An Experiment for Assimilating Different Type of Data Observations in Forecasting Heavy Rainfall over Central Highlands Region Due to the Impact of Hurricane Damrey

Abstract

This study investigates and assesses the impact of assimilating data types (observed data surface, sounding, and satellite-derived atmospheric motion vectors – AMVs) for the Weather Research and Forecasting (WRF) in forecasting heavy rainfall over Central Highlands region, due to the impact of hurricane Damrey. The WRF model combined with the Gridpoint Statistical Interpolation (GSI) system, was started running at 12Z 03/11/2017, and 84h forecasts in advance, with two kinds for running assimilation "cold start" and "warm start", and with the three-dimensional variational data assimilation (3D-Var) method. The results showed that assimilated cases have improved forecasting about spatial distribution and amount of rainfall at a 24-hour lead time, in which, the "warm start" for better forecasting. Notably, the assimilation of AMVs data with the "warm start" run has improved forecasting quality of heavy rainfall, the POD, FAR, and CSI indicators are the best at the 24-hour lead time, for rainfall thresholds greater than 80mm.