The Ohio River alluvial aquifer near Carrollton, Ky., is an important water resource for the cities of Carrollton and Ghent, as well as for several industries in the area. The groundwater of the aquifer is the primary source of drinking water in the region and a highly valued natural resource that attracts various water-dependent industries because of its quantity and quality. This report evaluates the performance of a numerical model of the groundwater-flow system in the Ohio River alluvial aquifer near Carrollton, Ky., published by the U.S. Geological Survey in 1999. The original model simulated conditions in November 1995 and was updated to simulate groundwater conditions estimated for September 2010. The files from the calibrated steady-state model of November 1995 conditions were imported into MODFLOW-2005 to update the model to conditions in September 2010. The model input files modified as part of this update were the well and recharge files. The design of the updated model and other input files are the same as the original model. The ability of the updated model to match hydrologic conditions for September 2010 was evaluated by comparing water levels measured in wells to those computed by the model. Water-level measurements were available for 48 wells in September 2010. Overall, the updated model underestimated the water levels at 36 of the 48 measured wells. The average difference between measured water levels and model-computed water levels was 3.4 feet and the maximum difference was 10.9 feet. The root-mean-square error of the simulation was 4.45 for all 48 measured water levels. The updated steady-state model could be improved by introducing more accurate and site-specific estimates of selected field parameters, refined model geometry, and additional numerical methods. Collection of field data to better estimate hydraulic parameters, together with continued review of available data and information from area well operators, could provide the model with revised estimates of conductance values for the riverbed and valley wall, hydraulic conductivities for the model layer, and target water levels for future simulations. Additional model layers, a redesigned model grid, and revised boundary conditions could provide a better framework for more accurate simulations. Additional numerical methods would identify possible parameter estimates and determine parameter sensitivities.