Joint risk of interbasin water transfer and impact of the window size of sampling low flows under environmental change
Academic Article
Overview
Research
Identity
Additional Document Info
Other
View All
Overview
abstract
2017 Elsevier B.V. Constructing a joint distribution of low flows between the donor and recipient basins and analyzing their joint risk are commonly required for implementing interbasin water transfer. In this study, daily streamflow data of bi-basin low flows were sampled at window sizes from 3 to183 days by using the annual minimum method. The stationarity of low flows was tested by a change point analysis and non-stationary low flows were reconstructed by using the moving mean method. Three bivariate Archimedean copulas and five common univariate distributions were applied to fit the joint and marginal distributions of bi-basin low flows. Then, by considering the window size of sampling low flows under environmental change, the change in the joint risk of interbasin water transfer was investigated. Results showed that the non-stationarity of low flows in the recipient basin at all window sizes was significant due to the regulation of water reservoirs. The general extreme value distribution was found to fit the marginal distributions of bi-basin low flows. Three Archimedean copulas satisfactorily fitted the joint distribution of bi-basin low flows and then the Frank copula was found to be the comparatively better. The moving mean method differentiated the location parameter of the GEV distribution, but did not differentiate the scale and shape parameters, and the copula parameters. Due to environmental change, in particular the regulation of water reservoirs in the recipient basin, the decrease of the joint synchronous risk of bi-basin water shortage was slight, but those of the synchronous assurance of water transfer from the donor were remarkable. With the enlargement of window size of sampling low flows, both the joint synchronous risk of bi-basin water shortage, and the joint synchronous assurance of water transfer from the donor basin when there was a water shortage in the recipient basin exhibited a decreasing trend, but their changes were with a slight fluctuation, in particular at shorter window sizes. The decreasing magnitude of individual water shortage risk in the recipient basin was significantly greater than that of joint bi-basin water shortage risk at all window sizes. Although the donor basin had much rich water, but its individual risk accounted for more than 25.3% within a window size of 30 days provided that its instream flow was maintained. In practical design and management of interbasin water transfer, impacts of the window size of sampling low flows under environmental change on the joint distribution of bi-basin low flows and the individual risk of water shortage in the donor basin should be paid more attention.
