Long-Term Variability of the Timing of Freezing and the Duration of Ice Phenomena in the White Sea Based on Satellite and In Situ Observations for 1980–2020

The long-term variability of the ice regime of the White Sea for 1980–2020 has been studied. The reliability of the satellite data was assessed by comparing them with in situ observation data. We use the regular hydrometeorological monitoring data from eight marine observation points, as well as satellite microwave passive sounding data (NSIDC) with a spatial resolution of 25 km and a time step of 1–2 days to form series of the main elements of the sea ice regime (the characteristic dates of ice regime and duration of ice phenomena). The average statistical dates of the start freezing and the ice breakup for the entire water area of the White Sea and its regions are obtained. Regression analysis of the data shows that the start of freezing and the ice breakup dates have shifted towards winter months over the past 40 years. The shifts occur at average rates of 7.4 days/10 years and 4.7 days/10 years, respectively, according to in situ observations, and 11.9 days/10 years and 4.1 days/10 years, respectively, according to satellite observations. Overall, over the past 40 years, the average duration of ice phenomena in the White Sea has decreased by 47 days according to in situ observations and by 62 days according to satellite observations. A comparative analysis of satellite and in situ data show significant differences in the average values of absolute deviations (up to 70 days) in determining the characteristic dates of the White Sea ice regime; however, the time series of characteristic dates are in good agreement with each other (pair correlation coefficients of 0.76/0.82 between the time series of dates of start freezing and dates of ice break up). This proves the possibility of using satellite data to calculate the regime indicators of ice phenomena over a long period in order to identify patterns in the development of ice processes, assess their climatic trends, and develop methods for forecasting ice conditions.