摘要
基于站点观测资料和欧洲细网格(ECMWF)、日本高分辨率(JMA)及多模式客观集成预报技术(OCF)模式预报资料,以2015年2-7月为例,评估这三种模式对浙江省6个水库流域客观面雨量预报效果。水库流域面雨量预报值采用网格算术平均法计算,面雨量观测值采用泰森多边形法计算。检验评分方法采用平均绝对误差、TS评分(Threat Score)和模糊评分法。初步检验结果表明:1)ECMWF对大雨暴雨的预报效果优于其他模式的,OCF对中雨的预报效果优于其他模式的,对小雨预报,三种模式效果相当。2)各模式预报效果均随着降水等级增大而下降。小雨时,空报率远高于漏报率,其中ECMWF和OCF漏报率不到10%;大雨暴雨时,漏报率远高于空报率,其中JMA漏报率达75%~90%。3)三种模式对浙江中部北部地区水库流域面雨量预报效果整体优于对浙江南部地区水库的,且随着降水等级升高,南北水库之间预报效果差异增大。4)ECMWF对大雨暴雨的预报效果随着时效临近表现为明显改善,OCF和JMA在72 h预报时效以外,预报效果随时效临近呈改善趋势,在72 h时效以内,预报效果不稳定。
Based on the observation data from conventional weather stations and the rainfall forecasts respectively obtained from three numerical weather prediction models of European Centre for MediumRange Weather Forecasts( ECMWF),Japan Meteorological Agency( JMA) and Objective Consensus Forecasting( OCF) during the period of February to July 2015,we evaluate the forecasting performances of the three numerical models mentioned above for the objective areal rainfall of six reservoir basins in Zhejiang province. The areal rainfall forecast of the six reservoir basins is calculated by the grid arithmetic average method,while the observed areal rainfall is calculated by Thiessen polygon method. Moreover,the scoring methods used to evaluate the performance include Mean Absolute Error( Ea),Threat Score( TS)and fuzzy comprehensive score. The evaluation results are as follows. 1) ECMWF performs better than other models for heavy and torrential rain,OCF performs better for moderate rain,and they are about the same for light rain. 2) The performances of the three models all decrease with the increasing grade of precipitation. The false alarm rates are much higher than the miss alarm rates for light rain,the miss alarm rates of ECMWF and OCF are lower than 10%,while the miss alarm rates are much higher than the false alarm rates for heavy and torrential rain,the miss alarm rates of JMA is up to 75% —90%. 3) The performances of the three models on the reservoir basins in northern and central area of Zhejiang are better than that in southern area,and they show a distinct increasing trend with the increasing grade of precipitation. 4) The performance of ECMWF for heavy and torrential rain shows a significant improving trend with forecast aging drawing near,while the performances of OCF and JMA show improving trend outside the 72 h forecast aging,but not stable inside the 72 h.
出处
《气象与环境科学》
2017年第3期99-106,共8页
Meteorological and Environmental Sciences
基金
气象灾害教育部重点实验室(南京信息工程大学)开放课题(KLME1407)
浙江省气象局重点项目(2017YB06)资助