基于MCDI的汉江流域季度干旱分析  

Analysis of quarterly drought in Hanjiang River Basin based on multivariate comprehensive drought index

作　　者：陈灏 赵小娥 熊佳 董前进[2] 银星黎 黑亮 CHEN Hao;ZHAO Xiao'e;XIONG Jia;DONG Qianjin;YIN Xingli;HEI Liang(Pearl River Hydrology and Water Resources Survey Center,Guangzhou 510370,China;State Key Laboratory of Water Resources Engineering and Management,Wuhan University,Wuhan 430072,China;Pearl River Water Resources Research Institute,Guangzhou 510610,China)

机构地区：[1]珠江水文水资源勘测中心,广东广州510370 [2]武汉大学水资源与水电工程科学国家重点实验室,湖北武汉430072 [3]珠江水利委员会珠江水利科学研究院,广东广州510610

出　　处：《武汉大学学报(工学版)》2025年第3期353-361,共9页Engineering Journal of Wuhan University

基　　金：国家自然科学基金项目(编号:51979198)。

摘　　要：利用Copula函数构建以气象干旱指数——标准化降水指数(standardized precipitation index,SPI)、水文干旱指数——径流异常百分比(runoff anomaly percentage,RAP)及农业干旱指数——帕默尔干旱指数(Palmer drought severity index,PDSI)为边缘分布函数的联合概率密度分布函数,利用该联合分布函数进行进一步的频率分析与计算,提出了一个多元综合干旱指数(multivariate comprehensive drought index,MCDI)。以汉江流域为例进行MCDI的计算,并利用MCDI将季度干旱分为轻度、中度、严重及极度干旱4个等级,在此基础上将MCDI与边缘分布指数SPI、RAP、PDSI对干旱的识别情况进行了对比分析,发现MCDI与SPI、RAP、PDSI具有相似的趋势,且能够识别出气象、水文及农业这3种干旱中2种及2种以上干旱同时发生的时段,结果表明:利用MCDI对干旱进行识别是合理可行的,且综合考虑气象、水文及农业干旱的MCDI比单一指数更加适用于季度干旱的识别。The joint probability density distribution function is constructed by using Copula function and taking a meteorological drought index—standardized precipitation index(SPI),a hydrological drought index—runoff anomaly percentage(RAP)and an agricultural drought index—Palmer drought severity index(PDSI)as the marginal distribution functions.Through further frequency analysis and calculation of the joint distribution function,a comprehensive evaluation index of seasonal drought,multivariate comprehensive drought index(MCDI),is proposed.Taking Hanjiang River Basin as an example,the MCDI is calculated,and the MCDI is used to divide the quarterly drought into four grades:mild,moderate,severe and extreme drought.On this basis,the identification of drought by MCDI,SPI,RAP and PDSI is compared and analyzed.It is found that the MCDI has similar trend to the marginal distribution indices,SPI,RAP and PDSI,and it can identify the time period when more than two kinds of droughts occur simultaneously among the three kinds of meteorological,hydrological and agricultural droughts.The results show that it is reasonable and feasible to use MCDI to identify drought,and the MCDI that comprehensively considers meteorological,hydrological and agricultural droughts is more suitable for the identification of quarterly drought than a single index.

关 键 词：COPULA函数 干旱指数 MCDI 季度干旱 汉江流域 

分 类 号：TV213.4[水利工程—水文学及水资源]