Nitate retention in riparian buffer strips is well documented in summer periods, but the potential of winter retention within these zones is poorly documented. Two sites, grass (Lolium perenne L.), and poplar (Populus italica)‐vegetated riparian strips, were investigated in southern England (River Leach). Groundwater flow was via subsurface pathways within the sites, NO−3 concentration gradients and loading rates were calculated over the winter period. Nitrate retention was found to be linearly dependent on load rate. Nitrate retention occurred at the edge of the riparian zone. This was most obvious in the poplar site where all hillslope‐derived NO−3 was absorbed within the first 5 m of flow within the riparian strip. When loading rates into the sites increased, NO−3 absorption migrated upslope from the riparian site. The poplar‐vegetated riparian zone was found to be more resilient (99% retention of NO−3) than the grass‐vegetated riparian zone (84% retention of NO−3) in the winter months. It is postulated that although vegetation has no active role in retaining NO−3 in the winter, above‐ground vegetative biomass does contribute C to the soil microbacterial biomass that is engaged in NO−3 reduction in the winter months, this accounted for the greater efficiency of the poplar vegetated site.

Soil saturation is known to be of crucial importance to denitrification and other nitrogen cycling processes within the riparian zone. Since denitrification potential generally increases towards the soil surface, water table elevation can control the degree to which nitrate reduction is optimised. Given their topographic location and sedimentary structure, most floodplains are characterised by high water tables. However, detailed field data on water table levels, hydraulic gradients and flow patterns within the riparian zone are generally lacking. This paper presents data collected as part of a pan-European study of nitrate buffer zones, the Nitrogen Control by Landscape Structures in Agricultural Environments project (NICOLAS). An identical experimental design was employed at each site, allowing riparian zone hydrology and nitrogen cycling processes to be explored across a wide range of temperate climates; only the hydrological data are discussed here. A grid of dipwells at 10-metre spacing was installed at each site and manual measurements made at least once a month for a minimum of one year. In addition, at least one dipwell in each grid was monitored continuously using a data logger. All the riparian zones studied displayed a clear annual cycle of water table elevation, although other factors seemed equally important in influencing the range of variation. Where the riparian zone was flat, the water level in the adjoining river or lake proved more significant in controlling water table levels within the riparian zone than was originally anticipated. q

Abstract:The broad purpose of the study described here was to assess the role of denitri®cation in riparian zones in ameliorating groundwater pollution through nitrate loss, and as a potential source of nitrous oxide to the atmosphere. A suitable riparian zone was identi®ed at Cuddesdon Mill on the River Thame¯oodplain near Oxford, England. Measurements were made of water and nitrate moving from arable land through the riparian zone and into the river. Techniques to measure denitri®cation were tested and applied, and the factors controlling denitri®cation measured. While there was considerable potential for denitri®cation at the site, this was not realized because much of the water moving o the farmland bypassed the riparian zone, entering the river directly via springs or through gravel lenses beneath the¯oodplain soil. Management of this site would not reduce nitrate leaching unless the¯oodplain hydrology could be substantially modi®ed, and the main conclusion is that nitrate buer zones will only operate eciently where the hydrology of the site is appropriate.

Discharge of groundwater from a limestone aquifer through floodplain sediments is associated with a large decrease in the nitrate concentration of the water. Results are presented to show that only a small amount of this reduction is caused by dilution of groundwater by water already present within the floodplain sediments; most of the effect is an active reduction process, most probably biological denitrification. The nitrate reduction process appears to operate independently of surface vegetation type and tends to be focused in specific regions of the floodplain where sediments are anaerobic and carbon-rich. The results suggest that active denitrification can operate throughout the winter, when nitrate concentrations in groundwater are at their highest and that the process remains effective even during periods of maximum run-off. The results show that undrained floodplains can be used as buffer zones to protect surface waters from groundwater polluted with agriculturally derived nitrate.

The broad purpose of the study described here was to assess the role of denitrification in riparian zones in ameliorating groundwater pollution through nitrate loss, and as a potential source of nitrous oxide to the atmosphere. A suitable riparian zone was identified at Cuddesdon Mill on the River Thame floodplain near Oxford, England. Measurements were made of water and nitrate moving from arable land through the riparian zone and into the river. Techniques to measure denitrification were tested and applied, and the factors controlling denitrification measured. While there was considerable potential for denitrification at the site, this was not realized because much of the water moving off the farmland bypassed the riparian zone, entering the river directly via springs or through gravel lenses beneath the floodplain soil. Management of this site would not reduce nitrate leaching unless the floodplain hydrology could be substantially modified, and the main conclusion is that nitrate buffer zones will only operate efficiently where the hydrology of the site is appropriate. Copyright © 1999 John Wiley & Sons, Ltd.