Air Emissions Risk Analysis (AERA) — Multi-pathway Risk Analysis

The AERA process includes conducting a screening level multi-pathway risk analysis to determine if volatile and persistent, bioaccumulative and toxic chemicals (PBTs) may be of concern. The process evolved from previous MPCA air toxics risk assessment efforts and follows risk assessment principles and guidance provided by the Minnesota Department of Health and U.S. EPA.  The results from this process are estimated numeric risk values from direct inhalation exposure and from indirect ingestion exposures. The indirect exposure scenarios involve the potential for nearby residents and farmers to be exposed to various exposure media such as soils, water and foods (multimedia).  This multi-pathway component of the AERA process is specifically influenced by EPA’s Human Health Risk Assessment Protocol for Hazardous Waste Combustion Facilities (HHRAP). Some portions of the AERA process are also influenced by EPA’s three-volume Air Toxics Risk Assessment Reference Library. The current AERA process focuses on adult receptors and does not assess children. Risk calculations made in the Risk Analysis Screening Spreadsheet and Q/CHI spreadsheet do not include the fish consumption pathway. Risks from mercury exposure through the fish consumption pathway can be estimated using the MPCA Mercury Risk Estimation Method (MMREM). The AERA process does, however, consider potentially sensitive receptors through the review of receptor location maps and other information provided by the project.

Who Should Complete a Multi-pathway Risk Analysis using the RASS and/or the Q/CHI Spreadsheet?

If the chemicals of potential interest (COPI) list indicates that PBTs are emitted from the facility and the exposure analysis indicates the presence of farmers and/or residents in the area of impact, a multipathway analysis will be done using MPCA’s multimedia factors.  Multimedia factors are defined as the ratio of the non-inhalation risks to the inhalation risks.  For many volatile compounds, the inhalation risks are higher than the non-inhalation risks, so the multimedia factor will be less than one.  On the other hand, chemicals that accumulate in the food chain may have high non-inhalation risks and a multimedia factor greater than one.  If the non-inhalation risk is equal to the inhalation risk, then the multimedia factor will equal one.

The multimedia factors were initially developed from air dispersion modeling files for a single generic facility imported into IRAP-h View © software (IRAP) for risk analysis.  The non-inhalation risks for each chemical were extracted directly from IRAP.  The inhalation risks were calculated by multiplying the air concentrations (extracted from IRAP) times the inhalation unit risk (taken from MPCA’s hierarchy of inhalation health benchmarks, which can be found in the RASS). 

Only those chemicals with a ratio of one (rounded to integer values) or higher were assigned multimedia factors.  The assigned values can be found on the MMFactors page of the RASS.  The factors represent a screening estimate of the number of times that the non-inhalation risk is greater than the inhalation risk.  Multimedia factors were developed for two scenarios, the adult farmer and the adult resident.

How are Multi-pathway Risks Calculated in the RASS and the Q/CHI Spreadsheet?

Risk results from the inhalation pathway are multiplied by the Multi-pathway Screening Factors (MPSF) on a pollutant specific basis to calculate a screening level risk from the ingestion pathway. Multi-pathway Screening Factors are defined as ratios of the maximum risk from the indirect (ingestion) exposure routes to the maximum risk from the direct (inhalation) exposure route. Multi-pathway screening factors were developed for two scenarios, the adult farmer and the adult resident

For many volatile compounds, the inhalation risks are higher than the non-inhalation risks, so the MPSF will be less than one. Only those pollutants with a MPSF ratio of one or higher will be assessed for the indirect exposure pathway.

Development of the new Multi-pathway Screening Factors (previously, “Multimedia Factors”)

Multi-pathway screening factors were only developed for those organic pollutants in the RASS that were determined to be persistent, bioaccumulative and toxic (PBTs) using the EPA PBT profiler model and RASS-listed metals included in a more comprehensive list adopted by the European Union. This MPCA PBT list was further pared by eliminating pollutants that did not have both oral and inhalation toxicity data and sufficient chemical and physical property data.

Ingestion to inhalation risk ratios (MPSFs) were then calculated using the AERMOD air dispersion model and the IRAP-h View TM risk model and are published in the study referenced below. The study includes a sensitivity analysis of the impact of various model input parameters found to affect the ingestion to inhalation risk ratios. Results from the 19 cases included in the sensitivity analysis were used to focus the MPSF calculations. Several exposure pathways were eliminated from the MPSF calculations, including drinking water, fish consumption, and infant exposure to polychlorinated dibenzodioxins, polychlorinated dibenzofurans, and polychlorinated biphenyls via breast milk ingestion.

This study has shown that the MPSFs are usually highest at the location of the maximum inhalation risk and therefore can be used as multipliers in risk analyses to estimate reasonable upper-bounds for ingestion risks. 

The recommended Multi-Pathway Screening Factors, as well as a detailed description of their derivation, are presented in the April 2009 issue of Journal of the Air & Waste Management Association (Pratt and Dymond 2009).  The MPSFactors worksheet of the updated RASS and Q/CHI spreadsheets (MPCA 2009) reflect these published values.

Information for Estimating Potential Human Health Risks using Central Tendency Values for Human Exposure Factors

An important element of a human health risk assessment is the transparent communication of uncertainty and variability. A portion of the uncertainty in a final risk estimate stems directly from the assumptions used to characterize potential human exposures. The USEPA and MPCA recommend estimating risks based on a set of default exposure assumptions, called the “Reasonable Maximum Exposure” (RME). One approach to communicate the uncertainty associated with the default exposure assumptions is to provide risk estimates using multiple human exposure assumptions. Thus, MPCA staff reviewed the human exposure data in the EPA Exposure Factor Handbooks (1997, updated Children’s EFH, 2008) and are providing guidance for estimating risk using central tendency human exposure factors. Risk results using central tendency human exposure factors should not replace risk estimates based on the RME and should not be considered a refinement to screening level risk assessments that follow MPCA’s Air Emissions Risk Analysis (AERA) guidance.

Human health risk assessments are conducted with the MPCA for permitting and environmental review purposes. Initial screening level risk assessments are sufficient for understanding the potential impacts for many projects. Some projects may require a broader scope and more extensive environmental review due to: high number of sources, matching pollutant concentrations in space and/or time, neighborhood concerns, interest or requests, significant emissions, etc. More complex multi-pathway risk modeling software is used for these more complex analyses (e.g. IRAP h-view from Lakes Environmental Software or Breeze from Trinity Consultants). The two multi-pathway risk assessment tools mentioned above are based on calculations described in the Human Health Risk Assessment Protocol (HHRAP, 2005). Including central tendency exposure estimates would generally be appropriate in these larger multi-pathway risk assessments, where more discussion of uncertainty is warranted. 

The exposure duration and consumption rates used in the default settings of multi-pathway risk software are chosen from national studies examining where people spend their time, how much they eat of certain foods or how often they inhale. These studies result in a range of data (including high, low and mid-range). Risk calculations based on central tendency exposure estimates are exactly the same as RME risk calculations except that they use central tendency estimates (such as means or medians) for exposure durations and consumption rates. Included in Table 1 are guidance values for calculating risk estimates using central tendency human exposure and maximum exposure rates.

Table 1. Suggested Exposure Factor Values for Risk Estimates based on Central Tendency Human Exposure Estimates

¹These data are based on national means that have been time-weighted for age and in the case of the hypothetical farmer adjusted with a factor for households who farm. Since the data are based on means for the RME, there is no justification to change this for the central tendency estimate.

²This factor describes the portion of the items produced on site that are considered contaminated. In general, the scale resolution for modeled deposition is not adequate to describe the portion of onsite food or soil contamination. The RME suggested value is 100%, and there is not justification at this time to change this value for the central tendency estimate. The amount of food that is grown onsite (i.e. contaminated) and consumed is accounted for in the consumption rate value development.

³Mean residency period reported in the EPA Exposure Factors Handbook.

⁴Mean time spent at home (California) from EPA Exposure Factors Handbook used in place of national average.

Generally, the above suggested exposure factors are used with modeled or measured air concentrations to estimate potential risks to human health based by a comparison to an air concentration that is unlikely to pose harm. An example equation is included below:

Groupings of assumptions about human exposures such as consumption rates and exposure durations by exposure pathways are called “Exposure Scenarios.” For example, a grouping of assumptions related to consuming produce grown at a specific location, accidentally ingesting onsite soil and breathing outdoor air is a hypothetical exposure scenario described using the label “resident.” A person inhaling outdoor air, consuming many more products grown on-site (e.g. beef, milk, produce, pork, chicken, eggs) and accidentally ingesting onsite soil is described with the label “farmer.” Consumption rates of each individual food product, etc. may vary depending on the intended population to be assessed (reasonable maximum, central tendency, offsite worker, etc.).

Table 2. Default Exposure Factors for Consumption Rates and Bodyweight from HHRAP, 2005

Notes: Farmers consume all products. Residents consume products with *. Body weight: adult=70 kg, child 15 kg

There are other sources of uncertainty and variation in risk assessments other than those associated with human exposure. Modeled or measured air concentrations fluctuate depending on factors such as wind direction. Toxicity values have specific uncertainty factors associated with them dependent on the study/ies used to derive the value. This guidance does not address these other sources of uncertainty and the MPCA does not currently advise deviating from the AERA guidance with respect to characterizing other sources of uncertainty in risk assessments.²

Reasonable Maximum Exposure - maximum exposure reasonably expected to occur in a population.

² Screening level AERAs report estimated risks using the MPCA Risk Analysis Screening Spreadsheet (RASS). The RASS calculations for estimated inhalation risks are based on a lifetime toxicity assessment at locations of maximal air concentrations. The calculations for indirect (ingestion-based) estimations of risk are based on the HHRAP default values (RME) and scaled from inhalation estimates using multi-pathway screening factors. Since the RASS is a risk screening tool, a central tendency estimation using a scaling factor for central tendency exposures is not planned at this time.

References

• U.S. EPA. Exposure Factors Handbook (Final Report) 1997. U.S. Environmental Protection Agency, Washington, DC, EPA/600/P-95/002F a-c, 1997
• U.S. EPA. Child-Specific Exposure Factors Handbook (Final Report) 2008. U.S. Environmental Protection Agency, Washington, D.C., EPA/600/R-06/096F, 2008.
• U.S. EPA, Office of Solid Waste EPA530-R-05-006 - “Human Health Risk Assessment Protocol (HHRAP) for Hazardous Waste Combustion Facilities (Final)” September 2005

To Learn More

Risk analysis and risk assessment are complex and this summary provides only a basic overview.  To learn more about risk assessment or the AERA process, please see the following:

• EPA’s Human Health Risk Assessment Protocol for Hazardous Waste Combustion Facilities
• Air Toxics Risk Assessment Reference Library
• MPCA AERA Guidance

Who to Contact

Environmental Analysis and Outcomes Division, Risk Evaluation/Air Modeling Unit

• Greg Pratt, 651-757-2655
• Mary Dymond, 651-757-2327
• Kristie Ellickson, 651-757-2336