Nitrate fluctuations in groundwater: review of potential mechanisms and application to case studies

This report describes work carried out as part of the BGS research project “Nitrate fluctuations in groundwater”. The aim of the project was to determine the mechanism(s) linking within-year fluctuations with groundwater level observed in many time series of nitrate concentrations in groundwater. Four conceptual models were prepared to represent scenarios where chalk recharge is taking place by only one mechanism. These mechanisms were winter piston flow through the unsaturated zone matrix, winter bypass flow from the base of the soil bringing high nitrate water directly to the water table, water table rise from water entering elsewhere in the catchment flushing out porewater and a change in flow path giving access to a greater percentage of shallow high nitrate water. These were evaluated in a very simplified way to determine whether different relationship between water levels and nitrate concentration arriving at the water table. It was concluded that flushing by a rising water table could potentially result in a delay between the water level rising and the nitrate concentration increasing, but the other mechanisms would be all likely to result in the more or less simultaneous rising of the water level and arrival of nitrate. However the distribution of nitrate in the porewater profile was predicted to be different and this may prove to be a moreuseful indicator. Three case study sites were selected for evaluation; two previous BGS research sites at Ogbourne St George, Wiltshire and at Bircham, Norfolk where there was considerable existing data and Morestead, Twyford, Hampshire which was selected as the research site for the current project. Ogbourne St George was found to be the most informative site, primarily due to the successful deployment of a multi-level sampler which obtained saturated zone samples from small discrete intervals and also to corresponding porewater data. At this site water with a nitrate concentration higher than the porewater was observed in the zone of water table fluctuation during periods of rising water levels. Additionally increases in nitrate concentration were possibly detected ahead of water level rises in a nearby observation borehole. These observations are inconsistent with piston flow through the matrix and correspond better with the fracture flow model or with lateral transfer of water from another part of the aquifer where nitrate concentrations are already higher. In the Bircham area four sites were chosen; Sedgeford, Fring, Osier Carr and Great Bircham. Two of these were looked at in more detail. The boreholes at Sedgeford had a very high response to water level variation, but in dry intervals nitrate peaks had either low amplitude or were lost. At Great Bircham the pattern was different with more subdued water level rises and a more protracted impact from drier periods. This was likely to be due to differences in the physical setting of the Bircham site which was in a shallow valley where the unsaturated zone was much thinner than at Sedgeford. The porewater profiles shape and the detection of higher nitrate water as soon as the water level rose into the zone of fluctuation were consistent with the winter high nitrate peaks being derived from increased permeability in the zone of water table fluctuation allowing a greater contribution of shallow, polluted groundwater to abstracted water. At Twyford where the porewater profile was measured after a long dry period, the distribution of nitrate in the unsaturated zone in the absence of other data did not rule out any of the mechanisms. It was concluded that measuring the difference between the water level rise and arrival of nitrate at a borehole is unlikely to be able to distinguish the mechanism operating and other information, such as porewater concentration, may be needed.