摘要
利用我国自主研发的多尺度通用的新一代同化与数值预报系统——GRAPEs模式,通过对4种积云参数化方案的比较,确定集合预报成员,对山西省夏季3类典型强降水过程进行了集合预报试验。试验结果表明:(1)对所模拟的个例,Betts-Miler-Janjic对流调整方案和Kain—Fritsch对流参数化方案优于对流调整方案和郭氏参数化方案;(2)积云参数化方案对降水的影响比边界层参数化方案对降水的影响大,不同个例对2种不同积云参数化方案和3种不同边界层参数化方案的6种组合的反应敏感程度也各不相同;(3)不同集合成员对降水的预报结果各不相同,其预报偏差也各不相同;(4)对于不同的天气系统,不同的方案其预报效果是不同的;(5)通过集合平均,降水预报偏差明显减小,降水预报偏大的情况得到改善,且其24~36h预报效果最好。
Using GRAPES model, by comparing four parameterization schemes, deciding ensemble members, and ensemble simulation of three typical heavy rain in summer in Shanxi province has been given. All results show that : ( 1 ) For these cases, Betts-Miler-Janjic scheme and Kain-Fritsch scheme are advantage over general convective adjustment scheme and Guo scheme. (2) The influence of cumulus parameterization scheme is larger than the influence of the PBL parameterization scheme on the precipitation. For cumulus parameterization scheme and the PBL parameterization scheme, the sensitive degree of different cases are different. (3) Different ensemble members have different forecasting results and different bias. (4) For different weather systems, the forecasting effects of different schemes are different. (5) After ensemble averaging, the bias of precipitation forecasting become smaller, and the condition of larger bias is improved. The forecasting effects for 24 h--36 h are the best.
出处
《气象科学》
CSCD
北大核心
2008年第B12期8-14,共7页
Journal of the Meteorological Sciences
基金
山西省重点科研项目(0509)
山西省科技厅攻关项目(051116)
关键词
强降水
集合预报
试验
Heavy rain Ensemble forecasting Simulation