Comparative Modeling of Nitrogen Losses in a Tile-Drained Watershed Using SWAT Model: Uncertainty and Calibration Considerations

Junyu Qi^1*Robert Malone²Kang Liang^3,4Kevin Cole²Bryan Emmett²Daniel Moriasi²Muhammad Rizwan Shahid⁵Michael Castellano⁵

• ¹University of Maryland, College Park, College Park, United States
• ²USDA-ARS National Laboratory for Agriculture and the Environment, Ames, United States
• ³USDA-ARS Beltsville Agricultural Research Center, Beltsville, United States
• ⁴Oak Ridge Institute for Science and Education, Oak Ridge, United States
• ⁵Iowa State University of Science and Technology, Ames, United States

Ecohydrological models are critical for understanding coupled hydrologic– biogeochemical processes in tile-drained watersheds and for assessing management options. Despite recent advances in SWAT's hydrological and biogeochemical processes, there has been limited evaluation of both the original and new tile drainage and nitrogen (N) modules. We therefore applied a comparative modeling approach in a typical Midwestern tile-drained watershed, evaluating eight configurations that vary tile-drainage module (original/new), tile parameter treatment (calibrated/default), and N module (original/new) to assess performance for N-loss simulation. Using daily streamflow and nitrate (NO3⁻) load records from three monitoring sites, we conducted calibration, validation, sensitivity analysis, and uncertainty assessment. Each configuration effectively reproduced daily and monthly dynamics, although high-flow and associated NO3⁻ load peaks were underestimated. We found that the new tile module generally improved streamflow simulations, particularly under tile parameter calibration conditions, while the new N module consistently enhanced NO3⁻ load simulations compared to the original module. Despite improvements in streamflow and NO3⁻ loads with the new tile and N modules, the additional processes in the new N module can magnify uncertainty in N-gas-flux estimates when calibration observations are scarce. We recommend applying the new N module in conjunction with additional measurements—such as soil moisture and nitrous oxide (N2O) fluxes—to constrain better N gas flux estimates beyond outlet NO3⁻ load data. If such observations are unavailable, careful calibration with reasonable estimates may still help constrain soil N cycling and improve overall N budget accuracy.