Distributed simulation and evolution characteristics of water cycle in the Mu Us Sandy Land under changing environment

Climate changes and human activities have significant impact on hydrological cycles, especially in arid and semi-arid zones． To investigate evolution characteristics of hydrological cycle in typical semi-arid ecologically fragile areas such as the Mu Us Sandy Land under changing environment, WEP-L (water and energy transfer processes in large river basin) surface water model and Visual MODFLOW (visual modular finite difference groundwater flow) groundwater model are used in this study to simulate the water cycle process． Model parameters are calibrated and validated with measured monthly runoff from Baijiachuan hydrological station and groundwater level data from monitoring wells． The correlation coefficient between simulated and measured monthly runoff in the surface water model calibration period and validation period are both above 0.5． The trends of the simulated and measured groundwater levels in the monitoring wells are more consistent with each other, with a correlation coefficient of > 0.56. Bias-corrected CMIP6 multi-model ensemble climate change scenario and local water withdrawal scenario are used to drive the coupled surface water-groundwater model, to quantify characteristics of changes in key water cycle elements in the Mu Us Sandy Land under changing environments from 2021-2040． In future scenario, average increase of annual precipitation and annual evapotranspiration in the Mu Us Sandy Land are found to be 8.0% and 7.4% respectively, with depth of annual runoff in a gradual decreasing trend． Groundwater level at the end of 2040 will decrease about 1 m compared with the end of 2021． Difference between total regional groundwater recharge and discharge under the two water use scenarios in the future is −182 million m³·a⁻¹ and −97 million m³·a⁻¹, difference between total surface runoff and surface water use being 183 million m³·a⁻¹ and 229 million m³·a⁻¹． Therefore, development and utilization of groundwater resources will face greater challenges． This work will support regional water resources planning and management in the Mu Us Sandy Land．