摘要
文章对 1 998年大洪水期间改进的 HL AFS[1]的降雨预报产品进行检验 .结果表明 ,改进的HLAFS对松嫩流域的暴雨预报较好 ,中到大雨则是长江中下游地区预报较好 ,而三峡库区相对较差 ;对于主要降雨过程降雨区的强度预报往往比实况弱 ,且其落区约有半数较实况偏北 ,但其轴向一般与实况一致 ;改进的 HLAFS对小雨预报比原 HLAFS好 ,对中雨和暴雨预报也有不同程度的改进 ;整体预报面积偏大和大雨以上降水预报准确率低仍是 HL AFS有待改进的问题 .对三峡库区面降水量预报 ,半数左右的预报较好 ;对大量级降水预报往往偏弱 。
The rainfall prediction products of improved HLAFS were verified for the catastrophic flood period of 1998. The results show that this model performed better in Songhuajiang Nenjiang River Valley for torrential rain forecast, better in the middle and lower reaches of the Yangtze River and worse in the Three Gorges area for the moderate and heavy rain forecasts. Usually, The predicted rainfall intensity is weaker than the observed for the main rainfall processes. In about half causes, the predicted rainfall area is to the north of the real while the predicted axis is consistent with the real. The improved HLAFS performs better for light rainfall prediction and shows improvement for moderate and torrential rainfall. However, the overall predicted rainfall area is too large and the accuracy of heavy and torrential rain prediction needs further improving. For the rainfall amount prediction around the Three Gorges area, in about half causes, the model performs well with less predicted amount for heavy rain and larger one for small rain.
出处
《应用气象学报》
CSCD
北大核心
2001年第1期118-122,共5页
Journal of Applied Meteorological Science