Riparian buffer strips are managed for the enhancement of water quality through control of non point source pollution. Denitrification in riparian buffer strips is thought to be the major process -with nitrate uptake by plant growth- that reduces nitrate input in surface water. We investigated the Denitrifier Enzyme Activity (DEA) to test how the buffer strip management modifies the denitrification process. The experimental site is composed of a crop field and a 11 m wide grassed buffer strip at the border between the crop field and a tributary to the Attert river, South-East Belgium. Soil samples from the crop field and the buffer strip have been submitted to different imposed conditions combining glucose, nitrate and water saturation to investigate how antecedent water regime, nitrate and carbon content affect denitrification. The work included DEA measurements on undisturbed soil cores freshly sampled. Experiments on undisturbed soil cores identify the buffer strip as more effective in denitrification (p < 0.001) than the cropped field. However, experiments on samples conditioned under imposed carbon and/or nitrate contents emphasised the importance of micro-topography.