The University of Maine DigitalCommons@UMaine
Earth Science Faculty Scholarship Earth Sciences
1995 Paleoglaciologys Grand Unsolved Problem Mikhail G. Grosswald
Terence J. Hughes University of Maine - Main, [email protected]
Follow this and additional works at: https://digitalcommons.library.umaine.edu/ers_facpub Part of the Earth Sciences Commons
Repository Citation Grosswald, Mikhail G. and Hughes, Terence J., "Paleoglaciologys Grand Unsolved Problem" (1995). Earth Science Faculty Scholarship. 66. https://digitalcommons.library.umaine.edu/ers_facpub/66
This Article is brought to you for free and open access by DigitalCommons@UMaine. It has been accepted for inclusion in Earth Science Faculty Scholarship by an authorized administrator of DigitalCommons@UMaine. For more information, please contact [email protected]. Journa/ ofG/aci% g)', \'01. 41, No, 138, 1995
Paleoglaciology's grand unsolved problenl
MIKHAlL G, GROSS\\".\LD
, , . lllslilule ojCeograj)/O', Russiall .I(({den�)' ofScil'lll"es, 29 SlarOI/lOllfil�J' Slrel'l, 1090/7 llo lcO/l ' RII.Isia
TERE:\'CE J. HUGHI�S lllslilllle jor Q.JIalerllal)" SllIdies alld 1111' De/Jarlmenl of Ge%giw/ ScieJICl's. L'lIiz'eI"si{J' of .Ilaille, OrOIlO, .Ilaille (4).69, C,S,.'I,
ABSTRACT, The palcoglaciological concept that during the Plcistocene glacial
hemi-cycles a super-large, structurally co mplex ice sheet de\"Cloped in the i\rctic and beha\"Cd as a single dynamic system, as the Antarctic ice sheet does today, has not yet been subjected to concerted studies designed to test the predictions of" this concept. Yct, it ma\' hold the keys to solutions of major problrms of' pa\eoglaciology, to understanding climate and sea-Ic\"C1 changes, The Russian Arctic is the Irast-kno\\'n region exposed to paleoglaciation by a h\'pothetical Arctic ice sheet but 110\\' it is Illore open to testing the concept. Implementation or these tests is a challenging task. as the
region is extensi\"C and the anlilable data are contro\'Crsial. \\'ell-planned and co ordinatcd field projects arc needed toe\;\\', as \\"ell as broad discussion of the kno\\"n
' e\ idence, existing interpretations and nc\\' field results, Here \\"C present the knO\\"Il
e\'idence for p aleoglaciation of' the Russian Arctic continental shelf' and reconstruct possible marine icc sheets that could han' produced that e\'idcnce,
INTRODUCTION colleaguc, G, H. Denton, \\c argued that this Il\ pothet ical f()rmcr ice sheet, \\"hich \\T called the Arctic ice sheet. \\'riting in ,\{dllre. \\'eertman (1976 stated "O\"Cr the \\'as an ,\rctic counterpart to the pn'sent-day Antarctic past t\\'O decades, our understanding of the beha\'ior of' ice shect, because it consistcd of' terrestrial ice domes glaciers, ice sheh-es. and ice sheets has progressed and grounded on land abo\"(' sea InTI and marine ice domes
increased \"Cry nice! y, \\' e kno\\' enough no\\" ro recognize grounded on con ti nental shekes belo\\' sea IC\"CI \\'i th
a grand glaciological problem that remains to be soh-ecl. marine ice domes drained by ice streams that supplied an The \\'est Antarctic lee Sheet is this problem," He then ice shelf that floated on the deep part of' the Ar ctic Ocean pointed out that it is a "marine" ice sheet grounded \\"ell and caked into the North Atlantic (Hughes and others, belo\\' sea 1r\"CI, so its stabilit\' depends on a stable sea 1977), \\'e f'urthcr suggested that the hypothetical Arctic 1e\"C1. Yet, sea le\"C1 has changed dramaticall\' in the last ice sheet interacted \'igorously \\'ith the ocean and 20000 \'ears, He asked "Ho\\' then did this ice sheet fmm? beha\"(:'d as a single dynamic system, as the Antarctic \\'hy does it remain in existence") Is it gro\\'ing or ice sheet does LOday, disintegrating at the present time?" He noted that the As \\'ith glaciology's grand unsoh-ed problem in the ' \\ est Antarctic ice sheet is drained iw ice streams, \\'hich Antarctic. the grand unsoh"Cd problem of' Arctic arc f�lst currents orice similar to great riwrs that drain the paleoglaeiology translatcs into such questions as: "Did other continents. Ice streams supply the large floating ice an Arctic ice sheet de\'clop during Quatcrnary glaciation shel\"('s that f'ringe \\'est Antarctica and, like surging C\TIeS? If' it did, \\'hat \\'as its specific extent and structure?
mountain glaciers, ice streams may also be unstable, In particular, did it include grounded marine ice domes Se\"l'J"al large-scale glaciologieal research programs are and floating ice sheh-es? Ho\\' long did it remain in 110\\' addressing all of' these questions, existence? \\'hat \\'as its disintegration histoJ"\'? Obtaining In the nearly t\\'O decades since \\'ccrtman ( 19 76 ) the ans\\'ers and thus testing the hypothesis is a difficult identified the marine \\'est Antarctic ice sheet as task because c\'idence [()I" the Arctic ice sheet is scattered "glaciology's grand unsoked prohlem·'. paleoglaciology O\"('r a huge, hardly accessii)le area and it is of'ten has also adnll1ced \Try nicely and \\'e kno\\' enough no\\' obscured by aCli\"C periglacial processes, or is ("oncea led
to recognize a grand paleoglaciological problem that under permafi'ost-bound ice-rich sand and silt. or is
remains to be soh-ed, That problem is the possibility that submerged by the sea on continental sheh-cs, Thc a largek marine ice sheet formed, gre\\' and collapsed in l1\'pothesis of' the Arctic ice sheet has not been subjected
the Arctic du ring Quaternary glaciatioll cycles (:\[ercer, to concerted studies pointedly designed to test its
19 70), As an unsokcd paleoglaciological problem, it is predictions, although the critical region [or this test, the
remarkably parallel to the modern Antarctic glaciological Russian Arctic, is no\\' open to international studies, and problem formula ted by \\' eertman, Along \\'i th our Russian components of' an Arctic ice sheet may hold keys
313 Journal oJ Glaciology to understanding global climatic and oceanographic changes. A sense of urgency in addressing paleoglaciology's grand unsolved problem exists, in \'iew of the contention by Miller and de Vernal ( 1992) that greenhouse warming in the decades ahead may initiate ice-sheet growth in High Arctic latitudes. They argued that greenhouse warming may duplicate the warmer winters and cooler summers in these latitudes that accompanied the onset of the last glaciation cycle. In particular, we suggest sea ice might not form in the Norvvegian and Greenland Seas during warmer Arctic winters, thereby allowing greater precipitation o\'er the permanent sea ice that does not melt during cooler Arctic summers. This would upset the precarious present-day mass balance in the High Arctic, so that sea ice would ground on shallow Arctic continental shelves and snowfields would grow on Arctic islands and uplands. This could impound the great rivers that drain into the Arctic from Eurasia and North America, creating a vast "vVhite Hole" within which precipitation could accumulate but not escape. The area of this "White Hole" is shown in Figure I which is Fig. 2. The proposed A rctic ice sheet during a QJwlenzmy comparable to the maximum areal extent of Northern glacial maximum. Presenl-day and glacial maximum Hemisphere glaciation during the last glaciation cycle. shorelines are dotted and solid lines, respective01. Black including ice sheets in Eurasia and North America, and areas are ice-dammed lakes a1/d glacier-Jed lakes at the ice shelves or sea ice o\'er the Arctic Ocean. The ""Vhite glacial maximum. For the ice sheet, solid lines are the Hole" would account for the drop in sea level at the maximum extenl oJ glaciation and dashed lines are ice-sheIJ beginning of the cycle that was as rapid as the rIse 111 sea grounding lines. level at its termination.
lakes that drained into either the Sea of Okhotsk or the Mediterranean Sea, and that would innuence the present day arid climate of Central Asia. Figure 2 is based on the glacial geological, geomorphological and periglacial features reported in Russia by Grosswald (1993) and located in Figure 3, on a hypothetical marine glaciation of cen tral Beringia pro posed by H ughes and H ughes (1994), and on a glaciological interpretation of oriented lakes on the Arctic coastal plains of Siberia and Alaska. This evidence will by summarized under the headings: ice dammed lakes, terrestrial ice sheets, marine ice sheets, ice streams and floating ice shelves. Figure 4 reproduces the glacial geological and geomorphological fe atures on Arctic continental sheh'es and adjacent mainlands shown in Figure 3 and adds similar features in northeast Siberia and northwest Alaska, all of which were used to construct flowlines fo r the marine ice-sheet reconstructions in Figures 5 and 6.
ICE-DAMMED LAKES
Fig. 1. The" White Hole" at incejJtioll oJ the Arctic ice An Arctic ice sheet would consist of an Eurasian ice sheet sheet. Present-day alld glacial maximum shorelines are that was joined to the ice sheets of Greenland and North
solid and dotted lines, resjJectively. The heavy solid line America by an ice shelf noating on the Arctic Ocean. The encloses the Arctic Ice sheet at inception. The heavy dashed most conclusive evidence for an Eurasian ice sheet would line encloses watersheds of rivers flowing toward Ihe Arctic be impoundment of the large Eurasian ri\"Crs that now ice sheet. Light solid lines are limits oJ gl aciation at a now into the Arctic Ocean, thereby creating Eurasian ice Qyalenza7]1 glacial maximum. dammed lakes during the last glaciation. These lakes constitute additional compelling evidence for the Eur asian ice sheet and their history constitutes a large-scale Figure 2 shows the hypothetical Arctic ice sheet just paleoglaciological problem in itself. pnor to termination. Note the extensive ice-dammed That a continuous Eurasian ice sheet, if it existed, had
314 Grosswald alld Hug/les: Paleoglaciology' 5 grand unsolved jJ1"oblem
(Se" Slf
�, L....�,o �H Lm, �:CJ " L..g " [L;1J.�" D .. o 100 200 lOO 400 5OO�",
120·
Fig. 3. Evidence Jor glaciatioll oJ Arclic Russia �)I a marine ice sheel. S)lI1bols on Ihe mal) are: /. Brillk of the colltillental shelf; 2. Direclions alld relative size oJ submarine troughs (channels); 3. BOllndmJ' of Ihe Scalldil/aviall and ft"am ice sheets during the lasl glacial maximum; 4. Inferred direclions oJ ice flow; 5. Ouler limit oJ the Scandil/aviall al/d It-ara ice sheets during Ihe 1051 glacial l1)(lIimulIl (moraine belt A): 6. Ice limits. lale glacial alld Ilolocme slages (moraines 13, C, D, E and illlermillml); 7. Hill-al/d-hole pairs: 8. Ice-shoved folding ill illicol/solidated sedilllmls; 9. Glacial grooves and strialiolls; 10. Drumlills, drtlllllilloids, fluling al/d olher lineations; 11. Glacial breaches (Ihrough uaLlfD's); 12. ,Ilcijor spillways. mellwaler-oue/flow chal/llels; 13. Ice-dammed lakes-Iheir lIl a.\iIllUm erlfllt al lhe Lasl Glacial Jllaximlllll; 14. " 1""edollla" areas (a([umlllal iOlls oJ ire-rich sill alld sal/d): J 5. " .. llas·· Ut/It')'.1 (jJarallel l'a{{�l'S crealed b.-J' meLLillg of groulld ice); 16. DiI'erlion oJ 10l/g aus oJ" orimted lakes". to impound the northlVard-flowing ri\'Crs of' Eurasia and Vilyui lake of East Siberia, as well as of the inter cause formation of' huge proglacial meltwater lakes was connected basins of the A ritl , Caspian and Black "Seas". understood as ear'" as the mid-nineteenth century. The The catchmen t area of the system amounted 6 2 to lakes were first postulated and much later confirmed on 21 x 10 km the basis of scattered pieces of geological e\'iclen The eastern system \\ 'as ce. impoun ded by the East Recently, a late \\' Siberian part of the Eurasian ice sheet and was eichselian palaeo-Iake reconstruction was published fo r European Russia (K\' represented not so much by lakes but mostly by asov, 1979) and fo r lVest Siberia (Volko\' yedomas, which are thick accumulations of ice-rich silt and others, 1978; Arkhipo\' and others. 1980; Grosswald, 1980) . A sketch of the melt and fine sa nd. Y cdomas fo rmed in ice-dammed proglacial water-drainage system at the Last Glacial Maximum, deltas under very cold and dry conditions, when each showing its status before erosional deepening of spillways, summer's meltwater turned into ice during the fo llowing is included in Figure 3. This picture is based on all the winter. Only a small proportion of the yedomas was a\ailable information on the occurrence of glacio fo rmed during the Last Glacial Maximum. Most of it lacustrine sediments and shorelines studied and, in some fo rmed during the Late Glacial and Holocene ice retreat. instances, dated in the field. The rest was reconstructed The ),eclomas arc still preserved on the Arctic coastal by geomorphologic methods, with particular attention to lowlands, where they apparently bury and conceal nearly the geography of flat expanses of paleo-Iake terraces, the whole bulk of ice-marginal land fo rms. Urslromlriler and spillways, as well as to spacing of yedoma The lakes experience draslic changes of' their extents, accumulations. As a result, two transcontinental melt outlines and levels, which were caused by the ice-sheet \\ 'ater-drainage systems, western and eastern, were retreats and re-advances, by isostalic crustal rebound and visualized. by erosional deepening or sediment infilling of the The western system was the largest. Tt consisted of the spill\\· ays. Today, we can only speculate about these Vychegda-Pechoran lakes of the northern Russian Plai n, changes. All wc know is that, during the Last Glacial the Mansi and Yenisei lakes of west Siberia (of' which the ylaximum, the westernsystem discharged meltwater into �1ansi was the world's largest ice-dammed lake) , Lena the eastern l\!Iedi terranean Sea and the eastern system
315 Journal of� ClJ aClology.
"I'r'l� ' -,.: I , ! ! \� X
< -'i >
: +
316 ><
x
x
S) c a '" '" '" ::;; t;;: ::: ;:;.. � (ic. :::- � 't + 60 :::, ;;;- \70 C> (J� ...,:::, c' c �
'" cr.s Fig. 5. FlOll'lines nnd elem/ioll (olllollr.1 al 200117 illler/'{//.I :2 ;:;.. Jor llie Hurasiall /IIarille ice dOilies of all IO'/Jollielical �'Jr(lic ice slieel, based all llie el'idmre ill F(gllre J- whm /Jreselll-d(�)' � '" SIIOU' lilies are Lowered �JI 1000 m ill (a) the westem '" "'-
F;/Irasiall Arctic, (b) tfle rell/ra! F;/Irasiall :Iretir alld ( () tfle � b c '-'" eastem Eumsiall Arc/ic. � ...... � w '--+-l '" '" co .A /'� /I�--'- , - \ ...... LA\ '-....(L{rv/... \'-./ /� , '::: " � � CJ � " C; �
a
c
Fig. 6. Flowlines and elevation contoun at 2001ll intervals for the Eurasian marine ice domes of an hypothetical Arctic ice sheet, based on the evidence ill Figure 4 when fHesent-day snow lines are lowered by 1200 m in (a) the westem Eurasian Arctic, (b) the central Eurasiall Arctic and (c) the eastern Eurasian Arctic. The Izacllllred lines ill (e) show the extent of mountain glaciatioll ill Siberia and Alaska that merged with the marine Cllllkehi ice dome. b GrosswaLd and HlIghes: PaLeoglacioLogfs grand unsolved jJrobLem discharged into the Sea of Okhotsk. At 14-13 kaBP, the high elevations near moisture sources. Computer simul western system turned latitudinal, Oowing into the ations based on these hypotheses ha\ 'e produced a southern North Sea. During the warm Bolling-Allerod Cordilleran ice sheet that formed sooner, grew bigger interval, when the Eurasian ice sheet was breached by an and lasted longer than the Laurentide ice sheet (Budd ice-free channel extending along the Bear Island Trough and Smith, 1981). (Bjornoyrenna in Figure 3) and the White Sea gorge, the The second problem is that highland glaciation on system's water was dumped into the northern Norwegian BafTin Island, in Scandinavia, in Svalbard, in Franz Josef Sea. The Younger Dryas cooling resulted in restoration of Land and in Novaya Zemlya would not be extensi\'e the system, which then probably resumed its 0011' \· enough to lower sea In'el sufTiciently to allow these ia the
Caspian Sea into the Mediterranean Sea. The pre-Boreal glaciers to cross the exposed Ooors of Foxe Basin, the warming entailed a new meltwater outburst (t'om the Baltic Sea and the Barents Sea, and then transgress onto northern Russian Plain and West Siberia into the the mainlands of Canada and Russia to become the Norwegian Sea. Laurentide and Eurasian ice sheets. Instead , the glaciers These re-organizations in proglacial lake systems would calve into these seas and never reach the mainland. resulted in tremendous en\'ironmental changes. They The numerous modeling studies that accomplish this account fo r modern distributions of fish and other island-hopping to the mainland do so by excluding ichthyo-faunas, some mammals (like seals), for the calving dynamics in their models, thereby ignoring the largest peat deposits and Arctic yedomas, and for many dominant ablation mechanism; e.g. see Budd and Smith fea tures of Eurasian geomorphology. Perhaps they can (1981). Only by invoking marine ice sheets and the also explain, by outbursts of the Eurasian ice-dammed marine ice-transgression hypothesis (MITH) can these lakes, the massive meltwater spikes established in the vast late Quaternary ice sheets of North America and 98 Atlantic Ocean, which are now attributed to outbursts of Eurasia be explained ( De nto n and Hughes, 1 1 h; the Agassiz paleo-Iake in North America ( Fairbanks, H ughes, 1987; G rosswald. 1988). 1989).
MARINE ICE SHEETS TERRESTRIAL ICE SHEETS The first clues that a number of marine ice sheets existed
Most of the terrestrial ice sheets 111 the .\Torthern on the Arctic continental shelves of Eurasia and North Hemisphere are relatively well documented ( Flint, 1971; America came (i'om such evidence as ice-motion proxies Denton and Hughes, 1981 b; Lundqvist, 1986). This is and dated raised beaches on the islands of the north true for the Laurentide ice sheet and the Cordilleran ice western Barents Sea and Arctic Canada. As a result, a sheet of Nort h America and for the Scandinavian ice sheet marine Barents ice sheet in the Barents Sea and a marine of Eurasia, but it does not hold fo r terrestrial ice sheets in lnnuitian ice sheet on Canada's Queen Elizabeth Islands Russia. For instance, the ice caps of Putorana Plateau, were postulated (Schytl and others, 1968; Blake, 1970). Anabar Plateau and Taimyr Peninsula, all located in The usual glaciological processes that produce prima facie cenrral Siberia, are sometimes described as independent glacial land fo rms and glacial deposits seem not to have formations (Faustova and VeJichko, 1992), but they been ac tive in many areas of the High Arctic, probably could in fact be terrestrial parts of the major marine Kara because the ice was frozen to its bed O\' Cr the islands and ice dome, along II'ith the glacier complex of the Ural parts of the sea Ooor there. As a consequence, glacial land Mountains (Arkhipo\ ' fo rms and deposits in such areas as the western Queen and others, 1980; Kind and Leonov, 1982). Similarly, a terrestrial glacier complex Elizabeth Islands or Sel·ernaya Zt'mlya, an island group el1\ 'isaged by Hopkins ( 1972) and by Bespaly and north of the Taimyr Peninsula (see Fig. 3), would be GlushkO\'a (A rkhipov and others, 1986) on the Chukchi confined to the inter-island channels occupied by ice Peninsula may have been part of a much larger marine streams sliding on thall ed deforming marine sediments. ice sheet centered on the adjacent Chukchi Sea (Gross In these conditions, particular attention should be paid to wald, 1988; Hughes and Hughes. 1994). The available the clues prO\'icled by geophysical studies and computer data on the chronology of the Putorana ice cap and the modeling, such as that by Clark (1980), who provided glaciers of the Taimyr and Chukchi Peninsulas are considerable support for the Innuitian ice sheet by meager, indirect and unreliable, permitting speculations showing that an ice dome o\'er the Qu een Elizabeth as to whether the glacial land fo rms there represent the Islands was a necessary prereq u isi te for enabli ng an Last Glacial NIaximum or one of the Late Glacial events. inverse computer model of glacioisostatie rebound to Accordingly, the maximum extent of the last glaciation in accoun t for the raised beaches mapped and date cl by these areas is still unclear. Blake (1970). Two major problems arise if Quaternary glaciation During the 1980s, Scandinavian geologists confirmed the cycles in the Northern Hemisphere are restricted to just existence of a Barents ice sheet at the Last Glacial terrestrial ice sheets. The first problem is that terrestrial �laximum and presented strong e\'idence that it was
ice sheets respond only to changes in snow-line elevations. grounded on the Ooor of the western Barents Sea and Therefore, terrestrial ice sheets can nucleate only in coalesced with the Scandinavian ice sheet. Field data on mountains, according to the highland-origin, windward which they based this conclusion consisted of isostatically (1971), 1981) growth hypothesis of FI i nL or on plateaux, raised beaches (Salvigsen, and the sedimen tary record according to the instantaneous glacierization hypothesis of the sea Ooor (Elverhoi and Solheim, 1983; Von'en and of Ives (1957), with glaciation being most extensive at KrisLO(fersen, 1986; Ekerhoi and others, 1992; Szettem and
319 Journal 0/ Glaciology others, 1992; \'orren, 1992) , Subsequently, Russian marinc geomorphological evidence, The other emphasizes dat geologists also presented strong evidence fo r thc existence of' ing results obtained from problematic materials, A this ice shcet in the eastcrn Barents Sea (Gataulli n and parallel situation existed in the glacial geology of Arctic others, 1 993) , Addi tional compelling e\'idence in thc eastern Canada during the 1970s and also resulted in Barents Sea, including data on a continuous sea-Ooor "maximum" and "minimum" models of Laurentide lodgcment-till cover, hummocky and ridgy bottom topog glaciation ( � [ ayewski and others, 1981) , These differ raphy and the widesprcad occurrence of glaciotectonic ences eventually resulted in the two irreconcilable models fe atures, has bcen unco\'cred in the coursc of oil and gas of "continuous" and "restricted" glaciations in the prospecting personal communication to .\1.G, G, fr om Eurasian Arctic as well. Although we fa \'or the first N, A, Polyako\'a) , approach, \\'e realize that the only possible way of It was also shown (Grosswald, 1980) that the resoh-ing the con trovers), is by i m plemen tation of a terrestrial glacial geological record, notably the south special program targeted at dating and rc-dating a facing end moraines of northeastern European Russia number of carefully selected key sections of the Siberian ( La\To\', 1977), \\'(st Siberia (Arkhipo\' and others, 1980) Qu aternary by means of the A�IS IIC method, surface and the Taimyr Peninsula ( K ind and LeonO\', 1982) . all exposure dating with cosmogenic nuclides and other shown in Figure 3, can be explained only if marine ice ach'anced techniques, It is noteworthy that, when these domes occupied the Barents and Kara Seas, and trans techniques \\' (re applied to Arctic Canada, it was the gressed landward from there onto the Eurasian mainland, maximum model of late Wisconsin glaciation that Later, a re\'ised model of the last ice sheet grounded on pre\'ailed, the Barents Kara continental shelf \\ as ad\'anced �[ost recently, geological signatures of a former ( Grosswald, 1988), This model was based upon new marine ice sheet werc also unco\'ered in areas of the evidence uncovered by Lavrov (Arslano\' and others, Laptev, East Siberian and Chukchi Seas, At first, they 1987), Andreyeva and Isayeva (1988) and Gonchari\ \\'ere confined to late \Veichselian ice-shoved fe atures on ( 1986) that pointed to a single major ice dome with its the New Siberian Islands and in the Tiksi Harbour area, ice-spreading center located in the south\\'estern Kara Sea just cast of the Lena Ri\'cr delta (see Figs 3 and 4), These (sec Fig, 3), Ice spreading fr om this Kara ice dome O\'er glacial tectonics permi tied reconstruction of the first flo\\' rode NO\'aya Zemlya in an east-to-west direction, band of a hypothetical east Siberian ice dome centered producing the glacial through valleys shown in Figure just nort h of the New Siberian Islands (Grosswald, 1988; 3, and ill\'aded the Barents Sea from the cast as confirmed Grossvvald and others, 1992) , Today, the supporti ng by the geological e\'idence ofGataullin and others ( 1993 ) , evidence also includes ( I ) upper and middle Pleisrocene
Kara ice also encroached upon large al�eas of European till sheets or the Vankarem Lowland ( Laukhin and others, Russia, west Siberia and central Siberia in directions 1989); (2) glacial en-atics on Cape Heart Rock; (3) north shown by the glacial geology in Figure 3, to-south oriented fj ords and U-shaped through \'alleys of According to the model based both on the above the Chukchi Peninsula; (4) traces of southerly ice Oow geological data and the theoretical conclusion by \\' eert through Bering Strait, all located in Figure 4 ( unpub man (1974) that marine ice sheets expand until they lished data of '\I.G, Grosswald and T.J, H ughes) ; and become aOoat, the Kara ice dome reached the outer edge (5) post-Sangamon en-a tics and ice-shO\'ed fe atures on St of' the Ba ren ts-K ara con ti n en tal shel f; it covered Lawrence Island (Brigham-Grette and others, 1992; 6 6 x I 0 km� and was the largest dome of the Eurasian Heiser and others, 1992 ) , This e\'idence implies that an ice sheet. The ice dome was remarkably stable and long ice sheet ach"anced southwards fi-om the Chukchi Sea, as li\'ed, and its melting history was prolonged and stepwise; well as fr om the East Siberian Sea, Also. modern specifically. the ice-dome retreat was not completed observations by Grosswald (unpublished) make clear before the mid-Holocene and was punctuated by the that \I\Trangel lsland, which lies between the East Siberian Younger Dryas and Boreal time re-ad\'ances of its ice and Chukchi Seas, is characterized by strikingly glacial margin (Grosswald, 1993) , According to a computer geomorphology which can be described as Lappland simulation of the last glaciation cycle by Fastook and style; that is, glacial geology produced beneath or near ice Hughes (1991), the Kara ice dome sUI'\'i\'ed longer than domes, as diagnosed by Kleman (1994) and Kleman and any of' the other major ice domes. Borgstrom (1994) fo r northern Sweden, Also, the north By con trast, Astakhov (1992) argued that the Kara ice ern lowland of Wrangel Island is blanketed by marine dome did not exist during the Last Glacial Maximum, sediments that could have been deposited prior to which seems to be consistent with "old" radiocarbon isostatic rebound after the ice dome collapsed and dates fro m surficial, not glacially disturbed, sediments of possibly prior to the Last Glacial Maximum, northern West Siberia, Howe\'er, Astakhov's model We expect that fu rther progress in establishing the disregards the evidence for recent ice flow directed extent of marine ice sheets in the Arctic will be reached by out\\"ards fr om the Kara Sea cen rer sho\\'n in Figure 3 mapping "oriented lakes" on the Arctic coastal lowlands (Gross\\'ald. 1993, 199+) , Grosswald. in his turn, chose to in northeastern Siberia and northern Alaska, and ignore the "old" dates, considering them spurious and searching fo r buried glacial land fo rms, These lakes and misleading as they appeared to have been obtained from associated elongated and parallel features are remarkably materials which had been recycled, mixed and contam well oriented in sub-mer idional directions (Figs 3 and 4), inated by older organics, The pre\'ailing \'iew is that the lake orientations resulted Epitomized in this contro\'ersy are two different from thermokarst processes operating under the strong approaches to paleoglaciological reconstructions fo r impact of prevailing northeasterly winds (Rex, 1961; nort hern Eurasia, One emphasizes geological and Carson and Hussey, 1962), This hypothesis seems to be
320 Grossl£'ald alld HlIghes: Paleoglariolog/s gralld IIl/jolted jJloblelll supported by hydrodynamic theory, mathematical mod seq uences to be a conseq uence of glaciological acti\'i t y els and field experiments, Ho\\'e\ n, the geomorphology of" associated with former marine ice domes surrounding the the lakes suggests that the wind-affected thermokarst deep Arctic Basin, namely by ice scouring the continental processes con tri bute not to orientin g the lakes but rather shell-cs and dumping the resulring debris into the basin? to obscuring their oricntation, Besides, the "wind" hypothesis is unable to accou n t [o r both longitudinal and tranS\TrSe alignments of the lakes and fo r the other ICE STREAMS linear features comprising the geomorphological com plexes of the Arctic coastal lowlands, So fa r, all attempts Figures 3 and 4- identify the sites of major ice streams thal to explain tht' complexes. not just the lakt's alone, hm't' drained marine ice domes on the Arctic continental shelf" resu ltt'd in a non-realistic hypothesis of "structural of Eurasia and Alaska, The larger ice streams produced control", Instead, liT propose that the linear complexes, saddles on rhe ice di\'ide by the down-draw mechanism, of" which oriented lakes arc just one manif'cstation, thereby lea\'i ng ice domes between the saddles, These ice inherited their orientations fi'om fo rmer glacial land domes arc not mass-balance fe atures created by prec fo rms, Specifically, relati\T to long axes of the lakes, the ipitation patterns; they arc dynamic features created by longitudinal lineations arc lI'hat remains of" glacial ice streams, Computer simulations of the Eurasian ice drumlinization and fluting after the drumlins and flutes sheet that do not allow ice-stream dynamics, such as those were disfigured by post-glacial thermokarst and solilluc by Lindstrom and i\Iau\yeal (1989) and Lindstrom tion processes, and tranS\TrSe lineations arc the remnants ( 1990) , prod uce fe wer ice domes, and those that arc or land forms produced along the f'dge of the retreating produced arc in locations not compatible lI'ith glacial ice-sheet margin, [n addition, our hypothesis explains the geology, Therefore, the glaciation cyelt's simulated by origin and thermal properties of yt'doma, These thick these moclels cannot replicate the fu ll complexity and sequences of ice-rich sand and silt that blanket Arctic I'ersatility of ice sheets, and exclude the ability of" ice lowlands are sediments deposited I]V aggradation in ice sheets to cfleet abrupt climatic change triggered by damming ell\'ironments, Hence liT argue that orient ed iceberg outbursts lI'hen internal ice dynamics actil'ate ice lakes inherited their orientation fr om the grOO\TS streams, or triggered by outbursts of \'ast glacially prod uced by a marine ice sheet tra nsgressing on to the impounded lakes whcn actil'(, ice streams collapse saddles ArCl ic coastal lowlancls from the adjacent seas, This seellls along the ice dil'icle, These outbursts may halt prod uction to be paleoelimatically justified because nort heastern of" North Atlantic Deep \\'ater. and thereby cause abrupt Siberia ancl northern ,\Iaska, I\'i th their "half\\'ay- to- Ice flip-Oops ri'om a glacial to an interglacial global climate, Age" elll'ironments, liTre and arc highl\" susceptible to Such llip-IIops hal"(, becn proposed fo r the last termin easy and early glacicrization ( \ ' erbitsky and Oglesby, ation, fo r example, by Charles and Fairbanks (1992) ane! 1992; :\f arsia t, 1 994) , Afean wh i le, thousands of" n('\\' icc by Broeckcr and others ( 1 992) , Ice saddles collapsed by IlOII'lines can 1101\' be deril"t'd rrom the lake orientations, ice streams could ha\"(' produced outbursts of"icebngs and lI'hich therefore prO\'ide pOII"l'rfu l tools for reconstructing impounded lakes in the Gulf" or Saint La\\Tence, the marine ice sheets in the East Siberian and Chukchi Seas, Labrador Sca, BafTin Bay, the North Sea, the Harents Sea, lI'here rht' usual glacial land forms halT bcen either the Kara Sea, thc Laptc\' Sea, the Chukchi Sea and the altered or buried, or nel'er cxisted beca ust' the bee! was Beau [ort Sea ( Fastook alld Hughes, 1991; GrosslVald, frozen, This interpretation prOl'ed justified by the 1993; Lindst!"om and l\laci\yeal, 1993) , l\lajor marine ice geomorphology or the 'l' ana I nd igirka LOII'land, \I'here streams lI'ould hal'e occupied Cabot Strait in the Gull' of the alignments ororicnlcd lakes and accompanying ridges Saint LallTcnce; Hudsoll Strait in the Labrador Sea; arc parallel to the fo rmer icc-flo\\' direClions deril'Cd [i"om Lancaster Sou nd, Jones Sou lld, l\ares Strait and i\fell'ille ice-shOl'Cd features on the :\Cll' Siberian Islands and on Bugt in BafTin Bay; the Norwcgian Trough in the North the Siberian mainland around Tiksi Harbor (see Figs 3 Sea; Bear Island Trough and Franz Victoria Trough ill ane! 4 Grossll'ald and Spt'ktor, 1993 the Barems Sca; St Anna Trough and Voronin Trough ill Still. liT realize that nearly all the clues for a Illarine the Kara Sea; Boris \'i1kitski Strait in the Laptel' Sea; cast Siberian ice dome arc argulllentatil"l' and therefore Charlie Gap in the Chukchi Sea; and Amundsen Gulfand our reconstruction should be considered tentatil'C, None .\1'Clure Strait ill the Beauf<)!"( Sea, theless, climate moe!elers arc increasingly placing marine Although the glaciological mechanisms that actil'ate ice domes on Arctic continental shell-cs, on the east and deactil'ate ice streams are unclear, thel' probably Siberian continental shell' in particular, because meteor tha\I' and rcfreeze till or sediments in inter-island ological conditions allo\\' ice to thicken OI'er time at these channels and submarine troughs where these deposits locations (T ushingham and Pel tier, 1991; Dcblone!e and accumulate and arc either ice-cemented or \\'a ter others, 1992) , saturated, Upon thawing, the deposits no longer prOl'ide Geologists, in their turn,should not OI'erlook e\'idence basal traction to the ol'Crlying ice, which then conl'Crt s (i'om non-traditional sources, For instance, no one has yet rr om sheet floll' to stream flow as it fu nnels into the acknol\'ledged the paleoglaciological implications of the channels and troughs, Thereforc, ice streams \\'ould be unusually large thickness of see!imentary seq uences acti\'ated by climatically l\"Cumed surf"ace ice that is blanketing the deep Arctic Ocean floor and the peculiar adl'ected downward , by fr ictional heating fr om ice fe atures of their distribution ( Sweeney, 1981) , These shearing OIT!" the beel ane! I))" geothermal heating of sequences arc normally OI'er 2 km thick and grol\' thicker basal ice, COIllTrscly, ice streams lI'ould be deactil'atee! by alongside tht' continental slopes ofI' the Baren ts, Kara and climatically cooled surface ice that is achTctecl dOll"llward, East Sibt'rian continental shell"l's, [s it not natural fo r the by com'erting sensiblt' heat into latent heat as basal
32 1 Journal of Glaciology meltwater refreezes and by conducting basal heat upward ation fr om G. jones, 1993). Also, numerous parallel more rapidly through ice that has been thinned by being plowmarks have been recorded on the southern Yermak down-drawn into ice streams. Although climatic warming Plateau, near the "exit" of the central Arctic Ocean and cooling at the ice surface is a fa ctor, the effe ct of the between Greenland and Svalbard, at water depths of bed can be delayed for millennia, whereas basal melting 850- 1000 m, perhaps even up to 2000 m (Vogt and and freezing that turn ice streams on and off can be others, 1994). These marks seem to be the first-known virtually instantaneous. Therefore, ice sheets have a signatures of the thick floating ice shelf or rather giant capaci ty for abrupt change that is largely independent icebergs either incorporated into the ice shelf or calved of exerted climatic fo rcing. {i-om it and carried across Yermak Plateau by southward On the other hand, changing sea level is external flow of the ice shelf. Marine biological activity north of forcing that can have an immediate effect on marine ice Yermak Plateau began shortly after the Last Glacial streams such as those located in the inter-island channels Maximum (Stein and others, 1994) . Perhaps this and submarine troughs in Figures 3 and 4. As shown by coincided with progressive disintegration of the ice shelf Weertman ( 1974) and applied to marine ice streams by after the calving front breached the ice-shelf bottle-neck Thomas (1977), Thomas and Bentley ( 1978) , Stuiver and in Fram Strait. As fo r the Norwegian-Greenland Sea, others (1981) and Fastook ( 1984) , the grounding line of there is microfaunal evidence that deep-sea life did not an ice shelf with a marine ice sheet can advance or retreat cease there during the Last Glaciation. However, this is irreversibly for small changes in sea level if the bed slopes not an unsurmountable hindrance to the concept of a downward up ice streams, due to isostatic depression Norwegian-Greenland Sea ice shelf, especially if the ice beneath the marine ice sheet. Since ice streams are the shelf thinned as it moved southward, interacted with the dynamic links connecting a marine ice sheet to its floating continental margins and entering ice streams, and ice shelves, grounding-line advance and retreat occur adjusted its shape to conform to inter-island passages, along the inter-island channels and submarine troughs all of which had to result in a multitude of crevasses and occupied by ice streams. Collapse of ice saddles along the rifts. The thinner the ice, the more likely top and bottom ice divides of marine ice sheets are caused by irreversible crevasses could meet and open rifts. Perhaps, the retreat of ice-shelf grounding lines along marine ice periodically forming crevasses and chasms provided streams. Floating ice shelves are therefore an important enough open water for initiation and development of dynamic component of a hypothetical Arctic ice sheet. deep-sea life recorded by oceanographers. Marine organisms live under the Ross Ice Shelf in Antarctica 450 km from the calving fr ont but only 70 km from rifts FLOATlNG ICE SHELVES through the ice shelf in the lee of Crary Ice Rise, a local pinning point (Stockton, 1982). vVi th terrestrial and marine ice domes grounded along nearly the whole circumference of the Arctic, only one additional interconnecting element, an ice shelf floating RECONSTRUCTING MARINE ICE SHEETS over the deep Arctic Ocean basins, would have been required to produce an Arctic ice sheet at the Last Glacial Most of the glacial geological evidence in Figure 4 Maximum. The idea of a floating ice shelf on the Arctic supports transgressions of marine ice sheets from the Ocean is not new. Thomson (1888) and Crary (1960) Arctic continental shelf of Eurasia southward on to the postulated the same ice shelf on thermophysical grounds, Russian mainland and northward into the Arctic Ocean. considering it a consequence of Ice Age changes in the If an Arctic ice sheet developed during late Qu aternary ocean's heat balance. Mercer (1970) postulated the same glaciation cycles, a major component would have been ice shelf on mechanical grounds, citing it as an important marine ice domes on the broad, shallow continental coun terbalance capable of buttr essing a marine ice dome shelves of the Barents, Kara, Laptev, East Siberian and grounded on the Barents Sea floor, checking its instability Chukchi Seas. In the marine-ice-transgression hypothesis and preventing the dome from collapsing. Broecker (MITH), marine ice sheets began when sea ice thickened (1975) and Williams and others (1981) proposed an from snow accumulating on its top surface and water Arctic ice shelf on the basis of oxygen-isotope ratios and freezing onto its bottom surface, until it grounded on sea-level records. Lindstrom and MacAyeal (1986, 1989) shallow Arctic continental shelves. Thereafter, continued modeled floating ice shelves in the Arctic Ocean and the snow accumulation on the top surface produced ice domes Norwegian-Greenland Sea, based on mass-balance that spread onto adjacent coastal lowlands, where marine computations in a finite-element computer simulation. ice may have merged with terrestrial ice fr om ice caps on With reasonable assumptions, they fo und that a dynamic coastal highlands, and spread to the edge of the finite-element ice-sheet model produced an ice shelf continental shelf, where marine ice either calved into
1000 m thick north of Fram Strait, up to 1500 m thick the ocean or became a floating ice shelf. further into the Canadian Basin and 1000-500 m thick in Initial sea-ice thickening rates have been given by the the Norwegian-Greenland Sea as fa r south as Iceland. formula (Crary, 1960) Their ice shelf was supplied by marine ice domes dhr . ( D/ hr) iJ.T - Q - = + -'----'----'----- grounded on the Arctic continental shelf of Eurasia. a (1) There is compelling evidence fo r abiotic conditions in dt prH the central Arctic Ocean during the Last Glacial
NI aximum, which is consistent with a continuous floating where hi is ice thickness, t is time, a IS the ice ice shelf between 33 and 13 ka BP (personal communic- accumulation rate on the top surface of sea ice, D is the
322 Crosswald and Hughes: Paleoglaciology's grand unsolved problem thermal conductivity coefficient fo r sea ice, jjT is the Tce elevations along surface Oowlines of an ice sheet temperature difference between the mean annual air can bc computed fr om Equation (3) by writing it in the temperature on the top surface and the freezing fo llowing numerical fo rm temperature of sea water on the bottom surface, Q is the rate at which heat is supplied to the bottom surface by (TO) jjx ( ---TO ) L'lx -- hi+l = hi + hi + (4) -hI -- = h - hR Arctic rivers and ocean currents, PT is ice density and H is 'i Pl9z i Pl9z the laten t heat of freezing fo r sea water. Integrating
Equation (I) for constant a, jjT and Q where a = L'lhl jjx, jjh= hi+l - hi is thc change in ice elevation in constant step length jjx along a flowline of
_ [ PrHhT ] _ prHDjjT [ (apIH - Q)hl] t- . [ .)]1 n1 + . length L divided into LI L'lx equal steps from the ice aPrH - Q (apIH - Q)- DjjT margin to the ice divide, integer i denoting the step, bed
(2) elevation hR = h - hr and [To/(h - hR )L m ust be specified at each step. Glacioisostatic depression of the 10 Taking Q, � cm a I and jjT� ISoC fo r present-day bed lowers the surface fr om elevation h to elevation h* 0 conditions and Q � for the "White Hole" in Figure I, above sea level and lowers the bed fr om hR to hR * above (2) Equation predicts that sea ice would ground in watcr or below sea level. Taking l' as the ratio of the lowered 100 m deep or less in SOO a or less. As seen by the 100 m surface to the lowered bed bathymetric contour in Figure 3, this would allow marine h* - hR ice domcs to grow in all the scas of the Arctic Eurasian 1'=--- (5) continental shelf. hR - hR* , Orographic effects increase a as the marine ice domes it can be shown that Equation (4) becomes (Hughes, get higher, thereby offsetting the cessation of ice thickening 1985) by bottom fj'eezing after sea ice becomes grounded. Eventually, the marine ice domes will become high enough * * jjx [ TO ] -- hi+1 hi + (6) so their gravitational potential energy becomes kinetic = * I (I + T)h - (1 + T» hR i PI9z encrgy of motion and they will then creep in directions having the greatest icc-surface slope. Initially, the Arctic and that hR * is linked to hR for present-day topography contincntal shelf of Eurasia would be frozen, except in the or bathymetry by the expression western Barents Sea, as is the case today (Gavrilova, 1981 ; Bigl, 1984). As marine-ice domes thicken and spread, (7) howcver, more of the bed will become thawed and glacial 917 3600 sliding will then begin to produce the glacial erosional and If PI = kg m 3 fo r ice densi ty and PR = kg m-3 depositional fe atures shown in Figure 4. fo r thc mean dcnsity of the Earth's mantle, isostatic Glacial erosional and depositional fe a tures that reveal equilibrium rcquires that the pattern of flowlines from the interior ice di\·ide to Pl marine and terrestrial margins of a fo rmer ice shect arc = TO (8) l' = PR - PI 3" . produced by traction between moving ice and its bed. If � the bed is frozen, these features arc rare and consist Taking to as the time constant fo r glacioisostatic primarily of glacial tectonics, such as thrust sheets and compensation beneath an ice sheet, the value of T fo r an hole-hill pairs. If the bed is thawed, these fe atures are ice sheet advancing over time t is ubiquitous and result primarily fr om areal scouring that 1'0 [1 - exp( -tlto)] (9) produces glacial Iineations ranging in scale fr om striated l' = pavements to fluted landscapes. If the bcd is freezing or and the value OfT fo r an ice sheet retreating over time t is melting, these fe atures are mostly confined to thawed parts in a mosaic of thawed and frozen patches where Ta �xp( -tlt ) (10) freeze-thaw quarrying produces a pitted landscape of l' = o lakes and hills elongatcd in the direction of ice flow. If sheet flow becomcs stream flow towards the ice margin, where Ta is the \alue Of T in Equation (9) when advance selective lincar erosion produces fo re-deepened troughs ends and retreat begins. beneath ice streams and the eroded material is deposited Equation (6) is an initial-value finite-difference along grounding lines beyond which marine ice streams recursive fo rmula. Thc initial value is grounding-line ice become afloat or is deposited as lobate moraines where thickness ha for i = 0 at a marine ice margin fo r which terrestrial ice streams end on land. Sugden and John the condition fo r ice just becoming aOoat in water of (1976) described these various glacial landscapes in detail density pw and depth hw is and the mechanisms that produced them . ho h (p I p (11) Basal traction that produces glacial landscapes is = \V w d· represen ted by basal shear stress TO defi ned as the prod uct of ice density PI , gravitational acceleration 9z, ice The initial value is ice thickness hI at i = I for the first thickness hI and ice-surface slope Cl: for a coordinate jj.T step in from a terrestrial ice margin for which
= = (3) system in which x is horizontal along Oowlines and z is hI hi jjh and Equation can be integrated fo r vertical a = jjhljjx and cons tan t TO to give
(3) (12)
323 Journal !ifGlaciol ogy , The finite difference is Llx, The reeursi\'e fe ature of of ice (J',r.r' is gi\'en by Equation (6) is that, once the initial hi is specified by ei ther Equation (11) or Equation (12), hi+l fo r the next (14) step is calculated readily and becomes hi fo r calculating l1i+ 1 i n the lollowing step, a process that recurs along a TO is given by flowline of length L until s tep i = LI Llx at the ice divide, ( _ ') 1 l\lodified Eulcr or Runge KUlla solutions of Equation (6) PI\ 111/(2111+1) are a\'ailable to reduee the dependencc oC h on the length TO -Tv I-- (15) PI Llxo f steps. These soluti ons introd uce a correction factor that depends on Llxand that can be applied to Equation (6) if it is sohed directly using a pocket calculator and a _ (P\\,) 1111/(2111+1) gi\'Cn Llx. Ts-Tv -P (16) \\'ith \'alues of hR at each Llx step obtained from I topographic or bathymetric maps of the deglaciated p\\' density, Pi landscape and \'alues of T obtained Crom Equations (9) PI is ice density, is water is basal ice and ( 1 0) fo r specified times of glacial advance or retreat, pressure, P\\, is basal water pressure, Tv is the \'iscoplastic
' only TO must be specified at each Llx step in Equation (6) in yield stress of ice and m is the \ iscopl asti e parameter i n order to reconstruct ice e1e\'ations along flowlines of fo rmcr the \Veertman (1957) sliding law of ice, modified to ice sheets, Values of TO depend on ice velocity and on how include sliding on soft water-soaked sediments or till. frozen and thawed areas are distri buted OI'er the bed. Since Eq ua lions (13) and (14) are results of the mass TO is a measure of bed traction, TO is nil beneath ice di\'ides balance and the force balance, Eq uation (15) assumes \I'here ice is bareh- mo\'i ng and beneath ice streams where a that bedrock bumps which control the sliding \'elocity for layer of basal \I'ater or muck greatly reduces traction, In sheet flOlo\'o\'er bedrock become progressively drowned by bet ween, TO increases from the ice di\'ide to the ice stream, basal melt\l'ater or blanketed by easily ' deformed because ice \'eloeity must increase in order to transport ice sediments or till fo r stream floll' , Equation (16) assumes accumulating on the ice-sheet surface. If sheet flow that side shear increases in proportion to the decrease in con tinues to the ice margi n, TO will increase in the basal shear. A solution fo r these eq uations is obtained by accu mulation zone and decrease in the ablation zone, assum i ng the fo llowi ng variation of PI\' I PI along normal because ice \'Cloci ty increases and then decreases as ice is ized length xl Ls of' the ice streams gained by accumulation or lost by ablation. Theoretical \'aria tions of TO fo r sheet flow along an ice (17) flowline ha\'e been deri\'Cd lor \'ariable accumulation and ablation rates, \'ariable cOIl\'Crgence and divergence of flo\\', and \'ariable frozen and thawed basal conditions \I'here h\\' is the heigh t to \\'hich p\\, would raise basal along the flO\I'line ( H ughes, 1985 ) , Gsing accumulation \I'ater in an imaginary borehole through ice of thickness
and ablation rates fo r marine ice domes on the Eurasian hi, and 0 < c < 00 represen ts the spectrum of basal Arctic continental shelf specified by Hughes (1985), and buoyancy along length Ls of stream flow that converts
using the glacial geology in Figure 4 to speci fy con sheet flow (c = (0) to shelf flow (c = 0), with x = Ls at \'('rgence or divergence of flo\\' and fr ozen or thawed basal the head and x = 0 at the fo ot of the ice stream, For
conditions, these theoretical TO values can be calculated aeti\'e ice streams, 0 < c < 1 . For inacti\'e ice streams,
and used to eom pu te floll'li ne-eleva tion profi les in 1 < c« 00. Equation (6) fo r sheet flow. 1 n applying Equation (13) along flowlines determined ' For stream flow along length Ls of a flowline of length by the glacial geology in Figure 4, the LlO',u. I Llx lerm L, the counterpart of Equation (6), using an imprOl'ed can be ignored because, except near the grounding line, it ' \'Crsion of the analysis by H ughes ( 1992b) , is is small compared to the other terms when 0',1','. is gi\'en by Equations (14) and (17), the term can be ignored [ ) ( )?] 'I., PI\'- ' PJ except in the fj ords of Norway and Svalbard, where hi+l = hi + (1 - - - P (h - hR )i is not insignificant compared to Wi , RJ,/ 1 if Pw I ::::::
Wi does not vary greatly along Ls, Tv :::::: 100 kPa for viscoplastic yielding in shear, Ls fo r active ice steams is the length of an inter-island channel or a submarine trough fo rmerly occupied by an ice stream, c is chosen to gi\'e the same ele\'ation at the ice divide lor floll'lines
dO\\'ll opposite flanks of' the ice di\'ide, and Uo is obtained fr om the mass flux of ice ha\'ing thickness ha and \I'idth Wo
where a is the ice-accumulation rate, ho and Uo are ice at x = 0, as specifi ed by the mass balance fo r ice thickness and \'eloeity ( negati\'e fo r x positi\'C upstream) cOl1\'Crging on the ice stream.
at the foot of the ice stream , W is the width of the ice The glacial ge ology in Figure 4, and the ice-sheet stream, R,,.,,' represents the stress field, is the reconstructions in Figures 5 and 6 that are based upon it, ' O'er,,.' longit udinal deviator stress, LlO',L,. I Llx is longitudinal are added to the map of the Arctic Region, sheet 14 of the
de\·iator stress gradient, TO is the basal shear stress, 'I., is world 1 : 5 000 000 series of topographic and bathymetric
the side shear stress, A and n arc the hardness parameter maps p rod uced by the American Geographical Society. and the \'iscoplastic parameter in the N ye ( 1953) flow law The isostatically depressed topography and bathymetry
324 G'rOHll,(//d and HlIghes: Paleog/aria/ag)'j grand 1IIlsah'N/ /Jra blem
GLACIA TION FOR SNOW LINES LOWER beneath terrestrial and marine components of the 1000 In reconstructed ice sheets arc obtained for T = Ta at t = ta in Equation (9), Upper and 101l'er limits of t" may be Figure 5 shows the extent of marine Eurasian glaciation
15 ka > ta > 5 ka, since this brackets the spacing of a!i.er 20 000 years fo r present-day precipitation rates when Heinrieh e\'ents fo r glacial unloading of Laurentide ice present-day sno\l' lines are lowered by 1000 m, According in the ;\IonhAtlantic ( Bond and others, 1992) , Sea le\'el to the highland-origin, II'ind\l'ard-grOlllh hypothesis began to rise rapidly about 14000 ),ears ago (Fairbanks, (Flint, 1971), the present-day ice cap on NO\'aya Zemlya 1989), after Il'hich marine ice domes on the Eurasian is the most probable nucleus for a marine ice dome in the continental shelfll ould halT begun to collapse if their ice Barents Sea, �loistLlre-bearing \I-esterly \I'inch, along \I'ith streams became afloat on a bed that sloped upII'ard shalloll' water on the west and deep \I'ateT on the cast. sea\l'ard (Thomas, 1977) , H the ice streams became afloat would allow this ice cap to ad\'ance into the Barents Sea on a bed that sloped dO\l'l1\1'LHd sea\l'ard, the marine ice but not into the Kara Sea . . \ marine ice dome would then domes would ha\"( remained stable despite rising sea le\TI de\'clop in the eastern Barcnts Sea and mCl'ge with a and ta would be longer. Any ta \' 1989: Pelto, 1992) . Both reconstructions arc obtained to oriented lakes and glacial tectonics (sce Fig. 3). The fi'om present-day precipitation rates, with t" = 20 ka in East Siberian dome extended to the edge of the Eq uation (10), allo\\'ing 20000 years of gro\l,th to continental shelf in the north alld formed a saddle with conform with the high-latitude hemi-cyclc of insolation the Chukchi dome in the cast. Figure .3bshows the ice dominated by the +1 ka cycle of' the Earth's axial till. flow pattern, for comparison wi lh glacial geological flow Insolation is assumed to control snoll' lines, Snow-line indicators in Figure +. Features II'C attribute to glacial IO\l'ering expands existing island ice caps in the eastern tectonics arc southll'ard about an ice dome nonh of the and nort hern Barents Sea, Howe\'er, these ice caps cannot Ke\l' Siberian lslands and imply north\l ard retreat orthe adl' 325 JournaloJ Claciology the general circulation-model simulation by lVIanabe fo r comparison with glacial geological flow indicators in and Broccoli ( 1985) . Marine Chukchi ice spilled over Figure 4. There are no major discrepancies. Chukchi Peninsula and through Bering Strait to produce The marine ice dome oyer the New Siberian Islands in glacial tectonics on St. Lawrence Island and to end as a the East Siberian Sea would ha\'e advanced southward to calving ice wall along the edge of the continental shelfin the fo othills of mountains in northeastern Siberia if snow the southwestern Bering Sea. Field studies by Heiser and lines were 1200 m lower. Therefore, we located the others (1992) and by Brigham-Grette and others (1992) southern margin of this ice dome along the edge of these show that the snow line during the Last Glaciation rose fo othills. In particular, we assumed that the Kolyma fr om 150 m high on St. Lawrence Island to 300 m high River was ice marginal at the glacial maximum, so that on Seward Peninsula and a submarine ridge, possibly a its course today marks the fo rmer ice margin, and we moraine, extended northeastward from St. Lawrence assumed that ice lobes occupied the valleys of the Island. These results limit the transgression of Chukchi I ndigirka Ri\'er and the Yana River. If true, the glacial marine ice to the western Bering Sea and to coastal geology produced by this advance is now buried under Alaska north of Seward Peninsula, perhaps as far as yedoma deposited in ice-dammed lakes during ice retreat. Point Barrow, according to oriented lakes. Ice from the In the Laptev Sea, the ice saddle between the Kara ice Chukchi dome may have spread over the Chukchi dome and the East Siberian ice dome was both higher fo reland, where it would have entered ice streams in and broader when marine ice advanced to the sou thern submarine troughs to the north and ended along the fo othills. For ice flowing northward from the East Northwind Escarpment to the east. Figure 5c shows the Siberian dome, we located the grounded ice margin at ice-flow pattern fo r comparison with glacial geological the 1000 m bathymetric contour west of Lomonosov flow indicators in Figure 4. The depositional wedge on Ridge and at the 1500 m bathymetric contour east of the continental slope in the southwesternBeri ng Sea in Lomonoso\' Ridge, since an ice shelf 1500 m thick in the Figure 4 may have been supplied by the cah-ing ice wall Canadian Basin of the Arctic Ocean could not cross in Figure 5c. Transgression of the marine Chukchi ice Lomonosov Ridge without thinning by 500 m. Figure 6b dome onto the northwestern Alaskan coast should shows these flow patterns fo r comparison with glacial isostatically depress the coastli ne. Raised beaches 10 geological flow indicators in Figure 4. Discrepancies are 14 12 m high are widespread along this coast and are C minor and are the same as those noted fo r Figure 5b. dated as older than 40 ka (Hamilton and Brigham The Chukchi dome is enlarged by moisture-bearing Grette, 1991). winds from the North Pacific and it expands in that direction, if snow lines are lowered by 1200 m. Mountain glaciation in northeastern Siberia and Alaska is also GLACIATION FOR SNOW LINES 1200 Ul LOWER expanded. In particular, glaciation of the Anadyr Range and the Koryak �Iountains in Siberia and the Brooks Figure 6 shows the extent of Eurasian glaciation after Range in Alaska allow marine Chukchi ice pouring over 20 000 years fo r present-day precipitation rates when Chukchi Peninsula, through Bering Strait and over present-day snow lines are 1200 m lower. This allows Seward Peninsula to be unusually thick because marine the marine ice dome in the eastern Barents Sea to and mountain ice merged, thereby eliminating lateral continue expanding eastward into the Kara Sea and ablation zones in this region. Marine Chukchi ice would southward on to the central Siberian mainland, even after then have been able to advance fu rther southward across its expansion westward and northward had halted at the the Bering Sea. Judging from glacial troughs cutting "the edge of the Eurasian Arctic continental shelf. Therefore, corners" of southeastern Chukchi Peninsula, as well as of the Barents ice dome migrated southeastward into the the Navarin, Olyntorskiy and Govena Peninsulas (all Kara Sea to become a Kara ice dome that was much three the seaward promontories of the Koryakskiy larger and was able to engulf highland ice caps on Taimyr Range) , that ice probably invaded the deep basin of the Peninsula, Putorana Plateau and Anabar Plateau (see Bering Sea, ending as an ice shelf that calved into the Fig. 3). Island ice caps on Severnaya Zemlya, nourished Pacifi c Ocean along the Aleutian-Commander Islands. It by moisture-bearing westerly winds, could expand is worth mentioning that this fu ndamentally new concept westward and thereby con tribute to forming a mari ne of the North Pacific glaciation is consistent with the Kara ice dome. Kara ice could then spread eastward sensational discoveries made on leg 145 of the "J oides across Severnaya Zemlya and through Boris Vilkitski Resolution" Ocean Drilling Program (Anon., 1992) . If Strait into the Laptev Sea. Kara ice could also spread longi tudinal lakes and tra nsverse ridges in the southern westward in through valleys across Novaya Zemlya and Yukon delta reveal glacial lineations, the lower Yukon cross the Baren ts Sea, ending as an ice stream in Bear River would have been ice marginal. In northern Alaska, Island Trough, as ice streams in submarine troughs and Chukchi ice may have occupied glacial through valleys in inter-island channels in the northwest Barents Sea and as the DeLong :vI.ountains, entered the Beaufort Sea and ice lobes following river valleys on the north Russian transgressed onto the Alaskan North Slope as fa r as the Plain. Kara ice moving northward would have entered Colville River, which would have been ice marginal. A ice streams in Saint Anna Trough and Voronin Trough. larger Chukchi dome may have advanced to the 1500 m Kara ice moving southward would have followed the Ob bathymetric contour in the Chukchi fo reland. This flow and Yenisey River estuaries into the central Siberian pattern is shown in Figure 6c fo r comparison with the Lowland and fo llowed glacial through valleys in Taimyr glacial geology in Figure 4. Oriented lakes in the southern Peninsula to continue across the Putorana Plateau and Yukon delta are from the Black and Kwiguk Qu adrangles the Anabar Plateau. Figure 6a shows these flow patterns of the United States Geological Survey (USGS) topo- 326 Crosswald and Hughes: Paleoglaciology's grand unsolved problem graphic maps of Alaska, \\'hereas other oriented lakes cycles longer than but comparable to surge cycles of some depicted in Figure 4 are from satellite imagery. \Ve have mountain glaciers? Preliminary studies suggest they do more confidence in gi\'ing a paleoglaciological interpreta ( Clarke. 1987) . Can these life cycles produce iceberg tion to oriented lakes in satellite imagery, because the outbursts in the Nort h Atlantic that correlate with associated geomorphology is more e\·ident. Through "Hcinrich e\'ents" of' glaciomarine scdimentation that valleys in the DeLong :-lountains and across Scward may date cessations of :\orth Atlantic Deep \\'ater Peninsula arc also fi'om USGS topographic quadrangle production and shu t-downs of o\'ert urning circulation? maps, so our interprctation that thcsc are glacial through This possibility exists ( \JacAyeal, 1993) and the ces \'alleys is questionable. With thesc reservations, the 001\' sations may trigger abrupt global climate change indicators in Figurc 4 are in reasonable agrcemcnt with ( H ughes, 1992a) . How was the dry interglacial ecolog the Oowlines in Figure 6c. Raised marine shorelincs up LO ical regime of Cen tral Asia transf'ormed by major changes 60 m high on thc northern and northwestern Alaska in atmospheric circulation and water-vapor transport coasts, and of apparent Pliocene age (Brigham-Grctte and ensuing fi'om the Arctic ice-sheet fo rmation" What role in Carter, 1992) , \yould be compati ble \\'i th glacioisostatic these changes was played by the \'ast proglacial lakes depression by the Chukchi ice dome in Figure 6. Wc impounded bv the ice sheet, especially by their outbursts propose that these raised shorelincs represented rcbound during deglaciation? On theoretical grounds, one may after collapse of the cosmogenic nuclides may identify expect profound ecological and gcomorphological re these shorelines as Pleistocenc fe atures. organizations. What arc the archeological implications fo r peopling the Americas if' the Bering Land Bridge between Siberia FUTURE RESEARCH GOALS and Alaska \\'as periodically blocked by a marine ice lobe extending across Bering Strait from a marine ice dome in Paleo-ice sheets controlled by present-day prccq)ltatlon the Chukchi Sea? Our finite-element ice-sheet model rates for snow lines either 1000 or 1200 m lowcr and produces this lobe with only modest changes of prescnt growing fo r 20 000 ycars arc, of course, not thc only day climate in Beringia (Hughcs and others, 1991). possibilities. Obvious alternatives are to confinc gro\\·th of Finally, ho\\' rapidly and how soon could an Arctic ice paleo-ice shccts to other timc spans and to allo\\' dirTcrent shcet fo rm again? precipitation rates and other snow lines. \lore realistic Answers to these compelling questions are hidden in reconstructions would allow prccipi tation ra tes to change the Arctic, fo remost in Arctic Russia. With Russia no\\' as an ice sheet gcts larger and it increasingly creates the more open to international co-operati\'(' scientific in\'es su rrounding metcorological cond i tions, because it be tiga tions than e\'er bcf()re. definiti\,(, answers to thcse comes the dominant geographical fe ature. It is also questions arc within reach but time may be running out possi ble that, exccpt for the Barcn ts Sea, thcre was no as we enter a new and perhaps a warmer century. In cxtensi\·c glaciation of the Eurasian Arctic contincntal particular, onc can speculate that if "greenhouse" shelf, especially during the Last Glacial \Iaximum, so warming eliminated vvinter sea ice in the Low Arctic, there \\'as no Arctic ice shcet. This possibility has thc allowing greater winter precipitation in the High Arctic. aeh'antage of' f'rccing us from the task of' reconstructing thickening sea ice in the High Arctic might ground on paleo-ice sheets in the Russian Arctic but it makes no shallO\\' Arctic continental shell-cs and impound rivers contribution toward stimulating Quaternary research in nowing into the Arctic Ocean. The "White Hole" might these remote rcgions, in the hopc that keys to under form almost instantaneously in the dccades ahead and standing global climatic change will be fo und there. enclose an area com parable in size to that cOl'Cred by an Wc expect that fu ture rescareh related to the Arctic Arctic ice sheet at its maximum stage (sce Fig. I), causing ice sheet will provide new insights into the \'ery basics of' sea level (0 drop almost as fa st as it rose during paleoglaciology. It will il1\'oh-e mapping the maximum termination of the last glacial cycle. Paleoglaciology's cxtent of marine ice domes on Arctic continental sheh-es, grand unsoll-ed problem may be answered much sooner mapping and dating the stages of their deglaciation, than lI'e expect and by Nature herself, not b\' us. determining the dynamics of iceberg and lacustrine outbursts as ice domes collapsed and ice-dammed lakes were drained, establishing whether the ice domes were IMMEDIATE RESEARCH GOALS connected across the Arctic Ocean by a noating ice shelf, locating the major ice streams that linked them \Vc do not need to wait until Nature answers palco dynamically to the ice shelf, and determining ho\\' glaciology's grand unsolved problem. I\Iuch can be done changes in one seCLOr or a unified Arctic ice sheet LO determine wh ether or not an Arctic ice sheet existed in alIccted ice dynamics in other sectors. On a more general the past and what arc its implication. for understanding scale, we will learn \\'hether existence of' the ic(' sheet Qu atemary climatic changes, including changes triggned compels recasting our knowledge and understanding or by "greenhouse" warming and abru pt changes. Quaternary glacial his(On·. In facl. the pa\,-of[ may be a scientific re\'olution in understanding global climate and Datable materials, including tusks and teeth of sea-le\'el changes, especially abrupt changes. woolly mammoths, are widespread on the Arctic Lessons learned fr om studying glaciology's grand coastal plains of Siberia and its ofT.<;hore islands, and unsoked problem in the Antarctic can be applied they are typically associated wi th both glacial and directly to understanding paleoglaciology's grand un marine deposits. So fa r, these materials are only soh'Cd problem in the Arctic. Do ice streams ha\'e life sporadically dated and some of the samples used for 327 ]ollmal of C'laciolog), dating appeared to be contaminated and thus yielded The nearly continuous raised marine shorelines spurious, misleading chronologies, Ob\'iously, these along the north\\'est Alaskan coast are older than chronologies should be subjected to scrupulous 40 ka BP (Hamilton and Brigham-Grette, 1991), This revision, based upon systematic dating and rc-dating coastline should be dated using exposure dating with of Siberian samples. taken in stratigraphically clear cosmogenic nuclides to determine whether ice from situations, combining A�lS I+C dating and surficial the marine Ch ukchi ice dome transgressed on to the dating of eosmogenic nuclides whene\'er possible, A Alaskan mainland during the Last Glaciation, history of marine ice-sheet transgressions onto the Arctic coastal plains can be unra\'c!ed by this Holocene raised beaches are 10\\' or absent on chronology-establishing program, Russian Arctic islands but are high and widespread on Canadian Arctic islands, This implies that either Dating lacustrine deposi ts of the fo rmer Y clogui marine ice domes \\'ere lo\\' or absent on the Russian ice-dammed lake in the Yenisey Yalley and a youngest Arctic continental shelf' but not on the Canadian moraine on the northern Taimyr Peninsula are of Arctic continental shelf during the Last Glacial highest priority, as their ages would shed light on the �Iaximum, or a floating ice shelf' or perennial sea ice maximum extent of the last Kara ice dome and the pre\'Cnted the wa\'e action needed to fo rm beaches in final stages of its history , the Russian Arctic until the late Holocene, The second explanation is the case in Antarctica, where the late Russian geologists, especially those working in \\'isconsin marine ice sheet on the Ross Sea con Siberia, have accomplished impressi\'C work on glacial tinental shelf' reached 700 m abo\'e present-day sea and Quaternary lithology, stratigraphy, paleontologv, le\'el on Ross Island and along the nearby Transant archeology and geocryology, HO\l'e\'Cr, glacial gco arctic �Iountains, and \\'as O\'Cr 1300 m thick in morphology, which is most important for paleoglaciol \Ic:'furdo Sound yet raised beaches are low or absent ogical reconstructions, has not similarly ad\'anced, (Stuiver and others, 1981), This region lies just north Hence, a special glaciogeomorphological program of' the calving front of the Ross Ice Shelf; which has should also be considered, Specifically, the existing retreated during the Holocene and perennial sea ice maps of cnd moraines, ice-shO\'ed features and ice extends to most shorelines e\'en today, Systematic motion-directional proxies, such as giant glacial gra\'ilY measurements should be made in the Russian groO\'es, drumlins and fl u tes, should be \'erified and Arctic to see whether a pattern of' negati\'e anomalies impro\'ed on the basis of air-space images and field exists that is compatible \\'ith the marine ice domes in reconnaissances, We suggest that special attention be Figures j and 6, Hit is, these ice domes probably existed paid to the Arctic "oriented lakes" as a particularly during the Last Glacial ]\laximum and ice grounded promising tool of ice-sheet reconstructions fo r Siberia, along Holocene shorelines pre\'ented beaches from fo rming during post-glacial isostatic rebound, �larine scientists are on the verge ofa major break through in sturiirs of the deep Arctic Ocean, We expect that the studies of unusually thick sediments DISCUSSION blanketi ng the ocean floor would yield e\'idence fo r past glacial-interglacial cycles, In this contexl. \I'e \\'e ha\'e a\'oided assigning times to the marine ice sheets should be ready to correlate the glacial records from reconstructed in Figures 5 and 6, These reconstructions the Arctic continental sheh-es of Eurasia and the are constrained by geomorphology and not by dates, Both Eurasian mainland with those of the ocean, espeeiallv reconstructions allow the ice sheets to grow fo r in Arctic basins just beyond troughs on the Arctic 20000 years for snow lines lowered by 1000 and 1200 m, continental shelf that are the most probable sites fo r respecti\'ely, \\'e expect the marine ice sheets to fo rm ice streams that drained the marine ice domes, whene\'er these conditions prevailed during the Quatern ary, Specifically, since oxygen-isotope records during the The origin of the ice-rich unconsolidated sediments last 800 000 years reveal glaciation cycles lasting about blanketing the Arctic coastal lowlands in northeastern 100 000 years each, with no great di[(' erences in ice \'01 ume Siberia and northern Alaska is still unclear and from one glacial maximum to the next if oxygen-isotope debatable, \Ieanll'hile, these sediments, containing ratios are a reliable ice proxy ( Hays and others, 1976) , we en'atics and taken either for signatures of eustatic bclie\'e that either reconstruction is possible fo r any mari ne transgression or fo r lacustri ne seq uences Pleistocene glacial maXimum, Furthermore, neither fo rmed in ice-dammed lakes, may testify as to reconstruction has attained a stead\' state for our \I'hether these lowlands de\'eloped under dominance specified mass balance, If' lime \\'as sufficient, a sno\\' 0(' marine oceanic transgressions and regressions I i ne 1000 m lower would cOI1\'en the Figure 5 reconstruc ("dilu\·ialistic" scenario) , or experienced recurrent tion into the Figure 6 reconstruction, and a sno\\' line glaciations by a marine ice sheet grounded on the 1200 m 10\l'er \I'ould make the Figure 6 reconstruction Arctic continental shelf, The latter "glacialistic" even larger, so that ice in Figure 6a would advance into scenario seems more predictable for the High Arctic, the Ukraine, and ice in Figure 6b would reach the Sea of \\'hich e\'en today is just half-\\'ay bet\\'een lee Age Okhotsk, The geomorphology in Figure 4 indicates that and interglacial condi tions, especially the broad time \\'as not sLifTicient to produce a steady-state ice sheet continental shelves and glaciated islands of the in Arctic Eurasia, The actual limils of glaciation, Russian Arctic. especially during the last glaciation cycle, will be 328 C;rossu)(tld alld Hughes: Paleoglaciologl" s gralld ul1solved problem established not by computer models but by careful and bed that partly replaced the former ice load, Spill \\'ays sustained field work using the latest dating methods, draini ng these lakes would be con tinuousl y reloca ted H o\\'e \'e r , our m od el simulations alert us to the duri ng ice retreat, so the isostatic response to deglaci ation possibilities, would be irregular bot h temporally and sp a t ial l y , The ice The ice-sheet reconstruction in Figures 5 and 6 a re ma rgin would respond to these instabilities \I'ith advances based on the \'ie\\' that sno\l' preci pi ta t ion borne by that were superi m posed on a general retreat, such as wes t erly winds was hea\')' on the western side of' Novaya occu rred along the landwar d margin of t h e marine ice Zemlya and light on the eastern side, and an ice dome sheet in the Barents and Kara Seas, \\'here a tha\\'ed bed \\'ould m igra te east\\'ard o\'Cr No\'a y a Zeml ya if the exi sted and so glacial geol ogy \\'as prod u ced , regional sno\l' line were 10ll'ered by an additional 200 m, Because the bed was thawed, basal sliding \\'as This \'ie\l' req uires that a single high ice dome de\'eloped continuously t ra nsporting marine shells land\\'ard and in the Barents Sea, HOWe\Tr, i ce spreading fr om a single each ad\'ance and retreat of the landward ice m a rgin high Kara Sea ice dome pro\ ides a bet te r fit to glacial \\'as accompa nied Iw dumping this mix 0(' m a rine shel ls geol ogy, near the Urals and else\l'here, than ice sp reading in progla cial lakes or as glaci al out\\'ash, The mixed f'rom a single high B a re nts Sea ice dome, During shells would have the ages of all the Pleistocene deglaciation, retreat o f' ice margins \\'ou ld Iea\'e a single intergla cia tions and many interstadials within glacia i ce dome O\'cr Nm'a ya Zemlya in Figure 5 and over t h e lions, This is why radiocarbon ages of' ma rine d eposi ts on Kara Sea in Figure 6, G lacial geology sho\\'s ice mo\'ing the Siberian lowlands are often non-finite or "o\'er \\'est\\'ard into the B a rent s Sea and there is no e\'idence 40 ka", These bogus dates and 10\\' raised beaches that an ice dome e\'er existed over No\'a y Barellls Sea ice domc depic ted in Figu re 5 ne\ er existcd , ice-dammed lakes to be discharged directly illlo the open then the Kara Sea ice dome d epict ed in Figure 6 sea, Raised beaches arc of'ten present bu t the ), are 10\\', de\'eloped in situ, not as a result of e ast \l'a rd migra tion This is consistent \\'ith final collapse of the E urasi an of' a hypot het ical Barents Sea ice d om e , marine ice domes late in the Holocene, \\'hen isostatic One could conclud e that no fl ight s of raised beac h es rebound \I'as nearh' com p lete and the coastlines were along the coast of the north Russia n Plai n prmTS tha t no season a lly fiTe of' sea ic e , marine ice sheet existed in the Barents Sea during the The concept that when marine ice sheets are Last Glacial \Iaximum, HO\\'C\'Cr, high raised beaches on grounded on polar continental sheh'es and coasts, so the islands of' S\'albard and \\'idespread g lac i al geology on that ice mo\'es O\'er preg lac ial marine sed iments which t he floor of' the Bare n ts Sea and on the north Russian arc then e rod ed , entrained and transport ed land\\'ard, t o Plain prm'e that a marine ice sheet did exist, \\'i thout be d um ped into ice-dammed lakes or added to glacial producing fl igh ts of' ra ised beaches on the north Russian out \\'ash . pm\'ides an ex planat ion f or t h e h uge lowland Plain, The sallle is t r ue of the Kara Sea, Further cast, the areas of Siberia that a re b lan k e ted \I'i t h o l d -age fossil-rich marine ice sheet \I'ould form in situ on a frozen bed, so sediments h a \'ing marine geochem istry , .\('ter many little glacial geology \\ ou ld be produced , The lack of' Qu ate rn a ry glaciation c)'cle s, contamination of' surficial \\'id es pread glacial geol ogy or flights of raised beaches on sedimen ts by no n-contem porary organ ics and i norganics the coas tal plai n s of' Siberia and Alaska is not, in itself� is now so o\'en\'helming, so omniprcsel ll and so di\'Crse proof' that the frozen continental shel\'es and adj acen t that e\'Cn t he A!\IS lie dating tech ni q ue may be unable coastal plains of' the Lap te\', East Siberian, Chukchi and to u n tangle the stratigraphic puzzles, let a lo ne cOI1\'('n Beaufort Seas \\'ere not cO\'ered by a manne Ice sheet tional rad iocarbon dati ng, Ignoring d epos i tion by a during the last glacia tion, ma ri ne ice sheet has led so m e Russians to embrace A solut ion to t his dilemma lies in recognizing the neodilu\'ianism as an explana t ion for lI'idespread marine distinction bet\\'Cen glaci a ted coastal zones of the .\Ior sediments or "old" age in t h e 10\l'er Pechora Basin and in \\'egia n type and of' the Siberian type, The .'\ol'\\'Cgian the \\'est S ibe ria n LO\l'land, coas tal zone, \\'hich inc l ud ed S\'alba rd, lay betll'een the The Dry Valleys of' Antarctica were il1\'aded by ice divide and the sea\\'ard margi n 0(' the ice sheet, so marine ice lobes \I'hen the marine \\'est Antarctic ice retreat of' t h e sea\\'ard ice margin was i m med iat ely sheet ach'a n eecl across the R oss Sea during the last accom panied by isostatic rebound of' a coastal zone that gl aciation , These ice lo bes clammed lakes in the Dry was in continuous contact wit h the open sea, Therefore, \' alle\'s just as \\'C propose fo r t he Siberian Arctic coastal fligh t s of raised beaches fu ll of contemporary marine pla i n , The co n f'using s tra ti graph y of' t he D ry \'alleys fa una \I'ould be p roduced , one beach for e\'er\' increment 1'C\'Caled a d eglaciati on his tor\' of t he marine ice s heet ' of retreat, and rad iocarbon dating of faunal shells \\'ould only after the lacust ri ne shorelines \\'CI'C m apped and \'ield a clear and prec ise chronology o f d eglaciation, organi cs in their associated deltas lI'erc dated ( S t ui \'er In sharp contrast, the Siberian coastal zone la y and ot h ers , 1 98 1 ) , This history placed gla cia I e )')"a tics O\'er bet\\'een the ice di\'ide and the landward m argi n of the 600 m abo\'e raised beaches o nly 6m high on n ea rby Ross ice sheet, so re t reat of' the landward m a rgi n w as Island, Therefore, \\T point to Antarctica to support our accompanied by isostatic re bound of l a nd that was not \'ie\\' of the glacial history of the :\rctic, We note, hO\\'C\'er, in contact \\'ith the open sea, and marine beaches would that 30 yea rs of sust ai ned field research were spent before not fo rm, Instead , ice-d a m med lakes alon g the ret rea t ing the record of the last deglacia tion in the Dry \' a lleys, a tiny lancl\\'ard ice marglll would fo rlll and be a load on the area, was understood, \\'e cannot expect t h e R ussian 329 Journal cif ClaciolofJ' Arctic, which spans II time zones, to reveal its deglaciation Early Holocenc geochronology in the northern Pcchora 10\\'land.J III Punning. Y. -:'d.K .. ed ..\ oIJJ e dal/l�lJ e 1)0 geokhronolog�)' Chelt'erlicilllogo history without a correspondingly larger effort. ' ' ' a Period . \el(' dala Oil geo(hrol/ologl' if Ihe Qjlfllemat)' 1\10scm,', "auka. Of more glaciologieal interest than the extent of 101 111. fo rmer Eurasian glaciation is the expanded horizon for ice .\stakhm·. \'. I. 1992. The last glaciation in \\'est Siberia. Suer, Geol. dynamics when Eurasian glaciation combines with North Clldas . . Ser, A 81, 21-30. Big!. S. R. 1984, Permaf'rost. seasonally f'rozen ground, snOll' COITr and American and Greenland glaciation to complete the ITgetation in the USSR, CRREL Rep. 84-36. circuit of marine ice domes on Arctic continental shelves. Blake, \\' ., Jr. 1970, Studies of' glacial history in Arclic Canada. I. Then the domes could sustain an ice shelf floating on the Pumice. radiocarbon dates, ancl clifTerenlial postglacial uplift in the Arctic Ocean, so that an Arctic ice sheet existed as a eastern Queen Elizabeth Islands. Call. J. Earlh Sri" 7 (2 , 634--66+. Bond. G. fllld 13 olh . 1992. E\'idence for massi,'c discharges of' icebergs unified dynamic system, just as does the Al1larctic ice m into the North Atialllic Ocean during the lasl glacial period. ,\(tlllre, sheet today. The heartbeat of the Arctic ice sheet would 360(640 I). 245 249. be cycles 0[' thawing and refreezing of the bed under the Brigham-G relle. J. and L. D. Carter. 1992. Pliocene marine transgres sions or northern \ Ia ka circulllarClic correlations and paleoclimatic major ice domes, with the concomitant reduced and . s : interpretations. :I t' li , 45 ( 1 ) , H 89, restored basal traction causing the sou them ice margins ( r Brigham-Grelle, J., S. Benson. D. ,\1. Hopkins, P. A. Heiser, \'. F. to advance and retreat, and sea le\ el to rise and fa ll. The hanol' and A. Basi"'an. 1992 . .\Iiddle and Late Plcistocene Russian arteries through which this heartbeat pumped ice were glacial ic(' extent in the Bering Strait region: results of recent field work. (;eol. Soc. /1111. 19.92 A 11111101 ,lleeling . lbslrarls lI'ilh i\346. the major ice streams radiating fr om the major ice domes . Programs 1975. and flanking the ice saddles between ice domes. V\,ith an Brocckn, W, S, Floating glacial ice caps in the Arctic Ocean. Sriellce. 188(4 193 . 1116-1 118. Arctic ice sheet beha vi ng as a single dynamic system, the Broecker, \\'. S. and G. H, Denton. 1989. The role of' ocean atmosphere a 0) . heartbeat of each ice dome communicated with the reorganizations in glacial cycles. Georhilll. Coslllorhim. Acl . 53( I heartbeats of all other ice domes through their ice 2465-250 I. streams, since they pumped ice into the sea through Broecker. \\' .. G. Bond. K. '\liecz)'sla\\'a. E. Clark and J. '\lc.\l 330 CrosswaLd and Hughes: Paleoglaciolog)!,sgrand unsolved problem Gataullin. \'., L. Polyak. O. Epstein and B. ROlllan"uk. 1993. Poleoce((lIograjJ/�)', 5 (2) . 207 227. Glacigenic deposits of the Central Deep: a kc) to the Late Qu aternary Lindstrom, D. R. and D. R. :'IlacAyeal. 1986. Paleoclimatic constraints c\'olution of the eastcrn Barellls Sea. Borem. 22 I , 4 7 58. on the maintenance of possible ice-shelf CO\'er in the Norwegian and Ca\TilQ\'a, 1\1.K. 1981. Sovrflllflll�O klimal i l'erllll(�J'a lIIeI"�/ola lIa Greenland seas. Poleo({'allogra/)/�)', 1 (3), 313 337. kOlllillelllokh Rami riilllale alld /JI'rmajrosl 011 cOlllillenl.1 NQ\'osibirsk, Lindstrom, D. R. and D. R. 1\lacAyeal. 1989. Scandinavian, Siberian. Izdatcl'st\·o "Nauka'·. and Arctic Ocean glaciation: effect of Holocene atlllospheric CO2 Goncharo\', S. B. 1986. Granitsa poslednogo oledeneni\'a na srednelll \·ariations. Sc;ellce, 245 (4918 . 628 631. Yeniscye polozhcni)e i \'ozraSt lLilllit of the last ice sheet in the Lindstrom. D. R. and D. R. i\lacAyeal. 1993. Death or an ice sheet. Illiddle reaches of the Yenisei ri\Tr: the posi tion and ageJ. Doki. :ikad. , \ (1/1111', 365 (644-3 ) ,21 4- 2 1 5. . \ allk SSSR, 290 6 , l·i-36 1+39. Lunck l\·ist. J. 198fl. La te \ \'eichsclian glaciation and deglaciation in Gross\I·ald. :'11.G. 1980. Late \\'cichselian ice sheet of northern Eurasia. Scandin",·ia. Qjwl. Sci. Rn' .. 5, 269-292. Q}lal. ReI .. 13 I I 32. i\/ac,\\eal. D. R. 1993. Binge/purge oscillations of the Laurenride ice Gross\I·ald. 1\1.G. 1988. Antarctic-style ice sheet in the Northern sheet as a cause of the North Atlantic's Heinrich e\·ents. Helllisphere: to\\"ard a nC\I' global glacial theory. Polar (:I'ogr. (;1'01.. I'aleoreallogra/)/�)', 8 (6), 775 784. 12 -, 239 267. :'IIanabe. S. and A. H. Broccoli. 1985. The influence of continental ice Gross\\"ald. :'I!.G. 1993. Extent and melting histol"\' of the Late sheclS on the climate of an ice age. J. (;eo/)/�rs. ReJ ., 90(D I ) . 2167 \\'eichsc-lian ice shect, the Harems Kara contincntal margin. III 2190. " Peltier, \\'. R., I'd. ICI' ill lhe dill/ale _:).111'111. Berlin. etc., Springcr-\'erlag, :'Iiarsiat, I. 1994. Simulation of the :\'orthcrn Hemisphere continental ice 1-20. , N . \TO .\SI Sel".. Scr. I. \'01. 12. shccts o\'('r the last glacial interglacial cycle: cxpcrinlcnts with a Gross\I·ald. :'11.G. 1994. The drumlin fields of the i\'(l\',,)"a I.emh-a Crals latitude longitude \'ertically integrated ice-sheet model coupled to a region and the Kara Sea ccnter of glaciation. Polar (,'('ogr. G(Jol .. 18\ 1), zonally a\"('raged climate model. PalaeociilllaleJ - Dola alld ,I/addlillg. 15-32. 1 1 1 , 59 98. Gross\I'alcl, 1\1.G. and \'. B. Spektol". 1993. [Glacial land lorms of the :'I1;l\T\l ski. P. A .. G. H. Denton and T..J. Hughes. 1981. Late Wisconsin Tiksi region \\"estern ("(last of the Buor Kha\ a Ba)". north Yakutia .J icc sheets or ;'\orth ,\merica. 11I Denton, G. H. and T.J. Hughes, I'd, . (;eoll/orjll logi)'a. 1. 72 82. 7 h,' llI.11 greal il"l' .lhl'l'i.\. i\'e\\" York. etc .. I\'iley-Inrcrscience, 67 178. Gross\I'ald, 1\1.G., \\' . Karlc'n. I.. Shishol'ina and 1\. Bodin. 1992. Glacial 1\iercer, J. H. 1970. i\ lormer ice sheet in the I\rctic Ocean? PaLaeogeogr. , ianclforn1s and the age of eleglaciation in the Tiksi area, east Siberia, Palafociilllollll. , 1'(i/(II'oeeol .. 81 1 I . 19-27. Geogr .11111., 74A 4 . 29:; 30-1. :'Ilillcl". G. H. and ,\. de \'ernal. 1992. \\' ill greenhouse \\"arming lead to Hamilton. T. I). and j. Brighalll-Creuc. 1991. The last illlerglaciation in '1onhern Helllisphere icc-shect gro\\"th? \·lIllIre. 355\6357) , 244-2+6. Alaska: stratigraphy and palco('cology of" potcntial sitcs. OJwlern(f}), '1"e. .J. I;. 1953. The f10\\' la\\" of ice rro m measurements in glacier IlIlemalillllal, 10-12, +9 71. tlll1l1el�. laboratory experiments and thc Jungfraufirn boreholc Ha"s, J. 0 .. J. Imbrie and ." .J. Shackleton. 1976. \'ariations in the experiment. Pro(. N. Soc. l,olJ(lolI. Ser. A. 219 1139 . 477 489. Earth"s orbit: pacemaker of the ice ages. Scima. 194 4270 ,11 21 1132. Pclto. :'11.S. 1992. Equilibrium line altitude \'ariations \\"ith latitude, Heiser, P. /\. alld 6 olitl'n . 1992. PleiltocClIC glacial geology of St. tocl,,) ancl duri ng tlte Late \\·isconsin. Polaeogeogr , I>alaeoclilllalol .. La\\"rence Island, .\Iaska. (;eol. Soc ..1111. 1992 .illlllllll .1 /1'I'lillg.. lb, lrarl Plllal'onal., 95 I 2 ) , ·t! 4·6. Il'ilit Progralll.l . A3+5 .\3+6. PC\I'C, T. L. 1975. Quaternan' gcologv of Alaska. C.S. (;I'oi. Sun'. Pro! Hopkins. 1). :'11. 1912. The pakogeograph) and climatic history of 1'''/1. 835. Beringia during Late C:enozoic time. loll'l"-. \·ord 12. 121 150. Rex, R. \ \'. 1961. Hyd rodynamic analvsis of circulation and orientation Hughcs. B. J\. alld '1'. .1. Hug- iles, 1994. Transgrcssioll�: ITlhinkillg" of lakes in nort hern Alaska. III Raasch, G. 0., ed. Geologl' oI lhl' lirclir. Bcringi'lll giaciation. Pa/afog('ogr,. JJa/aeoc/illlalol. . Palaeo{'(ol., 110 3 1"01. lIe Toronto. Cni\"CI"sit) "f Toronto Press. 1021 1043. + . 275 29+. Sa·Uclll. j.. D ..\. R. Pooic. L. Ellingsen and H. 1'. Sejrup. 1992. Glacial Hughes. T . . J 1985. Thc great (:enozoic ice sheet. Palaeugeogr .. gc"log\' of ou ter Bjorno\Tenna, southwestern Barents Sea . . Ifar. (,'1'01.. Polol'odilllalo!. , Pola!'ol'ro!., 50( I), <) ·13. 103 13) . 1 5 51. Hughes, T . .J. 1987. The marine ice transgressioll h) pOlhesis. (,'I'ogr.. il lll . . Sah-i!(sl"n, O. 1981. Radiocarbon dated raised beaches in Kong KarJs 69A 2 , 237 250. Lancl. S\'albard. and their conseq uences I()I' the glacial history or the Ilughes. T. 1992;1. .\brupt c1illlatic change relatcd to unstable ice-shect Barcnts Sea arca. (:eo,�r ..11111 .. 63A, 3 4, , 283 291 dyn Kleman . .1. 199+. Presel"\"ation of landlorms under ice shects and icc S\\"ecnc) . J. F. 1981 ..\r ctic sealloor structure and tectonic e\'olution. III caps. (;I'olllor/lholog)', 9, 19 32. Pa/eoreroll.)lruclioll oJ ' Ihe (olllillellls, \\'ashington. DC, .!- \nlt'fican Klcman, .l. and I. Borgst rolll. 199+. Ghcial land ro rllls incJicati\'C of K\·;\sO\·. D. I). 1979. 7 "1' 1,011' Qj(((Il'fI(ar)' "i..lul)' o/ I(II�l!,f lakl'J ((lid illllllld .,m.' Thomas. R. 11. and C. R. Be nill·Y. 1978. 1\ model 101' Holocene retreat of '!/({lIl l'fI( 1,·lIra/H. Helsinki. Suomalainen Ticdea katcmia. the West .\nt 331 Journal of Glaciology \·ogt. P. R .. K. Cranc and E. SunckoL 199+. Deep Pleistoccnc iccberg Yorrcn, T. O. and Y. Kristoflcrscll. 1986. Late Qu aternaryglaciation in plo\\'marks on the Yermilk Plateau: sidescan ancl 3.5 kHz evidcnce thc soulh-\\'cstcrn Barcnts Sea. Boreas, 15 ( 1 ) . 51 59. fo r thick caking ice fi'onts and a possible marine ice shcct in t h e \\·certman, J. 1 957. On the sliding of glilcicrs. J. G/a(lol., 3 ( 211 , 33 38. Arctic Occan. Gcoiogr, 22 5), 1·03- 40fi. \\'ecrt man, J. 197+. Stability of thc junction of an ice sheet and an icc \·olkO\·. 1. .\ .. :\1.G. Gross\\'ald ancl S. L. Troitskiy. 1980. The discharge shelf'. I (;/a(;ol.. 13 67 . 3 I I. of proglacial waters during the last glaciation of w('stern Siberia. \\'eertman, J. 1976. Giaciology's grand ul1Sokcd problem . . \alllre. 260. Polar (;eogr. Cco/., 4 I ) , I 1+. Translated from 1�1' . . Ikad. Ya uk ,',SSR, 284· 286. Sa. (,I'IW.. 1978, (+), 23 35. , \\'illiams, I). F . . \\ .S. :\Ioore and R. H. Fillon. 1981. Role of glacial \'orrcn. T. 0. 1992. Glaeiations of thc Barents Sea all O\·cn·ic\\,. S1'er Arctic Ocean ice shect in Pleistocelle o'ygcn isotope and sca IC\'CI Gcol. Clldm . . Sf!. Ca 81. 367 372. rccords. L'arlh P/allcl. Sri. Lell . , 56, 157 166. ,\IS received 22 . llarch 1994 and acce/lled ill revised Jo rm 21 Sejl/ember 1994 332