Estimation of spatially distributed groundwater recharge in data-scarce regions

Abstract

Study region: Upper Beles Basin, Ethiopia Focus: Data limitations significantly challenge accurate groundwater recharge estimation, especially in regions with sparse gauging stations and highly variable hydro-meteorological conditions, such as the Upper Beles Basin, Ethiopia. Using interpolated data often yields unreliable results. This study evaluates the use of remote sensing-based hydro-meteorological data to estimate spatially distributed groundwater recharge in such data-scarce regions, comparing it with point estimates derived from primary field data, using a case study of Upper Beles Basin. New hydrological insights for the region: The study employed TerraClimate and CHIRPS datasets to estimate spatially distributed recharge using the WetSpass model. Groundwater recharge was also estimated using the Water Table Fluctuation (WTF) and Chloride Mass Balance (CMB) methods, with data from 21 monitoring wells and 45 water samples. The estimated average annual recharge was 420 mm (WTF), 308 mm (CMB), and 365 mm (WetSpass). The substantial variability in point estimates reflects the basin's recharge heterogeneity, indicating the risks of relying solely on point data. A strong correlation (72 %) was found between the WTF-derived point estimates and WetSpass-generated values. Recharge variability is influenced by land use in the lowlands and by slope, soil, and rainfall in the highlands. This research demonstrates that remote sensing-based data can yield more reliable recharge estimates in regions with sparse gauging stations and highly variable hydro-meteorological conditions.

Keywords

• Spatial Groundwater recharge
• Global hydro-meteorological data
• WetSpass
• WTF
• CMB