###
气象:2019,45(7):893-907
←前一篇   |   后一篇→
本文二维码信息
码上扫一扫!
集合预报产品在江苏省暴雨预报中的应用评估
陈圣劼,刘梅,张涵斌,俞剑蔚,陈超辉
(中国气象局交通气象重点实验室,南京 210008; 江苏省气象台,南京 210008; 北京城市气象研究院,北京 100089; 国防科技大学气象海洋学院,南京 211101)
Evaluation on Forecasting Heavy Rainfall over Jiangsu Region Using Ensemble Forecast Techniques and Products
CHEN Shengjie,LIU Mei,ZHANG Hanbin,YU Jianwei,CHEN Chaohui
(Key Laboratory of Transportation Meteorology, CMA, Nanjing 210008; Jiangsu Meteorological Observatory, Nanjing 210008; Institute of Urban Meteorology, CMA, Beijing 100089; Institute of Meteorology and Oceanography, National University of Defense Technology, Nanjing 211101)
摘要
图/表
参考文献
相似文献
本文已被:浏览 1350次   下载 2655
投稿时间:2018-08-23    修订日期:2019-03-11
中文摘要: 利用2011—2015年6—8月TIGGE(THORPEX Interactive Grand Global Ensemble)数据集中欧洲中期天气预报中心(ECMWF,以下简称EC)的集合降水预报数据和江苏省70个基本站逐日24 h(20时至次日20时)降水数据,通过大量暴雨样本系统检验和评估了EC集合预报及多种后处理释用产品对江苏暴雨的预报能力。结果表明:作为集合预报的初级产品,集合平均对暴雨的预报存在明显的漏报率,TS预报评分尚不及EC确定性预报;集合预报不同成员间对暴雨的预报技巧差异大,其最优成员组合的预报能力显著优于EC确定性预报,表明集合预报具有较大的应用潜力;在多种集合预报后处理释用技术中,最大值、最优百分位、降水偏差订正频率匹配法、概率预报、集合异常预报法和杜 周排序法(最大值法)的平均TS评分均较高,超过10%,其次90%分位数、融合、融合 概率匹配和杜 周排序法(集合平均或中位值法)的预报效果也均优于EC确定性预报。集合中位值、概率匹配方法对江苏暴雨的预报评分低于集合平均预报,在暴雨预报上的参考价值相对较低。该评估结果进一步加深了对各集合预报产品区域暴雨预报能力的认识,为预报员更直接快速地选取有效的集合预报产品提供参考。
Abstract:Using global ensemble forecast of European Centre for Medium Range Weather Forecasts (ECMWF) from THORPEX Interactive Grand Global Ensemble (TIGGE) data, and the 24 h observed daily accumulated precipitation data from 20:00 BT to the next 20:00 BT, the applications of the EC ensemble forecast and multiple EC ensemble products and various forecast products of heavy rainfalls by post process techniques over Jiangsu Region are evaluated. The results show that the ensemble mean forecast exhibits a high missing forecast rate, and the threat score (TS) is lower than that of EC deterministic forecast. Distinct differences in the forecast skills of each member of EC members and greatly higher TS of the synthesis of the optimal forecasts indicate that the EC ensemble forecast holds great potential application for forecasting heavy rainfall. Multiple post process products of ensemble forecast have different performances. Maximum value, the optimal percentage, the frequency matched based on precipitation forecast bias correction, probabilistic forecasts, ensemble anomaly forecast approach and Du Zhou performance ranking method based on the maximum all have higher TS by more than 10%. Then, the TSs of 90% percentage, fusion, the fusion probability matching and the Du Zhou performance ranking method based on the mean or median values are all higher than the EC deterministic forecast. However, the median value and probability matching of EC ensemble forecast are less skillful than ensemble mean with small values in reference. The results of this evaluation may enhance the cognitions of ensemble forecast products and kinds of post process techniques, providing forecasters with a useful reference of the utilization of ensemble forecast products for the forecasting of heavy rainfall.
文章编号:     中图分类号:P456    文献标志码:
基金项目:江苏省自然科学基金面上项目(BK20161604)、南京大气科学联合研究中心面上项目(NJCAR2016MS02)和江苏省气象科研基金重点项目(KZ201601)共同资助
引用文本:
陈圣劼,刘梅,张涵斌,俞剑蔚,陈超辉,2019.集合预报产品在江苏省暴雨预报中的应用评估[J].气象,45(7):893-907.
CHEN Shengjie,LIU Mei,ZHANG Hanbin,YU Jianwei,CHEN Chaohui,2019.Evaluation on Forecasting Heavy Rainfall over Jiangsu Region Using Ensemble Forecast Techniques and Products[J].Meteor Mon,45(7):893-907.