摘要
对2018年3月—2019年6月ECMWF、GRAPES-GFS、JAPAN、GERMAN预报产品与对应相关格点值,采用相关系数法、均方根误差法、TS评分检验法进行检验。结果发现:四种模式中,降水预报均与本站实况有显著的相关性,但ECMWF模式降水产品24~48 h晴雨预报TS评分优于其他模式,均方根误差也小于其他模式,同时进一步发现各家模式季节性降水预报及0.1~9.9 mm预报具有较高一致性。
The correlation coefficient method,root mean square error method and TS score test method were used to test the ECMWF,GRAPES-GFS,JAPAN,GERMAN forecast products and their corresponding grid values from March 2018 to June 2019.The results showed that:among the four models,the precipitation forecast had significant correlation with the actual situation of the station,but the TS score of 24-48 h clear rain forecast of ECMWF model was better than that of other models with smallest mean square.At the same time,it is found that the seasonal precipitation forecast and 0.1-9.9 mm forecast of each model have high consistency.
作者
陈勇
王小虎
刘东
ChenYong;Wang Xiaohu;Liu Dong(Chongqing Qianjiang District Meteorological Bureau,Qianjiang 409000)
出处
《气象科技进展》
2021年第2期95-97,共3页
Advances in Meteorological Science and Technology
关键词
数值预报产品
武陵山区
降水性能检验
numerical forecast products
Wuling Mountain area
test of precipitation performance