Sea-ice variability in the subarctic North Pacific and adjacent Bering Sea during the past 25 ka: new insights from IP25 and Uk′37 proxy records
- 211 Downloads
This study focusses on the last glacial–deglacial–Holocene spatial and temporal variability in sea-ice cover based on organic geochemical analyses of marine sediment cores from the subarctic Pacific and the Bering Sea. By means of the sea-ice proxy “IP25” and phytoplankton-derived biomarkers (specific sterols and alkenones), we reconstruct the spring sea-ice conditions, (summer) sea-surface temperature (SST) and primary productivity, respectively. The large variability of sea ice was explained by a combination of local and global factors, such as solar insolation, global climate anomalies and sea-level changes controlling the oceanographic circulation and water mass exchange between the subarctic Pacific and the Bering Sea. During the Last Glacial Maximum, extensive sea-ice cover prevailed over large part of the subarctic Pacific and the Bering Sea. The following deglaciation is characterized by a rapid sea-ice advance and retreat. During cold periods (Heinrich Stadial 1 and Younger Dryas) seasonal sea-ice cover generally coincided with low alkenone SSTs and low primary productivity. Conversely, during warmer intervals (Bølling/Allerød, Early Holocene) reduced sea-ice or ice-free conditions prevailed in the study area. At the northern Bering Sea continental shelf a late-Early/Mid Holocene shift to marginal sea-ice conditions is in line with the simultaneous wide-spread sea-ice recovery observed in the other Arctic marginal seas and is likely initiated by the lower Northern Hemisphere insolation and surface-water cooling.
KeywordsLast glacial to Holocene Sea-ice cover Biomarkers North Pacific Bering Sea
One of the key characteristics of the Arctic Ocean and its marginal seas—including the northern North Pacific and the Bering Sea—is the sea-ice cover with its strong seasonal variability. The Arctic sea ice is a critical and very sensitive component in the global climate system, that contributes to changes in the Earth’s albedo, the ecosystem and the deep-water formation (e.g., [33, 51]). Over the last decades, this sea ice decreased dramatically in its extent and thickness [29, 86, 108, 114, 115], and the causes of these recent changes, i.e., natural versus anthropogenic forcings, are poorly understood. Here, records of past climate and sea-ice conditions going beyond instrumental records and representing times of different boundary conditions, may help to decipher the processes controlling Arctic climate and sea ice variability (cf., ). By this, such records also help to further improve climate models for a better prediction of future climate change on Earth. Despite its significance, however, this type of records giving detailed information about past Arctic sea-ice conditions are still very rare, especially due to the lack of precise proxies for sea-ice reconstructions (cf.  and references therein).
In the Bering Sea, various paleoceanographic studies have been carried out to reconstruct the past sea-ice distribution based on diatoms assemblages [19, 52, 53, 95] and radiolarians (i.e. [69, 116]). Kuehn et al.  performed an ultra-high-resolution micro-X-ray-fluorescence and sediment facies study of laminated sediments recovered at a sediment core from the north-eastern Bering Sea. They identified an alternation of terrigenous and diatomaceous laminae and interpreted the diatomaceous laminae as productivity events related to the retreating sea ice margin. All these studies have shown that the Bering Sea underwent major climatic changes during the late Quaternary, with rapid fluctuations between warm and cold periods, leading to spatial and temporal variations in sea ice. In this study, we report results of organic geochemical analyses from three sediment cores from the subarctic North Pacific and Bering Sea. The main objective is to reconstruct the spatial and temporal sea-surface temperature and sea-ice cover variability during the Last Glacial Maximum (LGM) and the following deglaciation in this area.
Regional setting of the study area
The Bering Sea is a subarctic semi-enclosed sea situated between the North Pacific and the Arctic Ocean and comprised a north-eastern shallow continental shelf and a southern deep basin (Fig. 1). The physical oceanography of the eastern Bering Sea is influenced by tides, winds, topography, flows through passages and the annual formation, drift and melting of sea ice .
Summary of locations and data source of additional cores discussed in the text
Emperor Seamount, N. Pacific
Detroit Seamount, N. Pacific
Gebhardt et al. 
Patton Seamount, N. Pacific
Gebhardt et al. 
Bowers Ridge, Bering Sea
Cook et al. 
Bowers Ridge, Bering Sea
Umnak Plateau, Bering Sea
Umnak Plateau, Bering Sea
Caissie et al. 
Shirshov Ridge, Bering Sea
Cook et al. 
The water mass circulation in the Bering Sea originate from two currents of nutrient-rich North Pacific waters, the Alaskan Current (AC) and the Alaskan Coastal Current (ACC) . Flowing through the Aleutian passages, they create the anticlockwise surface-water circulation. The water mass circulation in the eastern Bering Sea basin, is dominated by two principal currents, the Aleutian North Slope Current (ANSC)  and the Bering Slope Current (BSC) .
Sea ice, with its strong seasonal variability is a critical component of the Bering Sea system. After initial in situ ice formation in the northern coastal regions as early as October, the Bering Sea ice is rapidly transported southward by prevailing northeasterly winds in December, January, and February, reaching its maximum in March or early April [79, 85, 87, 106]. By July, the sea ice retreats through the Bering Strait, leaving the Bering Sea-ice free until September [77, 78] (Fig. 1). Besides the seasonal variability, a prominent interannual variability in summer and winter sea ice distribution has been recorded [21, 25, 75, 97].
Proxy reconstruction of past sea ice and sea surface temperatures
A large number of studies dealing with the reconstruction of past sea-ice cover are based on sedimentological, mineralogical, and geochemical data as well as microfossils such as diatoms, dinoflagellates, ostracods, and foraminifers (e.g. [31, 32, 28, 30, 58, 88]). These are all predominantly indirect sea-ice proxies, with different advantages and disadvantages (see ). In this study, we have used a biomarker approach based on the determination of a highly branched isoprenoid (HBI) with 25 carbons (C25 HBI monoene = IP25) . This biomarker is only biosynthesized by specific diatoms living in the Arctic ice [16, 17; 17], meaning the presence of IP25 in the sediment is a direct proof of the presence of past Arctic sea ice. Meanwhile, this proxy has been successfully used in numerous studies for spring sea-ice reconstructions in the Arctic and marginal seas, covering different time periods and temporal resolutions (e.g., [4, 7, 10, 11, 12, 40, 47, 57, 65, 67, 70, 72, 73, 110, 111, 112, 113, 118, 121, 122]; for overviews see [9, 111]). The ambiguity of this new proxy resides in the fact that IP25 values of about zero can indicate either the absence of sea ice, or the opposite scenario, permanent and thick sea ice, preventing light penetration and thus, ice algae growth. To circumvent this issue and distinguish between these two opposite scenarios, Müller et al.  compared IP25 with phytoplankton-derived biomarkers (i.e., brassicasterol and dinosterol). When IP25 is absent, lack or very low abundance of phytoplankton biomarkers reflects permanent sea-ice coverage, whereas elevated abundance of phytoplankton markers reflects ice-free conditions.
Another useful approach that help to distinguish between the two “IP25 = 0” extremes, i.e., ice-free versus thick closed ice coverage, is the determination of the sea-surface temperature (SST). SST values significantly above 0 °C give important information about surface water characteristics per se, but also clearly point to ice-free conditions if IP25 = 0 (cf. [57, 113]). Furthermore, the coincidence of SST values ≫ 0 °C and the presence of IP25 in the same samples indicate that the SST values represent summer conditions, whereas the IP25 values represent sea-ice conditions during the spring (or autumn) season (cf. ).
For SST reconstruction, a very promising biomarker tool is the alkenone thermometry [15, 89]. This tool evolved from the observation that certain microalgae of the class Prymnesiophyceae, notably the marine coccolithophorids Emiliania huxleyi and Gephyrocapsa oceanica (e.g. [26, 62, 119]), and presumably other living and extinct members of the family Gephyrocapsae [63, 71], have or had the capability to synthesize alkenones whose extent of unsaturation changes with growth temperature [15, 62, 89]. Based on this correlation, paleo-SST can be calculated from the so-called ketone unsaturation index UK37 [15, 71, 89] or the simplified version of the index UK′37 [71, 89].
Materials and methods
Lithology and age control of studied sediment cores
The three studied cores were recovered during the INOPEX RV Sonne Expedition 202 in 2009 with a kastenlot corer (Fig. 1; Gersonde ). Core SO202-07-6 was taken from the Detroit Seamount in the northwestern Pacific Ocean (Fig. 1; 51°16.29′N, 167°41.98′E, 2340 m water depth). The 4.69 m long sediment core presents a heterogeneous lithological composition and can be divided into five different units, including terrigenous mud, diatoms ooze and calcareous sediment. Core SO202-18-6, recovered from the Bering Sea continental slope (Fig. 1; 60°07.6′N, 179°26.1′W, water depth 1105 m) is 7.21 m long and shows lithological changes with an alternation between diatom-rich sediment and terrigenous mud. The section is characterized by the presence of laminations and well-preserved bivalves. Core SO202-27-6 is 2.91 m long and recovered from the Patton Seamounts, in the northeastern Pacific Ocean (Fig. 1; 54°117.8′N, 149°36.0′W, water depth 2919 m). It consists of biosiliceous oozes containing bioturbated calcareous biogenic remains, terrigenous mud and high amounts of high-rifted debris (IRD) probably originating from southern coastal Alaska .
The age models of all three cores resulted from a combined chronostratigraphic approach, including inter-core correlations using high-resolution spectrometry and X-ray fluorescence and lamination pattern as well as 14C-accelerator mass spectrometry radiocarbon dating of planktic foraminifera. For source, background and details about the chronology and age models of the three cores, including a discussion of potential changes in local reservoir ages, we refer to Kuehn et al.  (Core SO202-18-6), Maier et al.  and Serno et al.  (Core SO202-07-6), and Maier et al.  (Core SO202-27-6).
Analytical approach of biomarker determinations
After collection, the samples were stored at − 30 °C in glass bottles until further processing. Subsequently to the freeze-drying and homogenisation of the sediment, the TOC content was determined by means of a LECO CS-125 elemental analyser.
For biomarkers analysis, ca. 2–5 g of sediment were extracted with an Accelerator Solvent Extractor (DIONEX-ASE 200; 100 °C, 5 min, 1000 psi) using dichloromethane:methanol (2:1, v/v) as solvent. For quantification purpose, squalane (0.48 µg/sample), 7-hexylnonadecane (0.0766 µg/sample), and cholesterol-d6 (cholesterol-5-en-3β-ol-D6; 2.2 µg/sample) were added as internal standards before further analytical step. An aliquot of the total extract was used for analysing n-alkanes, IP25, alkenones and sterols. These different biomarkers were separated by column chromatography using SiO2 as stationary phase (for details, see ). The n-alkanes and alkenones were analyzed using GC (HP6890) as described by Fahl and Stein [38, 39], and IP25 and sterols were analyzed using gas chromatography–mass spectrometry (GC–MS; Agilent 6850; 30 m HP-5 MS column, 0.25 mm i.d., 0.25 µm film thickness; coupled to an Agilent 5975 C VL mass selective detector). The identification of the compounds was achieved on basis of GC retention time and fragmentation pattern obtained from mass spectrometry. These lipids were identified by comparison with the mass spectra published by Boon et al.  and Volkman  for sterols and by Belt et al. [5, 8] for IP25. Absolute biomarker concentrations were normalized to TOC. For further details about IP25 identification and quantification we refer to Fahl and Stein  and Belt et al. . All TOC and biomarker data are available under https://doi.pangaea.de/10.1594/PANGAEA.887833.
Bering Sea continental shelf (Core SO202-18-6)
The lowermost interval of Core SO202-18-6, corresponding to the Bølling–Allerød (B/A, ~ 14.7–12.9 ka; ), is characterized by moderate TOC (0.82–0.95 wt%), high long-chain n-alkanes values and variable but relatively high brassicasterol (avg. ~ 102.2 µg/g TOC) and low dinosterol (avg. ~ 14.6 µg/g TOC) values (Fig. 2). The concentrations of the sea-ice proxy IP25 (avg. ~ 0.1 µg/g TOC) show variable values with a slight general trend towards higher values, coinciding with the gradual decrease in Uk′37-based SSTs (Fig. 2; Supplementary Table S1).
The interval between 622 and 432 cm corresponding to the Younger Dryas (YD, ~ 12.9–11.7 ka; ), is characterised by rather low TOC values throughout this interval (0.74–0.93 wt%). IP25 concentrations remain relatively high (avg. ~ 0.11 µg/g TOC), reach maximum values around 570 cm (~ 12.5 ka) and decline abruptly until total disappearance at ~ 470 cm (~ 12 ka, Supplementary Table S1). This observation matches the short period of a pronounced decrease in SST, with a minimum of ~ 4.9 °C, reached around 560 cm (~ 12.5 ka) and the subsequent gradual SST increase. During this interval of elevated IP25 concentrations and low SST, concentrations of phytoplankton markers, brassicasterol and dinosterol display reduced values (avg. ~ 30.8 and ~ 15.6 µg/g TOC, respectively) (Fig. 2; Supplementary Table S1).
The early Holocene is marked by distinctly increased TOC contents, reaching maximum values (> 1 wt%) between ~ 400 and ~ 170 cm (~ 11.5–10 ka). Likewise, continuously increasing SST, reaching maximum values of ~ 9 °C at ~ 180 cm (~ 10 ka), strongly coincides with the absence of IP25 in the sediment. Simultaneously, phytoplankton markers (brassicasterol and dinosterol) show maximum values (avg. ~ 116 and ~ 31 µg/g TOC, respectively) (Fig. 2; Supplementary Table S1).
At about 160 cm (~ 10 ka) a major TOC drop to values < 0.7 wt% is observed. At the same time, also an abrupt decline in phytoplankton biomarkers and long-chain n-alkanes occurs, remaining at minimum values until ~ 5.4 ka (Fig. 2; Supplementary Table S1). Furthermore, this change coincides with a decrease in SST reaching minimum values of 6.5 °C around 60 cm (~ 7 ka) and an increase of IP25 to maximum concentrations of ~ 0.485 µg/g TOC around 2 cm (~ 5.4 ka).
North-eastern Pacific (Core SO202-27-6)
The lower half of the studied section of Core SO202-27-6, representing the LGM and Heinrich Stadial 1 (HS1), is characterized by TOC values around 0.5%, SST values between 7 and 9 °C (decreasing upwards), and high IP25 concentrations between 0.2 and 0.8 µg/g TOC with an absolute maximum of 0.89 µg/g TOC at 55 cm (~ 17 ka). During this period, minimum phytoplankton marker and relatively low long-chain n-alkanes concentrations are observed (Fig. 3; Supplementary Table S2). TOC contents reach highest values of 0.9 wt% at 35 cm (~ 14.9 ka). This period coincides with an absence of IP25, maximum concentrations of brassicasterol and dinosterol (~ 45 µg/g TOC at 14.2 ka and ~ 33 µg/g TOC at 14.9 ka, respectively) (Fig. 3; Supplementary Table S2).
The Holocene is characterized by maximum SST values of 12 °C. Relatively low TOC and phytoplankton markers prevailing during this period, coeval with the absence of IP25 (Fig. 3; Supplementary Table S2).
North-western Pacific (Core SO202-07-6)
The general patterns of TOC, SST and IP25, at Core SO202-07-6 are relatively similar to those of Core SO202-27-6 although differences are obvious when looking at the absolute values. In the lower part correlating with the LGM and HS1, TOC values of ~ 0.7% are typical, SST values decrease towards minimum values around 8 °C, and IP25 reaches high concentrations of 1–3 µg/g TOC with an absolute maximum of 8.4 µg/g TOC at 110 cm (23.6 ka) (Fig. 4, Supplementary Table S3). Across the boundary between HS1 and the B/A interval, TOC contents reach highest values of ~ 1.1 wt%, associated with an abrupt drop of IP25 to zero values. Concentrations of phytoplankton markers are highly variable within this interval and simultaneously reach maximum values of ~ 67 µg/g TOC for brassicasterol and ~ 18 µg/g TOC for dinosterol at the end of MIS 2 (55 cm; ~ 14.2 ka) (Fig. 4; Supplementary Table S3). High concentrations of long-chain n-alkanes, on the other hand, occur in the lower part of the record (pre-LGM and LGM).
Similarly to Core SO202-27-6, MIS 1 (Holocene) is marked by the absence of IP25, coinciding with low values of TOC, dinosterol and long chain n-alkanes (avg. ~0.5 wt%, ~ 4 µg/g TOC and 78 µg/g TOC, respectively). SST values increase to about 10 °C during the early Holocene.
Our comprehensive organic-geochemical study of sediments from cores SO202-18-6, SO202-07-6 and SO202-27-6, recovered from the Bering Sea, the north-western and the north-eastern subarctic Pacific (Fig. 1), respectively, clearly demonstrates glacial/interglacial climate and environmental variability in these areas over the last 25 ka. The obtained biomarker records provide new insights about the spatial and temporal sea-ice variations and related changes in primary production, SST and oceanic circulation patterns in the Bering Sea and the subarctic Pacific during last glacial to Holocene cold intervals (Figs. 5, 6; i.e., the Last Glacial Maximum—LGM, the Heinrich Stadial 1—HS1, and the Younger Dryas—YD) and warm intervals (Bölling/Alleröd—B/A, Early Holocene Thermal Maximum—HTM). At all sites and during warm as well as cold climatic intervals SST values are well above 0 °C (i.e., ranging between about 5 and 12 °C; Fig. 5), suggesting that the SST data represent more the summer situation with ice-free conditions. If IP25 is present contemporaneously (i.e., in the same sample), this supports that the IP25 proxy is indicative for sea-ice conditions during the spring season (cf. [8, 113]).
The Last Glacial Maximum (LGM)
The period corresponding to the LGM (22–18 ka) is characterised by elevated IP25 concentrations in cores SO202-07-6 and SO202-27-6 (Fig. 6), pointing to the occurrence of sea ice in the western and eastern subarctic Pacific. A more detailed reconstruction of sea-ice conditions is allowed by looking at the IP25/phytoplankton biomarker relationship (Fig. 7) and the PBIP25 and PDIP25 index values (Fig. 8). For Core SO202-27-6, both the IP25/phytoplankton biomarker relationship and maximum PBIP25 and PDIP25 index values point to an extended spring sea-ice cover in the eastern subarctic Pacific (Fig. 9a). However, in Core SO202-07-6, lower PIP25 values (Fig. 8) suggest marginal to variable sea-ice coverage in the western subarctic Pacific (Fig. 9a). The occurrence of sea ice in this area is also supported by the presence of ice-related diatoms in Core ES from the north-western Pacific (Fig. 1; Table 1) .
Although the LGM reconstruction of sea-ice condition using IP25 in the Bering was not possible, previous studies revealed sea-ice related diatoms occurrence in sediment cores from the Umnak Plateau [19, 27, 53] and the Shirshov Ridge , interpreted as thick, extensive sea-ice coverage in these regions during the LGM (Fig. 9a). However, the absence of sea-ice related diatoms on the western Bowers Ridge, indicate that sea ice was probably at the crest but not on the western side of the Bowers Ridge during this period  (Fig. 9a).
These observations are consistent with diatoms analyses performed by Katsuki et al.  in the Bering Sea and the subarctic Pacific. In this study, they showed that Core ES from the western subarctic Pacific (Table 1; Fig. 1) contained higher proportion of open-water diatoms and fewer ice-related diatoms than cores from the eastern Bering Sea. Based on these observations, they also concluded that seasonal sea ice prevailed in the western Bering Sea and subarctic Pacific, whereas in the eastern part of the study area more stable and long-lasting sea-ice conditions existed.
These conclusions align well with the divergent trends observed in phytoplankton markers in the two cores from the subarctic Pacific (Fig. 5). Indeed, high concentrations of sterols in Core SO202-07-6 indicate elevated phytoplankton productivity in the western subarctic Pacific, whereas low sterols values, recorded in Core SO202-27-6, point to reduced productivity in the eastern subarctic Pacific (Fig. 5). This dissimilarity in productivity between the eastern and western subarctic Pacific is likely due to the different spatial and temporal extent of seasonal sea ice in the subarctic Pacific during the LGM. Hence, extent and long-lasting sea ice in the eastern subarctic Pacific likely restricted the biological production period to a shorter and colder summer. However, in the western subarctic Pacific, earlier sea-ice retreat in the season probably allowed greater phytoplankton production. Moreover, during the LGM, higher abundance of long chain n-alkanes, resulting from greater contribution of land plant organic matter [34, 35], is registered in Core SO202-07-6 (Fig. 4; Appendix Table A3). Eolian dust input, by providing land-derived trace nutrients—known as a limiting factor for phytoplankton growth —probably supported greater phytoplankton productivity in the western subarctic Pacific . Enhanced glacial terrigenous input was also reported in previous studies in the Bering Sea [81, 82], the Okhotsk Sea , and the north-western Pacific  and interpreted as an increase in atmospheric mineral dust transportation.
During the LGM, extreme glacial conditions with extended sea ice in the Bering Sea and continental ice sheets close-by prevailed as shown by the predominance of sea-ice diatoms and high abundance of ice-rafted debris (IRD) in this area (e.g. [27, 53, 116, 123]). This extensive glaciation initiated a severe sea-level drop of 120–135 m below the present one  inducing major morphologic changes in the Bering Sea environment. Hence, the eastern Bering Sea shallow continental shelf was aerially exposed at that time, forming Beringia (Fig. 9a). The rapid sea-level changes also caused the closure of Bering Strait, impacting the water-mass circulation between the Bering Sea and the Arctic Ocean [46, 47, 48, 49, 50].
The resulting cut-off of the warm and nutrient-rich Alaskan Stream waters inflow, as reflected by the low abundance of open-water diatoms, an Alaskan Stream indicator species , probably allowed the establishment of a perennial sea-ice cover and low biological productivity [19, 95]. In the Bering Sea, low biological production during the LGM is revealed by low δ13C values from planktic foraminifera , probably caused by the combination of extended sea-ice cover, cold and low salinity surface-water  and the cut-off of warm and nutrient-rich North Pacific water supply.
The Heinrich Stadial 1 cold event (HS1)
The early phase of HS1 (~ 18–15 ka; e.g., Hemming ) is marked by a significant increase in IP25 concentrations in the north-eastern Pacific (Core SO202-27-6), and stable, elevated IP25 concentrations in the north-western Pacific, pointing to sea-ice occurrences in both regions (Fig. 6). Low but still > 0 °C SSTs seem to prevail in the western and eastern subarctic Pacific during this period (Fig. 5). This observation supports the occurrence of spring sea ice but ice-free summers in these two regions.
During HS1 cold event, surprisingly higher values of PBIP25 and PDIP25 indexes, compared to the LGM, are calculated in Core SO202-07-6 (Fig. 8), probably reflecting marginal conditions in the western subarctic Pacific (Fig. 9b). This is well in accordance with maximum IRD accumulation reported by Gebhardt et al.  between 15 and 19 ka at sites MD02-2489 and MD01-2416, in the northeast and northwest Pacific, respectively (Table 1; Fig. 6).
During HS1, relatively high IP25 concentration in cores SO201-2-12 and SO201-2-77 indicate that sea-ice coverage remained stable and extensive in the western Bering Sea  (Fig. 6). However, the decrease in sea-ice related diatoms abundance, observed on the Umnak Plateau at ~ 16.7 ka [19; Table 1], suggests the occurrence of short open-water periods during summer.
In the eastern subarctic Pacific (Core SO202-27-6) a transition period starting around 16 ka, is marked by a simultaneous ~ 1.5 °C raise in SSTs (Fig. 5), sudden drop in IP25 concentrations (Fig. 6) and lower PBIP25 and PDIP25 index values (Fig. 8). These fluctuations coinciding with a temperature increase over Greenland towards the end of the HS1 (Fig. 5), are pointing toward the initiation of warming and ice melting, leading to a shift from marginal sea-ice cover registered during the first part of HS1 to variable/less and subsequent ice-free conditions (as suggested by the absence of IP25) established at the end of the HS1 (Figs. 7, 8). This interpretation is in agreement with the sudden drop in IRD and the increase of SST recorded around 15 ka in Core MD02-2489 from the eastern subarctic Pacific (Fig. 6; ).
Meanwhile an increase in concentrations of phytoplankton in the eastern subarctic Pacific (Fig. 5), suggesting an enhanced phytoplankton productivity, is in accordance with the decrease in sea-ice coverage. Similarly, a massive rise in productivity and abrupt drop of sea-surface salinity registered during final HS1 in core MD02-2489 (Table 1; Fig. 6; ), confirm the input of fresh melt water induced by melting of sea ice in the eastern North Pacific .
The Bølling–Allerød warm period (B/A)
With the onset of the B/A, IP25 (Fig. 6) as well as related PBIP25 and PDIP25 index values drop down to zero in Core SO202-27-6 (Fig. 8) suggesting ice-free conditions during spring in the eastern subarctic Pacific that continued throughout the B/A warm period (Figs. 7, 9c). Similar conditions seem to have occurred in the western subarctic Pacific as shown by the (almost) absence of IP25 in Core SO202-07-6 (Fig. 6) and the IP25 versus phytoplankton biomarker relationship (Fig. 7). The one high PBIP25 value of > 0.4 near 13 ka (Fig. 8) has to be interpreted very cautiously as both IP25 and the brassicasterol concentrations are extremely low. Maximum TOC values (Figs. 3, 4) and increased phytoplankton biomarker concentrations (Fig. 5) in cores SO202-07-6 and SO202-27-6 reflect the enhancement of primary productivity in the subarctic Pacific. In previous studies, similar elevated early deglacial productivity characterized by laminated organic carbon and biogenic opal-rich sediment cores from the northwest [41, 55, 92] and northeast Pacific , was also reported.
In the north-eastern Bering Sea (Core SO202-18-6), ice-free condition and maximum productivity at 13.4 ka are reflected by the lack of IP25 (Fig. 6), consistent with the maximum SSTs (Fig. 5) and elevated phytoplankton biomarker concentrations (Fig. 5). Open-water conditions also prevailed in the western Bering Sea as demonstrated by the IP25 drop to concentrations close to zero (Fig. 6) in cores SO201-2-12 and SO201-2-77 . Similar to the North Pacific situation, high deglacial biological productivity in the Bering Sea is also reflected in the occurrence of laminated and biogenic opal-rich sediments [18, 19, 27, 59, 99] and the increase of open-water diatoms in sediments from the western Bering Sea [66, 92], the Umnak Plateau [19, 95], the Bowers Ridge [27, 53] and Shirshov Ridges . By combining ultra-high-resolution micro-X-ray-fluorescence data and sediment facies analysis of the laminated B/A interval at Core SO202-18-6, Kuehn et al.  identified an alternation between predominantly terrigenous layers and diatom-dominated opal sedimentation. These authors interpreted the diatomaceous laminae as spring/summer productivity events related to the retreating sea-ice margin. Under such conditions, increased flux and subsequent remineralization of the organic matter may have resulted in oxygen-deficient conditions causing laminae preservation.
In general, increased marine productivity in the Bering Sea and the subarctic Pacific is most likely due to the B/A climate warming resulting in modern summer-like conditions in these areas. Consecutive melting of sea ice and fresh water release in the Bering Sea resulted in stronger water stratification  and enhanced primary productivity.
At the end of the B/A (around ~ 13.5 ka), an increase of IP25 (Fig. 6), PBIP25 and PDIP25 (Fig. 8) associated with a drop of phytoplankton markers (Fig. 5), is observed in the northern Bering Sea (Core SO202-18-6) and correlate with an increase of IP25 in the western Bering Sea (cores SO201-2-112 and SO201-2-77) (Fig. 6). This probably reflects the initiation of sea-ice advance, likely due to the contemporary SSTs cooling leading to a shift from ice-free conditions at the onset of the B/A to variable sea-ice conditions at the end of this period. Similar phenomena of sea-ice advance, leading to ice-edge conditions, is observed in the western subarctic Pacific at the end of the B/A as reflected by an increase in IP25 (around ~ 13.5 ka) in core SO201-2-12 (Fig. 6) and in PBIP25 and PDIP25 in core SO202-07-6 (Fig. 8).
The Younger Dryas cold event (YD)
The occurrence of the YD-like cold event in the Bering Sea is demonstrated by minimum SSTs (Fig. 5), matching the millennial-scale temperature fluctuations recorded in Greenland ice core isotope records .
Meanwhile, high PIP25 index values are observed in Core SO202-18-6 and a significant increase of IP25 concentrations is registered in all three cores from the Bering Sea (SO202-18-6, SO201-2-114 and SO201-2-77; ) and in the western North Pacific (Core SO201-2-12). Maximum IP25 concentrations are simultaneously reached in all four cores at 12.5 ka (Figs. 6, 7, 8; Supplementary Table S1). This suggests that YD-colder climatic conditions lead to rapid sea-ice advance, resulting in extended spring sea-ice coverage spreading form the north-eastern Bering Sea continental shelf to the western Bering Sea and the western subarctic Pacific (Fig. 9d).
However, in core SO202-07-6, which compared to core SO201-02-12 is farther south and further from the Kamchatka coast, the increase in IP25 concentration is almost imperceptible (Fig. 6), but PDIP25 shows a net increase during this period (Fig. 8). Therefore, this site probably corresponds to the boundary of sea-ice advance with occasional sea-ice occurrences in the area. In the eastern subarctic Pacific (Core SO202-27-6), the lack of IP25 (Fig. 6) and minimal PIP25 values (Fig. 8) demonstrate well-established ice-free conditions during spring (Fig. 8d). As registered by the increasing temperatures (Fig. 5), the YD-like cold event is not reflected in this area.
Previous studies based on sea-ice diatoms abundance allow us to better constrain marginal ice zone in the Bering Sea. Hence, Cook et al.  and Caissie et al.  have shown that during the YD, sea ice diatoms were abundant in cores from the Umnak plateau, pointing to the presence of stable sea-ice cover in this area. However, according to Cook et al. , diatom abundance shows that it was unlikely that sea ice reached Bowers Ridge. Reduced concentrations of phytoplankton markers (Fig. 5) recorded in Core SO202-18-6, are probably the result of extended sea-ice cover limiting the primary productivity on the Bering Sea continental shelf.
From 12.5 ka onwards, decreasing IP25 concentrations is observed in the Bering Sea (cores SO202-18-6, SO201-2-114 and SO201-2-77; ) and in the western North Pacific (Core SO201-2-12). This reflects the initiation of the sea-ice retreat allowing longer period of ice-free conditions in these areas. Simultaneously, alkenone records indicate a warming likely triggered by the increase in Northern Hemisphere summer insolation .
In a recent study by Cabedo-Sanz et al. , comparable IP25 records with high IP25 and PBIP25 values at the onset of the YD shifting to lower values from the mid YD, were also observed in northern Norway. Hence, alike in the Bering Sea, those records reveal a shift from a reduced sea-ice cover to marginal ice-zone conditions in northern Norway that occurred at about 11.9 ka .
In the Bering Sea, the transition between the YD-like cold event and the Early Holocene is marked by the rapid disappearance of IP25 around 11.5 ka in both cores from the Bering Sea (Fig. 6) and minimum values of PIP25 in Core SO202-18-6 (Fig. 8). The absence of IP25 and minimum PIP25 index values is also registered in the subarctic Pacific during this period (Fig. 7). These observations demonstrate that year-round open-water conditions prevailed in these areas during the Early Holocene (Fig. 9e). These data are confirmed by an abrupt decrease in IRD in the southern Bering Sea  and in the Okhotsk Sea  observed during this period. Retreat of sea ice from the Bering Sea during the deglacial is also shown by a shift in the diatom assemblages, from one dominated by sea-ice diatoms to one dominated by high-productivity species [19, 53]. This aligns well with the warming event, recorded by the simultaneous temperatures increased by more than ~ 3 °C in the three studied cores (Fig. 5) and is also observed in previous studies in the subarctic Pacific , the Bering Sea [19, 66, 82] and the Okhotsk Sea .
In northern Norway, similar observations of establishment of ice-free conditions at the beginning of the Holocene were reflected by the absence of IP25 and decrease of PBIP25 to minimal values after 11.5 ka .
This warm interval, corresponding to the Holocene Thermal Maximum (HTM; 11.5-9 ka), was likely initiated by the maximum in Northern Hemisphere summer insolation . Melting of ice-sheets during the deglaciation induced rapid sea-level rise, causing the re-opening of the Bering Strait (dated to 11–12 ka; [22, 36, 37, 56]) and the Aleutian passages, allowing the inflow of the warm and nutrient-rich subarctic Pacific water into the Bering Sea. Increased abundances of N. seminae—known as a reliable tracer of the Alaskan Stream water —provide an evidence of enhanced inflow of the subarctic Pacific Water into the Bering Sea [53, 95]. The combination of increased insolation and inflow of warm Pacific water led to the observed warming in the Bering Sea. Those conditions also lead to high deglaciation productivity as shown by the increase of phytoplankton markers in the Bering Sea and in the western subarctic Pacific (cores SO202-16-6 and SO202-07-6, respectively; Fig. 5). High primary productivity in the Bering Sea during deglacial is also shown by the peaks in CaCO3  and the positive δ15N anomalies . This was probably triggered by the dominant summer conditions and the retreat of sea ice to the continental shelf regions [1, 80, 84, 93].
From 8.5 ka onwards, a significant increase in IP25 concentrations (Fig. 6), remaining high during the Mid Holocene, and a shift from minimal PBIP25 and PDIP25 index to high values occur in Core SO202-18-6 from the north-eastern Bering Sea continental shelf (Fig. 8). This coincides with an abrupt decrease in phytoplankton marker concentrations and a ~ 3 °C drop in SSTs (Fig. 5), in agreement with a prominent cooling event observed at high latitudes during this interval [2, 24]. These observations reflect the advance of sea ice leading to marginal/extended sea-ice cover on the Bering Sea continental shelf during the Mid Holocene (Fig. 9f).
This event of late-Early Holocene/Mid Holocene increase of spring sea-ice cover and decreasing phytoplankton marker accumulation was also documented in different regions of the Arctic. Hence, Müller et al. [72, 74] registered an increase in IP25 contents and IRD in sediment cores from the Fram Strait, interpreted as winter sea-ice extent reaching the West Spitsbergen continental coast during the Mid Holocene. Simultaneously, an increase in IP25 and PIP25 values in sediment cores from the Laptev, East Siberian, and Chukchi seas [40, 112] as well as IRD input at the western Barents Sea continental slope  point to a sea-ice advance in these areas. Finally, IP25 reconstructions in sediment cores from the Canadian Arctic Archipelago also showed an increase in sea ice since 6 ka .
The broad sea-ice reappearance in the Arctic and marginal seas is likely due to decrease in Northern Hemisphere insolation and general surface water cooling enhancing sea-ice formation during winter and delaying its spring retreat.
In this study, the last glacial–deglacial–Holocene variability in SST and sea-ice cover are reconstructed using alkenone-based SST and sea-ice IP25 and PIP25 proxies in sediment cores from the Bering Sea and the subarctic Pacific. At all sites and during warm as well as cold climatic intervals SST values are well above 0 °C, suggesting ice-free summer conditions throughout.
In line with low Northern Hemisphere insolation and a cold global climate, the LGM spring sea-ice coverage was extended, spreading into the entire Bering Sea, to the eastern subarctic Pacific and reaching Emperor Seamount in the western subarctic Pacific. Furthermore, resulting lowered sea level provoked the closure of the Bering Strait, preventing the warm Pacific Water inflow into the Bering Sea and, thus, allowing further sea-ice extension. At the onset of the HS1, sea-ice coverage remained extended in the Bering Sea and subarctic Pacific but a shift to stable marginal ice-zone conditions with longer period of open-water environments is observed in the subarctic Pacific at the end of HS1.
Deglaciation accelerated at the onset of the B/A, leaving the subarctic Pacific and the Bering Sea ice-free, coinciding with an increase in SSTs and primary productivity.
The YD-cold event experienced a rapid recovery of the spring sea-ice cover, in line with the cooling trend observed during this period. As shown by the sudden increase of IP25, an early and fast sea-ice advance led to extended sea-ice cover in the Bering Sea and variable sea ice in the western Pacific. However, the eastern subarctic Pacific remained ice-free during this cold interval.
A prominent decrease of IP25 and minimum values of PIP25 index, reflecting ice-free conditions in the subarctic Pacific and the Bering Sea during the Holocene Thermal Maximum, is in line with the SST warming. This is likely due to the maximum orbitally forced summer insolation, enhancing the melting of ice sheets and thus a sea-level rise. The resultant opening of Bering Strait (~ 11 ka) and the Aleutian passages allowed the inflow of Pacific Water into the Bering Sea. This warm-water inflow further enhanced the sea-ice melting, leading to ice-free conditions in the Bering Sea during this period.
During the late-Early/Mid Holocene, an increase in IP25 and PIP25 in sediments from the north-eastern Bering Sea reveals a shift to marginal sea-ice conditions that is in line with the recovery of extended sea ice observed in other Arctic marginal seas. The extension of Arctic sea ice during the Mid Holocene was likely triggered by the lower Northern Hemisphere insolation and a general widespread surface water-cooling.
This study is a contribution to the international INOPEX (Innovative North Pacific Experiment) Project funded by the German Ministry of Education and Science (Bundesministerium für Bildung und Forschung) and the German Research Foundation (DFG), project STE412/25. We thank the captain of the research vessel R/V Sonne, L. Mallon, as well as the crew members and the scientists on board for their remarkable work. Thanks also go to W. Luttmer and S. Wassmuth for technical support in the laboratory. We thank two anonymous reviewers for numerous constructive suggestions for improvement of the manuscript.
- 1.Ahagon N, Ohkushi K, Uchida M, Mishima T (2003) Mid-depth circulation in the northwest Pacific during the last deglaciation: evidence from foraminiferal radiocarbon ages. Geophys Res Lett 30(21):2097Google Scholar
- 2.Alley RB, Mayewski PA, Sowers T, Stuiver M, Taylor KC, Clark PU (1997) Holocene climatic instability: a prominent, widespread event 8200 year ago. Geology 25(6):483–486Google Scholar
- 3.Amo M, Minagawa M (2003) Sedimentary record of marine and terrigenous organic matter delivery to the Shatsky Rise, western North Pacific, over the last 130 ka. Org Geochem 34:1299–1312Google Scholar
- 4.Belt ST, Brown TA, Ampel L, Cabedo-Sanz P, Fahl K, Kocis JJ, Massé G, Navarro-Rodriguez A, Ruan J, Xu Y (2014) An inter-laboratory investigation of the Arctic sea ice biomarker proxy IP25 in marine sediments: key outcomes and recommendations. Clim Past 10:155–166Google Scholar
- 5.Belt ST, Brown TA, Navarro-Rodriguez A, Cabedo-Sanz P, Tonkin A, Ingle R (2012) A reproducible method for the extraction, identification and quantification of the Arctic sea ice proxy IP25 from marine sediments. Anal Methods 4:705–713Google Scholar
- 6.Belt ST, Brown TA, Ringrose AE, Cabedo-Sanz P, Mundy CJ, Gosselin M, Poulin M (2013) Quantitative measurement of the sea ice diatom biomarker IP25 and sterols in Arctic sea ice and underlying sediments: further considerations for palaeo sea ice reconstruction. Org Geochem 62:33–45Google Scholar
- 7.Belt ST, Cabedo-Sanz P, Smik L, Navarro-Rodriguez A, Berben SMP, Knies J, Husum K (2015) Identification of paleo Arctic winter sea ice limits and the marginal ice zone: optimised biomarker-based reconstructions of late Quaternary Arctic sea ice. Earth Planet Sci Lett 431:127–139Google Scholar
- 8.Belt ST, Massé G, Rowland SJ, Poulin M, Michel C, LeBlanc B (2007) A novel chemical fossil of palaeo sea ice: IP25. Org Geochem 38:16–27Google Scholar
- 9.Belt ST, Müller J (2013) The Arctic sea ice biomarker IP25: a review of current understanding, recommendations for future research and applications in palaeo sea ice reconstructions. Quat Sci Rev 79:9–25Google Scholar
- 10.Belt ST, Vare LL, Massé G, Manners HR, Price JC, MacLachlan SE, Andrews JT, Schmidt S (2010) Striking similarities in temporal changes to spring sea ice occurrence across the central Canadian Arctic Archipelago over the last 7000 years. Quatern Sci Rev 29(25–26):3489–3504Google Scholar
- 11.Berben SMP, Husum K, Navarro-Rodriguez A, Belt ST, Aagaard-Sørensen S (2017) Semi-quantitative reconstruction of early to late Holocene spring and summer sea ice conditions in the northern Barents Sea. J Quat Sci 32:587–603Google Scholar
- 12.Berben SMP, Husum K, Cabedo-Sanz P et al (2014) Holocene subcentennial evolution of Atlantic water inflow and sea ice distribution in the western Barents Sea. Clim Past 10:181–198Google Scholar
- 13.Blockley SPE, Lane CS, Hardiman M, Rasmussen SO, Seierstad IK, Steffensen JP, Svensson A, Lotter AF, Turney CSM, Ramsey CB, Members (2012) Synchronisation of palaeoenvironmental records over the last 60,000 years, and an extended event stratigraphy to 48,000 b2k. Quat Sci Rev 36, 2–10Google Scholar
- 14.Boon JJ, Rijpstra WIC, de Lange F, de Leeuw JW, Yoshioka M, Shimizu Y (1979) The Black Sea sterol—a molecular fossil for dinoflagellate blooms. Nature 277:125–127Google Scholar
- 15.Brassell SC, Eglinton G, Marlowe IT, Pflaumann U, Sarnthein M (1986) Molecular stratigraphy: a new tool for climatic assessment. Nature 320:129–133Google Scholar
- 16.Brown TA, Belt ST, Philippe B, Mundy CJ, Massé G, Poulin M, Gosselin M (2011) Temporal and vertical variations of lipid biomarkers during a bottom ice diatom bloom in the Canadian Beaufort Sea: further evidence for the use of the IP25 biomarker as a proxy for spring Arctic sea ice. Polar Biol 34:1857–1868Google Scholar
- 18.Brunelle BG, Sigman DM, Jaccard SL, Keigwin LD, Plessen B, Schettler G, Cook MS, Haug GH (2010) Glacial/interglacial changes in nutrient supply and stratification in the western subarctic North Pacific since the penultimate glacial maximum. Quat Sci Rev 29:2579–2590Google Scholar
- 19.Caissie BE, Brigham-Grette J, Lawrence KT, Herbert TD, Cook MS (2010) Last Glacial Maximum to Holocene sea surface conditions at Umnak Plateau, Bering Sea, as inferred from diatom, alkenone, and stable isotope records. Paleoceanography 25:PA1206Google Scholar
- 20.Cabedo-Sanz P, Belt ST, Knies J, Husum K (2013) Identification of contrasting seasonal sea ice conditions during the Younger Dryas. Quat Sci Rev 79:74–86Google Scholar
- 21.Cavalieri DJ, Parkinson CL (1987) On the relationship between atmospheric circulation and the fluctuations in the sea ice extents of the Bering and Okhotsk seas. J Geophys Res 92:7141–7162Google Scholar
- 22.Chappell J, Omura A, Esat T, Mcculloch M, Pandolfi J, Ota Y, Pillans B (1996) Reconciliation of late quaternary sea levels derived from coral terraces Huon Peninsula with deep sea oxygen isotope records. Earth Planet Sci Lett 141:227–236Google Scholar
- 23.Clark PU, Mix AC (2002) Ice sheets and sea level of the Last Glacial Maximum. Quat Sci Rev 21:1–7Google Scholar
- 24.Clarke GKC, Leverington DW, Teller JT, Dyke AS (2004) Paleohydraulics of the last outburst flood from glacial Lake Agassiz and the 8200 BP cold event. Quat Sci Rev 23(3–4):389–407Google Scholar
- 25.Close S, Houssais M-N, Herbaut C (2017) The Arctic winter sea ice quadrupole revisited. J Clim 30:3157–3167Google Scholar
- 26.Conte MH, Eglinton G, Madureira LAS (1992) Long-chain alkenones and alkyl alkenoates as palaeotemperature indicators: their production, flux and early sedimentary diagenesis in the eastern North Atlantic. Org Geochem 19:287–298Google Scholar
- 27.Cook MS, Keigwin LD, Sancetta CA (2005) The deglacial history of surface and intermediate water of the Bering Sea. Deep Sea Res II 52:2163–2173Google Scholar
- 28.Cronin TM, Gemery L, Briggs WM Jr, Jakobsson M, Polyak L, Brouwers EM (2010) Quaternary sea-ice history in the Arctic Ocean based on a new ostracode sea-ice proxy. Quat Sci Rev 29:3415–3429Google Scholar
- 29.Danielson S, Curchitser E, Hedstrom K, Weingartner T, Stabeno P (2011) On ocean and sea ice modes of variability in the Bering Sea. J Geophys Res 116:C12034Google Scholar
- 30.Darby DA, Ortiz JD, Grosch CE, Lund SP (2012) 1,500-year cycle in the Arctic Oscillation identified in Holocene Arctic sea-ice drift. Nat Geosci 5:897–900Google Scholar
- 31.de Vernal A, Eynaud F, Henry M, Hillaire-Marcel C, Londeix L, Mangin S, Matthiessen J, Marret F, Radi T, Rochon A, Solignac S, Turon J-L (2005) Reconstruction of sea-surface conditions at middle to high latitudes of the Northern Hemisphere during the Last Glacial Maximum (LGM) based on dinoflagellate cyst assemblages. Quat Sci Rev 24:897–924Google Scholar
- 32.de Vernal A, Gersonde R, Goosse H, Seidenkrantz M-S, Wolff EW (2013) Sea ice in the paleoclimate system: the challenge of reconstructing sea ice from proxies—an introduction. Quat Sci Rev 79:1–8Google Scholar
- 33.Dieckmann GS, Hellmer HH (2003) The importance of sea ice: an overview. In: Thomas DN, Dieckmann GS (eds) Sea ice. Blackwell Publishing, Oxford, pp 1–21Google Scholar
- 34.Eglinton G, Hamilton RJ (1963) The distributions of alkanes. In: Swain T (ed) Chemical plant taxonomy. Academic Press, New York, pp 187–217Google Scholar
- 35.Eglinton G, Hamilton RJ (1967) Leaf epicuticular waxes. Science 156:1322–1335Google Scholar
- 36.Elias SA, Short SK, Nelson CH, Birks HH (1996) Life and times of the Bering land bridge. Nature 382:60–63Google Scholar
- 37.Elias SA, Short SK, Birks HH (1997) Late Wisconsin environments of the Bering Land Bridge. Palaeogeogr Palaeoclimatol Palaeoecol 136:293–308Google Scholar
- 38.Fahl K, Stein R (1997) Modern organic carbon deposition in the Laptev Sea and the adjacent continental slope: surface-water productivity vs. terrigenous input. Org Geochem 26:379–390Google Scholar
- 39.Fahl K, Stein R (1999) Biomarkers as organic-carbon-source and environmental indicators in the Late Quaternary Arctic Ocean: problems and perspectives. Mar Chem 63(3–4):293–309Google Scholar
- 40.Fahl K, Stein R (2012) Modern seasonal variability and deglacial/Holocene change of central Arctic Ocean sea-ice cover: new insights from biomarker proxy records. Earth Planet Sci Lett 351–352:123–133Google Scholar
- 41.Gebhardt H, Sarnthein, M, Grootes, PM, Kiefer, T, Kuehn H, Schmieder F, Röhl U (2008) Paleonutrient and productivity records from the subarctic North Pacific for Pleistocene glaciations I to V. Paleoceanography 23:PA4212Google Scholar
- 42.Gersonde R (2012) The expedition of the research vessel “Sonne” to the subpolar North Pacific and the Bering Sea in 2009 (SO202-INOPEX). Report on polar and marine research, Bremerhaven, 643. ISSN: 1866-3192Google Scholar
- 43.Gorbarenko SA (1996) Stable isotope and lithologic evidence of late glacial and Holocene oceanography of the northwestern Pacific and its marginal sea. Quat Res 46:230–250Google Scholar
- 44.Gorbarenko SA, Nürnberg D, Derkachev AN, Astakhov AS, Southon JR, Kaiser A (2002) Magnetostratigraphy and tephrochronology of the Upper Quaternary sediments in the Okhotsk Sea: implication of terrigenous, volcanogenic and biogenic matter supply. Mar Geol 183:107–129 https://doi.org/10.1016/S0025-3227(02)00164-0 CrossRefGoogle Scholar
- 45.Gorbarenko SA, Wang P, Wang R, Cheng X (2010) Orbital and suborbital environmental changes in the southern Bering Sea during the last 50 kyr. Palaeogeogr Palaeoclim Palaeoecol 286:97–106Google Scholar
- 46.Hemming SR (2004) Heinrich events: massive late Pleistocene detritus layers of the North Atlantic and their global climate imprint. Rev Geophys 42(1):RG1005 (1–43) Google Scholar
- 48.Hopkins DM (1973) Sea level history in Beringia during the past 250,000 years. Quat Res 3:520–540Google Scholar
- 49.Hu A, Meehl GA, Otto-Bliesner BL, Waelbroeck L, Han W, Loutre MF, Lambeck K, Mitrovica JX, Rosenbloom N (2010) Influence of Bering Strait flow and North Atlantic circulation on glacial sea-level changes. Nat Geosci 3:118–121Google Scholar
- 50.Hu A, Meehl GA, Han W, Timmermann A, Otto-Bliesner B, Liu Z, Washington WM, Large W, Abe-Ouchi A, Kimoto M, Lambeck K, Wu B (2012) Role of the Bering Strait on the hysteresis of the ocean conveyor belt circulation and glacial climate stability. PNAS 109(17):6417–6422Google Scholar
- 51.Hunt GL, Phyllis Stabeno PJ, Walters G, Sinclair E, Brodeur RD, Napp JM, Nicholas A, Bond NA (2002) Climate change and control of the southeastern Bering Sea pelagic ecosystem. Deep Sea Res II 49:5821–5853Google Scholar
- 52.Katsuki K, Takahashi K, Okada M (2003) Diatom assemblage and productivity changes during the last 340,000 years in the subarctic Pacific. J Oceanogr 59:695–707Google Scholar
- 53.Katsuki K, Takahashi K (2005) Diatoms as paleoenvironmental proxies for seasonal productivity, sea-ice and surface circulation in the Bering Sea during the late Quaternary. Deep Sea Res II 52:2110–2130Google Scholar
- 55.Keigwin LD, Jones GA, Froelich PN (1992) A 15,000 year paleoenvironmental record from Meji Seamount far northwestern Pacific. Earth Planet Sci Lett 111:425–440Google Scholar
- 56.Keigwin LD, Donnelly JP, Cook MS, Driscoll NW, Brigham-Grette J (2006) Rapid sea-level rise and Holocene climate in the Chukchi Sea. Geology 34:861–864Google Scholar
- 58.Krawczyk DW, Witkowski A, Moros M, Lloyd JM, Høyer JL, Miettinen A, Kuijpers A (2017) Quantitative reconstruction of Holocene sea ice and sea surface temperature off West Greenland from the first regional diatom data set. Paleoceanography 32:18–40. https://doi.org/10.1002/2016PA003003 CrossRefGoogle Scholar
- 59.Kuehn H, Lembke-Jene L, Gersonde R, Esper O, Lamy F, Arz H, Kuhn G, Tiedemann R (2014) Laminated sediments in the Bering Sea reveal atmospheric teleconnections to Greenland climate on millennial to decadal timescales during the last deglaciation. Clim Past 10:2215–2236Google Scholar
- 60.Maier E, Méheust M, Abelmann A, Gersonde R, Chapligin B, Ren J, Stein R, Meyer H, Tiedemann R (2015) Deglacial subarctic Pacific surface water hydrography and nutrient dynamics and links to North Atlantic climate variability and atmospheric CO2. Paleoceanography 30:949–968Google Scholar
- 61.Manley WF (2002) Postglacial flooding of the Bering Land Bridge: a geospatial animation: INSTAAR, University of Colorado, vol 1. http://instaar.colorado.edu/QGISL/bering_land_bridge. Accessed 18 Oct 2017
- 62.Marlowe IT, Brassell SC, Eglinton G, Green JC (1984) Long chain unsaturated ketones and esters in living algae and marine sediments. Org Geochem 6:135–141Google Scholar
- 63.Marlowe IT, Brassell SC, Eglinton G, Green JC (1990) Long-chain alkenones and alkyl alkenoates and the fossil coccolith record of marine sediments. Chem Geol 88:349–375Google Scholar
- 64.Martin JH, Fitzwater SE (1988) Iron deficiency limits phytoplankton growth in the north-east Pacific subarctic. Nature 331:341–343Google Scholar
- 65.Massé G, Rowland SJ, Sicre M-A, Jacob J, Jansen E, Belt ST (2008) Abrupt climate changes for Iceland during the last millennium: evidence from high resolution sea ice reconstructions. Earth Planet Sci Lett 269(3–4):565–569Google Scholar
- 66.Max L, Riethdorf J-R, Tiedemann R, Smirnova M, Lembke-Jene L, Fahl K, Nürnberg D, Matul A, Mollenhauer G (2012) Sea-surface temperature variability and sea-ice extent in the subarctic northwest Pacific during the past 15,000 years. Paleoceanography 27:PA3213Google Scholar
- 67.Méheust M, Fahl K, Stein R (2013) Variability in modern sea surface temperature, sea ice and terrigenous input in the sub-polar North Pacific and Bering Sea: reconstruction from biomarker data. Org Geochem 57:54–64Google Scholar
- 68.Méheust M, Stein R, Fahl K, Max L, Riethdorf JR (2016) Sea-ice variability in the western North Pacific and Bering Sea during the past 18,000 years. Geo Mar Lett 36(2):1–11Google Scholar
- 69.Morley JJ, Robinson SW (1986) Improved method for correlating late Pleistocene/Holocene records from the Bering Sea: application of a biosiliceous geochemical stratigraphy. Deep Sea Res I 33(9):1203–1211Google Scholar
- 70.Müller J, Stein R (2014) High-resolution record of late glacial and deglacial sea ice changes in Fram Strait corroborates ice-ocean interactions during abrupt climate shifts. Earth Planet Sci Lett 403:446–455Google Scholar
- 71.Müller PJ, Kirst G, Rhuland G, von Storch I, Rosell-Melé A (1998) Calibration of the alkenone paleotemperature index UK′ 37 based on core-tops from the eastern South Atlantic and the global ocean (60°N–60°S). Geochim Cosmochim Acta 62:1757–1772Google Scholar
- 72.Müller J, Massé G, Stein R, Belt ST (2009) Variability of sea-ice conditions in the Fram Strait over the past 30,000 years. Nat Geosci 2:772–776Google Scholar
- 73.Müller J, Wagner A, Fahl K, Stein R, Prange M, Lohmann G (2011) Towards quantitative sea ice reconstructions in the northern North Atlantic: a combined biomarker and numerical modelling approach. Earth Planet Sci Lett 306:137–148Google Scholar
- 74.Müller J, Werner K, Stein R, Fahl K, Moros M, Jansen E (2012) Holocene cooling culminates in sea ice oscillations in Fram Strait. Quat Sci Rev 47:1–14Google Scholar
- 76.Nakatsuka T, Watanabe K, Handa N, Matsumoto E, Wada E (1995) Glacial to interglacial surface nutrient variations of Bering deep basins recorded by 13C and 15N of sedimentary organic matter. Paleoceanography 10:1047–1061Google Scholar
- 77.Niebauer HJ (1980) Sea ice and temperature variability in the eastern Bering Sea and the relation to atmospheric fluctuations. J Geophys Res 85:7507–7515Google Scholar
- 78.Niebauer HJ (1983) Multiyear sea ice variability in the eastern Bering Sea: an update. J Geophys Res 88:2733–2742Google Scholar
- 79.Niebauer HJ, Bond NA, Yakunin LP, Plotnikov VV (1999) An update on the climatology and sea ice of the Bering Sea. In: Loughlin TR, Ohtani K (eds) Dynamics of the Bering Sea, University of Alaska Sea Grant, Fairbanks, pp 29–59Google Scholar
- 80.Ohkushi K, Uchida M, Ahagon N, Mishima T, Kanematsu T (2004) Glacial intermediate water ventilation in the northwestern Pacific based on AMS radiocarbon dating. Nucl Instrum Methods Phys Res B 223–224:460–465Google Scholar
- 81.Okada M, Takagi M, Narita H, Takahashi K (2005) Chronostratigraphy of sediment cores from the Bering Sea and the subarctic Pacific based on paleomagnetic and oxygen isotopic analyses. Deep Sea Res II 52:2092–2109Google Scholar
- 82.Okazaki Y, Takahashi K, Asahi H, Katsuki K, Hori J, Yasuda H, Sagawa Y, Tokuyama H (2005) Productivity changes in the Bering Sea during the late Quaternary. Deep Sea Res II 52:2150–2162Google Scholar
- 83.Okazaki Y, Takahashi K, Katsuki K, Ono A, Hori J, Sakamoto T, Uchida M, Shibata Y, Ikehara M, Aoki K (2005) Late quaternary paleoceanographic changes in the southwestern Okhotsk Sea: evidence from geochemical, radiolarian, and diatoms records. Deep Sea Res II 52:2332–2350Google Scholar
- 84.Okazaki Y, Timmermann A, Menviel L, Harada N, Abe-Ouchi A, Chikamoto MO, Mouchet A, Asahi H (2010) Deepwater formation in the North Pacific during the Last Glacial Maximum. Science 329:200–204Google Scholar
- 85.Overland JE, Pease CH (1982) Cyclone climatology of the Bering Sea and its relation to sea ice extent. Mon Weather Rev 110:5–13Google Scholar
- 86.Overland JE, Stabeno PJ (2004) Is the climate of the Bering Sea warming and affecting the ecosystem? Eos Trans AGU 85(33):309–312Google Scholar
- 87.Pease CH (1980) Eastern Bering Sea ice processes. Mon Weather Rev 108:205–223Google Scholar
- 88.Polyak L, Alley RB, Andrews JT, Brigham-Grette J, Cronin TM, Darby DA, Dyke AS, Fitzpatrick JJ, Funder S, Holland M, Jennings AE, Miller GH, ORegan M, Savelle J, Serreze M, John K, White JWC, Wolff E (2010) History of sea ice in the Arctic. Quatern Sci Rev 29(15–16):1757–1778Google Scholar
- 89.Prahl FG, Wakeham SG (1987) Calibration of unsaturation patterns in long-chain ketone compositions for palaeotemperature assessment. Nature 330:367–369Google Scholar
- 90.Prahl FG, Muehlhausen LA, Zahnle DL (1988) Further evaluation of long-chain alkenones as indicators of paleoceanographic conditions. Geochim Cosmochim Acta 52:2303–2310Google Scholar
- 91.Rasmussen SO, Andersen KK, Svensson AM, Steffensen JP, Vinther BM, Clausen HB, Siggaard-Andersen M-L, Johnsen SJ, Larsen LB, Dahl-Jensen D, Bigler M, Röthlisberger R, Fischer H, Goto-Azuma K, Hansson ME, Ruth U (2006) A new Greenland ice core chronology for the last glacial termination. J Geophys Res 111:D06102Google Scholar
- 92.Riethdorf J-R, Max L, Nürnberg D, Lembke-Jene L, Tiedemann R (2013) Deglacial development of (sub) sea surface temperature and salinity in the subarctic northwest Pacific: implications for upper-ocean stratification. Paleoceanography 28:91–104Google Scholar
- 93.Sagawa T, Ikehara K (2008) Intermediate water ventilation change in the subarctic northwest Pacific during the last deglaciation. Geophys Res Lett 35:L24702Google Scholar
- 94.Sancetta C (1982) Distribution of diatom species in surface sediments of the Bering and Okhotsk seas. Micropaleontology 28:221–257Google Scholar
- 95.Sancetta C, Robinson SW (1983) Diatom evidence on Wisconsin and Holocene events in the Bering Sea. Quat Res 20:232–245Google Scholar
- 96.Sarnthein M, Van Kreveld S, Erlenkeuser H, Grootes PM, Kucera M, Pflaumann U, Schulz M (2003) Centennial-to-millennial-scale periodicities of Holocene climate and sediment injections off the western Barents shelf, 75 1N. Boreas 32(3):447–461Google Scholar
- 98.Serno S, Winckler G, Anderson RF, Maier E, Ren H, Gersonde R, Haug GH (2015) Comparing dust flux records from the Subarctic North Pacific and Greenland. Implications for atmospheric transport to Greenland and for the application of dust as a chronostratigraphic tool. Paleoceanography 30:583–600Google Scholar
- 99.Schlung SA, Ravelo AC, Aiello IW, Andreasen DH, Cook MS, Drake M, Dyez KA, Guilderson TP, LaRiviere JP, Stroynowski Z, Takahashi K (2013) Millennial-scale climate change and intermediate water circulation in the Bering Sea from 90 ka: a high-resolution record from IODP Site U1340. Paleoceanography 28:54–67, https://doi.org/10.1029/2012PA002365 CrossRefGoogle Scholar
- 100.Schumacher JD, Reed RK (1992) Characteristics of currents over the continental slope of the eastern Bering Sea. J Geophys Res 97:9423–9433Google Scholar
- 101.Schumacher JD, Bond NA, Brodeur RD, Livingston PA, Napp JM, Stabeno PJ (2003) Climate change in the southeastern Bering Sea and some consequences for biota. In: Hempel G, Sherman K (eds) Large marine ecosystems of the world. Elsevier, Amsterdam, pp 17–40Google Scholar
- 102.Sikes EL, Volkman JK, Robertson LG, Pichon J-J (1997) Alkenones and alkenes in surface waters and sediments of the Southern Ocean: implications for paleotemperature estimation in polar regions. Geochim Cosmochim Acta 61:1495–1505Google Scholar
- 103.Smik L, Cabedo-Sanz, Belt ST (2016) Semi-quantitative estimates of paleo Arctic sea ice concentration based on source-specific highly branched isoprenoid alkenes: a further development of the PIP25 index. Org Geochem 92:63–69Google Scholar
- 104.Smirnova MA, Kazarina GK, Matul AG, Max L (2015) Diatom evidence for paleoclimate changes in the Northwestern Pacific during the last 20000 years. Oceanology 15(3):383–389Google Scholar
- 105.Stabeno PJ, Reed RK (1994) Circulation in the Bering Sea basin by satellite tracked drifters. J Phys Oceanogr 24:848–854Google Scholar
- 106.Stabeno PJ, Bond NA, Kachel NB, Salo SA, Schumacher JD (2001) On the temporal variability of the physical environment over the south-eastern Bering Sea. Fish Oceanogr 10(1):81–98Google Scholar
- 107.Stabeno PJ, Bond NA, Hermann AJ, Kachel NB, Mordy CW, Overland JE (2004) Meteorology and oceanography of the Northern Gulf of Alaska. Cont Shelf Res 24:859–897Google Scholar
- 108.Stabeno PJ, Bond NA, Salo SA (2007) On the recent warming of the southeastern Bering Sea shelf. Deep Sea Res 54:2599–2618Google Scholar
- 111.Stein R, Fahl K, Müller J (2012) Proxy reconstruction of Arctic Ocean sea ice history: from IRD to IP25. Polarforschung 82:37–71Google Scholar
- 112.Stein R, Fahl K, Schade I, Manerung A, Wassmuth S, Niessen F, Nam S-I (2017) 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 Quat Sci 32:362–379Google Scholar
- 115.Stroeve JC, Serreze MC, Holland MM, Kay JE, Malanik J, Barrett AP (2012) The Arctic’s rapidly shrinking sea ice cover: a research synthesis. Clim Change 110:1005–1027Google Scholar
- 116.Tanaka S, Takahashi K (2005) Late quaternary paleoceanographic changes in the Bering Sea and the Western subarctic Pacific based on radiolarian assemblages. Deep Sea Res II 52:2131–2149Google Scholar
- 117.Ternois Y, Kawamura L, Keigwin L, Ohkouchi N, Nakatsuka T (2001) A biomarker approach for assessing marine and terrigenous inputs to the sediments of Sea of Okhotsk for the last 27,000 years. Geochim Cosmochim Acta 65(5):791–802Google Scholar
- 118.Vare LL, Massé G, Gregory TR, Smart CW, Belt ST (2009) Sea ice variations in the central Canadian Arctic Archipelago during the Holocene. Quat Sci Rev 28:1354–1366Google Scholar
- 119.Volkman JK, Eglinton G, Corner EDS, Forsberg TEV (1980) Long-chain alkenes and alkenones in the marine coccolithophorid Emiliania huxleyi. Phytochemistry 19:2619–2622Google Scholar
- 120.Volkman JK (1986) A review of sterol markers for marine and terrigenous organic matter. Org Geochem 9:83–89Google Scholar
- 121.Xiao X, Fahl K, Müller J, Stein R (2015) Sea-ice distribution in the modern Arctic Ocean: biomarker records from trans-Arctic Ocean surface sediments. Geochim Cosmochim Acta 155:16–29Google Scholar
- 123.Wang R, Xiao W, Li Q, Chen R (2006) Polycystine radiolarians in surface sediments from the Bering Sea Green Belt area and their ecological implication for paleoenvironmental reconstructions. Mar Micropaleontol 59:135–152Google Scholar
- 124.Wanner H, Beer J, Butikofer J, Crowley TJ, Cubasch U, Fluckiger J, Goosse H, Grosjean M, Joos F, Kaplan JO, Kuttel M, Muller SA, Prentice IC, Solomina O, Stocker TF, Tarasov P, Wagner M, Widmann M (2008) Mid- to Late Holocene climate change: an overview. Quat Sci Rev 27:1791–1828Google Scholar