不同对流参数化方案对长江中下游梅雨锋暴雨降水预报的评估

Assessment of Precipitation Prediction using Different Cumulus Parameterization Schemes in MeiYu Front Rainfall

选用不同的对流参数化方案对近几年典型的几次长江中下游梅雨锋暴雨过程进行数值模拟,利用量化的降水评估方法检验MM5模式中不同对流参数化方案对梅雨锋暴雨降水预报的能力,一方面作为实际预报的参考,另一方面也作为模式改进的依据,评估结果显示模式预报的降水往往偏大,对不同时段和不同量级的降水,没有普遍适用的对流参数化方案,可以将经验预报和数值模拟相结合。对暴雨中心的预报,认为采用Grell方案和Kuo方案较好,但时间上有滞后性。

Base on the mesoscale model MMS, we have simulated several MeiYu front heavy rainfall scenarias and assessed the model precipitation prediction capability with different cumulus parameterization schemes by using the precipitation evaluate methods. The result shows that the model compute outcome commonly overprediction. For different precipitation, there is no prevalent cumulus parameterization scheme, but we can combine the experience and the numerical simulation prediction. Grell and Kuo parameterization schemes are better to predict the heavy rainfall center than other schemes.