摘要
变化环境下,探究干旱多属性相依结构动态变化及其驱动力,对区域干旱风险管理具有重要意义。以渭河流域为研究对象,应用Copula函数的极大似然比(Copula-based Likelihood-ratio,CLR)方法,对水文干旱历时-烈度相依结构进行变异诊断,分析变异前后特征;从气候变化(气象因子、气象干旱特征)与人类活动的角度,阐明水文干旱历时-烈度相依结构可能的驱动因素,获得影响因子重要性排序。结果表明:(1)华县、张家山、状头站水文干旱历时-烈度相依结构分别在1993年、1982年、1994年发生变异;(2)在相对长的历时下(华县站9~12个月、状头站5~12个月),华县、状头站相同历时下,变异点前后的烈度增加,且随着历时变长,烈度增加的幅度上升,干旱加剧态势明显;(3)除气象干旱对水文干旱演变的影响外,水文干旱历时-烈度相依结构的动态变化主要受水分“需求侧”因子(人类取用水、潜在蒸散发)的驱动。总体而言,研究成果为干旱多属性相依结构的非平稳性检测提供了一种新的思路,所提框架可推广应用于其他流域。
In a changing environment,exploring the dynamic changes and driving forces of drought multi-attribute dependent structure is of great significance for regional drought risk management.This paper takes the Weihe River Basin as the research object and applies the Copula-based maximum likelihood ratio(CLR)method to diagnose the abrupt change of hydrological drought duration-severity dependence structure,along with analyzing the characteristics before and after the abrupt change.From the perspective of climate change(meteorological factors,characteristics of meteorological drought)and human activities,clarify the possible driving factors of hydrological drought duration-severity dependency structure dynamics,and obtain the ranking of the importance of the influencing factors.The results showed that:(1)The hydrological drought duration-severity dependence structure of Huaxian,Zhangjiashan,and Zhuangtou stations had abrupt change point in 1993,1982,and 1994 respectively;(2)Under relatively long duration(9-12 months at Huaxian station,5-12 months at Zhuangtou station),under the same duration of Huaxian and Zhuangtou station,the severity before and after the change point increases,and as the duration becomes longer,the magnitude of the increase in severity increases,and drought has increased significantly;(3)In addition to the impact of meteorological drought on the evolution of hydrological drought,the hydrological drought duration-severity dependency structure dynamics is mainly driven by water“demand-side”factors(human water consumption,potential evapotranspiration).In general,the research results provide a new idea for the non-stationarity detection of drought multi-attribute dependency structures,and the proposed framework can be extended to other watersheds.
作者
马川惠
黄生志
黄强
MA Chuanhui;HUANG Shengzhi;HUANG Qiang(State Key Laboratory of Eco-hydraulics in Northwest Arid Region of China,Xi’an University of Technology,Xi’an 710048,China)
出处
《水利学报》
EI
CSCD
北大核心
2022年第10期1180-1193,共14页
Journal of Hydraulic Engineering
基金
国家重点研发计划项目(2021YFC3000203)
陕西省高校科协青年人才托举计划项目(20190413)
国家自然科学基金项目(51879213)
陕西省水利科技计划项目(2017slkj-16)。
关键词
水文干旱
COPULA函数
干旱历时-烈度相依结构
驱动力
随机森林法
渭河流域
hydrological drought
Copula function
drought duration-severity dependency structure
driving force
Random Forest Method
Weihe River Basin
作者简介
马川惠(1994-),博士生,主要从事水文水资源研究。E-mail:2669174313@qq.com;通讯作者:黄强(1958-),博士,教授,主要从事水文水资源研究。E-mail:syhuangqiang@163.com。
