Today most hydrologic models use geographic information systems (GIS) to assist in the parameterization process. Increasingly, GIS-based applications are being developed that automate the parameterization process of a hydrologic model that allow users to acquire spatial data from an array of sources, delineate and parameterize a watershed, and execute a simulation through a few clicks. Users are becoming disconnect from the model parameterization process and lack the understanding of the input data that was once required. Spatial data for a given model input may be available at different resolutions, with each resolution potentially yielding a different model result. This paper will present the results of a study investigating the effect of input data resolution of soils, topography and land cover on the simulation of runoff from semiarid watersheds. The study was conducted on the USDA Agricultural Research Service (ARS)s Walnut Gulch Experimental Watershed (WGEW) located near Tombstone, Arizona. WGEW has an extensive archive of spatial data that was used in this study. The study used the Automated Geospatial Watershed Assessment (AGWA) tool developed in a cooperative research effort between the ARS, EPA and University of Arizona. AGWA is a GIS-based suite of programs that uses topography, soils, and classified remote sensing imagery to estimate parameters for KINEROS, a physically-based distributed hydrologic model. AGWA/KINEROS was used to simulate the rainfall/runoff response from four watersheds ranging in size from 827 to 11,371 ha. The effect on simulation results from two soil resolutions (SURRGO and STATSGO) and five digital elevation models (2.5m-SAR, 10m-SAR, 10m-USGS, 30m-USGS, and 90m-USGS was assessed for the four watershed scales. The paper will present effect of resolution on runoff volume and peak discharge for 29 rainfall events.