Deep Learning Based Modeling of Groundwater Storage Change

The understanding of water resource changes and a proper projection their future availability are necessary elements sustainable planning. Monitoring GWS change crucial, especially under changing climatic conditions. Traditional methods for in situ groundwater well measurement significant challenge due to data unavailability. present investigation utilized the Long Short Term Memory (LSTM) networks monitor forecast Terrestrial Water Storage Change (TWSC) Ground (GWSC) based on Gravity Recovery Climate Experiment (GRACE) datasets from 2003–2025 five basins Saudi Arabia. An attempt has been made assess effects rainfall, used, net budget modeling groundwater. Analysis GRACE-derived TWSC GWSC estimates indicates that all show depletion 2003–2020 with rate ranging −5.88 ± 1.2 mm/year −14.12 −3.5 1.5 −10.7 1.5, respectively. Forecasting developed LSTM model investigated likely experience serious at rates −7.78 −15.6 −4.97 −12.21 2020–2025. interesting observation was minor increase rainfall during study period three basins.