Modelling seasonal and intraseasonal variations of circulation, temperature, salinity and sea level in the Bellingshausen Sea and on the Antarctic Peninsula shelf

The objective of the study is to simulate using numerical methods the seasonal and intraseasonal variations of circulation, sea level, temperature and salinity in the Bellingshausen Sea and on the shelf of the western part of the Antarctic Peninsula (WAP). The Semi-implicit Cross-scale Hydroscience Integrated System Model with an unstructured triangular horizontal grid and a vertical local sigma coordinate system and ice dynamic-thermodynamic Finite-Element Sea Ice Model were applied. Heat, momentum and salt fluxes were set on the ocean surface using the ERA5 reanalysis data. At the open boundaries, the vertical distribution of temperature and salinity was determined according to the COPERNICUS reanalysis. At the western open boundary of the computational domain, the vertical distribution of velocity of currents from COPERNICUS was also specified, whereas at the open eastern boundary the level deviations were specified. Time variability analysis of sea level was performed using wavelet analysis. The results of modelling of the sea level, temperature, and salinity fields for 2014–2015 were compared with the available observational data on the shelf of the WAP, including data from the Ukrainian Antarctic Expedition. The simulated horizontal and vertical distributions of the subsurface layer with minimum of potential temperature Tmin (Winter Water) are given. The depth of the Tmin varies in the range of 10–100 m increasing to the north. The values of minimum of potential temperature Tmin also increase to the north from –1.8 to 1.2 °C. The intraseasonal oscillations of sea level computed by the model for 2014—2015 were analysed together with data of observations at the tidal stations Faraday/Akademik Vernadsky and Rothera located at the coast of WAP. In the range 1–150 days the largest amplitudes of the level scalegrams were found for a period of approximately 100 days in 2014 and 120 days in 2015 at both stations. The largest amplitudes of modelled level scalegrams were observed with a period of approximately 88 days in 2014 and 80 days in 2015 at both stations. The largest amplitudes of scalegrams for Antarctic Oscillation (AAO) were found for a period of 105 days in 2014 and 123 days in 2015. The corresponding correlation coefficients between observed sea level scalegrams and AAO for 2014 were 0.84 and 0.86, respectively, whereas for 2015 they were 0.87 and 0.90, respectively. It was concluded that the relationship between intraseasonal processes in the oce