Author(s): B. C. Kumar Raju, Nandagiri Lakshman
Linked Author(s): Kumar Raju B C
Keywords: Hydrological modeling, uncertainty analysis, curve number (CN), variable source area (VSA), SWAT
Abstract: Assessing uncertainty associated with distributed Variable Source Area (VSA) hydrological models are essential for accurate simulation of stream flows of the humid tropical watersheds. The present study is taken up to assess the uncertainties associated with the Soil and Water Assessment Tool � VSA (SWAT-VSA) model using Generalized Likelihood Uncertainty Estimation (GLUE) technique. The study is carried out in Harangi watershed (538. 8 km2) located in the humid region of Karnataka State, India. The goodness-of-fit and efficiency of the model have been tested using the Nash-Sutcliffe efficiency (ENS) as the objective function. Sensitivity analysis indicated that model parameters CN2, Alpha_Bf and Sol_Awc are important for simulating uncertainty associated with stream flows. The P-factor, which is the percentage of observed data bracketed by the 95% prediction uncertainty (95PPU), was 43% and 48% during calibration and validation periods. The results indicated that the GLUE technique applied on the SWAT-VSA model performed well in quantifying uncertainties in stream flow estimates at the outlet of the Harangi watershed
Year: 2017