Introduction The subsurface vapor migration to indoor air pathway is now being considered more commonly and in more detail during risk-based site evaluations at contaminated sites. The USEPA issued Draft Guidance for Evaluating the Vapor Intrusion to Indoor Air Pathway from Groundwater and Soils (November 2002) and many states also have recent or forthcoming guidance documents. Although the increased interest and guidance documents are relatively recent, the vapor intrusion pathway for Volatile Organic Compounds (VOCs) has been under study since the 1980’s (e.g., the 1989 Hillside School case in Massachusetts) and radon intrusion has been studied for even longer. This article provides a practitioner’s perspective on the complexities and challenges inherent in evaluating the vapor intrusion pathway and suggests strategies for minimizing the cost of characterization while obtaining technically defensible results.

Conceptual Model The key to assessment of the vapor intrusion pathway is to identify the processes and mechanisms contributing to vapor intrusion and attenuation and collect data strategically to quantify each important mechanism or process in sufficient detail for a defensible assessment of potential inhalation risks. The framework for this approach is the Site Conceptual Model (SCM). The most common conceptual model is upward diffusion through soil gas from soil or groundwater, with convection into the building and dilution within the building. This conceptual model forms the basis of the commonly applied Johnson and Ettinger (1991) mathematical model (the J&E Model) for calculating the ratio of indoor air vapor concentrations to soil vapor concentrations at a specified depth (i.e., the attenuation factor).

Other processes and conditions that are not considered in the J&E Model may be important in certain circumstances. These include lateral vapor diffusion through the vadose zone, barometric pumping, migration through preferential pathways, biodegradation, pressurized buildings, passively-ventilated crawl spaces, wet basements, the fresh water lens, and perched groundwater layers. Depending on the relative importance of any of these processes and conditions, the common SCM may not be relevant or appropriate.

Generalized Evaluation Approach Approaches to assess this pathway commonly follow a tiered system. The evaluation typically begins with a review of potential imminent hazards (explosions, acute health risks). If VOCs are present that are sufficiently volatile and toxic to pose a potential inhalation risk, measured groundwater and/or soil gas concentrations are compared to generic screening values derived using simple one-dimensional modeling and conservative model input values. The screening values can also be customized to match site-specific conditions by modifying the simple model with site-specific input values. If groundwater and/or soil gas concentrations are higher than the generic or site-specific screening values, supplemental data collection may be required, which may include multiple lines of evidence (e.g. soil gas or indoor air concentration data, soil properties, building ventilation, biological activity, or other site-specific model inputs).

The strategic approach to site evaluation is to assess the range of potential impacts considering the expected variability in the important processes and conditions, supported by a combination of model predictions and measurements. Measurements will often be needed in a representative number of locations and time-periods to establish spatial and temporal trends. It is generally better to assess site conditions in a selected few representative locations using a variety of techniques than to rely on a single technique applied more extensively. For example, in cases where VOCs in groundwater have migrated under multiple occupied buildings, it is generally not necessary to conduct exhaustive studies of each and every building.

Modeling An appropriately-formulated mathematical model is always a useful tool for assessing the sensitivity of various components of the SCM. Screening-level modeling can be initiated at the outset of any assessment to identify the components that contribute most to variability and uncertainty. The field data collection activities can then be selected strategically to provide the most relevant information. An iterative process of data collection and model refinement will often provide the most cost-effective assessment.

The J&E Model is a screening level model that predicts indoor air concentrations from subsurface vapor concentrations, often within approximately one order of magnitude. The inherent variability in subsurface vapor and indoor air concentrations is also often about one order of magnitude; therefore, it is best to compare model results with multiple measurements to address data variability.

Models other than the J&E Model have been developed that can be used to evaluate vadose-zone biodegradation and/or two and three-dimensional transport. However, the level of complexity of these models may make them challenging to apply, calibrate, and defend to reviewers. In some cases, there are no appropriately-formulated mathematical models, in which case, an empirical assessment of the transport mechanisms may be the best alternative.

Data Collection A compounding challenge in evaluating vapor intrusion is the diversity of opinions on appropriate technical approaches to be used. Typical investigative techniques include sampling and analysis of groundwater, soil gas, indoor air, and outdoor (or ambient) air. For soil gas and groundwater data, it may be important to assess the vertical profile of concentrations via depth-discrete sampling. Target indoor air, soil gas and groundwater concentrations for vapor intrusion assessments are typically very low. To date, limited guidance documents have been published to help meet the data quality objectives necessary at these low concentration levels, although there is work in progress on this topic.

Other data that may be collected includes soil properties (porosity, permeability, moisture content, fraction of organic carbon), water table fluctuations, foundation design and condition, and presence of consumer products or building materials that could contribute vapors to indoor air. Other less common methods (building ventilation, building pressure testing, building envelope leakance testing, dynamic and static flux chambers, and passive soil gas sampling) or experimental methods may be considered if warranted by site conditions.

Some techniques are not applicable at all sites, depending on the geologic materials, depth to groundwater, building design, access constraints, and regulatory acceptance. Therefore, it is generally difficult to establish a “cookbook” approach that will universally applicable for site characterization.

Sampling and analysis of indoor air itself may appear to be the most direct way to assess exposure; however, several complicating factors can confound efforts to accurately quantify subsurface vapor contributions to indoor air. First, target indoor air concentrations for some compounds are lower than typical laboratory detection limits. Second, indoor air usually contains dozens of detectable compounds attributable to building materials and consumer products in addition to any potential contributions from vapor intrusion. Third, even outdoor air quality in some locations can exceed target cleanup concentrations for some compounds. Finally, indoor air quality varies between buildings because of differences in design, construction, thermal efficiency and ventilation, and even within a single building over time because of changes in barometric pressure, wind, temperature, HVAC operation, and the occupant’s activities. Therefore, it is difficult to assess the vapor intrusion pathway with indoor air sampling and analysis alone.

Cases that Demonstrate the Strategic Approach Sometimes a single important factor dominates the vapor intrusion pathway assessment, and a focused study will demonstrate the impact of this factor in sufficient detail to provide a robust decision. A few examples from the authors’ experiences are presented in brief below to demonstrate this.

A site in Massachusetts has trichloroethene (TCE) in groundwater at concentrations above 10,000 g/L at a depth of only 30 feet beneath a residential neighborhood with basements. However, the rate of infiltration of rainfall, lawn watering and snowmelt is sufficient to form a blanket of fresh water about 5 feet thick at the water table that acts as a vapor barrier and prevents off-gassing of TCE from the groundwater. Focused data collection to demonstrate this included depth discrete groundwater sampling using the Waterloo Profiler ™ method, and selected soil gas and indoor air sampling.

A site in Florida has a former industrial building that has been redeveloped into mixed commercial and industrial uses. TCE is present in groundwater beneath the site at concentrations above 10,000 g/L. The building shields the groundwater from recharge, so there is not likely to be any fresh water lens. However, the building is air-conditioned year-round by about 100 rooftop units that cool atmospheric air and blow it into the building. The flows are balanced to provide 20% new air, which maintains a slight positive pressure to inhibit subsurface vapor intrusion, plus sufficient ventilation to dilute any vapors potentially entering by diffusion to well below concentrations of concern.

A site with residences built on layered geologic materials in a humid climate had a series of fine-textured layers impeding infiltration sufficiently to create saturated layers that act as a vapor barrier. Vapor monitoring wells were installed and sealed at different depths and a pneumatic pumping test was performed, which showed that the different intervals can be pneumatically isolated by the function of the fine-textured layers as vapor barriers.

In all of these cases, the key to understanding the vapor intrusion and attenuation processes required unique data in addition to conventional groundwater, soil gas, indoor air and outdoor air data. Together with the conventional data and mathematical models, the assessments provide multiple lines of evidence to support their conclusions. High quality data is essential, because if any one line of evidence has a systematic bias, the various lines of evidence may not be mutually supportive.

Risk-Based Decisions To make risk-based decisions, it is necessary to assess whether vapor intrusion is likely to pose unacceptable risks. If groundwater and soil gas concentrations are lower than cautious screening values, it may not be necessary to collect supplemental field data. Otherwise, some data will usually be justified to evaluate the sources of the compounds of concern. It may also be necessary to evaluate spatial and temporal variability of vapor concentrations in selected representative locations.

In areas where unacceptable vapor intrusion is not easily refuted, it may be preferable to install a sub-slab depressurization system (e.g. radon mitigation system) to protect indoor air quality proactively. These systems are effective and can often be installed for less than the cost of a detailed site investigation. However, long-term operation, maintenance and monitoring of applied vacuum is usually required, and this may be undesirable. It will often be necessary to demonstrate a reasonable understanding of the vapor fate and transport processes to justify which properties are included and excluded from mitigation. It may also be necessary to collect such data to justify any future decommissioning of any mitigation systems.

Conclusion The keys to successful vapor intrusion evaluations are selecting the tools and scope of investigation sufficient for protection of human health and regulatory compliance and communicating the findings logically and clearly to all stakeholders. High quality data is essential to decision-making and to maintaining professional credibility throughout the process. The knowledge base for vapor intrusion continues to expand and efforts are underway to compile and share the data being collected at an increasing number of sites. It is clear that the vapor intrusion pathway evaluations will be required at many sites and effective and efficient methods to assess this pathway are needed. The strategic approach described above will focus the assessment scope and methods to the most important processes and conditions, thereby providing a cost-effective and technically defensible outcome.

Todd McAlary and Robert Ettinger are with GeoSyntec Consultants, Inc., a leading provider of vapor intrusion services. For further details of their services and experience, see: http://www.geosyntec.com/UI/Default.aspx?m=ViewPractice&p=8 and http://www.geosyntec.com/UI/Default.aspx?m=ListProjects