【作者】 公颖； 王叶红； 赖安伟；

【Author】 GONG Ying , WANGYe-hong , LAI An-wei (.Wuhan institute of Heavy Rain , CMA , Wuhan 430074 ,China)

【机构】 中国气象局武汉暴雨研究所；

【摘要】 本文对AREM(包括业务和试验模式)、T213、JAPAN三个模式对于2008年5月到11月西南地区(25-34.5°N,97-110.5°E)的降水预报进行统计检验,并对三个模式对于2008年汛期两次强降水过程预报的效果做检验分析,以期得出各模式在西南地区的预报效果及特点,结果表明:(1)各模式在西南地区降水预报中均体现了较好的性能,同时各模式也各具特点,对小雨-大雨(以上)量级降水JAPAN模式预报情况较好,对暴雨(以上)量级T213、JAPAN、AREM都有较好的表现,而对于大暴雨(以上)量级AREM-SHIYAN模式表现出良好的预报效果。(2)三个模式对于西南地区降水走势的预报均较为可靠,相比之下T213降水预报偏强情况较多,AREM模式预报偏弱情况多,JAPAN模式预报相对适中。(3)在西南地区,降水陡增(实况降水量较前一天大一个量级以上且比前一天大5mm以上,下同)和其后强降水持续阶段,是各模式预报偏弱情况最可能发生的时期,同时也是各模式预报效果较差的时段,在此阶段,JAPAN模式预报可靠性相对较高,T213模式预报偏强可能性较高,AREM模式预报偏弱可能性极大,转折性天气的预报仍然是一个难点,而AREM模式对此表现更为突出。(4)三个模式在四川盆地西部边缘地区均有降水空报现象,AREM、JAPAN模式空报范围较小,T213空报范围较大,AREM模式在四川盆地西南部边缘地区的降水空报现象,可能源于两个原因,第一,模式对于中低层风场的预报偏强,第二,模式对于四川盆地处地形的处理与实况存在一定差别。(5)对于西南低涡这一较为特殊的天气系统,由于其生成地的特殊地形、其发生发展机制多种多样及尺度小、结构复杂等特征,当前模式要做到对其强度、位置和诱发的强降水做出较为准确的预报,仍然较为困难。更多还原

【Abstract】 The rain prediction effect of three models-AREM (including operation and testing version), T213, JAPAN in southwest (25-34.5 ° N ,97-110 .5 ° E) in May to November 2008 was statistically analyzed, and two heavy rain processes prediction was evaluated , in order to get the prediction character in Southwest China of every model. The results show that(:1)All of the three models exhibited preferable rain prediction effect in southwest area , and at the same time , they have specialties respectively, JAPAN predicted better for light to heavy rain , and T213 , JAPAN , AREM all had good effects in predicting torrential rain , and AREM-SHIYAN predicted better in large rain storm. (2) All of the three models predicted well for the patterns of precipitation in southwestern region, in contrast, T213 predicted heavier than observation, AREM predicted lighter than observation and JAPAN predicted relatively moderate. (3) In southwest region, in the period that rain increases suddenly (24-hour observation precipitation is over 5mm larger than that of yesterday) and the subsequent heavy rain persistence, models most likely predicted weaker than observation , and also most likely had poor rain prediction effect, at this stage, JAPAN prediction had relatively high reliability, T213 had strong possibility of heavier prediction, AREM had strong possibility of lighter prediction, the prediction for turning point of weather was still a difficult, AREM had evident performance for this. (4) All of the three models had false prediction in the southwest edge of the Sichuan Basin, the range of false prediction of AREM, JAPAN was relative small, but that of T213 was relative large, for the false prediction phenomenon of AREM in Sichuan Basin, perhaps it rooted two reasons: firstly, it predicted the low level wind stronger than observation, secondly, it’s treatment for the terrain in Sichuan Basin had some difference from the fact. (5) For the special weather system-southwest low vortex , because of the special terrain in its generation area, and its many kinds of occurrence and development mechanism, and the character of small scale, complex structure, it’s still hard today to make accurate prediction for its intensity, location and the heavy rain inducing by it.更多还原