Applying a 3DVAR-EnKF Hybrid Data Assimilation System to Typhoon Forecast Models

A GSI (Grid-point Statistical Interpolation)-based 3DVAR-EnKF hybrid data assimilation system was developed for the Weather Research and Forecasting (WRF) model and applied to typhoon track forecast with the real data including satellite radiance. Our previous studies showed that the hybrid system outperformed the 3DVAR system in tropical cyclone track forecast. Recently this hybrid system is extended and applied to the GRAPES (Global/Regional Assimilation and Prediction System) model which was originally developed by the China Meteorological Administration (CMA). Techniques to deal with the different coordinates between the GRAPES background and GSI analysis will be mentioned and discussed. The one-observation test shows the success of the ensemble background error capturing the typhoon vortex circulation information. Real-data case studies for typhoon Fitow (2013) show that the hybrid system results in significant improvement in precipitation forecast as well as TC track forecast, compared to the original 3DVAR system. The flow-dependent ensemble covariance is found to be potentially able to adjust the position of the typhoon vortex, better update the vortex structure and environmental field and produce a dynamically balanced typhoon initial field.