We evaluated nitrogen (N) removal efficiency by ri-parian buffers at 14 sites scattered throughout seven European countries subject to a wide range of climatic conditions. The sites also had a wide range of nitrate inputs, soil characteristics, and vegetation types. Dissolved forms of N in groundwater and associated hydrological parameters were measured at all sites; these data were used to calculate nitrate removal by the riparian buffers. Nitrate removal rates (expressed as the difference between the input and output nitrate concentration in relation to the width of the riparian zone) were mainly positive, ranging from 5% m 1 to 30% m 1 , except for a few sites where the values were close to zero. Average N removal rates were similar for herbaceous (4.43% m 1) and forested (4.21% m 1) sites. Nitrogen removal efficiency was not affected by climatic variation between sites, and no significant seasonal pattern was detected. When nitrate inputs were low, a very large range of nitrate removal efficiencies was found both in the forested and in the nonforested sites. However, sites receiving nitrate inputs above 5 mg N L 1 showed an exponential negative decay of nitrate removal efficiency (nitrate removal efficiency 33.6 e 0.11 NO 3 input , r 2 0.33, P 0.001). Hydraulic gradient was also negatively related to nitrate removal (r 0.27, P 0.05) at these sites. On the basis of this intersite comparison, we conclude that the removal of nitrate by biological mechanisms (for example, denitrification, plant uptake) in the riparian areas is related more closely to nitrate load and hydraulic gradient than to climatic parameters.