacial lakes, including surfacetopography, topographical changes, gravity survey, andseismic investigations. The aim of this paper is to detail all

the techniques available for the detection of subglaciallakes, and to pull together recent information regardinglake locations. In doing so, a revised inventory is estab-lished, from which an assessment of the dimensions andtopographic setting of subglacial lakes is updated [fromDowdeswell and Siegert, 2002].

2. DISCOVERY, IDENTIFICATION, ANDCHARACTERIZATION OF SUBGLACIAL LAKES

2.1. Radio Echo Sounding (RES)

2.1.1. Development of the technique. The technique ofRES takes advantage of a window in the radio part of theelectromagnetic spectrum within which emitted waves willtravel freely through both ice and air. As with all E-M waves,reflections occur at boundaries between materials with dif-ferent dielectric properties and therefore different speeds of


9

wave propagation. On entering ice, the speed of the radiowave drops by nearly half, from 300 to 168 m μs�1 [Glenand Paren, 1975]. An active transmit/receive radar antenna,mounted either near the surface [e.g., Popov et al., 2003] oron an airborne platform [e.g., Blankenship et al., 2001], can

therefore be used to detect reflections originating from bothwithin and at the base of a glacier or ice sheet.The initial investigations of the base of the Antarctic ice

sheet were carried out as part of a joint Scott Polar ResearchInstitute (SPRI), National Science Foundation (NSF), and

Figure 1.Methods of investigating subglacial lakes: (a) radio echo sounding (RES), a lake reflector is visible in the center;(b) free-air gravitational anomaly (with ice sheet effects removed) detected from the air above Vostok Subglacial Lake[Studinger et al., 2004b]; (c) prominent flat spot in the ice sheet extending north of Vostok Station, 10-m contours fromERS-1 radar altimetry [Siegert and Ridley, 1998b]; and (d) vertical surface elevation changes measured by ICESat used toidentify two lakes within the catchment of the Byrd Glacier, East Antarctica. Reprinted by permission from MacmillanPublishers Ltd: Nature Geoscience [Stearns et al., 2008], copyright 2008.

10 IDENTIFICATION AND SETTING OF ANTARCTIC SUBGLACIAL LAKES

Technical University of Denmark (TUD) airborne RES cam-paign. Between 1967 and 1979, this project completed over400,000 km of line transects, spaced by an average of 50 km,and covering approximately half of the total area of Antarc-tica [Robin et al., 1977; Drewry, 1983].The SPRI-NSF-TUD survey was able to penetrate even

the thickest ice in Antarctica using a 60-MHz frequencyradar. By measuring the time elapsed between transmit andreceive, it was shown that an ice thickness greater than4 km existed over large parts of East Antarctica [Drewry,1983]. Nearly 40 years later, this is still the greatest aerialcoverage of any single aerogeophysical survey in Antarc-tica. More recent airborne RES surveys have been charac-terized by smaller spatial coverage but much higher spatialresolution [e.g., Rémy and Tabacco, 2000; Popov et al.,2002; Rippin et al., 2003; Studinger et al., 2003a, 2004a;Holt et al., 2006a; Vaughan et al., 2006]. In several in-stances, these surveys have targeted areas not covered, oronly sparsely covered by the SPRI-NSF-TUD data, withthe result being a patchwork of bed information. Despitethese efforts, however, a number of large gaps still existwhere our knowledge of the ice sheet bed remains poor[e.g., Le Brocq et al., 2008].Several reviews detail both developments in the techni-

ques of radioglaciology and the resulting enhancements inour understanding of the ice sheets [e.g., Plewes and Hub-bard, 2001; Dowdeswell and Evans, 2004; Bingham andSiegert, 2007]. This subject will therefore not be discussedfurther here.

2.1.2. The discovery of subglacial lakes. The discovery ofthe first subglacial lake occurred near the Russian station atSovetskaya during the 1967/1968 season of SPRI-NSF-TUD radio echo sounding [Robin et al., 1970]. An area ofunusually low signal fading and short duration of the re-turned pulse, indicating a specular reflection, was found tocoincide with low attenuation of the transmitted signal and anear-horizontal, flat bed geometry. This was, at first tenta-tively, best explained as the result of a sub-ice water body[Robin et al., 1970]. During the 1971/1972 season, anextension of the RES survey over the Dome C area identi-fied a further 16 similar locations, indicating that the occur-rence of pockets of liquid water, or subglacial lakes, beneaththe central regions of the ice sheet might be relativelycommonplace [Oswald and Robin, 1973]. Owing to the lackof penetration through water of radio waves at megahertzfrequencies, the depths of these newly discovered featurescould not be determined, only that a sufficient depth (i.e., afew meters or more) must exist to permit the continuous,strong, and flat echo returns observed [Oswald and Robin,1973].

2.1.3. Characterization of bed reflections. Radio echosounding has been used to gather much of what is currentlyknown about the subglacial environment of Antarctica.When the velocity of the radar pulse is known, or can beestimated, the thickness of the ice can be calculated bymeasuring the time difference between echoes received fromthe air/surface and ice/bed interfaces. A time series of suchechoes recorded as the observer moves over the ice surfacecan be used to create a pseudo-cross-section of the ice sheetand of the underlying bed. These data combined with mea-surements of the surface elevation can be used to reconstructthe topography of the underside of the ice sheet [Drewry,1983; Lythe et al., 2001].It was soon realized that much more information about the

subglacial environment could be obtained from an analysisof the echo returns. In particular, the strength [Neal, 1976]and shape [Berry, 1973, 1975] of the returned pulse is relatedto the degree of scattering at the interface and therefore to themicrotopography of the subglacial surface. Early studieswere limited to the use of “incoherent” radar-sounding ap-paratus. Modern RES equipment can record both amplitudeand phase of reflected pulses (“coherent” radar) allowing amoving platform to operate in the Synthetic Aperture (SAR)mode [Gogineni et al., 1998]. Coherent integration bothallows the detection of radar reflections where they wouldotherwise be obscured by scattering from crevasses, etc. andimproves the ability to quantify reflection and scatteringfrom a subglacial interface [Peters et al., 2005].Radar reflections within ice are caused by changes in

dielectric permittivity (er) due to changing density or crystalfabric orientation and by changes in electrical conductivity,due largely to varying acidity associated with the fallout ofvolcanic aerosols [Fujita et al., 1999]. Basal reflections aregenerally caused by the large difference in dielectric imped-ance between ice and the basal material, the magnitude of thereflection being proportional to the change in impedance.Dielectric constants for the various types of bedrock, ob-served in Antarctica, range from a minimum of ~4 to amaximum of ~9. This is very much closer to the value forglacier ice (er = 3.2) than is the dielectric constant of purewater (er = 80). For this reason, a basal reflection from an ice-water subglacial interface is much brighter than the equiva-lent reflection from a dry interface or from frozen sediments[Bogorodskiy et al., 1985]. Shabtaie et al. [1987] showedthat the minimum sub-ice water thickness required for awater-dominated reflection is between a few tens of centi-meters to a few meters, depending on salinity.The shape of a smooth ice/bed or ice/water interface is an

additional factor which can affect the strength of the returnedecho in the same way that a concave or convex mirror acts tofocus light [Tabacco et al., 2000]. In tests conducted over a

WRIGHT AND SIEGERT 11

floating ice tongue, isolated geometrical effects have beenshown to influence the total received power by ±6–8 dB[Bianchi et al., 2004].

2.1.4. Identifying subglacial lakes by RES. The strength ofthe radio echo from the base of the ice sheet has been used byseveral authors to infer information about basal conditions invarious glaciated regions [e.g., Bentley et al., 1998; Gades etal., 2000; Catania et al., 2003; Peters et al., 2005, 2007]. Todo this, the proportion of energy reflected at the bed (thebasal reflection coefficient) must be distinguished from themany other factors which can affect the strength of the signalreceived at the antenna. Probably, the most significant ofthese is the dielectric power loss during transmission throughthe ice. This depends sensitively on ice temperature and canvary spatially by 15–20 dB km�1 [Peters et al., 2007]. Whilesubglacial water will always produce a bright radar reflec-tion, an “absolute brightness” criteria for lake identificationcan be misleading. Rather, it is the brightness of a particularfeature “relative” to its surroundings, which can be moreuseful in identifying subglacial lakes [Carter et al., 2007].Amplitude fading is the fluctuation in radio echo ampli-

tude as the observer moves at a fixed distance from aninterface, it is caused by interference from different scatter-ing centers fore and aft of the observers position and can beused to obtain useful information about bed roughness[Oswald, 1975]. Very low fading (or alternatively, a verylarge “fading distance”) implies a continuous, flat, mirror-like or “specular” reflection. A purely specular reflectioncan only occur where the interface is smooth on the scaleof the radar footprint. A substantial body of water at thebed of an ice sheet will exhibit a smooth ice-water interfacethat will also satisfy the criteria of hydrostatic equilibrium.This states that due to the different densities of ice andwater, and assuming that the water supports the full over-burden pressure of the ice, the ice-water interface will havea slope 11 times greater and in the opposite direction to theslope of the ice surface. Calculations of the hydrologicalpotential field can therefore be useful in evaluating subgla-cial lake candidates from their radar profile [Oswald, 1975;Carter et al., 2007].The electrical properties of liquid water act to inhibit the

transmission of electromagnetic waves. For this reason, REScannot normally be used to determine the depths of subgla-cial lakes. An exception to this has been found in shallowregions of some lakes surveyed with the SPRI-NSF-TUDradar, where bottom reflections have been recorded fromdepths of up to 21 m below the lake’s surface [Gorman andSiegert, 1999]. These observations confirm that these lakesare, in fact, substantial bodies of water, but indicate onlytheir minimum depths.

2.2. Identification of Subglacial Lakes From IceSurface Topography

Even before subglacial lakes had been firmly identified inRES records, their surface expressions had been noted by pilotstraversing the center of the continent. Unusually flat areas ofthe ice sheet were often referred to as “lakes” and were fre-quently used as landmarks for navigation before any connec-tionwasmade to the subglacial environment [Robinson, 1960].When an ice sheet flows over a localized body of water, the

weight of the ice is taken by the incompressible fluid. Pro-vided that there is no outlet channel for the water to escape,this will lead to the establishment of local hydrostatic equi-librium. This has a significant affect on the flow regime ofthe ice and, for a large enough lake, can result in the mor-phological expression of an extremely flat and featureless icesurface, similar to that of a floating ice shelf.Satellite observations with the Seasat radar altimeter iden-

tified a prominent flat area, in Terre Adelie, East Antarctica,the position of which was shown to correspond to a subglacialreflector identified in the SPRI-NSF-TUD radar record [Cu-dlip and McIntyre, 1987]. This technique achieved greatersuccess when several RES lake reflectors in the area to thenorth of Vostok Station [Robin et al., 1977] were shown to liebeneath a single, continuous flat surface area observed withthe ERS-1 satellite [Ridley et al., 1993; Kapitsa et al., 1996].A finding later confirmed and elaborated on using moresophisticated radar altimetry techniques [Roemer et al.,2007] and laser altimetry [Studinger et al., 2003a].Subsequent analysis of ERS-1 data identified flat surface

features associated with a further 28 subglacial lakes knownfrom RES records in the Dome C and Terre Adelie regions[Siegert and Ridley, 1998a]. Small subglacial lakes (dimen-sions <4 km) are generally not found to have a correspondingflat surface feature. Furthermore, flat areas of ice, meetingthe criteria for identification of a subglacial lake, have alsobeen shown to occur where no lake exists [Siegert andRidley, 1998a]. Water-saturated sediments can cause a simi-lar reduction in basal stress and, therefore, induce a similarsurface expression. For this reason, a surface flat area alone isnot normally sufficient evidence for a lake discovery [Siegertand Ridley, 1998a].In addition, for floating ice, the retarding force of the basal

shear stress is reduced to zero. As a result, ice flowing from asolid bed onto a subglacial lake experiences acceleration.The resulting extensional flow has been shown to cause alocal thinning of the ice and a lowering of the surface on theupstream side of the lake [Shoemaker, 1990; Gudmundsson,2003; Pattyn et al., 2004]. Conversely, a thickening of the icecan occur over the downstream lake shore, where the returnof basal drag causes compressive flow.


Until recently, it was thought that no other subglacial lakesof a similar scale to Vostok Subglacial Lake (hereinafterreferred to as Lake Vostok) existed beneath Antarctica [Sie-gert, 2000]. Imagery from the Moderate Resolution ImagingSpectroradiometer (MODIS) satellite, however, has nowbeen used to determine large surface areas for two lakes(90-E Lake and Lake Sovetskaya) that were previouslyknown only from relatively short sections of RES survey[Bell et al., 2006]. Further discoveries of four large lakes inthe upstream region of the Recovery Ice Stream have beenmade using MODIS imagery to locate the lake surfaces,which have then been shown to possess surface ridge andtrough features consistent with the direction of ice flow [Bellet al., 2007].

2.3. Discovery of Active Subglacial Lakes by Measurementsof Surface Height Change

It is now becoming widely recognized that subglacialdrainage systems are generally dynamic in nature. Changesassociated with the movement of water either betweenknown lakes or between lakes and a distributed hydrologicalsystem are apparently common occurrences [Siegert et al.,2007]. Movement of subglacial water has been known, orsuspected, as the cause of vertical displacements of the icesurface of valley glaciers for a number of years [e.g., Iken etal., 1983; Fatland and Lingle, 2002]. Not until recently,however, have the means, in the form of repeat satellitemeasurements, been available to observe such local surfaceheight changes in the remote regions of the Antarctic plateau.Gray et al. [2005] were the first to identify vertical move-

ment of the Antarctic ice sheet that could be attributed to themovement of subglacial water. They used the technique ofInterferometric Synthetic Aperture Radar with repeat-passesof the RADARSAT satellite to detect areas of vertical dis-placement in upstream areas of Ice Streams C (“Kamb IceStream”) and D (“Bindschadler Ice Stream”) in the SipleCoast region of West Antarctica. The surface height changesmeasured in this study have several features in common witha large number of events identified by other authors since. Themeasured vertical displacements averaged ~0.5 m, they oc-curred within an orbital period of 24 days, and were smoothlyvarying in amplitude over roughly spherical regions 10–20 km across. In the case of the Ice Stream D event, anupstream surface lowering was observed over the same timeperiod as a downstream surface rise representing an approx-imately equivalent change in volume [Gray et al., 2005].Further evidence for subglacial water movement beneath

the Siple Coast ice streams was presented by Fricker et al.[2007] and Fricker and Scambos [2009]. This was obtainedfrom repeat-pass laser altimetry with the ICESat mission to

identify areas of raising and lowering together withdifferencing of MODIS images in which discrete areas ofsurface change are visible. While these features demonstrat-ed the scale and activity of the subglacial water systembeneath the fast flow features of West Antarctica, Winghamet al. [2006] used the radar altimeter on the ERS-2 satelliteto identify similar phenomena occurring beneath the thickinterior ice of the Adventure Subglacial Trench in EastAntarctica. In this case, a single lake was discovered to bedeflating (~3 m drop over a surface area of ~600 km2)upstream along a predicted flow path from several smallerlakes found to be inflating. Unlike the RADARSAT data, thiswork showed lakes draining and filling over a period ofmany months and inferred water transport over a distance ofaround 260 km. Further work has shown that a significantfraction of the water discharged from the upstream lakeduring this event was retained by the downstream lakes andthat a distributed model of subglacial water transport isneeded to explain the observed travel times [Carter et al.,2009].The launch of the ICESat laser altimeter has greatly im-

proved the spatial coverage of Antarctica and has resulted ina large number of new lakes discovered by the effects of theirfilling and draining on the ice sheet surface. Smith et al.[2009] analyzed all ICESat repeat tracks between 2003 and2008 for indications of anomalous surface height changesthat could not be explained by normal glaciological pro-cesses [see Gudmundsson, 2003; Sergienko et al., 2007].They detected new lakes throughout the continent (130 intotal) with concentrations centered under several of the majoroutlet glaciers of both East and West Antarctica. Especiallysignificant among them are two large lakes in East Antarc-tica, the discharge of which was found to have coincided witha 10% increase in the flow speed of the Byrd Glacier, locateddirectly downstream of the lakes [Stearns et al., 2008].As a footnote to this section, Richter et al. [2008] con-

ducted a high-resolution ground-based GPS survey overLake Vostok over the period 2002–2007. Their results indi-cated no surface height change that could be attributed to anychange of the water level within the lake, a conclusion that issupported by 1-year period repeat laser altimetry from theICESat satellite [Shuman et al., 2006].

2.4. Seismic Survey

Unlike other methods described above, the analysis ofreflected pressure waves allows information to be collectedfrom below the surface of subglacial lakes. Longitudinalseismic or “p” waves have the unique ability to penetratelarge distances through ice, rock, and liquid water. Thismeans that by timing reflections from the various subglacial


interfaces, seismic studies can be used to infer the depth ofwater in subglacial lakes and the thickness of any unconsol-idated sedimentary layers at their bed.Seismic investigations, however, are time and labor inten-

sive to carry out and provide only one data point for each“shot” taken. High levels of background noise on the Ant-arctic plateau can also make interpretation of the returnsdifficult. During the 1960s, a seismic survey carried outaround Vostok Station [Kapitsa and Sorochtin, 1965] priorto the drilling of a deep ice core failed to identify thepresence of a subglacial lake. The same data set would laterprovide the first indications of the depth of Lake Vostok afterreexamination in the light of knowledge of the lake’s surfaceextent gained from satellite altimetry [Ridley et al., 1993].The initial data indicated a water depth of around 500 m at asite located a few kilometers from Vostok Station [Kapitsa etal., 1996]. More recent seismic investigations have revealedthe maximum depth of the lake to be over 1100 m [Masolovet al., 2006]. A separate basin, with a small surface area, butup to 680 m deep, has also been found to exist immediatelybelow Vostok Station [Masolov et al., 1999].Lake Vostok was the first, and for many years the only,

subglacial lake to have a direct depth measurement by seis-mic sounding. Recently, however, a subglacial lake situatedvery near the South Pole [Peters et al., 2008] and anothernear the Ellsworth Mountains in West Antarctica [Woodwardet al., 2010] have been the targets of seismic surveys. At theSouth Pole Lake, Peters et al. [2008] used the amplitudevariation with offset technique, which utilizes the observa-tion that seismic reflectivity varies as a function of angle indifferent ways for different subglacial materials. In this way,they were able to identify liquid water beneath the SouthPole independently of the RES data that had originallylocated the lake. In addition, they used traditional seismicprocessing to determine the depth of the water column inSouth Pole Lake to be around 32 m.As well as characterizing the water depth in subglacial

lakes, seismic sounding is currently the only method forinvestigating the physiography of lake beds and particularlyfor determining thicknesses of sediments. At Lake Vostok,several of the available seismic records show multiple reflec-tions from around the bed of the lake; these, however, can beinterpreted in different ways. Reflections spanning a range of0.1–0.5 s from a number of different sites above Lake Vostokare thought to represent between 100 and 350 m of sedimentswith seismic velocity between 1700 and 2100 m s�1 [Masolovet al., 1999, 2001]. However, it has also been suggested thatthese secondary bottom echoes may represent side reflec-tions due to either steep side slopes or high basal roughnessat the bed of the lake and, therefore, that the water layer isin direct contact with the seismic basement across the entire

lake bottom [Masolov et al., 2006; Popov et al., 2006].Filina [2007] tested these hypotheses and determined thatsome of the reflections could be due to a sloping lake bedbut that, nevertheless, the seismic data were best explainedby the presence of a 200- to 300-m thick layer of sediment.In 2007/2008, seismic measurements were made on Ells-

worth Subglacial Lake (hereinafter referred to as Lake Ells-worth) in West Antarctica [Woodward et al., 2010; Ross et al.,this volume]. Based on an assessment of sidewall slopes bor-dering the lake, the water depth was previously hypothesized tobe at least several tens of meters [Siegert et al., 2004]. Theseismic data upheld this hypothesis and recorded a depth ofaround 160 m in long location. This result has been used as thebasis for the forthcoming exploration of Lake Ellsworth(in 2012/2013), detailed by Ross et al. [this volume].

2.5. Characterization of Subglacial Lakes by the Surveyof Gravity Anomalies

Geologists use measurements of the free-air gravitationalanomaly in Antarctica to map density variations below thesurface that are the result of both geological structures and ofthe topography. If the surface and bed topographies of the icesheet over a subglacial lake are known (e.g., from RES sur-vey), then the effect of ice sheet geometry can be removedfrom the free-air gravity anomaly. For a large subglaciallake, such as Lake Vostok, the thickness of the subglacialwater column then dominates the remaining gravity signal[Studinger et al., 2004b].This method has been used to map the bathymetry of Lake

Vostok at a higher spatial resolution than is possible withseismic methods alone [Studinger et al., 2004b]. Two sepa-rate basins were identified in this way, a larger and deepersouthern basin separated by a 40-km-wide bedrock ridgefrom a smaller and shallower northern one. Roy et al.[2005] took the next step by using gravity data to producea complete model of Lake Vostok including both watercolumn and sediment thickness. Both these studies usedaerogeophysical data collected during the 2001/2002 Ant-arctic summer [Holt et al., 2006b]. Neither model, however,corresponded precisely with the water and sediment depthsindicated by more recent seismic studies [e.g., Masolov etal., 2006]. Further work with the original gravity data com-bined with recent seismic studies has produced a refinedbathymetry/sediment model for Lake Vostok [Filina et al.,2008]. This indicates a deep (~1200 m) maximum waterdepth underlain by ~300 m of unconsolidated sediments inthe southern basin and shallower water (~250 m) overlying athicker sedimentary layer (~350 m) in the northern basin.The only other subglacial lake to have been characterized

at any level of detail using gravity measurements is Lake


Concordia. A University of Texas/Support Office for Aero-geophysics (SOAR) airborne survey covered the region ofthe lake in 1999/2000 and collected gravity data [Studingeret al., 2004a; Filina et al., 2006]. Inversion modeling ofgravity data from the six survey lines crossing the lake wasthen undertaken by both Tikku et al. [2005] and Filina[2007]. Gravity results, along with simultaneously acquiredRES data, were then used to determine that the lake has alarge surface area of between 600 and 800 km2, but a rela-tively shallow average water column thickness of ~59 m(maximum depth 126 m), and consequently a volume of just70 km3 [Filina, 2007; Thoma et al., 2009]. Since the lake isrelatively shallow, any underlying sedimentary layer has notbeen resolved in the gravity data.The inherent noisiness of airborne gravimetry (due to the

sensor being subjected to nongravitational accelerations re-sulting from aircraft motion) means that extensive low-passfiltering is required to extract that part of the recorded signal,which is of geological origin [Childers et al., 1999]. Analysisof repeat lines and crossovers has determined an accuracy of1–2 mGal for the surveys of both Lake Vostok and LakeConcordia [Holt et al., 2006b]. This, combined with thethickness of the overlying ice (necessitating that data beacquired at an effective altitude of more than 4 km above theice base target), currently imposes a limit of around 8–10 kmon the resolution of aerogravity surveys over central Antarc-tica [Holt et al., 2006b]. Modern aerogravity equipment, asused in commercial exploration, has the potential to providesubmilligal accuracy to future surveys of the Antarctic sub-glacial environment [Studinger et al., 2008]; however, as yet,no subglacial lake studies have been published with this kindof resolution. Further errors can potentially be introducedinto lake bathymetries determined solely by gravity surveydue to the unknown nature of the underlying geology. Un-known or uncertain spatial variation in rock density can beindistinguishable from changes in lake volume in gravitysurveys; hence, the use of seismic data is desirable in verify-ing gravity inversion models [e.g., Filina et al., 2008].

3. GEOGRAPHICAL DISTRIBUTION, DIMENSIONS,AND RELATION TO TOPOGRAPHIC AND

GLACIOLOGICAL SETTING

3.1. Inventory of Known Subglacial Lakes

The first published inventory of subglacial lakes is thework of Oswald and Robin [1973]. At this time, just 17 lakeshad been discovered by the SPRI-NSF-TUD RES program,14 of which lay beneath the Dome C area of East Antarctica,a region which would later become known as the “AntarcticLake District.”

By the time another systematic catalog of all known lakeswas produced [Siegert et al., 1996], the total number hadincreased to 77. The large majority of these lakes were still tobe found under the thick ice of central East Antarctica withtwo clusters in particular, beneath the Dome C and Ridge Bareas, accounting for 77% of the total number [Siegert et al.,1996]. Lake Vostok was included in this inventory. Its largevolume, at least two orders of magnitude greater than anyother lake, would dominate calculations of water storagebeneath the ice sheet [Dowdeswell and Siegert, 1999].The most recent inventory of subglacial lakes [Siegert et

al., 2005], while still only incorporating lakes identified byRES, drew upon a larger number of studies including, mostsignificantly, the Italian survey of the Dome C area [Tabaccoet al., 2003], the Russian airborne RES of the Dome A andDome F regions [Popov and Masolov, 2003], and the U.S.surveys of the Vostok and Wilkes Basin areas [Studinger etal., 2003a, 2004a]. The tally for this inventory reached 145separate lakes, even after six individual reflectors from theprevious inventory had been reclassified as parts of largerlakes [Siegert and Ridley, 1998a].Since the publication of this inventory, a number of recent

studies have further added to our knowledge of subglaciallakes:

1. Popov and Masolov [2007] reported 29 new lakesidentified during ground-based RES work aimed at mappingthe shoreline of Lake Vostok. Most of these are small lakes(0.5- to 10-km-long reflectors) located within valleys severalhundred meters deep. Two further small lakes (4 and26.5 km2) were detected during an overland traverse betweenMirny and Vostok [Popov and Masolov, 2007].

2. A systematic categorization of RES reflections fromthe 1998–2001 SOAR campaigns covering the Pensacola-South Pole transect, the Wilkes Basin Dome C transect, anda survey of the Lake Vostok area by Carter et al. [2007]resulted in the identification of 22 new “definite” lakes anda further 58 classed by the authors as “dim lakes” (seesection 2.1.4). This reanalysis also determined that severalfeatures previously thought to be separate lakes, in fact,belonged to the same water bodies, thereby reducing thetotal of Siegert et al. [2005] by 5. A complete list of all“lake-type” reflectors identified in this study is given byBlankenship et al. [2009].

3. As mentioned in section 2.2, Bell et al. [2006] contribut-ed revised dimensions for two large lakes at 90-E (~2000 km2)and at Sovetskaya (~1600 km2). Similarly, four lakes withsurface areas 1500–4500 km2 have been identified at the headof the Recovery Ice Stream [Bell et al., 2007].

4. Five new lakes were reported by Cafarella et al. [2006]as a result of Italian RES work during 2003. Four of these aresmall (1- to 3-km-long reflectors) and are located within the


Belgica Subglacial Highlands, and one is larger (~10 km)and situated in the southern Aurora Basin.

5. Finally, several active subglacial lakes have been iden-tified in individual studies throughout East and West Antarc-

tica [Gray et al., 2005 (2); Wingham et al., 2006 (4); Frickeret al., 2007 (7); Stearns et al., 2008 (2)]. A systematiccatalog of all lakes that were active (and crossed by morethan one ICESat track) between 2003 and 2008 has been

Figure 2.Map of subglacial Antarctica showing the locations of all lakes known from radio echo sounding and all sites ofsurface height change consistent with lake activity detected by satellite. Larger lakes are shown in outline; smaller lakesand those of unknown surface area are indicated by triangular (RES) and circular markers (satellite). Lakes included in theprevious inventory of Siegert et al. [2005] are shown in black; lakes discovered since that time are shaded according to thepublication in which they were first identified.


produced by Smith et al. [2009]. This work added 113previously unrecorded lakes.The total number of lakes described in the literature as of

October 2009, therefore, stands at 387 (see Figure 2).

3.2. The Distribution of Subglacial Lakes

3.2.1. Distribution with ice thickness and with distancefrom subglacial flow divide. The direction of subglacialwater flow at any point on the ice sheet can be shown todepend only upon the slopes of the ice surface and of thebedrock [Shreve, 1972]. Thus, given a digital elevationmodel of the ice sheet surface [e.g., Bamber et al., 2009]and of the bed [Lythe et al., 2001], flow lines can be calcu-lated for subglacial water throughout the ice sheet [e.g.,Wright et al., 2008]. The slope of the ice surface has anorder of magnitude greater effect on the direction of subgla-cial water flow than does the topography of the bedrock[Shreve, 1972]. The flow lines therefore, largely follow thedirection of ice flow. As a consequence of this, an equivalentsubglacial watershed is situated approximately beneath theice divide.The upstream distance from a lake along the flow path to

the subglacial flow divide is a useful parameter in charac-terizing the distribution of subglacial lakes. Figure 3 showshistograms of this distance to the divide for (a) all lakes, (b)lakes detected by RES fieldwork, and (c) lakes detected bysatellite measurements of height change. Figure 3a clearlyshows that the densities of all known lakes increase nearlyexponentially with approach to the flow divide. Histogramsfor lakes detected by RES and for active lakes detectedfrom space, however, produce very different results. Thetotal number of lakes from RES surveys (256) is greaterthan the total for active lakes (128) and the distribution ismore heavily skewed; hence, the pattern for radar lakesdominates that for all lakes. (Four lakes beneath the headof the Recovery Ice Stream are known by their surfaceexpression only [see Bell et al., 2007]. One lake (LakeMercer) is identified by both surface height change and inRES records.) The active lakes, so far, reported do notshow a strong relationship to upstream distance to the flowdivide. The modal value for this distance is 750 km, whichfor lakes detected by RES falls within the upper 5% tail ofthe distribution. This demonstrates that the method of lake

detection, or perhaps the type of lake detected, has asignificant effect on the distribution of lakes with regard tothe flow divides.

Figure 3. (opposite) Distribution of subglacial lakes in terms ofdistance along the flow line to a major ice divide. The histogramsshow (a) all known subglacial lakes, (b) lakes identified by theirRES reflection, and (c) lakes identified by satellite measurements ofvertical surface movement.


Documents

Geophysical Monograph Series · Figures, tables and short excerpts may be reprinted in scientific books and journals if the source is properly cited. Authorization to photocopy items