Prediction Ability Assessment on SWCWARMS Model Using Different Starting Times during Case“18·05·21”

Base on the prediction results of southwest regional meso-scale business model for the first regional heavy rain weather process in Sichuan on 21 May 2018, to comparative analysis of forecast results by different starting times,found that the more near to get the forecast effect is better. The results show that: the description of the atmospheric state is more realistic when the observed data and reanalysis?data are used before the occurrence of precipitation. On the one hand,the stratification state can be changed by improving the temperature and humidity structure of the atmosphere to affect its stability; on the other hand,the cyclonic convergence and moisture transport in the lower layer can be enhanced by improving its circulation field,thus affecting the whole precipitation process from a dynamic perspective. The more realistic initial field starting model can more accurately simulate the thermodynamic process of energy/water vapor accumulation and release before and after precipitation,as well as the coordinated development process of dynamic factors such as vorticity,divergence and vertical rising motion,so it has a positive impact on the prediction effect.