Passive diffusion ground water samplers at a site with heterogeneous hydrostratigraphy: Pilot study results

Abstract

AbstractUse of passive diffusion (PD) ground water samplers has significantly increased in recent years because significant cost savings may be realized, primarily because sampling time is reduced and minimal investigation‐derived waste (IDW) is generated. However, PD samplers measure water quality over a relatively discrete interval, while traditional purge‐and‐sample methods are most influenced by the water quality in more permeable zones across the well screen. Consequently, these two sampling methods may potentially yield distinctly differing concentration measurements for a specific well if significant contaminant stratification and geologic heterogeneity exist. A large body of purge‐and‐sample historical monitoring data exist for the study site, and results from potential future full‐scale PD sampler application must be interpretable and comparable in the context of these historical data to be useful for evaluating achievement of long‐term site remediation objectives. Therefore, a pilot study was conducted to assess potential for PD samplers for monitoring chlorinated volatile organic compounds (VOCs) in ground water. PD samplers were installed in a select group of indicator wells, and their performance was compared to results from both purge‐and‐sample and low‐flow (LF; another discrete interval–type technique) methods. At several wells, multiple PD samplers were installed across the well screen interval to evaluate potential contaminant stratification. At one well, contaminant stratification was particularly significant: concentrations of total VOCs measured by two PD samplers installed ∼0.6 m (2 feet) apart differed by an order of magnitude. PD sampler and LF data interpreted in tandem with purge‐and‐sample data provide additional information, suggesting that contamination is located primarily in the more permeable zones at the study well locations. Additionally, low concentrations of some VOCs were measured in some PD and LF samples but not in the corresponding purge‐and‐bail samples. The results of this study underscore the importance of pilot studies and careful program design and regulatory agreement before implementing PD samplers (or any alternative sampling method) in a long‐term monitoring program, especially at sites with heterogeneous hydrostratigraphy.