官晓军,潘宁,黄待静,王琦,李玲. 2021. 基于降水极端预报指数的福建台风极端降水预报研究[J]. 气象学报, (0):-, doi:10.11676/qxxb2021.022
基于降水极端预报指数的福建台风极端降水预报研究
Study of Typhoon Extreme Precipitation Forecast In Fujian Based on Precipitation Extreme Forecast Index
投稿时间:2020-06-30  修订日期:2020-10-17
DOI:10.11676/qxxb2021.022
中文关键词:  极端降水,预报方法,EFI,台风
英文关键词:Extreme  precipitation,Forecast  method,EFI,Typhoon
基金项目:
作者单位E-mail
官晓军 福建省气象服务中心 40392689@qq.com 
潘宁 福建省气象台 pancathy@163.com 
黄待静  807725797@qq.com 
王琦  topmad@vip.qq.com 
李玲  heidicat1987@163.com 
摘要点击次数: 34
全文下载次数: 50
中文摘要:
      [资料和方法]利用1961-2018年中国气象局热带气旋最佳路径数据集、站点日降水观测资料和2015年8月-2018年12月ECMWF集合预报系统降水极端预报指数(extreme forecast index, EFI)数据,通过百分位法定义台风影响期间福建各站点的台风极端降水阈值,并采用最小阈值法(剔除极端降水EFI箱线图中的异常值,保留最小值作为阈值)[目的]建立台风极端降水预报方法。基于该方法,分别对2015年8月-2017年12月和2018年的影响台风进行回报和预报试验,并采用TS评分、空报率、漏报率对其进行检验。[结果]结果表明:福建台风极端降水阈值由沿海向西北内陆逐渐减小,其中闽东沿海和南部沿海地区的阈值最大,闽西北内陆最小;台风降水、极端降水都与降水EFI存在明显正相关性,降水EFI的箱线图差异指数(box difference index, Ibd)也表明EFI可以较好地区分台风极端降水和非极端降水;预报试验20时起报预报时效12-36h、36-60h、60-84h和84-108h的TS评分分别为0.26、0.22、0.20和0.19,预报试验评分结果略优于回报试验;台风的极端降水越显著时,极端降水预报效果越好,回报和预报中都存在的不足是空报率高,主要出现在极端降水不明显的个例中。[结论]降水EFI对台风极端降水具有较好的指示意义,基于降水EFI的台风极端降水客观预报方法可作为业务中台风极端降水预报的参考依据。
英文摘要:
      Typhoon extreme precipitation forecast is of great importance for social and economic development. hence, an objective forecast method of typhoon extreme precipitation is in urgent need. So in this study a forecast method of typhoon extreme precipitation is developed by percentile method and minimum threshold method based on typhoon best tracks and station rainfall data obtained from CMA both for the period of 1961-2018 and 24h precipitation extreme forecast index (EFI) obtained from ECMWF ensemble prediction system ranging from 2015 to December 2018. EFI is a product derived from ECMWF ensemble prediction system that quantitatively measure the difference between ensemble prediction and model climate and it’s found in previous research that EFI is a good indicator to forecast extreme weather events. Fujian typhoon extreme precipitation threshold is defined by percentile method which refers to the daily precipitation of each station during the time period affected by typhoon is sorted by ascending order and the 95 percentile is regarded as the threshold of this station. Typhoon extreme precipitation EFI threshold is defined by minimum threshold method that removes anomalous EFI values including the outliers of the box plots and meaningless values (negative 24h precipitation EFI) and takes the minimum value as threshold. Verification has been done to both the reforecast (from August 2015 to December 2017 with 20 typhoon cases) and forecast experiments (year 2018 with 7 typhoon cases). The results are as follows: (1) Typhoon extreme precipitation thresholds gradually reduce from coastal areas to northwest inland of Fujian. East and southern coastal areas of Fujian have the largest thresholds (greater than 100mm) while the northwest inland of Fujian have the smallest thresholds(less than 50mm). (2) Daily precipitation caused by typhoon especially extreme precipitation has significant positive correlation with 24h precipitation EFI. The box difference indexes of EFI also show great capability to distinguish typhoon extreme and non-extreme precipitation events.(3) TS scores of forecast experiment are 0.26, 0.22, 0.20 and 0.19 with lead time of 12-36h, 36-60h, 60-84h and 84-108h initiated at 20h. Forecast experiment has better forecast skills than reforecast on the whole and Typhoon cases with significant extreme precipitation tend to have better forecast results. The defects of both experiments are high false alarm rates which are more likely to occur in typhoon cases with no extreme precipitation. In summary, precipitation EFI is a good indicator to typhoon extreme precipitation, the objective forecast method based on precipitation EFI could be a reference to typhoon extreme precipitation forecast.
查看全文   查看/发表评论  下载PDF阅读器
分享按钮