at SciVerse ScienceDirect

Quaternary International 290-291 (2013) 57e67

Contents lists available

Quaternary International

journal homepage: www.elsevier .com/locate/quaint

Basin morphology and seismic stratigraphy of Lake Kotokel, Baikal region, Russia

Yongzhan Zhang a, Bernd Wünnemann a,b,*, Elena V. Bezrukova c, Egor V. Ivanov c,Alexander A. Shchetnikov d, Danis Nourgaliev e, Olga V. Levina c

a School of Geographic and Oceanographic Sciences, Nanjing University, 21 Hankou Road, Jiangsu 210093, Nanjing, PR Chinab Institute of Geographical Sciences, Freie Universitaet Berlin, Malteserstr. 74-100, 12249 Berlin, Germanyc Institute of Geochemistry, Siberian Branch Russian Academy of Sciences, Favorsky Str. 1a, Irkutsk, 664033, Russiad Institute of the Earth’s Crust, Siberian Branch Russian Academy of Sciences, Lermontov Str. 1a, Irkutsk, 664033, Russiae Faculty of Geology, Kazan State University, Lenin Str. 18, RU-420008 Kazan, Russia

a r t i c l e i n f o

Article history:Available online 3 December 2012

* Corresponding author. School of Geographic aNanjing University, 21 Hankou Road, Jiangsu 210093,

E-mail addresses: bwuenne@nju.edu.cn,(B. Wünnemann).

1040-6182/$ e see front matter � 2012 Elsevier Ltd ahttp://dx.doi.org/10.1016/j.quaint.2012.11.029

a b s t r a c t

The basin of Lake Kotokel, located along the eastern shore of Lake Baikal, Russia, has attracted severalscientific projects to investigate the climate, vegetation and lake history throughout the Late Pleistoceneand Holocene. However, little was known about its basin structure and sediment architecture. Echosounding and 3.5 kHz single frequency sub-bottom profiling were used to decipher the basinmorphology and seismic stratigraphy to a depth of approximately 50 m. The bathymetric map showsa very shallow lake of 4 m mean water depth and an almost flat lake bottom. A distinct elongated small-sized depression of up to 12 m water depth between the north-western coast and a small islanddeveloped along an NW-SE-oriented fault line. A total of 46 km of seismic profiles crossing the lake along12 transects shows that the bottom sediments consist of three different facies, which accords to previ-ously analyzed core sequences. Several distortions of sediment layers at various sites indicate tectonicallyinduced impact, which resulted in up to 3 m vertical offsets of sediment packages at local sites. Theoffsets indicate a probably still active fault along the western shoreline of the lake. Soft gyttja of theupper 6 m does not show distortions and may have obscured potential younger tectonic activity. Thesediments date to the Late Pleistocene. Small updoming features along the boundary between layers Iand II may be assigned to degassing processes or to seismic activity. River channel fills along the north-eastern coast are indicative of a lower lake level prior to 15 ka BP. The sediment stratigraphy indicatesthat suitable coring sites for paleoclimate studies are only located in the southern part of the basin wherealmost undisturbed sediments can be expected.

� 2012 Elsevier Ltd and INQUA. All rights reserved.

1. Introduction

Lake basins are of particular importance for paleoclimatestudies, as they are local sinks for a variety of sedimentary transportprocesses which contribute to the final deposition of materialwithin a lake. The geological setting andmorphology of a given lakebasin and its catchment, local to regional climate conditions andhuman activity influence such dynamic processes significantly (e.g.Wünnemann et al., 2010). They even may amplify site-specificdepositional patterns, if abrupt tectonic events, human-controlledlandscape development or extreme climate events are involved.

nd Oceanographic Sciences,Nanjing, PR China.wuenne@zedat.fu-berlin.de

nd INQUA. All rights reserved.

As a consequence, sediment architecture within a lake can varythrough space and time, as demonstrated by seismic surveys indifferent lake basins in China and other parts of the world (e.g.Lister et al., 1991; Vanneste et al., 2001; Colman et al., 2002;Colman, 2006; Dietze et al., 2010; Daut et al., 2010).

Lake records are widely applied to infer the lake’s history andrelated influencing factors by identifying various proxy data ob-tained from drilled sediment cores. They are preferably assigned toclimate change throughout the Late Quaternary. Many of the re-ported results refer to a single sediment record, assuming that theobtained data reflect the general depositional conditions within thesystem caused by climate impact. However, little is known howstrongly the basin morphology and the spatial variety in depositionmay have overprinted the site-specific sediment composition, thusinfluencing the interpretation of records.

During recent decades, many lake records from the Lake Baikalregion in southern Siberiawere investigated for paleoenvironmental

Y. Zhang et al. / Quaternary International 290-291 (2013) 57e6758

and paleoclimate purposes. They reveal millennial-scale oscillationsof the regional climate over several millions of years, particularlyduring the Late Pleistocene and Holocene (e.g. Colman et al., 1996;BDP-Members,1997, 2005; Karabanov et al., 1998; Prokopenko et al.,2001; Kashiwaya, 2003; Prokopenko and Williams, 2004; Demskeet al., 2005; Tarasov et al., 1994, 2005, 2007; Bezrukova et al., 2005;Swann et al., 2005). Krivonogov et al. (2004) provided a criticalcomparison of numerical-dated sediments around Lake Baikal andused the chronological framework for reconstructing Late Pleisto-cene and Holocene vegetation changes.

Lake research at the relatively small Lake Kotokel, located nearLake Baikal, became an important issue since two records from thesouthern basin of the lake and records from peat bogs around thelake reveal dramatic oscillations in vegetation history, climate andlake-internal bio-productivity (diatoms) throughout the last ca. 48ka BP (Bezrukova et al., 2008, 2010; Shichi et al., 2009; Tarasovet al., 2009). The two sediment cores KTK1 and KTK 2, recoveredabout 100 m separated from each other, show similar sedimentcomposition, but with some differences in the thickness and timeintervals of the individual sediment sequences. None of theretrieved cores reached the bottom sediments of the lake. To date,it remains unclear whether a further core could recover even oldersediments and might help to solve the problems of chronology anddifferent thickness of lithological sequences. Some of the differ-ences between the existing cores may be also attributed tounknown specific sedimentary conditions at the respective siteswhich could have influenced their stratigraphic order.

The application of seismic stratigraphy in lake basins isconsidered to reflect the depositional conditions through time andthus allows the reconstruction of syn- and post-depositionalprocesses, lake-level fluctuations and/or neotectonic impact(Niessen et al., 1999; Colman et al., 2002; D’Agostino et al., 2002;Schnellmann et al., 2002; Brooks et al., 2005; Anselmetti et al.,2006; Colman, 2006; Hofmann et al., 2006; Beres et al., 2008;Wagner et al., 2008; Oberhänsli et al., 2011). Therefore, the majorobjective was to identify the basin morphology and depositionalconditions of Lake Kotokel, potential distortions of sediments, andto evaluate the applicability of drill locations for new coringactivities.

Table 1Seismic profiles from Lake Kotokel, conducted in May 2011.

No Name Length (m) No Name Length (m)

1 S110522-1 2400 7 S110522-7 18062 S110522-2 676 8 S110522-8 9873 S110522-3 197 9 S110523-1 11,1894 S110522-4 566 10 S110523-2 58345 S110522-5 3586 11 S110523-3 40566 S110522-6 3410 12 S110523-4 11,687Total length 46,394

2. Study site

Lake Kotokel (52�490N, 108�090E; 453 m a.s.l, according to SRTMdata; 458 m a.s.l. according to Russian topographic maps) is a smalland shallow tectonic freshwater lake located close to the easterncoastline of Lake Baikal (Fig. 1). The lake basin of Cenozoic age(Florensov, 1960) is separated from Lake Baikal by a SW-NEstretching ridge (maximum height: 729 m a.s.l.). The shortestdistance between the lakes is about 2 km. The Ulan Burgasy Ridgewith elevations >2000 m forms the boundary to the east(Bezrukova et al., 2010). A 2 km2 rocky island is located in thenorth-western part of the lake. It belongs to the pre-Baikaliancrystalline complex (Galaziy, 1993). The closest distance to themain land is ca. 600 m. The area of the catchment is calculated as187 km2, of which the lake occupies 64 km2 (excluding the island).

The Kotokel rift basin is characterized by active recent geo-dynamics (Ufimtsev et al., 1998; Ten Brink and Taylor, 2002;Shchetnikov, 2007). It is surrounded by pre-Baikalian foldedcomplexes of metamorphic rocks and granitoides. Cenozoicformations occur in the southwestern and north-eastern part of thebasin catchment. According to Geological map 1:200,000, sheet N-49-XXV (Davidov, 1974) several SW-NE striking active faults crossthe lake area and its island and thus should have influenced thebasin morphology. However, according to the Baikal Atlas, scale

1:4000.000 (Galaziy, 1993), only a SW-NE directed major faulttouches the eastern part of the lake.

The vegetation in the area is a mixture of boreal coniferous anddeciduous forests with wetland communities including sphagnumbogs at flat near-shore locations on the southern side of the lake(Shichi et al., 2009). The densely vegetated area around the lakemay be also responsible for considerable input of organic matterand nutrients throughout the year, resulting in brownish waterwith high amounts of suspended particles and leading to a very lowtransparency of the water, as observed during the field survey.

3. Methods

As a detailed bathymetry of Lake Kotokel was lacking,a Hummingbird echo sounder with coupled GPS system and doublefrequency sonar signal transmission (80/200 kHz) was used. Intotal, more than 2000 measurements were conducted followingtransects according to Fig. 2. Several hundred data points from theseismic surveys were also included and enabled comparison ofmeasured water depth between both systems along parallel tran-sects. However, due to the very soft gyttja sediments at the inter-face between lake water and settled sediments, the measuredwater depth differed between 50 and 100 cm from manuallycontrolled measure points, depending on the sediment density atselected sites. The collected data points included the geographicposition, water temperature and water depth. They were exportedto MS Excel files for further processing by ArcGIS 9.3 software.Design of the bathymetric map used the Raster-to-Topo functionwhich produced the best and most reliable contour map comparedwith other procedures, e.g. spline or kriging. All elevation data arebased on SRTM elevation models which differ from the reportedelevations by about 5 m (Shichi et al., 2009; Bezrukova et al., 2010).

A geophysical survey was applied to reveal the shape of the lakebasin and to penetrate the sub-bottom sedimentary strata, usinga 3.5 kHz GeoPulse sub-bottom profiler, GeoAcoustics Ltd., UK,coupled with a high accuracy GPS (Trimble GPS SPS351). The sub-bottom structures are delineated using reflections from a select-able single frequency multi-cycle high power signal, transmittedfrom 4 transducers mounted on a towed platform.

Field exploration at Lake Kotokel in May 2011 used the followingmain parameters: signal duration time 100ms, sampling rate 60 ms,transmitter output 10 kW, 3.5 kHz, receiver low pass and high passfilter as 300 Hze7 kHz, gain as 6 or 9 dB, with TVG control. Thepenetration velocity of the transmitted signal was calculated toca. 1500 m s�1. Twelve tracks with a total length of more than46 km were obtained (Fig. 1, Table 1).

The entire collected profiles were played back in the worksta-tion to improve quality by using gradient filter, digital low and highpass filters, gamma correction, and TVG gain control with poly fitfunction. The playback profiles were compressed in the horizontalaxis and then changed to JPG files, processedwith threshold controland interpreted with the current software Photoshop CS4 andMapInfo.

Fig. 1. Overview of lake Kotokel and its catchment in lake Baikal region. A: Overview Russia; B: Lake Baikal with associated major faults after Shchetnikov et al. (2012), modified; C:Relief, morphology and bathymetry of the Kotokel basin with locations of the drill cores KTK 1 and 2, seismic tracks and locations of the figures. Red lines: Seismic tracks presentedin figures; black dotted lines: Seismic tracks not shown and lake catchment; boxes display the location of the figures. Thin arrows mark the direction of seismic tracks in therespective figures.

Y. Zhang et al. / Quaternary International 290-291 (2013) 57e67 59

4. Results and discussion

4.1. Basin morphology

Most of the relatively small-sized Kotokel Basin is occupied bythe lake. Its catchment shape is dominated by steep slopes of up to28% (w13� inclination). Fringes of alluvial plains between the ridgeand the lake shore are very small or even absent (Fig. 1), indicatinglimited sediment transport and deposition along the slopes.However, a larger flat area of alluvial/fluvial deposits at the north-eastern side of the basin has been formed by fluvial activity alongthe Istok and Kotochik Rivers. The slope of this plain to north andtowards the lake is about 2.7%, without a characteristic fan shape.The plain may be associated with floodplain development. Therivers are the only perennial drainages within the catchment whichpartly (seasonally) drain to the lake. Due to low velocity flow, theirtransport capacity is restricted to suspended load. A further smallperennial creek is located at the southern side of the lake. All othervalleys facing towards the lake provide water along small drainagesonly during rainy seasons.

As mentioned by Shichi et al. (2009), the southern side of thelake basin is bordered by an 8 m high fossil bluff, composed of

fluvio-lacustrine sand covered by eaolian (dune) deposits. Itslocation between peatland on both sides indicates the formation ofa sand bar during a former highstand of the lake. Several shorelinesaround the lake perimeter of up to 1.5e2 m above the present lakelevel (2011) are evidence of fluctuating levels during the pastdecades as reported by the local people. A 2-m increase in waterlevel would reach the foot of the bluff. Traces of higher lake standshave not been found, and might be impossible as the lake wouldflow out towards Lake Baikal via the Kotochik River if the levelexceeded 456 m a.s.l.

According to the bathymetric map (Fig. 1C), a morphologicalseparation of the lake into two sub-basins (Shichi et al., 2009) isunlikely. Both the southern and northern part are similar in basinmorphology, comprising an almost flat lake floor with mean waterdepth of ca. 4 m. However, the lake floor in the northern partinclines by about 1.7e1.8% to the west, reaching 5 m water depthclose to the island (Fig. 1C). As this slight dip corresponds with thealluvial plain at the north-eastern shore of the lake, it is assumedthat this plain formerly extended for about 2 km into the lakeduring a lake low stand. Hence, lake deposits there should berelated to fluvial transport activity, implying coarser sediments andhigher accumulation rates than in the central parts.

Fig. 2. Seismic stratigraphy along the transect S110523-1 with lithology and chronology of core KTK 2 (after Bezrukova et al., 2010). A: Original seismic profile with marked area fordetail 1; B: Continuation of profile with marked areas for details 2 and 3; C: Stratigraphy and chronology of core KTK 2.

Y. Zhang et al. / Quaternary International 290-291 (2013) 57e6760

The previously reported “complicated bottom relief” (Shichi et al.,2009, p. 99) of the northern sub-basinwith amaximumwater depthof 14 m can be only partly confirmed. The bathymetry map (Fig. 1C)shows an important elongated depression in thewestern part of thebasin between the main land and the island. Maximum measuredwater depthwas here close to 12m (441m a.s.l.). This comparativelydeep channel with symmetrically steep flanks consists of three iso-lated depressions, separated by barriers. They are definitely notformed by fluvial processes but probably by tectonic impact, asseveral faults were also detected in the seismic profiles (see below).The depressions terminate in the center of the lake (Fig.1C) and passinto a very flat lake floor towards south, not exceeding 4 m waterdepth. The center of this flat area was selected for coring of twosediment cores KTK 1 and 2, which have been recently analyzed (e.g.Tarasov et al., 2009; Bezrukova et al., 2010).

4.2. Seismic stratigraphy

From the 12 seismic profiles, comprising in total more than46 km length, the most representative ones were selected todemonstrate the general sediment stratigraphy within the lake andits architecture. These transects cross the lake from southwest tonortheast, passing the central parts of the basin and the drilllocations of KTK 1 and 2 (S11023-1: track A; S110523-2: track B,Figs. 1e3). A second transect surrounds the island and continuesSW along the western coastline of the lake (S110523-4: track C,Figs. 4e6).

According to the profiles, three different units of acousticreflections could be identified, developed in the upper 20e25 mbelow the lake surface. They are assigned to Units IeIII indescending order from the top of the lake sediments to the bottom.

Sediments in Unit I always appear as reflections with lowimpedance contrast (low density), involving partly parallel to sub-parallel reflections and diffuse backscatter signals (e.g. Fig. 2) whichmay be most likely assigned to acoustic artifacts due to veryshallow lake. This unit indicates an up to 8 m-thick layer of uniformsediments with low density. Generally, the interface betweenwaterand sediment forms a distinct boundary between high and lowimpedance contrast and thus can be easily defined. However, insome cases (e.g. Fig. 3) the boundary remains unclear, as themeasured water depth differs remarkably from the acoustic signal.Towards the basin margins the thickness of this unit decreases toless than 2 m and onlaps underlying sequences.

Layers of Unit II appear as densely-spaced high contrast reflec-tions of about 1e2 m thickness. Parallel-running reflections indi-cate almost undisturbed layers, forming an overall onlap pattern.However, differences occur along track C, where cross-beddedsignatures, sigmoidal clinoform and downlap structures dominate(Fig. 5C), indicating different sediment composition. They areassigned to sub-Units IIaed.

Unit III belongs to the downward parts of the entire sequenceand is partly masked by reflection multiples. Densely-spacedreflections occur as chaotic and occasionally hummocky structureof >10 m thickness, partly alternating with reflections of low

Fig. 4. Seismic transect S110523-2 along the north-eastern side of lake Kotokel. A: Original, B: Marked units with location of close-up figure, C: Detail of marked area in B.

Fig. 3. Detail 3 of seismic profile S110523-1, lake Kotokel, (see Fig. 2). A: Original; B: With marked layer boundaries.

Y. Zhang et al. / Quaternary International 290-291 (2013) 57e67 61

Fig. 5. Seismic transect S110523-4. A: Compressed original profile with boxes of detail Figs. 1 and 2a/2b; B: Detail 1 e original transect (S110523e4.2); C: Detail 1 with marked layerboundaries.

Y. Zhang et al. / Quaternary International 290-291 (2013) 57e6762

contrast. The latter may be masked by degassing processes, asdescribed by Dietze et al. (2010). The unit forms a distinct boundarytowards the overlying sequence.

4.2.1. Transect S110523-1 (Track A)This transect is about 11 km long and displays the sediment

structure crossing the central part of the lake and the core site KTK 2(Fig. 2). Themain part shows undisturbed and horizontally orientedunits with clear boundaries between them. A detailed close-upfigure (Detail 2 in Fig. 2B) displays the units around the drill siteKTK 2. According to the results of sediment analysis (Tarasov et al.,2009; Shichi et al., 2009; Bezrukova et al., 2008, 2010, Fig. 2C); thelithology comprises i) soft brownish-black gyttja (0e660 cm depth,Unit I), ii) grayish-black slightly laminated clay (660e740 cm depth,Unit II) and iii) grey silty clay followed by dark-grey silty clay (740e1010 and 1010e1253 cm depth, respectively, Unit III). Comparedwith the seismic stratigraphy, Units I and II can be assigned to theupper two layers of the sediment record. Low impedance contrast ofUnit I results from the occurrence of soft gyttja on top of the section.Any distortions of sediments do not exist or remain invisible.According to the chronology, this sequence has been depositedduring the Late Pleistocene transition and the Holocene (e.g.Bezrukova et al., 2010). Parallel-oriented reflections with highimpedance contrast in Unit II can be attributed to the appearance oflaminated dense clay, which seems to be almost undisturbed. They

represent a deposition during the Late Glacial period. Conversely,the two silty clay layers in the KTK 2 profile differ only in color andhence cannot be clearly distinguished by variations in reflectioncontrast but probably by the orientation of the reflections. Theupper silty clay layer seems slightly stratified (Unit IIIa), whereas thelower one appears as diffuse reflections without any orientation(Unit IIIb). The boundary between both layers, however, seems to beundulated and partly unclear. According to the seismic stratigraphyit is likely that the Unit III with a comparable lithology continuesdownward for several tens ofmeters, althoughmasked bymultiplesat ca 26 m depth.

Surprisingly, the layers of Units II and III at the beginning of trackA (Detail 1 in Fig.2A) are highly distorted with strongly undulatedboundaries along the units, whereas the toplap sediments of theupper Unit I remain unaffected. The seismic reflections in Unit IIindicate sub-parallel structures with signs of pseudo clinoformfacies. A distinct updoming of sediments occurs in the middle partand indicates locally vertical flow of sediments. Slightly upliftedsediments nearby are of minor intensity but display a sharp verticaloffset of approximately 1 m in height. As both structures also occurwithin the multiple reflection layers, they appear as significantfeatures within this part of the track. Undulated and distortedsequences within generally diffracted signal contrasts are charac-teristic for Unit III. Furthermore, the boundary of multiple reflec-tions do not follow the surface architecture of Unit II and thus may

Fig. 6. Details 2a and 2b within seismic transect S110523-4 (for location see Fig. 5). A: Original; B: With marked layer boundaries, C: Continuation of A; D: With marked layerboundaries.

Y. Zhang et al. / Quaternary International 290-291 (2013) 57e6764

include further sedimentary structures of Unit III, tracing down to>30 m depth.

As this part of the track is located in the center of the lake faraway (ca 1.5 km) from direct influence by near-shore or on-shoreprocesses, seismic events/tectonic impact has caused the distor-tion of the layers. Taking the current chronology and the almostunaffected drape layer into consideration, this event happenedbefore the deposition of the upper gyttja sediments and thus datesback to the Late Pleistocene (most likely younger than 15 ka BP).

Detail 3 of track A (Figs. 2A and 3) shows the southernmost partof the track, passing the near-shore location there. The moststriking features here are the close contact to near-shore deposi-tional conditions and a prominent vertical offset (normal fault) ofUnits II and III by about 3 m. The upper layer (Unit I) close to theshore comprises ca. 2 m thick sediments with alternations betweenlow and high contrast. Towards northern direction this unitincreases to about 6 m thickness. Unit II can be divided into twosub-units (II a and II b, Fig. 3) of potentially different composition.Whereas sediments in Unit IIa occur as densely-spaced highcontrast reflections (onlap feature), layers of Unit IIb showsigmoidal and sigmoid-oblique clinoforms close to the shore line.They indicate sediment transport from the land and sub-aqueousnear-shore deposition. They most likely differ from clay depositsdescribed at location KTK 2. Sediments of the underlying Unit IIIcannot be differentiated in detail and generally show diffractedreflections with few signs of undulation. Very low impedancecontrast beneath the normal fault may be due to gas appearancethat masks all reflections. The tectonically induced subsidence ofthe lower layers in Units II and III falls in the same period assumedfor the distortions in Detail 1 of this track.

4.2.2. Transect S110523-2 (Track B)This 5.8 km long transect covers the north-eastern part of the

lake from the near-shore region towards the beginning of track B(Figs. 1 and 4). Water depth did not exceed 3 m. A cursory viewalong the seismic profile indicates a very flat and homogenous lakebottom (Fig. 4A). Some important differences to the previoustransect are worth to be noted. First, the densely-spaced highcontrast of reflections marking the water column does not matchthe truewater depth. Parts of this parallel structure represent Unit I.Perhaps acoustic artifacts dominate this upper part of the profile.However, the proximity to the inflowing region of the Istok Rivermay have influenced the sediment composition of the gyttja whichnormally occurs as low impedance contrast as seen in the lowerpart of Unit I. This assumption however, needs field confirmation. Inthis transect, Unit I is only 2e4 m thick.

Unit II appears as reflections with high contrast, almost hori-zontally orientedwith clear unconformity surfaces at the upper andlower boundaries. However, sigmoidal clinoform structures aredeveloped close to the northern shoreline and indicate fluvialdeposition there. Two striking features occur close to the presentdrainage area of the Istok River, and thus may be attributed toformer river channels that emerged as drainage pathways duringa former low stand of the lake. The difference between the channelbottom and the present lake level is ca. 10 m. These channels weresuccessively filled by the sediments of Unit II. According to thechronology (Fig. 2C) these features may have formed during thefinal stage of the Last Glacial Maximum (LGM) between ca. 20 and15 ka BP.

Unit III can be divided into two sub-units mainly based on slightchanges in reflection contrast. A sub-aqueous prograding delta inchaotic clinoform seems to have developed close to the northernshore, along with the deposition of sediments in Unit III. A differentorientation of reflections at 20e30 m depth in the north-western(left) part of the track indicates the continuation of sediments in

Unit III down to>30 m depth, although this area is already maskedby multiples.

4.2.3. Transect S110523-4 (Track C)Transect S110523-4 runs over a distance of 11.6 km along the

western part of the lake from south-west to north-east and passesthe north-eastern shore-regions of the island (Fig. 5). Similar to thetransect crossing the center of the lake (Fig. 2), the first 3 kmpassageshows an almost flat lake bottom, and the same divisions of layers(Units IeIII) without remarkable distortions. However, the seismicstratigraphy close to the lake shore (middle part of the track, Fig. 5A:Detail 1) changes towards a series of high contrasts with differentstructure, indicating fluvial deposits with typical cross-beddedfeatures (Unit IIc) and downlap facies (Unit IIb) beneath a 4 m thicksequence of low contrast reflections (Unit I). These units indicate fandevelopment covering bedrock. Unit IIa (onlap/sigmoidal clinoform)is truncatedhere. Reflectionsofmediumcontrast inUnit IId ca.100mfarther south do not show strong differences to the underlying UnitIII and thus may be composed of similar material, which can beidentified as a sand bar. It developed lakeward close to the formercoast and in close proximity to the former fluvial fan. However, itsrelatively sharp boundaries at both sides may also indicate tectoni-cally induced slight offsets. All features can be considered as old (LateGlacial) and are not recently formed.

Approximately 300 m further north-east, Unit IIa (onlap facies)continues and appears as horizontally-oriented densely-spacedhigh contrast reflections as described before. However, the entiresequences experienced a distinct vertical offset of about 2 m overca. 400 m distance (Fig. 5C) due to a rapid uplift (perhaps singleevent). Slope material along the uplifted flanks is indicative ofa tectonic process prior to the deposition of Unit I.

Close to the island, a further 3 m high offset of the entire sedi-ment package appears as a sudden uplift, probably contemporarywith the previously described sequence and confirms importanttectonic impact. The uplift also affected nearby sediments bydistinct displacements along the fault zones (Fig. 6, Detail 2b), asrecognized by deformed structures of low/medium contrast.

Further striking features in this section are hummocky mounds(updoming features) which appear as high contrast reflectionswithin Unit I (Fig. 6 AeD) ca. 13 m below the lake surface. They areassociatedwith layers fromUnit II and formsinglemoundsof ca.10mwidth and 2e4 m height. It is uncertain how these mounds wereformed. A possible explanation could be that gas expulsions affectedthe upper layer of Unit II and forced the sediments to inject into theexisting soft gyttja. Deformed structures within the mounds areevidence of updoming from the bottom layer. As these mounds arerelatively large, ascending gas pressure must have been very strongto induce updoming. Low impedance contrast in the vicinity of thesemounds indicates the presence of degassing processes which havemasked respective sediment layers (Unit III) and are potential causesfor post-sedimentary changes in sedimentary structure. The pres-ence of abundant organic matter and its decomposition on the onehand aswell as ascending gases from the deeper underground alongfracture zones may be possible sources for gas emissions.

As these features only occur between the faults along thewesterntrack C (Figs. 5 and 6), seismic shaking/tectonic movement in thatregion not only caused the offset of sediment packages but alsoupdoming processes of layers from Unit II. This probably may havealso occurred before the upper layer (soft gyttja) was deposited. Inthis case the deformation happened during pre-Holocene time.

A further explanation could be that during a lake low stand (lessthan 1m thickwater body) or even complete desiccation periglacialprocesses could reach the lake bottom in winter time and inducedthe formation of frost mounds, which were preserved before theupper sediments (Unit I) were deposited. However, as the

Y. Zhang et al. / Quaternary International 290-291 (2013) 57e67 65

occurrence of these features are restricted to the fault zones and notwidespread over the entire basin, the hypothesis of periglacialprocesses seem to be unlikely, although a respective lake lowstandseem to have existed during the LGM.

4.3. Spatio-temporal sediment distribution pattern and tectonicimplications

The seismic profiles indicate that the three depositional unitsare distributed over the entire lake basin. The soft gyttja reachesmaximum thickness of 7e8 m in most parts of the lake, thinningout towards the shorelines. The 6 m isopach roughly follows the1.5 m isobath, except for the north-eastern part of the lake, wherethe gyttja is considered to be markedly thinner (Fig. 7), probablydue to the influence of the Istok River. This gyttja (Unit I) wasdeposited from the Late Glacial (ca. 15 ka BP) throughout theHolocene (Bezrukova et al., 2010) and leveled the older undulatedlake bottom relief. Conversely, Unit II appears within the entirelake basin as a 3e4 m thick layer of partly laminated clay and siltyclay or sandy material at near-shore locations. This layer followsthe pre-relief of the lake bottom and may have been formed aslake sediments during the Late Pleistocene between 25 androughly 15 ka BP, according to the chronology as shown in Fig. 2.Unit III sediments, composed of black silty clay, occur in the entirelake basin and form the bottom lake sediments of unknownthickness. The seismic profiles indicate that this sequence maycontinue down to 20 m sediment depth or even deeper. They dateto older than 32 ka BP (Bezrukova et al., 2010). As this unit ismasked by reflection multiples, a reliable calculation of the totalthickness remains uncertain. However, they indicate a long termdeposition within an existing lake during MIS 3 or even olderperiods. Channel formation at the north-eastern side of the lakeare evidence of higher fluvial activity of the Istok drainage thateroded the sediments of Unit III during a time of approximately10 m lower lake level than today. According to the chronology, this

Fig. 7. Lake Kotokel, bathymetry, faults and sediment distribution. A: Detected andassumed fault lines (with numbers) crossing the lake basin; B: Distribution pattern ofgyttja > 6 m thickness within the lake basin.

phase of fluvial activity should have occurred >25 ka BP and thusfalls into the period of pre-LGM.

The most striking features of all seismic profiles are related totectonic activity that also affected the basin structure and itssediments. This influence is not very surprising as the Baikal Riftsystem and its attached small rift basins such as Kotokel basin aredominated by extensional tectonics with strike-slip components(Ten Brink and Taylor, 2002) which are still active (Déverchère et al.,2001). Although mapping results of faults in the vicinity of LakeKotokel differ in location and strike direction (see Davidov, 1974;Galaziy, 1993; Shchetnikov et al., 2012), they all show the generalframework of SW-NE trending large-scale tectonic structures thatinfluence the entire region (see also Fig. 1B). The finding oftectonically induced sediment offsets in Lake Kotokel can beregarded as continuations of faults that cross the western part ofthe lake basin, documented by the sediment Units II and III alongthe track A (No. 1 in Fig. 7A). They exactly follow the fault south ofthe lake basin that was mapped by Davidov (1974) and Shchetnikovet al. (2012). A second fault (No. 2 in Fig. 7A) seems to run at 90� tofault no.1, following the elongated depression between the islandand land. This may indicate that the island was separated from theland by tectonic movement. A third fault (No. 3 in Fig. 7A) in theeastern part of the basin can be assumed from sediment distur-bance and offsets in the northern part of track A. Its directionremains unclear, but may continue southwestward towards thefault line on land which was mapped by Galaziy (1993). However,tectonic structures crossing the southern part of the lake and coresite KTK2 could not be traced. A further fault seems to occur in thenorth-western part of the basin as indicated by small sedimentoffsets in a short seismic track (Fig. 1C, not shown as figure).Following the bathymetry of the lake, it seems plausible that thisfault may continue southwestward towards fault No. 2. However,the direction of faults nos. 2 and 4 need to be confirmed by furtherinvestigations. All documented faults affected Units II and III byvertical offsets but not Unit I. The soft gyttja of Unit I mutedmovements and prevented their preservation. Furthermore thevery low seismic reflectivity of Unit I may be also a reason that theyare not displayed. Taking the recent seismic activity in the Baikalregion into consideration, it is assumed that the faults around LakeKotokel are still active. However, the oldest tectonic movement,preserved in the sediments must have started after the depositionof Unit II layers ca. 15 ka BP ago, as they are the youngest visiblesequences affected by offsets.

5. Conclusion

Detailed studies on basin morphology and seismic stratigraphyof Lake Kotokel along selected transects demonstrate that therelatively flat lake bottom is the result of Holocene sedimentationwhich leveled former undulated surfaces of Late Pleistocene age.The seismic stratigraphy at the drill site KTK 2 could be confirmedby lithological investigations (Bezrukova et al., 2010), which sup-ported the extrapolation of similar layers at many sites within thelake basin but with differences in individual thickness. However,the finding of tectonic impact, revealed by sequential verticaloffsets of entire sediment packages (up to 3.0 m) indicate thatknown faults along the western and eastern margins of the lake(e.g. Davidov, 1974) continue within the lake. Comparable verticaloffsets were also found in the other seismic profiles not shownhere. They were most likely active since about 15 ka BP. Howeverdeformations within the Holocene sediments could not be traced,although the area is considered to be still tectonically very active(Ufimtsev et al., 1998; Déverchère et al., 2001; Ten Brink and Taylor,2002; Shchetnikov, 2007). As a result, only a few sites containundisturbed sediment sequences covering the Holocene and Late

Y. Zhang et al. / Quaternary International 290-291 (2013) 57e6766

Pleistocene depositional units. The penetration depth of theacoustic signals andmasking effects bymultiples indicate sedimentthickness of minimum 30 m at least. Fossil fan deposits could beidentified along some near-shoreline locations. There is alsoevidence for fluvial impact from the Istok River towards the lake asremnants of river beds extend ca 2 km into the lake basin. Theywere most likely active during pre-LGM time under a much lowerlake level. According to seismic stratigraphy, promising sites forfurther sediment drilling are restricted to the southern part of thelake basin.

Acknowledgements

This work contributes to “Bridging Eurasia” research initiativesupported by the German Research Foundation (DFG TA 540/4-1,TA 540/5-1), Center of International Cooperation, Freie UniversitätBerlin and Russian Foundation for basic research (project N 12-05-00476) and by Baikal-Hokkaido Archaeology Project (BAHP)-theMajor Collaborative Research Initiative financed by Social Sciencesand Humanities Research Council (Canada).We are thankful to Prof.Pavel Tarasov who initiated the first Bridging Eurasia workshop atthe FU Berlin in May 2010 and helped to organize the cooperationbetween the Chinese, German, Russian and Canadian partners andsupported us in discussing the obtained results. Furthermore wethank Prof. Andrzej Weber and Andrea Hiob (University of Alberta,Canada) and Prof. Mikhail I. Kuzmin (Institute of Geochemistry,Russian Academy of Sciences, Irkutsk), for fruitful discussions andadditional financial support. Special thanks are also given toMarinaKhomutova, RAS Irkutsk who strongly supported us in adminis-tration and customs procedures.

References

Anselmetti, F.S., Ariztegui, D., Hodell, D.A., Hillesheim, M.B., Brenner, M., Gilli, A.,McKenzie, J.A., Mueller, A.D., 2006. Late Quaternary climate-induced lake levelvariations in Lake Petén Itzá, Guatemala, inferred from seismic stratigraphicanalysis. Palaeogeography, Palaeoclimatology, Palaeoecology 230, 52e69.

BDP-99 Baikal Drilling Project Members, 2005. A new Quaternary record of regionaltectonic, sedimentation and paleoclimate changes from drill core BDP-99 atPosolskaya Bank, Lake Baikal. Quaternary International 136, 105e121.

BDP-Members, 1997. Preliminary results of the first drilling on Lake Baikal, Bugul-deika site, southeastern Siberia. Quaternary International 37, 3e17.

Beres, M., Gilli, A., Ariztegui, D., Anselmetti, F.S., 2008. The Lago Cardiel basin,Argentina (49�S), origin and evolution revealed by high-resolution multi-channel seismic reflection studies. Journal of South American Earth Sciences 25,74e85.

Bezrukova, E.V., Krivonogov, S.K., Abzaeva, A.A., Vershinin, K.E., Letunova, P.P.,Orlova, L.A., Takahara, H., Miyoshi, N., Nakamura, T., Krapivina, S.M.,Kawamuro, K., 2005. Landscapes and climate of the Baikal region in the LateGlacial and Holocene (from results of complex studies of peat bogs). RussianGeology and Geophysics 46, 21e33.

Bezrukova, E.V., Krivonogov, S.K., Takahara, H., Letunova, P.P., Shichi, K.,Abzaeva, A.A., Kulagina, N.V., Zabelina, Yu. S., 2008. Lake Kotokel as a stratotypefor the late glacial and Holocene in southeastern Siberia. Doklady AcademiiNauk, Geografiya 420 (2), 1e6.

Bezrukova, E.V., Tarasov, P.E., Solovieva, N., Krivonogov, S.K., Riedel, F., 2010. Lastglacialeinterglacial vegetation and environmental dynamics in southern Sibe-ria: chronology, forcing and feedbacks. Palaeogeography, Palaeoclimatology,Palaeoecology 296, 185e198.

Brooks, K., Scholz, C.A., King, J.W., Peck, J., Overpeck, J.T., Russell, J.M.,Amoako, P.Y.O., 2005. Late-Quaternary lowstands of Lake Bosumtwi, Ghana,evidence from high-resolution seismic-reflection and sediment-core data.Palaeogeography, Palaeoclimatology, Palaeoecology 216, 235e249.

Colman, S.M., 2006. Acoustic stratigraphy of Bear Lake, Utah-Idaho, LateQuaternary sedimentation patterns in a simple half-graben. SedimentaryGeology 185, 113e125.

Colman, S.M., Jones, G.A., Rubin, M., King, J.W., Peck, J.A., Orem, W.H., 1996. AMSradiocarbon analyses from Lake Baikal, Siberia. Challenges of dating sedimentsfrom a large, oligotrophic lake. Quaternary Science Reviews 15, 669e684.

Colman, S.M., Kelts, K.R., Dinter, D.A., 2002. Depositional history and neotectonics inGreat Salt Lake, Utah, from high-resolution seismic stratigraphy. SedimentaryGeology 148, 61e78.

Daut, G., Mäusbacher, R., Baade, J., Gleixner, G., Kroemer, E., Mügler, I., Wallner, J.,Wang, J., Zhu, L., 2010. Late Quaternary hydrological changes inferred from lake

level fluctuations of Nam Co (Tibetan Plateau, China). Quaternary International218 (1e2), 86e93.

Davidov, V.I. 1074. Aerogeologiya, 1974. Geological map of the USSR, Scale 1:200000, Pribaykalskaya series, map sheet N-49-XXV, Moskow.

Demske, D., Heumann, G., Granoszewski, W., Nita, M., Mamakowa, K., Tarasov, P.E.,Oberhänsli, H., 2005. Late glacial and Holocene vegetation and regional climatevariability evidenced in high-resolution pollen records from Lake Baikal. Globaland Planetary Change 46, 255e279.

Déverchère, J., Petit, C., Gileva, N., Radziminovitch, N., Melnikova, V., San’kov, V.,2001. Depth distribution of earthquakes in the Baikal rift system and itsimplications for the rheology of the lithosphere. Geophysical Journal Interna-tional 146, 714e730.

Dietze, E., Wünnemann, B., Diekmann, B., Aichner, B., Hartmann, K., Herzschuh, U.,IJmker, J., Jin, H., Kopsch, C., Lehmkuhl, F., Li, S., Mischke, S., Niessen, F., Opitz, S.,Stauch, G., 2010. Basin morphology and seismic stratigraphy of Lake DonggiCona, north-eastern Tibetan Plateau, China. Quaternary International 218 (1e2),131e142.

D’Agostino, K., Seltzer, G., Baker, P., Fritz, S., Dunbar, R., 2002. Late-Quaternarylowstands of Lake Titicaca, evidence from high-resolution seismic data. Palae-ogeography, Palaeoclimatology, Palaeoecology 179, 97e111.

Florensov, N.A., 1960. Mesozoic and Cenozoic Basins of Pribaikalye, vol. 258. Publ.House Akad. Nauk SSSR, Moscow-Leningrad (in Russian).

Galaziy, G.I. (Ed.), 1993. Baikal Atlas. Federal Agency for Geodesy and Cartography ofRussia, Moscow (in Russian).

Hofmann, M.H., Hendrix, M.S., Moore, J.N., Sperazza, M., 2006. Late Pleistocene andHolocene depositional history of sediments in Flathead Lake, Montana.Evidence from high-resolution seismic reflection interpretation. SedimentaryGeology 184, 111e131.

Karabanov, E.B., Prokopenko, A.A., Williams, D.F., Colman, S.M., 1998. Evidence fromLake Baikal for Siberian glaciation during oxygen-isotope substage 5d.Quaternary Research 50, 46e55.

Kashiwaya, K. (Ed.), 2003. Long Continental Records from Lake Baikal. Springer,Tokyo, pp. 370.

Krivonogov, S.K., Takahara, H., Kuzmin, Y.V., Orlova, L.A., Timothy Jull, A.J.,Nakamura, T., Miyoshi, N., Kawamuro, K., Bezrukova, E.V., 2004. Radiocarbonchronology of the Late PleistoceneeHolocene paleogeographic events in LakeBaikal region (Siberia). Radiocarbon 46, 745e754.

Lister, G.S., Kelts, K., Zao, C.K., Yu, J.-Q., Niessen, F., 1991. Lake Qinghai, China, closedbasin lake levels and the oxygen isotope record for ostracoda since the latestPleistocene. Palaeogeography, Palaeoclimatology, Palaeoecology 84, 141e162.

Niessen, F., Ebel, T., Kopsch, C., Fedorov, G.B., 1999. High-resolution seismic stra-tigraphy of lake sediments on the Taymyr Peninsula, Central Siberia. In:Kassens, H.A., Bauch, I., DmitrenkoEicken, H., Hubberten, H.-W., Melles, M.,Thiede, J., Timokhov, L. (Eds.), Land-ocean Systems in the Siberian Arctic,Dynamics and History. Springer, Berlin, pp. 437e456.

Oberhänsli, H., Novotna, K., Piskova, A., Chabrillat, S., Nourgaliev, D.K.,Kurbaniyazov, A.K., Grygar, T.M., 2011. Variability in precipitation, temperatureand river runoff in W Central Asia during the past w2000 yrs. Global andPlanetary Change 76 (1e2), 96e104.

Prokopenko, A.A., Williams, D.F., Karabanov, E.B., Khursevich, G.K., 2001. Conti-nental response to Heinrich events and Bond cycles in sedimentary records ofLake Baikal, Siberia. Global and Planetary Change 28, 217e226.

Prokopenko, A.A., Williams, D.F., 2004. Deglacial methane emission signals in thecarbon isotopic record of Lake Baikal. Earth and Planetary Science Letters 218,135e147.

Shchetnikov, A.A., 2007. Morphotectonics of the Baikal rift zone lacustrine basins.Russian Journal of Pacific Geology 1 (2), 120e129 (in Russian).

Shchetnikov, A.A., White, D., Filinov, I.A., Rutter, N., 2012. Late Quaternary geology ofthe Tunka rift basin (lake Baikal region), Russia. Journal of Asian Earth Sciences46, 195e208.

Shichi, K., Takahara, H., Krivonogov, S.K., Bezrukova, E.V., Kashiwaya, K.,Takehara, A., Nakamura, T., 2009. Late Pleistocene and Holocene vegetation andclimate records from Lake Kotokel, central Baikal region. Quaternary Interna-tional 205, 98e110.

Schnellmann, M., Anselmetti, F.S., Giardini, D., McKenzie, J.A., Ward, S.N., 2002.Prehistoric earthquake history revealed by lacustrine slump deposits. Geology30, 1131e1134.

Swann, G.E.A., Mackay, A.W., Leng, M.J., Demory, F., 2005. Climatic change in CentralAsia during MIS 3/2: a case study using biological responses from Lake Baikal.Global and Planetary Change 46, 235e253.

Tarasov, P.E., Harrison, S.P., Saarse, L., Pushenko,M.Ya., Andreev, A.A., Aleshinskaya, Z.V.,Davydova, N.N., Dorofeyuk, N.I., Efremov, Yu.V., Khomutova, V.I., Sevastyanov, D.V.,Tamosaitis, J., Uspenskaya, O.N., Yakushko, O.F., Tarasova, I.V., 1994. Lake StatusRecords from the Former Soviet Union and Mongolia: Data Base Documentation.In: NOAA Paleoclimatology Publications Series Report, vol. 2, pp. 274.

Tarasov, P., Granoszewski, W., Bezrukova, E., Brewer, S., Nita, M., Abzaeva, A.,Oberhänsli, H., 2005. Quantitative reconstruction of the last interglacial vege-tation and climate based on the pollen record from Lake Baikal, Russia. ClimateDynamics 25, 625e637.

Tarasov, P., Bezrukova, E., Karabanov, E., Nakagawa, T., Wagner, M., Kulagina, N.,Letunova, P., Abzaeva, A., Granoszewski,W., Riedel, F., 2007. Vegetation and climatedynamics during the Holocene and Eemian interglacials derived from Lake Baikalpollen records. Palaeogeography. Palaeoclimatology, Palaeoecology 252, 440e457.

Tarasov, P.E., Bezrukova, E.V., Krivonogov, S.K., 2009. Late Glacial and Holocenechanges in vegetation cover and climate in southern Siberia derived

Y. Zhang et al. / Quaternary International 290-291 (2013) 57e67 67

from a 15 ka long pollen record from Lake Kotokel. Climate of the Past 5,285e295.

Ten Brink, U.S., Taylor, M.H., 2002. Crustal structure of central Lake Baikal: insightsinto intracontinental rifting. Journal of Geophysical Research 107 (B7), 2132.

Ufimtsev, G.F., Shchetnikov, A.A., Agafonov, B.P., 1998. The lakes around Baikal.Proceedings of Russian Geography.Society 130 (4), 36e46.

Vanneste, M., De Batist, M., Golmshtok, A., Kremlev, A., Versteeg, W., 2001. Multi-frequency seismic study of gas hydrate-bearing sediments in Lake Baikal,Siberia. Marine Geology 172, 1e21.

Wagner, B., Reicherter, K., Daut, G., Wessels, M., Matzinger, A., Schwalb, A.,Spirkovski, Z., Sanxhaku, M., 2008. The potential of Lake Ohrid for long-termpalaeoenvironmental reconstructions. Palaeogeography, Palaeoclimatology,Palaeoecology 259, 341e356.

Wünnemann, B., Demske, D., Tarasov, P.E., Kotlia, B.S., Bloemendal, J., Diekmann, B.,Hartmann, K., Reinhardt, C., Riedel, F., Arya, N., 2010. Hydrological evolutionduring the last 15 ka in the Tso Kar lake basin (Ladakh, India), derived fromgeomorphological, sedimentological and palynological records. QuaternaryScience Reviews 29, 1138e1155.