Spatiotemporal Variations of Groundwater Storage in arid region

¹ Institute of Hydrogeology and Engineering Geology of Gansu Bureau of Geology and Mineral Exploration and Development, Zhangye, 734000, Gansu, China;
² Heihe Water Resources and Ecological Protection Research Center, Lanzhou, 730030, P. R. China;
³ Gansu Inland Arid Area Water Cycle and Water Balance Field Scientific Observation and Research Station, Zhangye, 734000, Gansu, China;
⁴ Geological Society of China Technology Innovation Base for Water Cycle Monitoring and Water Balance Analysis in Inland Arid Areas, Zhangye, 734000, Gansu, China

^* Corresponding author: 602656943@qq.com

Abstract

Groundwater depletion is a critical issue in arid regions, where groundwater supports both agriculture and ecosystems. The Heihe River Basin (HRB) in northwestern China is a representative area with high groundwater dependence. This study estimates groundwater storage (GWS) changes in the HRB from 2003 to 2020 using multi-source remote sensing data and Bayesian statistical modeling. GRACE satellite data, GLDAS model outputs, and JRC surface water records were used to derive total water storage (TWS), soil moisture (SM), snow water equivalent (SWE), and surface water contributions. GWS was calculated by subtracting these components from TWS. A Bayesian framework was applied to quantify trends and associated uncertainties. Results show significant groundwater depletion, especially in the middle and lower reaches, where agricultural water demand is highest. The Bayesian model effectively captures long-term trends and seasonal variability while accounting for uncertainty, providing a robust basis for sustainable groundwater management in arid regions. Furthermore, these findings inform specific management practices and policy recommendations for groundwater conservation—such as dynamic extraction permits, managed aquifer recharge schemes, precision irrigation scheduling, tiered extraction fees, and the integration of climate projections into adaptive management plans—to ensure long-term water security in the HRB.