Impact of subwatershed partitioning on modeled source-and transport-limited sediment yields in an agricultural nonpoint source pollution model

Distributed parameter hydrologic models have a potential use as tools for supporting watershed management policy. However, proper model implementation will require an understanding of how to integrate data collection and models in watersheds that have differing characteristics. This study investigated the behavior of one such model, the Soil and Water Assessment Tool (SWAT), in relation to the level of spatial aggregation of input data and the extent to which the watershed is source- or transport-limited in terms of sediment yield. The approach here was to first test SWAT in an agricultural watershed in Southern Wisconsin. Then, a series of computer model applications were generated with a range of sediment source- and transport-limited conditions. SWAT was run for each watershed condition using eight watershed delineations, each with a different number of subwatersheds. Data aggregation affected model behavior differently depending on whether the watershed was sediment source-limited or transport-limited. This indicates that care is needed in selecting distributed sampling points, characterizing stream channel processes, and improving the selection of subwatershed sizes to match SWAT to watersheds with different characteristics.