地理学报 ›› 2009, Vol. 64 ›› Issue (3): 323-330.doi: 10.11821/xb200903007

• 水分循环 • 上一篇    下一篇

中国次雨量特征及其区域分异

谢云,林小鹃,殷水清,和清华   

  1. 北京师范大学地表过程与资源生态国家重点实验室,地理学与遥感科学学院,北京100875
  • 收稿日期:2008-08-29 修回日期:2009-01-09 出版日期:2009-03-25 发布日期:2010-08-03
  • 作者简介:谢云(1964-),女(回族),辽宁省大连市人,教授,博士生导师。主要从事土壤侵蚀及其影响评价研究。 E-mail: xieyun@bnu.edu.cn
  • 基金资助:

    国家重点基础研究发展规划项目(2007CB407203)

Characteristics of Individual Storms and Their Spatial Distributions in China

XIE Yun,LIN Xiaojuan,YIN Shuiqing,HE Qinghua   

  1. Beijing Normal University,State Key Laboratory of Earth Surface Processes and Resource Ecology, School of Geography,Beijing 100875,China
  • Received:2008-08-29 Revised:2009-01-09 Online:2009-03-25 Published:2010-08-03
  • Supported by:

    Natoinal Basic Research Program of China, No.2007CB407203

摘要:

基于统计方法的降雨随机模拟已成为目前许多地表过程模型的重要组成部分, 但多以 日降雨模拟为主。为了进行次降雨随机模拟, 必须揭示次降雨的统计特征。本文利用中国62 个气象站1954-2001 年5-9 月逐日小时降水资料, 整理出能够刻画次降雨特性的基本指标, 分析了我国次降雨的基本特征, 并利用卡方检验方法, 验证了我国次雨量主要服从两参数Γ 分布, 个别未通过检验的站点主要是由于对次雨量极值模拟效果不好所致。进一步选择Γ 分布两参数和90%、95%、99%次雨量百分位共5 个指标, 采用K 均值动态聚类法将我国次降 雨特征划分为5 个类型区域, 由东南向西北呈带状分布, 各区次雨量的偏态性、非均匀性和分散度依次减弱。利用5 个区Γ 分布两参数的区域化参数值, 可生成任一站点次雨量随机模 拟序列, 对于地表过程定量研究具有重要意义。

关键词: 次降雨,&Gamma, 分布, 次雨量

Abstract:

The statistical characteristics of individual storms are the basis for rainfall event simulation. This paper selected hourly precipitation data from May to September covering the period 1954-2001 from 62 weather stations throughout China to analyze the individual storm characteristics. The individual storms in China obey the gamma distribution by using chi-square test. Only several stations did not pass the test due to the poorly simulated effects for extreme events. Five indexes including two parameters in gamma distribution and the 90th, 95th and 99th percentiles of individual rainfall amount were used in the method of K means clustering. The results present a distribution of five zonal regions from southeast to northwest, with the decreasing ranking of the skewness, heterogeneity and dispersion for the individual rainfall amount during May to September in China. The regionalization of individual rainfall amount is the basis for estimating regional parameters in storm-based simulation models.

Key words: individual storm event, gamma distribution, individual rainfall amount