Chronology of the Holocene transgression at the North Siberian

Ž .Global and Planetary Change 31 2001 125–139www.elsevier.comrlocatergloplacha

Chronology of the Holocene transgression at theNorth Siberian margin

H.A. Bauch a,b,), T. Mueller-Lupp a, E. Taldenkova c, R.F. Spielhagen a, H. Kassens a,P.M. Grootes d, J. Thiede b, J. Heinemeier e, V.V. Petryashov f

a GEOMAR, Wischhofstrasse 1-3, 24148 Kiel, Germanyb Alfred-Wegener-Institute for Polar and Marine Research, Columbusstrasse, 27568 BremerhaÕen, Germany

c Geography Department, Moscow State UniÕersity, VorobieÕy Gory, 119899 Moscow, Russiad Leibniz Laboratory, Christian-Albrechts-UniÕersitat, Max-Eyth-Strasse 11, 24098 Kiel, Germany¨

e Institute for Physics and Astronomy, UniÕersity of Aarhus, 8000 Aarhus, Denmarkf Zoological Institute, Russian Academy of Sciences, 1 UniÕersitetskaya, 199034 St. Petersburg, Russia

Received 10 February 2000; accepted 23 May 2001

Abstract

To establish a chronology of the Holocene transgression in Arctic Siberia, a total of 14 sediment cores from the LaptevSea continental slope and shelf were studied covering the water depth range between 983 and 21 m. The age models of the

Žcores were derived from 119 radiocarbon datings, which were all analyzed on marine biogenic calcite mainly bivalve.shells . The oldest shell sample was found at the slope and dates back to about 15.3 cal. ka, indicating that the time interval

Ž .investigated starts prior to the onset of the meltwater pulse 1A ;14.2 cal. ka when global sea-level rose dramatically. TheŽ .inundation history was reconstructed mainly on the basis of major changes in average sedimentation rates ASR , but also

other sedimentological parameters were incorporated. A diachronous reduction in ASR from the outer to the inner shelfregion is recognized, which was related to the southward migration of the coastline as the primary sediment source. Weestimate that the flooding of the 50-, 43-, and 31-m isobaths was completed by approximately 11.1, 9.8, and 8.9 cal. ka, andthat Holocene sea-level highstand was approached near 5 cal. ka. Between these time intervals, sea level in the Laptev Searose by 5.4, 13.3, and 7.9 mmryear, respectively. q 2001 Elsevier Science B.V. All rights reserved.

Keywords: Arctic Siberia; Holocene sea level; land–ocean interaction; shelf sedimentation

1. Introduction

The extent of the Eurasian ice sheets during theWeichselian has recently been thoroughly revised by

) Corresponding author. Department of Paleoceanology, Re-search Center for Marine Geosciences, GEOMAR-University ofKiel, Wischhofstrasse 1-3, 24148 Kiel, Germany. Tel.: q49-431-600-2853; fax: q49-431-600-2941.

Ž .E-mail address: [email protected] H.A. Bauch .

Ž .Svendsen et al. 1999 . Even though many detailsstill remain to be clarified, now the eastern boundaryof the large Eurasian ice sheet never extended furthereast than the Taymyr Peninsula and the Central

ŽSiberian Uplands Forman et al., 1999a; Moller et¨.al., 1999; Svendsen et al., 1999 . Therefore, sedi-

ment cores from the western Eurasian shelvesŽ .Barents and Kara seas frequently show widespreadglaciogenic sediments underneath marine sediments

Žof Holocene age Polyak et al., 1995; Lubinski et al.,

0921-8181r01r$ - see front matter q 2001 Elsevier Science B.V. All rights reserved.Ž .PII: S0921-8181 01 00116-3

( )H.A. Bauch et al.rGlobal and Planetary Change 31 2001 125–139126

.1996; Hald et al., 1999 , whereas sedimentation onthe wide and shallow shelf seas further east remainsunaffected by glaciations and shows no such fea-tures.

Due to this uneven distribution of ice sheets alongthe Arctic Ocean’s margins, the sea level on theshallow Siberian shelf seas outside the limits of lastglacial ice-sheets is expected to have risen with someregional time differences, compared to those shelveswhich came under the effect of postglacial verticalcrustal movements. Because of the inundation offormerly exposed landmasses, sediment records fromcentral and eastern Siberian shelves can providecharacteristic lithological features of the Holocene

Ž .transgression Elias et al., 1996; Bauch et al., 1999 .For a long time, however, the general scarcity ofsuitable sediment cores from these shelves has ham-pered the establishment of an independent chronol-ogy of the last transgression in this part of the Arctic.A first, detailed insight into postglacial sedimentaryevolution came from a few radiocarbon-dated coresfrom the modern Laptev Sea. They showed a distinc-tive, sea level-related change in the input of terres-trial-derived sediment material due to the gradual

Žretreat of the paleocoastline Bauch et al., 1999;.Stein and Fahl, 2000; Mueller-Lupp et al., 2000 .

Meanwhile, more sediment cores have been re-covered from various water depths of the Laptev Seashelf, ranging from the continental slope to the shal-low, inner shelf region. This study, therefore, com-piles previously published sedimentological data to-gether with a large number of new radiocarbon-basedcore chronologies in order to make a detailed recon-struction of the Holocene transgression for the north-ern Siberian margin. Because this region was unaf-fected by ice sheets during the last glaciation, ourregional sea-level history may have direct implica-tions for reconstructing sea levels in other coastalareas of the Arctic, which, like the Laptev Sea shelf,also remained uncovered by the last glacial ice sheets.

2. Sedimentary environment, core material, andmethods

The submarine channels on the Laptev Sea shelfŽ .Fig. 1 are a distinctive topographic feature andwere probably formed by the large rivers of theLaptev Sea hinterland when sea level was signifi-

Ž .cantly lower than today Holmes and Creager, 1974 .During times of postglacial flooding of the shelf,these trough-like channels were gradually filled withsediments originating from various sources. Previousinvestigations indicate that since the time when theHolocene sea level had reached a near-modern level,

Žmarine sedimentation became dominant Bauch et.al., 1999; Fahl and Stein, 1999 . However, sediments

were mainly deposited along the submarine channels,Židentifying the rivers as major source Kuptsov and

.Lisitzin, 1996; Kleiber and Niessen, 1999 . Indeed,the large rivers draining into the Laptev Sea trans-port substantial amounts of organic and other sedi-

Žmentary matter onto the shelf each year Alabyan etal., 1995; Cauwet and Sidorov, 1996; Ivanov and

.Piskun, 1999; Rachold and Hubberten, 1999 . Butthe Laptev Sea shelf also underwent strong coastalerosion during the last transgression, a process whichwas of high significance to shelf sedimentation dueto the widespread occurrence of labile permafrost

Ž .coasts Are, 1999 . Because of these various pro-cesses, Holocene shelf sediments in the Laptev Seausually have a mixed terrestrial and marine originŽ .Fahl and Stein, 1999; Bauch et al., 1999 .

All 14 sediment cores dealt with in this studywere taken from the Laptev Sea during joint Ger-man–Russian expeditions conducted between 1993and 1998. The cores were obtained from water depths

Ž .ranging between 983 and 21 m Fig. 1 . Three of thecores studied were taken with a vibro kasten corer,allowing a maximum penetration of 6 m, whereas allother cores were obtained using a gravity coring

Žsystem eight giant kasten corer; three barrel gravity

Ž . Ž .Fig. 1. Bathymetric map in meters of the Laptev Sea shelf showing core positions top panel . Asterisks indicate sites of collected prebombŽ .bivalves that were used to determine the modern reservoir age see also Table 1 . Panel below shows sediment recovery and the different

core groups according to water depth. For convenience, the bold numbers in core names only are referred to on the map and throughout themain text.

( )H.A. Bauch et al.rGlobal and Planetary Change 31 2001 125–139 127


.corer . Total sediment recovery varied between 200and 900 cm per core, with the longest cores havingbeen taken from the continental slope. Experiencefrom these expeditions has shown that the best sedi-ment recovery on the shelf was usually achievedwithin the submarine channels.

To reconstruct the Holocene transgression of theLaptev Sea, reliable age determination of the sedi-ment cores is crucial. Previous investigations haveshown that dating these sediments with bulk sedi-ment samples usually yields unreliable results due tothe admixture of older and younger organic materialŽHolmes and Creager 1974; Kuptsov and Lisitzin,

.1996 . Therefore, the chronology of the cores in thisstudy is entirely based on radiocarbon ages from

Ž .marine biogenic calcite mostly from bivalve shells ,determined by means of accelerator mass spectrome-

Ž .try AMS .The modern bivalve species assemblage in the

Laptev Sea has distinct spatial patterns not only withregard of the water depth but also in dependence ofthe distribution of water mass salinities, which are

Žrelated to the influence of riverine waters Sirenko et.al., 1995 . A total of at least 10 species were used

14 Žfor C dating Astarte crenata, Leionucula bellotii,Lyonsia arenosa, Macoma cf. moesta, Macoma cf.calcarea, Macoma sp., Nuculana sp., Portlandia arc-tica, Yoldia amygdalea hyperborea, Yoldia sp.,

.Yoldiella cf. intermedia, Yoldiella cf. lenticula . Thequality of preservation of the dated shells was gener-ally good. They were either found in situ with bothvalves in place, or the periostracum was still pre-served, indicating insignificant lateral transportation.

All AMS 14C-dates were converted to calendarŽ .years B.P. cal. ka using the programme CALIB 4.3

Ž .Stuiver et al., 1998 and taking into account theresults of 14C reservoir determination for the Laptev

Ž .Sea see below . In order not to overinterpret abruptchanges in local sedimentation regime, and to over-come the problem of occasional age reversals, aver-

Ž .age sedimentation rates ASR were calculated forentire core section to unveil the major changes inoverall depositional regime. Since AbombB carbon,i.e., 14C ages younger than 1950, was often detectedin near-surface sediments, a zero age was assumedfor the sediment surface.

3. Results

3.1. ReserÕoir age

To determine specific marine 14C reservoir valuesfor the Laptev Sea region, recent samples of bivalves

Ž .species were radiocarbon-dated Table 1 . Five of thesix specimens originated from the central Laptev Sea

Ž .shelf, one from near the coast at Tiksi Fig. 1 . Themolluscs were collected alive during expeditions in1901 and 1937 and, with their soft parts still inplace, were kept in a solution of formaldehyde at theSt. Petersburg Zoological Museum. The 14C datingwas conducted at the AMS 14C facility in Aarhus,Denmark, on parts of the shells.

The results yielded reservoir ages ranging be-tween 860 and 295 years and averaging 451 yearsŽ .Table 1 . Four samples from the central shelf regionŽ .samples 1, 4, 5 and 6 gave ages significantly lessthan the modelled mean age of the global ocean

Žmixed layer, which is 402 years Stuiver et al.,.1998 . However, we also observed some other dis-

crepancies. For instance, there is a notable age offset

Table 1Ž .Results of radiocarbon analyses of prebomb shells age obtained from the Laptev Sea and the derived reservoir ages

aLaboratory Bivalve species Vessel, year Geographical Water depth Age Tree-ring age Reservoir age14 14 14w x w x w x w xnumberrsample position, NrE m C years C years C years

X XAAR-2769r1 Nuculana pernula RrV Sadko, 1937 76838 r118820 46 495"40 155"12 340"40X XAAR-2770r2 Nuculana pernula RrV Zarya, 1901 75842 r124841 31 615"45 77"11 538"45

AAR-2771r3 Portlandia aestuariorum RrV Sadko, 1937 Tiksi Bay 8 1015"55 155"12 860"55X XAAR-2772r4 Portlandia arctica RrV Sedov, 1937 74842 r134834 25 500"60 155"12 345"60X XAAR-2773r5 Leionucula bellotii RrV Sedov, 1937 75836 r130811 45 485"50 155"12 330"50X XAAR-2774r6 Leionucula bellotii RrV Sadko, 1937 76838 r118820 46 450"45 155"12 295"45

a Ž .Calculated by subtracting the tree-ring age 10-year average prior to collecting year from the radiocarbon age.


between different species from the same localityŽ .samples 1 and 6 , a feature that may be linked to adifferent ecological habitat preference of each of the

Ž .two species cf. Forman and Polyak, 1997 . Thecomparatively old age calculated for sample 3 fromthe low-saline Tiksi Bay, on the other hand, may beoverprinted by a local hardwater effect and is proba-bly not representative for the rest of the shelf.

The relatively high age of sample 2 comparedwith the others from the central shelf may be relatedto specific oceanographic circumstances. The LaptevSea remains ice-covered for about 10 months of the

Ž .year Kassens et al., 1998 , but develops a distincteast–west stretching flaw polynya during this timeŽ .Dmitrenko et al., 1998 . Because this flaw polynyais generated by off-shore winds, it enhances the rate

Žof vertical mixing of the shelf water Dethleff et al.,.1998 , thereby facilitating a rapid exchange with

atmospheric CO . This process may cause the rela-2

tively young reservoir ages in the central shelf regionŽ .Table 1 . However, intermediate waters from theArctic Ocean are known to penetrate onto the central

Ž .Laptev Sea shelf Dmitrenko et al., 1999 and mayexplain the relatively high reservoir age of sample 2Ž .Table 1 . Considering all this, we have calculatedthe average reservoir age on the basis of samples 1,2, 4, 5 and 6 to be 370"49 years. While convertingall radiocarbon dates obtained from the sediment

Žcores using the programme Calib 4.3 Stuiver et al.,.1998 , the determined reservoir age was subtracted.

3.2. Radiocarbon dating of sediment cores

On the basis of their water depth, the 14 coresŽ .were separated into four groups Fig. 1 : The first

group consists of four cores from the continentalslope and covers the water depths between 983 and114 m; the second group originates from the outershelf and comprises two cores from 77- and 60-mwater depth; a third group is made up of four cores

Žrecovered from the central shelf region 51–42-m.water depth ; the fourth group derives from the inner

shelf and incorporates the four most shallow-waterŽ .sediment cores studied 32–27-m water depth .

A total number of 119 downcore samples wereŽ .radiocarbon-dated Table 2 . Within five of the cores

investigated, a AbombB age was detected. The oldestanalyzed shell sample dates back to about 15.4 cal.

ka, indicating that the time interval covered by ourcores starts before the onset of meltwater pulse 1AŽ .MWP-1A , i.e. just before the time when globalsea-level experienced the most dramatic rise of the

Ž .deglaciation Fairbanks, 1989; Bard et al., 1996 .

3.3. Sedimentation at the continental slope

The oldest 14C-dated sample at the continentalslope is found in the western part of the Laptev Sea,

Ž .in core 54, and dates back to 15.3 cal. ka Table 2 .Below this sample, there is a considerable sedimentsequence, which could not be dated because of thelack of biogenic calcite. In this sequence, however,small-sized plant debris was noted together with theconspicuous occurrence of the minerals mica andvivianite. The two uppermost dates in the core indi-cate a significant sedimentary change above 138-cmcore depth when the ASR decreased from 92cmrkyear before 11.1 cal. ka to 12 cmrkyear untilpresent. A similar distinctive change in ASR asfound in core 54 is recognized in the cores from the

Ž .eastern slope area cores 18-2, 18-3, 58 , but thenthis change is identified about 2 kyears later, proba-bly due to a denser sample resolution. The timeinterval when the ASR became reduced is similar inall of the three eastern cores. However, in contrast tocore 58 from 983-m water depth, cores 18-2 and18-3 from the upper part of the slope reveal a muchhigher ASR prior to this event. It should be notedthat no datable marine biogenic calcite was found in

Ž .core 58 between 252-cm core depth 9.4 cal. ka andŽ .the sediment surface bomb age . Because the core is

from a steep slope, one cannot preclude that gravity-controlled processes have removed some of the sedi-ments deposited after 9.4 cal. ka.

3.4. Sedimentation in the outer shelf region

The two investigated cores from the outer shelfregion, cores 25 and 59, cover a water depth range of17 m. Core 25 from 77-m water depth basicallyshows the same sedimentation pattern as found at theeastern continental slope, where the ASR steeplydecreased after 8.9 cal. ka. The 14C dates in core 59Ž .from 60-m water depth , however, indicate morechanges in ASR. The first major change occurred


Table 2Ž .Radiocarbon dates, calibrated calendar years, and computed average sedimentation rates ASR of Laptev Sea sediment cores

after 9.6 cal ka and the second after 5.3 cal. kaŽ .Table 2 .

3.5. Sedimentation in the central shelf region

The four cores from the central shelf span a waterdepth range of 9 m. Core 35 from 51-m water depth

in the Yana paleoriver valley dates back to 11.3 cal.ka at the bottom of the core. The 14C dates furtherupward indicate a disturbance, either above 266-cmcore depth or somewhere between 143- and 80-cmcore depth. In comparison with nearby core 02,which exhibits a relatively reliable chronology, astrong similarity in depthrage relation is recognized


for the last ;7.6 cal. ka as well as for the timeinterval beyond 9.6 cal. ka. Therefore, it is assumedthat the two dated shells which gave the age rever-sals in core 35 have either been displaced from thecore surface along the metal casing during the coring

Žprocess, or that the upper section above 143 cm core.depth was deposited during rather recent time, per-

Žhaps displaced by iceberg gouging Lindemann,.1995 . Below 400-cm core depth, the bivalve fauna

in core 35 is rather monospecific, being composed ofthe brackish water species Portlandia arctica, whichin the Laptev Sea today dominates the shallow,

Ž .southeastern Laptev Sea Sirenko et al., 1995 , wherecontents of suspended matter is high and salinity islow due to strong riverine runoff.

Core 41, from outside the Lena paleovalley, wastaken from a topographic depression. As opposed tocores 35 and 02, which contain bivalves and biotur-bational structures throughout, core 41 below 112-cmcore depth consists of homogeneous, blackish siltyclay containing only few small bivalves. Below 395-cm core depth, no shells were found at all. A sandysilt layer at 119- to 112-cm core depth was dated to9.8 cal. ka containing numerous bivalves in its toppart. Because the sand content at this depth levelclearly reflects a much higher-energy depositionalregime than below, it may be related to the transgres-sion. The sediments above are clayey silt with bio-turbation features increasing upwards, particularly

Žabove 80-cm core depth interpolated age is ;6.5.cal. ka .

The chronology, sedimentology as well as geo-chemistry of core 99 from 48-m water depth in theKhatanga–Anabar paleovalley has been described in

Ž .detail by Bauch et al. 1999 . The lower part of theŽ .core up to ;165-cm core depth lacks biogenic

CaCO , but contains considerable amounts of small-3Ž .sized plant debris and mica Peregovich et al., 1999 .

It is a laminated, nonbioturbated, organic-rich, sandysilt with slightly coarser grain composition near thebase, ranging from silt to sand. The higher sandcontent in this lower core section indicates a stronger,more dynamic depositional regime than higher up in

Žthe core, probably because of fluvial influence Bauch.et al., 1999 . Datings of the screened plant debris

from this lower section unveiled ages all older than12.3 cal. ka. Above 165-cm core depth, a change insedimentary environment is marked by the first oc-

currence of marine bivalves in this core, which weredated to 11.1 cal. ka. The core exhibits a notable

Ž .decrease in ASR after ;7 cal. ka Table 2 . Acharacteristic change in the carbon isotopic signature

Ž 13 .of the bulk organic content d C towards signifi-org

cantly heavier values, i.e., an increasing proportionof marine organic carbon, was also previously identi-

Ž .fied at this core level Bauch et al., 1999 .

3.6. Sedimentation in the inner shelf region

The cores from water depths between 21 and 32m do not produce dates much older than ;9 cal. kaŽ .Table 2 . Cores 92-12 and 92-13 from the samelocality in the Lena paleovalley show a change inASR after 7.3 and 7.1 cal. ka, respectively. In core92-12, which has an extrapolated age of ; 9 cal. kaat the bottom, diatom assemblages indicate a domi-nance of freshwater species until about 7 cal. kaŽ .Polyakova and Bauch, 1999 .

Core 62 from the Yana paleovalley gives an ASRof 50 cmrkyear. On the basis of the 14C dates, nosignificant change in ASR is observed. However,previous investigations on this core have revealedsilty sand in the lowermost ;30 cm containing

Žsubstantial amounts of woody peat clasts Bauch et.al., 2001 . The lowermost mollusc shell dated in this

core was in situ, within the upper part of the sandylayer. A continuous decrease in total sediment accu-mulation is recognized in this core after ;7 cal. kaŽ .Bauch et al., 2001 .

Ž .The shallowest site core 80 investigated is lo-Ž .cated just to the northeast of the Lena Delta Fig. 1 ,

where the influence of water runoff from the LenaŽ .River is strong today Bauch et al., 2000 . The

chronology of core 80 indicates a significant de-Ž .crease in ASR after 5.1 cal. ka Table 2 . This

relatively small ASR in such close proximity to thedelta seems unexpected, considering the substantialamounts of suspended matter which is carried by the

Ž .Lena River Ivanov and Piskun, 1999 .

4. Depositional environment

Today, the Laptev Sea shelf down to the 100-misobath covers an area of approximately 450,000 km2


Ž .Fig. 1 . The continental slope breaks steeply nearthis water depth. On the basis of the modernbathymetry, it is apparent that once the transgressingsea had submerged the slope break, any further risein sea level caused very large increases in the arealflooding of formerly exposed land and a swift south-

Ž .ward retreat of the coastline Bauch et al., 1999 . Asrevealed by the chronology data, the cores from theslope mainly cover the time between 15.3 and 9 cal.ka, which is the interval when most parts of theLaurentide, Fennoscandian, Barents- and Kara Sea

Žice shields were melting away e.g., Andersen, 1980;

Elverhøi et al., 1995; Barber et al., 1999; Svendsen.et al., 1999 . Most of this deglaciation falls within

14 Žsignificant C plateaus Bard et al., 1990; Stuiver et.al., 1998 , and may explain the few age reversals

recorded in our cores for this time interval. Thus,estimating deglacial relative sea-level changes seemsmore difficult when based solely on ASR calcula-tions from the slope region. A more comprehensivedepositional reconstruction is therefore attempted onthe basis of ASR in combination with some impor-tant sedimentological features, including alreadypublished data.

Ž .Fig. 2. Major breaks in average sedimentation rates ASR as revealed by sediment cores from different water depths of the Laptev Sea.


One main feature in nearly all our cores is acertain age–depth threshold, recognizable as majorbreak in ASR from high to significantly lower valuesŽ .Fig. 2 . Taking into account some age uncertaintiesbetween the cores, the age of the earliest of theseASR breaks was estimated to lie between ;11.1

Ž .cal. ka in core 54 from the western slope and aboutŽ9 cal. ka eastern continental slope and the outer

.shelf . This younger age coincides in time with theend of the main phase of deglaciation in the Nordic

Ž .seas Bauch et al., 1996 and the establishment offull-marine conditions there and in the western

ŽEurasian Arctic shelf seas Polyak and Sølheim,1994; Forman et al., 1995; Polyak and Mikhailov,

.1996; Elverhøi et al., 1995; Hald et al., 1999 .A second break in ASR occurred around 7.2 cal.Ž .ka averaged from cores 92-12 and 13 . This is

coincident with a characteristic change in d13Corg

towards heavier, more marine values in cores 02 andŽ99 at about 7 cal. ka Bauch et al., 1999; Mueller-

.Lupp et al., 2000 . The cause for this change in theisotopic composition of organic matter most likely isthe sudden decrease in the input of terrestrial organicmatter from rivers and coastal erosion relative tothese sites. More evidence for a change in sedimenta-tion around 7 cal. ka is found in core 62 from 27-mwater depth. In this core, a decrease in total sedimentaccumulation and a significant reduction in the rela-tive content of sand occurred between 7.5 and 7 cal.

Ž .ka Bauch et al., 2001 . Furthermore, the main changefrom a freshwater-dominated diatom assemblage toone with a predominantly marine imprint is noted in

Žcore 92-12 at about 7 cal. ka Polyakova and Bauch,.1999 . All the evidence together implicate a major

change in hydrology and depositional environmentfor the modern 30-m isobath at a time when theglobal Holocene sea-level highstand was nearly ap-

Ž .proached Ingolfsson and Hjort, 1999 . However, a´modern depositional environment in the Laptev Seawas probably established no earlier than 5 cal. ka,

Ž .when in some cores cores 80, 135, 159 anotherŽ .break in ASR is recognized Fig. 2; Table 2 .

5. Time-slice reconstruction

The major changes observed in the stepwise re-duction of ASR in our studied Laptev Sea cores give

ample evidence of a north to south transgressing seaŽ .after the LGM Fig. 2 . This transgression led to

both a southward retreat of the coastline and achange in the sediment supply delivered to eachparticular paleodepth from coastal erosion and flu-vial systems at different times. The material that waseroded during transgression derived from a land-scape probably characterized by ice complexes,

Žthermokarst lakes, and fluvial settings, Alekseev,.1997; Romanovski et al., 1997 .

Major changes in depositional environment forthe interval of deglaciation have been also recog-nized in sediment cores from the northern Kara SeaŽ .Hald et al., 1999 ; however, reconstructing pale-odepths is complicated in this region because ofstrong isostatic rebound during and after the deglacial

Ž .phase Forman et al., 1999b . Because the LaptevSea shelf remained unglaciated and since no glacia-tion affected eastern Taymyr during the LGMŽ .Svendsen et al., 1999 , a reconstruction of theHolocene transgression on this shelf is relativelyunhampered by direct ice-sheet-related verticalcrustal movements. However, it must be assumedthat the increasing water load and the fact that theshelf seas to the west were covered by glacial icecaused some, but probably negligible, isostatic low-ering of the Laptev Sea shelf after the Holocenetransgression.

The cores from the continental slope and outershelf, as shown above, cannot be utilized on thebasis of their present chronology to resolve actualsea-level changes during the first half of thedeglaciation. The lack of marine calcite fossils below

Ž .570 cm in core 54 extrapolated age ;15.9 cal. kain combination with the occurrence of plant debris,mica, and vivianite indicate the proximity of the

Žpaleocoast to the slope break during this time Fig..3 . The mica probably derived from the Taymyr

ŽPeninsula where suitable rocktypes are present Per-.egovich et al., 1999 , whereas the occurrence of the

phosphate-rich mineral vivianite points to a lakeenvironment, in which this authigenic mineral fre-quently forms under organic-rich conditions.

To make estimates on the transgressional timingfor the period until about 11.1 cal. ka, we use the

Ž .Barbados sea-level curve from Fairbanks 1989 cali-Ž .brated into calendar years Fig. 3 . We realize that

such a comparison with the sea-level curve from the


Ž . ŽFig. 3. Comparison of Barbados sea-level curve calibrated into calendar years , modern Laptev Sea water depth profile drawn along 1308.longitude , and lithology logs of three key cores showing ages used for sea-level reconstruction. Bottom panel shows estimations of

Ž .sea-level rise and coastal retreat for three time intervals grey . Note, the advance of the transgressing sea between the onset of globalŽ . Ž .Meltwater Pulse 1A MWP-1A and 11.1 cal. ka arrows in bottom panel was calculated from the Barbados sea-level curve.

North Atlantic has some bias because of both thelack of clear evidence in our cores prior to MWP-1Aand the fact that sea-level reconstructions from dif-ferent ocean basins show some offsets in timing as

Žwell as in absolute magnitude Fairbanks, 1989;Peltier, 1994; Bard et al., 1996; Hanebuth et al.,

.2000 . However, the error seems negligible for theLaptev Sea because of the steep angle of the conti-

Ž .nental slope Fig. 3 . Due to this steep slope, it isapparent that until the end of MWP-1A, the esti-mated southward retreat of the paleocoast was minor

and the total flooded shelf area was rather smallcompared with the interval after MWP-1A and until

Ž .about 11.1 cal. ka Figs. 3 and 4 .The presence of conspicuous plant debris together

with mica was noted also in the central shelf regionŽnear the base of core 135 in combination with the

.low-salinity tolerant bivalve Portlandia arctica andŽbelow the first dated bivalve in core 99 Bauch et al.,

.1999 , allowing us to assume that the paleocoast at;11.1 cal. ka was located near the modern 50-m

Ž .isobath Fig. 4 , which is similar to sea-level rise in


Fig. 4. Reconstruction of the Lapev Sea transgression showing the difference in areal flooding between each time interval investigated. Thetopographic map is based on Russian navigation charts and bathymetric data obtained during several marine German–Russian expeditions.Note that the modern shelf topography does not reflect the actual paleosurface prior to inundation.


Ž .the North Atlantic Fig. 3 . The inundation of thiswater depth level is crucial for the evolution of sealevel and water mass circulation of the Arctic Inte-rior, as it also marks the sill depth of the BeringStrait, allowing for the first time, after the LGM, awater mass exchange between the Arctic and Pacific

Ž .oceans Elias et al., 1996 .Although, the lowermost date in core 41 yielded

an age of about 11.3 cal. ka for a calculated pale-oisobath of about 46 m, the three bivalves foundbelow 112-cm core depth were probably not in situ.

Ž .This is indicated by a reversed age Table 2 , by theŽ .unusual occurrence of silty clay Fig. 3 , and by a

13 Žclear terrestrial d C signature T. Mueller-Lupp,org.unpublished data . It is noteworthy that this core was

obtained not from the submarine Lena valley butfrom a small topographic depression, which, on thebasis of shipboard acoustic data, resembles in sizeand shape as well as in the pattern of sedimentaryinfill thermokarst lakes as they have been describedfrom the western part of the Lena Delta landmassŽ .Rachold and Grigoryev, 1999 . Regardless of thespecific environment, sedimentation in core 41 wasrelatively rapid prior to 9.8 cal. ka. After this time, asandy layer characterizes a sudden change in deposi-tional environment together with high abundance of

Ž .bivalves, bioturbational features Fig. 3 , as well as a13 Žsignificant increase in d C T. Mueller-Lupp,org

.unpublished data . These changes in different sedi-Žmentary features after 9.8 cal. ka i.e., above 112 cm

.core depth probably reflect the final flooding of the43-m paleoisobath. Using this time level as basis, sealevel in the Laptev Sea rose ;5.4 mmryear be-tween 11.1 and 9.8 cal. ka, forcing the coastline

Ž .southward by ;100 km Fig. 3 .The occurrence of peat debris embedded in a

sandy matrix in the lowermost part of core 62 indi-cates a fluvial-type deposition prior to ;8.9 cal. kaand the flooding of the 31-m paleoisobath around

Ž .this time Fig. 4 . The coast and rivers had movedŽsouthward another 70 km since 9.8 cal. ka sea-level

.rise ;13.3 mmryear and a total distance of ap-proximately 450 km from the shelf break in theeastern Laptev Sea. Probably, in combination with afurther decrease of the shelf gradient, this long dis-tance from the coast led to the strong reduction insediment transfer to the outermost shelf and slope

Ž .around 9 cal. ka Fig. 2 .

A further southward migration of the river de-pocenters can be interpreted from the change incomposition of diatom assemblages and from thedecrease in ASR after about 7 cal. ka, noted in thecores from the Lena paleovalley from around 30-mwater depth. Although sea level was probably closeto the modern level by this time, the reduction inASR at about 5.1 cal. ka in core 80 from the inner

Žshelf recorded in cores 159, 135 from the mid-depth.shelf at 5.3 cal ka is regarded by us as the time

when Holocene sea-level highstand in the LaptevŽ .Sea was nearly reached Fig. 3 . This is corroborated

by a rather constant sediment flux recorded in coreŽ .62 after about 5 cal. ka Bauch et al., 2001 . More-

over, time estimate of about 5 cal. ka for the estab-lishment of the modern sea-level in Arctic SiberiaŽ .Fig. 3 is consistent with sea-level records from

Žfurther east in Beringia Mason et al., 1995; Elias et. Žal., 1996 and from outside the Arctic e.g., Ingolfs-´

.son and Hjort, 1999 . Although sea-level rise in thisŽfinal phase seems not great in comparison ;7.5

.mmryear , the total area flooded was the largest ofthe entire transgression, due to the low angle of the

Ž .shelf topography Figs. 3 and 4 .

6. Summary

A total number of 14 sediment cores taken fromwater depths between about 1000 and 20 m wereused to reconstruct the Holocene transgression of theLaptev Sea shelf. The chronology of the cores isbased on 119 radiocarbon ages, which were analyzedon marine biogenic calcite. Assuming that the deter-mined modern reservoir age of 370 years for theLaptev Sea remained constant back in time, theoldest radiocarbon sample dates back to 15.3 cal. ka,i.e., into the first phase of the last deglaciation.According to their water depths, the cores could beseparated into four groups, of which each is markedby a distinctive, time-transgressive reduction in aver-age sedimentation rates and changes in sedimentarycomponents. The steep decrease in sedimentationrates is the direct result of the postglacial sea-levelrise, which gradually diminished sedimentation fromthe outer to inner shelf due to an increasing distancebetween the shelf areas and the coast as the primarysediment source.


On the basis of the presented data, we concludethat the general pattern in downcore sedimentationrates reflects the southward retreat of the coastlineduring the Holocene flooding of the Laptev Seashelf. Allowing for some uncertainties, we estimatethat the inundation of the present 50-, 43-, and 31-misobaths was concluded by about 11.1, 9.8, and 8.9cal. ka. The Holocene sea-level highstand wasreached near 5 cal. ka. The rate of sea-level risebetween these time constraints was calculated to 5.4,13.3, and 7.9 mmryear. For the inundation of the50-m isobath, we have found a rather parallel sea-level rise compared with the North Atlantic andBeringia. Moreover, the relatively high rate calcu-lated between 9.8 and 8.9 cal. ka implicates a rapidsea-level rise towards the end of the deglaciation thatseems consistent with the existence of a secondmajor deglacial meltwater pulse in the North At-lantic.

Acknowledgements

We thank the scientists, crews, and captains of thevarious expeditions to the Laptev Sea. The study ispart of the Russian–German cooperative projectALaptev Sea System 2000B, financed by German and

Ž .Russian ministries BMBF and Ministry of Science ,Žand contributes to QUEEN Quaternary Environment

.of the Eurasian North , a programme of the Euro-pean Science Foundation. The manuscript greatlybenefited from review comments made by J.T. An-drews, L. Polyak, and C. Hjort. Data are available

Žthrough the information system PANGAEA http:rr.www.pangaea.de .

References

Alabyan, A.M., Chalov, R.S., Korotaev, V.N., Sidorchuk, A.Y.,Zaitsev, A.A., 1995. Natural and technogenic water and sedi-ment supply to the Laptev Sea. Reports on Polar Research176, 265–271.

Alekseev, M.N., 1997. Paleogeography and geochronology in theRussian eastern arctic during the second half of the quaternary.Quaternary International 41r42, 11–15.

Andersen, B.G., 1980. The deglaciation of Norway after 10,000BP. Boreas 9, 211–216.

Are, F.E., 1999. The role of coastal retreat for sedimentation inthe Laptev Sea. In: Kassens, H., Bauch, H.A., Dmitrenko, I.,Eicken, H., Hubberten, H.-W., Melles, M., Thiede, J., Timo-

Ž .khov, L. Eds. , Land–Ocean Systems in the Siberian Arctic:Dynamics and History. Springer, New York, pp. 287–295.

Barber, D.C., Dyke, A., Hillaire-Marcel, C., Jennings, A.E., An-drews, J.T., Kerwin, M.W., Bilodeau, G., McNeely, R.,Southon, J., Morehead, M.D., Gagnon, J.-M., 1999. Forcing ofthe cold event of 8200 years ago by catastrophic drainage ofLaurentide lakes. Nature 400, 344–348.

Bard, E., Hamelin, B., Fairbanks, R.G., Zindler, A., 1990. Cali-bration of the 14 C timescale over the past 30,000 years usingmass spectrometric U–Th ages from Barbados corals. Nature345, 405–410.

Bard, E., Hamelin, B., Arnold, M., Montaggioni, L., Cabioch, G.,Faure, G., Rougerie, F., 1996. Deglacial sea-level record fromTahiti corals and the timing of global meltwater discharge.Nature 382, 241–244.

Bauch, H.A., Erlenkeuser, H., Grootes, P.M., Jouzel, J., 1996.Implications of stratigraphic and paleoclimatic records of thelast interglaciation from the Nordic Seas. Quaternary Research46, 260–269.

Bauch, H.A., Kassens, H., Erlenkeuser, H., Grootes, P.M., Thiede,ŽJ., 1999. Depositional environment of the Laptev Sea Arctic

.Siberia during the Holocene. Boreas 28, 194–204.Bauch, H.A., Cremer, H., Kunz-Pirrung, M., 2000. Siberian shelf

sediments contain clues to paleoclimate forcing. EOS Transac-tions 81, 233 and 238.

Bauch, H.A., Kassens, H., Naidina, O.D., Kunz-Pirrung, M.,Thiede, J., 2001. Composition and flux of Holocene sedimentson the eastern Laptev Sea shelf, Arctic Siberia. QuaternaryResearch 55, 344–351.

Cauwet, G., Sidorov, I., 1996. The biogeochemistry of LenaRiver: organic and nutrients distribution. Marine Chemistry53, 211–227.

Dethleff, D., Loewe, P., Kleine, E., 1998. The Laptev Sea flawlead—detailed investigations on ice formation an export dur-ing 1991r1992 winter season. Cold Regions and ScienceTechnology 27, 225–243.

Dmitrenko, I., Golovin, P., Gribanov, V., Kassens, H., Holemann,¨J., 1998. Influence of the summer river runoff on ice formation

Ž .in the Kara and Laptev Seas. In: Shen, H.T. Ed. , Ice inSurface Waters. Balkema, Rotterdam, pp. 251–257.

Dmitrenko, I., Golovin, P., Gribanov, V., Kassens, H., 1999.Oceanographic causes for transarctic ice transport of riverdischarge. In: Kassens, H., Bauch, H.A., Dmitrenko, I., Eicken,H., Hubberten, H.-W., Melles, M., Thiede, J., Timokhov, L.Ž .Eds. , Land–Ocean Systems in the Siberian Arctic: Dynamicsand History. Springer, New York, pp. 73–92.

Elias, S.A., Short, S.K., Nelson, C.H., Birks, H.H., 1996. Life andtimes of the Bering land bridge. Nature 382, 60–63.

Elverhøi, A., Andersen, E.S., Dokken, T., Hebbeln, D., Spielha-gen, R., Svendsen, J.I., Sørflaten, M., Rørnes, A., Hald, M.,Forsberg, C., 1995. The growth and decay of the late Weich-selian ice sheet in western Svalbard and adjacent areas basedon provenance studies of marine sediments. Quaternary Re-search 44, 303–316.


Fahl, K., Stein, R., 1999. Biomarkers as organic-carbon-sourceand environmental indicators in the late Quaternary ArcticOcean. Marine Chemistry 63, 293–309.

Fairbanks, R.G., 1989. A 17,000-year glacio-eustatic sea levelrecord: influence of glacial melting rates on the YoungerDryas event and deep ocean circulation. Nature 342, 637–642.

Forman, S.L., Polyak, L., 1997. Radiocarbon content of pre-bombmarine mollusks and variations in the 14C reservoir age for thecoastal areas of the Barents and Kara Seas. GeophysicalResearch Letters 24, 885–888.

Forman, S., Lubinski, D., Miller, G., Snyder, J., Matishov, G.,Korsun, S., Myslivets, V., 1995. Postglacial emergence anddistribution of late Weichselian ice-sheet loads in the northernBarents and Kara Seas, Russia. Geology 23, 113–116.

Forman, S.L., Ingolfsson, O., Gataullin, V., Manley, W.F.,Lokrantz, H., 1999a. Late Quaternary stratigraphy of westernYamal Peninsula, Russia: new constraints on the configurationof the Eurasian ice sheet. Geology 27, 807–810.

Forman, S.L., Lubinski, D.J., Zeeberg, J.J., Polyak, L., Miller,G.H., Matishov, G.G., Tarasov, G., 1999b. Postglacial emer-gence and later Quaternary glaciation on northern NovayaZemlya. Boreas 28, 133–145.

Hald, M., Kolstad, V., Polyak, L., Forman, S.L., Herlihy, F.A.,Ivanov, G., Nescheretov, A., 1999. Late-glacial and Holocenepaleoceanography and sedimentary environments in the St.Anna Trough, Eurasian Arctic Ocean margin. Palaeogeogra-phy, Palaeoclimatology, Palaeoecology 146, 229–249.

Hanebuth, T., Stattegger, K., Grootes, P.M., 2000. Rapid floodingof the Sunda Shelf: a late-glacial sea-level record. Science288, 1033–1035.

Holmes, M.L., Creager, J.S., 1974. Holocene history of the LaptevŽ .Sea continental shelf. In: Herman, Y. Ed. , Arctic Ocean

Sediments, Microfauna, and Climatic Record in the LateCenocoic Time. Springer, Berlin, pp. 211–229.

´Ingolfsson, O., Hjort, C., 1999. The Antarctic contribution to´Holocene global sea level rise. Polar Research 18, 323–330.

Ivanov, I.I., Piskun, A.A., 1999. Distribution of river water andsuspended sediment loads in the deltas of rivers and basins ofthe Laptev and East Siberian Seas. In: Kassens, H., Bauch,H.A., Dmitrenko, I., Eicken, H., Hubberten, H.-W., Melles,

Ž .M., Thiede, J., Timokhov, L. Eds. , Land–Ocean Systems inthe Siberian Arctic: Dynamics and History. Springer-Verlag,New York, pp. 239–250.

Kassens, H., Dmitrenko, I., Rachold, V., Thiede, J., Timokhov,L., 1998. Russian and German scientists explore the arctic’sLaptev Sea and its climate system. EOS Transactions Ameri-can Geophysical Union 79, 317–323.

Kleiber, H.P., Niessen, F., 1999. Late Pleistocene paleoriverchannels on the Laptev Sea Shelf — implications from sub-bottom profiling. In: Kassens, H., Bauch, H.A., Dmitrenko, I.,Eicken, H., Hubberten, H.-W., Melles, M., Thiede, J., Timo-

Ž .khov, L. Eds. , Land–Ocean Systems in the Siberian Arctic:Dynamics and History. Springer-Verlag, New York, pp. 635–656.

Kuptsov, V.M., Lisitzin, A.P., 1996. Radiocarbon of Quaternaryalong shore and bottom deposits of the Lena and the LaptevSea sediments. Marine Chemistry 53, 301–311.

Lindemann, F., 1995. Sonographische und sedimentologische Un-tersuchungen in der Laptevsee, sibirische Arktis M.Sc. thesisŽ .unpublished , Kiel University 1-75.

Lubinski, D.J., Korsun, S., Polyak, L., Forman, S.L., Lehman,S.J., Herlihy, F.A., Miller, G.H., 1996. The last deglaciation ofthe Franz Victoria Trough. Boreas 25, 89–100.

Mason, O.K., Jordan, J.W., Plug, L., 1995. Late Holocene stormand sea-level history in the Chukchi Sea. In: Finkl, C.W.Ž .Ed. , Holocene Cycles: Climate, Sea Levels, and Sedimenta-tion, vol. 17. CERF, Charlotteville, VA, pp. 173–180, SpecialIssue.

Moller, P., Bolshiyanov, D.Y., Bergsten, H., 1999. Weichselian¨geology and palaeoenvironmental history of the central TaymyrPeninsula, Siberia, indicating no glaciation during the lastglobal glacial maximum. Boreas 28, 115–132.

Mueller-Lupp, T., Erlenkeuser, H., Bauch, H.A., Hefter, J.,Kassens, H., Thiede, J., 2000. Input of terrestrial organicmatter into the Laptev Sea during the Holocene — evidencefrom stable carbon isotopes. Journal of Earth Sciences 89,563–568.

Peltier, W., 1994. Ice age paleotopography. Science 265, 195–201.Peregovich, B., Hoops, E., Rachold, V., 1999. Sediment transport

Ž .to the Laptev Sea Siberian Arctic during the Holocene —evidence from the heavy mineral composition of fluvial andmarine sediments. Boreas 28, 205–214.

Polyak, L., Mikhailov, V., 1996. Post-glacial environments of thesoutheastern Barents Sea: foraminiferal evidence. In: An-

Ž .drews, J.T. Ed. , Late Quaternary Paleoceanography of theNorth Atlantic Margins. Geological Society Special Publica-tion, vol. 111, pp. 323–337.

Polyak, L., Sølheim, A. et al., 1994. Late- and postglacial envi-ronments in the northern Barents Sea west of Franz JosefLand. Polar Research 13, 197–207.

Polyak, L., Lehman, S.J., Gataullin, V., Jull, T.A.J., 1995. Two-step deglaciation of the southeastern Barents Sea. Geology 23,567–571.

Polyakova, Ye.I., Bauch, H.A., 1999. Late Holocene variations inthe conditions of fluvial runoff and sea-ice cover in the LaptevSea. Terra Nostra 99r11, 62.

Rachold, V., Grigoryev, M.N., 1999. Russian–German coopera-tion system Laptev Sea 2000: the Lena Delta 1998 expedition.Reports on Polar Research 315, 1–260.

Rachold, V., Hubberten, H.-W., 1999. Carbon isotope composi-tion of particulate organic material in east Siberian rivers. In:Kassens, H., Bauch, H.A., Dmitrenko, I., Eicken, H., Hub-

Ž .berten, H.-W., Melles, M., Thiede, J., Timokhov, L. Eds. ,The Laptev Sea System: Dynamics and History. Springer,New York, pp. 223–238.

Romanovskii, N.N., Gavrilov, A.V., Hubberten, H.-W., Kholodov,A.L., Kassens, H., 1997. Reconstruction of paleogeographicconditions on the Laptev Sea shelf for the late Pleistocene–

ŽHolocene glacioeustatic cycle. Kryosphera Zemli 1, 42–49 in.Russian .

Sirenko, B.I., Petryashov, V.V., Rachor, E., Hinz, K., 1995.Bottom biocoenoses of the Laptev Sea and adjacent areas.Reports on Polar Research 176, 211–221.

Stein, R., Fahl, K., 2000. Holocene accumulation of organic


Ž .carbon at the Laptev Sea continental margin Arctic Ocean :sources, pathways, and sinks. Geo-Marine Letters 20, 27–36.

Stuiver, M., Reimer, P.J., Bard, E., Beck, J.W., Burr, G.S.,Hughen, K.A., Cromer, B., McCormic, G., van der Plicht, J.,Spurk, M., 1998. INTCAL 98 radiocarbon age calibration,24,000–0 cal BP. Radiocarbon 40, 1041–1083.

Svendsen, J.I., Astakhov, V.I., Bolshiyanov, D.Y., Demidov, I.,Dowdeswell, J.A., Gataullin, V., Hjort, C., Hubberten, H.W.,Larsen, E., Mangerud, J., Melles, M., Moller, P., Sarnistoo,¨M., Siegert, M.J., 1999. Maximum extent of the Eurasian icesheets in the Barents and Kara Sea region during the Weich-selian. Boreas 28, 234–242.

Chronology of the Holocene transgression at the North Siberian

Documents

Transcript of Chronology of the Holocene transgression at the North Siberian