Estimating exploitable groundwater for agricultural use under environmental flow constraints using an integrated SWAT-MODFLOW model

Traditionally, the impact of groundwater abstraction on streamflow and aquifers has been assessed primarily in terms of water availability and depletion, often neglecting critical environmental flow (Eflow) requirements. To address this gap, our study employed an Eflow perspective to evaluate exploitable groundwater for agricultural use at a watershed scale using an integrated modeling approach. We used the SWAT-MODFLOW model, integrating the Soil and Water Assessment Tool (SWAT) and the Modular Finite-Difference Flow Model (MODFLOW), to analyze water balance segments and stream-aquifer interactions in the study region. Eflow for different management classes was determined using the flow duration curve (FDC) approach. The FDC method classifies Environmental Flow Management Classes (EMC) from “A” (natural or minor alteration) to “F” (critically modified). The relationship between the selected EMC and simulated streamflow, under the pressure of agricultural groundwater pumping, served as a constraint in evaluating exploitable groundwater alongside groundwater level fluctuations. As the traditional approach depends on average recharge for determining exploitable groundwater, this study also assessed the spatiotemporal distribution of groundwater recharge. Simulation periods were aligned with agricultural pumping schedules. Results indicated that groundwater pumping significantly impacts streamflow at the start of the irrigation period. Under moderately affected EMC and with an average groundwater level fluctuation of one meter, the exploitable groundwater during the irrigation period can reach up to 2.96 Mm3/day. With this amount of pumping, the average groundwater discharge to streams declined by around 13 %. While groundwater level fluctuations near the watershed outlet were minimal, in the upstream area, levels could decrease by as much as 2.5 m. The groundwater level fluctuation underscores the necessity of accounting for both spatial and temporal factors, particularly in highly affected regions.