In order to check the precipitation prediction ability of T213 and AREM,precipitation prediction of T213 and AREM from 2004 to 2005 and grading TS score verification of 117 stations daily precipitation in Henan province are made.
为检验T213与AREM模式的降水预报效果,将2004—2005连续两年两个模式的预报结果与河南省117站逐日降水实况进行分级TS评分检验,结果发现:两个模式对小雨以上量级的预报能力基本相当,夏秋季节的预报效果好于冬春季节;中雨以上量级预报均无季节特征;大雨以上量级的预报评分夏季T213高于AREM模式;大暴雨预报AREM模式好于T213。
By contrasting the TS score of AREM from June to August in 2007 for every forecasting period and every forecasting area,contrasting the TS score among three models of AREM,T213 and JAPAN,and analyzing the prediction effect of AREM for important rainfall processes in the flood season of 2007,the prediction effect and characteristic of AREM in flood reason in 2007 was obtained.
过对2007年6～8月AREM模式降水预报做不同时效、不同区域的TS评分对比,比较AREM与T213、JAPAN三个模式降水预报TS评分,并对AREM模式2007年汛期主要降水过程预报效果进行检验分析,从而获得AREM模式2007年汛期降水预报效果和特点,结果表明:(1)从AREM模式不同时效降水预报TS评分对比可知,对长江中下游区域,AREM模式12～36 h预报效果好于0～24 h预报,24～48 h效果相对较差,对华南、华北、东北、西南东部区域的降水,AREM模式预报效果均随时效延长而减弱。
Especially when forecasting light rain,the TS score averaged over China using probability regression method is higher than that of regression precipitation categorical forecast,the false alarm ratio is obviously smaller,and also the forecast bias is closer to 1.
该文对比分析概率回归降水等级预报和回归降水等级预报的差异,2007年秋季至2008年夏季全国平均检验结果表明:概率回归降水等级预报效果好于回归降水等级预报,尤其是小雨预报,TS评分明显高于回归降水等级预报,同预报偏差过大情况也有很大改善。