Quantifying groundwater flows to streams using differential flow gaugings and water chemistry

Abstract

Summary While estimates of net groundwater inflow to streams (inflow minus outflow) can be made using differential flow gauging, the inclusion of water chemistry (tracer) measurements allows both inflow and outflow to be separately quantified. In this paper we assess how the estimates of net and gross groundwater inflows are affected by the choice of tracer at three contrasting field sites. Groundwater flows are first estimated with differential flow gauging and then with the sequential addition of natural tracer data – electrical conductivity, chloride concentration and radon activity measurements. The final analysis is where an injected tracer experiment is also conducted to constrain the gas transfer velocity for radon. Groundwater inflow rates were estimated by calibrating a numerical model which simulated flows and concentrations of tracers in the river. Although both the total groundwater inflow along the study reach and the spatial distribution of inflow depended on the data used for the model calibration, the difference between the estimates was less than the prediction error. The analysis also showed that prediction error for groundwater inflow decreases as additional tracers are included in the analysis. The magnitude of the error reduction is related to the properties of the specific catchment. Generally, for a tracer to reduce uncertainty substantially the concentration of the tracer in groundwater must be well defined, and the contrast between the concentration of the tracer in groundwater and the river must be high.