Soil erosion is one of the most important problems in the world and one of the main factors effecting ecosystems. Soil erosion in Iran is high. According to the available information 59 percent of the 17 basins that have been studied in Iran have been severely degraded. Erosion-prediction technology is a powerful tool used for more than half a century in policy development, erosion inventories, conservation planning and engineering design. Spatially distributed hydrological models are useful tools to support the design and evaluation of water management plans. The objective of this study was to use SWAT2000 (Soil Water Assessment Toll) to evaluate runoff and sediment estimation in Vanak. Vanak sub-basin (3198 km 2 ), located in southeast of north Karun watershed. SWAT is a conceptual, physically based long-term continuous watershed scale simulation model that operates on a daily time step. This model was developed to predict the impact of land management practices on water, sediment and agricultural chemical yields in large complex watersheds with varying soils, land use and management conditions over long periods of time. Model calibration and uncertainty analysis were performed with SUFI-2, which was interfaced with SWAT using the generic SWAT-SAI program. SUFI-2 is a semi-automated inverse modeling procedure for a combined calibration-uncertainty analysis. In this study, runoff and sediment data of four hydrometery stations Aghblagh, Kasgan, Soolgan and Vanak were used for calibration and validation model. Five factors were used to model evaluation, include percent differences factor, R 2 coefficient, Nash-Sutcliffe coefficient, d-factor and p-factor. First, the model considers basin, dam and springs was run, but because inaccuracy and discontinuity data, in runoff simulation, they were omitted. Data from period of 1989-2004 were used. Model calibration and validation were performed with ¾ and ¼ data, respectively. After calibration, model was used to simulate monthly runoff in Kasgan, Soolgan and Vanak stations satisfactory. R 2 coefficient, Nash-Sutcliffe coefficient, d-factor and p-factor were 0.68, 0.41, 1.14 and 0.78, for Aghblagh, 0.6, 0.58, 1.16 and 0.71 for Kasgan, 0.7, 0.64, 0.97 and 0.79 for Soolgan, and 0.85, 0.76, 0.77 and 0.8, and for Vanak, respectively. These factors after monthly runoff validation were 0.72, 0.7, 0.95 and 0.81 for Aghblagh, 0.7, -0.1, 2.04 and 0.79 for Kasgan, 0.84, 0.75, 1.12 and 0.77 for Soolgan and 0.87, 0.78, 0.86 and 0.89 for Vanak, respectively. These values for Aghblagh, Soolgan and Vanak have acceptable accuracy. Monthly runoff simulati
