ИСТИНА

The results of the Russian numerical weather prediction COSMO-Ru system were used for evaluating the effects of aerosol and cloud optical and microphysical parameters on solar radiation in both clear and overcast situations over European continental regions at the Moscow State University Meteorological Observatory (MSU MO, Russia), and Lindenberg Meteorological Observatory (Germany), over desert/semi-desert areas (Israel) and over the Arctic district at the international Hydrometeorological Observatory of Tiksi (Russia). Similar approach was used at all sites for studying the effects of aerosol on radiative processes. For evaluating pure aerosol effects we chose the days without cloudiness in different seasons and analyzed solar irradiance estimated with different aerosol climatologies (Tegen et al., 1997, Kinne et al., 2013, Tanre et al., 1984) and no aerosol input. In addition, using the accurate CLIRAD-SW radiative code we simulated the radiative feedback due to different climatologies and AERONET input aerosol data over these sites during a year . In the analysis we specified the situations with high surface albedo to estimate its effect on solar fluxes at different locations. Temperature effects were studied as well. For comparisons of cloud parameters we used Cloudnet data (Illingworth et al, 2007) for the Lindenberg Observatory as well as some other additional data. Model experiments were performed with a new (Blahak et al., 2016) and a default cloud radiation algorithm considering different aerosol climatologies as well as aerosol-free situations for clear sky and overcast conditions. We analyzed several cloud parameters such as liquid and ice water content, effective radius of droplets, location of cloud layers, and others obtained for both cloud radiation algorithms and compared them with the Cloudnet retrievals. For solar radiation fluxes the comparisons were made against the best BSRN type radiative observations. We also studied temperature effects due to the application of the new cloud-radiation algorithm.