The article presents the results of verification and application of the proposed automatic algorithm of upwelling detection and presents the results of its work based on the CMEMS Baltic Sea Physical Reanalysis product in the southeastern part of the Baltic Sea for 2010-2019. The algorithm was verified by using the data of observations on the offshore ice-resistant stationary platform D-6, ship data obtained in the 127th cruise of the research vessel “Professor Shtokman”, and the results of upwelling detection based on the sea surface temperature derived from measurements of the MODIS Terra/Aqua spectroradiometer. It is shown that the lowest frequency of upwellings is observed in August-September (5-6 days per year), the highest - in May - June and October (11-15 days per year). In the period 2010-2013 in the study area, on average, up to 10 % of negative thermal anomalies were observed during the warm period of the year. Since 2014 (with the exception of 2017), an increase in the frequency of upwellings has been noted - on average, about 20 % of days were days with upwelling. It is shown that the winter East Atlantic/West Russia and East Atlantic indices and the spring Scandinavia pattern index can be used to assess the characteristics of the future summer upwelling.