Application of the Regression Algorithm to the Problem of Studying Horizontal Inhomogeneity of the Cloud Liquid Water Path by Ground-Based Microwave Measurements in the Angular Scanning Mode

The results of the cloud liquid water path (LWP) “land–sea” gradient retrieval from ground-based measurements of the downwelling microwave radiation near the Gulf of Finland coastline in the suburbs of Saint-Petersburg are presented. The measurements were carried out at the Department of Physics, St. Petersburg State University, by an RPG-HATPRO radiometer operating in the angular scanning mode. The inverse problem is solved by linear regression with the use of different statistical models of cloudiness for training the algorithm. Seven-year average values of the gradient of LWP for summer and winter have been obtained. The results demonstrate the presence of a positive “land–sea” gradient of LWP (larger values over the land and smaller values over the sea) in both periods, which qualitatively agrees with available satellite data.