Assimilation of multiple-Doppler radar data with WRF-3DVAR system: Preliminary results in observing system simulation experiments

The objective of this study is to develop a method for assimilating radial velocity and reflectivity factor observed by Doppler radar and to investigate the performance of this method with simulated observations. Development of this method is based on three-dimensional variational data assimilation (3DVAR) designed for the Weather and Research Forecasting (WRF) model. Operators for both types of observation are incorporated into WRF 3DVAR. Especially for reflectivity factor assimilation, the increments of water vapor, cloud water, rain water, and temperature are balanced within the framework of warm rain processes. We investigate the performance of radar data assimilation via Observing System Simulation Experiments (OSSEs). In our OSSE, two fields are prepared from results of simulation with the WRF model. Multiple-radar observations are simulated from the one of simulation, while a background field is prepared from another simulation. Results indicate that WRF 3DVAR works reasonably well for retrieving key features of dynamical and microphysical fields in a scale larger than the convective scale. Even if only radial velocities are assimilated, positive impacts are found in precipitation forecasting, but they are limited during a initiation stage of convection. Reflectivity factor assimilation becomes crucially important during developing or mature stages of convection.