KangshengWu_SWAT_ForestedWatershed.pdfVIP

HYDROLOGICAL PROCESSES Hydrol. Process. 22, 1431–1442 (2008) Published online 13 August 2007 in Wiley InterScience () DOI: 10.1002/hyp.6695 Hydrologic comparison between a forested and a wetland/lake dominated watershed using SWAT Kangsheng Wu* and Carol A. Johnston Center for Biocomplexity Studies, Box 2207B, South Dakota State University, Brookings, South Dakota 57007, US Abstract: The Soil and Water Assessment Tool (SWAT) is a physically-based hydrologic model developed for agricultural watersheds, which has been infrequently validated for forested watersheds, particularly those with deep overwinter snow accumulation and abundant lakes and wetlands. The goal of this study was to determine the applicability of SWAT for modelling streamflow in two watersheds of the Ontonagon River basin of northern Michigan which differ in proportion of wetland and lake area. The forest-dominated East Branch watershed contains 17% wetland and lake area, whereas the wetland/lake-dominated Middle Branch watershed contains 26% wetland and lake area. The specific objectives were to: (1) calibrate and validate SWAT models for the East Branch and Middle Branch watersheds to simulate monthly stream flow, and (2) compare the effects of wetland and lake abundance on the magnitude and timing of streamflow. Model calibration and validation was satisfactory, as determined by deviation of discharge D and Nash and Sutcliffe coefficient values E that compared simulated monthly mean discharge versus measured monthly mean discharge. Streamflow simulation discrepancies occurred during summer and fall months and dry years. Several snow melting parameters were found to be critical for the SWAT simulation: TIMP (snow temperature lag factor) and SMFMX and SMFMN (melting factors). Snow melting parameters were not transferable between adjacent watersheds. Differences in seasonal pattern of long-term monthly streamflow were found, with the forest-dominated watershed having a higher peak flow during April b