Technical Abstract: Riparian ecosystems have the capacity to lower NO3- concentrations in groundwater entering from non-point agricultural sources. The microbial processes responsible for decreases in riparian groundwater NO3- concentrations in the Willamette Valley of Oregon are not well understood. Our objective was to determine if denitrification and/or dissimilatory NO3- reduction to NH4+ (DNRA) could explain decreases in groundwater NO3- moving from an established perennial ryegrass field into a mixed-herbaceous riparian area. In situ denitrification rates (DN) were not different between the riparian area (near-stream or near-cropping system) and cropping system the first year. In the second year, during the transition to a clover planting, DN was highest just inside of the riparian/cropping system border. Median denitrification enzyme activity (DEA) rates ranged from 29.5 to 44.6 mg N2O-N kg-1 d-1 for surface soils (0-15 cm) and 0.7 to 1.7 µg N2O-N kg-1 d-1 in the subsoil (135-150 cm). DEA rates were not different among the zones when dates were pooled and were most often correlated to soil moisture and NH4+. Nitrate additions to surface soils increased DEA rates, indicating a potential to denitrify additional NO3-. Based upon groundwater velocity estimates, NO3- (3.8 mg NO3--N L-1) entering the riparian surface soil could have been consumed in 0.2 to 7 m by denitrification and 0.03 to 1.0 m by DNRA. Denitrification rates measured in the subsoil could not explain the spatial decrease in NO3-. However, with the potentially slow movement of water in the subsoil, denitrification and DNRA (0 to 264 µg N kg-1 d-1) together could have completely consumed NO3- within 0.5 m of entering the riparian zone.