A SENSITIVITY STUDY OF THE NON-HYDROSTATIC REGIONAL CLIMATE MODEL REGCM-4.7.1 TO PHYSICAL PARAMETRIZATION SCHEMES OVER THE BALKAN PENINSULA AND BULGARIA

The aim of this study is to present preliminary results of testing the new non-hydrostatic dynamic core of the regional climate model RegCM4.7.1 by varying different parametrization schemes over the Balkan Peninsula and Bulgaria. A one-way double nesting technique was applied, employing a mother domain with 15 km grid spacing and a nested domain with 3 km grid spacing. Twenty test simulations were performed with a model resolution of 15 km for the year 2000 on a Lambert Conformal Conic projection and 12 test simulations with a model resolution of 3 km for one monthJanuary 2000. Two planetary boundary layer schemes (UW PBL and Holtslag), four cumulus schemes (Emanuel, Grell, Tiedke and Kain-Fritsch), three explicit moisture schemes (Nogherotto-Tompkins, WSM5 and SUBEX) and two shallow convection configurations (MM5 and Tiedke) were tested. We used ERA-Interim reanalysis as boundary conditions for the 15 km runs with 0.75-degree horizontal resolution. The results are presented for several model output parameters (precipitation, temperature, surface pressure and cloud fraction cover). The output is compared with E-OBS observational dataset, satellite-based observations and ERA-Interim reanalysis. Based on the results, the model is sensitive to the choice of different parametrization schemes. The 15 km simulations overestimate the precipitation and underestimate the temperature in all scheme configurations. The biggest biases in precipitation and surface pressure we found in simulations with combinations with Nogherotto-Tompkins moisture scheme. In the case of 3 km simulations, UW PBL scheme with WSM5 moisture configurations and Nogherotto-Tompkins configurations gave the largest biases for the studied territory.