Climate variability impacts on runoff projection under quantile mapping bias correction in the support CMIP6: An investigation in Lushi basin of China

In this study, we modelled the XAJ (Xin'anjiang) hydrologic model parameters for the Lushi basin of China, an important tributary of the Yellow River. Time series of daily precipitation, evaporation, and runoff data from 1976 to 1995 were used to calibrate the model, while the data from 1996 to 2000 were used for validation. Thereafter, the future climate variability was projected from 2021 to 2100 for 0.22° grids under CMIP6 CORDEX-East-Asia (Coordinated Regional Climate Downscaling Experiment-East-Asia) RCP2.6 and RCP8.5 climate scenarios. The non-parametric quantile mapping (QM) bias correction method was used to minimize the large differences between climate variability directly derived from the regional circulation model (RCM) and the historical observations, and the results were also compared with the baseline period (1976–2000). Overall, when compared to the baseline case, results showed a decline in the annual precipitation by 7.22 % and 5.01 %, while an increase was observed for the annual evaporation by 2.03 % and 3.58 % under RCP2.6 and RCP8.5 climate scenarios in the 21st century. Annual runoff projection by XAJ model indicate reduction, with 9.3 % and 31.2 % fall for short-term (2021–2040), 31.5 % and 41.2 % fall for mid-term (2051–2070) and 32.3 % and 77.4 % fall for long-term (2081–2100), respectively. The extreme streamflow projects increase during the dry period (November to June) with 76.5 m³/s and 84.6 m³/s under RCP2.6 and RCP8.5 compared to 45.6 m³/s from the baseline period. In contrast, there is a decrease during the wet period (July to October) by 390 m³/s and 237 m³/s compared to 405 m³/s in the baseline data of the same period. Such projections become more pronounced toward the end of the 21st century, and the reduction is likely to continue if future climate projections occur. The hydrological parameters can be applied to adjacent and similar landscape basins for disaster forecasting and early warning. The QM method is effective in bias correction. The findings can also be used for flood and drought mitigation planning and water resources management in the Yellow River Basin.