Available via license: CC BY 3.0
Content may be subject to copyright.
Clim. Past, 13, 1111–1127, 2017
https://doi.org/10.5194/cp-13-1111-2017
© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.
Holocene dynamics in the Bering Strait inflow to the Arctic
and the Beaufort Gyre circulation based on sedimentary
records from the Chukchi Sea
Masanobu Yamamoto1,2,3, Seung-Il Nam4, Leonid Polyak5, Daisuke Kobayashi3, Kenta Suzuki3, Tomohisa Irino1,3,
and Koji Shimada6
1Faculty of Environmental Earth Science, Hokkaido University, Kita-10, Nishi-5, Kita-ku, Sapporo 060-0810, Japan
2Global Institution for Collaborative Research and Education, Hokkaido University, Kita-10, Nishi-5, Kita-ku,
Sapporo 060-0810, Japan
3Gradute School of Environmental Science, Hokkaido University, Kita-10, Nishi-5, Kita-ku, Sapporo 060-0810, Japan
4Korea Polar Research Institute, 26 Songdomirae-ro, Yeonsu-gu, Incheon 21990, Republic of Korea
5Byrd Polar and Climate Research Center, The Ohio State University, Columbus, OH 43210, USA
6Tokyo University of Marine Science and Technology, 4-5-7, Konan, Minato-ku, Tokyo 108-8477, Japan
Correspondence to: Masanobu Yamamoto (myama@ees.hokudai.ac.jp)
Received: 1 April 2017 – Discussion started: 24 April 2017
Revised: 21 July 2017 – Accepted: 9 August 2017 – Published: 8 September 2017
Abstract. The Beaufort Gyre (BG) and the Bering Strait in-
flow (BSI) are important elements of the Arctic Ocean circu-
lation system and major controls on the distribution of Arctic
sea ice. We report records of the quartz /feldspar and chlo-
rite /illite ratios in three sediment cores from the northern
Chukchi Sea, providing insights into the long-term dynam-
ics of the BG circulation and the BSI during the Holocene.
The quartz /feldspar ratio, interpreted as a proxy of the BG
strength, gradually decreased during the Holocene, suggest-
ing a long-term decline in the BG strength, consistent with
an orbitally controlled decrease in summer insolation. We
propose that the BG rotation weakened as a result of the
increasing stability of sea-ice cover at the margins of the
Canada Basin, driven by decreasing insolation. Millennial to
multi-centennial variability in the quartz /feldspar ratio (the
BG circulation) is consistent with fluctuations in solar irradi-
ance, suggesting that solar activity affected the BG strength
on these timescales. The BSI approximation by the chlo-
rite /illite record, despite a considerable geographic variabil-
ity, consistently shows intensified flow from the Bering Sea
to the Arctic during the middle Holocene, which is attributed
primarily to the effect of higher atmospheric pressure over
the Aleutian Basin. The intensified BSI was associated with
decrease in sea-ice concentrations and increase in marine
production, as indicated by biomarker concentrations, sug-
gesting a major influence of the BSI on sea-ice and biological
conditions in the Chukchi Sea. Multi-century to millennial
fluctuations, presumably controlled by solar activity, were
also identified in a proxy-based BSI record characterized by
the highest age resolution.
1 Introduction
The Arctic currently faces rapid climate change caused by
global warming (e.g., Screen and Simmonds, 2010; Harada,
2016). Changes in the current system of the Arctic Ocean
regulate the state of Arctic sea ice and are involved in global
processes via ice albedo feedback and the delivery of fresh-
water to the North Atlantic Ocean (Miller et al., 2010; Screen
and Simmonds, 2010). The most significant consequence of
this climate change during recent decades is the retreat of
summer sea ice in the Pacific sector of the Arctic (e.g., Shi-
mada et al., 2006; Harada, 2016, and references therein). In-
flow of warm Pacific water through the Bering Strait (here-
after Bering Strait inflow – BSI) is suggested to have caused
catastrophic changes in sea-ice stability in the western Arctic
Ocean (Shimada et al., 2006). Comprehending these changes
requires the investigation of a longer-term history of circula-
tion in the western Arctic and its relationship to atmospheric
Published by Copernicus Publications on behalf of the European Geosciences Union.
1112 M. Yamamoto et al.: Holocene dynamics of western Arctic ocean circulation
Figure 1. Index map showing location of cores ARA02B 01A-GC (this study), HLY0501-05JPC/TC (this study and Farmer et al., 2011),
HLY0501-06JPC (this study and Ortiz et al., 2009), and HLY0205-GGC19 (Farmer et al., 2011), as well as surface sediment samples
(Kobayashi et al., 2016, with additions). The positions of added surface sediments are listed in Supplement Table S1. BC: Barrow Canyon;
HN: Hanna Shoal; HR: Herald Shoal. BG: Beaufort Gyre; ACC: Alaskan Coastal Current; SBC: Subsurface Boundary Current; ESCC: East
Siberian Coastal Current; TPD: Transpolar Drift. Red, yellow, and blue arrows indicate BSI branches. AO+and AO−indicate circulation in
the positive and negative phases of the Arctic Oscillation, respectively.
forcings. Within this context, the Chukchi Sea is a key re-
gion to understand the western Arctic current system as it is
located at the crossroads of the BSI and the Beaufort Gyre
(BG) circulation in the western Arctic Ocean (Fig. 1) (e.g.,
Winsor and Chapman, 2004; Weingartner et al., 2005).
In this paper we apply mineralogical proxies of the BG
and BSI to sediment cores with a century-scale resolution
from the northern margin of the Chukchi shelf. The gener-
ated record provides a new understanding of changes in the
BG circulation and BSI strength during most of the Holocene
(last ∼9 kyr). We discuss the possible causes and forcings of
the BG and BSI variability, as well as its relationship to sea-
ice history and biological production in the western Arctic.
2 Background information
2.1 Oceanographic settings
The wind-driven surface current system of the Arctic
Ocean consists of the BG and the Transpolar Drift (TPD)
(Proshutinsky and Johnson, 1997; Rigor et al., 2002). This
circulation is controlled by the atmospheric system known as
the Arctic Oscillation (AO) (Rigor et al., 2002). When the
AO is in the positive phase, the BG shrinks back into the
Beaufort Sea, the TPD expands to the western Arctic Ocean,
and the sea-ice transport from the eastern Arctic to the At-
lantic Ocean is intensified. When the AO is in negative phase,
the BG expands, the TPD is limited to the eastern Arctic, and
sea ice is exported efficiently from the Canada Basin to the
eastern Arctic. Thus, sea-ice distribution is closely related to
the current system.
A dramatic strengthening of the BG circulation occurred
during the last two decades (Shimada et al., 2006; Giles
et al., 2012). This change was attributed to a recent reduc-
tion in sea-ice cover along the margin of the Canada Basin,
which caused a more efficient transfer of the wind momen-
tum to the ice and underlying waters in the BG (Shimada
et al., 2006). The delayed development of sea ice in winter
enhanced the western branch of the Pacific Summer Water
across the Chukchi Sea. This anomalous heat flux into the
western part of the Canada Basin retarded sea-ice formation
during winter, thus further accelerating overall sea-ice reduc-
tion.
The BSI, an important carrier of heat and freshwater to
the Arctic, transports the Pacific water to and across the
Chukchi Sea and interacts with the BG circulation at the
Chukchi shelf margin (e.g., Shimada et al., 2006). Moor-
ing data suggest that an increase in the BSI volume by
Clim. Past, 13, 1111–1127, 2017 www.clim-past.net/13/1111/2017/
M. Yamamoto et al.: Holocene dynamics of western Arctic ocean circulation 1113
∼50 % from 2001 (∼0.7 Sv) to 2011 (∼1.1 Sv) has driven
an according increase in the heat flux from ∼3×1020 to
∼5×1020 J (Woodgate et al., 2012). After passing the Bering
Strait the BSI flows in three major branches. One branch, the
Alaskan Coastal Current (ACC), runs northeastward along
the Alaskan coast as a buoyancy-driven boundary current
(Red arrow in Fig. 1; Shimada et al., 2001; Pickart, 2004;
Weingartner et al., 2005). The second, central branch fol-
lows a seafloor depression between Herald and Hanna shoals,
then turns eastward and merges with the ACC (Yellow ar-
row in Fig. 1; Winsor and Chapman, 2004; Weingartner et
al., 2005). The third branch flows northwestward, especially
when easterly winds prevent the ACC from flowing (Winsor
and Chapman, 2004). This branch may then turn eastward
along the shelf break (Blue arrow in Fig. 1; Pickart et al.,
2010).
The BSI is driven by a northward dip in sea level be-
tween the North Pacific and the Arctic Ocean (Shtokman,
1957; Coachman and Aagaard, 1966). There has been a long-
standing debate whether this dipping is primarily controlled
by steric difference (Stigebrandt, 1984) or wind-driven cir-
culations (Gudkovitch, 1962). Stigebrandt (1984) assumed
that the salinity difference between the Pacific and Atlantic
oceans causes the steric height difference between the Bering
Sea and the Arctic Ocean. Aagaard et al. (2006) argued that
the local salinity in the northern Bering Sea controlled the
BSI, although wind can considerably modify the BSI on
a seasonal timescale. De Boer and Nof (2004) proposed a
model by which the mean sea level difference along the strait
is set up by the global winds, particularly the strong sub-
antarctic westerlies.
Recently, a conceptual model of the BSI controls has been
developed based on a decade of oceanographic observations
(Danielson et al., 2014). According to this model, storms
centered over the Bering Sea excite continental shelf waves
on the eastern Bering shelf that intensify the BSI on syn-
optic timescales, but the integrated effect of these storms
tends to decrease the BSI on annual to decadal timescales.
At the same time, an eastward shift and overall strengthen-
ing of the Aleutian Low pressure center during the period
between 2000–2005 and 2005–2011 increased the sea level
pressure in the Aleutian Basin south of the Bering Strait
by 5 hPa, in contrast to the overall decreased pressure of
the Aleutian Low system, thus decreasing the water column
density through isopycnal uplift by weaker Ekman suction.
This change thereby raised the dynamic sea surface height
by 4.2 m along the Bering Strait pressure gradient, result-
ing in the BSI increase by 4.5 cm s−1, or 0.2 Sv (calculated
based on the cross-section area of 4.25 ×106m2). This in-
crease constitutes about a quarter of the average long-term
BSI volume of ∼0.8 Sv (Roach et al., 1995). Such a large
contribution clearly identifies changes in the Aleutian Low
strength and position as a key factor regulating the BSI on
interannual timescales.
The BSI also transports nutrients from the Pacific to the
Arctic. A rough estimation suggests that the BSI waters
significantly contribute to marine production in the Arctic
(Yamamoto-Kawai et al., 2006). High marine production in
the Chukchi Sea of up to 400 g C m−2y−1in part is thought
to reflect the high nutrient fluxes by the BSI (Walsh and Di-
eterle, 1994; Sakshaug, 2004). A recent enhancement of bio-
logical productivity and the biological pump in the Beaufort
and Chukchi seas has been associated with the retreat of sea
ice (summarized by Harada, 2016). This phenomenon is at-
tributed to an increase in irradiance in the water column (Frey
et al., 2011; Lee and Whitledge, 2005), wind-induced mixing
that replenishes sea surface nutrients (Carmack et al., 2006),
and their combination (Nishino et al., 2009). However, the
nutrient flux into the Arctic Ocean was not evaluated in this
context. The investigation of BSI intensity and marine pro-
duction during the Holocene will be useful to understand on-
going changes in marine production in the Arctic Ocean.
2.2 Mineral distribution in the Chukchi Sea sediments
Spatial variation in mineral composition of surficial sedi-
ments along the western Arctic margin has been investi-
gated in a number of studies using different methodological
approaches but showing an overall consistent picture (e.g.,
Naidu et al., 1982; Naidu and Mowatt, 1983; Wahsner et al.,
1999; Kalinenko, 2001; Viscosi-Shirley et al., 2003; Darby
et al., 2011; Kobayashi et al., 2016). A recent study of min-
eral distribution in sediments from the Chukchi Sea and ad-
jacent areas of the Arctic Ocean and the Bering Sea sug-
gests that the quartz /feldspar (Q /F) ratio is higher on the
North American than on the Siberian side of the western Arc-
tic (Fig. 2; Kobayashi et al., 2016). These results are con-
sistent with earlier studies including mineral determinations
of shelf sediments and adjacent coasts (Vogt, 1997; Stein,
2008). Darby et al. (2011) show a trend of a decreasing Q /F
ratio in dirty sea ice from the North American margin to
the Chukchi Sea and further to the East Siberian Sea. This
zonal gradient of the Q /F ratio suggests that quartz-rich but
feldspar-poor sediments are derived from the North Ameri-
can margin by the BG circulation, whereas feldspar-rich sed-
iments are delivered to the Chukchi Sea from the Siberian
margin by currents along the East Siberian slope (Kobayashi
et al., 2016). Thus, this ratio can be used as a provenance in-
dex for the BG circulation reflecting changes in its intensity
in sediment-core records (Kobayashi et al., 2016).
Kaolinite is generally a minor component of clays in the
western Arctic but relatively abundant in the Northwind
Ridge and Mackenzie Delta areas where the BG circulation
exerts an influence (Naidu and Mowatt, 1983; Kobayashi et
al., 2016). Kaolinite in the Northwind Ridge originated from
ancient rocks exposed on the North Slope and was deliv-
ered by water or sea ice via the Beaufort Gyre circulation
(Kobayashi et al., 2016).
www.clim-past.net/13/1111/2017/ Clim. Past, 13, 1111–1127, 2017
1114 M. Yamamoto et al.: Holocene dynamics of western Arctic ocean circulation
Figure 2. Spatial distributions of the diffraction intensity ratios of (a) feldspar to quartz (Q/F) and (b) chlorite +kaolinite and (c) chlorite
to illite (CK /I and C /I, respectively) of bulk sediments; (d) the longitudinal distribution of the Q/F ratio in the western Arctic (>65◦N)
and (e) the latitudinal distribution of the CK /I and C /I ratios in the Bering Sea and the western Arctic (> 150◦W). The C /I ratio could
not be determined in some coarse-grained sediment samples. Data from Kobayashi et al. (2016) with additions for the Beaufort Sea (See
Supplement Table S1 in more detail). The regression lines in panel (e) show the geographic trends in mineral proxy distribution for the
Chukchi Sea. The Bering Sea sediments do not show a systematic pattern, probably reflecting multiple sources of chlorite, such as the Yukon
River, Aleutian Islands, etc. The enlarged maps of the Mackenzie River delta and Yukon River estuary are shown in Supplement Figs. S1 and
S2.
Kobayashi et al. (2016) also indicate that both the (chlo-
rite +kaolinite) /illite and chlorite /illite ratios (CK /I and
C/I ratios, respectively) are higher in the Bering Sea and de-
crease northward throughout the Chukchi Sea, reflecting the
diminishing strength of the BSI (Fig. 2). These results are
consistent with earlier studies showing that illite is a common
clay mineral in Arctic sediments (Kalinenko, 2001; Darby et
al., 2011), whereas chlorite is more abundant in the Bering
Sea and the Chukchi shelf areas influenced by the BSI (Naidu
and Mowatt, 1983; Kalinenko, 2001; Nwaodua et al., 2014;
Kobayashi et al., 2016). Chlorite occurs abundantly near the
Bering Sea coasts of Alaska, Canada, and the Aleutian Is-
lands (Griffin and Goldberg, 1963). The chlorite/illite ratio
is higher in the bed load of rivers and deltaic sediments from
southwestern Alaska than from northern Alaska and East
Siberia, reflecting differences in the geology of the drainage
basins (Naidu and Mowatt, 1983). Because chlorite grains
are more mobile than illite grains under conditions of intense
hydrodynamic activity, chlorite grains are transported a long
distance from the northern Bering Sea to the Chukchi Sea via
the Bering Strait (Kalinenko, 2001). In the surface sediments
of the Chukchi Sea, the CK /I ratio shows a good correlation
with the C /I ratio, indicating that both ratios can be used as
a provenance index for the BSI (Kobayashi et al., 2016).
Ortiz et al. (2009) constructed the first chlorite-based
Holocene record of the BSI by quantifying the total chlo-
Clim. Past, 13, 1111–1127, 2017 www.clim-past.net/13/1111/2017/
M. Yamamoto et al.: Holocene dynamics of western Arctic ocean circulation 1115
rite plus muscovite abundance based on diffuse spectral re-
flectance of sediments from a northeastern Chukchi Sea core.
The record shows a prominent intensification of the BSI in
the middle Holocene. However, a record from just one site is
clearly insufficient to characterize sedimentation and circula-
tion history in such a complex area. More records of mineral
proxy distribution covering various oceanographic and depo-
sitional environments are needed to further our understand-
ing of the evolution of the BSI.
The Holocene dynamics of the BG circulation is also
poorly understood. A study of sediment cores from the north-
eastern Chukchi slope identified centennial- to millennial-
scale variability in the occurrence of Siberian iron oxide
grains presumably delivered via the BG (Darby et al., 2012).
However, transport of these grains depends not only on the
BG but also on circulation and ice conditions in the Eurasian
basin, which complicates the interpretation and necessitates
further proxy studies of the BG history.
3 Samples and methods
3.1 Coring and sampling
This study uses three sediment cores from the northern and
northeastern margins of the Chukchi Sea: ARA02B 01A-
GC (gravity core; 563 cm long; 73◦37.890N, 166◦30.980W),
HLY0501-05JPC/TC (jumbo piston core/trigger; 1648 cm
long; 72◦41.680N, 157◦31.200W), and HLY0501-06JPC
(1554 cm long; 72◦30.710N, 157◦02.080W) collected from
111 m, 462 m, and 673 water depth, respectively (Fig. 1). The
sediments in 01A-GC and in the Holocene part of 05JPC/TC
(0–1300 cm) and 06JPC (0–935 cm) consist predominantly
of homogeneous clayey silt (fine-grained unit). This unit
of cores 05JPC and 06JPC is underlain by a more com-
plex lithostratigraphy with laminations and coarse ice-rafted
debris indicative of glaciomarine environments affected by
glacial/deglacial processes (“glaciomarine unit”; McKay et
al., 2008; Lisé-Pronovost et al., 2009; Polyak et al., 2009).
In total 110 samples were collected for mineralogical anal-
ysis from core 01A-GC at intervals averaging 5cm, equiva-
lent to approximately 80–90 years (see chronology descrip-
tion below), down to a depth of 545cm (ca. 9.3kaBP).
In core 05JPC/TC, 44 samples were collected from fine-
grained unit at intervals averaging 30cm (equivalent to ap-
proximately 210–220 years) down to a depth of 1286cm (ca.
9.3 ka) and 7 samples were collected from the underlying
glaciomarine sediments. In core 06JPC, 79 samples were col-
lected from fine-grained unit at intervals of 10 cm (equivalent
to approximately 90 years) down to a depth of 937 cm (ca.
8.0 ka) and 46 samples were collected from the underlying
glaciomarine unit.
We also analyzed 16 surface sediment samples (0–1 cm)
from the eastern Beaufort Sea near the Mackenzie River delta
and 3 surface sediment samples from the western Beaufort
Sea to fill the gaps in the dataset of Kobayashi et al. (2016)
(Fig. 2). These samples were obtained during the RV Araon
cruises in 2013 and 2014 (ARA04C and ARA05C, respec-
tively; Supplement Table S1).
3.2 Chronology
Age for core 01A-GC was constrained by seven accelerator
mass spectrometry (AMS) 14C ages of mollusc shells (Sup-
plement Table S2; Stein et al., 2017). The core top in ARA
01-GC may not represent the modern age due to some sed-
iment loss in the coring process. This is indicated by the
absence of oxidized brown sediment at the core top, as op-
posed to samples from a multi-corer collected at the same
site. Nevertheless, we believe that the top of 01-GC is close
to the sediment surface based on the biomarker distribution.
IP25 and brassicasterols show a downward decreasing trend
in their concentrations in the top 10 cm (Stein et al., 2017).
We suppose that this indicates their degradation with burial.
A similar extent of brassicasterol concentration decrease oc-
curs also in some of the deeper intervals, but it is unique for
the upper ∼200 cm, while the IP25 decrease at the top is
unique for the entire record. Therefore, the core top of 01A-
GC was assumed to represent sediment surface in the age–
depth model. 14C ages were converted to calendar ages using
the CALIB7.0 program and marine13 dataset (Reimer et al.,
2013). Local reservoir correction (1R) for 01A-GC sited in
surface waters was assumed to be 500 years (McNeely et al.,
2006; Darby et al., 2012). The age model was constructed
by linear interpolation between the 14C datings (3.1–8.6 ka).
Ages below the dated range were extrapolated to the bottom
of the core (9.3 ka).
In core 05JPC/TC, age was constrained by six AMS 14C
ages of mollusc shells from core 05JPC (Supplement Ta-
ble S2; Barletta et al., 2008; Darby et al., 2009). Local reser-
voir correction (1R) was assumed to be 0 years as the core
site is washed by Atlantic intermediate water (Darby et al.,
2012). Concurrent age constraints for 05JPC were provided
by 210Pb determinations in the upper part (05TC) and paleo-
magnetic analysis (Barletta et al., 2008; McKay et al., 2008).
The age model for core 05JPC/TC was constructed by linear
interpolation between the 14C datings (2.4–7.7 ka) as well as
the assumed modern age of the 05TC top, with the assump-
tion that the offset of JPC to TC is 75 cm (Darby et al., 2009).
Ages below the dated range were extrapolated to the bottom
of the homogeneous fine-grained unit at 1300 cm (9.4 ka).
In core 06JPC, age was tentatively constrained by 10
paleo-intensity datums based on regional paleomagnetic
chronology and a 14C age of benthic foraminifera (8.16 ka at
918 cm) (Supplement Table S2; Lisé-Pronovost et al., 2009),
with the assumption that the offset of JPC to TC is 147 cm
(Ortiz et al., 2009). The age model for core 06JPC was con-
structed by linear interpolation between the paleo-intensity
datums (2.0–7.9 ka).
www.clim-past.net/13/1111/2017/ Clim. Past, 13, 1111–1127, 2017
1116 M. Yamamoto et al.: Holocene dynamics of western Arctic ocean circulation
3.3 X-ray diffractometer mineralogy
Mineral composition was analyzed on an MX-Labo X-
ray diffractometer (XRD) equipped with a CuKαtube and
monochromator. The tube voltage and current were 40kV
and 20 mA, respectively. Scanning speed was 4◦2θmin−1,
and the data-sampling step was 0.02◦2θ. Each powdered
sample was mounted on a glass holder with a random orienta-
tion and X-rayed from 2 to 40◦2θ. An additional precise scan
with a scanning speed of 0.2◦2θmin−1and sampling step
of 0.01◦2θfrom 24 to 27◦2θwas conducted to distinguish
chlorite from kaolinite by evaluation of the peaks around
25.1◦2θ(Elvelhøi and Rønningsland, 1978). In this study,
the background-corrected diagnostic peak intensity was used
for evaluating the abundance of each mineral. The relative
XRD intensities of quartz at 26.6◦2θ(d=3.4 Å), feldspar
including both plagioclase and K feldspar at 27.7◦2θ(d=
3.2 Å), illite including mica at 8.8◦2θ(d=10.1 Å), chlo-
rite including kaolinite (called “chlorite+kaolinite” here-
after) at 12.4◦2θ(d=7.1 Å), kaolinite at 24.8◦2θ(d=
3.59 Å), chlorite at 25.1◦2θ(d=3.54 Å), and dolomite at
30.9◦2θ(d=2.9 Å) were determined using MacDiff soft-
ware (Petschick, 2000) based on the peak identification pro-
tocols of Biscaye (1965).
The mineral ratios used in this study are defined based on
XRD peak intensities (PIs) as
Q/F=quartz/feldspar = [PI at 26.6◦2θ]/[PI at 27.7◦2θ]
CK/I=(chlorite +kaolinite)/illite
= [PI at 12.4◦2θ]/[PI at 8.8◦2θ]
C/I=chlorite/illite = [PI at 25.1◦2θ]/[PI at 8.8◦2θ]
K/I=kaolinite/illite = [PI at 24.8◦2θ]/[PI at 8.8◦2θ].
The standard error of duplicate analyses in all samples av-
eraged 1.1, 0.08, and 0.05 for Q /F, CK /I, and C /I ratios,
respectively.
Clay minerals (less than 2 µm diameter) in core 01A-GC
were separated by the settling method based on the Stokes
law (Müller, 1967). To produce an oriented powder XRD
sample, the collected clay suspensions were vacuum-filtered
onto 0.45 µm nitrocellulose filters and dried. Ethylene glycol
(50 µL) was then soaked onto the oriented clay on the filters.
Glycolated sample filters were stored in an oven at 70◦C for
4 h and then immediately subjected to XRD analyses. Each
sample filter was placed directly on a glass slide and X-rayed
with a tube voltage of 40 kV and current of 20 mA. Scan-
ning speed was 0.5◦2θmin−1and the data-sampling step
was 0.02◦2θfrom 2 to 15◦2θ. An additional precise scan
with a scanning speed of 0.2◦2θmin−1and sampling step
of 0.01◦2θfrom 24 to 27◦2θwas conducted to distinguish
chlorite from kaolinite by evaluation of the peaks around
25.1◦2θ(Elvelhøi and Rønningsland, 1978). The standard
errors of duplicate analyses in all samples averaged 0.05 and
0.06 for CK /I and C /I ratios, respectively.
The diffraction intensity of chlorite+kaolinite at 7.1 Å
was significantly positively correlated with that of chlorite
at 3.54 Å (r=0.89) but not with that of kaolinite at 3.59 Å
(r=0.39) in western Arctic surface sediments (Kobayashi
et al., 2016), indicating that the diffraction intensity of
chlorite+kaolinite is governed by the amount of chlorite
rather than that of kaolinite.
Spectral analyses of the downcore Q/F and C /I variabil-
ity were performed using the maximum-entropy method pro-
vided in the Analyseries software package (Paillard et al.,
1996).
4 Results
4.1 Surface sediments of the Beaufort Sea
Because the dataset of Kobayashi et al. (2016) has only one
sample in the eastern Beaufort Sea, we added the data of
16 samples from the eastern Beaufort Sea near the Macken-
zie Delta and 3 samples from the western Beaufort Sea to fill
the gaps in their dataset. The new combined dataset shows
more clearly than Kobayashi et al. (2016) that the surface
sediments in the eastern Beaufort Sea have higher Q/F and
lower CK /I and C /I ratios than those in the Chukchi Sea
(Fig. 2a–c; Supplement Table S1).
The Q /F ratio showed a westward decreasing trend from
the eastern Beaufort Sea to the East Siberian Sea and its
offshore area (Fig. 2d). This supports a notion that quartz-
rich but feldspar-poor sediments are derived from the North
American margin by the BG circulation, whereas feldspar-
rich sediments are delivered to the Chukchi Sea from the
Siberian margin by currents along the East Siberian slope
(Vogt, 1997; Stein, 2008; Darby et al., 2011; Kobayashi et
al., 2016).
The CK /I and C /I ratios showed a northward decreas-
ing trend in the Chukchi Sea and the Chukchi Borderland
(Fig. 2e). These results are consistent with earlier studies
showing that illite is a common clay mineral in Arctic sedi-
ments (Kalinenko, 2001; Darby et al., 2011), whereas chlo-
rite is more abundant in the Bering Sea and the Chukchi shelf
areas influenced by the BSI (Naidu and Mowatt, 1983; Kali-
nenko, 2001; Nwaodua et al., 2014; Kobayashi et al., 2016).
These trends support the conclusion of Kobayashi et
al. (2016) mentioning that the Q /F ratio can be used as a
provenance index for the BG circulation reflecting a west-
ward decrease in its intensity, and the CK/I and C /I ra-
tios can be used as a provenance index for the BSI reflect-
ing a northward decrease in its intensity. The provenance and
transportation of these detrital minerals are discussed in de-
tail in Naidu and Mowatt (1983), Kalinenko (2001), Nwao-
dua et al. (2014), and Kobayashi et al. (2016).
Clim. Past, 13, 1111–1127, 2017 www.clim-past.net/13/1111/2017/
M. Yamamoto et al.: Holocene dynamics of western Arctic ocean circulation 1117
4.2 Cores 01A-GC, 05JPC/TC, and 06JPC
Quartz, feldspar, including plagioclase and K feldspar, illite,
chlorite, kaolinite, and dolomite were detected in the study
samples. Plagioclase comprises a variety of anorthite and al-
bite. Microscopic observations of smear slides for the study
samples revealed that quartz and feldspar are the two major
minerals in the composition of detrital grains.
The variation patterns of the Q /F, C /I, CK /I, and K /I
ratios are different for fine-grained and glaciomarine units in
cores 05JPC/TC and 06JPC (Fig. 3; Supplement Tables S3–
S5). The ratios of the fine-grained unit are relatively sta-
ble compared with those in glaciomarine units. The higher
Q/F ratio in glaciomarine units is consistent with the find-
ing of previous studies that quartz grains are abundant in
the western Arctic sediments delivered from the Laurentide
ice sheet during glacial and deglacial periods (Bischof et al.,
1996; Bischof and Darby, 1997; Phillips and Grantz, 2001;
Kobayashi et al., 2016). Some peaks correspond to dolomite-
rich layers (“D” in Fig. 3). Variation in the K /I ratio was
associated with that in the Q /F ratio (Fig. 3), which is in
harmony with an idea that kaolinite was delivered via the
Beaufort Gyre circulation (Kobayashi et al., 2016). The C/I
and CK /I ratios are lower in the glaciomarine unit than in
the fine-grained unit in 06JPC (Fig. 3c), which is consistent
with the closure of the Bering Strait in the last glacial (Elias
et al., 1992), but this difference is not significant in 05JPC
(Fig. 3b). High-amplitude fluctuations were observed in the
C/I and CK /I ratios in the fine-grained sediments in 01A-
GC and 06JPC (Fig. 3a and c). Similar fluctuations partly
appeared in 05JPC/TC despite its lower sampling resolution
(Fig. 3b).
The Q /F ratio in cores 01A-GC, 05JPC/TC, and 06JPC
shows a gradual long-term decrease throughout the Holocene
(Fig. 4a). In cores 01A-GC and 06JPC studied in more de-
tail, the Q /F ratio also indicates millennial- to century-scale
variability (Fig. 4a). Variations in the five-point running av-
erage highlight millennial-scale patterns (Fig. 4a). The varia-
tions are generally asynchronous between both cores on this
timescale, which strongly depends on their age–depth mod-
els.
In core 01A-GC, the CK /I and C /I ratios show a gen-
eral increase after ca. 9.5 ka with the highest values occur-
ring between 6 and 4 ka, and high ratios around 2.5 and
1 ka (Fig. 4b). In core 06JPC, the ratios show a general in-
crease after 9.2 ka, with higher values occurring between 6
and 3 ka (Fig. 4b). In core 05JPC/TC, slightly higher ratios
occur between 6 and 3 ka after a gradual increase from 9.3ka
(Fig. 4b).
400 8000 1200 1600
400 8000 1200 1600
Depth (cm)
Depth (cm)
1477 4656 5110 5584 6304 6871 Age (cal yr BP)
2030 3580 5730 6330 7868 Age (cal yr BP)
2150 6000
20
10
1.0
0
0.5
1.5
RatioClay ratio
30
20
40
10
1.0
0
0.5
1.5
RatioClay ratio
30
Fine-grained unit Glaciomarine unit
Fine-grained unit Glaciomarine unit
05JPC/TC
06JPC
(b)
(c)
Q/F
CK/I
C/I
K/I
Q/F
CK/I
C/I
K/I
D
D
Intensity (cps)
0
1000
2000
Intensity (cps)
1000
2000
0
200 3000 100 Depth (cm)
3057 3842 4482 4893 5991 7644 8642 Age (calendar year BP)
20
0
10
1.0
0
0.5
1.5
RatioClay ratio
30
01A-GC
(a)
Q/F
CK/I
C/I
K/I
600500400
Figure 3. Depth profile in the quartz /feldspar (Q /F) ratio, (chlo-
rite +kaolinite) /illite (CK /I) and the chlorite /illite (C /I) and
kaolinite /illite (K /I) ratios with 1σintervals (analytical error)
and the diffraction intensity of dolomite (D) in cores (a) ARA02B
01A-GC, (b) HLY0501-05JPC/TC, and (c) HLY0501-06JPC (Sup-
plement Tables S2–S4). Crosses indicate radiocarbon dates in 01-
GC and 5JPC and paleo-intensity datums in 06JPC. Open circles
in panel (b) indicate 05TC samples. Note that the depth scale for
01A-GC is doubled for presentation purposes.
www.clim-past.net/13/1111/2017/ Clim. Past, 13, 1111–1127, 2017
1118 M. Yamamoto et al.: Holocene dynamics of western Arctic ocean circulation
510
570
550
530
12
18
16
14
8
10
12
6
10
8
12
10
01A-GC
05JPC/TC
06JPC
Q/FQ/F
Q/F Insolation (W m-2)
21 June at 75° N
1.0
1.4
1.2
CK/I
1.6
1.8
1.4
1.2
CK/I
1.6
1.0
1.4
1.2
CK/I
1.0
0.6
0.8
C/I
0.6
C/I
0.8
0.6
0.8
C/I
01A-GC
05JPC/TC
06JPC
0
200
100
LSR (cm kyr )
-1
400
06JPC
05JPC/TC
01A-GC
0 642 10
8
Age (ka)
(a)
(b)
(c)
Intense BSI
Intense BG
8
Figure 4. Holocene changes in (a) the quartz /feldspar (Q /F) ra-
tio and the June insolation at 75◦N, (b) (chlorite +kaolinite) /illite
(CK /I) and chlorite /illite (C /I) ratios, and (c) linear sedimenta-
tion rates (LSRs) between age tie points in cores ARA02B 01A-
GC, HLY0501-05JPC/TC, and HLY0501-06JPC. Note that the age
model for 06JPC is very tentative, so that a peak in LSR at ca. 2ka
could be an artifact of spurious age controls.
5 Discussion
5.1 Holocene trend in the Beaufort Gyre circulation
The zonal gradient of the Q /F ratio in western Arctic sedi-
ments shown in Fig. 2 suggests that quartz-rich but feldspar-
poor sediments are derived from the North American mar-
gin by the BG circulation, whereas feldspar-rich sediments
are delivered to the Chukchi Sea from the Siberian mar-
gin by currents along the East Siberian slope, and the ratio
can be used as an index for the BG circulation reflecting
changes in its intensity in sediment-core records (Kobayashi
et al., 2016). A consistent decrease in the Q /F ratio in the
three different cores under study (Fig. 4a) suggests that the
BG weakened during the Holocene. This pattern is consis-
tent with an orbitally forced decrease in summer insolation
at northern high latitudes from the early Holocene to the
present. High summer insolation likely melted sea ice in the
Canada Basin, in particular in the coastal areas (Fig. 5). The
evidence of lower ice concentrations at the Canada Basin
margins in the early Holocene has been shown in the fos-
sil records of bowhead whale bones from the Beaufort Sea
coast (Dyke and Savelle, 2001) and driftwood from north-
ern Greenland (Funder et al., 2011). This condition could
decrease the stability of the ice cover at the margins of the
Canada Basin, which accelerated the rotation of the BG cir-
culation (Fig. 5) by comparison with observations from re-
cent decades (Shimada et al., 2006). A decrease in summer
insolation during the Holocene should have increased the sta-
bility of sea-ice cover along the coasts, resulting in the weak-
ening of the BG.
Recent observations show that the BG circulation is linked
to the AO (Proshutinsky and Johnson, 1997; Rigor et al.,
2002). In the negative phase of the AO, the Beaufort High
strengthens and intensifies the BG. If the gradual weaken-
ing of the BG during the Holocene were attributed to atmo-
spheric circulation only, a concurrent shift in the mean state
of the AO from the negative to positive phase would be ex-
pected. This view, however, contradicts the existing recon-
structions of the AO history showing multiple shifts between
the positive and negative phases during the Holocene (e.g.,
Rimbu et al., 2003; Olsen et al., 2012). We thus infer that the
decreasing Holocene trend of the BG circulation is attributed
not to changes in the AO pattern but rather to the increasing
stability of the sea-ice cover in the Canada Basin.
Based on a Holocene sediment record off the northeast-
ern Chukchi margin, Darby et al. (2012) suggested strong
positive AO-like conditions between 3 and 1.2 ka based
on abundant ice-rafted iron oxide grains from the West
Siberian shelf. In contrast, a mostly negative AO in the late
Holocene can be inferred from mineralogical proxy data in-
dicating a general decline in the BSI after 4 ka (Ortiz et al.,
2009), which could be attributed to a stronger Aleutian Low
(Danielson et al., 2014) that typically corresponds to the neg-
ative AO (Overland et al., 1999). Olsen et al. (2012) also con-
cluded that the AO tended to be mostly negative from 4.2 to
2.0 ka based on a redox proxy record from a Greenland lake.
In order to comprehend these patterns, we need to consider
not only the atmospheric circulation but also sea-ice condi-
tions. Based on the Q /F record in this study, summer Arctic
sea-ice cover shrank in the early to middle Holocene, so that
fast ice containing West Siberian grains could less effectively
reach the Canada Basin because sea ice would have melted
on the way to the BG (Fig. 5). Later in the Holocene the ice
cover expanded, and West Siberian fast ice could survive and
be incorporated into the BG (Fig. 5). We infer, therefore, that
sediment transportation in the BG is principally governed by
the distribution of summer sea ice and the resultant stability
of the ice cover in the Canada Basin.
Clim. Past, 13, 1111–1127, 2017 www.clim-past.net/13/1111/2017/
M. Yamamoto et al.: Holocene dynamics of western Arctic ocean circulation 1119
Early&Holocene&
(9.3–8 ka)
High%summer%insola/on
Middle&Holocene&
(8–3.5 ka)
Medium%summer%insola/on
Late&Holocene&
(after 3.5 ka)
Low%summer%insola/on
Weaker&BSIStronger&BSI
Stronger&BGWeaker&BG
Medium&BSI
Medium&BG
KSGKSGKSG
Figure 5. Conceptual map showing the distribution of summer sea ice and the rotation of the Beaufort Gyre (BG) in the early, middle, and
late Holocene, inferred from the quartz /feldspar (Q /F) proxy record. Also shown is the Bering Strait inflow (BSI) intensity inferred from
the (chlorite +kaolinite) /illite and chlorite /illite ratios. Red arrow indicates the drift path of Kara Sea grains (KSGs; Darby et al., 2012).
5.2 Millennial variability in the BG circulation
In addition to the decreasing long-term trend, the Q /F ra-
tio in 01A-GC and 06JPC clearly displays millennial- to
century-scale variability (Fig. 4a). Variation in the Q /F ra-
tio of both 01A-GC and 06JPC indicates a significant pe-
riodicity of ∼2100 and ∼1000 years with weak periodic-
ities of ∼500 and ∼360 years, consistent with prominent
periodicities in the variation in total solar irradiance (Fig. 6)
(Steinhilber et al., 2009). A comparison with the record of
total solar irradiance (Steinhilber et al., 2009) shows a gen-
eral correspondence, where stronger BG circulation (higher
Q/F ratio) corresponds to higher solar irradiance (Fig. 7).
A∼200-year phase lag between the solar irradiance and the
Q/F ratio in 01A-GC and 06JPC may be attributed to the
underestimation of the local carbon reservoir effect. This pat-
tern suggests that millennial-scale variability in the BG was
principally forced by changes in solar irradiance as the most
likely forcing. Proxy records consistent with solar forcing
were reported from a number of paleoclimatic archives, such
as Chinese stalagmites (Hu et al., 2008), Yukon lake sedi-
ments (Anderson et al., 2005), and ice cores (Fisher et al.,
2008), as well as marine sediments in the northwestern Pa-
cific (Sagawa et al., 2014) and the Chukchi Sea (Stein et al.,
2017). Because solar forcing is energetically much smaller
than changes in the summer insolation caused by orbital forc-
ing, we suppose that solar activity did not directly affect the
stability of ice cover in the Canada Basin. Alternatively, we
suggest that the solar activity signal was amplified by posi-
tive feedback mechanisms, possibly through changes in the
stability of sea-ice cover and/or the atmospheric circulation
in the northern high latitudes.
In addition to cycles consistent with the solar forcing,
Darby et al. (2012) reported a 1550-year cycle in the Siberian
grain variation in the Chukchi Sea record. This cycle was,
Spectral densitySpectral densitySpectral density
Spectral densitySpectral density
0
4
3
2
1
0
0.03
0.02
0.01
0
0.02
0.01
0
0.002
0.001
0
0.004
0.006
0.002
2.1
0.76
0.98
0.50 0.36
2.3 0.96
0.50 0.38
2.1 0.98
0.51 0.36
0.83
2.2
0.44
1.9 0.51
1.0
0.40
0.32
C/I
log Q/F
C/I
log Q/F
ΔTSI
0 1 2 3
Frequency (kyr
-1
)
01A-GC
6JPC
Figure 6. Maximum-entropy power spectra of variation in the
quartz /feldspar (Q /F) and chlorite /illite (C /I) ratios in core
ARA02B 01A-GC (N=85, m=21) and HYL0501-06JPC (N=
79, m=22) during 1.4–7.9 ka and the total solar irradiance (N=
932, m=140) (Steinhilber et al., 2009) during the last 9.3 kyr.
www.clim-past.net/13/1111/2017/ Clim. Past, 13, 1111–1127, 2017
1120 M. Yamamoto et al.: Holocene dynamics of western Arctic ocean circulation
-0.3%
-0.25%
-0.2%
-0.15%
-0.1%
-0.05%
0%
0.05%
0.1%
-0.3%
-0.1%
0.1%
0.3%
0.5%
0.7%
0% 1% 2% 3% 4% 5% 6% 7% 8% 9% 10%
log$Q/F$
C/I$
Age$(ka)$
-2#
-1.5#
-1#
-0.5#
0#
0.5#
1#
-0.2#
-0.1#
0#
0.1#
0.2#
0.3#
0.4#
0.5#
0# 1# 2# 3# 4# 5# 6# 7# 8# 9# 10#
!ΔTSI!(W!m-2)!
!log!Q/F!
Age!(ka)!
Detrended'ΔTSI
Detrended'log'Q/F'in'01A-GC
Detrended'log'Q/F'in'6JPC
Detrended'C/I'in'01A-GC
Figure 7. Detrended variations in the solar irradiance (TSI; Stein-
hilber et al., 2009), the quartz /feldspar (Q /F) ratio in logarith-
mic scale in cores ARA02B 01A-GC and HYL0501-06JPC, and
the chlorite /illite (C /I) ratio in core ARA02B 01A-GC during the
Holocene, with 400-year moving averages and 1000-year filtered
variations indicated by dark colored and black lines, respectively.
The detrended values were obtained by cubic polynomial regres-
sion.
however, not detected in our data, indicative of the BG varia-
tion (Fig. 6). This difference suggests that the occurrence of
Siberian grains in the Chukchi Sea sediments primarily re-
flects the formation and transportation of fast ice in the east-
ern Arctic Ocean rather than changes in the BG circulation.
5.3 Holocene changes in the Bering Strait inflow
Northward decreasing trends in the CK /I and C /I ratios in
surface sediments in the Chukchi Sea suggests that chlorite-
rich sediments are derived from the northern Bering Sea via
the Bering Strait, and the ratios can be used as an index
for the BSI reflecting changes in its intensity in sediment-
core records (Kobayashi et al., 2016). Although the varia-
tions in the CK /I and C /I ratios are not identical among
the three study cores (Fig. 4b), there is a common long-term
trend showing a gradual increase from 9 to 4.5ka and a de-
crease afterwards (Fig. 4b). Large fluctuations are significant
in 01A-GC from 6 to 4 ka, and this fluctuation is also seen in
6JPC to some extent (Fig. 4b).
The higher CK /I and C /I ratios in core 01A-GC in the
middle Holocene correspond to higher linear sedimentation
rates estimated by interpolation between 14C dating points,
but this correspondence is not seen in cores 05JPC/TC and
06JPC (Fig. 4c). We assume that these higher sedimentation
rates at 01A-GC indicate intensified BSI because fine sedi-
ment in the study area is mostly transported by currents from
the Bering Sea and shallow southern Chukchi shelf (Kali-
nenko, 2001; Darby et al., 2009; Kobayashi et al., 2016).
The difference in chlorite and sedimentation rate records be-
tween 01A-GC and 05JPC/06JPC may be related to either
(1) variable sediment focusing at different water depths or
(2) redistribution of the BSI water between different branches
after passing the Bering Strait. (1) A sediment-trap study
demonstrated that shelf-break eddies in winter are important
to carry fine-grained lithogenic material from the Chukchi
shelf to the slope areas (Watanabe et al., 2014). This rede-
position process may have weakened the BSI signal in slope
sediments of 05JPC/06JPC compared with outer-shelf sedi-
ments of 01A-GC. (2) Both the ACC and the central current
can transport sediment particles to the 05JPC/TC and 06JPC
area (red and yellow arrows, respectively, in Fig. 1; Winsor
and Chapman, 2004; Weingartner et al., 2005). In compar-
ison, the western branch is more likely to carry sediment
particles to the site of 01A-GC (blue arrow in Fig. 1). The
redistribution of the BSI water may have caused a different
response in BSI signals. Although it is not clear which pro-
cess made the difference to BSI signals between 01A-GC and
05JPC/06JPC cores, it is highly possible that the sedimenta-
tion rate and mineral composition of 01A-GC are more sen-
sitive to changes in BSI intensity than those of the other two
sites.
Diffuse spectral reflectance in core HLY0501-06JPC in-
dicated that chlorite +muscovite content is especially high
in the middle Holocene between ca. 4 and 6 ka (Supplement
Fig. S1; Ortiz et al., 2009). However, this pattern was not
confirmed by our XRD analysis, where XRD intensities of
chlorite and muscovite (detected as illite in this study) as well
as the C /I and CK /I ratios did not show an identifiable en-
richment between 4 and 6 ka (Supplement Fig. S1). We need
more research to understand the discrepancy between the re-
sults.
5.4 Millennial variability in the BSI
Variation in the C /I ratio of 01A-GC indicates a significant
periodicity of 1900, 1000, 510, 400 and 320 years (Fig. 6a).
The 1900-, 1000-, and 510-year periodicities are consistent
with prominent periodicities in the variation in total solar
irradiance (Fig. 6c) (Steinhilber et al., 2009). On the other
hand, variation in the C /I ratio of 06JPC indicates a period-
icity of 2200, 830, and 440 years (Fig. 6b). The periodicity
is different from that in 01A-GC (Fig. 6a). This suggests that
there are different agents of BSI signals in cores 01A-GC
and 06JPC. In core 01A-GC, 1000-year filtered variation in
the C /I ratio is nearly antiphase with those of the Q /F ratio
and total solar irradiance (Steinhilber et al., 2009) between 0
and 5 ka (Fig. 7). This suggests that millennial-scale variabil-
Clim. Past, 13, 1111–1127, 2017 www.clim-past.net/13/1111/2017/
M. Yamamoto et al.: Holocene dynamics of western Arctic ocean circulation 1121
ity in the western branch of the BSI was forced by changes in
solar irradiance after 5 ka. Recent observations demonstrated
that the BSI flows northwestward, especially when easterly
winds prevent the ACC (Winsor and Chapman, 2004). Be-
cause the easterly winds drive the BG circulation, this mech-
anism cannot explain the increase in BSI intensity when the
BG weakened. Alternatively, it is also possible that the solar
forcing could independently regulate the western branch of
the BSI via unknown atmospheric–oceanic dynamics.
5.5 Ocean circulation, sea ice, and biological production
The BSI, an important carrier of heat to the Arctic, af-
fects sea-ice extent in the Chukchi Sea (e.g., Shimada et
al., 2006). Sea-ice concentrations in the Chukchi Sea dur-
ing the Holocene were reconstructed by dinoflagellate cysts
(de Vernal et al., 2005, 2008, 2013; Farmer et al., 2011) and
biomarker IP25 (Polyak et al., 2016; Stein et al., 2017).
In central northern Chukchi Sea, IP25 records showed that
the sea-ice concentration indicated by the PIP25 index in core
01A-GC was lower in 9–7.5 and 5.5–4ka (Fig. 8a; Stein et
al., 2017), suggesting less sea-ice conditions in the periods.
The low sea-ice concentration during 9–7.5 ka is consistent
with the results of previous studies based on dinoflagellate
cyst and IP25 records showing the widespread sea-ice retreat
in the Arctic Ocean, which was attributed to higher sum-
mer insolation during the early Holocene (Dyke and Savelle,
2001; Vare et al., 2009; de Vernal et al., 2013; Stein et al.,
2017). On the other hand, the sea-ice retreat during 5.5–4 ka
cannot be explained by higher summer insolation. This pe-
riod corresponds to that of higher C /I and CK /I ratios in-
dicative of the stronger BSI at 01A-GC (Fig. 8a). This sug-
gests that the strengthened BSI during this period contributed
to sea-ice retreat in the central Chukchi Sea.
In the northeastern Chukchi Sea, dinoflagellate cyst and
biomarker IP25 records from several cores in the northeast-
ern Chukchi Sea, including 05JPC, demonstrate that sea-ice
concentration in this area was overall higher in the early
Holocene than in the middle and late Holocene (Fig. 8; de
Vernal et al., 2005, 2008, 2013; Farmer et al., 2011; Polyak
et al., 2016). This pattern contrasts with reconstructions from
other Arctic regions that show lower sea-ice concentrations
in the early Holocene (de Vernal et al., 2013). This discrep-
ancy suggests that the intensified BG circulation exported
more ice from the Beaufort Sea to the northeastern Chukchi
Sea margin. Furthermore, the heat transport from the North
Pacific to the Arctic Ocean by the BSI was likely weaker
in the early Holocene than at later times as indicated by the
C/I and CK /I ratios of cores 06JPC and 01A-GC (Fig. 8).
We infer that this combination of stronger BG circulation
and weaker BSI in the early Holocene resulted in increased
sea-ice concentration in the northeastern Chukchi Sea de-
spite high insolation levels (Fig. 5). In comparison, intense
BSI, a crucial agent of heat transport from the North Pacific
to the Arctic Ocean, along with a weaker BG in the middle
1.0
1.4
1.2
CK/I
1.6
1.8 1.0
0.6
0.8
C/I
01A-GC
0 642 10
8
Age (ka)
(a)
0.7
0.3
0.5
20
0
10
30 1.0
0
0.5
PIP25
IsoGDGTs (μg g-1)
Brassicasterol (μg g-1)
PBIP25
PDIP25
Central northern Chukchi Sea
(b) Northeastern Chukchi Sea
1.4
1.2
CK/I
1.6
1.0
1.4
1.2
CK/I
0.6
C/I
0.8
0.6
0.8
C/I
05JPC/TC
06JPC
8
0
6
2
4
05JPC
GGC19
Ice concentration (0–10)
More ice
More ice
0.02
0.01
0
IP
25
(μg cm
-3
)
05JPC/TC
High production
More ice
Intense BSI
Figure 8. Changes in (a) (chlorite +kaolinite) /illite (CK /I) and
chlorite /illite (C /I) ratios, PIP25 (PDIP25 and PBIP25 based on
IP25 and dinosterol or brassicasterol concentrations) indices (Stein
et al., 2017), and isoprenoid GDGT (Park et al., 2016) and brassi-
casterol concentrations (Stein et al., 2017) in core ARA02B 01A-
GC; (b) CK /I and C /I ratios in core HLY0510-5JPC/TC, IP25
concentrations in core HLY0510-5JPC (Polyak et al., 2016), and
mean annual sea ice cover concentration (scale from 0 to 10) es-
timated from dinoflagellate cyst assemblages in cores 05JPC and
GGC19 (de Vernal et al., 2013).
Holocene likely reduced sea-ice cover in the Chukchi Sea.
During the late Holocene, characterized by the weakest BG
and moderate BSI, sea-ice concentrations were intermediate
and strongly variable (Fig. 8; de Vernal et al., 2008, 2013;
Polyak et al., 2016).
The nutrient supply by the BSI potentially affects marine
production in the Chukchi Sea. We tested this possibility to
www.clim-past.net/13/1111/2017/ Clim. Past, 13, 1111–1127, 2017
1122 M. Yamamoto et al.: Holocene dynamics of western Arctic ocean circulation
compare our BSI record with marine production records from
cores 01A-GC (Park et al., 2016; Stein et al., 2017). Iso-
prenoid glycerol dialkyl glycerol tetraethers (GDGTs) and
brassicasterol showed concentration maxima during the pe-
riods between 8 and 7.5 ka and 6 and 4.5 ka (Fig. 8a). Iso-
prenoid GDGTs are produced by marine Archaea (Nishihara
et al., 1987) that use ammonia, urea, and organic matter in
the water column (Qin et al., 2014). Brassicasterol is known
as a sterol which is abundant in diatoms (Volkman, 1986).
Their abundance can, thus, be used as proxies to indicate ma-
rine production in the water column. The periods with abun-
dant isoprenoid GDGTs and brassicasterol corresponded to
the periods of low PIP25 indicative of less sea ice (Fig. 8a).
This correspondence suggests that the biological productiv-
ity increased with the retreat of sea ice in the Chukchi Sea
during the middle Holocene. The BSI indices, the C /I and
CK /I ratios, showed a maximum between 6 and 4ka, which
corresponded to the periods of high marine production, but
the corresponding maximum between 8 and 6.5 ka is not sig-
nificant. Also, correspondence between the BSI indices and
biomarker concentrations are not clear after 4 ka. This sug-
gests that marine production was not a simple response to
nutrient supply but was affected by other processes such as
the increase in irradiance in the water column (Frey et al.,
2011; Lee and Whitledge, 2005) and wind-induced mixing
that replenishes sea surface nutrients (Carmack et al., 2006).
5.6 Causes of BSI variations
Chukchi Sea sedimentary core records indicate a consider-
able variability in the BSI intensity, with a common long-
term trend of a gradual increase from 9 to 4.5 ka and a de-
crease afterwards (Fig. 4b). Below we discuss the possible
controls on this variability.
The timing of the initial postglacial flooding of the ∼50 m
deep Bering Strait was estimated as between ca. 12 and 11 ka
(Elias et al., 1992; Keigwin et al., 2006; Jakobsson et al.,
2017). The gradual intensification of the BSI inferred from
the increase in chlorite content from ca. 9 to 6 ka may have
been largely controlled by the widening and deepening of the
Bering Strait with rising sea level, although other factors as
discussed below still need to be tested. After the sea level
rose to nearly the present position by ca. 6 ka, its influence
on changes in the BSI volume was negligible.
The possible driving forces of the BSI at full interglacial
sea level may include several controls. One is related to the
sea surface height difference between the Pacific and Atlantic
oceans regulated by the atmospheric moisture transport from
the Atlantic to the Pacific Ocean across Central America
(Stigebrandt, 1984). An increase in this moisture transport
during warm climatic intervals (Leduc et al., 2007; Richter
and Xie, 2010; Singh et al., 2016) may have intensified the
BSI. Salinity proxy data for the last 90 kyr from the equato-
rial east Pacific confirm increased precipitation during warm
events but also show the moisture transport across Central
America may operate efficiently only during intervals with a
northerly position of the Intertropical Convergence Zone due
to orographic constraints (Leduc et al., 2007). The existing
Holocene salinity records from the North Pacific (e.g., Sarn-
thein et al., 2004) do not yet provide sufficient material to
test the impact of these changes on the BSI.
The interplay of the global wind field and the Atlantic
meridional overturning circulation has been proposed as an-
other potential control on the BSI (De Boer and Nof, 2004;
Ortiz et al., 2012). Results of an analytical ocean model-
ing experiment (Sandal and Nof, 2008) based on the island
rule (Godfrey, 1989) suggest that weaker subantarctic wester-
lies in the middle Holocene could decrease the near-surface,
cross-equatorial flow from the Southern Ocean to the North
Atlantic, thus enhancing the BSI and Arctic outflow into the
Atlantic. This hypothesis awaits more thorough testing, in-
cluding by robust proxy records of the subantarctic wester-
lies over the Southern Ocean.
Finally, BSI can be controlled by the regional wind pat-
terns in the Bering Sea (Danielson et al., 2014), as explained
above in Sect. 2.1. Oceanographic observations of 2000–
2011 clearly show a decadal response of the BSI to a change
in the sea level pressure in the Aleutian Basin affecting the
dynamic sea surface height along the Bering Strait pressure
gradient. In order to draw conclusions on whether this rela-
tionship holds on longer timescales, longer-term records are
needed from areas affected by the BSI and the Bering Sea
pressure system.
A number of proxy records from the Bering Sea and ad-
jacent regions, both marine and terrestrial, have been used
to characterize paleoclimatic conditions related to changes
in the Bering Sea pressure system (e.g., Barron et al., 2003;
Anderson et al., 2005; Katsuki et al., 2009; Barron and An-
derson, 2011; Osterberg et al., 2014). Various proxies used
in these records consistently show that the Aleutian Low
was overall weaker in the middle Holocene than in the late
Holocene, the opposite of the BSI strength inferred from our
Chukchi Sea data (Fig. 4b). For example, multi-proxy data
from the interior of Alaska and adjacent territories (Kaufman
et al., 2016, and references therein) indicate overall drier and
warmer conditions in the middle Holocene, consistent with
a weaker Aleutian Low and stronger BSI. Diatom records
from the southern Bering Sea indicate more abundant sea ice
in the middle Holocene, also suggestive of a weaker Aleu-
tian Low (Katsuki et al., 2009). Alkenone and diatom records
from the California margin show that the sea surface temper-
ature was lower in the middle Holocene, suggesting stronger
northerly winds indicative of a weaker Aleutian Low (Bar-
ron et al., 2003). The intensification of the Aleutian Low in
the late Holocene, which follows from these results, would
have decreased sea level pressure in the Aleutian Basin, and
thus the strength of the BSI, consistent with overall lower
BSI after ca. 4 ka inferred from the Chukchi Sea sediment-
core data (Fig. 4). The considerable climate variability in the
Bering Sea region captured in the upper Holocene records,
Clim. Past, 13, 1111–1127, 2017 www.clim-past.net/13/1111/2017/
M. Yamamoto et al.: Holocene dynamics of western Arctic ocean circulation 1123
Table 1. Summary of Holocene variability in the BG and BSI in the northern Chukchi Sea.
Current system Holocene trends Multi-centennial to millennial cyclicity
Beaufort Gyre
circulation
Gradual weakening in response to decreas-
ing summer insolation.
0.36, 0.5, 1, and 2 ky cycles paced by
changes in solar activity.
Bering Strait inflow Geographically variable. Mid-Holocene
strengthening evident at the 01A-GC site,
presumably due to weaker Aleutian Low.
Geographically variable; ∼0.36, 0.5, 1,
and 2 kyr cycles paced by changes in so-
lar activity are identifiable in 01A-GC.
some of which have a very high temporal resolution, is also
closely linked to the pressure system changes (Anderson et
al., 2005; Porter, 2013; Osterberg et al., 2014; Steinman et
al., 2014). In particular, the weakening of the Aleutian Low is
reflected in Alaskan ice (Porter, 2013; Osterberg et al., 2014)
and lake cores (Anderson et al., 2005; Steinman et al., 2014)
at intervals centered around ca. 2 and 1–0.5ka, which may
correspond to BSI increases in the Chukchi core 01A-GC at
ca. 2.5 and 1 ka (Fig. 4), considering the uncertainties of the
sparse age constraints in the upper Holocene and/or underes-
timation of reservoir ages. Overall, the Aleutian Low control
on the BSI on century to millennial timescales is corrobo-
rated by ample proxy data in comparison with the other po-
tential controls, although more evidence is still required for
a comprehensive interpretation.
6 Summary and conclusions
The distribution of minerals in surficial bottom sediments
from the Chukchi Sea shows two distinct trends: an east–
west gradient in quartz /feldspar ratios along the shelf mar-
gin and a northwards decrease in the chlorite contents. These
trends are consistent with the propagation of the Beaufort
Gyre circulation in the western Arctic Ocean and the Bering
Strait inflow to the Chukchi Sea, respectively. The appli-
cation of these lithological proxies to sedimentary records
from the north–central and northeastern parts of the Chukchi
Sea allows for an identification of the Holocene paleo-
oceanographic patterns with century to millennial resolution.
Results of the identified Holocene changes in the BG circu-
lation and the BSI are summarized in Table 1.
The inferred BG weakening during the Holocene, likely
driven by the orbitally controlled summer insolation de-
crease, indicates basin-wide changes in the Arctic current
system and suggests that the stability of sea ice is a key fac-
tor regulating the Arctic Ocean circulation on the long-term
(e.g., millennial) timescales. This conclusion helps to better
understand a dramatic change in the BG circulation during
the last decade, probably caused by sea-ice retreat along the
margin of the Canada Basin and a more efficient transfer of
the wind momentum to the ice and underlying waters (Shi-
mada et al., 2006). These results suggest that the rotation of
the BG is likely to be further accelerated by the projected
future retreat of summer Arctic sea ice.
The identified millennial to multi-centennial variability in
the BG circulation (quartz /feldspar ratio) is consistent with
Holocene fluctuations in solar irradiance, suggesting that so-
lar activity affected the BG strength on these timescales.
Changes in the BSI inferred from the proxy records show
a considerable variability between the investigated sedi-
ment cores, likely related to interactions of different current
branches and depositional processes. Overall, we conclude
that after the establishment of the full interglacial sea level
in the early Holocene, the BSI variability was largely con-
trolled by the Bering Sea pressure system (strength and po-
sition of the Aleutian Low). Details of this mechanism, as
well as contributions from other potential BSI controls, such
as climatically driven Atlantic–Pacific moisture transfer and
the impact of global wind stress, need to be further inves-
tigated. A consistent intensification of the BSI identified in
the middle Holocene was associated with a decrease in sea-
ice extent and an increase in marine production, indicating a
major influence of the BSI on sea ice and biological activ-
ity in the Chukchi Sea. In addition, multi-century to millen-
nial fluctuations, presumably controlled by solar activity, are
discernible in core 01A-GC, which is characterized by the
highest age resolution.
Data availability. All data are shown in the Supplement.
The Supplement related to this article is available online
at https://doi.org/10.5194/cp-13-1111-2017-supplement.
Competing interests. The authors declare that they have no con-
flict of interest.
Special issue statement. This article is part of the special is-
sue “Climate–carbon–cryosphere interactions in the East Siberian
Arctic Ocean: past, present and future (TC/BG/CP/OS inter-journal
SI)”. It is not associated with a conference.
Acknowledgements. We thank the captain, crew, and scientists
of RV Araon for their help during the sampling cruise. We
www.clim-past.net/13/1111/2017/ Clim. Past, 13, 1111–1127, 2017
1124 M. Yamamoto et al.: Holocene dynamics of western Arctic ocean circulation
also thank Yu-Hyeon Park, Anne de Vernal, Seth L. Danielson,
Julie Brigham-Grette, and Kaustubh Thirumalai for valuable dis-
cussion; So-Young Kim, Hyo-Sun Ji, Yeong-Ju Son, Duk-Ki Han,
and Hyoung-Jun Kim for assistance in coring and subsampling;
and Keiko Ohnishi for analytical assistance. Comments by Mar-
tin Jakobsson, Tomas M. Cronin, and an anonymous reviewer
greatly improved this paper. The study was supported by a Grant-
in-Aid for Scientific Research (B) from the Japan Society for the
Promotion of Science, No. 25287136 (to Masanobu Yamamoto), a
Basic Research Project (PE16062) of the Korean Polar Research
Institute, and the NRF of Korea Grant funded by the Korean
Government (NRF-2015M1A5A1037243) (to Seung-Il Nam).
Edited by: Martin Jakobsson
Reviewed by: Thomas M. Cronin and one anonymous referee
References
Aagaard, K., Weingartner, T. J., Danielson, S. L., Woodgate, R. A.,
Johnson, G. C., and Whitledge, T. E.: Some controls on flow
and salinity in Bering Strait, Geophys. Res. Lett., 33, L19602,
https://doi.org/10.1029/2006GL026612, 2006.
Anderson, L., Abbott, M. B., Finney, B. P., and Burns, S. J.: Re-
gional atmospheric circulation change in the North Pacific during
the Holocene inferred from lacustrine carbonate oxygen isotopes,
Yukon Territory, Canada, Quaternary Res., 64, 21–35, 2005.
Barletta, F., St-Onge, G., Channell, J. E. T., Rochon, A., Polyak, L.,
and Darby, D.: High resolution paleomagnetic secular variation
and relative paleointensity records from the western Canadian
Arctic: implication for Holocene stratigraphy and geomagnetic
field behaviour, Can. J. Earth Sci., 45, 1265–1281, 2008.
Barron, J. A. and Anderson, L.: Enhanced Late Holocene
ENSO/PDO expression along the margins of the eastern North
Pacific, Quatern. Int., 235, 3–12, 2011.
Barron, J. A., Heusser, L., Herbert, T., and Lyle, M.: High-
resolution climatic evolution of coastal northern California
during the past 16,000 years, Paleoceanography, 18, 1020,
https://doi.org/10.1029/2002PA000768, 2003.
Biscaye, P.: Mineralogy and sedimentation of recent deep-sea clay
in the Atlantic Ocean and adjacent seas and oceans, Geol. Soc.
Am. Bull., 76, 803–832, 1965.
Bischof, J. and Darby, D. A.: Mid- to Late Pleistocene ice drift in
the western Arctic Ocean: Evidence for a different circulation in
the Past, Science, 277, 74–78, 1997.
Bischof, J., Clark, D. L., and Vincent, J. S.: Origin of ice rafted de-
bris: Pleistocene paleoceanography in the western Arctic Ocean,
Paleoceanography, 11, 743–756, 1996.
Carmack, E., Barber, D., Christensen, J., Macdonald, R., Rudels, B.,
and Sakshaug, E.: Climate variability and physical forcing of the
food webs and the carbon budget on pan-Arctic shelves, Prog.
Oceanogr., 71, 145–181, 2006.
Coachman, L. K. and Aagaard, K.: On the water exchange through
Bering Strait, Limnol. Oceanogr., 11, 44–59, 1966.
Danielson, S. L., Weingartner, T. J., Hedstrom, K. S., Aargaard,
K., Woodgate, R., Curchister, E., and Stabeno, P. J.: Coupled
wind-forced controls of the Bering-Chukchi shelf circulation and
the Bering Strait throughflow: Ekman transport, continental shelf
waves, and variations of the Pacific-Arctic sea surface height gra-
dient, Prog. Oceanogr., 125, 40–61, 2014.
Darby, D. A., Ortiz, J. D., Polyak, L., Lund, S., Jakobsson, M., and
Woodgate, R. A.: The role of currents and sea ice in both slowly
deposited central Arctic and rapidly deposited Chukchi-Alaskan
margin sediments, Global Planet. Change, 68, 58–72, 2009.
Darby, D. A., Myers, W. B., Jakobsson, M., and Rigor, I.:
Modern dirty sea ice characteristic and sources: The
role of anchor ice, J. Geophys. Res., 116, C09008,
https://doi.org/10.1029/2010JC006675, 2011.
Darby, D. A., Ortiz, J. D., Grosch, C. E., and Lund, S. P.: 1,500-
year cycle in the Arctic Oscillation identified in Holocene Arctic
sea-ice drift, Nat. Geosci., 5, 897–900, 2012.
De Boer, A. M. and Nof, D.: The exhaust valve of the North At-
lantic, J. Climate, 17, 417–422, 2004.
de Vernal, A., Hillaire-Marcel, C., and Darby, D. A.: Variabil-
ity of sea ice cover in the Chukchi Sea (western Arctic
Ocean) during the Holocene, Paleoceanography, 20, PA4018,
https://doi.org/10.1029/2005PA001157, 2005.
de Vernal, A., Hillaire-Marcel, C., Solignac, S., Radi, T., and Ro-
chon, A.: Reconstructing sea ice conditions in the Arctic and
sub-Arctic prior to human observations, Geophysical Monograph
180, American Geophysical Union, Washington, USA, 27–45,
2008.
de Vernal, A., Hillaire-Marcel, C., Rochon, A., Fréchette, B., Henry,
M., Solignac, S., and Bonnet, S. : Dinocyst-based reconstructions
of sea ice cover concentration during the Holocene in the Arctic
Ocean, the northern North Atlantic Ocean and its adjacent seas,
Quaternary Sci. Rev., 79, 111–121, 2013.
Dyke, A. S. and Savelle, J. M.: Holocene history of the Bering Sea
bowhead whale (Balaena mysticetus) in Its Beaufort Sea sum-
mer grounds off southwestern Victoria Island, western Canadian
Arctic, Quaternary Res., 55, 371–379, 2001.
Elias, S., Short, S. K., and Phillips, R. L.: Paleoecology of late-
glacial peats from the Bering land bridge, Chukchi Sea shelf re-
gion, northwestern Alaska, Quaternary Res., 38, 371–378, 1992.
Elvelhøi, A. and Rønningsland, T. M.: Semiquantitative calculation
of the relative amounts of kaolinite and chlorite by X-ray diffrac-
tion, Mar. Geol., 27, M19–M23, https://doi.org/10.1016/0025-
3227(78)90070-1, 1978.
Farmer, J. R., Cronin, T. M., de Vernal, A., Dwyer, G. S., Keig-
win, L. D., and Thunell, R. C.: Western Arctic Ocean tempera-
ture variability during the last 8000 years, Geophys. Res. Lett.,
38, L24602, https://doi.org/10.1029/2011GL049714, 2011.
Fisher, D., Osterberg, E., Dyke, A., Dahl-Jensen, D., Demuth,
M., Zdanowicz, C., Bourgeois, J., Koerner, R. M., Mayewski,
P., Wake, C., Kreutz, K., Steig, E., Zheng, J., Yalcin, K.,
Goto-Azuma, K., Luckman, B., and Rupper, S.: The Mt Lo-
gan Holocene–late Wisconsinan isotope record: tropical Pacific–
Yukon connections, Holocene, 18, 667–677, 2008.
Frey, K. E., Perovich, D. K., and Light, B.: The spatial distribution
of solar radiation under a melting Arctic sea ice cover, Geophys.
Res. Lett., 38, L22501, https://doi.org/10.1029/2011GL049421,
2011.
Funder, S., Goosse, H., Jepsen, H., Kaas, E., Kjær, K. H., Kors-
gaard, N.J., Larsen, N. K., Linderson, H., Lyså, A., Möller, P.,
Olsen, J., and Willerslev, E.: A 10,000-year record of Arctic
Ocean sea-ice variability – View from the beach, Science, 333,
747–750, 2011.
Giles, K. A., Laxon, S. W., Ridout, A. L., Wingham, D. J., and Ba-
con, S.: Western Arctic Ocean freshwater storage increased by
Clim. Past, 13, 1111–1127, 2017 www.clim-past.net/13/1111/2017/
M. Yamamoto et al.: Holocene dynamics of western Arctic ocean circulation 1125
wind-driven spin-up of the Beaufort Gyre, Nat. Geosci., 5, 194–
197, 2012.
Godfrey, J. S.: A sverdrup model of the depth-integrated flow for the
ocean allowing for island circulations, Geophys. Astro. Fluid, 45,
89–112, 1989.
Griffin, G. M. and Goldberg, E. D.: Clay mineral distributions in the
Pacific Ocean, in: The sea, III, edited by: Hill, M. N., 728–741,
Interscience Pub., New York, USA, 1963.
Gudkovitch, Z. M.: On the nature of the Pacific current in Bering
Strait and the cause of its seasonal variations, Deep-Sea Res., 9,
507–510, 1962.
Harada, N.: Review: Potential catastrophic reduction of sea ice in
the western Arctic Ocean: its impact on biogeochemical cycles
and marine ecosystems, Global Planet. Change, 136, 1–17, 2016.
Hu, C., Henderson, G. M., Huang, J., Xie, S., Sun, Y., and John-
son, K. R.: Quantification of Holocene Asian monsoon rainfall
from spatially separated cave records, Earth Planet. Sc. Lett.,
266, 221–232, 2008.
Jakobsson, M., Pearce, C., Cronin, T. M., Backman, J., Ander-
son, L. G., Barrientos, N., Björk, G., Coxall, H., de Boer, A.,
Mayer, L. A., Mörth, C.-M., Nilsson, J., Rattray, J. E., Stranne,
C., Semiletov, I., and O’Regan, M.: Post-glacial flooding of
the Bering Land Bridge dated to 11 cal ka BP based on new
geophysical and sediment records, Clim. Past, 13, 991–1005,
https://doi.org/10.5194/cp-13-991-2017, 2017.
Kalinenko, V. V.: Clay minerals in sediments of the Arctic Seas,
Lith. Min. Res. 36, 362–372, translated from: Litologiya I
Poleznye Iskopaemye, 4, 418–429, 2001.
Katsuki, K., Khim, B.-K., Itaki, T., Harada, N., Sakai, H., Ikeda,
T., Takahashi, K., Okazaki, Y., and Asahi, H.: Land–sea linkage
of Holocene paleoclimate on the Southern Bering Continental
Shelf, The Holocene, 19, 747–756, 2009.
Kaufman, D. S., Axford, Y. L., Henderson, A. C. G., McKay, N.
P., Oswald, W. W., Saenger, C., Anderson, R. S., Bailey, H. L.,
Clegg, B., Gajewski, K., Hu, F. S., Jones, M. C., Massa, C., Rout-
son, C. C., Werner, A., Wooller, M. J., and Yu, Z.: Holocene cli-
mate changes in eastern Beringia (NW North America) – a sys-
tematic review of multi-proxy evidence, Quaternary Sci. Rev.,
147, 312–339, 2016.
Keigwin, L. D., Donelly, J. P., Cook, M. S., Driscoll, N. W., and
Brigham-Grette, J.: Rapid sea-level rise and Holocene climate in
the Chukchi Sea, Geology, 34, 861–864, 2006.
Kobayashi, D., Yamamoto, M., Irino, T., Nam, S.-I., Park, Y.-H.,
Harada, N., Nagashima, K., Chikita, K., and Saitoh, S.-I.: Distri-
bution of detrital minerals and sediment color in western Arctic
Ocean and northern Bering Sea sediments: Changes in the prove-
nance of western Arctic Ocean sediments since the last glacial
period, Polar Science, 10, 519–531, 2016.
Leduc, G., Vidal, L., Tachikawa, K., Rostek, F., Sonzogni, C., Beau-
fort, L., and Bard, E.: Moisture transport across Central America
as a positive feedback on abrupt climatic changes, Nature, 445,
908–911, https://doi.org/10.1038/nature05578, 2007.
Lee, S. H. and Whitledge, T. E.: Primary and new production in
the deep Canada Basin during summer 2002, Polar Biology, 28,
190–197, 2005.
Lisé-Pronovost, A., St-Onge, G., Brachfeld, S., Barletta, F., and
Darby, D.: Paleomagnetic constraints on the Holocene stratig-
raphy of the Arctic Alaskan margin, Global Planet. Change, 68,
85–99, 2009.
McKay, J. L., de Vernal, A., Hillaire-Marcel, C., Not, C., Polyak,
L., and Darby, D.: Holocene fluctuations in Arctic sea-ice cover:
dinocyst-based reconstructions for the eastern Chukchi Sea, Can.
J. Earth Sci., 45, 1377–1397, 2008.
McNeely, R., Dyke, A. S., and Southon, J. R.: Canadian marine
reservoir ages, preliminary data assessment, Open File Report
– Geological Survey of Canada, 5049, no. 3, Ottawa, Canada,
2006.
Miller, G. H., Alley, R. B., Brigham-Grette, J., Fitzpatrick, J. J.,
Polyak, L., Serreze, M. C., and White, J. W. C.: Arctic amplifi-
cation: can the past constrain the future?, Quaternary Sci. Rev.,
29, 1779–1790, 2010.
Müller, G.: Methods in Sedimentary Petrology, Scweizerbart Sci-
ence Publishers, 283 pp., Stuttgart, Germany, 1967.
Naidu, A. S. and Mowatt, T. C.: Sources and dispersal patterns of
clay minerals in surface sediments from the continental shelf ar-
eas off Alaska, Geol. Soc. Am. Bull., 94, 841–854, 1983.
Naidu, A. S., Creager, J. S., and Mowatt, T. C.: Clay mineral disper-
sal patterns in the North Bering and Chukchi Seas, Mar. Geol.,
47, 1–15, 1982.
Nishihara, M., Morri, H., and Koga, Y.: Structure determination of
a quartet of novel tetraether lipids from Methanobacterium ther-
moautotrophicum, J. Biochem., 101, 1007–1015, 1987.
Nishino, S., Shimada, K., Itoh, M., and Chiba, S.: Vertical double
silicate maxima in the sea-ice reduction region of the western
Arctic Ocean: implications for an enhanced biological pump due
to sea-ice reduction, J. Oceanogr., 60, 871–883, 2009.
Nwaodua, E., Ortiz, J. D., and Griffith, E. M.: Diffuse spectral re-
flectance of surficial sediments indicates sedimentary environ-
ments on the shelves of the Bering Sea and western Arctic, Mar.
Geol., 355, 218–233, 2014.
Olsen, J., Anderson, N. J., and Knudsen, M. F.: Variability of the
North Atlantic Oscillation over the past 5200 years, Nat. Geosci.,
5, 808–812, 2012.
Ortiz, J. D., Polyak, L., Grebmeier, J. M., Darby, D., Eberl, D. D.,
Naidu, S., and Nof, D.: Provenance of Holocene sediment on
the Chukchi-Alaskan margin based on combined diffuse spectral
reflectance and quantitative X-Ray Diffraction analysis, Global
Planet. Change, 68, 73–84, 2009.
Ortiz, J. D., Nof, D., Polyak, L., St-Onge, G., Lisé-Pronovost, A.,
Naidu, S., Darby, D., and Brachfeld, S.: The late Quaternary flow
through the Bering Strait has been forced by the Southern Ocean
winds, J. Phys. Oceanogr., 42, 2014–2029, 2012.
Osterberg, E. C., Mayewski, P. A., Fisher, D. A., Kreutz, K. J.,
Maasch, K. A., Sneed, S. B., and Kelsey, E.: Mount Logan ice
core record of tropical and solar influences on Aleutian Low vari-
ability: 500–1998 A.D., J. Geophys. Res.-Atmos., 119, 11189–
11204, https://doi.org/10.1002/2014JD021847, 2014.
Overland, J. O., Adams, J. M., and Bond, N.: Decadal variability of
the Aleutian Low and its relation to high-latitude circulation, J.
Climate, 12, 1542–1548, 1999.
Paillard, D., Labeyrie, L., and Yion, P.: Macintosh program per-
forms time-series analysis, EOS Trans. AGU 77, Washington,
D.C., USA, p. 379, 1996.
Park, Y.-H., Yamamoto, M., Polyak, L., and Nam, S.-I.: Glycerol di-
alkyl glycerol tetraether variations in the northern Chukchi Sea,
Arctic Ocean, during the Holocene, Biogeosciences Discuss.,
https://doi.org/10.5194/bg-2016-529, 2016.
www.clim-past.net/13/1111/2017/ Clim. Past, 13, 1111–1127, 2017
1126 M. Yamamoto et al.: Holocene dynamics of western Arctic ocean circulation
Petschick, R.: MacDiff 4.2.6., available at: http://www.geol-pal.
uni-frankfurt.de/Staff/Homepages/Petschick/classicsoftware.
html (last access: 2 September 2017), 2000.
Phillips, R. P. and Grantz, A.: Regional variations in provenance
and abundance of ice-rafted clasts in Arctic Ocean sediments:
implications for the configuration of late Quaternary oceanic and
atmospheric circulation in the Arctic, Mar. Geol. 172, 91–115,
2001.
Pickart, R. S.: Shelfbreak circulation in the Alaskan Beaufort Sea:
Mean structure and variability, J. Geophys. Res. 109, C04024,
https://doi.org/10.1029/2003JC001912, 2004.
Pickart, R. S., Pratt, L. J., Torres, D. J., Whitledge, T. E., Proshutin-
sky, A. Y., Aagaard, K., Agnewd, T. A., Moore, G. W. K., and
Dail, H. J.: Evolution and dynamics of the flow through Herald
Canyon in the western Chukchi Sea, Deep-Sea Res. Pt. II, 57,
5–26, 2010.
Polyak, L., Bischof, J., Ortiz, J. D., Darby, D. A., Channell, J. E. T.,
Xuan, C., Kaufman, D. S., Løvile, R., Schneider, D., Eberl, D.
D., Adler, R. E., and Council, E. A.: Late Quaternary stratigraphy
and sedimentation patterns in the western Arctic Ocean, Global
Planet. Change, 68, 5–17, 2009.
Polyak, L., Belt, S., Cabedo-Sanz, P., Yamamoto, M., and Park, Y.-
H.: Holocene sea-ice conditions and circulation at the Chukchi-
Alaskan margin, Arctic Ocean, inferred from biomarker proxies,
The Holocene, 26, 1810–1821, 2016.
Porter, S. E.: Assessing whether climate variability in the Pacific
Basin influences the climate over the North Atlantic and Green-
land and modulates sea ice extent, PhD Thesis, Ohio State Uni-
versity, Columbus, USA, 222 pp., 2013.
Proshutinsky, A. Y. and Johnson, M. A.: Two circulation regimes
of the wind-driven Arctic Ocean, J. Geophys. Res., 102, 12493–
12514, 1997.
Qin, W., Amin, S. A., Martens-Habbena, W., Walker, C. B.,
Urakawa, H., Devol, A. H., Ingalls, A. E., Moffett, J. M., Arm-
brust, E. V., and Stahl, D. A.: Marine ammonia-oxidizing achaeal
isolates display obligate mixotrophy and wide ecotypic variation,
P. Natl. Acad. Sci. USA, 111, 12504–12509, 2014.
Reimer, P. J., Bard, E., Bayliss, A., Beck, J. W., Blackwell, P. G.,
Ramsey, C. B., Buck, C. E., Cheng, H., Edwards, R. L., Friedrich,
M., Grootes, P. M., Guilderson, T. P., Haflidason, H., Hajdas, I.,
Hatté, C., Heaton, T. J., Hoffmann, D. L., Hogg, A. G., Hughen,
K. A., Kaiser, K. F., Kromer, B., Manning, S. W., Niu, M.,
Reimer, R. W., Richards, D. A., Scott, E. M., Southon, J. R.,
Staff, R. A., Turney, C. S. M., and van der Plicht, J.: Intcal13 and
Marine13 radiocarbon age calibration curves 0–50 000 years cal
BP, Radiocarbon, 55, 1869–1887, 2013.
Richter, I. and Xie, S.: Moisture transport from the Atlantic
to the Pacific basin and its response to North Atlantic
cooling and global warming, Clim. Dynam., 35, 551–566,
https://doi.org/10.1007/s00382-009-0708-3, 2010.
Rigor, I. G., Wallace, J. M., and Colony, R. L.: Response of sea ice
to the Arctic Oscillation, J. Climate, 15, 2648–2663, 2002.
Rimbu, N., Lohmann, G., Kim, J.-H., Arz, H. W., and
Schneider, R.: Arctic/North Atlantic Oscillation signa-
ture in Holocene sea surface temperature trends as ob-
tained from alkenone data, Geophys. Res. Lett., 30, 1280,
https://doi.org/10.1029/2002GL016570, 2003.
Roach, A. T., Aagaard, K., Pease, C. H., Salo, S. A., Weingartner,
T., Pavlov, V., and Kulakov, M.: Direct measurements of trans-
port and water properties through Bering Strait, J. Geophys. Res.,
100, 18433–18457, 1995.
Sagawa, T., Kuwae, M., Tsuruoka, K., Nakamura, Y., Ikehara, M.,
and Murayama, M.: Solar forcing of centennial-scale East Asian
winter monsoon variability in the mid-to late Holocene, Earth
Planet. Sc. Lett., 395, 124–135, 2014.
Sakshaug, E.: Primary and secondary production in the Arctic
ocean, in: The Organic Carbon Cycle in the Arctic Ocean, edited
by: Stein, R. and Macdonald, R. W., Springer, Berlin, Germany,
57–81, 2004.
Sandal, C. and Nof, D.: The Collapse of the Bering Strait Ice
Dam and the Abrupt Temperature Rise in the Beginning of the
Holocene, J. Phys. Oceanogr., 38, 1979–1991, 2008.
Sarnthein, M., Gebhardt, H., Kiefer, T., Kucera, M. Cook, M., and
Erlenkeuserd, H.: Mid Holocene origin of the sea-surface salin-
ity low in the subarctic North Pacific, Quaternary Sci. Rev., 23,
2089–2099, 2004.
Screen, J. A. and Simmonds, I.: The central role of diminishing
sea ice in recent Arctic temperature amplification, Nature, 464,
1334–1337, 2010.
Shimada, K., Carmack, E., Hatakeyama, K., and Takizawa, T.: Va-
rieties of shallow temperature maximum waters in the Western
Canadian Basin of the Arctic Ocean, Geophys. Res. Lett., 28,
3441–3444, 2001.
Shimada, K., Kamoshida, T., Itoh, M., Nishino, S., Carmack, E.,
McLaughlin, F., Zimmermann, S., and Proshutinsky, A.: Pacific
Ocean inflow: Influence on catastrophic reduction of sea ice
cover in the Arctic Ocean, Geophys. Res. Lett., 33, L08605,
https://doi.org/10.1029/2005GL025624, 2006.
Shtokman, V. B.: Vliyanie vetra na techeniya v Beringovo Prolive,
prichiny ikh bol’shikh skorostei i preobladayueshego severnogo
napravleniya, Trans. Inst. Okeanolog., Akad. Nauk SSSR, 25,
171–197, 1957.
Singh, H. K. A., Donohoe, A., Bitz, C. M., Nusbaumer, J., and
Noone, D. C.: Greater aerial moisture transport distances with
warming amplify interbasin salinity contrasts. Geophys. Res.
Lett. 43, 8677–8684, https://doi.org/10.1002/2016GL069796,
2016.
Stein, R.: Developments in Marine Geology: Arctic Ocean Sedi-
ments: Processes, Proxies, and Paleoenvironment, Elsevier, Am-
sterdam, the Netherlands, 529 pp., 2008.
Stein, R., Fahl, K., Schade, I., Nanerung, A., Wassmuth, S., Niessen,
F., and Nam, S.-I.: Holocene variability in sea ice cover, primary
production, and Pacific-Water inflow and climate change in the
Chukchi and East Siberian Seas (Arctic Ocean), J. Quaternary
Sci., 32, 362–379, 2017.
Steinhilber, F., Beer, J., and Fröhlich, C.: Total solar irradi-
ance during the Holocene, Geophys. Res. Lett., 36, L19704,
https://doi.org/10.1029/2009GL040142, 2009.
Steinman, B. A., Abbott, M. B., Mann, M. E., Ortiz, J. D., Feng,
S., Pompeani, D. P., Stansell, N. D., Anderson, L., Finney, B. P.,
and Bird, B. W.: Ocean-atmosphere forcing of centennial hydro-
climate variability in the Pacific Northwest, Geophys. Res. Lett.,
41, 2553–2560, https://doi.org/10.1002/2014GL059499, 2014.
Stigebrandt, A.: The North Pacific: A global-scale estuary, J. Phys.
Oceanogr., 14, 464–470, 1984.
Vare L. L., Masse G., and Gregory, T. R.: Sea ice variations in the
central Canadian Arctic Archipelago during the Holocene, Qua-
ternary Sci. Rev., 28, 1354–1366, 2009.
Clim. Past, 13, 1111–1127, 2017 www.clim-past.net/13/1111/2017/
M. Yamamoto et al.: Holocene dynamics of western Arctic ocean circulation 1127
Viscosi-Shirley, C., Mammone, K., Pisias, N., and Dymond, J.: Clay
mineralogy and multi-element chemistry of surface sediments on
Siberian-Arctic shelf: implications for sediment provenance and
grain size sorting, Cont. Shelf Res., 23, 1175–1200, 2003.
Vogt, C.: Regional and temporal variations of mineral assem-
blages in Arctic Ocean sediments as climatic indicator during
glacial/interglacial changes, Reports on Polar Research, 251, 1–
309, 1997.
Volkman, J. K.: A review of sterol markers for marine and terrige-
nous organic matter, Org. Geochem., 9, 83–99, 1986.
Wahsner, M., Müller, C., Stein, R., Ivanov, G., Levitan, M.,
Shekekhova, E., and Tarasov, G.: Clay-mineral distribution in
surface sediments of Eurasian Arctic Ocean and continental mar-
gin as indicator for source areas and transport pathways – a syn-
thesis, Boreas, 28, 216–233, 1999.
Walsh, J. J. and Dieterle, D. A.: CO2cycling in the coastal ocean.
I. A numerical analysis of the southeastern Bering Sea, with ap-
plications to the Chukchi sea and the northern Gulf of Mexico,
Prog. Oceanogr., 34, 335–392, 1994.
Watanabe, E., Onodera, J., Harada, N., Honda, M., Kimoto,
K., Kikuchi, T., Nishino, S., Matsuno, K., Yamaguchi, A.,
Ishida, A., and Kishi, J. M.: Enhanced role of eddies in
the Arctic marine biological pump, Nat. Commun., 5, 3950,
https://doi.org/10.1038/ncomms4950, 2014.
Weingartner, T., Aagaard, K., Woodgate, R., Danielson, S., Sasaki,
Y., and Cavalieri, D.: Circulation on the north central Chukchi
Sea shelf, Deep-Sea Res. Pt. II, 52, 3150–3174, 2005.
Winsor, P. and Chapman, D. C.: Pathways of Pacific water across
the Chukchi Sea: A numerical model study, J. Geophys. Res.,
109, C03002, https://doi.org/10.1029/2003JC001962, 2004.
Woodgate, R. A., Weingartner, T. J., and Lindsay, R.: Ob-
served increases in Bering Strait fluxes from the Pacific to
the Arctic from 2001 to 2011 and their impacts on the Arc-
tic Ocean water column, Geophys. Res. Lett., 39, L24603,
https://doi.org/10.1029/2012GL054092, 2012.
Yamamoto-Kawai, M., Carmack, E., and McLaughlin, F.: Ni-
trogen balance and Arctic throughflow, Nature, 443, p. 43,
https://doi.org/10.1038/443043a, 2006.
www.clim-past.net/13/1111/2017/ Clim. Past, 13, 1111–1127, 2017
