A modified SWAT module for estimating groundwater table at Lethbridge and Barons, Alberta, Canada

Abstract

Groundwater is a priceless resource in Alberta and therefore, estimating groundwater are crucial to identifying and promoting holistic and integrated management of groundwater-surface water. However, it is a challenge to simulate groundwater storage due to the current rudimentary representation of two-way groundwater-surface water exchange in current hydrologic models, such as Soil and Water Assessment Tool (SWAT), which, in turn, limits our ability to predict land-atmosphere processes and groundwater storage. In this study, we modified the SWAT model to improve module of evapotranspiration in two-way groundwater-surface water exchange. The modified SWAT was calibrated and validated against the groundwater table height and evapotranspiration from 2008 to 2011 period at two location (Lethbridge and Barons) Alberta, Canada. The results showed that the modified SWAT model predicts the groundwater table height very well at both locations. The modified model predicted the daily groundwater table height with R2 values of 0.86 and 0.89 in the calibration period (2008–2009), 0.81 and 0.83 for the validation period (2010–2011) at Lethbridge and Barons, respectively. The Nash-Sutcliffe model efficiency (NSE) for daily groundwater table height was 0.69 and 0.71 during calibration periods (2008–2009) while the model gives lower values of NSE 0.65 and 0.67 for validation periods (2010–2011) at Lethbridge and Barons, respectively. Similarly, the model estimates evapotranspiration well with correlation coefficient (R2) of 0.77 during the calibration period and 0.81 for validation period. Our result showed that the modified SWAT model did improve estimates to dynamic groundwater table heights.