Evaluating Predictability of Severe Convection using a WRF Ensemble Kalman Filter

While many advances have been made in Numerical Weather Prediction (NWP) over the past few decades, NWP models still often struggle with the initiation and progression of severe convective events. High resolution models are becoming more proficient in accurately predicting severe convection, but more progress can be made. The use of an ensemble Kalman filter in NWP models is becoming an increasingly popular method of data assimilation. The main goal of this study is to implement an ensemble Kalman filter in the Weather Research and Forecasting (WRF) model to evaluate its usefulness in predicting convective initiation and the downstream progression of severe convective events. Attention will be given to the mean of the ensemble members; specifically, how useful the mean is, when it loses its usefulness in making a deterministic forecast, and how the best member should be determined if strong storm-scale nonlinearity forces the mean off the model attractor. Another area explored here is the evaluation of the ensemble when different sets of physics schemes are employed among the ensemble members. In particular, resulting forecast probabilities will be examined to understand the role of physics uncertainty in the regional-scale prediction of severe convection and convective initiation. Preliminary results and future research goals will be discussed.