Uncertainties in Transient Capture‐Zone Estimates of Groundwater Supply Wells

Abstract

UCOWR JOURNAL OF CONTEMPORARY WATER RESEARCH & EDUCATION Capture zones of water-supply wells are a widely used analysis tool for protection of ground water resources. Transient analyses of capture zones provide a more complete assessment than the commonly applied steady-state analyses. Previously, we have demonstrated that advectiononly analyses can produce biased transient capture-zone estimates. Therefore, it is important to consider the dispersion of contaminant plumes. Here, we extend our study to incorporate temporal and spatial distribution in the contaminant sources and their respective uncertainties. Our analysis indicates that the capture-zone estimates can be very sensitive to the transients in the contaminant releases. Even relatively small uncertainties in the contaminant source, when combined with transient fl ow effects associated with natural variability of gradients or water-supply pumping, can cause signifi cant uncertainties in the capture-zone estimates. This conclusion has important practical implications. Furthermore, we investigate the impact of uncertainty in the longitudinal and transverse dispersivities on the transient capture estimates. Capture zones are important for the effi cient protection of ground water resources produced by wells and springs. Typically, the capture zones are delineated using mathematical models. The models are based on simplifying assumptions for representation of real hydrogeological systems. For example, the transients are commonly ignored in the fl ow and transport models assuming a steady-state fl ow. Actually, substantial transients might exist, for example, due to variability in the pumping rates of water-supply wells (Reilly and Pollock 1996, Festger and Walter 2002) and, as a result, there might be substantial bias in the steady-state capture-zone estimates (Figure 1). Furthermore, even if the transients are incorporated in the model, the groundwater transport might be represented by advection-only fl ow paths (Rock and Kupfersberger 2002). The advection-only analysis might not provide an acceptable representation of mean plume behavior of potential transport because of the impact of transients on the effective plume dispersion. As a result, we might have an additional bias in the capture-zone estimates (Vesselinov and Robinson 2006). Here we analyse numerically the impact of the transients in the ground water fl ow and transport on the capture zone estimates for a series of synthetic cases. We also investigate the impact of uncertainty in the longitudinal and transverse dispersivities on the transient capture estimates.