NUMERICAL PREDICTION OF AN EXTREME RAINSTORM OVER THE PEARL RIVER DELTA REGION ON 7 MAY 2017 BASED ON WRF-ENKF

Abstract

An ensemble Kalman filter based on the Weather Research and Forecasting Model (WRF-EnKF) is used to explore the effectiveness of the assimilation of surface observation data in an extreme local rainstorm over the Pearl River Delta region on 7 May 2017. Before the occurrence of rainstorm, the signals of weather forecasts in this case are too weak to be predicted by numerical weather model, but the surface temperature over the urban area are high. The results of this study show that the wind field, temperature, and water vapor are obviously adjusted by assimilating surface data of 10-m wind, 2-m temperature, and 2-m water vapor mixing ratio at 2300 BST 6 May, especially below the height of 2 km. The southerly wind over the Pearl River Delta region is enhanced, and the convergence of wind over the northern Guangzhou city is also enhanced. Additionally, temperature, water vapor mixing ratio and pseudoequivalent potential temperature are obviously increased over the urban region, providing favorable conditions for the occurrence of heavy precipitation. After assimilation, the predictions of 12-h rainfall amount, temperature, and relative humidity are significantly improved, and the rainfall intensity and distribution in this case can be successfully reproduced. Moreover, sensitivity tests suggest that the assimilation of 2-m temperature is the key to predict this extreme rainfall and just assimilating data of surface wind or water vapor is not workable, implying that urban heat island effect may be an important factor in this extreme rainstorm.