主办单位：中国气象局沈阳大气环境研究所
国际刊号：ISSN 1673-503X
国内刊号：CN 21-1531/P
气象与环境学报

气象与环境学报 ›› 2016, Vol. 32 ›› Issue (5): 32-39.doi: 10.3969/j.issn.1673-503X.2016.05.005

• 论文 • 上一篇    下一篇

基于天气分型和物理量诊断西北地区东部雷暴预报方法研究

孔德兵1,2, 汪结华3, 尚可政1, 叶伟2, 赵文婧1, 王式功1, 杨流贵2   

  1. 1. 兰州大学大气科学学院 半干旱气候变化教育部重点实验室, 甘肃 兰州 730000;
    2. 中国人民解放军95605部队气象台, 重庆 大足 402361;
    3. 中国人民解放军92493部队, 辽宁 葫芦岛 125000
  • 收稿日期:2015-07-21 修回日期:2015-09-07 出版日期:2016-10-31 发布日期:2016-10-31
  • 通讯作者: 尚可政 E-mail:shangkz@lzu.edu.cn
  • 作者简介:孔德兵,男,1990年生,在读硕士研究生,主要从事现代天气预报技术和极端天气气候研究,E-mail:540343856@qq.com。
  • 基金资助:

    公益性（气象）行业专项（GYHY201306047、GYHY201206004）和兰州大学中央高校基本科研业务费专项（lzujbky-2013-m03）共同资助。

Study on forecast method of thunderstorm based on weather types and physical diagnostic analysis in eastern region of Northwest China

KONG De-bing1,2, WANG Jie-hua3, SHANG Ke-zheng1, YE Wei2, ZHAO Wen-jing1, WANG Shi-gong1, YANG Liu-gui2   

  1. 1. Key Laboratory of Semi-Arid Climate Change of Ministry of Education, College of Atmospheric Sciences, Lanzhou University, Lanzhou 730000, China;
    2. Meteorological Observatory of 95605 Troops of People's Liberation Army, Dazu 402361, China;
    3. 92493 Troop People's Liberation Army, Huludao 125000, China
  • Received:2015-07-21 Revised:2015-09-07 Online:2016-10-31 Published:2016-10-31

PDF (PC)

432

摘要：

利用2008-2013年西北地区东部169个气象观测站的天气实况资料和NCEP/NCAR 1°×1°再分析资料计算的对流参数对西北地区东部雷暴日进行了天气分型，并应用分型统计的参数阈值对2013年西北地区东部雷暴日进行了试预报。结果表明：2008-2013年西北地区东部雷暴日集中出现在5-9月，占雷暴总日数的85.9%；雷暴日发生的天气形势可分为低涡型、低槽型、西北气流型和西南气流型4种。引入天气型强度指数，研究4种雷暴天气型的自动识别方法，通过天气分型检验表明，天气型自动识别方法可准确的识别雷暴发生的天气形势，且漏报较少。在天气型识别的基础上，进一步进行雷暴物理量诊断表明，消空效果明显。2008-2012年西北地区东部雷暴日回代预报的TS评分为54.1%，漏报雷暴日为4 d；2013年雷暴日试预报的TS评分为51.8%，漏报雷暴日为10 d；雷暴日回代预报与试预报的TS评分均超过气候概率，预报效果较理想，可为西北地区东部雷暴天气预报研究提供参考。

关键词: 雷暴, 天气型, 物理量诊断, 统计预报

Abstract:

Based on the meteorological observations from 169 meteorological stations and the NCEP/NCAR (National Centers for Environmental Prediction/National Center for Atmospheric Research) reanalysis data with resolution of 1°×1° in the eastern region of Northwest China from 2008 to 2013,weather types of thunderstorm days were classified using calculated convective parameters,and the operational forecast of thunderstorm in 2013 was carried out according to parameter thresholds of different weather types.The results show that thunderstorm days mostly occur from May to September and account for 85.9% of the total thunderstorm days during this study period.There are four weather types of thunderstorm days,including the lower vortex type,the lower trough type,the northwesterly flow type,and the southwesterly flow type.The automatic identification method of the four weather types is investigated by introducing a weather type intense index,and it is examined to be able to identify different weather types correctly with fewer omitted samples.The physical diagnostic analysis by weather type identification indicates that the omitted samples decrease significantly.The threat score (TS) of back-substitution forecast on thunderstorm days in the eastern region of Northwest China from 2008 to 2012 is 54.1%,with four days of omitted samples,while TS of operational forecast on thunderstorm days in 2013 is 51.8%,with ten days of omission.TS in back-substitution forecast and operational forecast is both larger than climatic probabilities,which means the forecast performs good and can provide a reference to the study on a thunderstorm in the eastern region of Northwest China.

Key words: Thunderstorm, Weather type, Physical diagnostic analysis, Statistical forecast

中图分类号: