Postglacial Paleoceanographic Environments in the Barents and Baltic Seas

Abstract

This paper presents reconstructions of ice sheet boundaries, lacustrine and marine paleobasins, as well as the connections of the Barents and Baltic seas with the North Atlantic from the Last Glacial Maximum to the Holocene. The reconstructions are based on original and published data obtained from the northern and western parts of the Barents Sea and Baltic depressions with account for the available regional schematic maps of deglaciation. The early deglaciation of the Scandinavian-Barents ice sheet culminated with the Bolling-Allerod interstadial (14.5-12.9 cal ka BP), which was characterized by a more vigorous Atlantic meridional overturning circulation (AMOC) and a corresponding increase in surface Atlantic water inflow into the Barents Sea through deep troughs. The Baltic Ice Lake (BIL) remained a dammed-up isolated basin during deglaciation from 16.0 to 11.7 cal ka BP. In the Younger Dryas (YD), the lake drained into the North Sea and was replaced by a brackish Yoldia Sea (YS) at the beginning of the Holocene (Preboreal, 11.7-10.7 cal ka BP), due to a limited connection between two basins through the Narke Strait. In the Barents Sea, the next increase in the Atlantic water influx into the deep basins corresponded to terminal YD and Preboreal events with a culmination in the Early Holocene. The Yoldia Sea became a lake again during the next stage, the Ancylus (similar to 10.7-8.8 cal ka BP). Atlantic water inflow both into the Barents and Baltic seas varied during the Holocene, with a maximum contribution in the Early Holocene, when the Littorina Sea (LS, 8-4 cal ka BP) connection with the North Sea via the Danish Straits was formed to replace the Ancylus Lake. The recent, post-Littorina stage (PS, the last 4 cal ka) of the Baltic Sea evolution began in the Late Holocene.

Full text

118
ISSN 0001-4370, Oceanology, 2016, Vol. 56, No. 1, pp. 118–130. © Pleiades Publishing, Inc., 2016.
Original Russian Text © E.V. Ivanova, I.O. Murdmaa, E.M. Emelyanov, E.A. Seitkalieva, E.P. Radionova, G.N. Alekhina, S.M. Sloistov, 2016, published in Okeanologiya, 2016,
Vol. 56, No. 1, pp. 125–138.
Postglacial Paleoceanographic Environments
in the Barents and Baltic Seas
E. V. Ivanovaa, I. O. Murdmaaa, E. M. Emelyanovb, E. A. Seitkalievaa, c,
E. P. Radionovad, G. N. Alekhinaa, and S. M. Sloistova
a Shirshov Institute of Oceanology, Russian Academy of Sciences, Moscow, Russia
b Atlantic Department, Institute of Oceanology, Russian Academy of Sciences, Kaliningrad, Russia
c Department of Geology, Moscow State University, Moscow, Russia
d Geological Institute, Russian Academy of Sciences, Moscow, Russia
e-mail: e_v_ivanova@ocean.ru
Received October 29, 2014; in final form, January 21, 2015
Abstract⎯This paper presents reconstructions of ice sheet boundaries, lacustrine and marine paleobasins, as
well as the connections of the Barents and Baltic seas with the North Atlantic from the Last Glacial Maxi-
mum to the Holocene. The reconstructions are based on original and published data obtained from the
northern and western parts of the Barents Sea and Baltic depressions with account for the available regional
schematic maps of deglaciation. The early deglaciation of the Scandinavian–Barents ice sheet culminated
with the Bølling-Allerød interstadial (14.5–12.9 cal ka BP), which was characterized by a more vigorous
Atlantic meridional overturning circulation (AMOC) and a corresponding increase in surface Atlantic water
inflow into the Barents Sea through deep troughs. The Baltic Ice Lake (BIL) remained a dammed-up isolated
basin during deglaciation from 16.0 to 11.7 cal ka BP. In the Younger Dryas (YD), the lake drained into the
North Sea and was replaced by a brackish Yoldia Sea (YS) at the beginning of the Holocene (Preboreal, 11.7–
10.7 cal ka BP), due to a limited connection between two basins through the Närke Strait. In the Barents Sea,
the next increase in the Atlantic water influx into the deep basins corresponded to terminal YD and Preboreal
events with a culmination in the Early Holocene. The Yoldia Sea became a lake again during the next stage,
the Ancylus (~10.7–8.8 cal ka BP). Atlantic water inflow both into the Barents and Baltic seas varied during
the Holocene, with a maximum contribution in the Early Holocene, when the Littorina Sea (LS, 8–4 cal ka BP)
connection with the North Sea via the Danish Straits was formed to replace the Ancylus Lake. The recent,
post-Littorina stage (PS, the last 4 cal ka) of the Baltic Sea evolution began in the Late Holocene.
DOI: 10.1134/S0001437016010057
INTRODUCTION
The evolution of the Baltic and Barents seas during
global postglacial warming (~18 to 9 cal ka BP) was
controlled by the decay and melting of the Scandina-
vian–Barents ice sheet [21, 28, 30, 33, 39, 55, 58], the
irregular glacio-isostatic rebound of different parts of
the region after removal of the ice load [1, 10, 14, 19,
20, 24, 29, 32], and the connection with the North
Atlantic [14, 17, 24, 32, 43, 45]. Inasmuch as the Bar-
ents Sea, located in high latitudes, is characterized by
a weak meridional heat transfer from the North Atlan-
tic via the Norwegian Sea and the high albedo of the
ice sheet, its deglaciation commenced later than that
of the continent south of the Baltic Sea. The postgla-
cial history of the Baltic and Barents seas includes
deglaciation (16.0–11.7 cal ka BP) with millennial
oscillations followed by intervals of the formation
(11.7–9.0 cal ka BP) and evolution (the last 9 cal ka BP)
of marine environments in the Holocene. This com-
plex process was responsible for significant variations
in paleoceanographic parameters, such as sea level
and bioproductivity, temperature, and salinity of sur-
face and bottom waters, which are reflected in the suc-
cession of planktonic and benthic microfossils and
changes in different lithological and geochemical
properties of bottom sediments.
In this work, we reviewed paleoceanographic envi-
ronments in the Baltic and Barents seas during the
deglaciation of the Scandinavian–Barents ice sheet
and the Holocene with consideration of the factors,
determining the different responses of regional cli-
matic processes in these two seas. Reconstructions are
based on new original data as well as available pub-
lished data. All the sediment cores used in this study
are dated by geochronological methods. A set of accel-
erator mass spectrometry radiocarbon (AMS-14C)
dates provides reliable interregional correlations. Par-
ticular attention is paid to changes in the boundaries of
the ice sheet during its decay and melting, reconstruc-
tions of intensity of Atlantic water inflow into the Bar-
MARINE GEOLOGY

OCEANOLOGY Vol. 56 No. 1 2016
PO STGL ACIAL PALEOCEANOGRAPHIC ENVIRONMENTS 119
ents Sea, and the water exchange of the Baltic Sea with
the Atlantic via the North Sea. The presented recon-
structions illustrating the configuration of basins allow
the development of deglaciation in the region to be
compared with variations in vigour of the Atlantic
meridional overturning cell of the global thermohaline
circulation that determined the influx of Atlantic
water into the Barents Sea.
PRESENT-DAY OCEANOGRAPHIC
ENVIRONMENTS
The Barents Sea is a shelf marginal basin with sev-
eral troughs up to 650 m deep, located in the path of
relatively warm and saline surface and subsurface
Atlantic waters to the Arctic Ocean via the Norwegian
Sea (Fig. 1). The main flow of Atlantic waters enters
the sea from the west, where they practically fill the
entire Bear Island trough from its surface to the bot-
tom. Subsurface Atlantic waters flowing via the Fram
Strait along the continental slope of Eurasia penetrate
into the sea from the north through the Kvitøya–Erik
Eriksen, Franz Victoria, and Svyataya Anna–Sedova
troughs at depths of 150–200 m. Surface circulation is
to a large extent determined by the interaction between
Atlantic waters and cold fresh Arctic waters flowing
into the sea from the north and east [31].
The mostly shallow (depths up to 300 m) intracon-
tinental Baltic Sea is connected with the North Atlan-
tic via the Danish Straits and the North Sea. Its surface
and deep circulation is in general anticyclonic (Fig. 1).
The renewal of deep-water masses in basins of th e Bal-
tic Sea occured during episodic ingressions of large
volumes of oxygen-enriched saline waters from the
North Sea [4, 36].
REGIONAL CHRONOSTRATIGRAPHY
When defining and correlating postglacial events
and ages of basin developments, the main problem
consists in the scarcity of geochronological dates and
the spatial–temporal uncertainty of changes in the
reservoir effect, which prevents calibration of mea-
sured radiocarbon ages (e. g., [24, 34, 51]). Neverthe-
less, we have correlated regional scales, considering
the currently available concepts of the absolute age of
stratigraphic units, climatic events (stadials and inter-
stadials during the postglacial history), and stages of
the development of a postglacial basin that preceded
the present-day Baltic Sea (Fig. 2).
In this work, we used an original previously devel-
oped chronostratigraphic scale of postglacial sedi-
ments of the Barents Sea substantiated by AMS-14C
dates [6, 24, 25, 39]. The latter is correlated with the
standard chronostratigraphy available for the North-
ern Hemisphere (e.g., [25, 28, 44, 45]) and Gudelis’
scale [3] developed for the Baltic Sea; the ages of
stages in development of the basin are specified by
subsequent investigations [2, 10, 13, 14, 19, 27, 60].
The chronostratigraphy of the Northern Hemisphere
is in fact a climatostratigraphy, since it reflects millen-
nial-scale warming and cooling episodes during the
degradation of continental ice sheets, which are read-
ily recognizable in oxygen isotope record of the
Greenland ice core (NGRIP) [40]. The process of ice
sheet decay and melting determined glacio-eustatic
sea level rise (e.g., [61]) and the level of the dammed-
up Baltic Ice Lake (BIL, [5, 13]).
MATERIALS, INDICATORS, AND METHODS
OF PALEORECONSTRUCTION
The paleoreconstructions presented in this work
are based on original (new and published) and avail-
able (in the literature) data on sediment cores obtained
in the Barents and Baltic seas and Fram Strait
(Table 1, Fig. 1). In addition to these data, available
reconstructions illustrating changes in the ice sheet
extent at different developmental stages [21, 28, 30–
32, 35, 41, 42, 54] and basins that existed formerly in
the place of the present-day Baltic Sea [1, 7–11, 13,
14, 19, 59, 60, 62, 63], and results of deglaciation
modeling [29, 55] were used. All the cores used and
investigated are dated by the AMS-14C; their age scales
are taken without any changes from corresponding
original works. Reconstructions of the intensity of
Atlantic water inflow into the Barents Sea are based on
micropaleontological and isotopic–geochemical indi-
cators such as, for example, estimates of paleotem-
peratures in the surface water layer (~0–100 m)
derived from alkenones and assemblages of planktonic
foraminifers using analogue methods or transfer func-
tions for cores PSh-5159R [6], PSh-5159N [50, 51],
MD95-2011 [15, 49, 51], and M 23258 [51, 53] (Fig. 1,
Table 1). The increase in the relative content of ben-
thic foraminifers Cassidulina teretis, Pullenia spp., Tri-
farina angulosa, and others, which are considered as
serving as indicators of Atlantic water on the Arctic
shelf [6, 24, 32], implies the distribution of these
waters in the bottom layer of shelf depressions,
troughs, and fjords of the Barents Sea at depths up to
650 m. The presence of marine and glaciomarine sed-
iments implies the absence of glaciers at sampling
sites, and the high content of gravel in sediments indi-
cates iceberg rafting.
New data obtained from Core S-2519 were taken in
the northwestern part of the sea; new petrographic and
lithological data are yielded by previously investigated
cores ASV-880 from the Franz Victoria Trough and
PSh-5159R from the Ingøydjupet Depression, respec-
tively (Tables 1, 2, Fig. 1).
Core S-2519, retrived in the Erik Eriksen Trough
during Cruise 25 of the R/V Akademik Nikolaj Stra-
khov in 2007, was sampled entirely every 3 cm. The
samples were kept in a refrigerator of Institute of
Oceanology at 4°C over several months and were then
weighed and subjected to the standard water–sieve
treatment with the extraction of fractions 0.05–

120
OCEANOLOGY Vol. 56 No. 1 2016
IVANOVA et al.
Fig. 1. Schematic circulation and location of sediment cores. Circulation: (1) surface, (2) subsurface (for the Baltic Sea, after [4]);
cores: (3) studied in this work, (4) studied with the participation of the authors of this work and published previously [6, 17, 24,
25, 39, 50, 51], (5) studied by other researchers [2, 32, 44–46, 49, 53, 56, 57]. (BS) Baltic Sea; (EET) Erik Eriksen Trough;
(DS) Danish Straits. Currents; (NC) Norwegian, (WSC) West Spitsbergen.
PL94-29
PL94-07
PL-94-67
Svyataya Anna Trough
Sedova
Trough
Kvitøya
Trough
Bear Island
Trough
Fram
Strait
Franz
Victoria
Trough
Franz Josef
Land
JPC5
NP94-51
NP05-21
JM10-10
JM02-440
JM02-460
JM03-373
WSC
M23258
MD95-2011
POS 303700
NC
PSh-5159
S-2519
ASV-880
EET
Svalbard
Novaya Zemlya I.
NORWEGIAN
SEA
BARENTS
SEA
NORTH
SEA DS
BS
12345
Scandinavia
10°40°70°E
N
80°
60°
70°
0.1 and >1 mm. Several samples were used for measur-
ing sediment water content. The obtained fractions
were dried, weighed, and used to calculate their rela-
tive contents. For a study of coarse-grained ice-rafted
material, a >0.1 mm fraction was sieved through a
2 mm-mesh sieve. Each >0.1-mm fraction was exam-
ined under a binocular to identify benthic foraminifers
from аn aliquot containing 200–300 specimens (if
available) and the relative abundance of the common
atlantic-affiliated species Cassidlina teretis was deter-
mined. Tests of benthic foraminifers from layers with
a high abundance, which were subjected to AMS-14C

OCEANOLOGY Vol. 56 No. 1 2016
POSTGLACIAL PALEOCEANOGRAPHIC ENVIRONMENTS 121
Fig. 2. Correlation of climatostratigraphic scales for the Barents [25] and Baltic [3] seas and the Northern Hemisphere [24] with
the curve of sea level fluctuations [61] and climatic curve of variations in the oxygen isotope composition in the Greenland Ice
Core NGRIP [40] for the last 20 ka. Dotted lines designate hypothetical level of the Baltic Ice Lake (BIL), lower boundary and
correlation of this interval with synchronous intervals in other basins. (LGM) Last Glacial Maximum; (ED) early deglaciation;
(B–A) Bølling–Allerød; (YD) Younger Dryas; (YS) Yoldia Sea; (AL) Ancylus Lake; (LS) Littorina Sea; (PLS) post-Littorina Sea.
−120 −80 −40 0
0
2
4
6
8
10
12
14
16
18
20
Sea level, m
−45 −41 −37 −33
NGRIP, δ18 O, ‰
Age, cal ka BP
Age, cal ka BP
Age, cal ka BP
Ocean
BIL?
YD
B–A
ED
LGM
Deglaciation
Glacier
Unit I
Holocene
Unit
III
Unit
II
Unit
IV
BIL
YS
AL
LS
PLS
4
8
10.7
11. 7
11.7
12.9
14.8
18
Stratigraphic scale
of the Northern
Hemisphere
Barents
Sea
Baltic
Sea
measurements in the Poznan Radiocarbon Labora-
tory of Poland, yielded radiocarbon dates (Table 2).
They were converted into calendar ages using the
Calib 6.0.2 program [47] and marine calibration curve
[23], accounting for local reservoir effect of 71 ± 21
years [34]. The time scale for this core was obtained by
linear interpolation between dated levels, with an
assumption of a zero age for the cortop (0 cm) and
extrapolation below the oldest date.
Rock types were determined visually and under a
binocular microscope in all the clasts constituting a
>2-mm fraction in five samples from each of the cores
S-2519 and ASV-880. Figure 3 presents the relative
contents of rock types in the fraction in the form of cir-
cular diagrams.
For the Baltic Sea, reference cores taken from the
Bornholm, Gdansk, West Gotland, and North Baltic
basins served as a basis for the previously published,
available reconstructions used in this work [1, 2, 19,
60, 63]. These cores, examined by different research-
ers, are dated by the 14C [1, 2, 5, 19] and 210Pb (in the
upper part, [2]) methods, which made it possible to
define the intervals in sedimentary sections corre-
sponding to lacustrine and marine stages of basin evo-
lution discussed in publications. For the Gdansk
Basin, Core POS 303700 was used to calculate
changes in the salinity of bottom waters based on Br
contents in bottom sediments [2].
For the comparison of reconstructed changes in
the intensity of Atlantic water flow into the Barents
Sea and rate of thermohaline circulation in the Atlan-
tic, we used a record illustrating the distribution of the
231Pa/230Th values in postglacial sediments of Core
OCE326–GGC5 (33°42 N, 57°35 W) taken at depth
of 4550 m on the Bermuda Rise (Fig. 4) [37]. This
record provides satisfactory time resolution (centen-
nial-scale) for the deglaciation interval and is widely
used as adequately reflecting postglacial circulation
intensity in the Atlantic, although it is substantially
less reliable over the Holocene.

122
OCEANOLOGY Vol. 56 No. 1 2016
IVANOVA et al.
PROPERTIES OF POSTGLACIAL SEDIMENTS
IN TROUGHS SERVING AS PATHS
FOR THE ATLANTIC WATER INFLOW
INTO THE BARENTS SEA
The best-investigated and dated cores S-2519 and
ASV-880, from the northern troughs through which
subsurface Atlantic water penetrates into the Barents
Sea, and Core PSh-5159R from the Ingøydjupet
Depression characterizing the Bear Island Through,
the passage of surface Atlantic waters, were selected as
reference sections in this work.
Core S-2519 recovered a relatively complete sec-
tion of postglacial sediments typical of the Barents Sea
[25, 39], not older than 17.3 ka. The core is subdivided
into three lithostratigraphic units with distinct bound-
aries between them (Fig. 3). Unit I (0–225 cm, Holo-
cene) is represented by greenish gray pelitic mud with
hydrotroillite. Unit II (225–273 cm, late deglaciation
phase) is composed of pelitic mud with layering in
color shades (lamination?): alternating dark gray and
brown laminae. The unit encloses a thin intercalation
(at 265 cm) enriched with coarse aleurite. Unit III
(273–355 cm, early deglaciation phase) consists of
glaciomarine diamicton: sandy–silty pelitic mud with
abundant gruss and coarse gravel. Diverse rock frag-
ments (>2 mm across) are dominated by dark gray (to
black) mudstones and light-colored limestones with
minor chert, granite, and rare quartzite, fine-grained
sandstone, micaceous schist, and gneiss (Fig. 3). This
diverse composition of rock clasts, their poor sorting,
and their variable roundness indicate the iceberg raft-
ing of coarse detrital material of diamicton trans-
ported by glaciers from the Svalbard Archipelago.
Intense iceberg rafting in the core area ceased only in
the Younger Dryas [26].
The similar section of Core ASV-880 from the
Franz Victoria Trough [17, 25, 39] differs from that
Table 1. Coordinates of stations mentioned in the text
Station Latitude, N Longitude, E Depth, m Source
POS 303700 54°49.22´ 19°11.0 7 ´ 10 5 .4 2
ASV-880 79°55.50´ 47°08.20´ 388 17, 25, 39, this work
S-2519 79°30.75´ 28°41.68´ 347 t his work
PSh-5159R 71°21.65´ 22°38.81´ 418 6, 24
PSh-5159N 71°21.80´ 22°38.77´ 422 50, 51
NP05-21 79°03´ 11°05.40´ 327 45
NP94-51 80°21.41´ 16°17.94´ 399 56
JM02-440 77°22´ 12°48´ 240 57
JM10-10 77°24.8´ 20°06´ 123 46
JPC5/PG5
PL94-07
PL94-29
PL94-67
81°07.1´
80°59.69´
79°59.59´
78°19.54´
43°25.9´
67°32.77´
69°56.96´
70°02.07´
463
633
605
443
32
JM02-460
JM03-373
76°03´
76°24´
16°00´
12°58´
389
1485
44
MD95-2011 66°58.19´ 7°38.36´ 1048 49, 51
M23258 75°14°1768 51, 53
Table 2. AMS radiocarbon dates obtained for Core S-2519
(Poz) Poznan Radiocarbon Laboratory (Poland).
Laboratory code Depth, cm Dated material Age 14C,
years BP ∆RCalibrated age
(±1σ), years BP
Calendar age,
years BP (relative
to 1950)
Poz-33367 155.5 Benthic
foraminifers 6390 ± 40 71 ± 21 6720–6845 6780
Poz-33358 250.5 Benthic
foraminifers 11040 ± 60 71 ± 21 12387–12482 12470

OCEANOLOGY Vol. 56 No. 1 2016
POSTGLACIAL PALEOCEANOGRAPHIC ENVIRONMENTS 123
above in thickness of lithostratigraphic units as well as
in recovery of Unit IV at its base, which possibly rep-
resents a bottom moraine deposit of the last glaciation
(Fig. 3). Holocene Unit I (0–310 cm), composed of
olive–gray pelitic mud is characterized by relatively
abundant foraminiferal tests, relatively high Corg con-
tent, and the presence of chitin Polychaeta tubes
pointing to increased biological productivity. Unit II
(310–404 cm) includes thin lenses of coarse silt.
Unit III (404–494 cm) is formed by diamicton with
coarse detrital material dominated by black (carbona-
ceous) mudstones similar to Lower Jurassic carbona-
ceous mudstones of the Franz Josef Land islands.
Detrital material is represented by abundant sandstone
and chert fragments accompanied by granitoids and
some other rock varieties. In Unit IV (494–508 cm), the
entire fraction > 2 mm consists of black mudstone
clasts (Fig. 3).
The section of Core PSh-5159R from the Ingøyd-
jupet Depression (Fig. 3), which recovered sediments
of the early deglaciation phase, is characterized by the
absence of the glaciomarine diamicton typical of
coeval sediments in the Barents Sea. Pre-Holocene
silty–pelitic mud contains up to 10–15% of fractions
>0.05 mm (from coarse silt to small gravel), forming
autonomous thin intercalations and lenses, which
indicate activity of suspension flows from melting gla-
ciers, not ice rafting. In addition, unlike in other areas
of the Barents Sea, deglaciation sediments of the
Ingøydjupet Depression contain a relatively diverse
assemblage of foraminifers, which implies an
increased bioproductivity of its surface waters [6].
The lower parts of core sections from northern
troughs (below 305 and 270 cm in Cores ASV-880 and
S-2519, respectively) contain only singe, most likely
reworked, tests of benthic and planktonic foraminifers
at some levels. Nevertheless, their assemblage in sedi-
Fig. 3. Lithology, radiocarbon age, and distribution of coarse grain-size fractions in sediment cores PSh-5159 from the Ingøyd-
jupet Depression (after [24] modified), S-2519 from the Erik Eriksen Trough (this work), and ASV-880 from the Franz Victoria
Trough (after [39], modified).For selected intervals in Cores S-2519 and ASV-880, circular diagrams of the coarse-grained mate-
rial (>2 mm ) are provided to show the approximate estimate of major rock type proportions. Units (after [24, 39], modified):
(I) Holocene, (II) late deglaciation, (III and III*) early deglaciation in the southern and northern parts of the basin, respectively,
(IV) Last Glacial Maximum.
Core depth, cm
0
010
0.05−0.1 mm
>0.1 mm
>2 mm
1—mudstone
2—granite
3—sandstone
4—chert
5—limestone
6—others
%
50
100
150
200
250
300
350
400
450
500
635 ± 30
14С, years
12150 ± 70
13550 ± 70
10010 ± 50
Core PSh-5159
Core depth, cm
0
0 10203040%
50
100
150
200
250
300
350
14С, years
110 40 ± 60
6390 ± 40
273 cm 410 cm 425 cm
480 cm445 cm
495 cm
299 cm
336 cm315 cm
348 cm
Core S-2519
Core depth, cm
0
0102030%
50
100
150
200
250
300
350
400
450
500
1550 ± 60
2440 ± 60
14С, years
7420 ± 80
9150 ± 90
8830 ± 90
6560 ± 80
Core ASV-880
I II III III* IV
123456

124
OCEANOLOGY Vol. 56 No. 1 2016
IVANOVA et al.
Fig. 4. Correlation of postglacial events in the North Atlantic, Baltic, and Barents seas. Marine (1) and lacustrine (2) stages of the
Baltic Sea are defined based on variations in salinity of interstitial water calculated from measured Br concentrations in sediments
of Core POS 303700 from the Gdansk Basin, taking into account radiocarbon dates (after [2], modified]. In the Barents Sea
cores, variations in the relative abundance of the benthic foraminiferal species Cassidulina teretis illustrate the intensity of the
Atlantic bottom water inflow (arrow below indicates its intensification) from the west (Core PSh-5159R) (after [6], modified,
(?) designates probable inaccuracy in age model) and north (cores S-2519 (this work) and JPC/PG5 [32]). For the North Atlan-
tic, intensity of the thermohaline circulation system is reconstructed on the basis of variations in the 231Pa/230Th values calculated
by two methods for Core OCE326-GGC5; horizontal scale is inverted [37]; intensification is shown below by a doubled arrow.
(3) lower boundary of the Holocene; (4) correlation of synchronous deglaciation intervals.
Age, cal ka BP
YD
Stratigraphy
of the Northern
Hemisphere
B–A
ED
LGM
Deglaciation
Holocene
11.7
12.9
14.8
18
4710 ± 130
14
С, years
Baltic Sea
Core POS 303700
00
2
2
1
4
3
4
6
8
10
12
14
16
18
20
?
20 40 60 80
481216
Salinity, ‰ Core PSh-5159R
C. teretis, %
0
2
4
6
8
10
12
14
16
18
20
0.080.10 0.06 0.04
Core OCE326-GGC5
231Pa/230Th
0
2
4
6
8
10
12
14
16
18
20 40 60
Core JPC5/PG5
C. teretis, %
0
2
4
6
8
10
12
14
16
20 40 60
Core S-2519
C. teretis, %
LSALYSBIL PLS
88−89
169−17 0
251−252
331−332
411−412
493−494
573−574
655−656
733−734
813−814
901−902
989−990
1069 −1070
114 9−115 0
1229 −1230
Core depth, cm
5080 ± 130
5750 ± 120
6450 ± 150
7110 ± 200
8520 ± 130
11800 ± 500
Barents Sea North Atlantic
ments corresponding to 0–13.3 and 0–9.8 cal ka BP in
Cores S-2519 and ASV-880, respectively, is diverse,
with a highly variable degree of preservation and abun-
dance. Core S-2519 demonstrates several peaks of a
high (>20%) share of the benthic species Cassidulina
teretis: 0.8–1.8, 4.8–5.5 and 7.7–7.9 cal ka BP as well as
peaks of its maximum content: 9.6 and 12.5–13.0 cal ka
BP (39 and 47–51%, respectively) (Fig. 4). These inter-
vals also exhibit elevated concentrations of accessory
species connected with Atlantic waters [6, 24], such as
Trifarina angulosa and Pullenia spp. In the Holocene
sediments of Core ASV-880, the share of species indi-
cating the influence of the Atlantic water mass is very
low (usually <10%, [17]); therefore, Figure 3 also pres-
ents data on Core JPC5/PG5 [32]. This core was taken
closer to the main midstream of the Atlantic water mass
flow and the foraminiferal assemblage contains indicat-
ing species, primarily С. teretis, almost throughout the
whole postglacial section (Figs. 1, 3).
PALEOCEANOGRAPHIC EVOLUTION
OF BASINS AND THEIR CONNECTION
WITH THE NORTH ATLANTIC
Six time spans with climatically contrasting climate
conditions were selected for reconstructing the pale-
oceanographic evolution of the region under consid-
eration during the period lasting from the Last Glacial
Maximum to Early Holocene. These time spans are
best characterized by seismostratigraphic, micropale-
ontological, and lithological data for the entire Bar-
ents Sea (Figs. 1, 5, Table 1) and schematic paleocean-
ographic reconstructions available for the predeces-
sors of the Baltic Sea [7–10, 59, 60]. The penetration
of Atlantic waters via the Fram Strait onto the conti-
nental slope of Eurasia and into the Barents Sea is
reconstructed using cores (designated by rhombs in
Fig. 5), where this “signal” is the best ref lected either
in the increase of surface water temperatures or in the
elevated share of indicating species of benthic fora-
minifers (for the bottom layer). Despite diachronous
development of the different parts of each basin estab-
lished by previous researchers (e.g., [1, 25]), several
common events may be defined and correlated.
We share the standpoint of participants of the
international QUEEN project [33, 58], who believe
that during the Last Glacial Maximum (LGM) from 22
to 18 cal ka BP, the entire Barents Sea–Baltic region
was covered by an ice sheet up to 1–2 km thick
(Fig. 5a), which extended also onto the northern part
of Western Europe and the Russian Plain. Neverthe-
less, at that time, warm surface Atlantic waters pene-

OCEANOLOGY Vol. 56 No. 1 2016
POSTGLACIAL PALEOCEANOGRAPHIC ENVIRONMENTS 125
trated northward up to Spitsbergen during the summer
seasons (Fig. 5a) [38, 54]. According to data obtained
in the Vistula valley [62] and southern Lithuania [48],
the ice sheet reached its maximum extent in the area
southeast of the Baltic Sea only 18.3–18.4 cal ka BP
due to advancement of glaciers from Scandinavia [27].
According to the model in [55], ice-sheet volume
increased in the period from the LGM to ~16 cal ka
BP and ice streams moved from the center of the sheet
toward its periphery (e.g. [41, 42]).
The onset of deglaciation in the region under con-
sideration occurred 2–4 ka later than in the Western
Europe. Approximately at 16 cal ka BP, an excess of
critical stability mass resulted in the rapid decay of the
ice sheet due to the development of ice streams and
icebergs calving along the edge of the continental
slope [55]. The degradation and melting of the ice
sheet was also stimulated by geothermal heating from
below and the inflow of warm subsurface Atlantic
water beneath the floating edge of the sheet, which
favored the formation of ice streams and icebergs. In
opinion of some authors, a “f lotilla of icebergs” calved
from the ice streams flowed into the Norwegian Sea
along the Barents Sea ice sheet periphery penetrating
into the North Atlantic [52]. The release of fresh water
from icebergs was responsible for a significant
decrease in salinity and a sharp decline of convection
or perhaps stagnation in the North Atlantic (Fig. 4).
This occurred during the so-called Heinrich event H1
(~17–15 cal ka BP, Fig. 4) [24, 37, 38, 52], which cor-
responded to the Oldest Dryas cooling in Europe
(Fig. 2). At the same time, the relatively high abun-
dance of C. teretis at the end of this event in
Cores NP94-51, JPC 5, PL94-29, and JM02-460 [32,
44, 56] indicates the northward penetration of a signif-
icant volume of Atlantic waters along the edge of the
decaying ice sheet, at least to the Svyataya Anna
Trough (Fig. 5b). Inasmuch as the surface layer in the
Fram Strait, over the northern troughs of the Barents
Sea and, likely, in the Norwegian Sea was represented
by cold and fresh water with icebergs, while thermoha-
line circulation was weak (Fig. 4), the inflow of warm
saline water at depths of 400–600 m (Fig. 5b) requires
additional investigation.
The global warming and sea level rise led to retreat of
the ice sheet margines from the deep troughs in the
western and northern Barents Sea (Fig. 5b), primarily
the Bear Island Trough. Ice streams promoted to the
release of shelf depressions and formation of small
marine basins freshened due to significant meltwater
inflow [39]. Glaciomarine diamicton with a high con-
tents of coarse detrital material accumulated in depres-
sions (Figs. 2, 3). Available dates of diamicton in differ-
ent parts of the Barents Sea indicate a diachronous
development of deglaciation [25, 39]. Judging from the
absence of diamicton in Core PSh-5159R and the
radiocarbon dates in this core, the Ingøydjupet Basin
became free of ice prior to 16 cal ka BP (Fig. 3) [6].
By about 16 cal ka BP, the retreating glacier
released a narrow band in the southwestern part of the
Baltic Sea (Fig. 5b), which became filled with meltwa-
ter. This marked the onset of the development stage of
the dammed Baltic Ice Lake (BIL), isolated from the
World Ocean. This stage in the development of the
Baltic Sea embraces practically the entire deglaciation
period from ~16 to 11.7 cal ka BP and ref lects the
oscillations of the front of the ice sheet, which sup-
pressed biological productivity and was accompanied
by accumulation of lacustrine sediments depleted in
organic matter. This basin extended from the Born-
holm Basin to Lake Onega. In the opinion of some
authors, the Baltic and White seas were separated from
each other by a land bridge and later on developed
according to their own patterns. However, other
researchers believe that the BIL discharge into the
Arctic Basin via the White Sea continued until the late
Allerød [1]. During most of this stage in development
of the basin, the BIL level was higher than that of the
World Ocean (Fig. 2) [1, 14]. According to recon-
structions by Blazhchishin [1], during the Old Dryas
the BIL was covered by a floating shelf glacier, later
replaced by drifting icebergs.
During the warm Bølling–Allerød (B–A) inter-
stadial, approximately 14.8–12.9 cal ka BP, most of
the Baltic Sea became free of ice, which was pre-
served only on archipelagoes and shoals (Fig. 5c) [18,
28, 30]. Unfortunately, the data on this period are
insufficient for the adequate areal reconstruction of
basins (Fig. 5c). The elevated contents of C. teretis,
Pullenia spp., and other characteristic species of ben-
thic foraminifers in the Bølling–Allerød sediments
indicates the strengthened influx (or temperature rise)
of Atlantic waters into the Fram Strait, on the conti-
nental slope, into Spitsbergen fjords, southwestern
part of the Barents Sea, and its northern troughs
(Fig. 4) [6, 32]. Atlantic water filled the Ingøydjupet
Depression up to its bottom (Station PSh-5159), and
northern troughs during the B–A episode and pene-
trated into them at the end of the Younger Dryas
(YD, ~12.9–11.7 cal ka BP) as well (Fig. 5d) [6, 26,
32]. The absence of the distinct bottom Atlantic water
“signal” in the cores from the Svyataya Anna Trough
at ~14 cal ka BP may be explained either by the intense
sea ice and local bottom water formation or the insuf-
ficient time resolution of the available data [32]. This
explanation is most likely appropriate also for later
intervals (~12.0–11.7 and ~10.5–10.3 cal ka BP) and,
probably, for cores taken near Svalbard.
During the YD interval, the sea surface was covered
by floating ice; as in the Fram Strait [38]. Bioproductiv-
ity remained low even during the B–A episode, which is
confirmed by the absence of planktonic foraminifers in
deglaciation sediments, except for occasional intercala-
tions that reflect brief periods with hydrological condi-
tions favorable for plankton development. Thus, during
the entire deglaciation interval the sea-surface layer was
colder than the subsurface and bottom layers, which

126
OCEANOLOGY Vol. 56 No. 1 2016
IVANOVA et al.
were warmer in areas influenced by Atlantic water, par-
ticularly during the Bølling–Allerød warming episode
with intensified global thermohaline circulation
(Figs. 4, 5c) [12, 37, 51]. The Younger Dryas cooling
was marked by a relative advance of the Scandinavian
ice sheet to the northern part of the Baltic Sea and gla-
ciers from the Spitsbergen, Franz Josef Land, and
Novaya Zemlya archipelagoes onto the shelf (Fig. 5d).
Fig. 5. Schematic maps of the Atlantic water inflow and deglaciation in the Barents Sea–Baltic Sea region for different time intervals
(cal ka BP): (a) 20–18 (LGM), (b) 17–15 (ED), (c) ~14 (B–A), (d) ~12.0–11.7 (YD, terminal BIL stage), (e) ~11.4–11.1 (Pre-
boreal, YS), (f) ~10.5–10.3 (Early Holocene, AL) accounting for original data and reconstructions in [2, 6, 10, 13, 15, 21, 22, 27,
28, 30, 32, 33, 35, 41, 42, 44–46, 48–51, 54, 56–60, 62] and modeling results from [29, 55]. Black dots designate stations with a
defined strong “signal” of Atlantic waters at the surface (white) and bottom (black) layers of the water column; for station coordinates
see Table 1. The solid black line shows boundaries of glaciers, and dotted line indicates boundaries of basins that preceded the Baltic
Sea. Solid black and gray arrows designate the inflow of surface and subsurface waters, respectively, into the Barents Sea; dotted
arrows mark the weak inflow. For abbreviations, see captions to Fig. 2.
~20−18 ca l ka BP
80°
N
60°
N
20° E
60°
E
40°
E
(a)
S
c
a
n
d
i
n
a
v
i
a
n
–
B
a
r
e
n
t
s
i
c
e
s
h
e
e
t
?
~17−15 ca l ka BP
80°
N
60°
N
20° E
60°
E
40°
E
(b)
S
c
a
n
d
i
n
a
v
i
a
n
–
B
a
r
e
n
t
s
i
c
e
s
h
e
e
t
?
?
?
?
JM02-460
PSh-5159
POS 303700
S-2519
JPC 5
PL94-29
ASV-880
PL94-07
NP94-51
~14 cal ka BP
80°
N
60°
N
20° E
60°
E
40°
E
(c)
?
?
?
?
JM02-460
MD95-2011
JM02-440
PSh-5159
POS 303700
S-2519
JPC 5
ASV-880
PL194-07
NP94-51
Scandinavian
ice sheet
?
~12−11.7 cal ka BP
80°
N
60°
N
20° E
60°
E
40°
E
(d)
?
JM02-460
M 23258
MD95-2011
JM02-440
PSh-5159
POS 303700
S-2519
JPC 5
ASV-880
NP94-51
NP05-21
Scandinavian
ice sheet
BIL
BIL

OCEANOLOGY Vol. 56 No. 1 2016
POSTGLACIAL PALEOCEANOGRAPHIC ENVIRONMENTS 127
The transition from glaciomarine environments of
deglaciation to the marine conditions of the Holocene
occurred against the background of a sea level rise and
an irregular glacio-isostatic rebound in different areas
of the Barents and Baltic seas (Fig. 2) [20, 32].
By the beginning of the Holocene, the Barents Sea
ice sheet was collapsed and the small glaciers retained
only on the surrounding archipelagoes, while the
Scandinavian sheet became smaller and retreated
northward from the BIL (Fig. 5d) [1, 7, 8, 14, 30, 60].
In both basins, global warming and glacio-eustatic sea
level rise at the beginning of the Holocene provoked
the transgression and development of marine environ-
ments instead of glaciomarine and lacustrine condi-
tions in the Barents and Baltic seas, respectively. Each
of them received a significant meltwater influx, which
determined low sea-surface temperatures and sea ice
expansion in the Barents Sea [26, 50, 51]. The Early
Holocene, approximately 11.3–9.0 cal ka BP, was
marked by a significantly intensified inf lux of Atlantic
water via the Norwegian Sea into the Barents Sea, fill-
ing it from the surface to bottom, and a further retreat
of the glaciers on the Svalbard, Franz Josef Land, and
Novaya Zemlya archipelagoes (Figs. 4, 5e, 5f) [6, 51].
This is evident from relatively high temperatures in the
surface water layer (0–100 m) at Stations MD95-2011,
M 23258, JM03-373, and PSh-5159 [15, 44, 49, 51]
and an increase in the relative content of “Atlantic”
species in benthic foraminiferal assemblages from
Cores PSh-5159R, NP05-21, and PL94-67 [6, 32, 45]
as compared to the YD interval.
At that time, the Baltic Ice Lake drained into
neighboring basins via the newly opened Närke Strait
in southern Scandinavia (Fig. 5d). As a result the lake
level subsided by 20–30 m and became equal to the
global sea level [1, 13]. Due to the relatively slow sea-
water inflow into the lake, the latter was gradually (for
about 300 years [1]) transformed into a brackish Yoldia
Sea (YS, Fig. 5e) [8]. This occurred at the very begin-
ning of the Holocene (11.7–10.7 cal ka BP [1, 14]) or
even earlier, in the terminal Younger Dryas–initial
Preboreal [2] (Figs. 4, 5e). The Yoldia Sea was still
highly influenced by meltwater from the retreating
Scandinavian ice sheet. Saline Atlantic waters started
to penetrate into the West Gotland Basin, although
their influx was insignificant and any considerable
increase in salinity of the YS had not occurred [19].
Salinity as well as bioproductivity were spatially highly
variable, remaining generally low (e.g. [1]); therefore,
the older stage may only conditionally be considered
as marine one. For example, in the Gdansk Basin, no
salinity increase is registered both by geochemical and
micropaleontological (diatoms) methods (Fig. 4) [2].
In its southern part, the sea was connected with the
Danish Straits via shallow-water thresholds [14, 60].
In response to the Early Holocene warming epi-
sode, rapid deglaciation, and high rates of glacio-iso-
static Baltic ice sheet uplift in southern and central
Sweden and southern Finland, the Närke Strait
became drained and the Yoldia Sea was replaced by
the dammed undrained Ancylus Lake (AL, ~10.7–
8.0 cal ka BP, Fig. 5f). Like the Yoldia Sea, the latter
was highly influenced by meltwater from the Scandi-
Fig. 5. (Contd.)
~10.5−10.3
cal ka BP
?
?
80°
N
60°
N
20° E
60°
E
40°
E
(f)
JM02-460
M 23258
JM03-373
MD95-2011
JM10-10
JM02-440
PSh-5159
POS 303700
S-2519
JPC 5
ASV-880
PL94-67
NP94-51
NP05-21
Scandinavian
ice sheet
AL
~11.4−11.1
cal ka BP
80°
N
60°
N
20° E
60°
E
40°
E
(e)
M 23258
JM02-460
JM03-373
MD95-2011
JM02-440
JM10-10
PSh-5159
POS 303700
S-2519
JPC 5
ASV-880
PL94-67
NP94-51
NP05-21
Scandinavian
ice sheet
Närke Strait
YS

128
OCEANOLOGY Vol. 56 No. 1 2016
IVANOVA et al.
navian ice sheet, and its level was higher than global
sea level [1, 7, 14, 19]. The subsequent fall of the
AL level to the global sea level was determined by both
the growth of the ice sheet and the activity of rivers,
which eroded thresholds that separated the lake from
the sea on the west. At the end of the Ancylus stage,
the salinity of waters in the lake increased to turn the
latter into a brackish-water basin (Fig. 4) [1, 2], and
subsequently, in response to the formation of the
Great Belt Straits in Denmark, the level of the lake
rapidly lowered with a simultaneous reduction of its
size [14]. The Ancylus Lake drained into the North
Sea at that time [1]. Further global sea level rise and
the isostatic submergence of the southern North and
southern Baltic seas (along with the uplift in Scandi-
navia) resulted in the flooding of the lake by marine
waters via the Danish Straits [5]. Some authors define
the initial phase of the transgression as the Mastogloia
stage, several hundred years long [1, 2]. According to
different researchers, the next Littorina Sea (LS) stage
includes from three to six transgressive phases [2].
According to different authors, the onset of the Litto-
rina stage is dated from 10 to 7 ca ka BP [2, 3, 14] and
is related to the influx of North Sea waters into the
Baltic basin via the Danish Strait in response to the
continuing glacio-eustatic sea level rise. In the Gdansk
Basin, salinity increased up to 17–18‰, and bottom
currents intensified during the third of three or four
phases of the Littorina transgression [2]. The pulsa-
tory influx of marine waters from the southwest into
the deeps of the Littorina Sea and their weak mixing
with fresh surface waters led to the halocline forma-
tion, significant growth of bioproductivity, intermit-
tent H2S contamination of bottom waters in the deeps,
and the accumulation of microlaminated mud [19].
Owing to this, Littorina sediments are readily recog-
nizable in examined core sections [1, 4].
The interval approximately 8–6 cal ka BP was
characterized by a rapid rise of the LS level against the
background of relatively slow glacio-isostatic uplift of
the sea floor at least in its southern part [14]. This
interval corresponds to a reliably established warming
episode in the Norwegian and Baltic seas resulted
either from an increase in insolation [51] or intensified
Atlantic water inflow [17, 32].
Some researchers define the recent stage in devel-
opment of the Baltic region as the post-Littorina Sea
(PLS, approximately the last 4 ka, Fig. 4) [1]. This
stage is characterized by bottom-water salinity lower-
ing up to 12‰ and its further stabilization due to the
weakened inflow of saline waters from the North Sea
(Fig. 4) [2]. Hampered water exchange with the ocean
was determined by neotectonic uplift in the area of the
Danish Straits and subsequent LS regression [2, 16].
CONCLUSIONS
The differences in the postglacial history of the two
basins are primarily explained by the following factors:
(a) the intracontinental geographic position of the
Baltic Sea in temperate latitudes and the marginal
position of the arctic Barents Sea; (b) the presence of
deep troughs serving as paths for Atlantic water inflow
into the Barents Sea, and restricted water exchange of
the Baltic Sea with the ocean via shallow straits;
(c) the different influence of glacio-isostatic bottom
uplift. Thus, diachronous postglacial development in
specific areas of basins covered by a common
grounded ice sheet during the Last Glacial Maximum
demonstrated different patterns. However, the evolu-
tion of both basins exhibited the impact of global cli-
mate change and ice sheet decay via the ice stream
mechanisms.
ACKNOWLEDGMENTS
We thank T.L. Rasmussen for the discussion of
results and E.I. Polyakova for her useful recommenda-
tions. This work was supported by the Russian Science
Foundation (grant no. 14-50-00095) (E.V. Ivanova,
I.O. Murdmaa) and Otto Schmidt Laboratory for
Marine and Polar Research of the Arctic and Antarctic
Research Institute, St. Petersburg (grant nos. OSL 14-
09 and OSL 15-08; micropaleontological, lithological,
and petrographic study of the Core S-2519).
REFERENCES
1. A. I. Blazhchishin, Paleogeography and Evolution of
Late Quaternary Sedimentation in the Baltic Sea (Yan-
tarnyi Skaz, Kaliningrad, 1998) [in Russian].
2. A. G. Grigor’ev, V. A. Zhamoida, M. A. Spiridonov,
et al., “New data on evolution of southeastern part of
the Baltic Sea for Later Glaciation period until pres-
ent,” Reg. Geol. Metallog., No. 40, 103–114 (2009).
3. V. K. Gudelis and E. M. Emelyanov, Geology of the Bal-
tic Sea (Moklas, Vilnius, 1976) [in Russian].
4. E. M. Emelyanov, “Power of littoral silts and near-bot-
tom long-period currents in depression of the Baltic Sea
below halocline,” in Complex Studies of the Processes,
Characteristics, and Resources of the Russian Seas of the
North European Basin (Apatity, 2007), No. 2, pp. 381–
398.
5. E. M. Emelyanov, “Deglaciation of Northern Europe
and its reflection in marine sediments,” in Proceedings
of the XX International Scientific Conference (School) on
Marine Geology “Geology of the Seas and Oceans”
(GEOS, Moscow, 2013), Vol. 1, pp. 205–207.
6. N. O. Chistyakova, E. V. Ivanova, B. Risebrobakken,
E. A. Ovsepyan, and Y. S. Ovsepyan, “Reconstruction
of the postglacial environments in the southwestern
Barents Sea based on foraminiferal assemblages,”
Oceanology (Engl. Transl.) 50 (4), 573–581 (2010).
7. T. Andrén, “Baltiska Issjön – eller hur det började,”
Havsutsikt 1, 4–5 (2003).
8. T. Andrén, “Yoldiahavet – en viktig parentes,” Hav-
sutsikt 2, 6–7 (2003).
9. T. Andrén, “Ancylussjön – fortfarande ett mysterium,”
Havsutsikt 3, 8–9 (2003).

OCEANOLOGY Vol. 56 No. 1 2016
POSTGLACIAL PALEOCEANOGRAPHIC ENVIRONMENTS 129
10. T. Andrén, S. Björck, E. Andrén, et al., “The Develop-
ment of the Baltic Sea Basin during the Last 130 ka,” in
The Baltic Sea Basin, Ed. by J. Harff, et al., (Springer-
Verlag, Berlin, 2011), pp. 75–98.
11. O. Bennike and J. B. Jensen, “A Baltic Ice Lake low-
stand of latest Allerød age in the Arkona basin, southern
Baltic Sea,” Geol. Surv. Den. Greenl. Bull. 28, 17–20
(2013).
12. H. A. Bauch, H. Erlenkeuser, R. F. Spielhagen, et al.,
“A multiproxy reconstruction of the evolution of deep
and surface waters in the subarctic Nordic seas over the
last 30 000 yr,” Quat. Sci. Rev. 20 (4), 659–678 (2001).
13. S. Björck, “A review of the history of the Baltic Sea,
13.0–8.0 ka BP,” Quat. Int. 27, 19–40 (1995).
14. S. Björck, “The late Quaternary development of the
Baltic Sea basin,” in Assessment of Climate Change for
the Baltic Sea Basin (Springer-Verlag, Berlin, 2008),
pp. 398–407.
15. E. Calvo, J. Grimalt, and E. Jansen, “High resolution
UK37 sea surface temperature reconstruction in the
Norwegian Sea during the Holocene,” Quat. Sci. Rev.
21, 1385–1394 (2002).
16. L. B. Clemmensen, A. S. Murray, and L. Nielsen,
“Quantitative constraints on the sea-level fall that ter-
minated the Littorina Sea Srage, southern Scandina-
via,” Quat. Sci. Rev., No. 40, 54–63 (2012).
17. J. C. Duplessy, E. V. Ivanova, I. O. Murdmaa, et al.,
“Holocene paleoceanography of the Northern Barents
Sea and variations of the northward heat transport by
the Atlantic Ocean,” Boreas 30 (1), 2–16 (2001).
18. A. Elverhoi, J. A. Dowdeswell, S. Funder, et al., “Gla-
cial and oceanic history of the Polar North Atlantic
margins: an overview,” Quat. Sci. Rev. 17 (1–3), 1–10
(1998).
19. E. M. Emelyanov and G. Vaikutienė, “Holocene envi-
ronmental changes during transition Ancylus-Litorina
stages in the Gdansk Basin, south-eastern Baltic Sea,”
Baltica 26 (1), 71–82 (2013).
20. J. Harff, A. Frischbutter, R. Lampe, and M. Meyer,
“Sea-level change in the Baltic Sea: interrelation of cli-
matic and geological processes,” in Geological Perspec-
tives of Global Climate Change, Ed. by L. C. Gerhard,
et al., (Tulsa, OK, 2001), Vol. 12, pp. 231–250.
21. V. Gataullin, J. Mangerud, and J. I. Svendsen, “The
extent of the Late Weichselian ice sheet in the south-
eastern Barents Sea,” Global Planet. Change 31, 453–
474 (2001).
22. R. Gyllencreutz, J. Backman, M. Jakobsson, et al.,
“Postglacial palaeoceanography in the Skagerrak,” The
Holocene 16 (7), 975–985 (2006).
23. K. A. Hughen, M. G. L. Baillie, E. Bard, et al.,
“Marine radiocarbon age calibration, 0–26 cal kyr
BP,” Radiocarbon 46, 1059–1086 (2004).
24. E. V. Ivanova, The Global Thermohaline Paleocircula-
tion (Springer-Verlag, New York, 2009).
25. E. V. Ivanova, I. O. Murdmaa, J. C. Duplessy, and
M. Paterne, “Late Weichselian to Holocene Paleoenvi-
ronments in the Barents Sea,” Global Planet. Change
34 (3–4), 69–78 (2002).
26. E. Ivanova, I. Murdmaa, A. de Vernal, et al., “Northern
Barents Sea environment during the last deglaciation:
response to changes in AMOC and sea-ice extent,” in
11th International Conference on Paleoceanography,
Abstracts of Papers (Sitges-Barcelona, Spain, 2013),
p. 76.
27. V. Kalm, “Ice-flow pattern and extent of the last Scan-
dinavian ice sheet southeast of the Baltic Sea,” Quat.
Sci. Rev. 44, 51–59 (2012).
28. D. Klitgaard-Kristensen, T. L. Rasmussen, and N. Koç,
“Palaeoceanographic changes in the northern Barents
Sea during the last 16 000 years – new constraints on the
last deglaciation of the Svalbard–Barents Sea Ice Sheet,”
Boreas 42, 798–813 (2013).
29. K. Lambeck, C. Smither, and P. Johnston, “Sea-level
change, glacial rebound and mantle viscosity for north-
ern Europe,” Geophys. J. Int. 134, 102–144 (1998).
30. J. Y. Landvik, S. Bondebik, A. Elyerhoi, et al., “The
last glacial maximum of Svalbard and the Barents Sea
area: Ice Sheet extent and configuration,” Quat. Sci.
Rev. 17 (1–3), 43–76 (1998).
31. H. Loeng, “Features of the physical oceanographic
conditions of the Barents Sea,” Polar Res. 10 (1), 5–18
(1991).
32. D. J. Lubinski, L. Polyak, and S. L. Forman, “Fresh-
water and Atlantic water inflows to the deep northern
Barents and Kara seas since ca 13–14C ka: foraminifera
and stable isotopes,” Quat. Sci. Rev. 20, 1851–1879
(2001).
33. J. Mangerud, M. Jakobsson, H. Alexanderson, et al.,
“Ice-dammed lakes and rerouting of the drainage of
northern Eurasia during the Last Glaciation,” Quat.
Sci. Rev. 23, 1313–1332 (2004).
34. J. Mangerud, S. Bondevik, S. Gulliksen, et al., “Marine
14C reservoir ages for 19th century whales and mollusks
from the North Atlantic,” Quat. Sci. Rev. 25, 3228–3245
(2006). doi: 10.1016/j.quascirev.2006.03.010
35. L. Marks, “Pleistocene glacial limits in the territory of
Poland,” Przegląd Geol. 53 (10/2), 988–993 (2005).
36. W. Matthäus, “Natural variability and human impacts
reflected in long-term changes in the Baltic Deep water
conditions – a brief review,” Dtsch. Hydrogr. Z. 47,
47–65 (1995).
37. J. F. McManus, R. Francois, J.-M. Gherardi, et al.,
“Collapse and rapid resumption of Atlantic meridional
circulation linked to deglacial climate changes,” Nature
428, 834–837 (2004).
38. J. Müller and R. Stein, “High-resolution record of late
glacial and deglacial sea ice changes in Fram Strait cor-
roborates ice–ocean interactions during abrupt climate
shifts,” Earth Planet. Sci. Lett. 403, 446–455 (2014).
39. I. O. Murdmaa, E. V. Ivanova, J. C. Duplessy, et al.,
“Facies system of the Central and Eastern Barents Sea
since the Last Glaciation to recent,” Mar. Geol. 230 (3–
4), 273–303 (2006).
40. North Greenland Ice Core Project members, “High-
resolution record of Northern Hemisphere climate
extending into the last interglacial period,” Nature 431,
147–151 (2004).
41. D. Ottesen, J. A. Dowdeswell, and L. Rise, “Subma-
rine landforms and the reconstruction of fast-f lowing
ice streams within a large Quaternary ice sheet: the
2500-km-long Norwegian-Svalbard margin (57°–80°
N),” GSA Bull. 117 (7/8), 1033–1050 (2005). doi:
10.1130/B25577.1.

130
OCEANOLOGY Vol. 56 No. 1 2016
IVANOVA et al.
42. D. Ottesen, C. R. Stokes, L. Risea, and L. Olsen, “Ice
sheet dynamics and ice streaming along the coastal
parts of northern Norway,” Quat. Sci. Rev. 27, 922–
940 (2008).
43. Yu. Pavlidis and E. I. Polyakova, “Late Pleistocene and
Holocene depositional environments and paleocean-
ography of the Barents Sea: evidence from seismic and
biostratigraphic data,” Mar. Geol. 143, 189–205
(1997).
44. T. L. Rasmussen, E. Thomsen, M. A. Ślubowska, et al.,
“Paleoceanographic evolution of the SW Svalbard mar-
gin (76° N) since 20 000 14C yr BP,” Quat. Res. 67,
100–114 (2007).
45. T. L. Rasmussen, E. Thomsen, K. Skirbekk, et al.,
“Spatial and temporal distribution of Holocene tem-
perature maxima in the northern Nordic seas: interplay
of Atlantic-, Arctic- and polar water masses,” Quat.
Sci. Rev. 92, 280–281 (2014).
46. T. L. Rasmussen and E. Thomsen, “Palaeoceano-
graphic development in Storfjorden, Svalbard, during
the deglaciation and Holocene: evidence from benthic
foraminiferal records,” Boreas 44, 24–44 (2015).
47. P. J. Reimer, M. G. L. Baillie, E. Bard, et al., “IntCal09
and Marine09 Radiocarbon Age calibration curves, 0–
50,000 years cal BP,” Radiocarbon 51 (4) , 1111 –1150
(2009).
48. V. R. Rinterknecht, A. Bitinas, P. U. Clark, et al.,
“Timing of the last deglaciation in Lithuania,” Boreas
37, 426–433 (2008).
49. B. Risebrobakken, E. Jansen, C. Andersson, et al., “A
high-resolution study of Holocene paleoclimatic and
paleoceanographic changes in the Nordic Seas,” Pale-
oceanography 18 (1), 1017 (2003). doi: 10.1029/
2002PA000764
50. B. Risebrobakken, M. Moros, E. Ivanova, et al., “Cli-
mate and oceanographic variability in the SW Barents
Sea during the Holocene,” Holocene 20 (4), 609–621
(2010). doi: 10.1177/0959683609356586
51. B. Risebrobakken, T. Dokken, L. H. Smedsrud, et al.,
“Early Holocene temperature variability in the Nordic
Seas: the role of oceanic heat advection versus changes
in orbital forcing,” Paleoceanography 26, PA4206
(2011). doi: 10.1029/2011PA002117
52.M. Sarnthein, J. P. Kennett, J. Chappel, et al.,
“Exploring Late Pleistocene climate variations,”
Trans., Am. Geophys. Union. 81 (51), 625, 629–630
(2000).
53. M. Sarnthein, S. van Kreveld, H. Erlenkauser, et al.,
“Centennial to millennial scale periodicities of Holo-
cene climate and sediment injections off the western
Barents shelf,” Boreas 32 (75), 447–461 (2003). doi:
10.1080/03009480301813
54. M. Sarnthein, U. Pflaumann, and M. Weinelt, “Past
extent of sea ice in the northern North Atlantic inferred
from foraminiferal paleotemperature estimates,” Pale-
oceanography 18 (2), 1047 (2003). doi: 10.1029/
2002pa000771
55. M. J. Siegert and J. A. Dowdeswell, “Late Weichselian
iceberg, surface-melt and sediment production from
the Eurasian Ice Sheet: results from numerical ice-
sheet modeling,” Mar. Geol. 188, 109–127 (2002).
56. M. A. Ślubowska, N. Koç, T. L. Rasmussen, and
D. Klitgaard-Kristensen, “Changes in the f low of
Atlantic water into the Arctic Ocean since the last
deglaciation: evidence from the northern Svalbard con-
tinental margin, 80˚ N,” Paleoceanography 20, PA4014
(2005).
57. M. Ślubowska-Woldengen, T. L. Rasmussen, N. Koç,
et al., “Advection of Atlantic water to the western and
northern Svalbard shelf since 17 500 cal yr BP,” Quat.
Sci. Rev. 26, 463–478 (2007).
58. J. I. Svendsen, H. Alexanderson, V. I. Astakhov, et al.,
“Late Quaternary ice sheet history of northern Eur-
asia,” Quat. Sci. Rev. 23, 1229–1271 (2004).
59. M. Tikkanen and J. Oksanen, “Late Weichselian and
Holocene shore displacement history of the Baltic Sea
in Finland,” Fennia 180, 9–20 (2002).
60. S. Uścinowicz, “Relative sea level changes, glacio-iso-
static rebound and shoreline displacement in the
Southern Baltic,” Pol. Geol. Inst. Spec. Pap. 10, 1–80
(2003).
61. C. Waelbroeck, L. Labeyrie, E. Michel, et al., “Sea-
level and deep water temperature changes derived from
benthic foraminifera isotopic records,” Quat. Sci. Rev.
21, 295–305 (2002).
62. W. Wysota, P. Molewski, and R. J. Soko1owski,
“Record of the Vistula ice lobe advances in the Late
Weichselian glacial sequence in north-central Poland,”
Quat. Int. 207, 26–41 (2009).
63. J. Zachowicz, G. Miotk-Szpiganowicz, R. Kramarska,
et al., “A critical review and reinterpretation of bio-,
litho- and seismostratigraphic data of the Southern
Baltic deposits,” Pol. Geol. Inst. Spec. Pap. 23, 117–
138 (2008).
Translated by I. Basov