Utility of Simple Models for Capture Zone Delineation in Heterogeneous Unconfined Aquifers

Abstract

Monte Carlo simulation is used to examine the utility of simple (homogeneous/isotropic) models for the delineation of wellhead protection areas (WHPAs) in heterogeneous, statistically anisotropic, unconfined aquifers. Capture zone geometry is investigated under two hydrologic conditions: regional gradient negligible compared to pumping and significant regional gradient compared to the local gradient due to pumping. Variability of capture zone geometry is quantified in terms of simple measurements (maximum transport distance and maximum width of the capture zone) and the probability of inclusion in the capture zone of a series of hypothetical contaminant source locations. These results indicate that substantial uncertainty exists in the delineation of a wellhead capture zone. This uncertainty is related, principally, to flow that occurs along preferential pathways associated with zones of high hydraulic conductivity. It is suggested that these uncertainties can be accommodated in one of three fashions depending on the value of land, the benefit of wellhead protection, and the availability of funds to perform more thorough field assessment. The first option is to modify the estimated capture zone by a factor of safety. In the absence of significant regional flow, a factor of safety of approximately three applied to the calculated fixed radius estimate of the capture zone appears to be sufficient for the conditions simulated in this project. In the case of a capture zone within a significant regional gradient, uncertainty in the location of the real capture zone will require safety factors on the width of the estimated capture zone on the order of 30, with a safety factor of approximately three in the longitudinal direction. The second option reduces these factors of safety by increasing the risk that the estimated capture zone will not cover the entire area of the real capture zone. Work available in the literature provides insight into the trade-off between the size of the estimated capture zone and the risk endured. The third option is to reduce uncertainty in the vicinity of a wellhead through additional expenditure to allow the application of classic hydrologic techniques involving conditioning or inverse modeling.