国家级强对流天气分类预报检验分析
|
摘要
预报产品的客观检验是记录、考量各种预报业务质量,促进预报水平提高的重要手段,也是整个天气预报过程中的重要环节。本文采用"点对面"Threat Score(TS)、漏报率、空报率等客观指标首次对2010—2015年4—9月国家级强对流天气预报中雷暴、短时强降雨以及雷暴大风和冰雹等分类预报进行了检验。同时,本文也对强对流天气落区分类预报客观检验存在的问题以及未来发展进行了讨论。检验结果表明:过去6年间,6~24 h时效预报,雷暴TS评分为0.22~0.34,短时强降水为0.18~0.24,雷暴大风和冰雹为0.01~0.07;48、72 h时效预报、雷暴TS评分为0.30~0.40,强对流天气TS评分为0.16~0.23,除雷暴预报TS评分在2012—2013年有所回落外,其他类别的强对流天气预报总体上TS评分呈上升趋势,雷暴大风和冰雹预报评分明显低于其他两个类别。雷暴空报率是漏报率的2~3倍,短时强降水漏报率与空报率接近,雷暴大风和冰雹天气漏报率和空报率都在0.8以上。与美国风暴预报中心(SPC)2000—2010年定期发布的1 d对流展望产品检验结果比较,强天气预报中心雷暴和短时强降水落区预报TS评分较高,雷暴大风和冰雹评分较低。典型个例预报检验结果表明,系统性大范围的风雹天气可预报性较强,评分要显著高于平均预报水平;对于非过程性的、分散的风雹天气,预报难度大,TS评分低。
The objective assessment of forecast is an important part in the whole weather forecasting cycle.Based on the assessment,qualities of different operational forecasts can be noted and considered,so enhancing the level of forecast.The verification of national severe convective weather categorical forecasts from April to September during 2010-2015 is displayed by adopting the objective indices of the "point-to-area" thret score(TS),the false alarm rate(FAR) and the missing alarm rate(MAR) in this paper.At the same time,analysis is carried out for different time interval forecasts of the annual mean verification index(mainly is from April to September of the year).Furthermore,the problems existing in the verification of severe convective weather categorical forecasts and its development in the future are also discussed.Over the past six years,in addition to the decrease of thunderstorm forecast TS during 2012-2013,there was a rising trend for TS of severe convective weather forecasting.For the 6-24 h forecasts,the TS of thunderstorm was 0.22-0.34,the TS of the short-time heavy rain was 0.18-0.24,the TS of thunderstorm gale and hailstone was 0.01-0.07;for the 48-72 h forecasts,the TS of thunderstorm was 0.30- 0.40,the TS of severe convective weather was 0.16- 0.23.The TS of thunderstorm gale and hailstone was lower than that of the other two kinds of severe convective weather.The FAR of thunderstorm was twice as large as MAR of thunderstorm;the FAR of short-time heavy rain was close to its MAR,while the FAR and MAR of thunderstorm gale and hailstone were all larger than 0.8.Compared with the TSs of similar products in Storm Prediction Center(SPC) of USA,the TSs of thunderstorm and short-time heavy rain are higher than that of SPC and the TS of thunderstorm gale and hailstone is lower than that of SPC.Verification of typical case forecasting shows that the TS of systemic and widespread thunderstorm gale and hailstone is higher than that in other situations.The reason is that the predictability of systemic and widespread thunderstorm gale and hailstone is higher.
The objective assessment of forecast is an important part in the whole weather forecasting cycle.Based on the assessment,qualities of different operational forecasts can be noted and considered,so enhancing the level of forecast.The verification of national severe convective weather categorical forecasts from April to September during 2010-2015 is displayed by adopting the objective indices of the "point-to-area" thret score(TS),the false alarm rate(FAR) and the missing alarm rate(MAR) in this paper.At the same time,analysis is carried out for different time interval forecasts of the annual mean verification index(mainly is from April to September of the year).Furthermore,the problems existing in the verification of severe convective weather categorical forecasts and its development in the future are also discussed.Over the past six years,in addition to the decrease of thunderstorm forecast TS during 2012-2013,there was a rising trend for TS of severe convective weather forecasting.For the 6-24 h forecasts,the TS of thunderstorm was 0.22-0.34,the TS of the short-time heavy rain was 0.18-0.24,the TS of thunderstorm gale and hailstone was 0.01-0.07;for the 48-72 h forecasts,the TS of thunderstorm was 0.30- 0.40,the TS of severe convective weather was 0.16- 0.23.The TS of thunderstorm gale and hailstone was lower than that of the other two kinds of severe convective weather.The FAR of thunderstorm was twice as large as MAR of thunderstorm;the FAR of short-time heavy rain was close to its MAR,while the FAR and MAR of thunderstorm gale and hailstone were all larger than 0.8.Compared with the TSs of similar products in Storm Prediction Center(SPC) of USA,the TSs of thunderstorm and short-time heavy rain are higher than that of SPC and the TS of thunderstorm gale and hailstone is lower than that of SPC.Verification of typical case forecasting shows that the TS of systemic and widespread thunderstorm gale and hailstone is higher than that in other situations.The reason is that the predictability of systemic and widespread thunderstorm gale and hailstone is higher.
引文
戴建华,茅懋,邵玲玲,等,2013.强对流天预报检验新方法在上海的应用尝试[J].气象科技进展,3(3):40-45.
董克勤,杨麟美,周江兴,1986.台风路径预报现状分析[J].气象,12(7):2-6.
李佰平,戴建华,张欣,等,2016.三类强对流天气临近预报的模糊检验试验与对比[J].气象,42(2):129-143.
林明智,1997.美国国家气象中心定量降雨预报[J].气象,23(11):3-6.
林明智,毕宝贵,乔林,1995.中央气象台短期降雨预报水平初步分析[J].应用气象学报,6(4):392-399.
刘凑华,牛若芸,2013.基于目标的降水检验方法及其应用[J].气象,39(6):681-690.
茅懋,戴建华,李佰平,等,2016.不同类型强对流预报产品的目标对象检验与分析评价[J].气象,42(4):389-397.
牟惟丰,1988.中央气象台一年来预报评分结果分析[J].气象,14(11):49-51.
束家鑫,王志烈,1983.我国台风研究的十年进展/台风会议文集(1981)[M].上海:上海科学技术出版社:1-4.
许映龙,张玲,高拴柱,2010.我国台风预报业务的现状及思考[J].气象,36(7):43-49.
郑永光,周康辉,盛杰,等,2015.强对流天气监测预报预警技术进展[J].应用气象学报,26(6):641-657.
Anna G,Elizabeth E,2008.Special issue on forecast verification[J].Meteor Appl,15(1),1-1.
Brown B G,Bullock R R,Davis C A,et al,2004.New verification approaches for convective weather forecasts//Proceedings of the22nd Conference on Severe Local Storms[R].
Casati B,Wilson L J,Stephenson D B,et al,2008.Forecast verification:current status and future directions[J].Meteor Appl,15(1):3-18.
Ferro C A T,Stephenson D B,2011.Extremal dependence indices:Improved verification measures for deterministic forecasts of rare binary events[J].Wea Forecasting,26(5):699-713.
FinleyJP,1884.Tornado predictions[J].Amer Meteor J,1:85-88.
Hitchens N M,Brooks H E,2012.Evaluation of the storm prediction center's day 1 convective outlooks[J].Wea Forecasting,27(6):1580-1585.
Hitchens N M,Brooks H E,2014.Evaluation of the Storm Prediction Center's convective outlooks from day 3 through day 1[J].Wea Forecasting,29(5):1134-1142.
Kumar K,2005.Forecast Verification:a Practitioner's Guide in Atmospheric Sciences[J].J Roy Statis Soc:Series A(Statistics in Society),168(1):255-255.
Mason I,1989.Dependence of the critical success index on sample climate and threshold probability[J].Aust Meteor Mag,37:75-81.
Roberts N M,Lean H W,2008.Scale-selective verification of rainfall accumulations from high-resolution forecasts of convective events[J].Mon Wea Rev,136(1):78-97.
Schaefer J T,1990.The critical success index as an indicator of warning skill[J].Wea Forecasting,5(4):570-575.
Stephenson D B,Casati B,Ferro CAT,et al,2008.The extreme dependency score:a non-vanishing measure for forecasts of rare events[J].Meteor Appl,15(1):41-50.
董克勤,杨麟美,周江兴,1986.台风路径预报现状分析[J].气象,12(7):2-6.
李佰平,戴建华,张欣,等,2016.三类强对流天气临近预报的模糊检验试验与对比[J].气象,42(2):129-143.
林明智,1997.美国国家气象中心定量降雨预报[J].气象,23(11):3-6.
林明智,毕宝贵,乔林,1995.中央气象台短期降雨预报水平初步分析[J].应用气象学报,6(4):392-399.
刘凑华,牛若芸,2013.基于目标的降水检验方法及其应用[J].气象,39(6):681-690.
茅懋,戴建华,李佰平,等,2016.不同类型强对流预报产品的目标对象检验与分析评价[J].气象,42(4):389-397.
牟惟丰,1988.中央气象台一年来预报评分结果分析[J].气象,14(11):49-51.
束家鑫,王志烈,1983.我国台风研究的十年进展/台风会议文集(1981)[M].上海:上海科学技术出版社:1-4.
许映龙,张玲,高拴柱,2010.我国台风预报业务的现状及思考[J].气象,36(7):43-49.
郑永光,周康辉,盛杰,等,2015.强对流天气监测预报预警技术进展[J].应用气象学报,26(6):641-657.
Anna G,Elizabeth E,2008.Special issue on forecast verification[J].Meteor Appl,15(1),1-1.
Brown B G,Bullock R R,Davis C A,et al,2004.New verification approaches for convective weather forecasts//Proceedings of the22nd Conference on Severe Local Storms[R].
Casati B,Wilson L J,Stephenson D B,et al,2008.Forecast verification:current status and future directions[J].Meteor Appl,15(1):3-18.
Ferro C A T,Stephenson D B,2011.Extremal dependence indices:Improved verification measures for deterministic forecasts of rare binary events[J].Wea Forecasting,26(5):699-713.
FinleyJP,1884.Tornado predictions[J].Amer Meteor J,1:85-88.
Hitchens N M,Brooks H E,2012.Evaluation of the storm prediction center's day 1 convective outlooks[J].Wea Forecasting,27(6):1580-1585.
Hitchens N M,Brooks H E,2014.Evaluation of the Storm Prediction Center's convective outlooks from day 3 through day 1[J].Wea Forecasting,29(5):1134-1142.
Kumar K,2005.Forecast Verification:a Practitioner's Guide in Atmospheric Sciences[J].J Roy Statis Soc:Series A(Statistics in Society),168(1):255-255.
Mason I,1989.Dependence of the critical success index on sample climate and threshold probability[J].Aust Meteor Mag,37:75-81.
Roberts N M,Lean H W,2008.Scale-selective verification of rainfall accumulations from high-resolution forecasts of convective events[J].Mon Wea Rev,136(1):78-97.
Schaefer J T,1990.The critical success index as an indicator of warning skill[J].Wea Forecasting,5(4):570-575.
Stephenson D B,Casati B,Ferro CAT,et al,2008.The extreme dependency score:a non-vanishing measure for forecasts of rare events[J].Meteor Appl,15(1):41-50.