干旱是主要的自然灾害之一,它对环境和经济有着显著的影响。文中根据长春气象站106年的气象数据,基于标准化降水蒸散指数(SPEI)定义了干旱的两个主要特征:干旱历时和干旱程度,并利用Clayton Copula函数构造了干旱历时和干旱程度的联合分布函数,进而计算了干旱历时和干旱程度在不同组合条件下的联合超越概率、联合重现期、同现重现期及条件重现期。结果表明:(1)干旱历时和干旱程度的联合超越概率值随干旱历时和程度的值增大而呈现减小趋势,且递减速度逐渐变小;(2)单变量边缘分布的重现期介于联合重现期和同现重现期之间;(3)干旱历时(干旱程度)的条件重现期与干旱程度(干旱历时)的值成正比。研究结果可以为吉林省干旱风险的评估、水资源规划和管理提供有用的信息。
Droughts are of major natural hazards with significant environmental and economic impacts. In this study, two - dimensional Copulas were applied to the analysis of the meteorological drought characteristics of the Changchun gauge station with long time series of meteorological data. Two major drought characteristics, duration and severity, as defined by the Standardized Precipitation Evapotranspiration Index, were abstracted from observed drought events, and Clayton Copula was selected to construct the joint distribution function of drought characteristics. Based on the derived Copula - based joint distribution, some bivariate probabilistic properties of droughts, such as, joint probabilities, bivariate return periods, and conditional return periods were also investi- gated. The results showed that : ( 1 ) The joint probabilities of duration (DD) and drought severity (DS) showed a decreasing trend with the increase of DD or DS, and the rate became slow gradually. (2) The univariate marginal distribution of return period are between joint return period and co - occurrence return period. (3) The conditional return period of DD/DS is proportional to the DS/DD. In addition, the Changchun gauge station is located in the central part of Jilin province, and its climatic conditions can well represent that in most parts of Jilin province, so the results of this study can provide useful information for drought risk assessment, water resource planning and management in Jilin province.