Advancing aquifer recharge forecasting through hybrid explainable AI and hydrological modeling

• Develops a hybrid explainable AI-hydrological framework for recharge prediction. • Integrates physical modeling and data-driven inference for transparent analysis. • Identifies precipitation and antecedent moisture as key recharge controls. • Improves detection of low-magnitude recharge events in long-term datasets. • Projects declining recharge extremes under future climate scenarios. Reliable aquifer recharge prediction is essential for climate-resilient and sustainable groundwater management, yet uncertainty persists due to subsurface heterogeneity and the lack of direct basin-scale recharge measurements. We present a serial hybrid eXplainable Artificial Intelligence (XAI) framework that leverages hydrological model-derived recharge estimates to train AI models, improving prediction accuracy, transparency, and interpretability. The framework was applied to two basins within the karstic Edwards Aquifer system in Texas, USA. The XAI models identified subtle recharge events in the test dataset that were missed by the hydrological model, with findings corroborated by in-situ hydroclimatic records, HSPF recharge estimates, GRACE-derived groundwater storage anomalies, and bootstrap analyses. The results demonstrated the XAI model’s superior learning capability beyond emulators to identify limitations in the training model and test data while robustly predicting high and low aquifer recharge events. Using long-term (1946–2023) hydroclimatic records and SHapley Additive exPlanations (SHAP), the best-performing AI model (Extremely Randomized Trees) identified basin-specific recharge drivers: antecedent soil moisture dominated in the larger, warmer, and drier basin with perennial streams, while current-month precipitation was the primary driver in the smaller urbanizing basin characterized with small ephemeral streams and highly fractured zones. Each driver explained ∼32% of the variability in recharge estimates, underscoring the model’s generalizability. SHAP-based analysis further enabled probabilistic identification of hydroclimatic conditions conducive to enhanced recharge. Projections based on downscaled CMIP6 climate data under intermediate- and high-emission scenarios indicate a decline in large recharge events in both basins through 2100, highlighting potential risks to groundwater sustainability