|BIEGER, K - Texas A&M University|
|RATHJENS, H - Purdue University|
|VOLK, M - Helmholtz Centre For Environmental Research|
|Bosch, David - Dave|
|CHAUBEY, I - Purdue University|
|ALLEN, P - Baylor University|
|SRINIVASAN, R - Texas A&M University|
Submitted to: ASABE Annual International Meeting
Publication Type: Abstract Only
Publication Acceptance Date: 3/8/2015
Publication Date: 7/26/2015
Citation: Bieger, K., Rathjens, H., Volk, M., Bosch, D.D., Chaubey, I., Allen, P.M., Srinivasan, R., Arnold, J.G. 2015. Application of the new modular SWAT code to three watersheds in the United States. [abstract] American Society of Agricultural and Biological Engineers Annual International Meeting, New Orleans, Louisiana, July 26-29, 2015..
Technical Abstract: Due to the integration of new variables and subroutines over the past two decades since the release of the first version of the Soil and Water Assessment Tool (SWAT), the model has become difficult to manipulate and maintain. Therefore, the SWAT code and the input and output file structure have recently been completely revised. The new modular SWAT code simplifies the maintenance of the code, the integration of new variables, and the sharing of data. Also, it gives the user more flexibility in defining the spatial interactions within a watershed, since hydrologic response units, subbasins, aquifers, reservoirs, and reaches are treated as separate spatial objects. Thereby, it facilitates the implementation of landscape units to improve the spatial representation of hydrological processes in the watershed. Three watersheds in the United States were used to test and debug the new code and compare model outputs from the old and the new SWAT versions: the Little River Experimental Watershed (GA), the South Fork Wildcat Creek Watershed (IN) and the Middle Bosque River Watershed (TX). First, two models with the same subbasin delineation and the same number of Hydrologic Response Units were constructed for each watershed, one using the current release code, the other one using the new modular code. Results indicate that the new SWAT version performs equally well as the old one in simulating the water balance, stream flow and sediment and nutrient transport in all three watersheds. Second, the subbasin delineation in all three watersheds was modified to distinguish between floodplain and upland areas, as these landscape units differ considerably with regard to their hydrological function in the watershed. The delineation of the landscape units was done using a DEM-based slope position method. The implementation of two landscape units in the model setup results in an improved representation of routing and transport processes across the landscape and of deposition of sediments and nutrients in the floodplain. Also, the interactions between river and floodplain can be simulated more realistically.