There is a fundamental shortcoming in existing vapor intrusion policies. Most significant, those policies focus solely on the indoor air pathway, instead of addressing the full range of sources, pathways, and receptors at vapor intrusion sites. The data at such sites, the concerns of the residents and employees of such areas, and analysis by technical experts all suggest the broader approach.

According to a presentation by Robert Ettinger at a recent Groundwater Resources Association of California technical conference, background levels of indoor air contaminants often exceed air contamination from underlying groundwater plumes and even may be above health-based standards. He described this as a policy question, and neither Ettinger nor any of the other experts presenting at the conference were willing to tackle this issue.

In my community of Mountain View, California - a test case for U.S. EPA's Vapor Intrusion Guidance - we can't escape the question, however. Ambient - I prefer this term to "background" - levels of TCE and PCE, above EPA screening levels, have been found at local school and playground sites. Responsible parties suggest that if outdoor air concentrations exceed indoor air contamination, then no further action is necessary, even though my neighbors continue to breathe unsafe air. Furthermore, test results at one dwelling unit with significant indoor air contamination show that the common "remedy" of venting actually increases outdoor air contamination.

Since ambient concentrations of TCE and PCE do not arise naturally, it is necessary to determine the sources, analyze the risks, and if concentrations exceed health-based goals, take steps to eliminate or reduce exposures. Based upon preliminary data in Mountain View, I identify three POTENTIAL descriptions of our vapor intrusion problem.

Three Scenarios

1. Elevated health risks may exist ONLY in buildings directly above or within 100 feet of shallow groundwater plumes of volatile organic compounds. We have some buildings where concentrations have been measured well above the old cleanup goal, so we know that there are indoor risks that should be addressed promptly, but that's not enough.
2. Ambient levels throughout Silicon Valley might be elevated. This has been suggested by some regulators and some responsible parties, and it's a scary possibility. Imagine a million or more people, including the most susceptible populations, exposed to unhealthy concentrations of TCE and/or PCE around the clock, for decades. However, I don't think the sampling supports this conclusion.
3. Most likely, TCE and maybe PCE at levels above EPA screening levels are present in the neighborhood of shallow groundwater plumes. This would mean that tens of thousands of people are probably being exposed a good percentage of the time. Contamination is rising inside buildings and outside buildings, as well as coming from other unknown sources, at a fairly steady rate, because these compounds break down and diffuse rather rapidly. That is, the presence of ambient concentrations suggests that there is are continuing sources, and those sources should be addressed.

If I'm right, then current strategies for both sampling and remediation are either wrong or incomplete. As currently practiced, outdoor air sampling is conducted purely for reference purposes. It is not designed to determine the extent or timing of outdoor concentrations of volatile organic compounds, nor to identify potential sources. If the outdoor contamination is treated as a potential risk factor, then sampling should attempt to bound the atmospheric plume, and samples should be timed to evaluate the impact of weather and industrial activity on readings. Sampling should help evaluate whether groundwater treatment systems or continuing industrial activity are among the likely sources of contamination. While this may mean that cleanup program need to go "out of bounds," measuring emissions from industry not under its jurisdiction, I believe it can be justified as part of the effort to rule out non-waste sources.

If indeed, as I believe, outdoor air must be addressed, then it may NOT make sense to solve indoor air problems by venting indoor or subslab contamination. Moving contamination around is not a solution. Instead, the cleanup team should consider removing contaminated soil or using techniques, perhaps air sparging or bioremediation, designed to remove or break down near-surface contamination. Much of the effort to improve the cleanup to TCE and PCE has focused on their tendency to sink within aquifers. That's still an important problem, but near-surface cleanup may require earlier responses.

Preferential Pathways and Land Use Controls

Moreover, evidence from the field suggests that preferential pathways are often the primary pathway. For example, at a new building (with engineering controls) here in Mountain View, recent sampling discovered unexpectedly high levels of indoor TCE. The likely culprit: a connection to an unprotected utility vault. The models for predicting indoor air contamination break down where there are preferential pathways, and there may be such pathways at virtually every site. Furthermore, such pathways exist outdoors as well as indoors. Even if the soil above plumes tends to trap or degrade volatile organic compounds, preferential pathways may be large enough to drive a Hummer through. That is, a small hole or crack may defeat the mechanisms that might prevent vertical transport.

Other than at the building with the unprotected utility vault, the new commercial/office buildings in Mountain View, built directly above high concentrations of volatile organic compounds, report levels of indoor contamination comparable to outdoor readings. That is, the engineering controls of impermeable slabs and positive air pressure appear to be effective. I don't believe the studies have shown which of those is most responsible. Still, in earthquake country there is no guarantee that the slabs will remains impermeable, and over the life of the buildings there is no guarantee that the ventilation systems will not be changed.

Thus, even effective engineering controls must be integrated into long-term stewardship plans that 1) legally require the controls as part of the remedy, 2) provide for the long-term monitoring of their protectiveness, and 3) provide for additional remedial action if they prove unprotective. Yet to my knowledge none of the Mountain View sites have any registered institutional controls related to vapor intrusion. If construction and ventilation are regarded as ways to limit human exposure, then they should be legally incorporated into the cleanup. Detection Limits and Other Compounds

In addition, it is difficult to project risk and design remedies when it is difficult if not impossible to measure air contamination down to preliminary remediation goals. Yet in Mountain View, EPA reports a detection limit roughly an order of magnitude above that goal. One of the responsible parties uses a lab that claims a lower detection limit, but it is still above the health-based goal. This suggests a technical priority: A new detection method or technology should be developed to bring the detection limit in line with the levels necessary to characterize and remediate these sites. It may be beyond the reach of foreseeable technologies, but community members seek - for their own assurance - real-time or near-real time monitoring systems that detect low concentrations of such contaminants, as well.

The focus of the Mountain View investigations is TCE, even though PCE is found in the outdoor air in similar concentrations. One of the reasons is that EPA has completed a draft toxicity assessment for TCE, but not for PCE. It would be a mistake to focus resources on one contaminant while ignoring other, similar compounds if there is a chance that further toxicity assessments would lower the health-based goals for the other compounds. At least, those responding to vapor intrusion should consider approaches that deal with a wide range of compounds, to avoid the chance that they may have to come back to a site later to deal with those ignored the first time around.