Constraining a 3D-Var Radar Data Assimilation System with Large-scale Analysis to Improve Short-range Precipitation Forecast

It is known from previous studies that radar data assimilation can improve the short-range forecast of precipitation, mainly when radial wind and reflectivity are available. However, from our experience the radar data assimilation, when using the 3D-Var technique, can produce spurious precipitation and large errors on the position and amount of precipitation. One possible reason for the problem is attributed to the lack of proper balance in the dynamical and microphysical fields. This work attempts to minimize this problem by adding a large-scale analysis constraint in the cost function. The large-scale analysis constraint is defined by the departure of the high resolution 3D-Var analysis from a coarser resolution large-scale analysis. It is found that this constraint is able to guide the assimilation process in such a way that the final result still maintains the large-scale pattern, while adding the convective characteristics where radar data are available. As a result, the 3D-Var analysis with the constraint is more accurate when verified against an independent dataset. The performance of this new constraint on improving precipitation forecast is tested using six convective cases and verified against radar-derived precipitation by four skill indices. All skill indices show improved forecast when using the methodology presented in this study.