APPLICATION OF RADAR DATA ASSIMILATION IN HEAVY RAINFALL FORECASTING IN HO CHI MINH CITY AREA

Abstract

This study evaluated the ability to simulate 15 heavy rainfall days in 2018 in Ho Chi Minh city of the WRF model when assimilating Nha Be radar’s data. The impacts of initial assimilating process on the analysis field is investigated through three assimilation modes including: cold start, warm start and cycling. Results show that reflectivity in the cold start mode is identical with the case of no data assimilation after one hour of integration. Under the warm start mode, the discrepancy in the analysis field lasted longer which highlighted the role of background field from the previous forecast step. In comparison with precipitation observation at 11 meteorological stations, forecast results present the improvement of FBI, POD, CSI indices in the cycling mode. This can be seen not only at three thresfold of 1,5 and 10mm but also at leadtimes of 6h and 12h.