Groundwater recharge estimation in data-limited water-scarce regions

ABSTRACT Many groundwater recharge estimation methods require extensive data, costly equipment, sensors, and human effort. These activities include monitoring groundwater level changes, water balance assessments, or isotopic tracers. However, in many regions, the high cost of data collection makes these methods infeasible. This study presents an observation-constrained LSM to estimate groundwater recharge for a large, data-limited water scarcity region in the Rift Valley basin in Ethiopia. This model leverages publicly available data from the National Aeronautics and Space Administration (NASA) GLDAS Noah model, spanning the years 2000 to 2022, to provide a cost-effective means for groundwater recharge estimation for larger regions. We conduct a comprehensive analysis of groundwater recharge, precipitation, land cover, and land use while integrating both temporal and spatial variability. Validating this estimation is challenging due to data limitations, so we compare our findings with literature on rainfall conversion to groundwater recharge in similar regions. Our method estimates that only 5% of rainfall is converted to groundwater recharge in the study area. This value is consistent with reported estimates in similar regions. This framework could make groundwater recharge estimation feasible in data and water-limited regions around the world.