OSSE Experiments for Stormscale Radar Data Assimilation with an Ensemble of 3DVAR System

In this study, an ensemble of data assimilation method named as En3DA has been developed which consists of an ensemble of 3DEnVAR assimilations and forecasts which differ from each other member by perturbing initial conditions and/or perturbing observations. The method is applied to the assimilation of simulated radar data for a supercell storm. It is shown that the flow-dependent ensemble covariances derived from the En3DA method is effective in producing quality analysis. Most important features of the simulated storm including low-level cold pool and mid-level mesocyclone are well analyzed. Several groups of sensitivity experiments are conducted to test the robust of the method. The first group of sensitivity experiments demonstrates that incorporating mass continuity equation as a weak constraint into the En3DA algorithm can improve the quality of the analysis when radial velocity observations contain large errors. In the second group of experiments, we examine the sensitivity of the analysis to microphysics scheme. It is found that the analysis results are quite sensitivity to different microphysics schemes. For this reason, ensemble forecasts with multiple microphysics schemes could be used to reduce uncertainty related to model physics errors. Four sensitivity experiments show that assimilating reflectivity can reduce spin-up time, but has a small positive impact on the quality of analysis for wind fields. The experiment with the threshold 15 dBZ provides the best analysis.