Marine Geophysical Researches 22: 445–464, 2001.© 2002 Kluwer Academic Publishers. Printed in the Netherlands.
445
Seismic evidence of small-scale lacustrine drifts in Lake Baikal (Russia)
S. Ceramicola1,2,∗, M. Rebesco2, M. De Batist1 and O. Khlystov3
1Renard Centre of Marine Geology - RCMG, Gent, Belgium;2Istituto Nazionale di Oceanografia e di Geofisica Sperimentale – OGS, Trieste, Italy;3Limnological Institute, Irkutsk, Russia;∗Author for correspondence (Tel: +39-040-2140341;Fax: +39-0404-327307; E-mail: [email protected])
Received 12 July 2001; accepted 20 December 2001
Key words: contourite, drift, fault, seismics, Lake Baikal, Russia
Abstract
High resolution, single-channel seismic sparker profiles across the Akademichesky Ridge, an intra-basin structural high in Lake Baikal (Russia),reveal the presence of small sediment mounds and intervening moats in the upper part of the sedimentary cover. Such features interrupt thegenerally uniform and even acoustic facies and are not consistent with the hemipelagic sedimentation, which is expected on such an isolatedhigh and which would produce a uniform sediment drape over bottom irregularities. The influence of turbidity currents is excluded since theridge is an isolated high elevated more than 600-1000 m above adjacent basins. The mounded seismic facies, including migrating sedimentwaves and non-depositional/erosional incisions, strongly suggest that sediment accumulation was controlled by bottom-current activity. Weinterpret the mounds as small-scale (< few tens of km2 in area) lacustrine drifts. Four basic types of geometry are identified: 1) slope-plasteredpatch sheets; 2) patch drifts; 3) confined drifts; 4) fault-controlled drifts. The general asymmetry in the sedimentary cover of the ridge, showingthicker deposits on the NW flank, and the common location of patch drifts on the northeast side of small basement knolls indicate that depositiontook preferentially place at the lee sides of obstacles in a current flowing northward or sub-parallel to the main contours. Deep-water circulationin the ridge area is not known in detail, but there are indications that relatively cold saline water masses are presently flowing out of the CentralBasin and plunging into the deep parts of the North Basin across the ridge, a process that appears to be driven mainly by small differencesin salinity. We infer that the process responsible for the observed bottom-current-controlled sedimentary features has to be sought in theselarge-scale water-mass movements and their past equivalents. The age of the onset of the bottom-current-controlled sedimentation, based on anaverage sedimentation rate of 4.0 cm/ky, is roughly estimated to be as least as old as 3.5 Ma, which is generally regarded as the age of the onsetof the last major tectonic pulse of rift basin development in the Baikal region.
Introduction
The term ‘drift’ or ‘contourite drift’ is commonly usedto define deep-sea accumulations of sediment that aredeposited or significantly affected by bottom currents,and to distinguish this type of deposits from those con-trolled by other deep-sea sedimentation processes (i.e.,turbidites, pelagites, debrites, etc.). It may encompassmany types of sediment accumulations of variablesizes and shapes (mounds, sediment waves, etc.) andof erosional features (channels, furrows. . . ), and itmay involve many different types of bottom currents:contour currents s.s., thermohaline or wind-driven cur-rents, and other types of flow in deep water (i.e.,>300 m water depth) such as internal tides and down-or up-welling flows along slopes or within canyons.For a better understanding of all processes involved incurrent-controlled sedimentation, it is therefore essen-
tial to accurately describe and study all the varietiesof contourite drifts, including those occurring in lesstypical environments, such as on continental shelvesor in lakes.
Johnson et al. (1980) and Halfman and John-son (1984) already reported the presence of current-controlled sedimentary features in Lake Superior(USA), consisting of a.o. furrows, scours and a thincurrent-controlled sedimentary cover (see also Floodand Johnson, 1984). Other examples of current-controlled deposits, such as sediment waves and ero-sional moats, have been observed in some of the largerEast-African Rift lakes (Johnson, 1996, and referencestherein). In this paper, we report the existence of sedi-ment drifts in another large lake (Lake Baikal, Russia),with dimensions that are significantly larger than thosedescribed in Lake Superior and the African lakes. To
446
Figure 1. Location map. Track lines of the seismic survey are superimposed upon bathymetry (USSR Ministry of Defence, 1992). Contoursevery 100 m. Thick lines indicate the profiles shown in this article. Approximate location of the drifts is outlined by the grey shaded pattern.The grey arrows indicate the inferred direction of bottom-current flows: (1) sub-parallel to the contours in the sill between NWHB and SEHB;and (2) across the sill to the southwest of the NWHB.
our knowledge, this is the first report of lacustrinedrifts with dimensions comparable to those of theirdeep-sea counterparts. Another particular characteris-tic of the Lake Baikal drifts is that their occurrenceappears to be restricted to the crest of an intra-basinhigh, and that they do not occur on the deep basinfloors or along the basin margins.
Study area
Lake Baikal, the world’s largest and deepest fresh-water lake, occupies the central third of the Baikal RiftZone. It is roughly 600 km long, 30 to 87 km wide,and contains a total water volume of about 23,000 km3
(Galazy, 1993). The lake surface lies at an elevation of
465 m. The lake is subdivided into three major bathy-metric basins: the North, Central, and South Basins,about 900, 1,600, and 1,400 m deep, respectively(Figure 1). The basins, each about 200 km long, areseparated by structural highs that rise to a depth ofabout 300 m: the Akademichesky Ridge between theNorth and Central Basins, and the Selenga Delta areabetween the Central and South Basins.
The investigated area is the Akademichesky Ridge(i.e., the underwater prolongation of Olk’hon Island).It is 60 km long and 20 km wide, and is composedof two SW–NE striking structural blocks (Ceramicolaet al., 1998; Ceramicola, 2001; Mats et al., 2000): theNWHB (NorthWestern Horst Block) and the SEHB(SouthEastern Horst Block) separated by an elongated
447
Figure 2. Dip seismic profile 53 across the NWHB (above), and schematic line drawing with reflector terminations indicated by short ar-rows (below). For location see Figure 1. Crossing with profile 24 (Figure 3) is shown. The filling facies of Unit A shows discontinuousreflectors onlapping the acoustic basement and erosional truncation at the upper boundary. The patch drift within Unit B is characterised byhigher-amplitude, laterally continuous reflectors showing bi-directional downlap.
sill (Figure 1). The depth of the NWHB ranges from300 m in the SW to 500 m in the NE; that of the SEHBfrom 300 m in the SW to 200 m in the NE.
The ridge is an isolated structural high elevatedmore than 600–1,000 m above the adjacent basinfloors, which separate it from the lake borders andentirely shield it from turbidites and underflow de-posits. All sediments are delivered to the ridge through
hemipelagic settling. Sediments consist of alternationsbetween diatomaceous oozes and diatom-barren muds(Colman et al., 1995), deposited with an average sedi-mentation rate of 4 cm/ky (BDP Members, 1998), andshow no traces of turbidites (graded beds, consistentrepetition of facies. . . ) or any other mass-movementdeposits. For this reason, the area has been a primetarget for sediment coring and deep drilling for several
448
Figure 3. Strike seismic profile 24 across the NWHB (above), and schematic line drawing with reflector terminations indicated by short arrows(below). For location see Figure 1. Crossing with profile 53 (Figure 2) is shown. The patchy, mounded sedimentary cover within Unit B isshown to be asymmetrically distributed over the basement highs (a), to have no apparent connection with the basement morphology (b), or tomaintain the underlying topography (c).
years, because this type of sedimentary environmentis believed to hold a valuable record of paleoclimatechanges (BDP Members, 1998; Grachev et al., 1997,1998; Kuzmin et al., 1997, 2000).
In contrast, the sediments accumulating in the deepbasins of Lake Baikal are predominantly turbiditesand underflow deposits, which are controlled by thetectonic setting, the basin morphology and climate
changes (Nelson et al., 1995; Back et al., 1998, 1999).Basin-floor sedimentation rates vary from 16 cm/ky(Northern Baikal Basin) to 74 cm/ky (Southern BaikalBasin), with maximum values of 120 cm/ky near theSelenga Delta (Edgington et al., 1991).
The Baikal catchment area is approximately570,000 km2 (Galazy, 1993). The Selenga, Upper An-gara and Barguzin Rivers are the three main inflows
449
and deliver 72% of the total water supply. The AngaraRiver, at the southwestern end of the lake, is the onlyoutflow. The Barguzin River, the only tributary locatednear Akademichesky Ridge, has a catchment area ofabout 22,000 km2 and flows into the northern part ofthe Central Basin.
Lake Baikal behaves as a typical north-temperatedimictic lake with normal overturns in spring and au-tumn within the upper 200–300 m of its depth. Deep-water renewal is relatively fast and does not exceed19 years (Hohmann et al., 1997). In the deeper partof the water column, localised deep mixing of densewater may occur due to a strong wind effect (Brad-bury et al., 1994). Moreover, Hohmann et al. (1997)suggested the presence of important inter-basin water-mass exchanges caused by salinity differences. Theyidentified two main processes: (1) spring inflow ofcold and relatively saline water from the Selenga Riverinto the Central Basin; and (2) deep-water formationtriggered by the small salinity difference between theCentral and North Basins. Thus, the AkademicheskyRidge is considered a threshold across which the coldsaline water flows from the Central Basin into the deeppart of the North Basin (Hohmann et al., 1997).
Methods
The geophysical dataset used for this study consistsof about 1,000 km of single-channel high-resolutionreflection seismic profiles that have been acquired bythe Renard Centre of Marine Geology (Belgium), incollaboration with the Limnological Institute (Rus-sia). The profiles are generally oriented perpendicularand parallel to the ridge axis (Figure 1). The distancebetween profiles is on average 5 km.
The seismics were shot using RCMG’s ‘Cen-tipede’ multi-electrode sparker as a seismic source,and a single-channel streamer as receiver. The seis-mic source was operated at 500 J. During the sur-vey, an analog bandpass-filter of 200–2,000 Hz wasused to eliminate electric and ship’s noise. The signalwas recorded with the ELICS Delph2 digital data-acquisition system. Shot spacing was 25 m and thesampling interval was 0.25 ms.
Data show little ghost or bubble effects and atypical penetration of up to 400 ms TWT. The theo-retical vertical resolution (Rayleigh criterion) is about0.75 m, although in practice the length of the sparkerpulse may shift this to slightly higher values of around1.0 to 1.2 m, in particular for the lake-floor return and
to a lesser extent for deeper reflections. No deconvolu-tion was applied to the data. The good signal-to-noiseratio of the data required only minimal post-cruiseprocessing, which included automatic gain control(AGC) and water column muting, both essentially fordisplay purposes. Some selected profiles were alsomigrated.
Description of the results
The acoustic basement on our seismic data is markedon top by a few very high amplitude, laterally discon-tinuous reflections. Above it, two sedimentary unitsare identified in the seismic succession: Unit A andUnit B (the latter directly overlays the basement wherethe former is absent).
Unit A (Figure 2) is observed only on the NWHB(Ceramicola et al., 1998; Ceramicola, 2001; Matset al., 2000). The average thickness of Unit A is about30 ms∗ TWT and it never exceeds 100 ms TWT. Inter-nal reflectors generally show low amplitude and lowlateral continuity. Internal geometry is complex andincludes internal unconformities. It shows both a fill-ing facies within depressions and a prograding facieson the flanks of basement blocks. In half-graben-like structural depressions, the former may have awedge shape thickening towards the bounding faultand shows sub-parallel, concave and undulated re-flectors. The latter has a clinoform shape formed bysub-parallel, oblique to sigmoidal reflectors. Both fa-cies show distinct onlap at the lower boundary andlocal erosional truncations at the upper boundary.
Unit B extends across both the NWHB and SEHB,and in the Maloye More basin (Ceramicola et al.,1998; Ceramicola, 2001; Mats et al., 2000). Thicknessof Unit B ranges from a few milliseconds above highs,to about 1 sec above depressions. Most of the unitshows a draping facies characterised by the presenceof continuous, sub-parallel, concordant reflectors. Theamplitudes of the seismic reflections are variable butgenerally much higher than in Unit A (Figure 2).
However, the sediments on the crest of the ridgehave a unique undulating and patchy morphology,punctuated by several sediment mounds (Figures 2and 3). Moreover, sediment thickness of Unit B isasymmetric on the two ridge’s flanks, with a largerthickness on the northern flank than on the south-ern (Figure 4). Conversely, no comparable thickness
∗Thickness is given in this paper as milliseconds (ms) two-way-travel-time (TWT).
450
Figure 4. Dip seismic profile 48 across the top of the NWHB (above), and schematic line drawing with reflector terminations indicated byshort arrows (below). For location see Figure 1. The prograding facies of Unit A shows oblique to sigmoidal reflectors onlapping the acousticbasement. Erosional truncation at the upper boundary is more evident on the SE side of the ridge, whereas downlap terminations withinsigmoidal units are more common on the NW side. A small asymmetric drift controlled by active faulting is present within Unit B at a slopechange on the SE side of the NWHB. This patch drift is characterised by higher-amplitude, laterally continuous reflectors. Downlap of reflectorsis widespread on the SE side and in the lower part of the NW side, where a small moat is present between the drift and the ridge.
asymmetry is observed within Unit A. The patchymorphology of the ridge is produced where the gen-eral uniform configuration of Unit B is interrupted bythe presence of sedimentary mounds overlying lap-out (both onlap and downlap) terminations (Figure 5),sediment waves (Figures 5 and 6), and linear incisions
(Figure 7) which are all observed intermittently alongthe entire length of the ridge.
Sedimentary mounds are observed both on theslopes and on the crest of the NWHB, and on thecrest of the SEHB, whereas they are absent on theSE side of the SEHB. They are generally a few kmwide and about 100 ms TWT high. Since the size
451
Fig
ure
5.D
ipse
ism
icpr
ofile
42ac
ross
the
NW
HB
(lef
t),
and
sche
mat
iclin
edr
awin
gw
ithre
flect
orte
rmin
atio
nsin
dica
ted
bysh
orta
rrow
s(r
ight
).Fo
rlo
catio
nse
eFi
gure
1.T
heas
ymm
etri
cdr
ifts
how
sab
rupt
term
inat
ion
ofre
flect
ors
onth
eE
SEsi
deco
ntro
lled
byan
activ
efa
ult.
Min
ordo
wnl
apte
rmin
atio
nsm
aybe
pres
ent
asw
ell.
Dow
nlap
term
inat
ions
are
evid
ent
abov
eU
nitA
inth
ere
lativ
ely
flatp
arto
fth
eba
sem
ent.
Onl
apte
rmin
atio
nsar
epr
esen
tels
ewhe
rein
rela
tions
hip
toth
ere
mar
kabl
yst
eep
flank
sof
the
ridg
e.In
the
vici
nity
ofth
eba
sem
ent
high
moa
tsoc
cur
onbo
thsi
des,
asw
ella
sm
igra
ting
sedi
men
twav
eson
the
top.
452
of most of these features is below the resolution ofthe seismic grid, we have not been able to map themvery precisely. For this reason, only in a few cases wewere able to identify with some confidence their exactshape and the direction of elongation with respect tothe ridge’s slope.
In the absence of consistent mapping, the exactmound distribution is still largely unknown. How-ever, it has been observed that many mounds areassociated with prominent topographic relief producedby faults affecting the basement. In some cases, thebasement faults also affect the overlying sedimen-tary cover, hence interfering with the growth of themounds. Some of the mounds along the boundary be-tween the NWHB and SEHB are evidently influencedby active faulting. A small mound located at the baseof the SE slope of the NWHB is situated betweentwo active branches of the Akademichesky Fault (Fig-ure 4). A highly asymmetric mound, the geometry ofwhich may be explained only partly by the presence ofa basement fault affecting the sedimentary section upto the seafloor, is observed on the crest of the NW flankof the NWHB (Figure 5). While the presence of thefault may explain the steepness of the SE side of themound, it may not account for the slight divergence ofreflectors and associated gradual sediment thickeningon the NW side toward the mound crest. No moundsare observed in association with faults affecting onlythe sedimentary cover. So it seems that only thosefaults that played a role in controlling basement topog-raphy may also have played a role in the formation ofthe mounds.
It is evident that Unit B sediments do not forma uniform drape above the underlying morphology.Though sedimentary mounds are sometimes observedin association with faulting, and/or their locations oc-casionally bear obvious relationships with basementhighs, sediment thickness is often unpredictably re-lated to the underlying morphology: some of themounds maintain the underlying topography, someshow asymmetric sedimentary cover, and some appearto have no connection at all with the morphology ofthe underlying basement topography (Figure 3).
At a smaller scale, sediment waves (several hun-dreds of m wavelength, tens of ms high) have beenidentified both on top and on the slopes of prominentbasement highs. They may occur both as single wavesand as a set of migrating waves. Sediment waves onthe slopes of prominent topographic relief are gener-ally very localised, and they disappear by assuminglower amplitudes or longer wavelength when the topo-
graphic relief is overflowed by sediment deposition(Figure 6). When occurring on top of structural highs,the horizontal extension of the sediment waves is ofthe same order as the underlying relief, originating onone flank of the high and disappearing on the oppo-site flank (Figure 5). Though the actual orientationof the crests is not constrained by crossing profiles,the waves show an evident up-slope component ofmigration along the NW side of the NWHB.
Sediments on top of the NWHB frequently shownumerous small linear incisions ranging from a few to20 ms in depth and from 10 to 100 m in width. Theirlength and direction are difficult to establish with thepresent data set. However, from tentative correlationbetween adjacent profiles, their direction seems to beslightly oblique to the ridge axis. Though their actualshape is masked by hyperbolic diffractions on non-migrated data, they appear to be symmetric incisionsthinning at the base and diverging at both sides. How-ever, on migrated profiles they do not appear to besimple erosive features, since stratigraphic thinning ofstrata towards the incisions is prevailing on reflectiontruncations (Figure 7). These patterns seem to affectmostly the upper packages of the ridge’s sedimentarysequence. Their distribution is restricted to the NWHBand they occur more frequently on the crest of theridge. These patterns contribute to the characteristicundulating geometry to the patchily distributed Unit Bon the crest of the ridge.
Discussion
Seismic stratigraphy and sedimentary environments
Although the grid spacing of the seismic survey wasnot dense enough to fully image the small-scale de-positional geometry and morphologies, informationderived from seismic stratigraphic analysis in combi-nation with data from sediment cores and deep drillingin the area allows a reconstruction of the sedimen-tary paleo-environments of the Akademichesky Ridgearea.
The acoustic basement is interpreted to representthe base of the sedimentary succession, and corre-sponds to crystalline basement consisting of gneiss,granite, crystalline schist, and marble (Zonenshainet al., 1992). The very high amplitude, laterally dis-continuous reflections at the top of the acoustic base-ment correspond to very hard and highly compactedsediments formed as a sub-aerial weathering crust(Goldyrev, 1982).
453
The oldest deposits belong to Unit A and form adiscontinuous cover, including prograding clinoforms,and filling of half-graben-like depressions (Cerami-cola et al., 1998; Ceramicola, 2001; Mats et al., 2000)(Figures 2 and 3). This unit locally shows downwardincreasing high amplitudes interpreted as gravel lagdeposits corresponding to the fining-upward sequencewith coarse to massive very poorly sorted gravel re-covered in USGS core 341 (Colman et al., 1993; Nel-son et al., 1995). The sigmoid-prograding reflectorsare interpreted to represent deposition during submer-sion of the crest of the Ridge, when it passed fromsub-aerial to submerged conditions. The sedimentsrepresented by such reflectors are probably MiddleMiocene (17–10 Ma) in age (Mats et al., 2000; Ce-ramicola, 2001). The diverging wedge-shaped fillinggeometry testifies the syn-tectonic deposition. Unit Ais defined as a high-energy deposit characterised bytransitional environments (from littoral to shallow-lake) involving re-mobilisation and re-deposition ofthe initial sedimentary cover (Ceramicola, 2001).
The boundary between Unit A and Unit B cor-responds to a remarkable angular unconformity fre-quently highlighted by numerous lap-out terminationsof basal Unit B reflectors (Ceramicola et al., 1998;Ceramicola, 2001; Mats et al., 2000) (Figures 2,5 and 6). Apart from its base, Unit B is mostlycharacterised by continuous, sub-parallel, concordantreflectors with alternating high and low amplitudes.Cores from Unit B contain fine-grained sedimentsconsisting of cyclic alternations of diatom ooze anddiatom-barren mud (Colman et al., 1993; Zonenshainand Kazmin, 1995), suggesting pelagic/hemipelagicdeposition under a low-energy sedimentary regime.Analysis of the rhythmic signature of these deposits(Colman et al., 1995) yields an average sedimenta-tion rates of 4.0 cm/ky, which is in agreement withthe results of the BDP-96 borehole (Kuzmin et al.,1997, 2000; BDP Members, 1998). On the basis ofthe above information, Unit B is interpreted as havingbeen deposited once the Ridge was completely sub-merged (from Late Miocene times onwards accordingto Mats et al., 2000 and Ceramicola, 2001) and in adeep lacustrine environment characterised by a low-energy and relatively undisturbed sedimentary regime(Ceramicola, 2001).
Current-controlled sediment mounds
The continuous, sub-parallel, concordant facies ofUnit B, as well as the sediment cores retrieved from
this unit, clearly indicate that it is composed primarilyof hemipelagic sediments: a sediment drape depositedby slow settling through the water column in the ab-sence of any substantial bottom- or turbidity-currentactivity. However, the occurrence of localised moatsand mounds interrupting the generally uniform andeven acoustic facies is not consistent with a draping-like deposition which, by definition, would show amore or less uniform, acoustically transparent sedi-ment sheet over bottom irregularities (Moore, 1969).Remarkable lap-out terminations are observed bothin association with faults and, more unpredictably,in coincidence with moats and mounds. In addition,the sedimentary cover shows a pronounced thicknessasymmetry across the ridge. The possibility of suchdeposits having been produced by turbidity processesis excluded as the ridge is an isolated high separatedfrom the turbidite-generating lake borders by the deepadjacent basins. This inference is moreover supportedby core lithologies, which do not show sedimentaryfeatures typical of turbidites (graded beds, consistentrepetition of facies, etc. . . ). Furthermore the possi-ble interpretation of some of the observed thicknesschanges and lap-out terminations as due to gravita-tional instabilities and mass-wasting processes, relatedto the steep slopes of the Ridge and the high seis-micity of the area, was taken into account. Carefulchecking of crossing profiles through these featuresand elsewhere on the Ridge show no indication forthe presence of slide scars, slide or slump deposits ordebris flows. Based on these observations, this possi-bility was ruled out. Therefore we infer that depositionof Unit B was controlled by the effect of bottomcurrents.
It is fundamental to assess the advantages and lim-itations of the seismic system used, and to recognisewhat type and scale of information can be derivedfrom it (Faugères et al., 1999). Different seismicsources allow different scales of resolution and there-fore allow different scales of drifts to be recognisedand compared. Examples of lacustrine contouritesfrom Lake Superior (Johnson et al., 1980) were im-aged with a 3.5 kHz echosounder source (penetrationfrom a few to tens of meters), which highlighted thesurface morphology and the detailed geometry of thesub-recent environment. These records are not easilycomparable with the scale of resolution of our single-channel sparker data (frequency: 200–2,000 Hz). Onthe contrary, our data can be compared with therecords of classic larger-scale drift deposits obtainedby comparable seismic systems.
454
Figure 6. Strike seismic profile 24 within the sill between NWHB and SEHB (above), and schematic line drawing with reflector terminationsindicated by short arrows (below). For location see Figure 1. The moat-and-mound facies within Unit B is limited to the lower part of theslope and substituted by a relatively even sediment drape once the relief is overflowed. The moat-and-mound facies is interpreted as being dueto consistent northeastward migration of a complex contourite system, including small subsidiary drift, within the moat of a larger plasteredsheeted drift.
455
The external shape and internal geometry of theobserved moats and mounds show many similaritieswith larger-scale deep-sea sediment drifts (Mitchum,1985; McCave and Tucholke, 1986; Faugères andStow, 1993) since both have: (1) mounded shape withaxial thickening; (2) thickness exceeding the adjacentsedimentary cover; (3) more or less pronounced asym-metry; (4) high-frequency layering with high lateralcontinuity along both strike and dip; (5) marked lap-out (both downlap and onlap) above the underlyingreflectors; (6) presence of boundary channels (wherethe current flow is supposedly higher); (7) underlyingerosive surface providing high-amplitude reflectors.
On the basis of the above observations and inagreement with the diagnostic criteria defined by Stowand Lovell (1979), Johnson et al. (1980), Stow andPiper (1984), Stow et al. (1998), and Faugères et al.(1999), we interpret the described moat-and-moundfeatures of Unit B as small-scale lacustrine drifts.Drifts in the deep ocean generally have dimensionsranging in area from <100 km2 for small patch driftsto >100,000 km2 for abyssal sheets (Stow et al.,1998). Dimensions of the Akademichesky Ridge driftsare often difficult to assess but they appear to be com-parable in size to those of deep-sea patch drifts, whichare typically a few tens of km2 in area, 10–150 m thickand either irregular in shape or elongated in the direc-tion of flow (Faugères et al., 1999). On the other hand,the range of dimensions shown by the AkademicheskyRidge drifts is relatively small when compared to thatshown by deep-sea drifts, ranging over 4 orders ofmagnitude (Stow et al., 1998).
Four distinct geometries are known for deep-sea drifts: sheets (including patch sheets), elongatemounds, channel-related drifts (including patch driftsin the lee of knolls), and confined drifts (Faug’̀reset al., 1999). Akademichesky Rift drifts occur in fourpartially overlapping basic types: (1) slope-plasteredpatch sheets; (2) patch drifts; (3) confined drifts;(4) fault-controlled (or relief-controlled) drifts.
Slope-plastered patch sheets.According to the classification proposed by Faugèreset al. (1999), most of Akademichesky Ridge drifts be-long the slope-plastered patch sheet type. Two driftsplastered against both sides of a structural high ontowhich they are prograding are shown in Figure 8.These patch sheets contain continuous semi-paralleloblique reflections and their gross geometry is simi-lar to many deep-sea plastered drifts. Similar to thegeneral cases illustrated by Faugères et al. (1999),
these drifts are mostly aggradational. However, a mi-nor up-slope progradation probably corresponding tothe up-slope oblique component of a slight down-current progradation is visible. Unfortunately this in-terpretation could not be checked since the expecteddown-current migration is not verifiable on any exist-ing profile crossing one of these two plastered sheets.The only nearby transverse profile crosses the struc-tural high in between the two sheets and shows onlya very thin sedimentary cover with slightly enhancedthickness on the NE side (Figure 3). The observedgeometry therefore suggests deposition controlled bya current oriented sub-parallel to the ridge contours,either flowing in NE or SW direction.
Patch drifts, plastered against the slope (and on thetop) of small basement knolls.Many of the smaller drifts on the crest of the Aka-demichesky Ridge (Figures 2 and 3) show a lenticulargeometry, sometimes limited at both sides by bound-ary channels along which the bottom currents areconstrained. Similar to many deep-sea patch drifts(e.g., Hikurangi Plateau drifts, Carter and McCave,1994), the growth of such drifts is generally relatedto enhanced bottom-current deposition in the lee sidesof pronounced relief, from which the drifts are usu-ally separated by a moat. They show a more or lessasymmetric mounded shape. Unfortunately, we haveno direct data on bottom-current activity. Moreover,though the spacing of the seismic survey is relativelydense, the detailed relationships between such smallpatch drifts and the isolated basement knolls are stillnot adequately constrained. However, all of the patchdrifts appear to be on the north side of the knolls(e.g., the drift A in Figure 3 which lies on the NEside of a small knoll and which is also imaged onthe profile shown in Figure 2). The distribution of theAkademichesky Ridge patch drifts hence suggests thatthe north side of a basement knolls lies in the lee of thelocal bottom current. This information therefore pointsto a northward flow direction.
Confined drifts.The third type of drift in the Akademichesky Ridgearea is that of drifts confined between structural scarps.Figure 9 shows an example of a sediment drift con-fined within a depression on the NW side of the ridge.This drift is plastered on a highly uneven slope, whichevidently controls its geometry. It shows a convexaggradation pattern and lenticular internal geometries.The action of bottom currents may be envisaged in
456
Figure 7. Migrated version of the dip seismic profile 47 across the SW end of the NWHB. For location see Figure 1. The incisions within thecontourite drifts appear mainly to be the product of non deposition. Nevertheless some minor erosion seems to be present as well. Incisions andintervening drifts on both sides of the high show a moderate though consistent upslope migration as far as the base of the Unit B. Most of theincisions are related to changes in slope of the basement morphology.
the erosional/non-depositional character of the moatsadjacent to the slope. Similar to deep-sea patch driftsconfined in troughs, limited lateral migration is ob-served (Faugères et al., 1999).
Fault-controlled drifts.The growth of many of the Akademichesky Ridgedrifts appears to be controlled by faulting. The driftshown in Figure 4 shares some similarities with deep-sea drifts (asymmetric mounded shape with diver-gence of reflectors, high internal stratification, separa-tion from the slope, etc.. . . ). However, the steeper side(where downlapping reflectors are observed) is facingbasinward, and not facing the slope. This is the oppo-site of what is generally found in drifts. This differenceprobably comes from the evident tectonic control thatis present in the Akademichesky Ridge drift. In anycase, it has to be noted that tectonic processes alonemay not account for the observed geometries. Theasymmetric drift shown in Figure 5 is also confinedby a major fault that displaces the basement at theNWHB–SEHB boundary. The presence of this faultaffecting the seafloor has an obvious control on theSE side of the drift and may explain its steepness.However it may not account for the gradual sediment
thickening occurring on the opposite side (NW) ofthe drift. Conversely, the large-scale geometry of thisdrift, with the crest along the NE prolongation of thecrest of the NWHB and gradual thinning towards itssides is compatible with a possible control exerted by anortheastward flowing current. Such current would befavoured to flow sub-parallel to the contours along theNW side of the NWHB or in the sill between NWHBand SEHB. Sediment accumulation would hence befacilitated in the lee side of the NE prolongation ofthe NWHB crest. Moreover, the presence of up-slopemigrating sediment waves on the NW side of the driftis also in support of bottom-current activity.
Additional evidence of bottom-current activity isevident from a number of other features on the NWside of the ridge, such as buried confined drift and(subsidiary) patch drifts separated by moats underly-ing a relatively even sediment drape once the adjacentbasement relief is overflowed (e.g., Figures 9 and 10).This evidence and the asymmetric distribution of thesediments on the opposite flanks of the ridge (thickeron the NW side) leads us to interpret the entire NWside of the ridge as a large drift with complex morphol-ogy. However, the lack of consistent mapping does
457
Fig
ure
8.D
ipse
ism
icpr
ofile
31ac
ross
the
NW
HB
(lef
t),
and
sche
mat
iclin
edr
awin
gw
ithre
flect
orte
rmin
atio
nsin
dica
ted
bysh
orta
rrow
s(r
ight
).Fo
rlo
catio
nse
eFi
gure
1.Pa
tch-
shee
t-ty
peco
ntou
rite
drif
tsar
epl
aste
red
agai
nst
both
side
sof
the
base
men
thig
h,at
diff
eren
tde
pths
.The
ysh
owsl
ight
upsl
ope
mig
ratio
nw
ithdo
wns
lope
term
inat
ions
onth
elo
wer
end
and
onla
pon
the
uppe
r,st
eepe
ren
d.C
onve
rsel
y,th
esm
allp
atch
drif
tson
the
gent
lepl
atea
uN
Wof
the
top
are
esse
ntia
llyag
grad
ing
and
show
bi-d
irec
tiona
ldo
wnl
ap.
458
not allow us to make a sound interpretation of thelarge-scale geometries.
As a consequence of the complicated ridge mor-phology, including the sill between the two blocksand the presence of numerous knolls, AkademicheskyRidge drifts often have complex geometries. The driftshown in Figure 6b, for example, consists of buried,apparently irregularly stacked packages of moat andmound structure on the eastern slope of the ridge. It ispossible that a certain instability in the activity of thebottom currents generated adjacent packages growingalong-axis with different orientations. Alternatively,the complicated geometry might be explained if weconsider these mounds as patch drifts confined by theunderlying basement morphology and showing typicalirregular progradational geometry. However, an al-most consistent northeastward migration of the entiremoat-and-mound system can be observed (Figure 6b).This is similar to the configuration described by Howeet al. (1994), Stoker (1998) and Stoker et al. (1998) forthe northern Rockall Trough, including a subsidiarydrift within the moat of a giant elongated drift. We in-terpret the small mound (about 0.5 km wide) as a smallsubsidiary drift within the moat separating the ridgeslope from a larger (some km wide) plastered sheeteddrift located further east. The main difference betweenthe Rockall Trough example and the AkademicheskyRidge one is the direction of migration of the drifts:landward (up-slope) in the former and apparently bas-inward (away from the slope) in the latter case. How-ever, this difference may be explained by the differenttype of primary drift: while detached elongated drifts(like in the Rockall Trough case) are known to gener-ally prograde up-slope, sheet drifts plastered on theslope (like in the Akademichesky Ridge case) gen-erally show down-current progradation, with either abasinward or oblique landward component. The up-slope continuation of the profile shown in Figure 6shows that the moat-and-mound geometry disappearsand is replaced by a more even sediment drape whenthe adjacent relief is filled. This evidence suggests thatthe growth of the sheeted and subsidiary drifts is pro-duced by the interaction between the bottom currentsand the lake bottom irregularities. As soon as the re-lief has been levelled, the overlying sediment coverassumes a draping configuration.
A similar juxtaposition of facies characterised bya more complex geometry (including mounds andmoats) with an overlying more uniform sedimentdrape is also shown by Figures 9 and 10. Also in thiscase, the surface separating the two facies is at the
same level of the top of the adjacent high (a in Fig-ure 9). A change in the local bottom current regime(due to the filling of the intervening basin and thelevelling of the relief) is inferred to be the cause ofthe described geometry. The small patch (subsidiary)drift to the south of the confined drift in Figure 9 isapparently deposited in the lee of a small knoll by aNE flowing current. The confined drift itself, whenobserved on a dip profile (Figure 10), shows a limitedup-slope, southeastward migration as a result of thedeposition on the lee of the ridge slope. The informa-tion on bottom-current direction that may be gainedfrom this drift is that it should have flown towardsNNE, almost parallel with the NE–SW direction of thecontours.
High-amplitude reflectors
High-amplitude reflectors coinciding with basin-wideerosive surfaces are commonly observed underlyingdeep-sea drift accumulations and are hence considereda diagnostic criterion for the identification of current-controlled sedimentation (Faugères et al., 1999). Al-though the reflector amplitudes on our seismic profileshave to a certain extent been modified by the AGCthat was applied for display and printing purposes,the presence of such high-amplitude reflectors withinUnit B and at its base may provide additional evidenceof bottom-current activity and its capability to preventdeposition.
High-amplitude reflectors associated with wide-spread baselap terminations are not consistently ob-served at the base of the Akademichesky Ridge drifts.However, within the drift succession, packages oflow-amplitude reflectors are alternating with relativelyhigh-amplitude packages, which are often associatedwith local downlap terminations in the moats or onthe steep side of the drifts (e.g., Figure 6b). Suchhigh-amplitude reflectors are thought to mark areasof reduced sedimentation rates in the vicinity of themain axis of the bottom currents. Moreover, levelsenriched with Fe-Mn crusts and concretions, probablyrepresenting periods of non-deposition or of stronglyreduced sedimentation rates, have been described incores taken from the crest of the NWHB where the to-tal thickness of Unit B is often less than 10 m (Graninaet al., 1994; Deike et al., 1997). High-amplitude re-flectors are also generally observed in association withangular unconformities, both at the boundary betweenUnit A and Unit B (e.g., Figures 5 and 6) and withinUnit B (Figure 9). The high-amplitude reflectors at
459
the boundary between Unit A and Unit B are difficultto interpret unambiguously, but those associated withcurrent-related surfaces within Unit B (Figure 9) canbe attributed to the action of strong bottom currents, inanalogy to what has been described in deep-sea drifts(Stow and Piper, 1984).
Linear incisions on the crest of the Ridge
Linear lake-floor incisions highlighted by hyperbolicdiffractions were observed on the crest of Akademich-esky Ridge, particularly in the NWHB where the waterdepth is shallower (Figures 4, 6 and 8). When analysedin detail on migrated seismic data (Figure 7b), it isclear that these incisions were not created by ero-sion (no erosional truncation of strata in the incisions)but rather by consistent local winnowing or non-deposition (stratigraphic thinning of strata towardsthe incisions). Such incisions border the adjacent up-slope-migrating drifts, similar to moats observed else-where. However, they are slightly smaller in scale thanthe observed moats and do not seem to surround topo-graphic highs such as e.g. basement knolls. In thisregard, they bear some similarity with the long, linear,parallel sedimentary furrows that are often observed indeep-sea environments. Such furrows primarily formin fine-grained cohesive sediments where the bottomis swept by directionally stable currents (Flood, 1983;Dyer, 1982; Hollister et al., 1974; Flood and Johnson,1984). Sedimentary furrows have also been describedfrom parts of Lake Ontario (D.E. Hutchinson, per-sonal communication in Flood and Johnson, 1984).However, compared to these furrows, the Akademich-esky Ridge examples are not arranged in a regularlyspaced pattern and have larger dimensions. Thoughbottom-current action is proposed for the origin ofthese features, the factors controlling their locationand spacing remain unclear.
A tentative mapping of the incision pattern sug-gests that they have a NE–SW orientation, slightlyoblique to the ridge’s axis.
Bottom currents
From the above-described observations and interpreta-tions of our new seismic data, it is clear that the Aka-demichesky Ridge area in Lake Baikal is characterisedby several, previously unreported depositional featurespointing to the presence of bottom currents influenc-ing the ‘normal’ pelagic/hemipelagic sedimentationon this intra-basin high. In general, bottom-currentcontrol is more consistently observed on the crest of
the ridge, on both the NWHB and SEHB, where thesediments of Unit B are thinner and less continuous.Here bottom currents may have acted on the lake floorby winnowing and scouring the sediments in someareas and allowing deposition elsewhere as waves ordrifts. All described features are consistent with a gen-eral NNE bottom-current direction, slightly obliquewith respect to the main bathymetric contours of theridge (NE–SW).
Very little is known about the current regime onAkademichesky Ridge, and the presence of bottomcurrents important enough to modulate the sedimen-tary record is highly unexpected, and not straight-forward to explain. It may be related to the present-day observed process of deep-water mixing due tothe salinity difference between the North and CentralBasins (Hohmann et al., 1997). Hereby, saline waterflows from the Central Basin and sinks into the deeppart of the North Basin. Our data now indicate thatthis flow process seems to have been affecting differ-ent areas of the ridge in different ways and intensitiesduring the deposition of all Unit B until today. Thuswe interpret the sediments belonging to unit B in thestudy area (on the crest of the ridge) as a contouritesystem in which different drifts have been producedby the interaction of the bottom current flow with thecomplex morphologies of the ridge. We infer that in-fluence on flow direction and intensity exerted by ridgemorphologies was crucial for drift growth.
There are no direct data to constrain the age of theonset of bottom-current controlled sedimentation inthis area. However, a rough estimate may be obtainedwith the ‘Reflection Seismics RadioCarbon (RSRM)Method’ (Cohen et al., 1993; Lezzar et al., 1996).If the average sedimentation rate of 4.0 cm/ky calcu-lated for the late Early Pleistocene to Holocene period(Kuzmin et al., 1997, 2000; BDP Members, 1998) isalso assumed to be appropriate for the previous period,a minimum age of 3.5 Ma can be inferred for the baseof Unit B on the crest of the Ridge and hence for theonset of bottom-current controlled sedimentation inthis area (Ceramicola et al., 2000; Ceramicola, 2001).This very rough estimate was obtained without correc-tion for compaction on the sediment package close tothe crest of the ridge showed in Figure 6. This estimatesuggests that the present-day processes of water circu-lation in Lake Baikal have been existing since at leastthe onset of the last major rifting pulse, i.e., the begin-ning of the Late Rift Stage of Mats et al. (2000). At thistime the Baikal Rift basins underwent a major tectonicre-organisation, which was accompanied by a sudden
460
Figure 9. Strike seismic profile 26 across the NWHB (above), and schematic line drawing with reflector terminations indicated by short arrows(below). For location see Figure 1. Crossing with profile 12 is shown (Figure 10). A buried drift confined between two highs of the NW slopeof the ridge and minor (subsidiary) patch drifts are shown. Moats are present between the drifts and the basement relief. The mound-and-moatfacies only occurs below a surface (‘a’) that is at the same level of the highest of the two confining relief, whereas a mostly uniform sedimentdrape is observed above.
increase in basin subsidence and a strong uplift ofthe rift shoulders (the latter undoubtedly affecting theprecipitation and drainage patterns. . . ). In the deeperparts of the ridge, progressively subsided as a conse-quence of earlier rifting pulses, Unit B may span upto an age of about 9 Ma. There, substratum morpholo-gies are less irregular and reflections of Unit B assume
a more even and sub-parallel character. However theinfluence of weak bottom currents during depositionof the older part of Unit B may not be ruled out. Inthis area, the absence of drifts may be related to theabsence of prominent substratum irregularities capableof effectively influence the direction and intensity ofthe bottom-current flow.
461
Fig
ure
10.
Dip
seis
mic
profi
le12
acro
ssth
eN
WH
B(l
eft)
,an
dsc
hem
atic
line
draw
ing
with
refle
ctor
term
inat
ions
indi
cate
dby
shor
tar
row
s(r
ight
).Fo
rlo
catio
nse
eFi
gure
1.C
ross
ing
with
profi
le26
issh
own
(Fig
ure
9).T
hedr
ift(
confi
ned
ina
NW
-SE
dire
ctio
nas
show
nin
the
cros
sing
profi
le26
)sh
ows
alim
ited
upsl
ope
mig
ratio
nal
ong
the
stee
psi
deof
the
ridg
efr
omw
hich
itis
sepa
rate
dby
am
oat.
Dow
nlap
term
inat
ions
ofre
flect
ors
are
visi
ble
with
inth
em
oata
ndin
the
rela
tivel
yfla
tare
abe
neat
hth
em
ain
body
ofth
edr
ift.
Term
inat
ion
ofre
flect
ors
else
whe
rear
eof
onla
p-ty
peag
ains
tthe
rem
arka
bly
stee
pri
dge
(gen
eral
lybe
twee
n10
◦ and
20◦ o
nav
erag
e).
462
Conclusions
Sedimentary mounds, and sedimentary thickness vari-ations unpredictably related to the underlying mor-phology, observed on high-resolution seismic pro-files through the sedimentary cover of AkademicheskyRidge, an isolated structural high in the central partof Lake Baikal (Russia), are interpreted to belong toa contourite system generated under the influence ofbottom currents. We hence interpret the observed fea-tures as small-scale, lacustrine drifts, comparable inorigin, shape and overall sedimentary characteristics,to the deep-sea drifts.
Four basic types of geometries are identified:(1) slope-plastered patch sheets; (2) patch drifts;(3) confined drifts; (4) fault-controlled drifts.
Bottom-current activity is also evidenced by thepresence of sediment waves within the drift deposits.Non-depositional features include: high-amplitude re-flectors associated with baselap terminations of reflec-tors, moats intervening between drifts and topographichighs, and randomly occurring linear incisions.
The importance of the steepness of the underlyingbasement in controlling the geometry of the reflec-tor terminations has been stressed by showing thatthe widespread downlapping usually observed at thebase of deep-sea contourite drifts is often replaced byonlapping on the steep (10◦–20◦)flanks of the Ridge.
From the analysis of baselap terminations, mi-gration directions, and the preferential location ofdrifts behind basement knolls and other topographi-cal anomalies, a NNE-ward flowing bottom current isinferred to have controlled the deposition of the drifts.
A pattern of thermohaline circulation similar tothat presently responsible for deep water mixing be-tween the North and Central Basins is inferred to beat the origin of bottom currents that have affecteddeposition on the Akademichesky Ridge.
The onset of the bottom currents that are inferredto presently affect sedimentation on the crest of theridge is roughly estimated to be at least 3.5 Ma basedon an average sedimentation rate of 4.0 cm/ky. Thisage is generally regarded as the start of the last majortectonic pulse of rift basin development in the Baikalregion. The influence of weak bottom currents duringdeposition of older part of Unit B (since about 9 Ma)in the deeper parts of the ridge (outside the study area)may not be ruled out.
Several different drift morphologies have beendocumented in this paper and their formation spec-ulated upon. Nevertheless, this is far from giving a
complete picture of the lacustrine contourite drifts oc-curring in the Akademichesky Province, because thereare many geometries that require further investigation.A broader analyses of the interrelated factors control-ling these geometries is needed as well as detailedmapping of the small-scale drift bodies, moreover anextensive swath multibeam bathymetry survey wouldbe indispensable. Current velocity and variability,length of time over which the bottom current processeshave operated, detailed estimation of the direction ofthe currents, are some of the factors principally con-trolling the drift geometries were necessarily left outfrom the present study.
Acknowledgements
This study has been conducted in the framework of S.Ceramicola’s Ph.D. which was carried out of RCMGwith the support of the EC-Human Capital and Mo-bility Programme. Seismic data were collected in theframework of projects funded by the Belgian SciencePolicy Office and by INTAS, and endorsed by BICER.We thank the captain and crew of R/V Vereshchaginfor their help during data acquisition, W. Versteeg forthe processing of the seismic data, M. Grachev andJ. Klerkx for continuous logistic and scientific sup-port, and M. Sturm and R. Kipfer for stimulating andlively discussions about drifts and currents in LakeBaikal. We would like also to thank the two reviewersR. Flood and R. Wynn for their comments and sug-gestions, which helped us to significantly improve themanuscript.
References
Back, S., De Batist, M., Kirillov, P., Strecker, M., and Vanhauwaert,P., 1998, Frolikha Fan: A large Pleistocene glacio-lacustrineoutwash-fan in northern Lake Baikal, Siberia, J. Sed. Res. 68(5),841–849.
Back, S., De Batist, M., Vanhauwaert, P. and, Strecker, M.,1999, Quaternary depositional systems in Northern Lake Baikal,Siberia, J. Geol. 107(1), 1–12.
BDP (Baikal Drilling Project) Members, 1998, Continuous recordof climate changes in Lake Baikal sediments for last 5 My,Geologiya i Geofizika 39, 139–156 (in Russian).
Bradbury, J. P., Bezrukova, Y. V., Chernyaeva, G. P., Colman, S. M.,Khursevich, G., King J. W., and Likoshway, Y. V., 1994. Asynthesis of post-glacial diatom records from Lake Baikal, J.Paleolimnol. 10, 213–252.
Ceramicola, S., 2001, Lake Baikal (Siberia): The depositional,stratigraphic and structural response to rifting activity in Aka-demichesky Ridge accommodation zone using high-resolutionseismic reflection and core data. Unpublished PhD thesis, Fac-ulty of Sciences, University of Gent (Belgium), 183 pp.
463
Ceramicola, S., Khlystov, O., Grachev, M., De Batist, M., and Hen-riet, J.-P., 1998, Tectonic Evolution and Depositional Processesin Akademichesky Ridge, Lake Baikal (Siberia) from High-Resolution Reflection Seismics and Drilling/Coring Data, 15th
IAS International sedimentological Congress (Alicante, Spain),Abstracts, 245–246.
Ceramicola, S., De Batist, M., Rebesco, M., and Grachev M., 2000,Drift Deposits in a Lacustrine Environment? A Case Study fromLake Baikal (Russia), in Rebesco, M. and Marchetto, A. (eds.),Seismic expression of contourites and related deposits: a seismicworkshop (16–18th October, Trieste, Italy), Abstracts, IstitutoNazionale di Oceanografia e di Geofisica Sperimentale (OGS),Trieste, Italy, 27–28.
Carter, L., and McCave, I. N., 1994, Development of sedimentdrifts approaching an active plate margin under the SW Pa-cific Deep Western Boundary Undercurrent, Paleoceanography9, 1061–1085.
Cohen, A. S., Soreghan, M. J., and Scholz, C. A., 1993, Estimatingthe age of formation of lakes: An example from Lake Tanganyka,Geology 21: 511–514.
Colman, S. M., Carter, S. J., Hatton, J., and Haskell, B., 1993,Cores collected in Lake Baikal, Siberia, by the U.S. Geolog-ical Survey, 1990 to 1992: Visual description, photographs,x-radiographs, bulk-density measurements, and grain-size analy-ses, USGS Open-File Report 94–445, 39 pp.
Colman, S. M., Peck, J. A., Karabanov, E. B., Carter, S. J., Brad-bury, J. P., King, J. W., and Williams, D. F., 1995, Continentalclimate response to orbital forcing from biogenic silica recordsin Lake Baikal, Nature 378, 769–771.
Deike, R. G., Granina, L., Callender, E., and McGee, J. J., 1997,Formation of ferric iron crust in Quaternary sediments of LakeBaikal, Russia, and implications for paleoclimate. Mar. Geol.139, 21–46.
Dyer, K. R., 1982, The initiation of sedimentary furrows by standinginternal waves, Sedimentology 29, 73–95.
Edgington, D. N., Klump, J. V., Robbins, J. A., Kusner, Y. S., Pam-pura, V. D., and Sandimirov, I. V., 1991, Sedimentation rates,residence times and radionuclide inventories in Lake Baikal from137Cs and 210Pb in sediment cores, Nature 350, 601–604.
Faugères, J. C. and Stow, D., 1993, Bottom-current-controlled sedi-mentation: A synthesis of the contourite problem, Sed. Geol. 82,287–297.
Faugères, J.-C., Stox, D. A. V., Imbert, P., and Viana, A., 1999,Seismic features diagnostic of contourite drifts, Mar. Geol. 162,1–38.
Flood, R. D., 1983, Classification of sedimentary furrows and amodel for furrow initiation and evolution, Bull. Geol. Soc. Am.94, 630–639.
Flood, R. D. and Johnson, T. C., 1984, Side-scan targets inLake Superior – Evidence for bedforms and sediment transport,Sedimentology 31, 311–333.
Galazy, G. I. (ed.), 1993, Baikal Atlas. Federal Agency for Geodesyand Cartography, Moscow (in Russian).
Goldyrev, G. S., 1982, Sediments formation and Quaternary historyof Baikal Depression, Nauka, Novosibirsk (in Russian).
Grachev, M. A., Likhoshway, Y. E. V., Khlystov, O. M., Bezrukova,Y. E. V., Veinberg, E. V., Goldberg, E. L., Granina, L. Z., Kor-nakova, E. G., Lazo, F. I., Levina, O. V., Letunova, P. P., Otinov,P. V., Pirog, V. V., Fedotov, A. P., Yaskevich, S. A., Bobrov, V. A.,Sukhorukov, F. V., Rerchikov, V. I., Fedorin, M. A., Zolotatyov,K. V., Kravchinsky, V. A., 1997, Signals of paleoclimates of Up-per Pleistocene in Lake Baikal sediments, Geologiya i Geofizika38, 957–980 (in Russian).
Grachev, M. A., Vorobyova, S. S., Likhoshway, Y. E. V., Goldberg,E. L., Ziborova, G. A., Levina, O. V., and Khlystov, O. M.,1998, A high-resolution diatom record of the paleoclimates ofEast Siberia for the last 2.5 My from Lake Baikal, Quater. Sci.Rev. 17, 1101–1106.
Granina, L., Karabanov, E., and Callender, E., 1994, Relics of fer-romagnese formations in the bottom sediments of Lake Baikal,Quater. Sci. Rev. 1, 1101–1106.
Halfman, J. D. and Johnson, T. C., 1984, The sediment textureof contourites in lake Superior, in Stow, D. A. V. and Piper,D. J. W. (eds.), Fine-Grained Sediments: Deep-Water Processesand Facies, Geol. Soc. Spec. Publ. 15, 293–307.
Hohmann, R., Kipfer, R., Peeters, F., Piepke, G., and Imboden,D. M., 1997, Processes of deep-water renewal in Lake Baikal,Limnol. Oceanogr. 42(5), 841–855.
Hollister, C. D., Flood, R. D., Johnson, D. A., Lonsdale, P., andSouthard, J. B., 1974, Abyssal Furrows and Hyperbolic EchoTraces on the Bahama Outer Ridge, Geology 12, 395–400.
Howe, J., Stoker, M. S., and Stow, D. A. V., 1994, LateCenozoic sediment drift complex, Northeast Rockall Trough,Paleoceanography 9, 989–999.
Johnson, T. C., 1996, Sedimentary Processes and Signals of PastClimatic Change in the Large Lakes of the East African RiftValley, in Johnson, T. and Odada, E. (eds.), The Limnology,Climatology and Paleoclimatology of the East African Lakes,Gordon & Breach Pub. Inc., Newark, pp. 367–412.
Johnson, T. C., Carlson, T. W., and Evans, J. E., 1980, Contouritesin Lake Superior, Geology 8, 437–441.
Kuzmin, M. I., Grachev, M. A., Williams, D. F., Kawai, T., Horie,S., and Oberhaensli, H., 1997, A continuous record of paleocli-mates of the last 4.5 million years from Lake Baikal, Geologiyai Geofizika 38, 1021–1023 (in Russian).
Kuzmin, M. I., Karabanov, E. B., Prokopenko, A. A., Gelety, V. F.,Antipin, V. S., Williams, D. F., and Gvozdkov, A. N. 2000, Sedi-mentation processes and new age constraints on rifting stages inLake Baikal: Results of deep-water drilling, Internl. J. Earth Sci.89, 183–192.
Lezzar, K. E., Tiercelin, J. J., De Batist, M., Cohen, A., Bandora,Th., Van Rensbergen, P., Mifundu, W., and Klerkx, J., 1996, Newseismic stratigraphy and Late Tertiary history of the North Tan-ganyika Basin, East African Rift system, deduced from multifoldreflection and high-resolution data and piston core evidence,Basin Res. 8, 1–28.
Mats, V. D., Khlystov, O. M., De Batist, M., Ceramicola, S.,Lomonosova, T. K., and Klimansky, A., 2000, Evolution of theAcademician Ridge Accommodation Zone in the central part ofthe Baikal Rift, from high-resolution reflection seismic profilingand geological field investigations, Internl. J. Earth Sci. 89(2),229–250.
McCave, I. N. and Tucholke, B. E., 1986, Deep current-controlledsedimentation in the western North Atlantic, in Vogt, P. and Tu-cholke, B. (eds.), The geology of North America, The westernNorth Atlantic region, Decade of North America, GeologicalSociety of America, Boulder, Colorado, Vol M, pp. 451–468.
Mitchum, R. M., 1985, Seismic stratigraphic expression of sub-marine fans, in Berg, O. R. and Woolverton, D. G. (eds.),Seismic Stratigraphy II – An integrated approach to hydrocarbonexploration, AAPG Mem. 39, Tulsa, Oklahoma, pp. 117–138.
Moore, D. J., 1969, Reflection Profiling Studies of the CalifornianContinental Borderland: Structure and Quaternary TurbiditeBasin, Geol. Soc. Am. Spec. Publ. 107, 1–42.
Nelson, C. H., Karabonov, E. B., and Colman, S. M., 1995, LateQuaternary turbidite systems in Lake Baikal, Russia, in Pick-ering, K. T., Lucchi, F. R., Smith, R. U., Hiscott, R. N.,
464
and Kenyon, N. (eds.), An Atlas of Deep-Water Environments,Chapman and Hall, London, pp. 29–33.
Stoker, M. S., 1998, Sediment drift development on the Rockallcontinental margin, off NW Britain, in Stoker, M. S., Evans,D., and Cramp, A. (eds.), Geological Processes on ContinentalMargins: Sedimentation, Mass-Wasting and Stability, Geol. Soc.Spec. Publ. 129, 229–254.
Stoker, M. S., Akhurst, C., Howe, J. A., and Stow, D. A. V., 1998,Sediment drifts and contourites on the continental margin, offNorthwest Britain, Sed. Geol. 115, 33–52.
Stow, D. A. V., Faugères, J. C., Viana, A., and Gonthier, E., 1998,Fossil contourites, a critical review, Sed. Geol. 115, 3–32.
Stow, D. A. V. and Lovell, J. B. P., 1979, Contourites: Their recog-nition in Modern and Ancient Sediments, Earth-Sci. Rev. 14,251–291.
Stow, D. A. V. and Piper, D. J. W., 1984, Deep-water fine-grainedsediments: Facies models, in Stow, D. A. V. and Piper, D. J. W.(eds.), Fine Grained Sediments, Deep water Processes andFacies, Geol. Soc. Spec. Publ. 15, 611–646.
USSR, Ministry of Defence, 1992, Bathymetry charts of LakeBaikal, Siberia. Chief Directirate of Navigation and Oceanigra-phy, Moscow (Scale 1:200,000).
Zonenshain, L. P., Golmshtok, A. Y. A., Hutchinson, D., 1992,Baikal Rift structure, Geotectonics 6(5), 396–403.
Zonenshain, L. P. and Kazmin, V. G., 1995, New data on thehistory of Lake Baikal: Results of surveys by submersibles,Geotectonics 29(3), 235–247 (English translation from RussianEdition).
Recommended
