Soil Survey Geographic Database (SSURGO) data are available for the entire United States, so are incorporated in many regional and national models of hydrology and environmental management. However, SSURGO does not provide an understanding of spatial variability and only includes saturated hydraulic conductivity (K_sat) values estimated from particle size analysis (PSA). This study showed model sensitivity to the substitution of SSURGO data with locally described soil properties or alternate methods of measuring K_sat. Incorporation of these different soil data sets significantly changed the results of hydrologic modeling as a consequence of the amount of space available to store soil water and how this soil water is moved downslope. Locally described soil profiles indicated a difference in K_sat when measured in the field vs. being estimated from PSA. This, in turn, caused a difference in which soil layers were incorporated in the hydrologic simulations using TOPMODEL, ultimately affecting how soil water storage was simulated. Simulations of free-flowing soil water, the amount of water traveling through pores too large to retain water against gravity, were compared with field observations of water in wells at five slope positions along a catena. Comparison of the simulated data with the observed data showed that the ability to model the range of conditions observed in the field varied as a function of three soil data sets (SSURGO and local field descriptions using PSA-derived K_sat or field-measured K_sat) and that comparison of absolute values of soil water storage are not valid if different characterizations of soil properties are used.