Assessing performance of a permeable biobarrier

While the past 15 years have seen rapid development of permeable reactive barriers and other in situ systems for the treatment of groundwater contamination, assessing performance has not received the same attention. In some cases the long-screened monitoring wells installed as part of initial site investigations are used to monitor treatment, while in others a few additional wells may be installed to serve as sentinels. Performance is assessed based on differences in contaminant concentrations, which is unreliable where plumes are spatially complex and/or temporally unstable. In this work, a standard concentration performance metric is compared with that based on mass flux, using data collected during an evaluation of a discontinuous aerobic biobarrier treating a near-source (and spatially complex) BTEX plume. The data were derived from high-resolution multilevel samplers installed up- and down-gradient of the treatment system. Concentrations in screened monitoring wells of various lengths were approximated by interval-weighted concentrations from these multilevels. The accuracy of the different performance assessment methods was measured based on estimated uncertainty generated as a result of observed spatial plume complexity. While more data intensive, mass flux proved to be a more reliable performance metric. Assessment using data from few long-screened wells could not reliably match that of the flux fences, but did come within acceptable limits when the number of wells approached the number of multilevels.