The link between the Penn State/NCAR Mesoscale Meteorological Model (MM5) and the Hydrologic Model System (HMS) was implemented to simulate the river-basin response to single-storm events. Three MM5-domain setups, a single domain (36 km grid increment), one nest (36+12 km grid increments), and two nests (36+12+4 km grid increments) were used to simulate two single-storm events from April 1986 and May 1988 over the Susquehanna River basin (SRB). The MM5-simulated precipitation at various resolutions and the analyzed observed precipitation were used to drive the hydrologic simulation of direct surface runoff for the SRB and two subbasins (the upper west branch and the upper north branch) and of streamflow for the upper west branch. The streamflow simulation driven with the observed precipitation was first calibrated to the observed streamflow. The simulated streamflow with the observed precipitation and the MM5-simulated precipitation at various resolutions were compared to the observed streamflow. The discrepancies in modeling the streamflow response and the direct surface runoff with the MM5-simulated precipitation at various resolutions were evaluated. The simulated results improved as the resolution of the MM5-simulated precipitation increased. The MM5 simulations for all three domain setups captured the general pattern of the storm systems. On the basis of the quality of the simulated results and the required computing time, the MM5 setup with two nests appears to be the most appropriate to provide the simulated precipitation for the hydrologic simulation of single-storm events over the SRB. ¿ 1999 American Geophysical Union