滁河流域汛期多模式集成面雨量预报检验评估

对2013—2016年汛期ECMWF、JMA及中尺度WRF模式的预报结果进行检验,基于合理的时间尺度,制作了动态权重集成面雨量预报(DWI)、分步集成面雨量预报(SI)和等权重集成面雨量预报(EWI)产品,并对2017—2020年汛期降水期间多种集成面雨量预报产品和ECMWF、JMA、WRF的单个模式面雨量预报产品进行对比评估.结果表明:3种集成面雨量预报效果整体上优于单一数值预报模式,尤其是预报致灾严重的暴雨等级降水优势明显;DWI和WRF对强降水的面雨量预报正确率最高,其他2种集成面雨量预报结果次之,ECMWF及JMA较差;模式集成预报弱降水过程的优势不明显.

Verification and evaluation of multi model integrated forcasting of areal rainfall for Chuhe River Basin in flood period

Based on the precipitation forcasting verification results of ECMWF, JMA and mesoscale WRF models in flood season from 2013 to 2016, dynamic weight integrated areal rainfall forecast (DWI), stepwise integrated areal rainfall forecast (SI) and equal weight integrated areal rainfall forecast (EWI) products were made by reasonable time scales. Comparative verification and evaluation were carried out on multiple integrated areal rainfall forecast products and single model areal rainfall forecast products based on the precipitation in flood season from 2017 to 2020. The results showed that the areal rainfall forecast of the three integrated models was better than single model, especially in the case of severe rainstorms. Comparing each product according to the forecast of heavy precipitation and weak precipitation, the DWI integrated product and WRF model product had the highest accuracy in heavy precipitation areal rainfall forecast, followed by the other two integrated products, and the ECMWF and JMA model products were poor. For weak precipitation process, the advantage of model integrated area rainfall forecasting was not obvious.