Application of radar data assimilation on convective precipitation forecasts based on water vapor retrieval

Abstract Based on a short-time heavy rainfall in Anhui and the weather research forecasting (WRF) model, water vapor initial field of model is retrieved using statistical relationships reflectivity factor from Doppler radar with relative humidity hydrometeor. Three-dimensional variational (3DVAR) assimilation method used to assimilate radial velocity, then impact assimilating analysis forecast torrential rain assessed. The results show that, after vapor, significantly improved. content middle layer analyzed increased, corresponding well convective region. Meanwhile, precipitation distribution during this process successfully simulated. mesoscale characteristics are better presented by imageries factor, false echoes partially reduced. Besides, prediction regions closer actual observations. After simulated one-hour accumulated observation, fraction skill score (FSS) higher.