The present study was designed to model the adsorption of geosmin from water under laboratory conditions using the Freundlich isotherm model. This model was used to compare the efficiency of sugarcane bagasse and pecan shell-based powdered activated carbon to the efficiency of a coal-based commercial activated carbon (Calgon Filtrasorb 400). When data were generated from Freundlich isotherms, Calgon Filtrasorb 400 had greater geosmin adsorption at all geosmin concentrations studied than the laboratory produced steam-activated pecan shell carbon, steam-activated bagasse carbon, and the CO(-)activated pecan shell carbon. At geosmin concentrations < 0.07 mug/l for the phosphoric acid-activated pecan shell carbon and below 0.08 mug/l for a commercially produced steam-activated pecan shell carbon obtained from Scientific Carbons, these two carbons had a higher calculated geosmin adsorption than Filtrasorb 400. While the commercial carbon was more efficient than some laboratory prepared carbons at most geosmin concentrations, the results indicate that when the amount of geosmin was below the threshold level of human taste (about 0.1 mug/l), the phosphoric acid-activated pecan shell carbon and the Scientific Carbons sample were more efficient than Filtrasorb 400 at geosmin removal.