Although many studies have measured denitrification in stream sediments, few have utilized these data with local water column and sediment measurements to develop a predictive model for NO3- uptake. In this study, sediment denitrification was measured from cores in five streams under various land uses in south-central Minnesota using denitrification enzyme activity (DEA) assays and amplification of the nirS gene via real-time, quantitative polymerase chain reaction. Hydraulic and environmental variables were measured in the vicinity of the sediment cores to evaluate the influence of fluid flow and chemical variables on denitrification activity. Potential denitrification rates measured using DEA assays ranged from 0.02 to 10.1 mg N m-2 h-1, and the abundance of the denitrifier gene nirS was positively correlated with these measurements (R2 = 0.79, P < 0.001) for most of the streams studied. A predictive model to determine NO3- uptake via denitrification was derived, implementing dimensional analysis of variables that mediate denitrification in sand-bed streams. The proposed model explained 75% of the variability in DEA rates. The results of this study show that denitrification is most dependent on the distribution of sediment organic matter, interstitial pore space, and stream hydraulic characteristics, including shear velocity at the sediment-water interface and stream depth.