In the paper that is the foundation for this study, VanderKwaak and Loague (2001. Water Resources Research37: 999–1013) reported a demonstration of a fully coupled comprehensive physics-based hydrologic-response model, InHM (Integrated Hydrology Model), for two rainfall-runoff events from the small rangeland catchment known as R-5. The InHM simulations reported herein address (in three phases) limitations in the VanderKwaak and Loague (2001. Water Resources Research37: 999–1013) simulations. In Phase I, a new finite-element mesh was selected to represent R-5. In Phase II, with the new mesh in place, evaporation was considered for the R-5 events. In Phase III, with the new mesh in place and evaporation considered, the geology of R-5 was approximated. Each phase, compared with the results reported by VanderKwaak and Loague (2001. Water Resources Research37: 999–1013), shows a change in the simulated near-surface response. The performance of InHM for 15 R-5 events is also reported herein. The results from two stages of model calibration are presented. The uncertainty in initial soil-water content estimates for event-based simulation is shown to be a major limitation for physics-based models. The performance of InHM, relative to past event-based simulation efforts with a quasi-physically based rainfall-runoff model, is better for both peak stormflow and the time to peak stormflow, but worse for stormflow depth. The InHM simulations reported here set the stage for continuous simulation of near-surface response for the R-5 catchment with InHM. Copyright © 2004 John Wiley & Sons, Ltd.