ECMWF降水极端天气指数在安徽省的应用评估

详细信息   下载全文 | 推荐本文 |

英文篇名：Assessment and applications of precipitation Extreme Forecast Index(EFI)from ECMWF to Anhui province 作者：朱鹏飞 ; 邱学兴 ; 王东勇 ; 朱红芳 英文作者：ZHU Pengfei ; QIU Xuexing ; WANG Dongyong ; ZHU Hongfang ; Anhui Meteorological Observatory 关键词：极端天气指数 ; 暴雨 ; 气候距平 ; 集合预报 英文关键词：Extreme Forecast Index(EFI) ; heavy rainfall ; climate anomaly ; ensemble prediction 刊名：暴雨灾害 英文刊名：Torrential Rain and Disasters 出版日期：2015-12-15 出版单位：暴雨灾害 机构：安徽省气象台; 年：2015 期：04 基金：中国气象局2014年“全国集合预报应用业务系统建设(一期)”项目 CN：42-1771/P ISSN：1004-9045 页：26-33 页数：8 分类号：P456.7 语种：中文; 文件大小：1429K

摘要

使用2012年10月—2014年9月欧洲中期天气预报中心(ECMWF)全球集合预报系统逐日降水、5 d和10 d累计降水极端天气指数(EFI)预报资料,分析了降水EFI和强降水、降水气候距平的统计关系。结果表明:整体而言,降水EFI大值区和强降水具有较好对应关系,EFI越大,发生强降水的可能性越大,但随着预报时效增加,EFI的指示作用逐渐下降;对逐日降水EFI大值区和降水落区TS评分表明,在最优TS评分条件下,EFI阈值随着降水量级增加而增大,随着预报时效增加而降低;冬半年(10—3月)5 d和10 d累计降水EFI对过程降水的指示意义优于夏半年(4—9月),但当EFI超过0.4时,随其增加,不同季节对应降水气候距平均迅速增加;另外,5 d累计降水EFI对过程降水的指示作用优于10 d累计降水EFI。
        In this study, the precipitation Extreme Forecast Index(EFI) from ECMWF global ensemble prediction system(EPS) from October 2012 to September 2014, including 1d, 5d and 10 d accumulated precipitation EFI, were used to evaluate the statistical relationship between EFI and severe precipitation and its anomaly, respectively. Results indicate that the high value center of EFI corresponds to severe precipitation zone. The greater the EFI is, the higher the probability that severe precipitation events occurred is. But as the forecast time length increases, the indicative significance of EFI becomes less evident. TS scores of the high value center of EFI for daily precipitation and the precipitation area show that to obtain the best TS score, the thresholds of EFI increase with the increase of precipitation grades and decrease with the increase of forecast time length. The indicative significance of EFI for 5 d and 10 d accumulated precipitation to heavy rainfall in the cold seasons(from October to March the following year) is better than that in the warm seasons(from April to September). When EFI is greater than 0.4, the precipitation climate anomalies increase quickly with the increase both 5 d and 10 d accumulated precipitation. In addition, the indicative significance of EFI for 5 d accumulated precipitation is better than that for 10 d accumulated precipitation.

引文

[1]杜钧,陈静.单一值预报向概率预报转变的基础:谈谈集合预报带来的变革[J].气象,2010,36(11):1-11
    [2]杜钧,邓国.单一值预报向概率预报转变的价值:谈谈概率预报的检验和应用[J].气象,2010,36(12):10-18
    [3]邓国,龚建东,邓莲堂,等.国家级区域集合预报系统研发和性能检验[J].热带气象学报,2010,21(5):513-523
    [4]张涵斌,陈静,智协飞.GRAPES区域集合预报系统应用研究[J].气象,2014,40(9):1 076-1 087
    [5]张涵斌,陈静,智协飞,等.基于GRAPES_Meso的集合预报扰动方案设计与比较[J].大气科学学报,2014,37(3):276-284
    [6]李俊,王明欢,公颖,等.AREM短期集合预报系统及其降水预报检验[J].暴雨灾害,2010,29(1):30-37
    [7]林春泽,祁海霞,智协飞,等.中国夏季降水多模式集成概率预报研究[J].暴雨灾害,2013,32(4):354-359
    [8]陈朝平,冯汉中,陈静.基于贝叶斯方法的四川暴雨集合概率预报产品释用[J].气象,2010,36(5):22-39
    [9]刘琳,陈静,程龙,等.基于集合预报的中国极端强降水预报方法研究[J].气象学报,2013,71(5):853-866
    [10]杜钧,Grumm R H,邓国.预报异常极端高影响天气的“集合异常预报法”:以北京2012年7月21日特大暴雨为例[J].大气科学,2014,38(4):685-699
    [11]Lalaurette F.Early detection of abnormal weather conditions using a probabilistic extreme forecast index[J].Quart J Roy Meteor Soc,2002,129(594):3 037-3 057
    [12]Lalaurette F.Two proposals to enhance the EFI response near the tails of the climate distribution[EB/OL].http://www.ecmwf.int/products/forecasts/efi_guicle.pdf,2003:1-8
    [13]Zsoter E.Recent developments in extreme weather forecasting[J].ECMWF Newsletter,2006,107:8-17
    [14]Tsonevsky I,Richardson D.Application of the new EFI products to a case of early snowfall in Central Europe[J].ECMWF Newsletter,2012,133:4-5
    [15]Dutra E,Diamantakis M,Tsonevsky I,et al.The extreme forecast index at the seasonal scale[J].Atmosph Sci Lett,2013,14:256-262
    [16]夏凡,陈静.基于T213集合预报的极端天气预报指数及温度预报应用试验[J].气象,2012,38(12):1 492-1 501
    [17]肖丹,邓莲堂,陈静,等.T213与T639资料驱动WRF的预报初步检验比较[J].暴雨灾害,2010,29(1):20-29
    [18]陈超君,李俊,王明欢.2013年汛期华中区域业务数值模式降水预报检验[J].气象与环境学报,2015,31(2):1-8
    [19]公颖,李俊,鞠晓慧.2009年我国汛期降水形势总结与三个常用模式预报效果检验[J].热带气象学报,2011,27(6):823-833

目录