Modeling spatial and temporal groundwater recharge dynamics using the newly physically-based coupled model in Tana sub-basin, Ethiopia

Spatiotemporal evaluation of groundwater recharge is indispensable for sustainable groundwater resource and socioeconomic development of a region. However, estimation of spatiotemporal groundwater recharge using vigorous modeling approach that integrates surface water and rainfall recharge sources was not given much attention. The present paper aims to assess spatiotemporal dynamics of groundwater recharge in the Tana sub-basin employing SWAT + gwflow, a newly coupled surface water and groundwater flow model. The coupled model was calibrated using a parameter estimation tool using observed streamflow data until its performance was in the acceptable range. Trend analysis of groundwater recharge was also carried out using the modified Mann–Kendall method. Annual average volumetric recharge of 16.575, 68.786, 105.641, and 469.903 million cubic meters (MCM) were received in Megech, Ribb, Gumara, and Gilgel Abbay major watersheds of the Tana sub-basin from 1997 to 2015. This implied that the Tana sub-basin gained 661.904 MCM mean annual renewable water from these watersheds. The temporal fluctuation, trends, and magnitude of recharge depend mainly on the rainfall of the watersheds. The total simulated recharge in the main watersheds of the Tana sub-basin showed significant (p < 0.05) decreasing trend from 2003 to 2012. In terms of spatial distribution, recharge rates ranging from 0.011 to 121.053, 0.039 to 55.547, 0.143 to 66.826, and 0.0 to 165.842 m³/day in Megech, Ribb, Gumara, and Gilgel Abbay, respectively. The higher recharge rates covered the smallest area in each watershed and found along river corridors. Additionally, the long-term annual average groundwater head was evaluated and showed a strong match with observed data. Stakeholders may apply the findings of this study for more comprehensive water policy strategic plans that incorporate both surface and groundwater management.