摘要
Winter soil-temperature simulations with soil and water assessment tool (SWAT) are generally inaccurate in snow-dominated areas of the word, because the existing empirically-based soil-temperature module in SWAT does not account for snow-insulation effects. This problem, in turn, leads to all subsequent projections including water flow and nutrient loading biased. To address this issue, a physically-based soil-temperature module was developed and incorporated into SWAT as an alternative to the empirical formulation, which was found generate better overall estimates of winter soil temperature in a previous study. In the current study, we continue to examine the modified version of SWAT performance on watershed modeling by making pairwise comparisons between the outputs from the original and modified versions of SWAT against related field measurements in a testing watershed. Results show that the physically-based soil-temperature formulation helps to dramatically improve SWAT estimations of base flow discharge and NO3-N loading for the watershed as a result of improvement in modeled winter soil temperatures. Compared with the original version of SWAT, the new version of SWAT predicted overall lower surface runoff and soil moisture content, as well as higher percolation and lateral flow in winter, resulting in clear differences in modeled flow paths and fate of chemical pollutants. Adding a physically-based treatment of soil temperature in SWAT is an important generalization that has potential of making the model more relevant to snow-dominated areas of the world.