Pb-210 Dating of Ice Scour in the Kara Sea

Abstract

Ice scours are formed when the keels of floating icebergs or sea ice hummocks penetrate unlithified seabed sediments. Until now, ice scours have been divided into “relict” and “modern” according to the water depth that corresponds with the possible maximum vertical dimensions of the keels of modern floating icebergs. However, this approach does not consider climatic changes at the present sea level, which affect the maximum depth of ice keels. We present an application of ²¹⁰Pb dating of the largest ice scour in the Baydaratskaya Bay area (Kara Sea), located at depths of about 28–32 m. Two sediment cores were studied; these were taken on 2 November 2021 from the R/V Akademik Nikolay Strakhov directly in the ice scour and on the “background” seabed surface, not processed via ice scouring. According to the results of ²¹⁰Pb dating, the studied ice scour was formed no later than the end of the Little Ice Age. Based on the extrapolation of possible sedimentation rates prior to 1917 (0.22–0.38 cm/year), the age of the ice scour is estimated to be 1810 ± 30 AD. The mean rate of ice scour filling with 70 cm thick sediments from the moment of its formation is around 0.33 cm/year.

1. Introduction

Ice scours (ice gouges, ice ploughmarks) are widespread on the bottom of high- and mid-latitude seas and big freezing lakes at different water depths up to 500 m [1,2,3,4,5,6,7,8,9,10,11,12] and occasionally up to 1200 m [13,14]. They are formed when the keels of floating icebergs or sea ice hummocks (pressure ridges) collide with unlithified seabed sediments. The study of the ice scouring process has become especially important in recent times due to climate change. Additionally, these kinds of processes are a natural risk for the development of Arctic shelf oil and gas fields and the Northern Sea Route, adversely affecting the construction and operation of structures and the movement of ships in the Barents–Kara region.

For the purposes of marine engineering, it is important to understand in which areas ice scouring can still occur, and in which areas ice scours on the seabed are relicts. During the construction of submarine structures such as pipelines, it is also important to understand under what conditions (sea level, climate) the depth of penetration of ice keels into the bottom sediments is observed. To answer these questions, the dating of ice scours, which to our knowledge has not yet been achieved, could help. Only the dating of completely buried iceberg scour relicts (with the age around 31 kyr cal BP) in the subtropical North Atlantic is known [15]. Their age has been estimated based on radiocarbon dating of sediment samples above and below the iceberg scour surface.

To estimate the age of ice scours, the water depth at which they are observed is now the most widely used [3,7,8,12,13]. According to this, ice scours are divided into “relict” (formed at another relative sea level other than the present) and “modern” (formed at the present sea level). Ice scour relicts are located at a water depth significantly exceeding the possible maximum vertical dimensions of the keels of modern floating icebergs. The threshold between modern iceberg scours and iceberg scour relicts varies widely depending on the area and in the range of 40 to 550 m [3,4,7,8]. However, this approach does not take into account climatic changes at the present sea level which affect the maximum depth of ice keels and, accordingly, the depth of the seabed at which the ice scours were formed and located in the corresponding climatic period.

Another attempt to date ice scours is based on the identification of newly formed ice gouges because of a comparison of multitemporal bottom surveys carried out during monitoring [16,17]. This approach has its disadvantages: it requires many years of observations, does not consider the extreme sizes of ice keels and is limited in space.

The method of marine sediment dating using non-equilibrium lead (²¹⁰Pb_ex), which is widely used to determine sedimentation rates over the past 100–120 years, could expand the possibilities for estimating the age of ice scours [18]. However, we are not aware of any attempts to date ice scours based on measurements of the specific activity of the natural radionuclide ²¹⁰Pb. In this regard, the main objective of the present study is to show the possibility of the well-established ²¹⁰Pb geochronology approach to ice scour dating.

2. Study Area

The study area is located in front of the entrance to Baydaratskaya Bay in the southwestern semi-enclosed part of the Kara Sea. The ice scours of the bay are the most studied in the entire Kara Sea [19]. They began to be investigated in 1988 as part of the design of a submarine gas pipeline crossing the bay. The largest ice scour known in this area was chosen for dating using the non-equilibrium lead method.

The studied ice scour is located at depths of about 28–32 m (Figure 1), which are close to the estimates of the maximum keel depths of sea ice features (stamukhi and hummocks) in the Kara Sea—no more than 28 m [20]. The maximum keel depth of modern icebergs in the Kara Sea is estimated to be no more than 180–200 m and the average is about 50 m [21]. There are known facts of iceberg penetration into Baydaratskaya Bay with depths of less than 20 m during periods of minimal ice cover in the Kara Sea in 1932 and 2007 [22].

From the point of view of the geomorphological structure of the bottom, the study area is located on a gently sloping coastal plain which extends to depths of 50–60 m and is bounded from the open sea side by a tectonic scarp [23]. By the beginning of the postglacial transgression, this plain was in subaerial conditions. During transgression, its surface from the scarp crest to the modern coastline was reworked by wave processes. It is assumed that about 10 kyr BP, there was a period of relative sea level stabilization at the modern sea depth of 32–34 m [23].

The Baydaratskaya Bay area is characterized by low average annual temperatures (from −7 to −10 °C) and a relatively short ice-free period (1–4 months a year) [19]. Over the past 50 years, there has been a tendency for the ice-free period to increase [24]. The sea ice of Baydaratskaya Bay consists of fast ice along the coast and drift ice offshore [25]. The maximum fast ice rim position generally varies in the range of sea depths of 10–20 m [26].

The wind and currents are the driving forces of ice drift. During the winter season, southerly and southwesterly winds prevail in the Baydaratskaya Bay area [25]. The currents here are associated with semi-diurnal tides and they are practically reversive and aligned along the axis of the bay. The maximum speed of the tidal current during the tidal cycle is 0.5 m/s, while the measured maximum current speed is 1 m/s [25]. Sea level fluctuations are associated with tides (up to 1.1 m) and storm surges (up to 2 m).

Ice scours are widespread in Baydaratskaya Bay to depths of more than 12 m and can be up to 2 m deep, 50 m wide and several kilometers long [5]. In shallower areas, ice scouring processes are also active, but wind waves destroy their evidence during strong summer and autumn storms. The most intensive and deep ice scours occur at the range of depths from 16 to 19 m next to the Yamal coast fast ice rim, where ice hummocks continue during the whole of the cold season [5]. The predominant orientation of the ice scours corresponds to the direction of reverse tidal currents along the axis of the bay.

Sedimentation rates in the Kara Sea are presented mainly as average values for the thickness of the Holocene deposits [27]. More precisely, the sedimentation rate was determined using the method of accelerator mass spectrometry (AMS ¹⁴C) using carbonate shells of mollusks and foraminifers only for a few single horizons of the sedimentary section [28,29].

Sedimentation rates determined by ²¹⁰Pb in the Kara Sea are presented in [30,31,32]. Fourteen cores of sediments up to 50 cm long were studied in key areas of the bottom deposit accumulation in the Kara Sea: the estuaries of the Ob and Yenisei, the East Novaya Zemlya Trench, the Voronin Trench and also in one of the northern bays of the Novaya Zemlya (Sedov Bay) [33]. However, all of them differ from Baydaratskaya Bay in terms of sedimentation conditions. The sedimentation rates were determined by measuring the specific activity of the natural ²¹⁰Pb and technogenic ¹³⁷Cs radionuclides. A close relationship was shown between the sedimentation rates and types of seabed sediments. The highest sedimentation rates are typical of terrigenous–estuary sediments, which are divided into traction sediments with sedimentation rates of 0.4–0.7 cm/year and sediments of “silt banks” with sedimentation rates of 0.7–1.0 cm/year [33].

3. Materials and Methods

The present study is based on the results of the seabed topography and sediment investigations of the Kara Sea by the Shirshov Institute of Oceanology RAS cruise 52 of the R/V Akademik Nikolai Strakhov (ANS) [34], which took place from 15 October to 10 November 2021. During the cruise, complex geological and geophysical, geomorphological and hydrophysical studies were carried out. One of the objectives of the cruise was a detailed study of the large ice scours in the Kara Sea.

3.1. Multibeam Echo Sounding

Multibeam echo sounding was carried out for the seabed topography mapping to obtain information about the morphology, morphometry and configuration of the ice scour. For this, a multibeam echo sounding system of the R/V Akademik Nikolaj Strakhov was used. It consists of a Reson 100 kHz SeaBat 8111, an Applanix POSMV integrating motion sensor and gyrocompass data. The multibeam transducers are installed in a hull-mounted gondola. The swath-bathymetric data were collected and processed using the PDS2000 software from Teledyne. The processed data were gridded at cell sizes of 4×4 m, and visualizations were carried out using PDS2000, ArcGIS and QGIS software. The survey area consisted of a series of single survey lines with a total length of about 35 km along and across the studied ice scour (Figure 2).

3.2. Seismic Profiling

Seismic profiling was carried out to assess the thickness of the deposits filling the ice scour and to obtain information about the structure of the sedimentary strata in which the ice scour was formed. For this, an EdgeTech 3300 high-resolution seismic profiler (sub-bottom chirp) and the seismoacoustic complex Geont-shelf with a sparker source were used. The sub-bottom transducers were installed in a hull-mounted gondola. In addition, the high-resolution parametric sub-bottom profiler SES-2000 Standard with frequencies of 8–10 kHz was installed on a pole on the starboard. The studies were carried out using a single-channel seismic streamer.

Seismic profiling was carried out along all multibeam echo sounding lines with a total length of about 35 km, but in this work, only seismic profiles located near the sampling site of the sediment core were used (Figure 2). In addition, profiles obtained near the sampling site during previous cruises were involved in the analysis.

3.3. Sediment Cores

In the present study, two sediment cores (ANS-52-16 and ANS-52-17) were studied (

Table 1. Sediment cores studied in the present study.

Sediment Core	Position	Sampler	Latitude (°N)	Longitude (°E)	Cruise	Water Depth, m	Length, cm
ANS-52-16	Ice scour	147 mm gravity corer	69.797683	65.40115	ANS-52	33.6	43
ANS-52-17	Background	147 mm gravity corer	69.796467	65.40185	ANS-52	31.6	271

). They were taken on 2 November 2021 (2021.8 AD) using a 147 mm gravity corer from R/V Akademik Nikolaj Strakhov near the entrance of Baydaratskaya Bay in the Kara Sea (Figure 1). The first core (ANS-52-16, length 43 cm) was taken directly in the ice scour at a water depth of 33.6 m; the second one (ANS-52-17, length 271 cm) was taken at a depth of 31.6 m from the “background” seabed surface south of the ice scour, not reworked by ice ploughing (Figure 2). In the present study, the first core is called “ice scour”, and the second is called “background”. The distance between cores is about 140 m.

Since coring from a relatively narrow ice scour constitutes a complex task, the results of the core analysis from the “background” seabed surface that had not been reworked by ice ploughing should confirm or disprove that the first core was indeed taken from the ice scour. This approach is based on the fact that the sedimentation conditions within and outside of the ice scour should be different and therefore the results from the analyses should show different patterns in these cores.

The following technique was used to increase the probability of core sampling directly from the ice scour. The vessel approached the ice scour with a predetermined spatial configuration and a preselected sampling site (according to multibeam echo sounding and seismic profiling) at an angle that provided greater distance during the ice scour crossing. During the vessel drift, the gravity corer was hung out at a given horizon (approximately 20 m from the bottom); then, when the vessel was in position at the planned sampling point, on command from the Captain’s Bridge, the gravity corer was released on the winch freewheel. At the moment when the gravity corer touched the seabed (determined by the sagging of the cable) the precise coordinates of the sampling point were recorded.

After the cores were recovered onboard the vessel, they were photographed and their lithology was visually described. For further laboratory analysis of the ice scour dating, the cores were packed into plastic cable channels 50 cm long, either completely (43 cm of “ice scour” core) or only the upper part of the core (50 cm of “background” core). Cores packed in cable channels were subjected to quick freezing (−18 °C) onboard the vessel and sent frozen to the laboratory of the Murmansk Marine Biological Institute of the Russian Academy of Sciences.

The sampling for laboratory analysis of only the upper 50 cm of cores is explained by the fact that the expected age of deposits below 50 cm should be significantly more than 100–120 years, which exceeds the dating capabilities of the non-equilibrium lead method.

3.4. Grain Size Analysis

Grain size analysis was carried out in the laboratory of Murmansk Marine Biological Institute with wet samples using the wet sieve and decantation methods, which are considered to be the most accurate for fine sediments, as they are based on the hydraulic settling of particles according to the Stokes formula [35]. The accuracy of settling in each stage was controlled by a microscope. The classification of M.V. Klenova was used to interpret the types of bottom sediments based on the results of grain size analysis [36].

The sampled cores were divided into layers of 1–5 cm for detailed studies. A total of 44 samples were analyzed: 20 in the “ice scour” core (0–30 cm) and 22 in the “background” core (0–50 cm). Grain size is given in weight percent of the total sample. Moisture content was determined for each sample together with grain size analysis.

3.5. Radiometric Measurements ²¹⁰Pb, ²²⁶Ra, ¹³⁷Cs and ⁷Be

The measurement of specific activity of ²¹⁰Pb, ²²⁶Ra, ¹³⁷Cs and ⁷Be radionuclides was carried out in the laboratory of the Murmansk Marine Biological Institute using a multichannel gamma-ray spectrometer for measuring X-ray and gamma radiation (Canberra Semiconductors NV, Olen, Belgium) with lead screen protection of the HPGe detector Ekran-2P manufactured by Aspect (Dubna, Russia). For the recording part, a BE5030 broadband detector made of ultrapure germanium planar type with a thin “carbon epoxy” entrance window (0.6 mm wide) and a crystal (diameter of 80 mm, an area of 5000 mm² and a thickness of 30.5 mm) was used. It records gamma quanta with energy from 3 keV to 3 MeV. The energy resolution along the 1332 keV ⁶⁰Co line is no less than 2.2 keV; along the ⁵⁷Co 122 keV isotope line, it is no less than 0.75 keV, and along the ⁵⁵Fe isotope line, it is no less than 0.5 keV. Spectral information was collected using a DSA-1000 pulse analyzer (Canberra Industries, Inc., Loches, France) with a resolution of 16K channels, corresponding to modern digital signal processing technology. Spectra processing took place and radionuclides were identified using Genie-2000 software (version 3.3).

The sampled cores were divided into layers of 1–2 cm for detailed studies. A total of 56 samples were analyzed: 36 in the “ice scour” core (0–43 cm) and 20 in the “background” core (0–25.5 cm). All sediment samples were dried before measurement, homogenized and left for 30 days in hermetically sealed vessels to ensure secular equilibrium between ²²⁶Ra and ²²²Rn, as well as ²¹⁴Pb and ²¹⁴Bi [37,38]. The measurement of each layer was carried out in the same vessels after the onset of equilibrium between the indicated radionuclides. The measurement time was 85,000 s, which gave an error calculation of 5–15% in the upper layers of the cores. The activity of the supported ²¹⁰Pb determined from the main ²²⁶Ra lines was subtracted to determine the excess ²¹⁰Pb activity (²¹⁰Pb_ex) from the total ²¹⁰Pb specific activity measured from its γ-line (46.5 keV). The quantification of ²²⁶Ra was carried out by ²¹⁴Pb (295.2 keV and 351.9 keV) and ²¹⁴Bi (1120 keV). The ²¹⁰Pb_ex activity was corrected for the date of core collection (2021.8 AD) and the results are given on a dry weight basis, corrected for self-absorption and sample geometry [39]. The short-lived radionuclide ⁷Be was measured in the uppermost layer to confirm that the surface layer was sampled correctly.

A quality check of the gamma spectrometer efficiency was carried out regularly with the help of two volumetric activity measures issued for special purposes by the D.I. Mendeleyev Institute for Metrology (VNIIM). The density of the the volumetric activity measure fillers was 1.01 g/cm³. The activity of ²¹⁰Pb deposited on the filler was 1.64 kBq; the activity of ¹³⁷Cs was 1.4 kBq. The specific activity of ²²⁶Ra was determined via the source of ¹⁵²Eu, and the activity of ¹⁵²Eu in the source was 3.78 kBq.

3.6. Calculation of the Calendar Age and the Sediment Accumulation Rates Based on the Results of Radiometric Measurements

²¹⁰Pb dating of marine sediments is an accessible and widely used method used to estimate the rates of modern sedimentation in the Arctic seas [31,32,33,40,41,42,43,44]. The simultaneous determination of natural ²¹⁰Pb and ²²⁶Ra and technogenic ¹³⁷Cs in a sample became possible after the advent of modern gamma spectrometers for measuring X-ray and gamma radiation and detecting gamma rays with energies from 3 keV to 3 MeV. This significantly reduced the analysis time and made it possible to use the measured material for other studies.

The natural radionuclide ²¹⁰Pb has a half-life (T_1/2) of 22.3 years and is a member of the ²²⁶Ra decay chain (T_1/2 = 1600 years) that is produced via the successive decay of the parent isotope ²³⁸U. ²¹⁰Pb in marine sediments consists of supported ²¹⁰Pb, which is continuously formed and, presumably, is in secular equilibrium with its initial radionuclide ²²⁶Ra and excess ²¹⁰Pb. The last one enters the surface of water bodies and the surrounding drainage area, and then makes its way into the bottom sediments as a result of atmospheric deposition. The excess ²¹⁰Pb is determined by subtracting the supported ²²⁶Ra activity from the total ²¹⁰Pb activity and is used to determine the age of sediments in a sedimentation basin. However, compared to other isotopes used for geochronology, such as ¹⁴C (T_1/2 = 5730 years), excess ²¹⁰Pb can only be used to date the age of recent deposits formed within the last 100–120 years [45,46].

Many studies have shown that the analysis of data and interpretation of the results obtained by dating via excess ²¹⁰Pb are often associated with certain assumptions and limitations [47,48,49,50]. The age of the sediments is calculated using an exponential equation that describes the decrease in the total specific activity of ²¹⁰Pb with depth, only if the flow of excess ²¹⁰Pb to the sediment surface is constant and there are no processes leading to its mixing or redistribution (constant sedimentation) [49]. Often, these conditions are not met and significant fluctuations in the ²¹⁰Pb content can be found in the vertical profiles of the deposits. To avoid questionable or inaccurate interpretations of ²¹⁰Pb profiles, the chronology of ²¹⁰Pb should always be confirmed using independent time markers (e.g., ¹³⁷Cs) or any other available geochemical indicators [49,51].

The calendar age and the sediment accumulation rates of layers in the studied sediment cores were calculated using the constant flux (CF) model based on the data of the excess ²¹⁰Pb activity [49]. The CF model makes it possible to consider the uneven supply of ²¹⁰Pb with the flow of sediment mass to the seabed surface. Layers lying below the accepted equilibrium boundary are not considered during sediment age calculations. The formulae used for the calculation are presented in the Supplementary Materials.

The independent verification of chronology is essential to ensure a high level of confidence in the results. Therefore, the known independent key date method was used. Artificial radionuclides make it possible to determine several key dates in the last 70 years. The ¹³⁷Cs chronostratigraphic marker is widely used to date sediment cores in the Arctic, as the radioactive fallout from atmospheric nuclear weapon testing was global. Thus, the presence of a concentration peak in most cases makes it possible to determine the depth of contamination with a known event date [52]. The calculated ages of studied sedimentary layers are verified using the results of measuring the specific activity of the technogenic radionuclide ¹³⁷Cs, especially since the nuclear tests were carried out relatively close to the study area—on the Novaya Zemlya archipelago.

3.7. Estimation of Ice Scour Age

Ice scour age estimation is based on the fact that the sedimentation in it commences immediately after its formation as a result of the impact of drifting ice on the seabed. Therefore, if we determine the age of the lower sediment layers filling the ice scour, we can determine the age of the ice scour itself. The extrapolation of the sedimentation rate, determined from ²¹⁰Pb in the upper layers, was used to estimate the age of the lower sediment layer of the ice scour in case it exceeded the capabilities of the ²¹⁰Pb dating method (100–120 years).

4. Results

4.1. Ice Scour Structure

At depths of 28–32 m, a multibeam echo sounder mapped an almost rectilinear segment 10.8 km long of the ice scour (in front of the Baydaratskaya Bay entrance, Figure 2), oriented NW–SE at an angle to the isobaths and located almost on the continuation of the bay axis (Figure 1). The maximum depth of the ice scour cutting into the background surface was 3.2 m, and the maximum width did not exceed 35 m. Side berms up to 1.5 m high could be traced almost along the entire ice scour.

The ice scour made a U-turn of almost 180 degrees, at the SE end of the segment, and after which the survey was interrupted, making it impossible to trace the further configuration of the ice scour. However, on the tacks transverse to the main ice scour, the minor fragments of some ice scours were recorded, which presumably could be a continuation of the main ice scour. In this case, the ice scour had a serpentine-shaped plan configuration, changing its direction 2–3 times and reaching a length of at least 30–35 km (dashed black line in Figure 2A). Such a configuration indicates that the main driving force in the formation of the ice scour was reverse tidal currents. At the SE end of the segment where the U-turn occurs, the ice scour becomes double.

Core sampling was carried out in the NW part of the studied segment of the ice scour. Here, it had an asymmetric structure—the southern side was 0.5–1.0 m higher than the northern one (Figure 2B). At the core sampling site, the apparent depth of the ice scour cutting into the background seabed surface was about 2.4 m. According to seismic profiles, the thickness of sediments filling the ice scour was estimated at about 0.6–0.8 m (Figure 3). Thus, the initial depth of the ice scour cutting into the background seabed surface at this location could be up to 3.2 m.

Two sediment units are clearly visible on the seismic profiles crossing the ice scour in the core sampling area (Figure 3). The upper unit, 1.5–2 m thick, covers the underlying layers like a mantle and is displayed on the seismic profiles mainly as an acoustically transparent stratum, which is due to its water saturation and the relatively fine composition of sedimentary material. The almost-complete absence of acoustically pronounced boundaries in this unit may indicate that it was formed under the conditions of a stable water basin, i.e., without significant changes in sedimentation. This allowed us to assume their formation over the past 5–6 kyr BP, when the sea level became close to modern. Thus, we interpreted the age of the upper unit as Late Holocene.

The lower unit has a “chaotic” wavefield pattern, which is characterized by numerous short sub-horizontal and inclined axes of commonality (Figure 3). These deposits are characterized by a complex facies composition laterally and in depth. They are represented by lithological differences, from sands and sandy loams to clayey silts. We interpreted the age of the lower unit as Late Pleistocene. Thus, as can be seen in the seismic profiles (Figure 3), the ice scour completely cuts through Late Holocene deposits and cuts into Late Pleistocene deposits by 1.5–2 m (Figure 3).

4.2. Sediment Grain Size Distribution

Grain size analyses show that the sediments of both cores are composed exclusively of clayey silt. The pelite content (<0.01 mm) varies from 52.9 to 68.4% in the “ice scour” core, silt content (0.05–0.01 mm) ranges from 23.4 to 29.5% and heterogeneous sand (1–0.05 mm) is from 4.2 to 16.4% (

Table 2. Grain size composition and natural moisture content (W) of the “ice scour” core sediments (ANS-52-16).

Layer, cm	Thickness, cm	>1	1–0.5	0.5–0.25	0.25–0.1	0.1–0.05	0.05–0.01	<0.01	W, %
0–1	1	0	0	2.1	4.2	10.1	26.4	57.2	130
1–2	1	12.5	0.4	0.4	4.7	5.7	23.4	52.9	90
2–3	1	0	0	0.5	2.8	3.7	26.2	66.8	122
3–4	1	0	0	1.0	4.0	6.5	29.5	59.0	101
4–5	1	0	0	0	3.3	3.7	28.6	64.4	94
5–6	1	0	0	0.5	3.75	7.5	29.25	59.0	94
6–7	1	0	0	0.3	3.2	6.3	28.4	61.8	90
7–8	1	0	0	0	3.8	6.6	28.4	61.2	94
8–9	1	0	0	0	1.3	3.7	28.0	67.0	94
9–10	1	0	0	0	1.7	3.0	26.9	68.4	90
10–11	1	0	0	0	1.6	2.6	27.9	67.9	94
11–12	1	0	0	0	2.0	3.5	27.3	67.2	98
12–15	3	0	0	1.0	2.8	4.9	25.4	65.9	86
15–17	2	0.5	0.2	0.8	4.4	3.9	33.8	56.4	81
17–19	2	0.2	0.2	0.7	3.0	2.2	33.0	60.7	69
19–21	2	0	0	0.8	3.1	3.1	30.1	62.9	67
21–23	2	0.2	0.2	0.8	3.6	3.1	26.5	65.6	63
23–25	2	0	0	0.6	2.9	3.0	25.5	68.0	69
25–27	2	0.1	0.1	0.8	3.8	3.8	26.9	64.5	67
27–30	3	0	0	0.4	2.7	5.2	28.0	62.7	71

). Some layers (1–2 and 15–27 cm) contain gravel grains. The highest content of fine gravel inclusions occurs in the upper part of the core (1–2 cm).

In the “background” core, the pelite content (<0.01 mm) varies from 56.6 to 72.7%, silt content (0.05–0.01 mm) ranges from 27.1 to 39.4% and the inclusion of heterogeneous sand and gravel is not significant (

Table 3. Grain size composition and natural moisture content (W) of the “background” core sediments (ANS-52-17).

Layer, cm	Thickness, cm	>1	1–0.5	0.5–0.25	0.25–0.1	0.1–0.05	0.05–0.01	<0.01	W, %
0–2	2	0	0	0	0	1.2	36.3	62.5	77
2–5	3	0	0	0	0	1.9	34.5	63.5	68
5–8	3	0	0	0.3	2.3	3.5	30.5	63.4	71
8–11	3	0.4	0.1	0.4	2.5	4.0	36.0	56.6	69
11–15	4	0	0	0	0.8	1.7	32.7	64.8	70
15–16	1	3.8	0	0	0	1.3	27.0	67.9	72
16–17	1	0	0	0	1.0	1.5	30.6	66.9	72
17–18	1	0	0	0	0	1.1	27.1	71.8	65
18–19	1	0	0	0	1.0	2.1	28.1	68.9	69
19–20	1	0	0	0	0.9	2.1	30.0	67.0	67
20–21	1	0	0	0	0	1.9	31.8	64.3	65
21–22	1	0	0	0	0.8	1.7	33.3	64.2	66
22–23	1	0	0	0	0	1.5	31.0	67.5	72
23–24	1	0	0	0	0	0.7	27.6	71.7	67
24–25	1	0	0	0	0	1.4	26.2	72.4	71
25–26	1	0	0	0	0.5	1.5	23.5	74.5	68
26–27	1	0	0	0	0.9	1.9	24.5	72.7	72
27–30	3	0	0	0	1.3	4.0	38.7	55.8	69
30–35	5	0	0	0.1	1.2	2.2	37.5	59.0	48
35–40	5	0	0	0.3	2.0	2.3	39.4	56.0	55
40–45	5	0	0	0.3	1.5	2.3	36.9	59.0	55
45–50	5	0.6	0.2	0.2	1.7	2.4	36.7	58.2	63

). Thus, the concentration of the coarse fraction in the sediments accumulated inside the ice scour is higher than in the sediments outside it, which is the main difference in the grain size composition of the studied cores. The natural moisture content of sediments (W) decreases from the upper layers to the lower ones in both cores. However, it is generally higher in the “ice scour” core (from 130 to 67%; Table 2) than in the “background” core (from 77 to 48%; Table 3).

4.3. Distribution of ²¹⁰Pb and ¹³⁷Cs in Sediments of the Cores

The specific activity of ²¹⁰Pb reaches a maximum value of 108 Bq/kg in the upper part of the “ice scour” core sediments (

Table 4. The results of measurements of the specific activity of short-lived radionuclides and the calculation of excess ²¹⁰Pb (C_i) in the “ice scour” core (ANS-52-16), Bq/kg.

Layer, cm	Thickness, cm	137Cs	U 1 (137Cs)	210Pb	U 1 (210Pb)	226Ra	U 1 (226Ra)	Ci	U 1 (Ci)
0–1	1	6.0	2.4	93.7	9.2	22.1	5.6	71.6	10.8
1–2	1	9.3	0.4	74.8	3.0	24.2	2.7	50.6	4.0
2–3	1	10.3	0.6	107.0	4.9	24.1	5.4	82.9	7.3
3–4	1	11.3	0.8	90.7	5.4	31.3	3.8	59.4	6.6
4–5	1	10.9	0.8	73.1	5.4	28.0	3.7	45.1	6.5
5–6	1	11.6	0.7	78.9	5.1	30.0	3.6	48.9	6.2
6–7	1	9.1	0.8	73.0	5.6	17.7	3.8	55.3	6.8
7–8	1	10.6	0.8	87.7	6.0	35.4	4.3	52.3	7.4
8–9	1	12.6	0.8	92.7	5.4	34.9	4.0	57.8	6.7
9–10	1	13.7	0.9	108.0	6.3	43.2	4.6	64.8	7.8
10–11	1	14.9	0.9	102.0	5.9	39.5	4.6	62.5	7.5
11–12	1	12.1	0.5	82.4	4.0	45.7	4.3	36.7	5.9
12–13	1	11.8	0.9	83.9	6.2	41.9	4.6	42.0	7.7
13–14	1	9.4	1.4	83.8	10.3	25.5	5.0	58.3	11.4
14–15	1	13.6	0.9	72.5	6.4	31.0	4.3	41.5	7.7
15–16	1	15.3	2.2	83.0	14.7	20.8	5.9	62.2	15.8
16–17	1	17	1.1	92.3	8.2	23.2	4.1	69.1	9.2
17–18	1	17.4	1.8	89.4	12.1	26.1	5.3	63.3	13.2
18–19	1	13.4	0.9	84.9	6.7	27.9	3.6	57.0	7.6
19–20	1	11.9	1.8	66.1	12.7	22.7	4.6	43.4	13.5
20–21	1	2.9	0.6	71.5	5.5	24.4	3.5	47.1	6.5
21–22	1	<MDA	-	66.7	4.7	25.8	3.9	40.9	6.1
22–23	1	<MDA	-	57.3	5.9	22.4	4.8	34.9	7.6
23–24	1	<MDA	-	47.3	4.9	17.4	4.5	29.9	6.7
24–25	1	<MDA	-	43.7	4.9	14.4	3.5	29.3	6.0
25–26	1	<MDA	-	40.0	4.8	17.9	3.3	22.1	5.8
26–27	1	8.1	0.4	45.5	2.9	21.0	3.4	24.5	4.5
27–28	1	<MDA	-	56.8	5.9	26.4	4.2	30.4	7.2
28–30	2	<MDA	-	52.1	4.9	26.6	4.1	25.5	6.4
30–32	2	47.3	0.9	56.1	3.6	22.5	2.4	33.6	4.3
32–34	2	2.4	0.6	59.3	4.1	24.2	2.7	35.1	4.9
34–36	2	2.3	0.4	41.7	3.5	19.3	2.7	22.4	4.4
36–38	2	<MDA	-	43.9	6.3	23.7	3.3	20.2	7.1
38–40	2	<MDA	-	48.0	4.8	20.0	2.1	28.0	5.2
40–42	2	<MDA	-	55.9	4.8	25.0	3.9	30.9	6.2
42–43	1	<MDA	-	52.5	4.0	24.5	4.4	28.0	5.9

¹ U—standard uncertainty.

). It decreases to 48–55.9 Bq/kg in the lower layers (38–43 cm) without any significant outliers. The ²²⁶Ra content varies from 14.7 to 45.7 Bq/kg throughout the sediment core, with an average value of 28.6 Bq/kg. Equilibrium with the initial radionuclide ²²⁶Ra at the layer of 42–43 cm was not revealed in the “ice scour” core, but there was a tendency to approach it. The technogenic radionuclide ¹³⁷Cs was determined in most layers. The values ranged from less than the minimum detectable activity (<MDA) to 47.3 Bq/kg. The MDA for ¹³⁷Cs was 0.2 Bq/kg. The radionuclide was not found in sediment layers below 36 cm (Table 4). In the uppermost layer of the “ice scour” core (0–1 cm), short-lived radionuclide ⁷Be was found with specific activity of 18.4 ± 8.3 Bq/kg, which confirms that the surface layer was sampled correctly.

In the “background” core sediments, the highest specific activity of ²¹⁰Pb is in the upper part, reaching 60 Bq/kg (

Table 5. Results of measurements of the specific activity of short-lived radionuclides and calculation of excess ²¹⁰Pb (C_i) in the “background” core (ANS-52-17), Bq/kg.

Layer, cm	Thickness, cm	137Cs	U 1 (137Cs)	210Pb	U 1 (210Pb)	226Ra	U 1 (226Ra)	Ci	U 1 (Ci)
0–1	1	<MDA	-	60.0	0.8	26.2	6.0	33.8	6.1
1–2	1	2.2	0.3	50.1	1.5	24.2	3.4	25.9	3.7
2–3	1	<MDA	-	36.0	5.0	25.3	4.9	10.7	7.0
3–4	1	<MDA	-	55.4	6.0	20.7	4.9	34.7	7.7
4–5	1	<MDA	-	35.5	3.6	16.1	4.3	19.4	5.6
5–6	1	<MDA	-	49.9	6.6	28.0	3.3	21.9	7.4
6–7	1	<MDA	-	49.3	4.6	28.0	5.2	21.3	6.9
7–8	1	<MDA	-	27.6	6.1	22.3	4.2	5.3	7.4
8–9	1	<MDA	-	45.3	3.6	21.5	5.0	23.8	6.2
9–10	1	<MDA	-	54.6	6.6	23.7	4.0	30.9	7.7
10–11	1	1.1	0.3	53.9	3.3	21.9	5.1	32.0	6.1
11–12	1	<MDA	-	47.0	7.3	30.4	4.4	16.6	8.5
12–13	1	<MDA	-	32.2	5.8	25.9	5.6	6.3	8.1
13–14	1	<MDA	-	30.1	4.7	25.9	5.5	4.2	7.2
14–15.5	1.5	<MDA	-	28.0	3.5	21.2	3.9	6.8	5.2
15.5–17.5	2	<MDA	-	34.6	2.5	25.5	3.9	9.1	4.6
17.5–19.5	2	<MDA	-	26.7	2.0	21.7	2.0	5.0	2.8
19.5–21.5	2	<MDA	-	32.9	3.2	31.1	3.0	1.8	4.4
21.5–23.5	2	<MDA	-	33.4	3.9	28.8	2.3	4.6	4.5
23.5–25.5	2	<MDA	-	28.6	3.0	23.7	2.7	4.9	4.0

¹ U—standard uncertainty.

). It decreases to 26.7–33.4 Bq/kg in the lower layers (17.5–25.5 cm). The ²²⁶Ra content varies from 20.5 to 30.4 Bq/kg throughout the sediment core, with an average value of 23.5 Bq/g. The equilibrium of ²¹⁰Pb (32.9 Bq/kg) with the initial radionuclide ²²⁶Ra (31.1 Bq/kg) is observed in the layer of 19.5–21.5 cm. The underlying two layers (21.5–25.5 cm) show C_i values (Bq/kg) of less than 1.0, considering the standard uncertainty. This allows us to take 19.5 cm as the equilibrium boundary for ²¹⁰Pb and ²²⁶Ra. The layers below the accepted equilibrium boundary were not considered during sediment age calculations [49]. The ¹³⁷Cs specific activity in all layers was below the MDA, except for layers of 1–2 cm and 10–11 cm (Table 5). The distribution of ²¹⁰Pb excess in both sediment cores demonstrates the same pattern: an exponential decrease in depth with some fluctuations, whereby the peaks of which are correlated between the cores (Figure 4).

4.4. The Calendar Age and the Sediment Accumulation Rates

The calendar age and the sediment accumulation rate (SAR) of each sedimentary layer top in the studied sediment cores were calculated using the CF model [49], based on the measurements of the excess ²¹⁰Pb activity (Figure 5 and Figure 6, Tables S1 and S2 in Supplementary Materials). In the “ice scour” core, the calculations were based on the assumption that the equilibrium between ²¹⁰Pb and ²²⁶Ra must be achieved in the underlying layers not penetrated by the corer, i.e., the supposed equilibrium border is at 43 cm (curve “Pb-210 age, ice scour, CF/43” in Figure 5). In the “background” core, the calculations were carried out taking into account the accepted equilibrium border at 19.5 cm (curve “Pb-210 age, background, CF/19.5” in Figure 5). The obtained calendar ages have the form of decimal fractions since they are calculated mathematically, relative to the date of the core selections—2 November 2021 (2021.8 AD).

The calendar age of the lowest reliably dated levels is 1916.8 ± 4.9 AD (40 cm) in the “ice scour” core and 1910.4 ± 17.9 AD (17.5 cm) in the “background” core (Figure 5, Tables S1 and S2 in Supplementary Materials). The age value of 1884.1 ± 7.0 AD per 42 cm in the “ice scour” core is not considered, as it is beyond the limits of the method capabilities. Thus, the upper 40 cm in the “ice scour” core accumulated over the last 105 ± 4.9 years and the upper 17.5 cm in the “background” core accumulated over the last 111.4 ± 17.9 years.

The SARs in the “ice scour” core varied within the range of 0.09–0.95 cm/year in different years (Figure 6A, Table S1 in Supplementary Materials). The average SAR value for a thickness of 40 cm was 0.38 cm/year. The SARs in the “background” core were lower (0.04–0.38 cm/year), with an average value for a thickness of 17.5 cm of 0.16 cm/year (Figure 6B, Table S2 in Supplementary Materials). Comparing the distribution of the calendar ages and SAR values in both of the cores (Figure 5 and Figure 6), one can notice the same trends in the change of rates but they occur at different times. The increase in the sedimentation rates in the ice scour occurs 10–20 years later compared to the background surface. For example, in the “ice scour” core, a sharp SAR increase occurred between 2000 ± 1 and 2002 ± 1, and in the “background” core it was between 1989 ± 4 and 1995 ± 3.

5. Discussion

5.1. Estimation of Missing Inventory of ²¹⁰Pb in the Ice Scour

Since there is no ²¹⁰Pb equilibrium with the initial radionuclide ²²⁶Ra in the “ice scour” core, we estimated the missing inventory of ²¹⁰Pb (A(j), Bq/m²) in the layers below 43 cm, not penetrated by the corer. To do this, we used two methods: the formula of the exponential dependence of the total ²¹⁰Pb on the mass (m_i, g/cm²) y = 159.44e^−0.035x and ¹³⁷Cs time marker estimation [49].

Using the formula of the first method, the total ²¹⁰Pb was reconstructed in the range of 43–51 cm. The mass of the layers was taken from the average value of the layers in the range of 32–40 cm (21.26 g). ²²⁶Ra was calculated as the average value of the specific activity for all of the layers and amounted to 26.4 ± 7 Bq/kg. Subsequently, the calculation was carried out according to the CF model with discreteness of 2 cm. Equilibrium was reached on a layer of 50 cm, which dates back to 1806 AD, which cannot be taken as representative information, as it is beyond the limits of the method capabilities. However, the dates in the range of 0–40 cm did not change significantly.

As a time marker of the second method, we used the ¹³⁷Cs peak (47.3 ± 0.9 Bq/kg), which was found in the 30–32 cm layer. According to the results of our measurements (Table S1 in Supplementary Materials), the inventory of non-equilibrium ²¹⁰Pb (δA₁, Bq/m²) that accumulated over 58 years from 1963 to 2021 (0–32 cm) was 15,341 Bq/m². The inventory of ²¹⁰Pb (A(j)₁ = 2915 Bq/m²) that accumulated before 1963 (layers from 32 cm to the boundary of equilibrium between ²¹⁰Pb and ²²⁶Ra) was calculated using the formula A(j)₁ = δA₁/(e^λt − 1), where t = 58 yr and λ = 0.03118 yr⁻¹ (disintegration constant). Based on the obtained values, the total inventory of non-equilibrium ²¹⁰Pb from the seabed surface to the equilibrium boundary was 18,256 Bq/m² (A(0) = δA₁ + A(j)₁). According to the results of our measurements (Table S1 in Supplementary Materials), the inventory of non-equilibrium ²¹⁰Pb (δA₂, Bq/m²) accumulated in the layers (0–43 cm) was 17,879.5 Bq/m². Then, the missing inventory of ²¹⁰Pb (A(j)₂ = A(0) − δA₂) is 376.5 Bq/m². Based on the activity of non-equilibrium ²¹⁰Pb in the 42–43 cm layer, which is equal to 244 Bq/m², it can be assumed that the missing inventory of 377 Bq/m² accumulated in the 43–45 cm layer. In this case, the equilibrium boundary should be lowered by 3 cm to 46 cm.

The results of the layers’ age calculations using the CF model, taking into account the estimation of the underlying missing inventory (curve “Pb-210 age, ice scour, CF/46”), are shown in the graph (Figure 5). This calculation makes the age a little younger: for layers in the range of 0–26 cm—no more than 2 years, in the range of 26–36 cm—no more than 5 years and in the range of 36–40 cm—no more than 8.5 years. The greatest age change is observed for the 40–42 cm layer, whose age has become 13.5 years younger.

5.2. Application of the Constant Flux Model in the Ice Scour Conditions

The CF model was designed for flat bottom sedimentation without any lateral limitations [49]. However, the accumulation of sediments in an ice scour occurs in a constrained environment. The sedimentation area at the bottom of the ice scour increases with time. In other words, as the ice scour is filled with sediments, the width of the ice scour bottom increases. This phenomenon is expressed in the cross section of the ice scour as an increase in the “width” of the layers upwards. Therefore, we calculated how the layer “width” changes with a step of 1 cm in an ice scour with a geometry such as ours, in order to assess the possible error in the results obtained from this model.

Calculations were made for an ice scour with w₀/S = 15, where w₀—the ice scour width at the present bottom surface, and S—the thickness of the ice scour filling sediments (in our case, w₀ = 12 m and S = 0.8 m). This ratio shows the absolute value of the change in the ice scour width, which is constant and amounts to 15 cm per 1 cm. With this ratio, the angle of inclination of the ice scour sides is 7.6 degrees. The ice scour width at a given depth relative to the present bottom surface was calculated using the formula w_i = 12–15 z_i, where w_i is the ice scour width at a given horizon, and z_i is the depth of horizon. The 0–1 cm layer (the uppermost dated layer) will have a roof “width” of 12.0 m and a base “width” of 11.85 m. The change is 1.25%. In the 38–40 cm layer (the lowest dated layer), the “width” of the roof will be 6.3 m, and the “width” of the base will be 6.0 m. The change is 4.8%. There is a downward trend of increasing the magnitude of the layers’ “width” relative changes which can reach 50–100% at the base of the filling sediments under the condition of an ideal cross section of the ice scour in the form of an isosceles triangle. However, we dated only the upper half of the filling sediments (40 cm out of 80 cm), where the “width” of 1 cm layers in the cross section changes by only 1.25–2.4% per 1 cm.

Such an insignificant change in the 1 cm layers’ “width” in the cross section gives grounds in our case to neglect the influence of the ice scour geometry on sedimentation. Therefore, we consider it reasonable to use the CF model to calculate the age in both cores: on the “background” surface, where the layers’ “width” remains constant, and in 0–40 cm of the ice scour, where the layers’ “width” changes no more than 2.4% per 1 cm. It is important to note that we realize that the results may have additional uncertainty, but it is not significant and does not exceed the standard uncertainty in the measurements themselves. The uncertainty increases with the increase in the thickness of the analyzed layers.

Another important fact is worth noting. As the ice scour fills with sediments, each successive layer becomes “wider”. Therefore, at a constant sediment flux entering the ice scour, the sedimentation rate should decrease with time. However, we observe the opposite picture—the sedimentation rate increases with time. From this, it can be concluded that the sediment flux entering the ice scour has a greater effect on the sedimentation rate than the ice scour layers’ “width”, whose influence is insignificant.

An additional control of the reliability of the ice scour dating results can be undertaken by comparing them with the results of the “background” core dating. Due to the close location (only 140 m between the cores), the general tendencies of sedimentation in this bottom area should be traced in both of the cores, while the sedimentation rates can be determined according to local conditions (in our case, according to the features of the bottom topography). Indeed, in both cores, an exponential distribution of ²¹⁰Pb was observed, whereby the fluctuations (peaks) of which are correlated between the cores (Figure 4). However, there is a 10–20 year lag of the sharp SAR changes in the ice scour relative to the background surface (Figure 6), which may be due to uncertainty due to the layers’ width changes in the ice scour.

To eliminate this uncertainty, an attempt was made to modify the formulae of the CF model to take into account the change with depth in the sedimentary layers’ width in the scour. The calculation using these modified formulae was designated as CFw (Table S3 in Supplementary Materials). Whereas the CF model calculation is based on the measured specific activity of radionuclides in the core, the CFw calculation is based on an estimation of the excess lead activity in all of the sediments filling the ice scour. To extrapolate the excess lead activity that had accumulated over the entire cross section of the ice scour, the excess lead activity in each layer was multiplied by the proportionality factor (k_i) of the ice scour width at a given depth to the ice scour width at the lowest horizon with the excess lead activity. Thus, at the lowest horizon (44 cm), k_i is equal to 1, and at the highest horizon (the present surface of the bottom of the ice scour), k_i reaches a value of 2.22. It was decided that k_i be used, because it was noted that the results of the age and SAR calculations were affected by the rate of increase in values with each subsequent step in the horizon depth change, and not by the absolute values themselves, and by which, the excess lead activity in the core was multiplied. Thus, the calculation results were the same both when the excess lead activity was multiplied by the ice scour width at a given horizon depth, and when the excess lead activity was multiplied by the ratio of the ice scour width at a given horizon depth to the core diameter. The modified formulae used for the CFw calculation are presented in the Supplementary Materials.

The calendar age obtained from the CFw calculation turned out to be no more than 10 years earlier than the age obtained from the CF model (Figure 5, curve “Pb-210 age, ice scour, CFw/45″). Compared to the background surface, the lag of sharp changes in the SAR in the ice scour has decreased, but it still remains and has reached about 10 years (Figure 6C). This may be due to the lower detail and higher uncertainty in the “background” core compared to the “ice scour” core. The uncertainty in the calendar age calculation increases from the seabed surface down the core and reaches ±5 years in the ice scour and ±18 years on the background surface. Thus, the modified formulae (CFw) correct the age within the method uncertainty; so, this, together with what was noted above, gives us a reason to use the results obtained from the measured specific activities of radionuclides without taking into account the changing width of the ice scour (CF model) to estimate the ice scour age.

5.3. Verification of Results with ¹³⁷Cs

The lowest layer from the “ice scour” core, where ¹³⁷Cs was found (34–36 cm), can be attributed to 1954 AD (1950.5 ± 3.2 AD at 34 cm according to the CF model, Figure 5). This is the year when the first high-performance thermonuclear weapon was detonated and atmospheric levels of ¹³⁷Cs in the Northern Hemisphere rose high enough to create a measurable signal that could be detected in soils and sediments in water bodies [53]. The ¹³⁷Cs specific activity increased approximately 20 times in the overlying layer (30–32 cm). Seventy-five atmospheric nuclear tests were carried out on the Novaya Zemlya archipelago between 1954 and 1962, including the largest thermonuclear charge ever detonated (“Tsar Bomba”) [54]. The global number of atmospheric nuclear tests peaked at 110 explosions in 1962 [55]. The peak concentration of atmospheric radioactivity was recorded in 1963 as a result of atmospheric nuclear testing [56]; so, the calendar age of the 30–32 cm layer should be attributed to this year, which is comparable to the chronological data obtained using the CF model (1967.1 ± 2.2 AD at 30 cm and 1959.2 ± 2.6 AD at 32 cm, Figure 5).

The further distribution of ¹³⁷Cs in the “ice scour” core from a layer 26–27 cm to the surface layer did not make it possible to identify the increased specific activity of the radionuclide designated as a reference in the literature, for example, the transoceanic transport of radioactive discharges of nuclear fuel reprocessing plant Sellafield, UK, to the Irish Sea (1982–1983) and atmospheric fallout after the accident at the Chernobyl Atomic Electric Power Station (1986) [57]. Thus, according to the ¹³⁷Cs data, the upper 34 cm in the “ice scour” core accumulated after 1954 AD over the last 67.8 years, and the average SAR value was 0.5 cm/year for a thickness of 34 cm, which is higher, but for a lower thickness than according to ²¹⁰Pb (0.38 cm/year for 40 cm).

The “background” core is characterized by low ¹³⁷Cs specific activity (<MDA), except for two layers (1–2 and 10–11 cm). This is probably due to periodic seabed erosion, which prevents the accumulation of ¹³⁷Cs in seabed sediments. The appearance of ¹³⁷Cs in the 10–11 cm layer allows us to interpret the calendar age of this layer as 1963 AD, which is comparable with the age obtained from ²¹⁰Pb (1965.6 ± 8.0 AD at 11 cm). Thus, according to the ¹³⁷Cs data, the upper 10 cm in the “background” core accumulated after 1963 AD over the last 58.8 years, and the average SAR value was 0.17 cm/year for a thickness of 10 cm, which is comparable to ²¹⁰Pb results (0.16 cm/year for 17.5 cm) even for a lower thickness.

5.4. Formation of Sediments and Dynamics of Their Accumulation

The “background” core is more than six times longer than the “ice scour” core. We explain the small length of the “ice scour” core by the presence of a compacted horizon, which is associated with the pressure during gouging. Probably, the corer did not hit the ice scour axis, but the lower part of the ice scour wall, which is covered by only 43 cm (or so) of filling post-gouging sediments. The GPS coordinates of the sampling point that were correlated with the DEM of the bottom also show the displacement of the sampling site toward the ice scour wall relative to its axis (Figure 2B). Thus, we have reason to believe that 43 cm of the sampled sediments lie on pre-exaration deposits. Older post-exaration deposits that fill the ice scour are located in the axial part of the ice scour at a depth of 43–80 cm below the modern bottom of the ice scour.

According to the grain size analysis (Table 2 and Table 3), sediment accumulation in the ice scour and at the “background” surface occurs mainly due to fine sediments (pelite and silt). In the ice scour, in contrast to the “background” surface, there is an inclusion of heterogeneous sand and fine gravel, which accumulated here as a result of slope subaqueous processes, saltation or ice rafting. The low content of sand and gravel, as well as the lower natural moisture content in the sediments of the “background” surface, may indicate higher hydrodynamics outside of the ice scour. Since the studied sediment cores are located quite close to each other (only 140 m), the quantity and quality (grain size composition) of sedimentary material entered in the ice scour and on the “background” surface are the same; however, active hydrodynamics outside of the ice scour lead to periodic erosion of the “background” surface and to the removal of sand and gravel. Under conditions of low hydrodynamics in the ice scour, coarse particles originating from ice rafting, slope processes and saltation are not able to move laterally and remain where they settled on the bottom. This fact explains the absence of the ¹³⁷Cs marker in the sediments of the “background” core and the availability of this isotope in high concentrations in the ice scour (Table 4 and Table 5). In other words, the sediments that have accumulated into the ice scour are stored there as in a sedimentological trap and the sediments of the “background” surface pass in transit, only being partially deposited. Thus, the obtained results additionally confirm that one of the sediment cores was indeed taken directly from the ice scour.

To quantify the sedimentological trap effect, the excess lead activity accumulated on the present seabed surface in the “background” core was multiplied by the proportionality factor of the ice scour width at the level of the present seabed surface to the ice scour width at the lowest horizon with excess lead activity (k₀ = 2.22). The obtained value was 6684 Bq/m² on the background surface, while in the ice scour it was 31,497 Bq/m². Thus, the studied ice scour accumulated 4.7 times more lead than the background surface.

The main sources of fine sediments accumulated in Baydaratskaya Bay are the erosion of its coasts and partly the runoff of the Ust-Kara River (Figure 1). The northern coast of the bay is mainly composed of homogeneous sandy deposits with relatively low ice content [58,59] and is actively eroded at an average rate of 0.3 to 1.2 m/year [60]. The southern coast of the bay is composed of medium ice-rich sandy loam deposits, underlain by medium or, in some places, strongly ice-rich boulder loams and clays [61], and is also actively eroded at an average rate of 0.5 to 2.4 m/year [60]. Taking into account the grain size composition of the studied sediment cores, it can be assumed that sedimentation in the ice scour and near it occurs due to the erosion of the southern coast of Baydaratskaya Bay.

Sediments of the last 100–120 years and older are composed of the same grain size. Consequently, the source of sedimentation material (erosion products of the southern coast of the bay) and lithodynamics (secondary seabed erosion) have not changed in the studied area. Taking into account that sedimentation in the study area is closely related to coastal erosion, one can note a certain correlation between the coastal dynamics of Baydaratskaya Bay and the sedimentation rates in the studied sediment cores. According to our data (Figure 6), the SARs have reached their maximum values in recent years, since the 2000s. The maximum coastal erosion of the bay also belongs to the same period [60]. In the 1990s, when coastal erosion of the bay slowed down, we noted a decrease in the sedimentation rate. At the same time, it is known that the dynamics of coastal erosion are closely related to the climate and depend on combinations of wind–wave and thermal factors within the region [60]. Consequently, the rates of marine sedimentation also depend on the climate.

5.5. The Age of the Ice Scour

The ²¹⁰Pb method allowed us to date the upper part (0–0.4 m) of the sediments filling the ice scour, whereby thickness was estimated from seismic profiles to be about 0.6–0.8 m. We estimated the age of the ice scour for a filling sediment thickness of 0.7 m, assuming that age variations can occur both upwards and downwards due to variations in the actual thickness of the filling sediments. The age estimation of sediments in the range of 0.4–0.7 m is based on the extrapolation of the maximum and minimum possible mean sedimentation rates under certain sedimentation conditions.

It is known that marine sedimentation is determined by a few factors: climate, water salinity, basin depth, gas regime, the presence and nature of currents and biological activity [62]. At the same time, air temperature, as an important indicator of the climate, is a significant factor determining basin sedimentation. Often, there is a strong correlation between climatic components and sedimentation [63].

According to our calculations, the top 40 cm of sediment accumulated in the ice scour over about 105 ± 5 years, starting around 1917 ± 5 AD (mean SAR 0.38 cm/year). At the same time, more than 1/3 of the dated sediments (0–15 cm) accumulated over the past 20 ± 1 years (since around 2002 ± 1 AD) at a mean rate of about 0.75 cm/year. This period is characterized by an increase in air temperature in the Baydaratskaya Bay area, which accelerated the processes of coastal erosion [64], increasing the sediment supply in the accumulation basin. Conversely, a decrease in the rate of sedimentation is observed with a decrease in the sum of daily positive air temperatures in the period of 1975–1985. In the 1960s, the dependence of sedimentation rates on air temperatures persisted. The sum of daily positive air temperatures increased and sedimentation proceeded more dynamically than in the next decade.

Thus, based on the past temperature regime, we can indirectly assume the mean sedimentation rate in the ice scour before 1917. Due to the lack of meteorological observations in the work area, we decided to rely on data on climatic anomalies of the northern hemisphere from 1850 to the present [65,66,67]. According to these data, the value of the temperature anomaly in 1900–1930 was unchanged. Considering this, it can be assumed that the mean SAR in the ice scour in that time was not less than the mean SAR established for the period 1916.8–1934.3 (38–40 cm), i.e., around 0.11 cm/year. In this case, in 1900–1916.8, a layer of 40–41.8 cm was formed; however, we found that a 42–43 cm layer contained excess ²¹⁰Pb. Therefore, its age cannot theoretically exceed 120 years (1901.8 AD) due to the dating limit of the method. Rounding the age of the beginning of the 42–43 cm layer formation to 1900 AD, it can be assumed that in 1900–1916.8, a layer of 40–43 cm was formed at a mean rate of 0.17 cm/year, rather than 40–41.8 cm. The values of temperature anomalies in 1850–1900 approximately corresponded to the temperature regime in 1930–1960. Therefore, we can assume that the mean SAR in the ice scour in 1850–1900 was not less than the mean SAR established for the period 1934.3–1959.2 (32–38 cm), i.e., around 0.24 cm/year. In this case, a layer of 43–55 cm was formed in 1850–1900. Thus, the mean sedimentation rate of the 15 cm stratum (40–55 cm) for 66.8 years (1850–1916.8) could be 0.22 cm/year. We consider this value as the minimum possible sedimentation rate before 1917. Considering that the warming since 2000 has been unprecedented, the mean rate of 0.38 cm/year determined for the thickness of 0–40 cm can be the maximum possible sedimentation rate for the period before 1917.

The extrapolation of the minimum (0.22 cm/year) and maximum (0.38 cm/year) possible sedimentation rates in the ice scour before 1917 allows us to estimate the time range at which the accumulation of the first sediments in the ice scour at a depth of 70 cm below the current seabed surface began. This time range is defined as 1780–1840. Thus, the age of the ice scour can be determined as 1810 ± 30 AD. The mean rate of ice scour filling with 70 cm thick sediments from the moment of its formation is around 0.33 cm/year.

6. Conclusions

• The studied ice scour, located in front of the entrance to Baydaratskaya Bay of the Kara Sea at a sea depth of about 28–32 m, has a serpentine-shaped plan configuration, changing its direction 2–3 times and reaching a length of at least 30–35 km. The maximum visible depth reaches 3.2 m, and the maximum width is up to 35 m. At present, it is the largest ice scour among those known in this region of the Kara Sea.
• Two sediment cores were studied, which were taken on 2 November 2021 (2021.8 A.D.) using a gravity corer directly in the ice scour and on the “background” seabed surface, which was not processed via ice scouring, and 140 m south of the first one. At the core sampling site, the apparent depth of the ice scour cutting into the background seabed surface was about 2.4 m. According to the seismic profiles, the thickness of sediments filling the ice scour was estimated at about 0.6–0.8 m, whereby the top 30 cm of which was presented exclusively by clayey silt. The pelite content varied from 52.9 to 68.4%, the silt content varied from 23.4 to 29.5% and the heterogeneous sand content varied from 4.2 to 16.4%. Some layers contained gravel grains. The highest content of fine gravel inclusions occurred in the upper part of the core. On the “background” seabed surface, the upper 50 cm of sediments was also represented by clayey silt. The pelite content varied from 56.6 to 72.7%, the silt content varied from 27.1 to 39.4% and the inclusion of heterogeneous sand and gravel was not significant. The low content of the sand and gravel in the sediments of the background surface is explained by higher hydrodynamics outside of the ice scour, which leads to the periodic erosion of sediments and the removal of rare sand and gravel particles.
• The excess ²¹⁰Pb was found in all of the analyzed layers (up to 43 cm) of the ice scour sediments, reaching maximum values of specific activity (108 Bq/kg) in the upper horizons and decreasing to 48–55.9 Bq/kg toward the lower ones. The equilibrium of ²¹⁰Pb with the initial radionuclide ²²⁶Ra was not revealed at the layer of 42–43 cm, but there was a tendency to approach it. The technogenic radionuclide ¹³⁷Cs below 36 cm was not detected, while above its content it ranged from 47.3 Bq/kg to values less than the minimum detectable activity (<0.2 Bq/kg).
• In deposits on the “background” seabed surface, the excess ²¹⁰Pb was only found in the upper 25.5 cm. Its specific activity decreased from top to bottom from 60 Bq/kg to 26.7 Bq/kg. The equilibrium of ²¹⁰Pb with the initial radionuclide ²²⁶Ra was observed in the layer of 19.5–21.5 cm. The ¹³⁷Cs specific activity was below 0.2 Bq/kg at all horizons, except for the 1–2 cm and 10–11 cm layers. The low content of the technogenic radionuclide ¹³⁷Cs also indicates the periodic erosion of sediments.
• Based on ²¹⁰Pb dating, the time of the beginning of sediment accumulation in the ice scour at a depth of 15 cm was estimated to be around 2002 AD; at a depth of 38 cm—around 1934 AD; at a depth of 40 cm—around 1917 AD and at a depth of 43 cm—around 1900 AD. Thus, over the past 120 years, there has been an increase in the mean SARs: 0.79 cm/year for the 0–15 cm horizon, 0.43 cm/year for the 0–38 cm horizon, 0.38 cm/year for the 0–40 cm thickness and not less than 0.35 cm/year for the 0–43 cm horizon.
• On the “background” seabed surface outside of the ice scour, the mean sedimentation rate over the past 110 years has been two times lower. The time of the beginning of sediment accumulation outside of the ice scour at a depth of 17.5 cm is estimated to be around 1910 AD (0.16 cm/year). Fluctuations in the mean SARs are not pronounced, which may be due to the periodic erosion of sediments outside of the ice scour.
• There is a close correlation between the marine sedimentation rates and air temperature fluctuations, as well as the coastal retreat rates of Baydaratskaya Bay, whereby the erosion products of which are the main source of seabed sediments due to the absence of large rivers in the area. Thus, since 2002, in the Baydaratskaya Bay area, there has been a sharp increase in air temperature, the rate of coastal retreat and the rate of sedimentation in the largest ice scour of this region.
• According to the results of ²¹⁰Pb dating, the studied ice scour was formed no later than the end of the Little Ice Age (LIA) in the Arctic (turn of the 19th and 20th centuries). The age of the ice scour is estimated to be 1810 ± 30 AD based on the extrapolation of possible sedimentation rates prior to 1917 (0.22–0.38 cm/year). The mean rate of ice scour filling with 70 cm thick sediments from the moment of its formation is around 0.33 cm/year. Assuming that after the end of the LIA, the size of icebergs decreased, their penetration into Baydaratskaya Bay improved. Therefore, the ice scours of Baydaratskaya Bay were probably formed mainly after the end of the LIA, i.e., in the 20th century.
• Further study of the sedimentation chronology in ice scours will help to establish the periods of active ice scouring on the glaciated continental margins and to supplement knowledge about sedimentation on the Arctic shelf.