Climate and ice conditions of East Siberian Sea during Holocene: Reconstructions based on sedimentary geochemical multiproxy

Abstract

Reconstruction of ice conditions and climate of the East Siberian Sea during the Late Middle Holocene was implemented for the first time using a transfer function method based on marine sedimentation proxies. Transfer functions were developed based on a hydrometeorological time series of changes in mean annual air temperature (ΔT₁₀) and the ice-free period (IF₁₀) for the last 70–90 years and a geochemical time series of bottom sediments accumulated during this time, which were obtained using a submillimeter scanning of the core in the X-ray fluorescence analysis of synchrotron radiation. Geochemical time series of Holocene deposits were obtained from X-ray fluorescence analysis of the core in a step of 1–2 cm. Reconstruction of ice conditions and average annual air temperature revealed a decreasing trend of ΔT₁₀ and IF₁₀, which started from the middle Holocene up to the Little Ice Age due to an orbitally determined decrease in insolation. A synchronous periodicity of changes in ice cover and air temperature was revealed and approximated by periods of 1600 years for IF10 and 1740 years for ΔT₁₀, which are comparable to Bond’s climatic cycles. These results indicates the telecommunication of the Atlantic processes, climate of the East Siberian Sea, and predominance of the cyclonic type of atmospheric circulation during the Holocene.

Keywords

East Siberian Sea Holocene paleoclimatology sea ice sediment chemistry surface air temperature

Introduction

In recent decades and throughout the instrumental era, rapid changes in the Arctic temperature regime (Neukom et al., 2019; Räisänen, 2001; Stone, 1997) and the degradation of ice cover (Brennan et al., 2020; Stroeve et al., 2007; Walsh et al., 2017; Wood et al., 2015) have garnered significant interest of the world scientific community for studying changes in the climate-dependent components of the natural environment at high latitudes (sea ice, snow cover, river runoff, coastal abrasion, glaciers, and permafrost) and their relationship with climate change. Unfortunately, instrumental hydrometeorological observations in the Arctic seas began only in the last 60–100 years, which makes it impossible to simultaneously consider many of the main cycles of climate change. Therefore, it is necessary to understand the causes of the current changes in the natural environment, including the data on their changes that have been reconstructed from various natural records. The longest records of the natural environment can be obtained from sedimentary layers, which makes them a necessary element for paleoclimate reconstruction.

Most of the climate reconstructions performed in the Arctic have used absolute or relative changes in surface air temperature (SAT) to characterize the climate, while variations over the instrumental era have been well studied (Brennan et al., 2020; Johannessen et al., 2004; Moritz et al., 2002; Stott et al., 2000; Walsh et al., 2017). Over the past centuries, SAT variations in the Arctic have been reconstructed using continental proxies (Crespin et al., 2009; Neukom et al., 2019), while arctic-wide temperature reconstruction models have been compiled at various scales over the past centuries and millennia according to the season (McKay and Kaufman, 2014; Overpeck et al., 1997). Additionally, no reconstruction of Holocene temperatures has been conducted in marine areas.

Reconstructions of Arctic ice conditions are based on various proxies in marine sediments, dinocyst species composition (de Vernal, 2017; Farmer et al., 2011; Stroeve et al., 2007), and other biomarkers (Hörner et al., 2016; Polyak et al., 2016; Stein et al., 2017; Su et al., 2022). However, these methods provide only a qualitative assessment of ice condition changes, without a comparison with digital hydrometeorological data. A quantitative reconstruction of Arctic ice cover over the past two millennia (Kinnard et al., 2011) was performed using a set of proxies on the adjacent coasts (Bonnet et al., 2010) by comparing them with satellite instrumental data on changes in Arctic ice cover over the period of observation as well as development transfer functions. Previous studies have established that in addition to the ice cover in the late 20th and early 21st centuries, a strongly reduced ice cover occurred during the 15th century Arctic warming (Crespin et al., 2009) and during the pre-industrial minimum at approximately AD 640 in the Fram Strait (Bonnet et al., 2010).

The method of “transfer functions” was used by the authors of this article to reconstruct the duration of the ice-free period (IF₁₀) and anomalies of the mean annual air temperature (∆T₁₀) over the past centuries and millennia at several stations in the Chukchi Sea (Astakhov et al., 2019a, 2020). Transfer functions were developed by comparing the time series of the chemical composition of the bottom sediments accumulated over the observation period with the hydrometeorological data. The choice of chemical composition was determined by the technical possibility of its determination, with the required discreteness (fractions of a millimeter) and accuracy (Astakhov et al., 2021). Currently, this method is the only method that allows SAT reconstruction in marine areas.

Few paleoreconstructions of the Holocene climate and ice conditions in the East Siberian Sea (Stein et al., 2017) have been conducted, owing to the difficulty in obtaining cores with normal sedimentation that are suitable for a detailed study. Because of the ice and iceberg impact on the bottom, cores often exhibit loss of layers, specific mixing patterns, and duplication of individual layers (Keskitalo et al., 2017; Rekant et al., 2015). The effects of bottom erosion by ice were studied in core GC58 (Figure 1).

Figure 1.

Locations of station LV77-36 and previously studied cores with elements of the recent oceanography of the East Siberian Shelf (a), seismic profiles obtained by GeoPulse Subbottom Profilier (b and c), and its location (d). On the sea: the yellow dotted line shows the average minimum ice boundary in September 1981–2010 (NSIDS); red lines with arrows - Pacific water inflow (Stein et al., 2017); blue arrows – Siberian Coastal current; pink arrows – direction of ice transfer (Serreze et al., 2016); black dotted and white lines – coastlines 8.4 and 7.2 thousand years ago, respectively (Dudarev et al., 2016); green dotted line – paleoIndigirka Valley; wide dotted yellow line – boundary of Pacific-derived and Atlantic-derived shelf; isobaths 50, 100, 200, 2000 m are shown. On the coast, wide purple lines show areas of intense coastline erosion (>0.5 m/year) (Lantuit, 2012); hatch - coastal areas containing ice complex deposits (yedoma) (Ershov, 1989). For blocks (b and c) the vertical axis represents two-way travel time (ms).

The aim of this study was to reconstruct the climate (SAT) and ice conditions (duration of the ice-free period) of the Holocene in the western part of the East Siberian Sea by using a method in the literature called “transfer functions.” For this purpose, cores were specially obtained on the inner shelf at the bottom of the paleovalley of the Indigirka River (in the zone with the absence of ice/iceberg gouging), as determined from high-resolution seismic profiling data (Figure 1).

Material and methods

A 376 cm long core, LV77-36-1 (74°05.279′N; 155°35.872′W, sea depth 36 m), was obtained using a gravity corer during the Russian-Chinese expedition of the 77th cruise of the R/V Akademik M.A. Lavrentyev in 2016. At the same station (74°04.743′N; 155°36.449′W), a 34 cm long core, LV77-36-2, was obtained using a multicorer for performing a detailed study on the surface layer of sediments (Figure 1).

The age of the LV77-36-1 core was determined according to a age model developed based on calibrated AMS ¹⁴C radiocarbon dates of mollusk shells (Dong et al., 2022). The age of the LV77-36-2 sediments was determined from the distribution of the non-equilibrium (atmospheric) ²¹⁰Pb isotope. Measurements of ²¹⁰Pb and ¹³⁷Cs radioactivity in the sediments from the LV77-36-2 core (Supplemental Table SD1, available online) were conducted according to a method in the literature (Astakhov et al., 2019a, 2021). The age model of the upper part of the core was based on the age of individual horizons determined using the ²¹⁰Pb constant rate of supply model (Appleby et al., 1979; Melgunov et al., 2019) (Supplemental Table SD1, available online). For the lower part of the core, age was conventionally assumed based on the minimum detected sedimentation rate in the dated interval of the core (0.44 mm/year).

The chemical composition of LV77-36-2 sediments was studied via X-ray fluorescence scanning by using synchrotron radiation (XRF SR) of the storage ring (VEPP-3; the Budker Institute of Nuclear Physics, Novosibirsk) for specially prepared monoliths of sediments (Darin et al., 2013). The scanning step was 0.51 mm (Figure 2, Supplemental Tables SD2, SD3, available online). The concentrations of Ca, K, Ti, Mn, Fe, Ni, Zn, Ga, Pb, Rb, Sr, Y, Zr, V, Cr, As, Br, Nb, and Mo were determined along with the X-ray density of the samples (XRD). The limit of detection for elements was 0.5 g/t (Br, Rb, Sr, Nb), 1 g/t (Zr, Y), 2 g/t (Zn), 5 g/t (Ni, Mn, Pb), 10 g/t (Fe), 15 g/t (Ti), and 100 g/t (Ca, K).

Figure 2.

Variation in rubidium normalized contents of some lithogenic, redox-sensitive, and biogenic elements in core LV77-36-2, and reconstruction of the ice-free period duration (IF₁₀) and the annual air temperature anomalies (∆T₁₀). The red line shows the measured IF₁₀ and ∆T₁₀ mean for 10-year intervals; blue dashed bold lines show variations in reconstructed IF₁₀, and ∆T₁₀; thin dashed black lines show boundaries of the 95% confidence interval reconstructions. On the diagram with element/Rb, thin blue lines show variations in the contents according to a scan with a 0.5 mm pitch; bold brown, purple, and green lines show 9-point running average values for lithogenic, redox-sensitive, and biogenic elements, respectively. The diagram shows the content of grain size fractions coarser than 16 µm; the yellow fill is the sediment layer with anomalous chemical composition due to silt and sand enrichment.

The chemical composition of core LV77-36-1 sediments was determined by the X-ray fluorescence method on the energy dispersion spectrometer ARL Quant’X, (Thermo Fisher) using SGH5, MAN, JH-1 standards and control analyzes by other methods (Astakhov et al., 2020) from discrete samples with an interval of 1–2 cm (Figure 3, Supplemental Table SD4, available online). The upper 150 cm of the core (5.5 ka) was characterized by a low sedimentation rate (Dong et al., 2022) and analyzed with a step of 1 cm; however, the lower 150 cm of the core exhibited a step of 2 cm. Other sediment characteristics (grain size, TOC, and opal), total diatom content, and diatom assemblages were studied using a method in the literature (Astakhov et al., 2020; Tsoy et al., 2017) with a step of 2–4 cm.

Figure 3.

Variations in the chemical composition of LV77-36-1 core sediments, diatom content, and the results of reconstruction of the duration of the ice-free period (IF₁₀) and the average annual air temperature anomalies (∆T₁₀). Thin blue lines show variations in reconstructed IF₁₀ and ∆T₁₀; bold red and green lines show 9-points running average values; layers: 1 – delta or lagoon deposits, 2 – shallow bay deposits; 3 – deposits of the inner shelf; postglacial sea level position (PSL) (Hörner et al., 2016).

Information on the duration of the ice-free period was obtained from the electronic climatic oceanographic atlas of the Arctic Ocean (Tanis and Timokhov, 1997, 1998) and from the website www.natice.noaa.gov/products/miz.html. This data were used for recorded the average ice edge for 10-day periods (decades) (Astakhov et al., 2021; Plotnikov and Pustoshnova, 2012). The main climatic parameter of the reconstruction was the variation in the mean annual air temperature (ΔT₁₀); the deviation smoothed over a 10-year interval (average value for the period from 1886 to 2005) (Hansen et al., 2010) was sourced from the global open-access climate database KNMI GISTEMP1200 (https://climexp.knmi.nl/plot_atlas_form.py) (Trouet and Van Oldenborgh, 2013).

Results

Features of element contents and sedimentary environments

Sediments of cores LV77-36-1 and LV77-36-2 are clayey silts, with a relatively small admixture of sand (0.5%–4.8%); they are notable for their slight variability along the core. In the LV77-36-1 core, the mean median size varies from 5.6 to 8.8 µm and continues to increase up to the top of the core (Dong et al., 2022). The finest-grained sediments with minimum admixture of sand accumulated up to 7000 years ago. In sediments accumulated after 5 ka that exhibited a general trend of increasing sand content and median diameter, interlayers with different silt/clay ratios alternated (Dong et al., 2022).

According to the chemical composition of the LV77-36-1 sediments, an anomalous layer was revealed in the interval of 49–59 mm (1910–1940); low K, Ca, Ga, and Ti contents; and high Fe, As, and Ni contents (Figure 2). It exhibited increased sand and silt contents (fraction >16 µm, Figure 2) and comprised an admixture of iron hydroxides in its upper part, as revealed by the smear-slide analysis. This “atypical” layer was not used to develop the transfer function and reconstructions.

Based on the chemical composition of the LV77-36-1 sediments (Figure 3), at least three layers accumulated under different conditions were distinguished. Sediments with an age of more than 8.2 ka (layer 1, Figure 3) are characterized by sharply anomalous Fe and S contents at elevated TOC levels, which indicates their formation under anoxic conditions. Taking note of the sea level change curve (Figure 3) and paleogeographic reconstructions for this period (Figure 1), we can conclude that sediments of layer 1 accumulated in the semi-closed estuary of paleo-Indigirka. It is confirmed by a low bromine content indicates low water salinity (Astakhov et al., 2015; Mayer et al., 2007), and an anomalously high sedimentation rate (Dong et al., 2022).

From 8.2 to 7.2 ka (layer 2, Figure 3), sediments accumulated at a high rate (Dong et al., 2022) in a basin with low salinity (Figure 3, Br diagram) at a low relative to the present-day sea level; they are distinguished by an anomalously high content of P and increased content of As, Fe, and TOC. The total diatom content was minimal, and their assemblages were dominated by marine species (60%–100%) with a small number of freshwater species. Diatom and geochemical data suggests that sediments were accumulated in the semi-closed bay of the mouth shelf.

Sediments of layer 3 (from 7.2 thousand years ago until now) accumulated at a sea level close to the modern one (Figure 3) under modern conditions (inner shelf). They are characterized by a relatively homogeneous chemical composition (Figure 3). There was a stable and uniform increase in the Br (increase in salinity), K, and Zr contents, while the Fe, As, and S contents decreased.

Starting from 5 ka, the content of diatoms in the sediments increased, especially in the last 3 ka. The Th. hyperborea content (Figure 3), which is typically found in ice communities, the subglacial water layer of the study area, and the surface sediments of the southeastern part of the Laptev Sea (Obrezkova et al., 2014, 2019), increased synchronously. Additionally, the content of the marine benthic-neritic species Paralia sulcata, which dominates to the north in the zone of influence of Atlantic waters (Ilyash and Zhitina, 2009) and in the eastern part of the Chukchi Sea (Obrezkova et al., 2014), remains stable. Thus, diatom analysis data may indicate an increase in the ice conditions, which is consistent with the results of the reconstruction of the duration of the ice-free period (Figure 3).

Development of transfer functions

Transfer functions for reconstruction of SAT (ΔT₁₀) and ice conditions (IF₁₀) were developed according to a method in the literature (Astakhov et al., 2019a, 2021), based on the chemical composition of the upper part of the core LV77-36-2 scanned with a step of 0.5 mm (Figure 2).

The following steps to create transfer functions ΔT₁₀ and IF₁₀ based on this data are used:

- Develop multiple year by year time series of hydrometeorological data for the observation period (1950–2016 for IF₁₀ and 1934–2016 for ΔT₁₀), and implement their smoothing using the moving average method with a window of 10.

- Create multiple timescale geochemical series of the LV77-36-2 core for normalized elements in rubidium (Figure 2) and normalized minimax (Astakhov et al., 2020); then, smooth them for 10-year intervals, transforming the time scale into a yearly integer-timescale geochemical series.

- Compare the meteorological and geochemical time series using multiple regression methods (Astakhov et al., 2019a, 2021; Babich, 1980; Babich et al., 2016; Kalugin et al., 2007) and the development of transfer functions.

- Implement reconstruction using the developed transfer functions of IF₁₀ and ΔT₁₀ for the core interval 1800–2016; then, calculate the correlation coefficient between the reconstructed and measured values of the parameters (K_k) and the confidence interval of the reconstructions (ϭ95).

As a result, transfer functions for the reconstruction of the duration of the ice-free period (IF₁₀) were developed and tested as follows:

(1) IF₁₀ = −17.522*K/Rb_m + 8.265*Fe/Rb_m + 3.678*Br/Rb_m−0.897*Nb/Rb_m−0.203*Y/Rb_m + 8,337

The anomalies of mean annual air temperature (ΔT₁₀) are as follows:

(2) ΔT₁₀ = 3.525*Fe/Rb_m−2.031*Nb/Rb_m − 1.525*Zr/Rb_m + 1.332*Br/Rb_m + 1,28

where El/Rbm is the ratio of the elemental content to the rubidium content normalized by minimax (reduced to a value from 0 to 1). Kk for ΔT10 was 0.83, with a ϭ95 value of 0.73°C (0.90% and 0.5°C); meanwhile, Kk for IF10 was 0.89 and 0.93 days*10.

Reconstructions of SAT and ice conditions in the Middle and Late Holocene

Reconstruction of ΔT₁₀ and IF₁₀ was performed based on the chemical composition of sediments from core LV77 to 36-1, which was studied using discrete samples (Figure 3). For the reconstruction based on the available core age model (Dong et al., 2022), age was determined for each sample and normalized for rubidium and minimax (Supplemental Table SD3, available online). Thus, a temporary geochemical time series was created, including the elements in formulas (1) and (2), and the values of ΔT₁₀ and IF₁₀ were determined for each sample. Considering the large differences in the sedimentation rates in the core (Dong et al., 2022), the details of the reconstruction on the time scale differed (Figure 3) when considering the bottom of the core. To improve clarity, the values of ΔT₁₀ and IF₁₀ were smoothed using the nine-point running average method; consequently, subperiodic variability was clearly manifested in the distribution of ΔT₁₀ and IF₁₀ (Figure 4c and d). The spectral decomposition of their time series, which was performed using the correlogram additive decomposition method (Babich et al., 2016), revealed low-frequency quasi-periodicities approximated by periods of 1600 years for IF10 and 1730 years for ΔT₁₀. The second quasi-periodicity was approximated by periods of 225 and 385 years.

Figure 4.

Reconstruction of ice extent (d, e, f, g, h) and anomalies of air temperature at the surface (a–c) for the Middle and Late Holocene in the Arctic, based on the results of the study of cores LV77-36-1 and LV77-36-2 (c and d) and other sources. (a) reconstruction of air temperature anomalies on the surface (SAT) (Marcott et al., 2013); (b) pollen-based reconstruction of summer air temperatures (SAS) of lake Dolgoe (Lena River Delta) (Klemm et al., 2013); (c) anomalies of average annual air temperature (ΔT₁₀): results of observations (red line) and reconstruction on cores LV77-36-2 (green line) and LV77-36-1 (blue thin and blue bold (9 – point running average) lines); (d) duration of ice-free period at station LV77-36 based on observations (red line) and reconstruction on cores LV77-36-1 (blue and purple [9 – point running average] lines) and LV77-36-2 (green line); (e) the linearity-detrended IP₂₅ record (green) and the band-pass filtered 1800 years cycle (red) in core PS51/159 from Laptev Sea (Hörner et al., 2016); (f) biomarker PIP₂₅ values for core PS72/350 from East Siberian Sea and ARA2B-1A from Chukchi Sea (Stein et al., 2017); (g) reconstruction of ice-free period duration (IF₁₀) on core LV77-3 from Chukchi Sea (Astakhov et al., 2020); (h) reconstruction of Arctic ice cover for the last 1450 years (Kinnard et al., 2011). The yellow fill highlights the periods of increased ice coverage (cold stages of the 1800-year Bond cycles) of the Laptev Sea (Hörner et al., 2016).

Discussion

Features of transfer functions

The transfer functions for the reconstruction of ice cover and air temperature based on the variations in the elemental composition of sediments in the core LV77-36-2 (Figure 2), include biogenic (Sr), redox-sensitive (Fe), and lithogenic (K, Nb, Zr, Y) elements. The main processes responsible for the ice cover effect on the elemental composition of sediments accumulated on the East Siberian shelf were characterized earlier (Astakhov et al., 2019a, 2021, 2022; Su et al., 2022). They are changes in the primary bioproductivity of surface waters, as a result of a longer vegetation period, and changes in the redox conditions of bottom waters toward more oxide ones due to better ventilation of the water column with a decreased ice cover. It has also been established that the variability of the ice cover in the region is largely determined by atmospheric temperature variations (Alley et al., 2001; Astakhov et al., 2022; Hoff et al., 2016; Kinnard et al., 2011). Accordingly, sediments will be enriched by biogenic elements (Br, Ca, Sr) and redox-sensitive elements of oxide environments (Fe, As, Mn) during periods of a decrease in the sea-ice cover, which usually occurs with an increased average annual air temperature. These elements in various proportions are usually included with a positive sign in the transfer functions ΔT₁₀ and IF₁₀ developed for different seas (Astakhov et al., 2019a, 2021), as formulas 1 and 2 of this work. Redox-sensitive elements of anoxic environments (Zn, V, Ni, Cr), the conditions for the accumulation of which in sediments arise with an increased ice cover, and lithogenic elements of clay (K, Ga) and clastic (Zr, Nb, Y) minerals usually include to the transfer functions with a negative sign (Astakhov et al., 2021, 2022).

Such a value of biogenic and redox-sensitive elements to the transfer functions is observed everywhere, but with some variations. The bromine is the main biogenic element in the formulas for the Chukchi and East Siberian Seas, and in the Laptev Sea, where the water is highly desalinated by the Lena River inflow, it replace by strontium (Astakhov et al., 2021). The accumulation of bromine in sediments strongly depends on water salinity (Mayer et al., 2007), so bioproductivity variations is lesser significant in its distribution in the Laptev Sea. At the station LV77-3 from the southern part of the Chukchi Sea, biogenic elements were not included in the transfer functions, which is explained by the location of the station near the Bering Strait. Variations in bioproductivity here are determined by the changes of the temperature and volume of Pacific waters entering through the strait, and not by the ice coverage (Astakhov et al., 2020).

The influence of climate and sea-ice cover on the accumulation of lithogenic elements in sediments is determined by complex and diverse processes, including hydraulic differentiation of matter during sedimentation, different composition of terrigenous matter supplied from land, ice rafting, etc. The most common lithogenic element in the transfer functions ΔT₁₀ and IF₁₀ is potassium, an element of clay minerals in the finest-grained fraction of bottom sediments. With a long ice-free period, the wave mixing of shelf waters intensifies, and fine-grained material enriched with potassium is transported out from the shelf (Astakhov et al., 2021). Thus, potassium is more responsive to changes in ice cover than in air temperature. In our case, it is part of the transfer function IF₁₀ (formula 1), but did not part of the ΔT₁₀ (formula 2).

The lithogenic elements of clastic minerals in our case (formulas 1, 2) and at some other stations (Astakhov et al., 2021) are included in the transfer functions with a negative sign. It is explained by a more intensive accumulation of fine-grained river runoff fractions enriched with these elements (Rachold, 1999) under the ice, as in the case of potassium, although there are some regional differences. In the Laptev Sea at stations LV83-16 and LV83-32 (Figure 1), niobium and yttrium enter the IF₁₀ formulas with a positive sign (Astakhov et al., 2021), that is, their accumulation in sediments increases with a decrease in ice cover. Previously (Astakhov et al., 2021), it was explained by the proximity of these stations to the coasts, where the ice complex (yedoma) is intensively eroded (Figure 1). During the warming and a decrease of ice cover, the rate of coastal abrasion increases and silty material enriched with stable heavy minerals accumulated on the shelf. At the same time, in the eastern part of the Chukchi Sea, these elements, along with strontium, are included in the ΔT₁₀ formulas with a positive sign, which was explained by an increase in the input of ice/iceberg rafted material from the Canadian archipelago during warm periods (Astakhov et al., 2021, 2022).

The specified variety of conditions, which is necessary to preserve in the bottom sediments a signal of changes in ice coverage and, moreover, in average annual air temperature, determines the impossibility of developing a uniform set of elemental proxies applicable in various settings. The accumulation of most of the elements included in the transfer functions and their distribution in bottom sediments on a spatial scale are determined by processes whose geochemical signal significantly exceeds the ice cover and air temperature signals. For bromine (the biogenic element most often included in the transfer functions) the main factor of lateral distribution is the salinity of sea waters. It increases from river mouths toward the ocean, that is, from south to north, and the spatial distribution of bromine in modern sediments can be directly opposite to bioproductivity data. The distribution of redox-sensitive elements is determined by the change in reducing conditions on the shelf to oxidizing ones in deep-sea basins and, partially, on the outer shelf. The main factors in the lateral distribution of lithogenic elements are the grain-size composition of sediments (decreases from south to north) and differences in the composition of material supplied from the continent. It is known that the contents of Zr, Nb, and Y in sediments decrease from west to east (Astakhov et al., 2019b). An exception is the spatial distribution of such biogenic elements as calcium and strontium in shelf sediments (Supplemental Figure SD1, available online). It is largely determined by bioproductivity, the dependence of which on ice conditions was shown earlier (Su et al., 2022). However, the use of these elements as the only proxies on the East Siberian shelf is also incorrect, since changes in redox conditions are important in their accumulation, which determine the intensity of dissolution of biogenic carbonate fossils. This may be a possible reason for the absence of these elements in the transfer functions (formulas 1 and 2) for station LV77-36.

Variations in ice conditions

Considering that changes in sedimentation conditions caused by the Holocene transgression could affect the accuracy of paleoreconstructions based on the chemical composition of sediments, reconstructions over a time interval of 0–7 ka were used for comparison with known regional and global variations in these parameters (Figure 4). This time interval corresponds to the period of the current sea level position determined by Horner and coauthors in the Laptev Sea (Hörner et al., 2016). By this time, the opening of the straits between the mainland and the New Siberian Islands had been completed, and a modern system of currents with alongshore transport (Figure 1) had formed from the Laptev Sea and from the Lena River delta to the East Siberian Sea.

The IF₁₀ variations reconstructed from core LV77-36-1 (Figure 3), which were supplemented by instrumental observations and reconstruction data for the last 200 years from core LV77-36-2 (Figure 4d), demonstrated a general trend of increasing ice cover (decrease in IF₁₀) from the average (Figure 4d) and even the early Holocene (Figure 3), which corresponds with the insolation record (Laskar et al., 2004). This result was confirmed by variations in some indirect indicators: an increase in the content of ice-rafted debris (IRD) in sediments (Dong et al., 2022) and cryophilic diatom species (Figure 3).

The established periodicity of IF₁₀ changes, which is approximated by a period of 1600 years, has been observed to be similar to the variability of ice conditions (with a periodicity of 1800 years) identified in the cores of the Laptev Sea (Figure 4e). Hörner et al. (2016) compared it with the Bond climate cycles, which were identified in the North Atlantic by using the IRD composition in sediment cores and either exhibited a duration of 1470 ± 500 years (Bond et al., 1997) or an 1800-year component 5000-year cycle of lunisolar oceanic tides (Keeling and Whorf, 2000). The periodicity of changes in the ice conditions during the middle and Late Holocene, which exhibits close cyclicity and correlation with climate variations, can be assumed from the distribution of IP₂₅ in the sediments of core PS72/350 from the northern part of the East Siberian Sea (Figure 4f). Subsynchronous changes in IF₁₀ and ΔT₁₀ over the last centuries have been identified in cores LV83-16 and LV83-32 (Figure 1) of the Laptev Sea (Astakhov et al., 2022).

Additionally, according to the distribution of PIP₂₅ in core ARA2B-1A (Figure 4f), IF₁₀ variations in core LV77-3 (Figure 4g), and variations in the area of ice cover in the Arctic as a whole (Figure 4h), periodicity and synchronism with climate change were not traced in ice cover reconstructions based on dinocysts in the cores from the northern part of the Chukchi Sea (de Vernal, 2017). Earlier reconstruction of IF₁₀ for the last century in the northern part of the Chukchi Sea (Astakhov et al., 2019a) and for core LV77-12 (Astakhov et al., 2021) from the southeastern part of the East Siberian Sea (Figure 1) revealed a lower ice cover in the Little Ice Age than that at the beginning of the 20th century, while demonstrating the absence of a correlation between climate (air temperature) and ice conditions. These features are determined by the existence of a predominant influence on the ice conditions of the Chukchi Sea and the adjacent water areas of the Pacific waters entering through the Bering Strait (Polyak et al., 2016; Stein et al., 2017; Woodgate et al., 2010).

Variations in mean annual air temperature

The average annual air temperature variations, which were reconstructed from core LV77-36-1 (Figure 4d) according to the general trend determined by insolation (Laskar et al., 2004), coincided with the summary reconstructions of Holocene temperature anomalies (Marcott et al., 2013). They are characterized by a maximum in the Holocene thermal optimum (Kaufman et al., 2004), a gradual decrease to a minimum in the Little Ice Age, and a subsequent sharp increase in the late 20th–21st centuries (Figure 4a). Our reconstruction on core LV77-36-1 considered the data of observations and reconstruction for the last 200 years in core LV77-36-2 (Figure 4d), also shows a clear decrease in temperature in the 19th–20th centuries, which is typical for the Northern Hemisphere (Figure 4a). The amplitude of SAT changes according to our and other regional reconstructions (2.0°C–2.5°C) is much larger than the global changes (0.8°C–1.0°C), which is evident due to the generalization over a large region for the latter. A large amplitude of summer air temperature variation (SAS) was also established (Klemm et al., 2013) based on pollen-based assemblages in the sediments of Lake Dolgoe in the Lena River delta (Figure 4b). Other pollen-based reconstructions in lacustrine deposits of Siberia exhibited negative values of anomalies in the Late Holocene (Klemm et al., 2016), which implies that in the second half of the 20th century (including 1961–1990, which is used as the zero period), the rise in temperatures was faster than the average for the Northern Hemisphere and in the East Siberian Sea.

Synchronicity of climate change and ice cover, the role of the Arctic oscillation

The data obtained from core LV77-36-1 indicate that in the western part of the East Siberian Sea and in the Laptev Sea (Astakhov et al., 2022; Hörner et al., 2016), ice conditions changed depending on climate variations (air temperature) during the middle and Late Holocene; these variations were determined by a decrease in insolation over the past 7.0 ka (Laskar et al., 2004). Such relationship between variations in ice conditions and climate is established in the North Atlantic and Eurasian sectors of the Arctic (Alley et al., 2001; Brierley and Zhang, 2021; Hoff et al., 2016). During variations in air temperature and associated ice cover, in addition to the general orbitally determined trend, quasi-periodic changes were revealed with a cycle of 1600 and 1740 years. Considering that such ice coverage cyclicity (1800 years) was revealed in the Laptev Sea and determined from IRD variations in the North Atlantic and Eurasian Arctic (Bond et al., 1997), there is every reason to compare it with the Bond cycles and consider it as an Atlantic climate signal in the amerasion sector of the Arctic.

The absence of such cyclicity in the reconstruction of climate and ice conditions in the Chukchi Sea and east of the East Siberian Sea can be explained by Morrison’s Arctic circulation model (Morison et al., 2012). The Laptev Sea belongs to the zone with Atlantic-derived ocean circulation, the Chukchi Sea belongs to the Pacific-derived zone, and the East Siberian Sea can correspond to both the zones depending on the prevailing type of atmospheric circulation (cyclonic–anticyclonic) in various years. Our data indicate the predominant influence of Atlantic-derived processes in the East Siberian Sea, which suggests the predominance of cyclonic atmospheric circulation in the Northern Hemisphere during the Holocene. Such circulation is typical for years with high (positive) values of the Arctic Oscillation (AO) index (Thompson and Wallace, 1998): when the west-east air transport is activated, the inflow of warmer and saline waters from the Atlantic increases; however, to some extent, it also decreases the ice cover in the Arctic as a whole (Rigor et al., 2002). Additionally, ice conditions in the Chukchi Sea become more severe than in periods of low (negative) AO values when interlatitudinal exchange is activated, thus leading to the outflow of warm air masses from the south (Kholoptsev and Podporin, 2021; Plotnikov et al., 2020). The conditional western boundary between the Pacific- and Atlantic-derived Arctic shelves can be set along the 170°E meridian (Figure 1).

Conclusion

Reconstruction of ice conditions (IF₁₀) and average annual air temperature (ΔT₁₀₎ made it possible to reveal the main features of their variations during the Holocene and to compare them with regional and global reconstructions. Variations in ΔT₁₀ and IF₁₀ show a general trend of decreasing air temperature and increasing ice coverage in the western part of the East Siberian Sea from the middle and early Holocene up to the Little Ice Age, owing to an orbitally determined decrease in insolation. A characteristic feature of their distribution is subperiodic variability, which is approximated by periods of 1600 years for IF10 and 1740 years for ΔT₁₀; this feature is comparable to Bond climate cycles. The synchronism of changes in air temperature and ice coverage west of the East Siberian Sea, which were similar to those detected in the Laptev Sea and on the Eurasian Arctic shelf, allowed us to consider the identified cyclicity as an Atlantic climatic signal that is not manifested in the more eastern regions of the Amerasian Arctic shelf. The data obtained indicate the predominance of the cyclonic type of atmospheric circulation in the Northern Hemisphere during the Holocene, with high (positive) values of the Arctic oscillation index that activates latitudinal air transport, increases the inflow of warmer and more saline waters from the Atlantic, and augments the transport of fresh waters from the great Siberian rivers to the east.

Supplemental Material

sj-xlsx-1-hol-10.1177_09596836221126049 – Supplemental material for Climate and ice conditions of East Siberian Sea during Holocene: Reconstructions based on sedimentary geochemical multiproxy

Supplemental material, sj-xlsx-1-hol-10.1177_09596836221126049 for Climate and ice conditions of East Siberian Sea during Holocene: Reconstructions based on sedimentary geochemical multiproxy by Anatolii S Astakhov, Valeriy V Babich, Xuefa Shi, Limin Hu, Maria S Obrezkova, Kirill I Aksentov, Alexander V Alatortsev, Andrey V Darin, Ivan A Kalugin, Viktor N Karnaukh and Mikhail S Melgunov in The Holocene

Footnotes

Acknowledgements

The authors are grateful to A.A. Maryash for providing TOC data and Jiang Dong for age model adaptation.

Authorship contribution

AA: Conceptualization, Writing – original draft; VB: Data curation, Writing – review & editing, XS: Funding acquisition, Data curation; LH: Data curation Writing – review & editing; MO: Formal analysis, Writing – review & editing; KA: Data curation; AlA: Formal analysis; AD: Formal analysis; IK: Data curation; VK: Formal analysis, Data curation; MM: Formal analysis.

Data availability

The entire dataset is provided in the Supplement.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was funded by the Russian Science Foundation (Project 21-17-00081). Expedition works were supported by the Ministry of Sciences and Education of the Russian Federation (project АААА-А17-117030110033-0), and the National Natural Science Foundation of China (Grants no. U160641 and 41420104005).

ORCID iDs

Anatolii S Astakhov

Kirill I Aksentov

Supplemental material

Supplemental material for this article is available online.

References

Alley

Anandakrishnan

Jung

(2001) Stochastic resonance in the North Atlantic. Paleoceanography 16(2): 190–198.

Appleby

Oldfield

Thompson

, et al. (1979) ²¹⁰Pb dating of annually laminated lake sediments from Finland. Nature 280: 53–55.

Astakhov

Bosin

Liu

, et al. (2019a) Reconstruction of ice conditions in the northern Chukchi Sea during recent centuries: Geochemical proxy compared with observed data. Quaternary International 522: 23–37.

Astakhov

Sattarova

Xuefa

, et al. (2019b) Distribution and sources of rare earth elements in sediments of the Chukchi and East Siberian Seas. Polar Science 20: 148–159.

Astakhov

Aksentov

Babich

, et al. (2022) Ice coverage of the Laptev Sea and air temperature variation during Recent centuries: Observed data and reconstructions using a geochemical proxy. Current Chinese Sciences 2(3): 198–212.

Astakhov

Kalugin

Aksentov

, et al. (2015) Geochemical indicators of paleo-typhoons in shelf sediments. Geochemistry International 53(4): 383–388.

Astakhov

Kalugin

Shi

, et al. (2021) The role of ice cover in the formation of bottom sediment chemical composition on the East Siberian Shelf. Geochemistry International 59(6): 585–598.

Astakhov

Shi

Darin

, et al. (2020) Reconstructing ice conditions in the southern Chukchi Sea during the last millenniums based on chemical composition of sediments and diatom assemblages. Marine Geology 427: 106220.

Babich

(1980) Iterative method of target classification and ordering of objects. In: Dmitriev AN (ed.) Recognition of Images in Problems of Qualitative Forecasting of Ore Deposits. Novosibirsk: Nauka, pp.59–67.

10.

Babich

Darin

Kalugin

, et al. (2016) Climate prediction for the extratropical northern hemisphere for the next 500 years based on periodic natural processes. Russian Meteorology and Hydrology 41(9): 593–600.

11.

Bond

Showers

Cheseby

, et al. (1997) A pervasive millennial-scale cycle in North Atlantic Holocene and glacial climates. Science 278(5341): 1257–1266.

12.

Bonnet

de Vernal

Hillaire-Marcel

, et al. (2010) Variability of sea-surface temperature and sea-ice cover in the Fram Strait over the last two millennia. Marine Micropaleontology 74(3–4): 59–74.

13.

Brennan

Hakim

Blanchard-Wrigglesworth

(2020) Arctic Sea-ice variability during the instrumental era. Geophysical Research Letters 47: 1–9.

14.

Brierley

Zhang

(2021) Simulating the mid-Holocene in PMIP. PAGES Magazine 29(2): 74–75.

15.

Crespin

Goosse

Fichefet

, et al. (2009) The 15th century Arctic warming in coupled model simulations with data assimilation. Climate of the Past 5: 389–401.

16.

Darin

Kalugin

Rakshun

(2013) Scanning X-ray microanalysis of bottom sediments using synchrotron radiation from the BINP VEPP-3 storage ring. Bulletin of the Russian Academy of Sciences Physics 77(2): 182–184.

17.

de Vernal

(2017) Variability of Arctic sea-ice cover at decadal to millennial scales: The proxy records. Pages Magazine 25(3): 144–145.

18.

Dong

Shi

Gong

, et al. (2022) Enhanced Arctic sea ice melting controlled by larger heat discharge of mid-Holocene rivers. Nature Communications 13: 5368.

19.

Dudarev

Charkin

Shakhova

, et al. (2016) Modern Lithomorphogenesis on the Eastern Arctic Shelf of Russia. Tomsk: Tomsk Polytechnic University.

20.

Ershov

(1989) Geocryology of the USSR. Eastern Siberia and the Far East. Moscow: Nedra.

21.

Farmer

Cronin

de Vernal

, et al. (2011) Western Arctic Ocean temperature variability during the last 8000 years. Geophysical Research Letters 38(24): 24602.

22.

Hansen

Ruedy

Sato

, et al. (2010) Global surface temperature change. Reviews of Geophysics 48(4): 29.

23.

Hoff

Rasmussen

Stein

, et al. (2016) Sea ice and millennial-scale climate variability in the Nordic seas 90 kyr ago to present. Nature Communications 7(1): 12247.

24.

Hörner

Stein

Fahl

, et al. (2016) Post-glacial variability of sea ice cover, river run-off and biological production in the western Laptev Sea (Arctic Ocean) – A high-resolution biomarker study. Quaternary Science Reviews 143: 133–149.

25.

Ilyash

Zhitina

(2009) Comparative analysis of sea-ice diatom species composition in the seas of Russian Arctic. Biology bulletin reviews 70: 143–154.

26.

Johannessen

Bengtsson

Miles

, et al. (2004) Arctic climate change: Observed and modelled temperature and sea-ice variability. Tellus A: Dynamic Meteorology and Oceanography 56(4): 328–341.

27.

Kalugin

Daryin

Smolyaninova

, et al. (2007) 800-yr-long records of annual air temperature and precipitation over southern Siberia inferred from Teletskoye Lake sediments. Quaternary Research 67(3): 400–410.

28.

Kaufman

Ager

Anderson

, et al. (2004) Holocene thermal maximum in the western Arctic (0–180°W). Quaternary Science Reviews 23: 529–560.

29.

Keeling

Whorf

(2000) The 1,800-year oceanic tidal cycle: A possible cause of rapid climate change. Proceedings of the National Academy of Sciences 97(8): 3814–3819.

30.

Keskitalo

Tesi

Bröder

, et al. (2017) Sources and characteristics of terrestrial carbon in Holocene-scale sediments of the East Siberian Sea. Climate of the Past 13(9): 1213–1226.

31.

Kholoptsev

Podporin

(2021) Meridional components of atmospheric circulation and Arctic ice cover in summer. Russian Meteorology and Hydrology 46(6): 372–378.

32.

Kinnard

Zdanowicz

Fisher

, et al. (2011) Reconstructed changes in Arctic sea ice over the past 1,450 years. Nature 479(7374): 509–512.

33.

Klemm

Herzschuh

Pestryakova

(2016) Vegetation, climate and lake changes over the last 7000 years at the boreal treeline in north-central Siberia. Quaternary Science Reviews 147: 422–434.

34.

Klemm

Herzschuh

Pisaric

MFJ

, et al. (2013) A pollen-climate transfer function from the tundra and taiga vegetation in Arctic Siberia and its applicability to a Holocene record. Palaeogeography, Palaeoclimatology, Palaeoecology 386(15): 702–713.

35.

Lantuit

(2012) The Arctic coastal dynamics database: a new classification scheme and statistics on Arctic permafrost coastlines. Estuaries and Coasts 35: 383–400.

36.

Laskar

Robutel

Joutel

, et al. (2004) A long-term numerical solution for the insolation quantities of the Earth. Astronomy and Astrophysics 428(1): 261–285.

37.

Marcott

Shakun

Clark

, et al. (2013) A reconstruction of regional and global temperature for the past 11,300 years. Science 339: 1198–1201.

38.

Mayer

Schick

Allison

, et al. (2007) Marine vs. Terrigenous organic matter in Louisiana coastal sediments: The uses of bromine:organic carbon ratios. Marine Chemistry 107: 244–254.

39.

McKay

Kaufman

(2014) An extended Arctic proxy temperature database for the past 2,000 years. Scientific Data 1(1): 140026.

40.

Melgunov

Mezina

Sherbov

, et al. (2019) Radioactive elements in atmospheric precipitations of the Western Siberia. E3S Web of Conferences 98: 10003.

41.

Morison

Kwok

Peralta-Ferriz

, et al. (2012) Changing Arctic Ocean freshwater pathways. Nature 481(7379): 66–70.

42.

Moritz

Bitz

Steig

(2002) Dynamics of recent climate change in the Arctic. Science 297: 1497–1502.

43.

Neukom

Barboza

Erb

, et al. (2019) Consistent multidecadal variability in global temperature reconstructions and simulations over the Common Era. Nature Geoscience 12(8): 643–649.

44.

Obrezkova

Kolesnik

Semiletov

(2014) The diatom distribution in the surface sediments of the Eastern Arctic seas of Russia. Russian Journal of Marine Biology 40(6): 465–472.

45.

Obrezkova

Kolyada

Tsoy

(2019) Distribution of diatoms in the surface sediments of the Eastern Arctic seas (based on the 77th and 83rd cruises of the R/V Akademik MA Lavrentiev, 2016, 2018). Issues of Modern Algology 2((20): 50–54.

46.

Overpeck

Hughen

Hardy

, et al. (1997) Arctic environmental change of the last four centuries. Science 278(5341): 1251–1256.

47.

Plotnikov

Pustoshnova

(2012) Variability and conjugacy of ice conditions in the system of East arctic seas (the Laptev, East Siberian, and Chukchi Seas). Russian Meteorology and Hydrology 37(7): 468–476.

48.

Plotnikov

Vakulskaya

Mezentseva

, et al. (2020) Variability of the ice conditions in the Chukchi Sea and their links with Arctic oscillation. Izvestiya TINRO 200(1): 155–167.

49.

Polyak

Belt

Cabedo-Sanz

, et al. (2016) Holocene sea-ice conditions and circulation at the Chukchi-Alaskan margin, Arctic Ocean, inferred from biomarker proxies. The Holocene 26(11): 1810–1821.

50.

Rachold

(1999) Major, Trace and rare earth element geochemistry of suspended particulate material of East Siberian rivers draining to the Arctic Ocean/. In: Kassens

Bauch

Dmitrenko

(eds) Land-Ocean Systems in the Siberian Arctic: Dynamics and History. Berlin: Springer-Verlag, pp.199–222.

51.

Räisänen

(2001) CO₂-induced climate change in CMIP2 experiments: Quantification of agreement and role of internal variability. Journal of Climate 14: 2088–2104.

52.

Rekant

Bauch

Schwenk

, et al. (2015) Evolution of subsea permafrost landscapes in Arctic Siberia since the Late Pleistocene: A synoptic insight from acoustic data of the Laptev Sea. Arktos 1(1): 1–15.

53.

Rigor

Wallace

Colony

(2002) Response of sea ice to the Arctic oscillation. Journal of Climate 15(18): 2648–2663.

54.

Serreze

Crawford

Stroeve

, et al. (2016) Variability, trends, and predictability of seasonal sea ice retreat and advance in the Chukchi Sea. Journal of Geophysical Research 121(10): 7,308–7,325.

55.

Stein

Fahl

Schade

, et al. (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). Journal of Quaternary Science 32(3): 362–379.

56.

Stone

(1997) Variations in western Arctic temperatures in response to cloud radiative and synoptic-scale influences. Journal of Geophysical Research Atmospheres 102(D18): 21769–21776.

57.

Stott

Tett

Jones

, et al. (2000) External control of 20th century temperature by natural and anthropogenic forcings. Science 290(5499): 2133–2137.

58.

Stroeve

Holland

Meier

, et al. (2007) Arctic sea ice decline: Faster than forecast. Geophysical Research Letters 34(9): L09501.

59.

Ren

Sicre

, et al. (2022) HBIs and sterols in surface sediments across the East Siberian Sea: Implications for palaeo sea-ice reconstructions. Geochemistry Geophysics Geosystems 23: e2021GC009940.

60.

Tanis

Timokhov

(1997) Joint US-Russian Atlas of the Arctic Ocean, Oceanography Atlas for the Winter Period. Boulder, CO: Environmental Working Group, University of Colorado.

61.

Tanis

Timokhov

(1998) Joint US-Russian Atlas of the Arctic Ocean, Oceanography Atlas for the Summer Period. Boulder, CO: Environmental Working Group, University of Colorado.

62.

Thompson

DWJ

Wallace

(1998) The Arctic oscillation signature in the wintertime geopotential height and temperature fields. Geophysical Research Letters 25(9): 1297–1300.

63.

Trouet

Van Oldenborgh

(2013) KNMI climate explorer: A web-based research tool for high-resolution paleoclimatology. Tree-Ring Research 69(1): 3–13.

64.

Tsoy

Obrezkova

Aksentov

, et al. (2017) Late holocene environmental changes in the southwestern Chukchi Sea inferred from diatom analysis. Russian Journal of Marine Biology 43(4): 276–285.

65.

Walsh

Fetterer

Scott Stewart

, et al. (2017) A database for depicting Arctic sea ice variations back to 1850. Geography Review 107(1): 89–107.

66.

Woodgate

Weingartner

Lindsay

(2010) The 2007 Bering Strait oceanic heat flux and anomalous Arctic sea-ice retreat. Geophysical Research Letters 37(1): L01602.

67.

Wood

Wang

Salo

, et al. (2015) The climate of the Pacific Arctic: During the first RUSALCA Decade 2004–2013. Oceanography 28: 24–35.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.33 MB