Abstract

Experiments involving the injection of tracers into two 0.1 ha field plots are used to demonstrate the effects of measurement, spatial and conceptual uncertainty on the identification of the dominant flow-pathway of weakly reactive contaminants through soil and drift. One field plot constitutes a 6 m deep glaciolacustrine drift with interbeded clay loam and sand layers. A gleyed brown earth developed on +10 m of glacial till and under-drained at 0.7 m comprises the second plot. Flow-paths within each plot are identified using both hydrometric and tracer methods. The hydrometric approach centres on distributed measurements of capillary potential and hydraulic conductivity (using permeametry, borehole tests and moisture release). The water-tracing involves the point injection weakly sorbing sodium chloride, Rhodamine WT and strontium chloride. Tracer migration is then identified by the integration of methods of water-extraction and laboratory analysis, logged water-resistivity within bores, and the measurement of bulk or ground resistivity. The two approaches of hydrometry and water-tracing are sensitive to very different but equally critical measurement, spatial and conceptual errors. A synergism is apparent however when both approaches are used simultaneously.