Summary Forest managers need methods to evaluate the impacts of management at the watershed scale. The Water Erosion Prediction Project (WEPP) has the ability to model disturbed forested hillslopes, but has difficulty addressing some of the critical processes that are important at a watershed scale, including baseflow and water yield. In order to apply WEPP to forested watersheds, we developed and assessed new approaches for simulating streamflow and sediment transport from large watersheds using WEPP. We created specific algorithms to spatially distribute soil, climate, and management input files for all the subwatersheds within the basin. The model enhancements were tested on five geologically and climatically diverse watersheds in the Lake Tahoe basin, USA. The model was run with minimal calibration to assess WEPP’s ability as a physically-based model to predict streamflow and sediment delivery. The performance of the model was examined against 17 years of observed snow water equivalent depth, streamflow, and sediment load data. Only region-wide baseflow recession parameters related to the geology of the basin were calibrated with observed streamflow data. Close agreement between simulated and observed snow water equivalent, streamflow, and the distribution of fine ( 20 μm) sediments was achieved at each of the major watersheds located in the high-precipitation regions of the basin. Sediment load was adequately simulated in the drier watersheds; however, annual streamflow was overestimated. With the exception of the drier eastern region, the model demonstrated no loss in accuracy when applied without calibration to multiple watersheds across Lake Tahoe basin demonstrating the utility of the model as a management tool in gauged and ungauged basins.