2005年AREM模式汛期试验结果评估分析

AREM2．3模式2005年在中国气象局武汉暴雨研究所汛期试验过程中计算稳定，计算了该模式降水预报的分区域（长江中下游、华南、华北、东北、西南、湖北省）汛期（6、7、8三个月）TS评分，全国范围汛期漏报率、空报率、预报偏差，及各区域年内重要降水过程预报和实况的对比及TS评分，计算该模式形势场预报的平均误差、均方根误差、倾向相关系数、误差标准差，并与持续性预报（将前一天的分析场当作前一天对当天的预报场，以此类推）对比，以此对模式预报效果进行分析、评估，并做出总结，为模式的进一步开发和应用提供参考。分析结果表明：AREM2．3模式在2005年汛期试验期问48小时内预报稳定；对于长江中下游、华北、华南、东北、西南、湖北省以至全国范围均有较好的预报水平，但总体上对于强降水中心位置的预报情况不是很好；模式对于高空形势场也具有较好的预报能力，对500hPa位势高度的预报好于对500hPa温度的预报。

Evaluation and Analysis of AREM Prediction Products in Flood Season of 2005

The operation of AREM2.3 in Wuhan Institute of Heavy Rain, CMA, in the flood season of 2005 is steady. The TS scores of the precipitation forecast are calculated in various areas, such as, the middle and lower reaches of the Changjiang River, South China, North China, Northeast China, Southwest China and Hubei Province in flood season (June, July and August). The fail rate, false rate, predicting deviation of the whole China in flood season are calculated. The comparison between the forecast and real rain field is made. And the TS scores of important rainfall processes of those areas are calculated. The evaluation of the model prediction is made by comparing with the persistent forecast. It indicates that the prediction of AREM2.3 in the flood season of 2005 is steady in 48 hours, and AREM2.3 has relative high prediction ability for rain in the middle and lower reaches of the Changjiang River, South China, North China, Northeast China, Southwest China, Hubei Province and the whole China, but the prediction for heavy rain centers is not perfect, AREM2.3 has high prediction ability for meteorological elements in high-level, too, its predicting ability for 500hPa height is better than that for 500hPa temperature.