How do vapors get from contaminated soil or groundwater into a building?
Volatile chemicals partition out of a subsurface source into soil gas, move upward through the unsaturated zone, and enter a building through openings in the foundation, where they mix with indoor air. EPA's OSWER Technical Guide for Assessing and Mitigating the Vapor Intrusion Pathway from Subsurface Vapor Sources to Indoor Air (OSWER Publication 9200.2-154, June 2015) describes the attenuation along that route as "the reduction in volatile chemical concentrations that occurs during vapor migration in the subsurface, coupled with the dilution that can occur when the vapors enter a building and mix with indoor air." Which source is present changes the assessment: where the vadose zone soil itself holds a source, the guide states that groundwater screening levels and groundwater attenuation factors "are not recommended for estimating potential upper-bound indoor air concentrations, because they have been derived assuming no other vapor sources exist between the water table and the building foundation."
The building is not a passive receiver. Soil gas enters because the building is generally at slightly lower pressure than the soil beneath it, and that difference varies with weather, heating and how the building is operated. The June 2015 guide observes that in many geographic areas in the continental United States, “indoor air sampling during the heating season may yield higher indoor air concentrations than at other periods, because stack effects are generally more significant.” Two similar houses over the same plume can produce different indoor results for reasons unrelated to the plume.
Preferential migration routes break the generic model outright. The guide defines one as "a naturally occurring subsurface feature or anthropogenic (human-made) subsurface conduit that is expected to exhibit little resistance to vapor flow in the vadose zone," and lists as examples "sewer lines and manholes, utility vaults and corridors, elevator shafts, subsurface drains, permeable fill, and underground mine workings." It notes that significant horizontal preferential routes "may result in elevated concentrations in areas on the periphery of subsurface contamination," and that where such routes exist their presence "is likely to render inappropriate the use of any of these generic attenuation factors." How migration is modeled at a specific site is covered at fate and transport modeling.
What does EPA's vapor intrusion guidance actually require?
By its own terms, nothing. The June 2015 OSWER Technical Guide carries a disclaimer stating that it "does not impose any requirements or obligations on the EPA, the states or tribal governments, or the regulated community," that "the sources of authority and requirements for addressing subsurface vapor intrusion are the relevant statutes and regulations," and that EPA decision-makers "retain the discretion to adopt or approve approaches on a case-by-case basis that differ from this guidance document" where the administrative record supports it. It is written for sites evaluated under CERCLA or under RCRA corrective action, and it states that it and the companion petroleum guide together “supersede and replace EPA's Draft Guidance for Evaluating the Vapor Intrusion to Indoor Air Pathway from Groundwater and Soils (EPA 2002c).”
What it supplies is a method, not a threshold: a conceptual site model, identification of preferential migration routes, building surveys, characterization of ambient air, and multiple lines of evidence weighed together. EPA "generally recommends that multiple lines of evidence be developed and their results weighed together when evaluating and making risk-informed decisions pertaining to vapor intrusion." One medium, sampled once, is not what the framework contemplates.
A companion document issued the same month covers petroleum releases separately, and its numbers are screening distances rather than safety margins. The Technical Guide For Addressing Petroleum Vapor Intrusion At Leaking Underground Storage Tank Sites, EPA 510-R-15-001 (June 2015), states that "[a]dditional investigation is generally unnecessary if the vertical separation distance is greater than 6 feet for dissolved contamination beneath buildings of any size, or 15 feet for light non-aqueous phase liquid (LNAPL) if the overlying building has at least one side shorter than 66 feet in length." Those distances are specific to petroleum hydrocarbons at leaking underground storage tank sites, where vadose zone biodegradation does much of the attenuating; they do not transfer to chlorinated solvents, which are covered at solvents and VOCs. Many states run their own vapor intrusion programs with their own values, and which framework governs a site is a question for the regulator and for counsel.
What are attenuation factors and vapor intrusion screening levels, and are they health limits?
An attenuation factor is the ratio of the indoor air concentration arising from vapor intrusion to the subsurface concentration, and a vapor intrusion screening level is a subsurface concentration derived by dividing a target indoor air concentration by that factor. Neither is a health-based limit and neither is a cleanup standard. Table 6-1 of the June 2015 OSWER Technical Guide recommends, for residential buildings, 0.03 for sub-slab soil gas; 0.03 for "near-source" exterior soil gas, except where the source is less than five feet below the foundation or preferential migration routes are present; 0.001 for groundwater, subject to the same exceptions and to a shallow water table; 0.0005 for groundwater where laterally extensive fine-grained vadose zone soils are documented; and 1.0 for crawl space air.
Those values are conservative by construction and bounded by their data. Except for the near-source exterior soil gas factor, which the guide says "corresponds to approximately the estimated 75th percentile value," the recommended residential values are "the estimated 95th percentile values, rounded to one significant figure," drawn from EPA's Vapor Intrusion Database: Evaluation and Characterization of Attenuation Factors for Chlorinated Volatile Organic Compounds and Residential Buildings, EPA-530-R-10-002 (March 2012) — 913 buildings at 41 sites, chlorinated volatile organic compounds and residential buildings. The groundwater and near-source soil gas factors, the guide adds, "do not, however, include the effects of biodegradation."
The guidance says plainly what an exceedance means. Section 6.5.4 states that "any individual subsurface sampling result that exceeds the respective, chronic screening level does not establish that vapor intrusion will pose an unacceptable human health risk to building occupants," because the subsurface screening levels "are expected to be conservative ... due to the use of a high-end attenuation factor," because for carcinogens they use a one-per-million lifetime cancer risk while EPA "recommends consideration of a cancer risk range when making risk management decisions," and because results vary spatially and temporally. The conservatism runs the other way too: single-chemical screening levels "do not account for the cumulative effect of all vapor-forming chemicals that may be present." The guide is explicit that these are not cleanup numbers: the vapor intrusion screening levels “are not automatically response action levels, although EPA recommends that similar calculation algorithms be employed to derive cleanup levels,” and comparing a sample to them “is only one factor recommended for use in determining the need for a response action at a site.” The numeric levels also move, because EPA's Vapor Intrusion Screening Level calculator draws on the same toxicity values as the Regional Screening Levels, which are revised periodically — so a screening value quoted without its table version and date is not a citation. Related data usability questions are covered at sampling and data quality.
Why is indoor air sampling for vapor intrusion so heavily contested?
Because indoor air measures what the case is about and is also the medium a defendant can most credibly attribute to something else. The June 2015 OSWER Technical Guide states that "indoor air is likely to contain detectable levels of a number of vapor-forming chemicals regardless of whether the building overlies a subsurface vapor source, because indoor air can be impacted by a variety of indoor and outdoor vapor sources unrelated to site contamination," and notes that in some buildings those background sources "by themselves can cause building occupants and visitors to experience significant exposures to vapor-forming chemicals."
The separation is attempted through a building survey, contemporaneous sub-slab, indoor and ambient air sampling, and removal of indoor sources beforehand — in residences, "at least 24 to 72 hours prior to the start of sampling, based on an approximate air exchange rate of 0.25 to 1.0 per hour" — while the guide concedes "it may not be possible to remove all indoor sources." One diagnostic it offers is the concentration ratio: on the generic sub-slab factor of 0.03, "sub-slab soil gas concentrations can be expected to typically exceed indoor air concentrations by 33 times or more in residences that are impacted by vapor intrusion," where background is negligible and the building is under-pressurized.
Temporal variability is the second front. Footnote 154 of the guide records the estimate, from the Utah dataset cited above, that a single randomly drawn one-day sample had a forty percent chance of falling below the true mean, and that where the true mean exceeded the action level by two or five times, one day's sample had a twenty percent or six percent chance respectively of missing the exceedance. EPA concludes that "multiple rounds (and often several rounds) of indoor air sampling is generally recommended," to guard against false negatives and false positives alike. One shortcut the guidance forecloses appears in reports on both sides: after publishing a compilation of North American residential indoor air concentrations for 1990 to 2005 (EPA-530-R-10-001, June 2011), EPA said it "does not recommend the use of generic values of historical background concentrations, even those cited in peer-reviewed publications or available from databases maintained by regulatory agencies, to characterize current levels in any building," and asks for site-specific data instead.
How does vapor intrusion show up in litigation and in property transactions?
In litigation it shows up as the injury mechanism and as the variable that decides whether a community can be tried together. In Ebert v. General Mills, Inc., 823 F.3d 472 (8th Cir. 20 May 2016), owners of residential property in one Minneapolis neighborhood alleged that General Mills had released trichloroethylene at a former facility in the same neighborhood and that trichloroethylene vapors had migrated into the surrounding residential area, threatening residents' health and decreasing property values; the Eighth Circuit reversed the certification order, holding that “the class lacks the requisite commonality and cohesiveness to satisfy Rule 23.” In Martin v. Behr Dayton Thermal Products LLC, 896 F.3d 405 (6th Cir. 16 July 2018), arising from the McCook Field neighborhood of Dayton, Ohio, the district court denied Rule 23(b)(3) certification of liability-only classes because Ohio law on injury-in-fact and causation defeated predominance, but certified seven issues under Rule 23(c)(4) — including whether the defendants' contamination and inaction "caused class members to incur the potential for vapor intrusion" — and the Sixth Circuit affirmed, stressing that "whether Defendants created the risk of vapor intrusion is distinct from the ultimate question of whether they caused an actual injury to property owners." Aggregation is covered at community exposure and class certification.
The work that decides those cases is exposure work. Whether vapors from a particular source reached a particular building, at what concentration and over what period, is established through the conceptual site model, the transport analysis, the sampling program and source attribution — which is why certification briefing is often where the modeling gets its first serious test. Whether the resulting record satisfies a legal element is for the court; whether any exposure affected anyone's health is a question this Institute does not answer and no screening value answers. See exposure assessment and dose reconstruction and source attribution and forensics.
In property transactions the pathway enters through due diligence. A purchaser seeking CERCLA's innocent landowner protection must show that "on or before the date on which the defendant acquired the facility, the defendant carried out all appropriate inquiries" into previous ownership and uses (42 U.S.C. § 9601(35)(B)), and EPA's all appropriate inquiries regulation sits at 40 C.F.R. Part 312. On 15 December 2022 EPA published a final rule, Standards and Practices for All Appropriate Inquiries, 87 Fed. Reg. 76578, effective 13 February 2023, listing ASTM E1527-21 at 40 C.F.R. § 312.11(a) as a standard that may be used to comply with the regulation's requirements; the same rule left E1527-13 listed at 40 C.F.R. § 312.11(c) only “[u]ntil February 13, 2024.” The evidence is also perishable and the pathway is mitigable: foundations get sealed, sub-slab depressurization systems get installed, buildings come down, and once a system is running the indoor air it produces is not the indoor air that existed before. See evidence preservation. What any of it is worth as a property loss belongs to our Economic Damages Institute.