Whitfield, C. J. and Reid, C. 2013. Predicting surface area of coarse-textured soils: Implications for weathering rates. Can. J. Soil Sci. 93: 621-630. The surface area of soil is an important determinant of mineral weathering rates, but is infrequently measured. Simple texture-based pedotransfer functions (PTFs) have been used to predict the specific surface area (SSA) of coarse-textured soils. Detailed physicochemical properties of 40 upland forest mineral soils from northeastern Alberta were used to evaluate three texture-based PTFs and to calculate weathering rates using a process-oriented soil-chemical model. Evaluation of the PTFs demonstrated that these equations predict only across a limited range of (low) surface areas. Moreover, the fit between predicted and measured SSA was generally poor for soils in this region of Alberta. Improved prediction of SSA was possible using a texture-based PTF calibrated for the region, although differences between measured and predicted values were often large. Mineralogy terms were used in a more comprehensive PTF to account for mineral-specific differences in surface area. This approach proved superior to texture-only approaches; however, it could not be used reliably for site-specific predictions (NRMSE=0.41). Soil-chemical model-generated weathering rates were strongly influenced by the SSA method used in parameterization; weathering estimates and corresponding critical load assessments based on measured SSA (and to a lesser extent SSA derived from the regional PTF) were the most robust. Methods for SSA prediction should be used with caution, particularly in cases where they are applied to soils with different character than those for which they were developed.