Distributed hydrological models are widely used to simulate the runoff process and to evaluate the hydrologic effect of
various management scenarios, It is not easy to immediately identify the similarities and differences between the
different models, few studies have compared the results of different models when applied to the same catchments. Two
hydrological models, Three-layer Variable Infiltration Capacity (VIC-3L) and Soil Water Assessment Tool (SWAT) are
compared in their abilities to simulate runoff in Weihe River Basin in China over a six-year validation period.VIC-3L is
a macro-scale hydrologic model based on soil-vegetation-atmosphere transfer scheme (SVATS), which is designed to
describe the land surface in numerical weather prediction and climate. SWAT model is a complex, conceptual,
hydrologic, semi-distributed model with spatially explicit parameterization; it is a continuous time model that operates
on a daily time step over long time periods.
Two simulated results clear show minimal differences between two models throughout the validation period, although
two models have some differences in structure and runoff mechanism. The runoff simulated by two models in spring and
winter is underestimated than observed stream flow. The simulated runoff in summer by SWAT is bigger than that by
VIC-3L, but is smaller in winter, the simulate runoff by VIC-3L is more flat than that by SWAT.
An important difference between model results is the ability to simulate the response of runoff to LUCC, SWAT is a
watershed based hydrological model, and it's more evident and accurate to study the response of runoff to LUCC than
VIC-3L. The simulated runoff under the land use type in 2000 decreases 6.41% than that in 1986. The runoff has
decreased by 15.18% in dry periods and shows an evidently decreasing tendency.