Comparison of SWAT with HSPF Model in Predicting Hydrologic Processes of a Small Multivegetated Watershed

Abstract

Information on similarities and differences between models is essential for the users to select the most suitable model for monitoring the hydrologic processes on watershed basis, as well as for developing the soil and water management strategies. We have examined the relative predictability of Soil and Water Assessment Tool (SWAT) and Hydrologic Simulation Programme-Fortran (HSPF) models in response to seasonal variation of rainfall by comparing the simulated monthly runoff and sediment yield with their measured counterparts in a small multivegetated watershed named Banha, DVC, Hazaribagh, Jharkhand, India, for six years (1996-2001) under sub-humid sub-tropical climate. The results indicated that the monthly runoff and sediment yield simulated by SWAT were closer to the measured values compared to those simulated by HSPF in most months of all the years studied, particularly, for low ranges. However, for high ranges, HSPF prediction was relatively closer to the measured values. Statistical tests performed on the whole range of runoff and sediment yield showed high values of Coefficient of determination and Nash-Sutcliffe simulation efficiencies as well as lower values of Root mean square error for SWAT than HSPF simulations. Based on the statistical test results, SWAT was adjudged superior to HSPF for the watershed studied. Better predictions of SWAT than HSPF further revealed that the saturation-excess overland flow mechanism of SWAT was superior to the infiltration-excess mechanism of HSPF for the simulation of runoff. Similarly, the modified universal soil loss equation (MUSLE) of SWAT proved better than the kinetic energy based erosion equation of HSPF in estimating sediment yield of the studied watershed.