Environmental tracer transport (H-3 and SF6) in the saturated and unsaturated zones and its use in nitrate pollution management

An important quantity in groundwater protection is the residence time of wa ter in an aquifer. It relates to both the travel time of a pollutant to arr ive at a well and the time span required for self-purification of a pollute d aquifer after removal of pollutant inputs. Time scales for aquifers can b e gained from artificial tracer experiments or from environmental tracer da ta, the latter offering the only realistic alternative if time scales of ye ars or decades have to be taken into account. Different tracers show different time scales due to their different transpo rt mechanisms especially in the unsaturated zone. While solute tracers are moved advectively with the seepage water, gas tracers pass the unsaturated zone diffusively through the air phase. Depending on the properties of the unsaturated zone (hydraulic properties, thickness) this difference in behav ior can be used to separate the subsurface transport process into the unsat urated and the saturated parts. In a field study in Germany, SF6 and H-3 were used as environmental tracers . Both have a relatively well-known input function. Interpretation of data from observation wells by a box model approach led to spatially and tempora lly varying residence times. This was an indication that the influence of t he unsaturated zone could not be neglected. While the gas tracer SF6 shows only residence times in the saturated zone, the tracer H-3 reflects the who le travel time of water including both the unsaturated and saturated zones. Using a one-dimensional plug-flow model for the unsaturated zone combined with a detailed two-dimensional flow and transport model for the saturated zone leads to a holistic and consistent interpretation of the measured trac er concentrations. The observed pattern of old water under thick loess cove r and younger water under areas where the fractured basalt aquifer crops ou t is reproduced after adjusting only two parameters: the effective porosity of the saturated aquifer and the product of field capacity and thickness o f the unsaturated zone. While the effective porosity of the saturated zone is adjusted by means of the SF6 data, the field capacity of the loess layer is adjusted by means of the H-3 observations. The thickness of the unsatur ated zone is deduced from geological and pedological maps. All flow data ar e obtained from a calibrated flow model, which is based on geological data, observed heads and pumping tests only. The transport model for the saturated zone was calibrated by fitting the po rosity by means of gaseous tracer concentrations (SF6). The combined satura ted-unsaturated zone model was then calibrated by fitting the held capacity of the unsaturated zone by means of H-3 concentrations. With this model it was possible to verify the observed NO3 concentrations at the drinking wat er wells and to develop predictions for their future development under vari ous scenarios of fertilizer input reduction in specific areas. (C) 2001 Els evier Science B.V. All rights reserved.