Abstract
Recent studies show that data assimilation improves the efficiency of weather forecast models, however, it is not properly quantified the impacts of radar data assimilation related to the physical model parameterizations, especially the microphysics. The goal of this study was to study the impact of the use of radar data with different microphysics parameterizations of the Weather Research and Forecasting (WRF) model with its data assimilation system (WRFDA-3DVAR) for cases of intense precipitation. The study area covers the South-western Brazil and Southeastern Paraguay. The simulations were done for three cases in 2014. The comparison is performed through the statistical metrics Fractional Skill Score (FSS) and Local Root Mean Square Error (LRMSE). Different microphysics parameterizations were tested when assimilating conventional and radar data for three events. Thus, we evaluated nine microphysical parameterizations in order to determine which one provides the most realistic short-term forecasts of meteorological fields over the radar coverage, as well as the relative impact of different microphysical parameterization and the assimilation of conventional and radar data. The positive impact of the radar data assimilation was in the average up to 20% in the FSS, while the positive impact among the microphysics options reached 70% in the FSS.
Keywords:
microphysics parameterization; radar data assimilation; WRF and WRFDA
