Estimation of Waters Vertical Structure in the Barents and Kara Seas

Abstract

Investigation of the features of spatial-temporary distribution of the Brunt-Väisäla frequency maximum and its occurrence depth in the Barents and Kara Seas is considered on the basis of the World Ocean Atlas 2013 reanalysis for the period 1955–2012 with 0.25° × 0.25° grid. The regional features of the vertical structure of the buoyancy frequency are revealed. Correlations between intra-annual variability of the Brunt-Väisäla maximum frequency and climatic indexes are investigated.

1 Introduction

Climate change occurring in the modern epoch is especially noticeable in the Polar Regions, the most sensitive to changes in hydrometeorological conditions and in turn having a serious impact on regional and global climate processes (Frolov et al. 2010; Matishov 2008). In recent decades, in the Arctic region there are significant fluctuations in the area of sea ice and an increase in surface air temperature, also changes in the vertical structure of the upper Ocean layer are observed (Frolov et al. 2010; Bukatov et al. 2017; The second estimating report on climate changes and their consequences in the territory of the Russian Federation 2014; Tsaturov and Klepikov 2012). The Barents and Kara Seas are the two most western seas of the Russian Arctic. They are separated by the Novaya Zemlya archipelago, which serves as a barrier on the way from the Atlantic of warm air masses and warm and saline Ocean waters. Owing to the North Cape Current a significant part of the Barents Sea area from the coast to 75° N does not freeze. The Kara Sea is climatically more severe than the Barents Sea: it is covered with ice completely or over a large area (Bukatov et al. 2017; Dobrovolsky and Zalogin 1982; Karklin et al. 2017).

A major role in the hydrological regime formation of the Arctic seas is played by the flow of fresh continental waters. The river flow of the Barents Sea is small relative to the area of the sea, its maximum is observed at the end of spring. The Kara Sea has about 55% of the total fresh water which flow into all seas of the Siberian Arctic. It is considerably higher than the volume of river water entering in the Barents Sea. The maximum of river water in the Kara Sea is noted at the end of summer—beginning of autumn. Almost half of the sea area is under the influence of continental waters (Dobrovolsky and Zalogin 1982).

The most important characteristic of the hydrological regime is vertical stratification. A comprehensive study of the vertical structure of the Barents and Kara Seas waters is relevant for understanding of the functioning of their ecosystems, as well as ensuring the sustainable economic development of the Russian Arctic. Climatic conditions, the presence of currents, continental runoff and composite bottom relief contribute to the appearance of features of the water vertical structure. The indicator characterizing the stratification of the water column is the buoyancy frequency (the Brunt-Väisäla frequency), which is also called the frequency of stratification. It expresses the magnitude of stability and it is a fundamental variable in the dynamics of a stratified fluid. The shape of the vertical profile of the buoyancy frequency is an indicator of the presence of certain water masses and the intensity of various hydrodynamic processes (Sherstyankin and Kuimova 2009).

The purpose of this work is to study the spatial-temporary variability of the vertical structure of the Brunt-Väisäla frequency in the Barents and Kara Seas.

2 Method and Data

The study was carried out on the base of World Ocean Atlas 2013. The calculations used the values of temperature and salinity obtained for the time period 1955–2012 in the 0.25° × 0.25° grid with a step of 5 m in depth to the horizon of 100 and 25 m—to the horizon 150 m (Locarnini et al. 2013; Zweng et al. 2013). Field of study is from 16° to 105° E and from 66° to 82° N.

According to monthly average data on temperature and salinity the density was calculated. The obtained density arrays were checked for the presence of inversions. If necessary they were corrected by replacing the inversion values with values interpolated over neighboring horizons (Artamonov et al. 2004). Based on the formed density arrays the buoyancy frequency (N) was calculated. The maximum Brunt-Väisäla frequency by depth (Nmax(z)) was determined and its occurrence depth (HNmax(z)) too. To identify consistency between the maximum of Brunt-Väisäla frequency and climatic indices NAO (North Atlantic Oscillation), SOI (Southern Oscillation), AO (Arctic Oscillation), TDO (Pacific Decade Oscillation) (https://www.cpc.ncep.noaa.gov/products/site_index.shtml), GSNW (Gulfstream Current Index) (https://web.pml.ac.uk/gulfstream/data.htm) the calculations of the correlation coefficients R_Nmax~NAO, R_Nmax~SOI, R_Nmax~AO, R_Nmax~TDO, R_Nmax~GSNW were made.

3 Analysis of Results

To characterize the horizontal distribution of the waters stability of the Barents and Kara Seas the maps of the geographical distribution of the monthly mean climatic values of the maximum of the Brunt-Väisäla frequency and its occurrence depth were obtained. Analysis of the maps for the cold months shows that in these months the waters stability of the Barents Sea is small or close to an unstable state. The values of the maximum of buoyancy frequency of in the Kara Sea are times higher than Nmax(z) in the Barents Sea. As an illustration, the map of the spatial distribution of Nmax(z) in the Barents and Kara Seas in the coldest month (February) of the year in this region is presented (Fig. 1).

Fig. 1

The distribution of the maximum of the Brunt-Väisäla frequency (cycle/hour) in the Barents and Kara Seas in February

The geographical distribution of the Brunt-Väisäla frequency maximum in the Barents and Kara Seas in June is presented in Fig. 2. It is seen, that in June waters with maximum stability are located in the coastal areas: along the northeast coast of the Kola Peninsula and the western coast of the Yamal Peninsula, in the mouth areas of the Ob and Yenisei. The central part of the Barents Sea is occupied by waters with small values of Nmax(z) (less than 5 cycles/hour). To the north of 75 parallel, the maximum of buoyancy frequency rises to 10 cycle/hour. The central part of the Kara Sea is also occupied by waters with relatively low values of Nmax(z). From 75° to 80° N the values of the Brunt-Väisäla frequency maximum are about 15 cycles/hour. The increased stability in June in the southern part of the Barents and Kara Seas is due to the influx of fresh continental waters creating a salinity gradient in the coastal zone. Surface heating also creates a temperature gradient at this time of the year, but it does not have such a large effect on stability which has a salinity gradient (Timofeev 1946). The high stability of the waters in the Barents Sea in June to the north of the 75th parallel is caused exceptionally by the salinity gradient which is result of the ice melting. The low stability in the central part of the Barents Sea (contour of Nmax(z) = 5 cycle/hour) is explained by the fact that the Atlantic waters of the North Cape Current, which have high salinity and temperature, pass through here. While these waters moving eastward they are cooled which leads to an increase in density and to a smoothing of the difference in density between the layers (Timofeev 1946).

Fig. 2

The distribution of the maximum of the Brunt-Väisäla frequency (cycle/hour) in the Barents and Kara Seas in June

The relatively low values of the buoyancy frequency maximum in the central part of the Kara Sea in June are due to the fact that the coastal waters desalinated by river flow have not yet spread far to the north.

Stability in July increases over the entire area of the Seas (Fig. 3). Especially large increase is observed in the southeastern part of the Barents Sea (in the southern tip of Novaya Zemlya) and the southwestern part of the Kara Sea (the western coast of the Yamal Peninsula), the deltas of the Ob and Yenisei Rivers, and the Pyasinsky Bay. The rise in water stratification in these zones is explained by the increase in the flow of Siberian Rivers. The increase in stability from June to July in the central part of the Barents Sea is primarily due to the more intensive influx of freshened waters from the north to this region, as well as by the weaker surface cooling of warm Atlantic waters coming from the west.

Fig. 3

The distribution of the maximum of the Brunt-Väisäla frequency (cycle/hour) in the Barents and Kara Seas in July

The distribution of the buoyancy frequency maximum in August–September over the Barents and Kara Seas remains the same as in July, although the values of Nmax(z) changes.

In order to assess and compare the annual behavior of sustainability in the Barents and Kara Seas, graphs of the climatic intra-annual variability of the Brunt-Väisäla frequency maximum at the Kola (32° E) and Yamal (73° E) meridians were created (Fig. 4a, b). From the presented Figures it can be seen that the highest values of water stability on the Kola meridian occur in May–September and can reach 15 cycle/hour in the region of 70° N and 17 cycle/hour in the region of 78° N. The maximum stability at 70 parallels is due to the May river flow, at 78° N—intense ice melting in August. The maximum values of Nmax(z) on the Yamal meridian are observed from June to October and can be 85 cycle/hour in the region of 73° N (influence of the river Ob flow). Such a significant excess of Nmax(z) in the Kara Sea over Nmax(z) in the Barents Sea is due to the difference in the flow of fresh continental waters. River flow in the Barents Sea is about 163 km³ per year; in the Kara Sea is about 1300 km³ per year (Dobrovolsky and Zalogin 1982).

Fig. 4

Annual variation of the Brunt-Väisäla frequency maximum in a Kola (32° E), b Yamal (73° E) meridians

For a quantitative characteristic of the waters stability fluctuations of the Barents and Kara Seas, the average value of the Brunt-Väisäla frequency $\left( {\overline{N} max(z)} \right)$ and the average occurrence depth of the maximum on the meridional sections $\left( {\overline{HN} max(z)} \right)$ were calculated (Figs. 5 and 6). The values $\overline{N}max(z)$ and $\overline{HN} max(z)$ are obtained as arithmetic averages on sections.

Fig. 5

The average along the meridian buoyancy frequency maximum

Fig. 6

The average along the meridian occurrence depth of the buoyancy frequency maximum

It seen that the maximum stability in the Barents Sea comes in July–August, and in the Kara Sea in August–September. In the zone of influence of the Ob and Yenisei Rivers, the values of $\overline{N}max(z)$ reach 40–45 cycle/hour, and in the Barents Sea $\overline{N}max(z)$ does not exceed 10–12 cycles/hour. In addition, there is one more of stability maximum in November in the meridional sector 70–80° E. In October, in the Kara Sea with the beginning of the process of ice formation, stability decreases everywhere. In November, with the formation of an ice cover, the river waters begin to spread in the under-ice horizons. It is leads to an increase in stability in the mouths of the Ob and Yenisei. From Fig. 6 it can be seen that in summer the layer of density jump rises to the surface, and in winter it deepens. In summer, the average depth of the pycnocline is 20–25 m in the Barents Sea and 10–15 m in the Kara Sea. In winter, the average depth of the density jump layer reaches 100 min the Barents Sea and 40 min the Kara Sea. The calculated correlation coefficient between the average for the meridian of the buoyancy frequency maximum and the average for the meridian of occurrence depth of the density jump layer is equal to −0.7. This indicates that there is an inverse relationship between $\overline{N}max(z)$ and $\overline{HN} max(z)$: the large (summer) values of the density gradient are observed at shallow depths (10–25 m), and vice versa, the small values (winter) at relatively large depths (40–100 m).

The intra-annual variability of the average value of the buoyancy frequency maximum and the depth of its occurrence by sectors are shown on Fig. 7. Sector 1—20°–50° E (Barents Sea), sector 2—50°–70° E (Novozemelsky district), sector 3—70°–100° E (Kara Sea). It is seen from the figures that in the Barents Sea, the maximum of the waters stability is reached in August, in the Novaya Zemlya sector—in July and November, in the Kara Sea—in September and November. At the same time, the differences in the annual behavior of sustainability associated with the regional features of the hydrological waters structure are clearly visible. Average values of the buoyancy frequency maximum in the Barents Sea are relatively small (4–12 cycles/hour), in the Novaya Zemlya region they are already 8–22 cycles/hour, and in the Kara Sea—16–34 cycles/hour. The minimum occurrence depth of $\overline{N}max(z)$ is observed in July: in the Barents Sea it is 19 m, near Novaya Zemlya—10 m, in the Kara Sea—12 m.

Fig. 7

Intra-annual variability of the average value of the buoyancy frequency maximum (a) and its occurrence depth (b) by sectors: curve 1—sector 20°–50° E, curve 2—sector 50°–70° E, curve 3—sector 70°–100° E

For a more detailed analysis of the vertical structure of the Barents and Kara Seas density field, horizontally averaged Brunt-Väisäla frequency profiles were created in the physiographic regions proposed in Gorbatsky (1970). In the Barents Sea 5 areas are allocated: (1) Southwest or Medvezhinsky; (2) Northwest or Spitsbergen; (3) North; (4) Northeast; (5) Southeast or Kolguevo-Vaigachsky (Fig. 8).

Fig. 8

Averaged profiles of Brunt-Väisäla frequency in the regions of the Barents Sea: 1—Southwest, 2—Northwest, 3—North, 4—Northeast, 5—Southeast

The basis for allocate of the Southwest region is the influence of the North Cape Current, which is decisive in climate forming and the hydrological regime of this part of the Barents Sea. Northwest and North regions are considered as regions of the Arctic Zone. The Northeast region is influenced by the Atlantic waters. It is characterized by a maritime climate with anomalously mild for high latitudes winter. In a relatively shallow Southeast region there is a vast waters zone with low salinity, formed under the influence of the Pechora River flow (Gorbatsky 1970).

In the Kara Sea 4 regions are allocated: (6) Southwest, (7) Northwest, (8) Central, (9) North (Fig. 9). The boundaries of the regions are given approximately. In the southwestern part the Kara Sea is affected by warmer and saltier Barents Sea waters through the straits of the Yugorsky Shar, the Karsky Gates, what is the basis for allocate the Southwestern region.

Fig. 9

Averaged profiles of the Brunt-Väisäla frequency in the Kara Sea regions: 6–Southwest, 7—Northwest, 8—Central, 9—North

The hydrological regime of the Northwest region of the Kara Sea is formed under the influence of the Barents Sea waters (Matochkin Strait) and waters of the Ob and Yenisei. Desalination of the surface layer by river water affects on hundreds of kilometers north of river mouths. The hydrological regime of the Central region of the Kara Sea is affected by the runoff of the Siberian Rivers (Ob, Yenisei, Pyasina). More salty and warmer Atlantic waters penetrate into the North region of the Kara Sea through the troughs of St. Anne and Voronin between the Franz Josef Land and Novaya Zemlya (Petrov 2008).

The vertical structure of the buoyancy frequency space-averaged in the Barents Sea is shown in Fig. 8. It can be seen that in each of the selected areas the vertical structure of the buoyancy frequency has regional features. Thus, the structure of the Southwest (1) and Northwest (2) regions is of the same type (zone of Atlantic waters): the density jump layer is expressed from July to October, the occurrence depth of $\overline{N}max(z)$ is 20–30 m in summer and 40–50 m—in the autumn. In the North region (3) the seasonal density jump layer is practically not observed in the winter months. In spring, summer and autumn, seasonal pycnocline is distinctly, $\overline{N}max(z)$ values can exceed 10 cycle/hour, $\overline{HN} max(z)$ is 30–60 m. A constant pycnocline is traced at a depth of 90–110 m, values $\overline{N}max(z)$ here are 3–5 cycle/hour. In the Northeast region of the Barents Sea (4) the stability in winter and spring is close to indifferent, from July to September a mild pycnocline is traced: the $\overline{N}max(z)$ values at a depth of about 30 m do not exceed 3 cycles/hour. At the same time throughout all the year at the depth of about 100 m there is a slight constant density jump, which is most distinctly in winter. In the Southeast (5) district, seasonal pycnocline is traced from May to October.

On Fig. 9 the profiles of averaged values of the Brunt-Väisäla frequency in the Kara Sea is shown. It is seen that in the Southwest (6) and Northwest (7) regions, the density jump layer persists throughout the year. In May–June a seasonal pycnocline begins to form. From June to December the layer of density jump is distinctly.

The average values of the Brunt-Väisäla frequency reach 30 cycles/hour. The occurrence depth of the averaged Nmax(z) is 15–20 m. The Brunt-Väisäla frequency profile in regions 6, 7, 8 has a special view in September. At a depth of about 15 m there is a decrease of $\overline{N}$ and then at 20–25 m, the buoyancy frequency reaches its maximum. Perhaps this is due to the beginning of the processes of convective mixing and ice formation in the zone of water low salinity.

To assess the consistency of the maximum of the Brunt-Väisäla frequency with climate indices the correlation coefficients between Nmax(z) and NAO, SOI, AO, TDO, GSNW (R_Nmax~NAO, R_Nmax~SOI, R_Nmax~AO, R_Nmax~TDO, R_Nmax~GSNW) were obtained (Fig. 10). At a 90% confidence interval significant values R > 0.497 (Brooks and Karuzers 1963). It is obtained, that between the maximum of the Brunt-Väisäla frequency and the North-Atlantic Oscillation index a positive correlation is observed on the large part of the Barents Sea (Northeast physiographic region) with a maximum correlation coefficient of 0.9. In the Kara Sea near the northern tip of Novaya Zemlya and at the mouth of the Ob R_Nmax~NAO reach ~0.8. There is a positive correlation between Nmax(z) and AO in the North physiographic region of the Barents Sea with R_Nmax~AO ~ 0.8 and negative correlation in the North and Central regions of the Kara Sea with R_Nmax~AO ~ −0.7. In the North region of the Barents Sea and in the North and Central regions of the Kara Sea, there is a negative correlation between Nmax(z) and SOI, Nmax(z) and TDO. The values of R_Nmax~SOI, R_Nmax~TDO reach ~−0.9. A positive correlation is observed between Nmax(z) and GSNW in the North and Central regions of the Kara Sea with maximum values of R_Nmax~GSNW ~ 0.9.

Fig. 10

Regional distribution of correlation coefficients Nmax(z) with climatic indices: NAO, SOI, AO, TDO and GSNW

4 Conclusions

Based on thermohaline data for the period 1955–2012 with a resolution of 0.25° × 0.25°, the features of the spatial-temporary distribution of the buoyancy frequency maximum and its occurrence depth in the Barents and Kara Seas were studied. Regional peculiarities of the vertical structure of Brunt-Väisäla frequency are revealed. The correlation relationships of the intra-annual variability of the buoyancy frequency maximum with climatic indices: NAO, SOI, AO, TDO and GSNW were studied.

It is shown, that the values of the buoyancy frequency maximum in the Kara Sea are several times higher than Nmax(z) in the Barents Sea, due to the difference in the volume of freshwater flow of continental waters.

It was revealed that the maximum waters stability in the Barents Sea comes in July–August, in the Kara Sea—in September, November. These months the averaged over the meridian values of the buoyancy frequency maximum in the Barents Sea is 10–12 cycles/hour, in the Kara Sea—35–40 cycles/hour.

It is established, that between the maximum of the Brunt-Väisäla frequency and the occurrence depth of the density jump layer there is distinctly relationship with R ~ −0.7. The large (summer) values of Nmax(z) are observed at relatively shallow depths (10–25 m), and vice versa, the small (winter) values—at large depths (40–100 m).

It is shown, that between Nmax(z) and NAO, Nmax(z) and AO there is a significant positive correlation in the North and Northeast regions of the Barents Sea. The correlation coefficients reach ~0.8. There is a positive correlation with a maximum coefficient ~0.9 between Nmax(z) and GSNW in the North and Central regions of the Kara Sea. In the North region of the Barents Sea, in the North and Central regions of the Kara Sea there is a negative correlation relationship between Nmax(z) and SOI, Nmax(z) and TDO with R ~ −0.9.