The amount and distribution of impervious surfaces are important input parameters of hydrological models, especially in highly urbanized basins. This study tests three different methods to input impervious surface area information to a semi-distributed hydrological model in order to examine their effects on storm flow. The three methods being evaluated include: (1) a constant value for impervious surfaces in the entire urban area, (2) constant values of imperviousness for commercial and residential land uses, respectively, and (3) different imperviousness for the residential land use in each subbasin. Storm flow of the Milwaukee River Basin in southeastern Wisconsin (USA) was modeled using the Hydrological Simulation Program-Fortran. The results show that the three methods resulted in substantially different amounts of storm flow. The storm flow simulated with the third method was the largest and had the largest variability between the subbasins. The differences between the scenarios are generally larger in subbasins with high percentage of urban land use types. The results suggest that the effect of different input methods is amplified in urbanized subbasins and the spatial variability of imperviousness should be commensurate with the spatial variability of the model configuration.
Keywords: Hydrological model; Impervious surface; Runoff; Storm flow; Urban land use.