Assimilation of Multiple-doppler Radar Data with the Wrf 3-d Var and a Cloud Analysis

A reliable application of a numerical weather prediction to short-range quantitative precipitation forecasting (QPF) is needed to forecast disastrous severe storms. The success of the dynamic, thermodynamic, and microphysical retrievals at the convective scale using Doppler radar observations are important but still great challenges. The use of three-dimensional variational data assimilation (3-D Var) is a suitable approach for the retrievals in a large domain. Sugimoto et al. (2005) demonstrated the performance of the WRF 3-D Var through observing system simulation experiments (OSSEs) and concluded that radar data assimilation with the WRF 3-D Var works reasonably well in recovering key features at scales larger than the convective scale. Improvement of large-scale forcing leads to positive impact on short-range QPF. The purpose of this study is to evaluate the performance of radar data assimilation with the WRF 3-D Var through an application of real radar data. A main concern is whether the performance evaluated is similar to the one obtained from an OSSE study, because an application to a real world suffers from issues related with data quality (i.e. clutters, aliases, and data density) and assumptions in the WRF 3-D Var (i.e. background error, and drop size distribution).