This is an unpublished and unfinished manuscript, begun in the spirit of east-west collaboration, later abandoned in a fog of linguistic confusion and poor communications. But the subject seems too interesting to keep under wraps, so in the interests of promoting wider discussion here is the unfinished manuscript, warts and all. Like it, hate it, let me know what you think about the ideas contained within it.


A GIANT SIBERIAN LAKE DURING THE LAST GLACIAL: EVIDENCE AND IMPLICATIONS

Lioubimtseva E.U. (1.), Gorshkov S.P. (1.) & Adams J.M. (2.)

1). Department of Physical Geography and Geoecology, Faculty of Geography, Moscow State University, Moscow, Russia. (E.U. Lioubimtseva is now at: Central Washington University. Ellensburg, WA 98926-7550, lioubime@cwu.edu)

2). Environmental Sciences Division, Oak Ridge National Laboratory, USA


Abstract

Radiocarbon dating evidence of fluvio-lacustrine sediments has confirmed the existence of a single giant freshwater lake covering most of the West Siberian Plain at around the time of the Last Glacial Maximum. Stretching some 1500 km from north to south, and a similar distance east to west at its widest points, at its maximum extent it would have had a surface area at least twice that of the Caspian Sea. Varying from several tens of metres to over 100 m in depth according to local topography, it would have contained of the order of hundreds of thousands of cubic km of water. Formed by the damming of the Yenisei and Ob rivers by an eastward lobe of the Ural and Putorana ice sheets, this mega-lake appears from the available dates to have reached its maximum extent by around 24,000 years ago, and to have existed in some form up until around 12,000 or 13,000 radiocarbon years ago.

The implications of the existence of this lake are manifold. It is likely that its presence would have greatly affected regional climates, and regional biogeography. Furthermore, the progressive retreat of the west Siberian lobe of ice would eventually have allowed the lake to drain northwards into the Arctic Sea, perhaps catastrophically and in several stages. If so, the volume of water released could have exceeded that from the late-glacial Lake Agassiz (Broecker et al 1989), or any other ice-dammed lake previously known to have existed. It is interesting to consider the effect that this rapid delivery of freshwater might have had on sea ice extent, on deep water formation in the north Atlantic, and perhaps on climate.


Introduction

Many aspects of the geology of the former Soviet Union remain obscure to Western scientists, to whom Russian is an unknown language and to whom this vast land area was formerly forbidden. Of this 'unknown' geology, few things could be more suprising than the possible existence of a vast freshwater lake, covering the major part of western Siberia north to 62 Deg.N, up until perhaps 13,000 years ago.

Modifying the ideas of earlier authors, Grosswald (1983) first put forward the idea that a vast system of glacial dammed lakes existed through northern Eurasia had existed relatively recently, during the LGM, with an outflow into the Black Sea and via the Mediterranean into the Atlantic Ocean. In doing so, he was building upon an earlier hypothesis of possible outflow during the maximum Quaternary glaciation from Siberia to Kazakstan, Central Asia, and the Aral, Caspian and Black Seas. However, this lake and outflow stage was usually attributed to the Early (Sigov 1958) or Mid (Zarrina et al. 1961) Quaternary.

Grosswald and Hughes (1995) concentrate only on the evidence for ice sheet extent at the last glaciation, with the existence and extent of lakes being an incidental result predicted from the distribution of ice dams. This paper, in contrast, aims to summarize the whole palaeogeographical context in terms of evidence for lake extent and duration as well as the existence of ice dams. Here, we review the most recent evidence concerning the debate over the nature and duration of this lake, and consider some of the possible implications which it may hold for understanding the ice-age world.

The Quaternary Geology of the West Siberian Plain, and its origins.

A vast area of the Western Siberian plain is covered mainly by Pleistocene alluvial and lacustrine deposits, which gives way to loess and loess-like formations on its southern margin (Fig.1, some key sites studied in previous published work of Gorshkov and others. Fig.2: Map indicating extent of the alluvial sediments).

In terms of its geomorphology and tectonics, this is a homogenous region of sedimentary accumulation lying above a Hercynian platform. The height above sea level rarely exceeds 150 m, with the exception of the Siberian Uval hills between the latitudes of 61° and 63° , reaching 300 m altitude. In the south-west, the Turgaj valley links the plain of the Western Siberia with those of Kazakhstan and Central Asia, while the northern edge of the plain slopes gently down to the level of the Kara sea.

Already at the beginning of the present century, various geologists had suggested that the alluvial and lacustrine sediments that mantle the West Siberian Plain could have been the product of a sequence of gigantic ice-dammed lakes, formed within the periglacial zone during the Quaternary continental glaciations. This idea has been further developed in more recent studies. For example, detailed reconstructions of the Middle and Late Pleistocene lakes over the whole Western Siberian plain were put forward by Zarrina et al. (1961) whilst Arkhipov and Lavrushin (1957) studied the evidence of ice-dammed lakes formed within the Yenisei river basin. Palynological and diatom analyses of lacustrine, alluvial and fluvio-glacial deposits from the Middle and Late Pleistocene exposures (Fig. 1) on the terraces of the Yenisei valley, were carried out by Aleshinskaya et al. (1964). Gorshkov (1967, 1986) collected vast amounts of data at sites across a very wide area of the Late Quaternary lacustrine and alluvio-lacustrine deposits over the West Siberian Plain in the zone between 55°,00' and 61°,5' N, indicating the general distribution of sediments across this area that is summarized in the map by Arkipov & Volkov (1991).

The idea that a vast system of glacial dammed lakes existed throughout Northern Eurasia during the LGM - rather than being confined to an earlier stage during the Quaternary - with an outflow into the Black Sea and further through the Mediterranean into the Atlantic ocean, is supported by the recent research of M.G. Grosswald (Grosswald,1983; Grosswald & Hughes, 1995). In this work, Grosswald has developed an earlier existing hypothesis of possible outflow during the maximum Pleistocene glaciation from Siberia to Kazakhstan, Middle Asia, the Aral, Caspian and Black seas. However, this event was usually attributed to the Early (Sigov, 1958) or Middle Pleistocene (Zarrina et al. 1961). In their recent paper, Grosswald and Hughes (1995) show two glacial dammed lakes on palaeogeographical maps of Western Siberia during the LGM. These lakes are shown as occupying the the Ob and Yenisei valleys, connected by Kas-Keta spillway only during the interval of the maximum extension of the ice sheet to the south and was not far from the Podkamennaya Tunguska river (see fig. 1). Still this reconstruction is based on the idea, that the Sartan (= Late Wurm) ice sheet never covered the Siberian Uvali mountains although it extended close to them.

Today there are increasing quantities of data demanding a revision of the age of the maximum glaciation - and hence the date of the maximum potential ice dam that could hold a lake in place - in Western Siberia. Goncharov (1986, 1991) found a layer with peat mass lenses of probable allochtanous origin under the upper moraine in the Yenisei„s valley. Peat samples were dated by 14C. Three dates were obtained from the outcrop in the lower course of Niznaja Sarchiha river (62 30 Deg.N): 40 270+- 450 BP, 51 500+-800 BP, 51 300 +_ 1600 BP, and 2 dates (43 400+-1080 BP and 41 200 +- 420 BP from the outcrop in the Komsa river mouth (61 50 Deg.N). Goncharov suggests that the age of the upper moraine should be younger than the youngest of the obtained dates taking in account not only stratigraphical position of the moraine but also a possibility of reburial of the ancient organics, which could increase their age.

Glacial deposits, overlaying the dated intermoraine horizon, are attributed by Goncharov to the Sartan Glaciation (the most recent, or oxygen isotope stage 2, glaciation). This allows the limit of the maximum ice sheet to be drawn at least as far east as the southern edge of the Siberian Uvali, in the Yenisei region. However, according to information of Grosswald on the Siberian Uvali, the maximum ice extent occurred during the Sartan Glaciation. In the light of these new views on the palaeogeography of ice sheets of Western Siberia during the Last Glacial many data on the pattern of deposition of sediments across west Siberia could find a logical explanation, in terms of a maximum lake phase occurring at that time.

The debate over the existence of the lake and its ice dam:

There is essentially no disagreement over the fact that some form of giant ice-dammed lake was present intermittently across western Siberia during the Quaternary period. The only uncertainty has been over the timing of the lake phases, and particularly those times when the lake reached its maximum extent.

It has generally been assumed that the giant lake phases in western Siberia occurred during the mid-Quaternary glaciations and no later. This view became prevalent because 1) the necessary ice dam was thought only to have been in place during the mid-Quaternary glaciations, with much smaller ice extent during late-Quaternary glaciations and 2) the 'terrace stratigraphy' chronology used to date the sediments from the lake phases also suggested a mid-Quaternary age. These issues must be dealt with here in turn, to set out the main arguments and sources of evidence:

1) The age of the ice dam. From the middle of this century up until the present, two opposing points of view on the extent of glaciation in Siberia have been manifest in the Russian palaeogeographical literature (Saks, 1953; Lavrushin, 1963; Arkhipov, 1971; Grosswald, 1977, 1983; Astakhov, 1976; 1989; Arslanov et al., 1983; Gorshkov , 1986; Arkhipov & Votakh, 1989; Spasskaja et al, 1993). According to some authors, during the Quaternary glaciations vast continuous ice dams formed across the northern part of the West Siberian Plain, completely blocking the riverine outflow from the Yenisei Basin northward to the Kara sea (Grosswald, 1983; Goncharov, 1989). Others believed that only minor glacial sheets were formed only over the most elevated parts of north-western Siberia, descending from the Putorana and Uralian centres of glaciation and leaving free the northward flow of the Siberian rivers (Astakhov & Isaeva,1985). This latter point of view, of less widespread glaciation, was dominant during the 1960-1970s until it was widely recognised that the configuration of glacial formations covering the deposits with a marine boreal fauna could not be explained without assuming the existence of a glaciation centre over northern part of the West Siberian Plain and the shelf zone of the Kara sea.

The existence at some time in the past of almost complete glaciation of the northern part of west Siberia having been to some extent accepted, the main points of the discussion then shifted to the precise age of maximum glaciation and of the accompanying ice-dammed lakes, and towards discovering exactly how large an area these covered (Spasskaja, Astakhov et al, 1993). However, until recently no good dating evidence had been forthcoming.

The dominant view came to be that the maximum glaciation in western Siberia dated back to the mid-Pleistocene, and more specifically the second - or Taz - stage of maximum glaciation. Both moraines, of the Samarovo (last glaciation) and Taz glaciations, appear in the outcrops on the slopes of the Yenisei valley in the most southern part of the former periglacial zone. The most extensive moraines were assumed to be only of Taz age.

This idea, that the maximum spread of the glacial sheet occurred only during the mid-Pleistcene, remained dominant for many years, and could be shaken only when a detailed geological survey of the left bank of the Yenisei river between 60° and 63° N was carried out. This work resulted in dating of intermoraine peats that fall between 51,300+1,600 BP. (GIN- 3349) to 41,800 +1,200 BP. (GIN- 3348) Goncharov, 1989). These dates enabled the upper moraine, which overlies the peat layer, to be attributed to the Sartan (= the last glacial or isotope stage 2) stage. However, on his palaeoreconstruction maps (and in contradiction to this dating evidence), Goncharov more conservatively assigned a Mid-Pleistocene age to the moraine, which is located on the most southern edge of the study area. It is in fact doubtful that this conclusion was correct.

The cause of the problem is that there are two moraines present in the outcrop named Oplivny Yar, 7 km south of the mouth of the Podkamennaya Tunguska river, a right bank tributary of the Yenisei (fig.1). On the left bank of the Yenisei the moraine deposits lay even higher hypsometricaly than on its right bank in Oplivny Yar. After this site the Yenisei makes a loop and then, before turning northward again, washes out a terrace scarp (called Zavalny Yar) on the left bank of the river. In this outcrop, like in Oplivny Yar, the upper moraine lays in the close parageneses with lacustrine deposits. But here, moreover, one can see the traces of glacial tectonics, interrupting both the glacial deposits and lacustrine formations. For example, in the outcrops there are overturned strata of boulder-pebble deposits, cemented by ferric oxides and hydroxides, crumpled in folds with fine layered lacustrine deposits. There are also steep inclined contacts due to faulting between sediments of different composition. It is noteworthy that crumpled strata, resembling isoclinal folds, also appear in the upper part of Hahalevsky Yar outcrop (fig.1) in lacustrine clays and silts of dammed-lake sediments that occur up to the southern limit of moraine deposits. It is this faulting and folding that has caused confusion in terms of the dating of these moraine and lacustrine deposits.

2) The age of the lake deposits. Crucial to reconstructing the history of the Quaternary lake phases have been the outcrops of Quaternary deposits in the Yenisei river valley (e.g. Panteleevsky Yar, Hahalevsky Yar, Oplivny Yar, Zavalny Yar, Bely Yar and Bakhtinsky Yar) between 61° and 63° N. There are other large outcrops in the Yenisei River tributary valleys, similar to those that occur in the Yenisei River valley itself.

Stratigraphical subdivisions of glacial dammed lakes in Siberia were based until recently on a generally accepted scheme of "terrace stratigraphy", for the terraces which occur in the Quaternary deposits of the Yenisei and its tributaries. This chronological framework assumed that the oldest and highest terraces in the valley were of mid-Pleistocene age. Thus, the largest glacial dammed lacustrine basin, defined by the highest terrace, was considered to date back to the maximum of Samarovo glaciation (Mid-Pleistocene). According to this chronology, the next lake phase, defined by a lower terrace, was medium in terms of its expansion and attributed to the Taz stage of the same glaciation, whilst the smallest lake phase was considered to date from the Zyrianka Ice Age (Late Quaternary). However, there are now good reasons for disputing this stratigraphical scheme.

Of key interest are studies of the 5-6m thick Urtamsk lacustrine deposits which cover several different terraces at different altitudinal levels in the Yenisei valley. These clays, silts and fine sands are also to be found at altitudes of 80-150 m or higher, in several parts of the Ob valley. Numerous radiocarbon dates from sites in both river valleys have attested that the age of the Urtamsk deposits is between 22,000 and 12,300 BP (Arkhipov et al 1973). This suggests that in fact the whole of the area delimited by these terraces has been covered by a lake during the last glacial phase.

Gorshkov (1986) has carried out a detailed study of sedimentation conditions of lacustrine clays and silts (about 3-4 m thick) from a series of cores on the right bank of the Yenisei valley 30 km south of the town of Yeniseisk (fig.2). This study revealed that analogous to the Urtamsk deposits, loess-like clay and silt sediments with bedded structure form the upper layer of the core on an 18 - 25 m high terrace. These same deposits can also be identified further up the steep valley slope and onto the second 35 - 50 m high terrace. Furthermore, they continue over the gentle slope and watershed area eastward from the Yenisei valley (fig.3). Thus, a local analogue to the Urtamsk deposits can be followed continuously from the level of the lowest terrace up to an altitude of 230-250 m. Within these deposits a tooth of Mammuthus primigenius (late form) was found at the depth of 3 m, suggesting a Late Pleistocene Karginsk to Sartan age (see table 1; Gorshkov, 1967). The age suggested by the mammoth tooth was confirmed by radiocarbon dates from 26,300 BP to 21,300 BP, since obtained for the upper layer of alluvial deposits on the first terrace (Gorshkov, 1986).

This lower terrace is covered by lacustrine loess-like silts and clays that contain large pseudomorphs along the ice veins. The presence of pseudomorphs is almost always associated with a very well preserved polygonal periglacial relief visible on the terrace surface. However, the upper layer of lacustrine deposits continues up and over the slopes, and the higher up the slopes one looks, the less pronouned these periglacial features are, and they are completely absent on the upper parts of the watershed. This suggests that the area of the uppermost lacustrine deposits was covered by a lake during the most recent maximum glaciation.

One may assume, from the studied exposures, that the covering lacustrine deposits are widespread in the valleys of the Yenisei, Ob and their tributaries, and they are probably developed also over lower levels of the watershed areas of the West Siberian Plain. More ancient generations of dammed-lake sediments are also present in this region, but this is another topic to be dealt with in a subsequent review.

In the basin of the Yenisei River, lacustrine deposits also cover low spurs of the Yenisei hills chain. In the zone between 57° and 58° N the plain surface that they form, rises up to and occasionally beyond 260-270 m, according to most up-to-date topographic maps. On the left bank of the river, the dammed-lake deposits continue over the Ob-Yenisei watershed and then go down into the valleys of the Ob and its tributaries.

It is thus noteworthy that the results on dating of the youngest dammed-lake deposits in the periglacial zone, using dating and correlation methods different from those of the terrace stratigraphy school, happen to coincide with a revision of views on the age of the upper moraine.

Distribution of lacustrine deposits in the Yenisei River valley.

In 1979-82 S.P.Gorshkov discovered conclusive evidence that the Žblue clay' sediments of the Yenisei river valley were of lacustrine origin, independently of their geomorphologic position within the valley. On the right slope of the Yenisei river valley, which is formed by Proterozoic granite-gneiss, the series of cores revealed green-brown, brown and grey-brown clays and silts, covering two low terraces and two high terrace-like surfaces in the Yenisei river valley and slopes dividing them (fig 4).

The sediments are of uneven thickness, and often have disturbed or deformed stratification. Tracing of horizontal extent features clays and silts on low terraces (WHAT DOES THIS MEAN???). The sediment layers differ from one another in terms of the quantity of dust paricles, sand grains and carbonate content. It seems probable thet this rock contained fine concretions of ice. The confused final pattern of stratification would have resulted from the melting of these ice bodies.

On the slope above the third terrace, clays and silts have fine, inclined bedding, parallel to the surface. The thickness of the layers is 1 mm and less.

The former view that the terrace and slope deposits were locally derived solifluction or deluvial deposits can be shown to be incorrect from the fact that their lithological composition is quite distinct from the metamorphic rocks that underly them.

It is noteworthy that on the left side of the Yenisei Valley the covering deposits are also underlain by rocks lithologically completely different from them: white quartz sands of the late Cretaceous. This lithological contrast is a good argument against the idea of a local origin of the silts and clays covering the valley slopes. Forming a single geological body, the Žblue clay' suite covers the area as a blanket, going down in the valleys and up over the slopes and extending on low watersheds.

Correlation of lacustrine and glacial deposits.

To the south of the former ice-sheet limit in west Siberia (i.e. about 61 30 Deg.N), a complex of glacial deposits gives way to lacustrine clays and aleurites. It is worth mentioning the character of this change. Close to this limit in the outcrop Zavalny Yar (61 35 Deg.N) the upper part of the glacial deposits consists of fine-stratified clays and silts with layers of sands and rare cobbles. Thickness of lacustrine-glacial deposits amounts to 75 m here. Under them in the depth interval 125-150 m there is a very cobbley red-brown continental moraine, which soon gives way to grey stratified clays and silts with isolated cobbles. On the border of glacial and non-glacial zones some glaciodislocationes were found in some horizons. E.g. in the outcrop of Zavalny Yar (left bank of the Yenisei river, 20 km lower to the mouth of the Podkamennaya Tunguska river) in the interval of absolute heights 130-137 m bizarre complex folds of fine sands have very steep inclined contact with underlying lacustrine-glacial clays. 1 km to the south, close to the outcrop with the above mentioned red-brown moraine the layer of fluvio-glacial pebble has vertical instead of horizontal position. Such glaciodislocations are found also in the lacustrine clays, silts and fine sands, exposed in the upper part of Khahalevsky Yar (60 km upper from 60 km upper from the mouth of the Podkamennaja Tunguska river) in the interval of absolute heights 100-130 m, e.g. on the same level that of the similar dislocations in the glacial complex.

About 125 km southward from Kahalevsky Yar, there is Simski Yar (right bank of the Sim river, 22 km from the mouth). In the interval of absolute heights 60-85m are grey clays with fine horizontal bedding, underlain by peats and below these by fine sands with silt layers. The total thickness of the whole lacustrine formation is 25 m.

The clays and peats contain diatom fossils. Studying the diatom flora in these sediments, Z.V.Aleshinskaja determined 68 fresh-water forms, belonging to 43 species. Bog forms predominate: 1. Eunotian praerupta Ehr., 2. Eunotia bigibba Kutz., 3. Eunotia monodon Ehr., 4. Ehr., 5 Neidium bisulcatum (Lagerst.) Cl., 6. Cymbella heteropleura Ehr. The absence of planktonic forms leads to the conclusion that clay sedimentation was occurring in a shallow dystrophic (oligotrophic??) lake. The presence of boreal and arctic-boreal species (NN 1,3,5,6) speaks of much colder conditions during this period. Pollen evidence also supports this conclusion (Aleshinskaja et al. 1964).

The lacustrine clays and silts feature an unusually high carbonate contents (1.2-8.6% CaCO4), containing marly concretions and white layers with high content of authigenic carbonate minerals. Some clay layers contain up to 40% of grains smaller than 0,001 mm, suggesting deposition in a rather deep part of the lake.

Coming back to the problem of correlation between glacial and lacustrine deposits, one can conclude that at least in the Yenisei River area at the south of the glaciation zone, lacustrine glacial deposits form the upper part of the glacial sediment complex, giving way to a Žblue clay' suite of lacustrine origin in the non-glacial zone. The upper altitudinal limit of both of these types of deposits was determined on the low offshoots of the Yenisei Kriaz: it does not exceed absolute heights of 250-260m. On the eastern edge of the Western Siberian Lowland (left bank of the Yenisei) the highest tops covered by lacustrine-glacial deposits are not higher than 220m. Further south, the Ob-Yenisei watershed - which is almost completely covered by lacustrine deposits between 61ƒ30 N and 57ƒ00N - is mainly from 150 to 180 m high, and only to the south of Lesosibirsk town attains to 260-270 m a.s.l. The latter figures seem to fix the upper limit of the glacial-dammed lake.

On the age of the lake deposits. The 'blue clay' layers have yielded several radiocarbon dates which specifically point to a Last Glacial age for the maximum lake phase. The base of the Žblue clay' suite does not cover the flood-plain and the first terrace of the Yenisei river, whose formation is dated by the end of Wurm. Moreover, on the second terrace of the Yenisei River, to the south of Yukseevo village (57ƒN), on the right bank a cover formed by fine-grain sands from 5 to 15 m thick is widespread. Besides this, on the second terrace, which is 18-25 m high, this suite forms vast fans gently descending from higher elements of the relief. At the bottom the terrace is featured by the presence of alluvium with the residuals of wood, dated by C14 as 26 300+-900 BP from the outcrop near Novonazimovo village and as 24 100+-300 BP near Yukseevo village (fig5).

Alluvium was formed at the end of Karginsky interstadial, which corresponds to the ??Upper Pleniglacial (Isotope Stage 2???). Lacustrine silts (Žblue clay' suite), and the covering alluvium, include large pseudomorphs along ice veins which accumulated during the Sartan (=Late Wurm) glaciation (Gorshkov, 1986).

The study of the structure of covering deposits on the third terrace and higher geomorphologic levels shows that the lacustrine sedimentation which covered the relief by lake deposits, took place also earlier than at the late Wurm times (WHAT IS THIS IN TERMS OF THOUSANDS OF YEARS AGO?? DID SEDIMENTION FINISH BEFORE THE LGM?). The above mentioned terrace is usually 35-45 m high near its edge. Besides altitudinal characteristics it can be identified also by the presence of socle (EXPLAIN THIS TERM???) formed by precainozoic rocks. the socle elevates over the Yenisei river for ca.5-8 m. It is covered by alluvial pebbles 9-10 m thick, giving place higher to sands sometimes, which are covered by lacustrine silts having very distinctive border with underlying deposits. These silts, to the south of Barobanovo village (56ƒ20íN) include from one to three horizons of buried palaeosols. Systems of pseudomorphs along the ice veins are identified under the present day soils. Because the Kazantsevo (Eem~ca 125 kyr) age of the lowest alluvial member of the third terrace is rather well proved (Gorshkov, 1986), one can believe that during the Late Pleistocene several inundations of the Yenisei valley by lake waters could took place.

I DON"T UNDERSTAND THE FOLLOWING BIT; HOW IS IT THAT THESE PROVE A LATE SARTAN AGE FOR THE LAKE DEPOSITS? This is a problem needing special discussion. For the goals of this research it is important to use a stratigraphic and palaeogeographical indicator, such as the upper system of pseudomorphs along the ice veins in combination with polygonal knobs micro-relief. The later consists of small knobs (microhills?) isometric in plane and divided by small cross-like linear depressions (fig6) FIND BETTER WORDS. Such microrelief appeared as a result of development and degradation of polygonal vein ices. Under depressions, which form a polygonal grid in plane, one can always meet pseudomorphs. Within the depression grid there are small knobs - places where the ground was pressed out during the periods of growth, development and extension of ice veins, and where its excess became visible after ice melting. For instance, on the left bank of the Yenisei near Yukseevo village (57N) in the outcrop of 16m terrace, it is visible that green-grey silts of the late Wurm 8-10 m thick, are crossed by pseudomorphs along ice veins. Their vertical length is 6-7 m. Here knobs, located between depressions with pseudomorphs, are elevated by ca 1.6m. Distances between tops of neighbouring knobs is ca 15 m. Such a microrelief appears on the second and third terraces and sometimes higher over gentle slopes. Total identity and excellent conservation of polygonal microrelief on the second and third terraces speak on the recent melting of ice veins from them. Moreover, data on the growth of ice veins synergetic with sedimentation throughout the same areas also prove that covering clays and silts on the second terrace and the upper part of covering deposits on the third terrace belong to the same - Sartan cycle of accumulation.

Conclusions: a last-glacial age for the most extensive lake phase.

1) Radiocarbon dating shows that the outermost complex of glacial moraine deposits in western Siberia is associated with the Late Pleistocene, and not with the Mid Pleistocene as was previously believed. The upper moraine correspond in age to a period several thousand years before and after the Last Glacial Maximum. Thus, the size of the potential ice dam seems to have been at its greatest at around this time.

2) Radiocarbon and zoological dating of the lacustrine deposits also seems to indicate that the maximum lake extent occurred during the last glacial period, and not the mid-Pleistocene as was previously believed.

From the extent of the lacustrine deposits, it appears that the lake which existed would have covered most of western Siberia, stretching about 1500 km from north to south (see map Fig.3), with several large islands of higher ground emerging from it. It is indeed suprising that such a striking feature of the late Quaternary world could have gone unrecognized.

The precise limits of this vast lake, or lake system, have yet to be defined. It is probable that it was fed, as at present, by the rivers from the mountains of southern Siberia and the watersheds of central Siberia. Whether it flowed northwards past the ice dam is unclear. From the existing topography of the area, and the reconstructed depth of the lake, it might have been connected to the south with the Caspian and Aral Seas through the Turgai Valley.

Given the strong evidence for greater-than-present aridity in the region during the last glacial (Velichko et al. 1992), it is perhaps suprising that such a large lake could form. It would seem that emphemeral rainfall (in heavy storms?) combined with very low evaporation rates was able to provide enough water to keep the lake filled. It is also notable that lakes tend to create a local climate of high humidity and rainfall which favours their own persistance; the larger the lake, the greater the reinforcing effect (Valdes P.. University of Reading, Pers. comm.). However, the lake may well have been covered by ice for most of the year.

Some possible implications of the existence of the Siberian giant lake phase during the last glacial.

The potential implications of the existence of a giant Siberian lake are manifold. As a modifier of the Northern Hemisphere climate system, it might have been significant. In addition to the localised effects described above, its effects on albedo might have contributed to the global climate of the Last Glacial. Presumably being completely frozen over in winter (all the indications from the molluscan and diatom fauna and flora are of boreal freshwater conditions; freshwater of course freezes somewhat more readily than brine), it would have acted as a major winter heat sink for the Northern Hemisphere, comparable perhaps with a large ice sheet in terms of its albedo effects. In summer, if all or some of the surface ice melted, the lake would possibly act as a major solar heat absorber by virtue of its lower albedo. There is clearly a need for GCM experiments to consider the effects of this lake on the LGM climate. Indeed, inclusion of this realistic feature of the ice-age world may help to clear up some of the disagreements between observed and predicted LGM climates.

Furthermore, the lake must at some stage have begun to drain northwards as the ice dam receded and as the northward-draining rivers regained their present drainage pattern. Whether this occurred suddenly (in catastrophic drainage events comparable to or exceding the Lake Agassiz events in North America: (e.g. Broecker et al 1989) or gradually is not clear. However, the existence of several lower terraces may suggest that it was concentrated in stages separated by longer periods of stable lake level. The exit of the lake's freshwater northwards into the Arctic Sea could potentially have had effects on sea ice formation. Minor variations in continental runoff into the Arctic sea at present are known to significantly affect sea ice and climate (Darby & Mysak 1993), and this influence could have operated on a far larger scale following lake drainage events. From the melting of the sea ice, there might also have been effects on deep water formation in the north Atlantic (Broecker et al 1989). The possibility of correlation between drainage events from the Siberian megalake and sudden changes in the Northern Hemisphere climate and ice flux (e.g. the sea-ice and ice sheet surges known as Heinrich events) should be considered as a hypothesis to be investigated.

There is a need for further fieldwork that will help clarify the detailed history of the lake's formation, its internal functioning and its subsequent drainage. Evidence (or lack of evidence) of catastrophic drainage events should be sought in the most likely areas of north-western Siberia and in the adjacent parts of the Arctic Sea. Much work clearly needs to be done before the true significance of this vast lake for the world of the Late Glacial can be fully understood.

((FURTHER REFS TO ADD INGrosswald M.G. (1983) Pokrovnije ledniki kontinentalnih shelfov) Ice sheets of the continental shelves.) Moscow: Nauka, 213 p. (in Russ.) Grosswald M.G. and Hughues T.J. (1995) Paleoglaciology's grand unsolved problem. Journ. of Glaciology, vol. 41, No.138, pp.313-332 Ananjev A.R. and Ivanovski L.N. (1949) Geologicheskije issledovanija v levobereznoj chasti r.Yenisei k severo-zapadu ot g.Yeniseisk. (Geological studies of the left-bank side of the Yenisei river valeey to the north-west of Yeniseisk town) - Report of geological survey 1:1M scale, Tomsk, 250 pp, unpubl.(in Russ.)

Goncharov S.V. (1991) Poslednije lednikovo-podprudnije ozera dolini Yeniseja. (The last glacial dammed lakes of the Yenisei river valley), Bull. of the Quaternary comission, No. 60, pp. 62-67 (in Russ.)

Gorshkov S.P. (1967) O stroenii i uslovijah formirovanija samarovskoi accumulativnoi ravnini vnelednikovoi zoni Priyenisejskoi Sibiri. (On a structure and formation conditions of Samarovo accumulative plaine of the extraglacial zone of the Yenisei adjacent Siberia). Bull. of the Quaternary Commission, No., pp.133-120 (in Russ.)

Gorshov S.P. (1973) Report on structural-geomorphological studies of 1:200K scale (geomorphological survey of the right bank of the Yenisei river between the Glotiha and Bahta rivers), Moscow, 115 p. unpubl. (in Russ)

Sigov A.P. (1958) Rannechetvertichnaja pereuglublennaja Tobolo-Uboganskaja dolina (Early Quaternary Tobolo-Ubogan overdeepened valley), Ed. All-Union Geographical society, No10 ))

References:

Aleshinskaya Z.V., S.P. Gorshkov, A.V. Minervin, L.A. Ragozin, L.G. Rekshinskaja, N.O. Ribakova and M.B. Sadikova (1964) - Quaternary core sites in the Yenisei river valley from Yartsevo to Podkamennaya Tunguska river (in Russ.) Tchetvertitchnije opornije razrezi dolini Yeniseja ot Yartseva do r. Podkamennaya Tunguska. - In: Questions of Krasnojarsk Region Geology (Voprosi geologii Krasnojarskogo kraja), pp.263-282

Arkhipov S.A. (1971) - Quaternary period in Western Siberia (in Russ.) Tchetverticnij period v Zapadnoj Sibiri. Novosibirsk: Nauka, pp. 332

Arkhipov S.A., L.V. Firsov , V.A. Panichev and L.A. Orlova (1973) - New data on the stratigraphy and geochronology of the Middle Ob terraces (in Russ.) Novije dannije po stratigrafii i geohronologii terras Srednei Obi. In: The Pleistocene of Siberia and adjacent areas (Pleistocen Sibiri i smeznih oblastei). Moscow: Nauka, pp. 21-34

Arkhipov S.A. and Yu.A. Lavrushin (1957) - On the question of the Yenisei river runoff during the maximum and Zirianka glaciations (in Russ.) K voprosu o stoke r.Yeniseia v periodi maximalnogo i ziranskogo oledenenij. - Izvestia Ac. Sc. USSR, ser. geol., N6, pp. 33-42

Arkhipov S.A. and M.R. Votakh (1989) - On the problem of subdividing Early Zirianka interstadials in Western Siberia (in Russ.) K probleme videlenija rannezirianskih mezstadialov v Zapadnoj Sibiri. In: Cainozoic in Siberia and North East (Kajnozoj Sibiri i Severo-vostoka). Novosibirsk: Nauka, pp. 24-46

Arslanov H.A., A.S. Lavrov and L.M. Potapenko (1983) - New data of the Late Pleistocene glaciation on the north of Western Siberia (in Russ.) Novije dannije o pozdnepleistocenovom oledenenii severa Zapadnoj Sibiri. In: Gaciations and palaeoclimates of Siberia during the Pleistocene (Oledenenija i paleoclimati Sibiri v plleistocene). Novosibirsk: Nauka, pp.27-35

Astakhov V.I. (1976) - Geological conditions of the Pleistocene glacial ice conservation (in Russ.) Geologicheskije uslovija zahoronenija plejstocenovogo lednikovogo lda na Yeniseie. In: Materials of glacialogical research: chronics and discussions. (Materiali glaciologicheskih issledovanij: khronika i obsusdenija) Moscow.: VINITI, vol.55, pp. 72-78

Astakhov V.I. (1989) Late Glacial sedementation conditions in the center of Western Siberia (in Russ.) Pozdnepleistocenovaja obstanovka osadkonakoplenija v centre Zapadnoi Sibiri. Novosibirsk: Nauka, 1989, pp.120-132

Astakhov V.I. and L.L. Isaeva (1985) On the age of glaciation down the Yenisei valley (in Russ.) O vozraste oledenenija nizoviev Yeniseia. / DAN USSR, vol. 283, N2, pp. 438-440

Broecker W.S., Kennett J.P., Teller J., Trumbore S., Bonnani G. & Wolfli W. (1989) - The routing of Laurentide ice-sheet meltwater during the Younger Dryas cooling event. Nature.

Darby M.S. & Mysak L.A. (1993). A Boolean delay equation model of an interdecadal Arctic climate cycle. Climate Dynamics v.8 p.241-246.

Goncharov O.V. (1989) - The Last Glaciation of Western Siberia and glacial dammed lakes in the Middle Yenisei basin. PhD. Theses (in Russ.) Posledneje oledenenie Zapadnoi Sibiri i lednikovo-podprudnie ozera v baseine Srednego Yeniseia (diss. cand. geogr.nauk). Moscow, 1989, 190 pp.

Gorshkov S.P (1967) - On structure and formation conditions of the Samarov accumulative plaine out of glacial zone in the Yenisei basin of Siberia (in Russ.) O stroenii and uslovijah formirovanija Samarovskoi accumulativnoi ravnini vnelednikovoi zoni Preyeniseiskoi Sibiri. Bull. of Quaternary Research Commission N 33, pp. 113-120

Gorshkov S.P. (1986) - Problem of the Pleistocene deposits correlation in the periglacial zone of Yenisei Siberia with those in the glacial zone (in Russ.) Problema sopostavlenija pleistocenovih otlozenij vnelednikovoj zoni Priyeniseiskoi Sibiri s otlozenijami v lednikovoi zone. - In: Quaternary glaciations in Middle Siberia (Tchetvertitchnie oledenenija Sredneii Sibiri). Moscow: Nauka, pp.45-101

Grosvald M.G. (1977) - The Last Euroasian Glacial Sheet (in Russ.) Poslednij Evroasiatskij lednikovij pokrov. - "Materials of Glaciological research: chronics and discussions" (Materiali glaciologicheskih issledivanij: khronoka, obsuzdenija). Moscow: VINITI, pp. 46-60

Grosvald M.G. (1983) - Glacial sheets on continental shelfs (in Russ.) Pokrovnie oledenenija kontinentalnih shelfov. Moscow: Nauka, 216 p.

Lavrushin Yu.A. (1963) - Alluvium of plain rivers in subarctic zone and periglacial areas of continental glaciations. - Moscow: Ed. by Ac.Sc.USSR, 266 pp.

Saks V.N. (1953) - Quaternary period in Soviet Arctic (in Russ.) Tchetvertitchnij period v Sovetskoi Arctike. 2d edition, Leningrad, Tr. NIIGA, vol 77, 120 pp.

Spasskaja I.I., V.I.Astakhov, O.Yu. Glushko et al. (1993) - Evolution of Landscapes and Climates of the Northern Eurasia. Late Pleistocene - Holocene; Elements of Prognosis. Vol.1 Regional Paleogeography (Ed. by A.A. Velichko). Moscow: Nauka, 102 pp.

Zarrina E.P., F.A. Kapelianskaja, I.I. Krasnov, Yu.M. Mikhankov, V.D. Tarnogradskij (1961) - Periglacial formation of the West Siberian plain (in Russ.) Periglacialnaja formatsija Zapadno-Sibirskoi nizmennosti. - "Trudi Vsegei", vol.4 pp. 25-31


"Who the heck is this Jonathan Adams, anyway?"

Anything to add? Send your comments, criticisms and suggestions to Jonathan Adams (jonathan@elvis.esd.ornl.gov)