Renmin Zhujiang (Jan 2023)
Sensitivity Study of WRF Parameterization Schemes and Initial Fields on Rainstorm Simulation in Minjiang River Basin
Abstract
As one of the most dominant weather forecasting tools,numerical models have been widely employed for early warning and prediction of strong weather processes such as extreme rainfalls.In practical forecasting operations,numerical models are affected by initial fields,boundary conditions,and parameterization schemes,which makes it difficult to conduct long-time accurate forecasts.Based on the new-generation mesoscale numerical weather forecasting model,i.e.,weather research and forecasting (WRF) model,16 sets of microphysical processes and cumulus convection parameterization schemes are cross-combined to simulate the sensitivity of a continuous rainstorm process in the Minjiang River Basin from May 17 to 24,2021.The results show that the WRF model can reproduce the spatio-temporal distribution characteristics of this rainfall process,but it significantly overestimates the intensity and extent of the storm center,with the overall better performance of EXP9 (Lin-NT combination).For the mass sensitivity test of initial fields,the simulation results based on the optimal combination of parameterization schemes reveal that the fifth-generation atmospheric reanalysis of the European Center for Medium-Range Weather Forecasts (ERA5) has a sound effect for all magnitudes of rainfalls.Meanwhile,it best matches the spatio-temporal characteristics of the observed data,followed by global data assimilation system/final operational global analysis (GDAS/FNL),while climate forecast system operational analysis (CFS) is the worst.This study can provide methodological references for improving the accuracy of extreme rainfall simulation and forecasting.
