Radar-based precipitation estimates and nowcasts can serve as input to distributed hydrological models to issue flood and flash-flood forecasts in mountain watersheds at various lead times. Improvements in flood predictability across catchment scales are expected as the quality of the radar-based observations and forecasts is enhanced. This study is oriented towards assessing the effects of different radar-based precipitation products on the flood forecasting skill in the Colorado Front Range. We utilize NEXRAD Multisensor Precipitation Estimates (MPE, 1 hr and 4 km), NEXRAD Level II reflectivity (15 min and 1 km) and nowcasts issued from these products to determine the impacts of temporal and spatial resolution on the flood forecast skill. Our modelling study is performed using the TIN-based Real-time Integrated Basin Simulator (tRIBS), which is a distributed hydrological model that explicitly accounts for spatial variation in topography, surface characteristics, and meteorological forcing. Flood predictability is assessed as a function of the precipitation product through comparisons to discharge observations at the selected basin outlets in the Colorado Front Range during the summer of 2004. Several metrics such as the difference in time to peak, total response volume and peak discharge will be used to evaluate a basin response to the rainfall products in the hydrological model. In addition, an analysis of the propagation of the precipitation forcing to the spatial hydrological response will be conducted. This preliminary analysis will determine the differences in hydrologic response and forecast skill for a range of observed and predicted forcings in forested mountain catchments subject to summer convective events.