Responses of hydrological drought properties and water shortage under changing environments in Dongjiang River basin

Modeling the joint distribution of hydrological drought associated with non-stationary drought properties under the effects of uncertain drought events and low flow variability is difficult. Daily streamflow data from gauging stations in Dongjiang River basin, South China are used. On the basis of run theory for truncating drought events, alterations are analyzed in terms of mean values for hydrological drought properties and their non-stationary statistics with changes in pooling and exclusion ratios. Bivariate copula functions are built for modeling the joint distribution of hydrological drought according to the test method of Cramer-von Mises statistics associated with Rosenblatt transfer. Furthermore, a pair of designed drought values for predefined joint return periods is calculated using the equalized probability method. Changes in drought properties associated with dependence and joint distribution, as well as water shortage under a changing environment in Dongjiang River basin, are investigated by comparing two subseries of droughts separated by the change point in low flow. Results demonstrate that the pooling and exclusion ratios set to 0.1 and 0.3, respectively, are rational. Despite positive dependence between drought properties, the bivariate and univariate models for hydrological droughts vary under different time series and variable pairs. The remarkable decrease in drought duration and water shortage in the middle and lower streams is attributed to the regulation of water reservoirs, particularly the Xinfengjiang Reservoir. When the joint return period exceeds the threshold by 2 years, the designed duration periods, total volumes, and daily maximum water shortage levels decrease by 63%-71%, 71%-84%, and 30%-47%, respectively. However, they still account for 12-18 d, 61.14-90.30 million, and 7.150-9.290 million m³, respectively, provided that instream flow satisfies the water demand in Dongjiang River basin.