根据 Anderson 亲爱原则,为极其重的降雨(缩短的同样极端的降雨 /降水) 的预报的一个方法基于中国气象学的管理(CMA ) 的 T213 全球整体预言系统(EPS ) 的整体预报数据被开发。在 20072010 期间使用 T213 预报降水数据并且在 20012010 的 JuneAugust 的观察降雨数据,累积分发功能(CDF ) 的特征观察, T213 EPS 预报降水被分析。因此,根据在模型之间的 CDF 的连续差别,气候和 EPS 预报,极端降水预报索引(EPFI ) 的一个数学模型被建立并且适用在 1731 2011 年 7 月期间在中国预报几个极端降雨事件的实验。结果证明 EPFI 利用了模型的尾巴信息气候的 CDF 和极端降雨的提供的令人喜悦的的预报。EPFI 为预先发给极端降雨的早警告 37 天基于 T213 EPS 是有用的。与预报铅时间的扩展, EPFI 变得不太熟练。结果也证明模型气候 CDF 的合理性具有到 EPFI 的技巧的重要重要性。
According to the Anderson-Darling principle, a method for forecast of extremely heavy rainfall (abbre- viated as extreme rainfall/precipitation) was developed based on the ensemble forecast data of the T213 global ensemble prediction system (EPS) of the China Meteorological Administration (CMA). Using the T213 forecast precipitation data during 2007-2010 and the observed rainfall data in June-August of 2001 2010, characteristics of the cumulative distribution functions (CDFs) of the observed and the T213 EPS forecast precipitation were analyzed. Accordingly, in the light of the continuous differences of the CDFs between model climate and EPS forecasts, a mathematical model of Extreme Precipitation Forecast Index (EPFI) was established and applied to forecast experiments of several extreme rainfall events in China during 17-31 July 2011. The results show that the EPFI has taken advantage of the tail information of the model climatic CDF and provided agreeable forecasts of extreme rainfalls. The EPFI based on the T213 EPS is useful for issuing early warnings of extreme rainfalls 3 7 days in advance. With extension of the forecast lead time, the EPFI becomes less skillful. The results also demonstrate that the rationality of the model climate CDF was of vital importance to the skill of EPFI.