When perchlorate degradation in the riverbank cannot impede the contamination of drinking water wells

Abstract

Health issues related to aquifer contamination with perchlorate are a growing concern in drinking water management. This study describes perchlorate transport and degradation processes from a contaminated stream toward drinking water pumping wells. Investigations are based on laboratory experiments and field measurements conducted at a well field near Bordeaux (France) in a heterogeneous carbonate aquifer interacting with a stream. Field measurements facilitated the characterization of perchlorate contamination and stream-to-aquifer flow. Experiments on columns of streambed sediments conducted in the laboratory confirmed that perchlorate had been degraded in the hyporheic zone. A one-dimensional reactive transport model was implemented to estimate Monod kinetic rates, which account for the inhibition of perchlorate degradation by nitrate. The estimated half-saturation constant for perchlorate $$ \Big({k}_{1/2}^{\mathrm{Pcl}} $$ ) is 6.93 10−9 mol L−1 and the estimated maximum specific degradation rate ( $$ {k}_{\mathrm{max}}^{\mathrm{Pcl}} $$ ) ranges between 10−5 and 4.0 10−3 mol L−1 day−1. Despite degradation in the hyporheic zone, perchlorate-contaminated stream water reaches drinking-water-production units. Such contamination highlights the effects of preferential flow paths between the stream and the pumping wells and significant hydraulic gradients caused by drawdowns. In such contexts, in spite of a good potential for degradation, riverbank filtration may not be effective for the protection of drinking water wells. Lessons from this study also reveal that contamination monitoring can be misleading: low concentrations can be reported in monitoring wells between the contaminant source and the production wells, but the latter may yet be contaminated.