Estimating Basin Recharge and Pumping Rates Using Hydrological-Groundwater Coupled Model (SWAT-MODFLOW) - A Case Study of the Zhuoshui River Alluvial Fan in Taiwan

Abstract

The interaction between surface water and groundwater is a crucial factor in assessing regional water resource utilization. As population growth and urban development escalate, the effective allocation of surface and groundwater sources has become a pressing concern. This study employs a coupled model that combines surface water and groundwater simulations, SWAT-MODFLOW, for estimating groundwater recharge and pumping rates in the expansive alluvial fan of the Zhuoshui River Basin in central Taiwan (4,995.76 km2). Utilizing the SWAT model, we derive key parameters related to groundwater recharge, such as rainfall-runoff, recharge rates, and river water levels. These parameters are integrated into the MODFLOW model for the alluvial fan, facilitating a comprehensive understanding of the complex dynamics. To address uncertainties arising from insufficient well pumping data, further calibration of the MODFLOW model is undertaken. The simulation spans from 2002 to 2021, with 2002-2011 and 2012-2020 designed as calibration and validation periods, respectively. Our daily flow simulations demonstrate the SWAT model's effectiveness in accurately capturing hydrological processes in the basin (R2 = 0.59-0.80, NSE=0.43-0.77, PBIAS=-49-55.9%). Through the linkage of the groundwater model and daily-scale simulations, we estimate monthly groundwater levels, revealing an annual recharge of approximately 214 million m3 and groundwater pumping rates ranging from 127 to 186 million m3 within the basin. These findings provide valuable insights for regional water resource management strategies and serve as a reference for decision-makers grappling with the challenges of regional water resource allocation within the context of the Zhuoshui River Basin.