Report Cover   Table of Contents   Sec. 1   Sec. 2   Sec. 3   Sec. 4   Sec. 5   Sec. 6   Sec. 7   Sec. 8

Download the Report PDF (874 KB) ("Right-click", then "Save as..." to your desktop)
View Report Appendices

SECTION 7

CONCLUSIONS AND RECOMMENDATIONS

The overall objective of this demonstration was to evaluate and demonstrate the use of selected diffusion and grab sampling devices that potentially represent useful and costeffective alternatives to conventional groundwater sampling approaches (e.g., 3-volume purge/sample and low-flow purge/sample) for analytes other than VOCs. Specifically, devices that potentially can be used to sample for metals, anions, and 1,4 dioxane were evaluated. Expansion of the suite of accepted no-purge sampling methods could be useful in augmenting or possibly substituting for the PDBS method in certain applications.

From a performance perspective, the HydraSleeve® and Snap Sampler™ methods typically produced results that are most similar to the more conservative (i.e., higherconcentration) results obtained from the two conventional sampling methods. Both of these methods are characterized as grab-type samplers, and although they do allow for well equilibration, they do not use diffusion as the operative mechanism, therefore the results obtained are more of a “snapshot” in time. It should be noted that all of the diffusion and grab samplers collect samples over a ‘short’ time frame with respect to the groundwater velocity at many sites. Of these two sampling methods, the HydraSleeve® was substantially less expensive based on the assumptions used in the cost analysis, although both methods were less expensive than the conventional approaches. The HydraSleeve® was simpler to deploy and retrieve, and permits a larger volume of water to be collected. Comparisons involving the Snap Sampler™ on the other hand indicate that the VOC data set for this sampler may be more consistently representative of the actual VOC concentrations in the well at the time of sample collection. A fully nonmetallic version of the Snap Sampler™ is available in the event that the metal construction of the Snap Sampler™ tested in this study is of concern.

For the diffusion-based methods, the PDBS provided the most conservative results (i.e., highest concentrations) for VOCs, but this device is only appropriate for monitoring most VOCs. The other diffusion-based devices evaluated occasionally produced results that were, on average, lower in concentration than the conventional and/or grab sampler results. Although much less expensive than the conventional sampling methods, these devices were generally more expensive to use than the HydraSleeve® based on the assumptions used in the cost analysis (the PDBS and HydraSleeve® costs were very similar, see Table 5.2).

Finally, although the conventional methods evaluated are well accepted throughout the industry (Newell et al., 2000), they did not always provide the most conservative (i.e., highest-magnitude) results. These methods also were more expensive than the diffusion and grab samplers used in the demonstration.

The American Petroleum Institute (Newell et al., 2000) concluded that three-volume purge/sample data are broadly accepted by regulatory agencies, suggesting a consensus that these data are adequately representative of formation conditions to ensure protection of human health and the environment. One of the primary objectives of this demonstration is to identify cost-effective alternatives to conventional groundwater sampling approaches for analytes other than VOCs. Accordingly, one critical question is whether the diffusion and grab devices are also adequately representative and therefore protective (or more conservative), but not systematically biased low relative to conventional methods. A particular grab or diffusion sampling device was considered to be viable when the analyte concentrations obtained using that device were similar to or higher than those obtained using conventional sampling methods. Conversely, if the results obtained using the grab or diffusion device exhibited low bias relative to the conventional results, then the ability of that device to accurately detect concentrations of that particular analyte group was considered to be suspect. In summary, high bias was considered to be acceptable, but low bias was not. Of the four diffusion and two grab sampling devices evaluated, the following conclusions were derived based on the results of this demonstration. In addition, Table 7.1 contains a summary of key conclusions and observations derived from this technology demonstration.

• The HydraSleeve® appears to be a technically viable method for monitoring all of the compounds included in this demonstration. Concentrations of metals obtained using this device tended to be lower than low-flow-purge concentrations, but the low-flow concentrations appear to be anomalously high.
• The Snap Sampler™ appears to be a technically viable method for monitoring all of the compounds it was tested for in this demonstration (i.e., anions, 1,4 dioxane, and VOCs). This method was not tested for metals and hexavalent chromium.
• The PDBS is a technically viable method for monitoring VOCs only. PDB samples may be advantageous when sampling for VOCs in waters where matrix interferences, such as highly turbid or alkaline conditions, may compromise results.
• The RPPS appears to be a technically viable method for monitoring hexavalent chromium, metals, and anions. Although concentrations of VOCs and 1,4 dioxane obtained using this method are statistically similar to low-flow concentrations of these analytes, they tended to be biased low relative to concentrations obtained using the three-volume-purge method. Further development of this technology may be warranted.
• The RCS appears to be a technically viable method for monitoring anions and possibly 1,4 dioxane and hexavalent chromium. Although other studies have shown that the RCS is appropriate for monitoring VOCs (Ehlke et al., 2004; Imbrigiotta et al., 2002; Vroblesky et al., 2002), that observation was not validated during this demonstration.
• The PsMS appears to be a technically viable method for monitoring hexavalent chromium, metals, and anions, and may be technically viable for monitoring VOCs (although VOC results using this method were typically less than those obtained using the three-volume purge and HydraSleeve® methods and were not compared with the Snap Sampler™ or RCS results because, with one exception, they were not installed in the same wells). Although all comparative tests indicate that 1,4 dioxane concentrations in the PsMS are similar to concentrations of this analyte obtained using conventional sampling methods, a more definitive endorsement regarding the use of the PsMS method for dioxane sampling was not reached due to the relatively low number of comparisons. Metal concentrations obtained using the PsMS method tended to be lower than low-flow concentrations, but the low-flow concentrations tended to be anomalously high. Further development of this technology may be warranted.
  
  

TABLE 7.1
SUMMARY OF CONCLUSIONS AND RECOMMENDATIONS
NO-PURGE SAMPLER DEMONSTRATION
McCLELLAN AFB, CALIFORNIA
View TABLE 7.1

Report Cover   Table of Contents   Sec. 1   Sec. 2   Sec. 3   Sec. 4   Sec. 5   Sec. 6   Sec. 7   Sec. 8