A Multi-Source Geospatial Framework for Quantifying Groundwater Recharge and Storage Change in Data-Scarce Arid Regions: Application to the Mandera Sub-Basin, Northern Kenya

Groundwater is the primary water source in arid and semi-arid regions, but its assessment is often hindered by limited observational data. This study presents a comprehensive geospatial framework to estimate groundwater recharge, changes in storage, and available resources in the Mandera sub-basin in northeastern Kenya. The framework combines GIS-based recharge calculations and GRACE/GRACE-FO mascon data (JPL RL06) satellite-derived groundwater storage estimates. Recharge was estimated using an inverse water-balance approach, combining effective rainfall from CHIRPS precipitation and FAO Penman-Monteith evapotranspiration with hydrogeological factors, including lithology, soil texture, and slope. Results show significant spatial variability: high recharge occurs in the southern and eastern sub-basins, where coarse, permeable sediments are present, while the northern and western regions, with shale and dolomite formations, exhibit low recharge. GRACE-derived groundwater storage anomalies indicate interannual variability across the Mandera sub-basin, with interannual variability driven by episodic rainfall events. Although Mann-Kendall trend analysis reveals a weak positive trend (τ = 0.134, p = 0.398), it is not statistically significant and likely reflects short-term climatic variability rather than long-term aquifer recovery. The study demonstrates that integrating GIS-based recharge modeling with satellite observations provides a dependable, transferable framework for assessing groundwater variability in data-scarce arid regions, supporting evidence-based water management and planning.