干旱气象

• 论文 • 上一篇    下一篇

基于前期海温异常的宁夏5~9月候降水量客观预测方法及检验评估

郑广芬1,2,王素艳1,2,杨建玲1,丁小谨2,李欣2   

  1. 1.宁夏气象防灾减灾重点实验室,宁夏银川750002;
    2.宁夏气候中心,宁夏银川750002
  • 出版日期:2016-02-29 发布日期:2016-02-29
  • 通讯作者: 王素艳(1974-),女,宁夏中宁人,高级工程师,主要从事气候与气候变化研究.
  • 作者简介:郑广芬(1965-),女,河北献县人,正研级高工,主要从事气候与气候变化研究. E-mail:ych_zgf@163.com
  • 基金资助:

    公益性行业(气象)科研专项(GYHY201306027)、公益性行业(气象)科研重大专项(GYHY201506001)、宁夏自然科学基金项目(NZ12277)、中国气象局气象关键技术集成与应用项目(CMAGJ2015M65)及宁夏气象局科研项目“月内强降水过程趋势预测系统本地化应用与改进”共同资助

Study on the Objective Prediction Method and Its Verification andAssessment for Pentad Precipitation from May to September in Ningxia Based on the Preceding SST Anomaly

ZHENG Guangfen 1,2WANG Suyan 1,2YANG Jianling1DING Xiaojin2LI Xin2   

  1. 1. Key Laboratory of Meteorological Disaster Preventing and Reducing in Ningxia,Yinchuan 750002,China;
    2.Ningxia Climate Center, Yinchuan 750002, China
  • Online:2016-02-29 Published:2016-02-29

摘要:

利用宁夏1981~2014年降水资料,通过相关分析,发现与宁夏降水关系较好的主要是赤道太平洋、印度洋、北大西洋及西太平洋海温,其关键的影响时段为5月各候、6月第3候、8月1~3候、9月4~6候。利用多元回归方法,建立基于前期海温异常的候降水量客观预测模型。总体看来,模型对宁夏引黄灌区和南部山区拟合效果好于中部干旱带,对5月、8月、9月拟合效果最好。模型对各区域候降水量≥10 mm的准强降水过程预测,总体评分高于“异常相似释用预测方法”;南部山区预测效果最好,引黄灌区预测能力较弱;从各月看,6月和9月的预测准确率相对较高, 5月预测准确率最低。模型对引黄灌区候降水量≥5 mm的过程预测,6月效果最好。利用模型预测的2014年5~9月各站准强降水过程,ZS评分高于实时发布的业务产品。

关键词: 海温异常, 候降水量, 客观预测模型, 检验评估

Abstract:

By using the rainfall data in Ningxia from 1981 to 2014, the relationship between the preceding sea surface temperature anomaly (SSTA) and the precipitation has been analyzed firstly. The results show that the preceding sea surface temperature anomaly(SSTA) in the equatorial Pacific, the Indian Ocean, the North Atlantic and the Western Pacific had good corresponding correlation with the precipitation, and it was best in all pentads in May, the 3rd pentad in June, the 1st to 3rd pentad in August and the 4th to 6th pentad in September. Secondly, according to the multiple regression method, the objective prediction models for pentad precipitation based on the early SST anomaly were established. The model fitted effect was better in the Yellow River irrigation area and the southern mountain area than that in the middle arid area. The model predicted effect for the processes of precipitation more than or equal to 10 mm in Ningxia was better than that of “Similar Interpretation Method”, and it was better in the southern mountain area and worst in the Yellow River irrigation area. On monthly time scale, the accuracy rate of prediction was relatively higher in September and June, and the worst in May. The predicted effect for the processes of precipitation more than or equal to 5 mm in the Yellow River irrigation area in June was the best. ZS score of the processes of precipitation more than or equal to 10 mm in 2014 was higher than that of real-time published business products.

Key words: SST anomaly, pentad precipitation, objective prediction method, verification and assessment

中图分类号: