Abstract
The Yarlung Tsangpo-Brahmaputra River (YBR) originating from the Tibetan Plateau (TP), is an important water source for many domestic and agricultural practices in countries including China, India, Bhutan and Bangladesh. To date, only a few studies have investigated the impacts of climate change on water resources in this river basin with dispersed results. In this study, we provide a comprehensive and updated assessment of the impacts of climate change on YBR streamflow by integrating a physically based hydrological model, regional climate integrations from CORDEX (Coordinated Regional Climate Downscaling Experiment), different bias correction methods, and Bayesian model averaging method. We find that (i) bias correction is able to reduce systematic biases in regional climate integrations and thus benefits hydrological projections over YBR Basin; (ii) Bayesian model averaging, which optimally combines individual hydrological simulations obtained from different bias correction methods, tends to provide hydrological time series superior over individual ones. We show that by the year 2035, the annual mean streamflow is projected to change respectively by 6.8%, -0.4%, and -4.1% under RCP4.5 relative to the historical period (1980-2001) at the Bahadurabad in Bangladesh, the upper Brahmaputra outlet, and Nuxia in China. Under RCP8.5, these percentage changes will substantially increase to 12.9%, 13.1%, and 19.9%. Therefore, the change rate of streamflow shows strong spatial variability along the YBR from downstream to upstream. The increasing rate of streamflow shows an augmented trend from downstream to upstream under RCP8.5 compared to an attenuated pattern under RCP4.5.