重庆风暴尺度集合预报系统降水预报性能分析

Prediction Performance Analysis of Chongqing Storm-Scale Ensemble Prediction System

本文利用重庆风暴尺度集合预报系统业务存档的2017年8月1~31日逐日08时起报的模式预报资料及相应的观测资料对该系统的降水预报进行了检验评估及综合分析,综合各种检验结果,总体而言:该系统集合平均和概率匹配平均等集合预报产品相对于控制预报表现出了较明显的优势;小雨和中雨量级降水集合平均的TS评分优于概率匹配平均;大雨和暴雨量级降水概率匹配平均的TS评分优于集合平均;各预报时效的Talagrand分布均表现出实况落在第11个概率区间的概率明显高于其他概率区间,需要在今后的科研和业务中加以关注;Outlier评分介于0.12~0.26;降水概率预报检验方面,各个降水量级的预报失误概率Brier评分和相对作用特征技巧评分AROC均较为理想。总体而言,该系统在降水预报方面相对于单一的确定性预报而言表现出了一定的优势。

In this paper, the prediction performance of the Chongqing Store1 - scale Ensemble Prediction System is comprehensively analyzed by using the forecast data and corresponding observation data started at 8am From August 1, 2017 to August 31 st. Result show that： in the aspect of Precipitation forecast. The ensemble mean forecast and the probability matching mean forecast have obvious advantages over the control forecast. The TS score of light rain and moderate rain precipitation of ensemble mean is better than that of probability matching mean. The TS score of rainfall and rainstorm precipitation of probability matching mean is better than that of ensemble mean. The talagrand distribution of each torecast time show that the probability of falling in eleventh probability intervals is obviously higher than that of other probability intervals, it need to pay attention to future research and operation. The outlier score is between 0. 12 and 0.26. In the aspect of precipitation probability forecast, the brier score and AROC of each precipitation magnitude are impressive. In general, the system shows a certain advantage over the single deterministic forecast in the precipitation torecast.