地理研究

地理研究 ›› 2011, Vol. 30 ›› Issue (9): 1693-1701.doi: 10.11821/yj2011090013

• 水文与水资源 • 上一篇    下一篇

东江流域枯水期最长连续无降水日数的变化特征

王兆礼1, 覃杰香1, 陈晓宏2   

  1. 1. 华南理工大学土木与交通学院,广州 510641;
    2. 中山大学水资源与环境研究中心,广州 510275
  • 收稿日期:2011-01-20 修回日期:2011-04-15 出版日期:2011-09-20 发布日期:2011-09-20
  • 通讯作者: 陈晓宏(1963-),男,湖北公安人,博士,教授,主要从事水文水环境研究。 E-mail: eescxh@sysu.edu.cn
  • 作者简介:王兆礼(1979-),男,江苏徐州人,博士,讲师,主要从事气候与土地利用变化引起的水文效应及GIS模拟等方面的研究。E-mail: wangzhl@scut.edu.cn
  • 基金资助:

    国家自然科学基金重点项目(50839005);华南理工大学中央高校基本科研业务费专项资金(2009ZM0186);水资源与水电工程科学国家重点实验室开放研究基金资助项目(2010B065)

Variations in longest consecutive dry days in dry season in the Dongjiang River Basin

WANG Zhao-li1, QIN Jie-xiang1, CHEN Xiao-hong2   

  1. 1. School of Civil and Transportation Engineering, South China University of Technology, Guangzhou 510641, China;
    2. Center for Water Resources and Environment Research, SUN Yat-sen University, Guangzhou 510275, China
  • Received:2011-01-20 Revised:2011-04-15 Online:2011-09-20 Published:2011-09-20

PDF (PC)

赞

被引次数

16 | 6

摘要： 利用1956~2008年东江流域各降雨站点逐日降雨资料,建立枯水期最长连续无降水日数时间序列(LCDD),采用经验模态分解(EMD)方法分析该序列的多尺度振荡变化及趋势变化,应用Mann-Kendall法识别突变点,采用R/S分析方法探讨变化趋势的持续性特征,并初步分析LCDD变化的可能原因。结果表明:近53年来,最长连续无降水日数时间序列具有以2.3 a为主的多个波动周期,表现出复杂的多时间尺度性;LCDD序列呈显著的增加趋势,增加速率为0.26 d/a,并于1982年发生了由少到多的突变;LCDD序列存在状态持续性和长期记忆性,未来仍将呈显著的增加趋势。东江流域枯水期LCDD的多少可能受ENSO信号的调节,即厄尔尼诺现象出现的当年LCDD普遍减少;厄尔尼诺现象出现的次年LCDD普遍增加。

关键词: 最长连续无降水日数, 变化特征, 经验模态分解, 东江流域

Abstract: Droughts can cause major economic and human losses, which affects hundreds of millions of people, and numerous studies have highlighted the need for drought prevention. Longest consecutive dry days (LCDD), defined as the longest period during the year when no measurable precipitation was recorded, are used as a simple indicator for identifying years with long dry periods and potential drought conditions. Based on daily precipitation data in dry season during 1956-2008 in the Dongjiang River Basin, this article constructs the time series of longest consecutive dry days (LCDD). The empirical mode decomposition (EMD), rescaled range (R/S) analysis and Mann-Kendall (M-K) test methods are used to explore the variation laws of LCDD and its influencing factors. The EMD method is used to obtain intrinsic mode functions (IMFs), by which variations of the original series are analyzed on various time scales. The results of intrinsic mode functions by EMD method indicate that there are obvious periodic variations with scales of 2.3, 4.0, 6.4, 10.7, 16.0 and 21.3 years for LCDD, and 2.3-year scale is the first period. The residual component decomposed by EMD shows that there is a significant changing trend with 0.26 d/a. The results of M-K test show that there exists an abrupt point in 1982 for the time series of LCDD. The Hurst exponent computed by R/S analysis method indicates that a long-term memory characteristic exists in LCDD time series. The Hurst exponent is found to be greater than 0.5, implying that the future tendency of LCDD is consistent with that of the past. LCDD is closely related with the occurrence of ENSO (El Nino-Southern Oscillation) event, that is, the year of El Nino phenomenon generally sees the LCDD reduction, while El Nino in the following year has a general increase in LCDD. The impact of ENSO on the LCDD is a very complex physical process. So further studies should be done to find the physical meaning of this process. Finally, in the Dongjiang River Basin, the results reported here are of great significance to research on drought and salinity upstream resistance, water supply planning, water environment protection and river health maintenance.

Key words: longest consecutive dry days, variation, empirical mode decomposition, the Dongjiang river basin