Assessment of Artificial Intelligence and Remote Sensing-Based groundwater storage management workflow

Groundwater is a vital freshwater reserve that is increasingly threatened by climate change and mounting anthropogenic pressure on global water resources. While arti- ficial intelligence (AI) has shown promise in groundwater monitoring in conjunction with remote sensing (RS), its integration with traditional technology remains bound to outdated hydrological assumptions, limiting the adaptability of the integrated approach across diverse regions and conditions. Here, we develop a hydrology- independent workflow using an explainable AI framework based on satellite obser- vations to monitor and forecast groundwater storage dynamics. Tested in a case study on a dataset of Morocco, our approach measured groundwater storage variations with high accuracy, without relying on conventional drivers such as precipitation. Model performance revealed robust spatiotemporal scalability and interpretabil- ity, promising broader applications across data-scarce environments. These findings demonstrate that moving away from classical hydrological dependencies and inte- grating AI-based frameworks can enhance the flexibility of groundwater modeling, offering a pathway toward more adaptive and scalable groundwater management, especially under climate uncertainty.