Abstract:The RUNOFF block of EPA's storm water management model (SWMM) was used to simulate the quantity and quality of urban storm water runo from four relatively small sites (i.e. 5 . 97±23 . 56 ha) in South Florida, each with a speci®c predominant land use (i.e. low density residential, high density residential, highway and commercial). The objectives of the study were to test the applicability of this model in small subtropical urban catchments and provide modellers with a way to select appropriate input parameters to be used in planning studies. A total of 58 storm events, measured by the US Geological Survey (USGS), provided hyetographs, hydrographs and pollutant loadings for biological oxygen demand (BOD 5 ), total suspended solids (TSS), total Kjeldahl nitrogen (TKN) and lead (Pb), and were used for calibration of the model. Several other catchment characteristics, also measured or estimated by USGS, were used in model input preparation. Application of the model was done using the Green±Ampt equation for in®ltration loss computation, a pollutant accumulation equation using a power build-up equation dependent on the number of dry days, and a power wash-o equation dependent on the predicted runo rate. Calibrated quantity input parameters are presented and compared with suggested values in the literature. The impervious depression storage was generally found to be the most sensitive calibration parameter, followed by the Manning's roughness coecients of conduit and overland¯ow, the Green±Ampt in®ltration parameters and, ®nally, the pervious depression storage. Calibrated quality input parameters are presented in the form of regression equations, as a function of rainfall depth and the number of antecedent dry days. A total of 16 independent rainfall events were used for veri®cation of the model, which showed a good comparison with observed data for both hydrographs and pollutant loadings.