An exposure assessment identifies the means by which humans can come into contact with environmental pollutants. Exposure depends on several elements: the activity patterns of people living and frequenting the vicinity surrounding the facility; how the pollutant gets from the source to the individual (exposure pathway), how much of the pollutant is available to get into the body (exposure concentration), how the pollutant gets into the body (exposure route), and how much of the pollutant is absorbed and available for interaction with biological receptors, organs or cells within the body (dose).
The first step of the exposure assessment is characterizing the neighborhood and potential receptors surrounding the facility.
A description of the general locale of the proposed project needs to be included in the AERA_02 Qualitative Information:
- AERA-02 Qualitative Information (August 2011)
The description needs to identify neighborhood characteristics and areas of industry and other air emission sources of significance in the area. Of specific interest may be:
- Population demographics within appropriate census tracts surrounding a facility
- Air Emission Point Sources identified by MPCA’s air toxics emissions inventory
- Other air emission sources, industrial facilities, or environmentally sensitive areas
- Locations of sensitive receptors
Maps can be useful in clarifying available information. Possible resources for locating much of the information recommended in this section are provided in the qualitative risk characterization section.
Buffer distances based on stack height are used to determine appropriate distances for evaluating qualitative information about the setting of an emissions source. Incremental ambient air concentrations and risk estimate from an emissions source occur within a distance that depends upon stack height (among other factors). As a rule of thumb, the greater the stack height, the greater the distance to the maximum modeled air toxic concentration, deposition, and risk estimate.
The buffer distances are judgments of areas around an emission source that will encompass most emissions of concern and are based on MPCA staff’s experience in estimating air toxic concentrations and deposition fluxes. For the purposes of AERA guidance, the MPCA recommends the following buffer distances for maps showing sensitive receptors, general neighborhood information, and nearby permitted air emission facilities.
Table 2: Buffer distances for placement of receptors, based on stack height
||1.5 (approximately 1 mile)
|50 to 100
||3 (approximately 2 miles)
||10 (approximately 6 miles)
Zoning and land use
Zoning and land use maps need to be based on a 10-kilometer radius regardless of stack height. If zoning or land use information exists for a city, township, or county that does not specifically include the 10-kilometer radius surrounding the facility, this information may be considered inadequate. Maps can be supplemented with relevant ordinances that would inform potential exposures, (e.g., raising chickens in town or prohibitions of livestock). The MPCA recognizes that some areas of the state do not have specific zoning information available.
Figure 1: Zoning map example
A land use map shows current land use within 10 kilometers of the facility. Land use maps include information such as areas of residential, commercial and industrial use, farms, forests, and waterways. If no map is provided, the most restrictive land use will be assumed. The project proposer needs to also state whether the land is used for purposes other than those designated on the land use maps.
The MPCA considers “reasonable potential future land use” when evaluating potential impacts to nearby property. Definitions for “reasonable potential future land use” come from U.S. Environmental Protection Agency’s (USEPA) Human Health Risk Assessment Protocol for Hazardous Waste Combustion Facilities (HHRAP).
Figure 2: Zoning and land use map example
Map created by Barr Engineering for purposes of AERA submittal.
For purposes of an AERA, sensitive receptors are groups of people who, due to their age or health status, are sensitive to air pollutants. Sensitive receptors may include infants, children, pregnant women, elderly, asthmatics, athletes, or immuno-compromised people.
The project proposer’s submittals need to include maps identifying schools, daycare facilities, hospitals, nursing homes, recreational areas (including parks, tennis courts and swimming pools), senior centers, and other public or private facilities at which sensitive people may be congregated. The maps may be a sketch with distances and receptor locations identified. If a map is not readily available or feasible, these types of potential receptors need to be described in writing and identified in the area around the facility. The maps or descriptions of sensitive receptor locations need to include the area within a radius of at least 1.5 kilometers from the facility.
Figure 3: Sensitive receptor map example
Provided by Natural Resources Group, Inc. for the purposes of an AERA submittal
Various types of farming (e.g., beef farming, dairy cows, chickens, urban gardening) in the vicinity may generate foods that can be consumed by people living on the farms or by nearby residents. In addition to existing farming locations, the MPCA considers “reasonable potential future land use” in assessing potential risks from farms. According to U.S. EPA’s Human Health Risk Assessment Protocol for Hazardous Waste Combustion Facilities (HHRAP), three examples of reasonable potential future land use are:
- Rural area characterized as undeveloped open fields that could reasonably be expected to become farmland if it can support agricultural activities.
- Rural area currently characterized by open fields and intermittent housing that could reasonably be expected to become a residential subdivision.
- An area currently characterized as an industrial area would not reasonably be expected to become farmland.
If no information is available regarding land use, the default assumption will be that a farmer could be impacted by facility emissions, and the farmer scenario risks will be used as a basis for decisions. If land use indicates that farms do not exist within the appropriate radius, only resident risks will be assessed. Resident exposures could include ingesting chickens, eggs, or other livestock that are raised on the property if allowed by ordinances.
Water bodies in the vicinity of the facility may be impacted by the deposition of facility emissions. The distance from the source to where air pollutants deposit depends in part on the stack or release height.
The MPCA recommends the following buffer distances for maps that show lakes, rivers and streams. Water bodies outside the specified area that may be fed by rivers and streams lying within the radius of interest also need to be shown.
Table 3: Buffer distances for maps with water bodies
|< 100 m
||3 km (approximately 2 miles)
|> 100 m
||10 km (approximately 6 miles)
Fishable water bodies
A "fishable water body” typically contains water year-round in a year that receives at least 75 percent of the normal annual precipitation for that area. Whether a water body has public access is also an important consideration.
Any fishable water body occurring at the area of maximum deposition needs to be evaluated in a MMREM based analyses. If the area of maximum deposition does not fall on a fishable water body, the project proposer needs to determine which water body is nearest to the area of maximum deposition. The nearest water body may represent the worst-case impacts at the screening level; however it also may not be clear which water body would be most impacted. There may be a water body with more impact because it has less dilution from its watershed and more fishing. If it is not clear which water bodies need to be evaluated, MPCA staff should be contacted.
Nearby permitted air emission facilities
The project proposer will be asked to provide a map and/or list of permitted air emission facilities and sources within the proper radius of the facility.
- List of nearby air emissions sources: MPCA’s What’s in My Neighborhood website provides a list of facilities within a specified radius. Note: The user must zoom in close enough to see the facilities on the screen.
- Map of nearby air emission sources: Using MPCA’s Air data webpage, click “search using a map” on the bottom left of the pop-up window, then click on the data you would like to search (e.g., point sources emissions, air monitor location
Figures 4 a and b: Maps of nearby air emission sources
National data on human consumption and behavior have been used by the U.S. EPA and other regulatory agencies to develop methods for assessing exposure of humans to environmental contaminants for use in regulatory risk assessments.
The following exposure scenarios are based on the default exposure scenarios recommended in the U.S. EPA Human Health Risk Assessment Protocol for Hazardous Waste Combustion Facilities (HHRAP, 2005). Receptor types evaluated in AERAs and recommended exposure pathways are summarized below.
Table 4: Recommended exposure scenarios
||Recommended exposure scenariosa
|Inhalation of vapors and particles
|Incidental ingestion of soil
|Ingestion of drinking water from surface water sources
|Ingestion of homegrown produce
|Ingestion of homegrown beef
|Ingestion of milk from homegrown cows
|Ingestion of homegrown chicken
|Ingestion of eggs from homegrown chickens
|Ingestion of homegrown pork
|Ingestion of fish
* Pathway is included in exposure scenario.
-- Pathway is not included in exposure scenario.
a Exposure scenarios are defined as a combination of exposure pathways evaluated for a receptor at a specific location.
b Site-specific exposure setting characteristics (e.g., presence of ponds on farms, or presence of ponds or small livestock within semi-rural residential areas) warrants the permitting authority to consider adding this exposure pathway to the scenario.
Table 5: The exposure routes and durations evaluated in AERA
||2 weeks to 3 months
(approximate lifetime assessments)
||30 years, resident
40 years, farmer
The acute inhalation exposure scenario is used to describe potential adverse effects from breathing hourly maximal air concentrations of facility air toxics at locations where this exposure could possibly occur. This type of exposure includes those living or working nearby; someone running or biking near a facility; snowmobiling along the facility boundary; or a delivery person waiting for their truck to be emptied. An assessment of acute inhalation is rarely scoped out of an AERA.
A “resident” is a USEPA-developed exposure scenario assessed over an approximate adult human lifetime for inhalation and 30 years for ingestion. This hypothetical “Resident” inhales air, indirectly ingests soil, and ingests home-grown produce that could be affected by facility air emissions. Maximum annual average air concentrations, derived from 5 years of meteorological data, are considered inhalation exposure concentrations (IEC) in this exposure scenario. The exposure durations may be limited to the life of a project (e.g., a 20 year mine plan, if that is acceptable upon MPCA review); however, the “resident” scenario generally assumes a lifetime exposure for the inhalation pathway. Less than lifetime exposures consistent with USEPA guidance are assumed for the ingestion pathways. An assessment of a potential resident is rarely scoped out of an AERA.
A “farmer” is an USEPA-developed exposure scenario assessed over an approximate adult human lifetime for inhalation and 40 years for ingestion. This hypothetical “Farmer” inhales air, indirectly ingests soil, ingests home-grown produce, drinks home-produced milk, and eats home-grown meat products (pork, beef, chicken eggs, and chicken) that could be impacted by facility air emissions. Maximum annual average air concentrations, derived from 5 years of meteorological data, are considered in this exposure scenario. The exposure duration may be limited to the life of a project (e.g., a 20 year mine plan, if that is acceptable upon MPCA review); however, the “farmer” scenario generally assumes a lifetime exposure for the inhalation pathway. Less than lifetime exposures consistent with USEPA guidance are assumed for the ingestion pathways. An assessment of a potential “farmer” is not relevant to an AERA conducted for an area without production of food products or animal husbandry. Zoning and/or land use information may be required if the “Farmer” scenario is scoped out of an AERA.
In some cases where adequate land use documentation is provided, consideration of the “Farmer” is not appropriate. This is the case in densely populated urban areas where animal husbandry is not allowable. For this reason an “Urban Gardener” exposure scenario was developed by MPCA staff so that some reasonable assessment of ingestion-based exposures in urban areas was possible. The ingestion rates used in this exposure scenario were drawn from those in the U. S. EPA guidance for the “farmer”. An “urban gardener” exposure scenario assumes a hypothetical person inhales air, indirectly ingests soil, ingests home-grown produce, and eats home-raised chicken eggs. Maximum annual average air concentrations, derived from 5 years of meteorological data, are considered in this scenario. The exposure durations may be limited to the life of a project (e.g., 20 year mine plan, if that is acceptable upon MPCA review); however, the “Urban Gardener” scenario generally assumes a lifetime exposure for the inhalation pathway. Less than lifetime exposures consistent with USEPA guidance are assumed for the ingestion pathways.
A fisher is considered in cases where one or more fishable water bodies may be impacted by emissions from a facility that emits persistent and bioaccumulative toxics (PBTs). Ingestion risks from consuming fish from an impacted water body are estimated independently from the other exposure scenario risks and may be added to risks estimated for the resident, farmer or urban gardener if it is reasonable.
A subchronic inhalation exposure scenario is employed in order to assess a mid-term exposure duration. This exposure scenario may be considered as being approximately a month long vacation, or work related assignment. Maximum monthly air concentrations, derived from 5 years of meteorological data, are considered in this exposure scenario.
Inhalation exposure for non-continuous lifetime exposures (e.g., offsite worker)
After air concentrations are estimated, it may be reasonable under some circumstances to adjust the annual estimated air concentration for less than lifetime continuous exposure. Circumstances where this might be reasonable would be when assessing risks to workers in an area zoned commercially next to the facility under evaluation. The inhalation exposure concentration (IEC) of the pollutant in air would be:
Exposure scenario refinement: Eliminating pathways, scenarios, and additional information
Exposure scenarios may be scoped out of quantitative assessment based on land use designations. If there is not currently the potential for, nor the future potential for a certain land use connected to a specific exposure scenario, that exposure scenario may be scoped out of the quantitative risk results. Examples of this type of scoping include:
- A residential exposure scenario will not be conducted in a land use area zoned as industrial.
- A farmer exposure scenario, including all of the related exposure pathways, will not be included in an urban area that is not zoned for extensive animal husbandry.
- The likelihood of a farmstead existing at the location of maximum air concentration is small because of the large distance between farms. Therefore, the Farmer risk may be greatly over-predicted if this type of receptor is placed at the maximum air concentration. In that case, the project proposer may choose to evaluate risks at the location of the closest actual farm in addition to a hypothetical farmer at the location of maximum air concentration.
- Acute inhalation exposures are rarely, if ever, scoped out of an analysis.
Reasonable maximum and central tendency exposure assumptions in a Level 3 AERA
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 U.S. EPA and MPCA recommend estimating risks based on a set of default exposure assumptions called the “Reasonable Maximum Exposure” (RME = the maximum exposure reasonably expected to occur in a population) (see Table 6 below.). The goal of RME is to combine upper-bound and mid-range exposure factors so that the result represents an exposure scenario that is both protective and reasonable; not the worst possible case (USEPA OSWER directive). Some of these factors are central tendency (ingestion rates), and other factors used in final risk estimates are maximal values (air concentrations).
The recommended exposure concentration is a “conservative estimate of the media average contacted over the exposure period”. MPCA recommends using the maximum annual average air concentration and the maximum hourly average air concentration as exposure point concentrations. These concentrations fall within USEPA’s definition of a maximally exposed individual. The final AERA results reflect a RME exposure scenario by combining these conservative concentration estimates with refined air dispersion modeling and central tendency ingestion rates.
One approach to communicate the uncertainty associated with the default exposure assumptions is to provide risk estimates using multiple human exposure assumptions. MPCA staff reviewed the human exposure data in the USEPA Exposure Factor Handbooks (1997, updated Children’s EFH, 2008, 2011) and provide the following guidance for estimating risk using central tendency human exposure factors. However, risk results using central tendency human exposure factors need not replace risk estimates based on the RME. Furthermore, they need not be considered a refinement to screening level risk assessments that follow MPCA’s AERA guidance, unless under very rare circumstances, there is appropriate and adequate site specific human exposure data. Presenting central tendency exposure estimates may be most appropriate in larger, more complicated, 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 (HHRAP, 2005) are chosen from US national studies examining where people spend their time, how much they eat of certain foods, or with what frequency 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 frequency. Included in Table 6 are guidance values for calculating risk estimates using both reasonable maximum and central tendency exposure assumptions.
Table 6: Default exposure assumptions for RME human exposure estimates and suggested exposure factor values for risk estimates based on central tendency estimates
||Reasonable maximum exposure
||Central tendency exposure
|Consumption rates (Table 7)
|Percent contaminated food
|Body weight (Table 7)
|Exposure duration (adult)
||30 years resident
40 years farmer
|Inhalation exposure time
||24 hours/day adult and child
||18 hours/day child
19 hours/day adult
1These data are based on US national means that have been time-weighted for age and in the case of the 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 no 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 USEPA Exposure Factors Handbook.
4Mean time spent at home (California) from USEPA Exposure Factors Handbook used in place of a US national average.
Intake assumptions and estimation
Generally, the above suggested exposure factors are multiplied by modeled or measured media concentrations to estimate human pollutant exposure (intake). An example equation is included below:
Consumption rate studies are used to estimate doses from ingesting pollutants in foods and from incidental ingestion of soil. Consumption rates may vary depending on the hypothetical exposure scenario or population under consideration (resident, farmer, adult or child, etc.). The default consumption rates recommended in the HHRAP, 2005 are included in the Table 2 below.
Default ingestion exposure assumptions
Table 7: Default exposure factors for consumption rates and bodyweight from HHRAP, 2005
|HHRAP default (approximate pounds per week)
|Exposed vegetables (e.g., tomatoes, peppers)*
|Root vegetables consumption (e.g., potatoes, turnips, carrots)*
|Protected vegetables consumption (e.g., winter squash)*
|| 15 pints/week
|Indirect Soil Ingestion*
|Farmers Consume All Products. Residents Consume Products identified by " * "
Fish consumption rates
Fish consumption rates that are more representative of Minnesota fishers are used rather than the HHRAP default values for both HHRAP-based analyses and the Minnesota Mercury Risk Estimation Method (MMREM).
The subsistence fisher ingestion rate was taken from USEPA’s Guidance for Assessing Chemical Contaminant Data for Use in Fish Advisories (USEPA, 2000). The recreational fisher ingestion rate is consistent with Minnesota Department of Health fish consumption advice. Adult daily doses assume body weights of 70 kg. Child fish consumption rates are calculated using the HHRAP ratio of adult to child fish consumption rates.
MPCA may request a higher level fish consumption rate that is based on Native American Treaty rights. This higher level exposure assessment will be requested if facility emissions may impact tribal land, or waters of interest.
Table 8: Ingestion rates recommended for the subsistence, recreational, and Native American fishers.
|Raw fish tissue consumption rate(g/day)
||Approx. ½ lb fish 4-5 times a week (adult)
||Approx. ½ lb freshwater fish per week (adult)
|Native American treaty rights-based
||Approx. ½ lb fish 7 times a week (adult)
Ambient air monitoring is a measurement of air pollutant concentrations at a specific time and location: Air monitoring stations are sited according to criteria established by the USEPA (CFR 40 Part 58 Appendix E). These include requirements to be within ambient air (i.e., areas where the public has access, outside facility fenceline) as well as requirements related to the effective operation of the monitoring equipment (spacing from obstructions, horizontal and vertical placement, etc). Monitoring is limited by expense, equipment capabilities, availability of a power source, security, and time.
Ambient air monitoring data are used in cumulative AERAs to represent air toxics concentrations and risks from surrounding sources or to confirm modeled air concentrations. However, identifying specific sources of air pollution from monitoring data is complex and requires many air measurements over time with site specific meteorological data.
Ambient air monitoring data are generally not preferred to represent exposure concentrations in AERAs because of the difficulty in ensuring a combination of the worst case operating and meteorological conditions that are specific to the facility under review.
Air dispersion modeling
Air dispersion modeling is the preferred method for estimating air concentrations used in AERAs. Air dispersion modeling is the process of simulating the movement of air pollutants after they are emitted from a source in order to estimate the concentrations of pollutants at locations around the source.
In AERAs, air dispersion modeling is used to estimate air pollutant concentrations on the fenceline and in the area surrounding a facility. Air dispersion modeling is done using an air quality modeling system that has been developed and refined over many years by the USEPA and partners. The models have been tested against measurements to verify accuracy.
Air dispersion modeling information is preferred by the MPCA for use in AERAs because air dispersion modeling: can combine worst case meteorological conditions and worst case operational conditions, is not limited by equipment detection limits, and can include information about specific air toxics that are traceable to emission units. This allows the results to inform final permit limits on specific sources and emission units within a facility.
A modeling analysis requires inputs of pollutant emission rates along with the parameters that characterize the release from each source (e.g., height, temperature, exit velocity), plus data on surrounding terrain, buildings, meteorology and receptor locations (i.e., where exposure concentration calculations will be made). The air dispersion model provides estimations of air concentrations (and deposition if needed) at each selected location.
General air dispersion modeling guidance for AERAs
In general, recommendations and guidance for air dispersion modeling follow the MPCA Air Dispersion Modeling Guidance. When dispersion modeling is performed for use in AERAs, the maximum annual modeled air concentrations and the first high modeled hourly air concentrations are used.
Air dispersion and deposition modeling refinement levels
Air dispersion modeling can be done at a screening level or with increasing levels of refinement. Since all of the AERA tools listed at the beginning of this guidance document typically use the same toxicity values and exposure scenarios and any level of emissions/operating assumptions, the main differences between them are how air concentrations and other media concentrations (soil, water, food) are calculated.
The more refined analyses require more data and are more resource intensive. In general there are three levels of air dispersion and deposition modeling completed for MPCA AERAs:
- Level 1/Initial screening: Screening dispersion modeling and food chain analysis using MPCA Risk Assessment Screening Spreadsheet (RASS) and the embedded dispersion factor and multi-pathway screening factor look-up tables
- Level 2: Refined dispersion modeling using AERMOD, and screening food chain analysis using spreadsheet tools (RASS and Q/CHI)
- Level 3: Refined dispersion modeling using AERMOD, and refined food chain analysis using commercially available software that follows USEPA’s Human Health Risk Assessment Protocol (HHRAP, 2005)
Level 1: Using the RASS default screen
Level one AERA analyses are most successful when there is one, or only a few emission stacks, when there are relatively low levels of emissions, or the facility fenceline ensures receptors are well removed from the facility. When evaluation of the fish pathway is necessary, Level 2 or 3 refined dispersion and deposition modeling is required.
RASS dispersion factors
The RASS contains a look-up table of default hourly and annual dispersion factors (in terms of μg/m3 per g/s). These were generated from many AERMOD modeling runs generally reflecting worst-case conditions including but not limited to: stack diameter, stack exit velocity, stack exit temperature, meteorological conditions, and stack-to-building geometry. [Note: In some very specific circumstances, the use of a RASS may provide results that are nearly as refined as Level 2 and 3 assessments (e.g., for a single short stack that extends just above building height)].
The default dispersion factors in the RASS look-up table are based on the stack height and receptor/fenceline distance input by the user on the “StackDisp” tab of the RASS. The factors are combined within the spreadsheet with hourly or annual emissions to estimate worst-case air concentrations at or beyond the receptor distance input by the RASS user. Concentrations are estimated at ground level receptors only; receptors at elevated levels are not considered in the RASS at this time.
Generally, the nearest receptor distance input to the RASS (on the “StkDisp” tab) is assumed to be at the facility’s fenceline, or at the owned and controlled boundary of the facility. For AERA guidance purposes, fenceline will mean either a physical barrier or a boundary controlled by other means (e.g., fence, security guards). In AERAs, a receptor represents a hypothetical person who is potentially exposed to air pollution. In air dispersion modeling, however, a receptor is a location where the model calculates concentrations and provides results.
If the facility is accessible to the public, the distance to the fenceline or receptor for the acute exposure scenario may be different from the distance considered for sub-chronic and chronic exposures. If physical access to a facility’s property is not restricted, acute impacts need to be assessed at the location of maximum hourly air concentration predicted anywhere (unless it falls over a building, in which case it need only be considered if there is public rooftop access). Chronic risks need to be computed for potential receptors located at the maximum annual air concentration at or beyond the property fenceline. If the facility is not accessible to the public, only one receptor/fenceline distance is entered, and the RASS is only run one time.
Merging Stacks with Similar Dispersion Characteristics
To accommodate multiple stacks more efficiently, it may be helpful to merge stacks with similar dispersion characteristics such as stack height, stack diameter, exit velocity, exit temperature, and proximity to similarly sized buildings. Stacks must be located within approximately 100 meters of each other, near similar sized buildings, and have stack parameters that vary less than 20 percent (USEPA 1992). The equation for merging stacks can be found in the MPCA Air Dispersion Modeling (ADM) Guidance in the Nearby Source Characterization section. The calculation of “M” is what assists in determining if stack parameters vary less than 20%.
Defining ‘Stacks’ for fugitive sources
The RASS evaluates air emissions impacts based on releases through ‘stacks’. Thus the characteristics of emission sources or points that are not stacks (windows and doors or fugitive emission sources) must be modified in some way to allow the RASS to estimate dispersion and risk. Options for modification include entering fugitive emissions in the RASS as though they are emitted through a one-meter stack, or using facility-specific refined dispersion modeling to estimate ambient air concentrations. More detailed options for the modification of fugitive sources are included in the MPCA Air Dispersion Guidance. (K2 83)
Level 2: using the RASS or Q/Chi with site-specific dispersion factors and air concentrations
There are several combinations of methods and tools that can be used to provide more accurate dispersion and deposition modeling than can be provided by the RASS alone. Possible tools for use in Level 2 Refinement include: RASS, AERMOD, Q/Chi spreadsheet.
Using the RASS with AERMOD site-specific dispersion factors
AERMOD may be used to generate unitized dispersion factors for each stack, which are then entered directly into the RASS in the “StkDisp” tab. The MPCA Air Dispersion Modeling Guidance needs to be used to determine receptor placement, meteorological data, and source characterization. Emission rates of 1 g/s are entered into AERMOD for each air emission source.
Using the RASS with site-and-pollutant-specific concentration modeling using AERMOD-Multi-Chem
The Multi-Chem function of AERMOD may be used to calculate specific air toxics concentrations. These air concentrations are entered directly into an unprotected MPCA RASS in a worksheet tab set up by the analyst. The project proposer needs to request an unprotected RASS if this type of modeling is proposed. Multi-Chem is a desirable function for complex facilities with many pollutants and stacks (>50), as it avoids a model run for each air toxic.
Using the Q/CHI spreadsheet
Hourly and annual air toxics emissions must be input into the Q/CHI spreadsheet on the “Emissions” tab. The Q/CHI sums are then calculated in this spreadsheet, and reported on the “Q_CHIs for ADM” tab. These Q/CHI sums are modeled in AERMOD in place of air toxic-specific or unitized emission rates. The AERMOD results are risk estimates, not air concentrations. AERMOD must be run once for each exposure scenario being assessed: acute, chronic non-cancer inhalation, chronic non-cancer indirect, chronic non-cancer total. It may be more efficient to run AERMOD for the indirect and inhalation pathways, and then sum the results for the total risks. This eliminates one model run, and saves resources.
Level 3: Dispersion and deposition modeling for HHRAP-based tools
More refined multi-pathway modeling involves different air modeling options and assumptions than are typically used for screening-level and criteria pollutant modeling. This type of modeling involves the calculation of dispersion factors as well as wet and dry deposition-related factors in AERMOD. These dispersion and deposition factors are then input into a multi-pathway risk model to calculate media concentrations and multi-pathway risk estimates. In general, multi-pathway risk models follow the USEPA’s Human Health Risk Assessment Protocol (HHRAP, 2005), and include model software such as BreezeR and IRAP-h ViewTM.
Guidance Documents for HHRAP-based Tools
Consult MPCA to develop a refined multi-pathway dispersion and deposition modeling protocol. The following information will be helpful in its development.
- MPCA default pollutant characteristics for gas and particle size distribution are provided in the AERA-26 Refined HHRAP-Based Analysis Form at http://www.pca.state.mn.us/index.php/view-document.html?gid=16063
- Guidance for the setup of AERMOD to calculate dispersion (vapor phase) and deposition (wet and dry particulate) are included in Chapter 3 of the HHRAP, 2005 Protocol. This guidance is directed toward the older ISCST3 model; however the basic steps are the same.
- Basic deposition modeling guidance in the MPCA ADM guidance documents and website
- General air toxics modeling guidance for deposition is included in the USEPA SCRAM website.
Fish pathway air dispersion modeling
If non-mercury bioaccumulative pollutants such as dioxins and PAHs are emitted from the facility near a fishable water body, a Level 3 analysis using HHRAP-based tools needs to be performed. The fish consumption pathway is not evaluated in either the RASS or the Q/Chi spreadsheet since air concentrations over water bodies and watersheds must be modeled and deposition from the air into the receiving media must be estimated. If mercury emissions are greater than 3 pounds/year, AERMOD results for mercury can be used in combination with the MMREM spreadsheet to evaluate this pathway.
The RASS or Q/CHI spreadsheet is typically used for the acute inhalation analysis even when completing a refined multi-pathway risk analysis. However, for facilities where it can be demonstrated that there is very little variability in hourly emission rates, HHRAP-based software may be used. In this case the acute inhalation health benchmarks must be entered into the HHRAP based software because MPCA’s acute IHB values are not typically used in HHRAP-based software. In these cases the software must be run separately for the acute and chronic analyses.
Plume depletion is a default assumption embedded in the AERMOD model. Adjustments may be made to the algorithm if there are facility-specific data for particle size distributions. Any changes to this algorithm will require additional MPCA review.
Dispersion modeling information needed for MPCA review
- AERA modeling protocol: This may be the same or a modified version of the criteria pollutant modeling protocol (AQDM-01 and AQDM-02 forms); or may be submitted as the AERA-03 form. For more complex modeling analyses, MPCA requests that the AERA modeling protocol be submitted prior to completion of the analysis. This may eliminate multiple modeling runs. More refined modeling needs to follow standard USEPA and MPCA guidance and practices. The MPCA air dispersion modeling guidance can be found on the Air Dispersion Modeling web page, and includes discussion of the air dispersion models generally accepted by MPCA.
- Information requested in the Air Dispersion Modeling Analysis Form to Support AERA ( AERA-03), including but not limited to:
- Input and output files
- Descriptions of non-default assumptions, level of refinement
- Maps showing property boundaries and fencelines.
- A screening RASS, if used.
Nitrogen dioxide modeling: Special considerations for modeling hourly concentrations
The MPCA assesses an acute inhalation hazard quotient from short term exposures to nitrogen dioxide (NO2). This is conducted in addition to the criteria pollutant modeling completed for comparison to the NAAQS. In general, nitrogen dioxide emissions are available as total NOx which is a combination of NO2, N2O, NOy and, NO. Nitrogen dioxide may be directly emitted from sources, but to a greater extent is formed through atmospheric chemical reactions. Therefore assumptions are made about the percentage of NO2 from NOx that is directly emitted from the stack, as well as the percentage of NO2 that is formed once emitted.
Default assumptions for the nitrogen dioxide modeling tiers are described in the MPCA Air Dispersion Modeling Guidance under the Pollutant Considerations/NO2 section.
Basic air dispersion modeling is expanded to include deposition in order to assess ingestion-based risks from air emissions that deposit into other environmental media. This additional modeling facilitates a multi-pathway risk assessment, in that air particles are allowed (within the model) to deposit from the air onto other environmental media (e.g., soil, water, crops) over time according to their density and particle size. Food chain analyses then use these modeled deposition rates along with other scientific data to estimate uptake of the pollutants into soil, water, produce, fish, livestock and related food products (eggs and milk).
As discussed in the Air dispersion modeling section, there are the following 3 levels of refinement recommended in AERAs. Levels 1 and 2 entail using results from default dispersion/deposition modeling and exposure scenarios, hence are considered more screening.
- Level 1/Initial Screening: Screening dispersion modeling and food chain analysis using MPCA Risk Assessment Screening Spreadsheet (RASS), the embedded look-up table and Multi-Pathway Screening Factors.
- Level 2: Refined dispersion modeling using AERMOD, and screening Food Chain analysis using spreadsheet tools (RASS and Q/CHI) with the Multi-Pathway Screening Factors.
- Level 3: Refined dispersion using AERMOD and refined Food Chain analysis using commercially available software that follows USEPA’s Human Health Risk Assessment Protocol (HHRAP, 2005)
Level 3 Estimating concentrations of environmental media and food-stuffs
Additional facility and MPCA recommended data and information will be required if a Level 3 AERA is being conducted. This information is listed and described in the AERA 26-Refined HHRAP-based Analysis form. In order to inform the exposure scenarios discussed in the next section, and depending on the scope of the AERA, the following environmental media and food concentrations will need to be estimated:
In only very specific cases are drinking water concentrations estimated in an AERA. Most Minnesotans drink groundwater or treated water from municipal water systems; additionally, most facilities under review do not include direct discharge to drinking water supplies.
Watershed and water body parameters
Many Minnesota-specific parameters are available for use in HHRAP-based tools for estimating pollutant water concentrations. These parameters and their sources are summarized below and need to be used unless more relevant site-specific information is available. Other types of information that may be more relevant to the facility location should be proposed in the AERA-26 form along with rationale for its use.
Table 9: Minnesota-specific sources and parameters for use in HHRAP-based tools for estimating pollutant water concentrations
||MN specific value
||HHRAP input location
|Average annual wind speed
||Risk receptor site parameters
||Professional judgment. Based on meteorological data from the MSP airport. Wind speeds found at other locations around the state do not have high variability.
|Fraction (percentage) of watershed that is impervious
||Watershed site parameters
||Professional judgment. Represents the fraction of the watershed that is impervious, such as roadways, pavement. The default value is 5%, which would be appropriate for most applications. This value underestimates the amount of water delivered by watersheds located in urban areas. http://land.umn.edu/index.html
|USLE erodibility factor
||Watershed site parameters
||Value of 0.39 is typical/conservative of average soil types. Used in Universal Soil Loss Equation. Consistent with HHRAP-based software (NC DEHNR 1997, USEPA 1994). This default value is based on a soil organic content of 1%.
|USLE length slope factor
||Watershed site parameters
||Value of 1.5 appropriate for moderately steep slopes; lower values likely for mildly steep slopes. Dependent on the nature of the watershed. HHRAP-based software suggests a default value consistent with NC DEHNR 1997 and USEPA 1994. However, they recommend “using current guidance (U.S. Department of Agriculture 1997; USEPA 1985) in determining watershed specific values for this variable based on site specific information.”
|Air viscosity (temp corrected)
||Risk receptor site parameters
||Used in gas phase transfer coefficient. The air viscosity was calculated for a temperature of 6 oC, the estimated average air temperature of Minnesota.
|Water viscosity (temp corrected)
||Watershed site parameters
||Used in liquid phase transfer coefficient. The value provided is 10 oC and 1 atm, as approximately 10 oC is average temperature of water bodies in Minnesota.
|Sediment delivery empirical slope coefficient
||Risk receptor site parameters
||Vanoni 1975 Used in calculating the sediment delivery to the water body.
|Dry particle deposition velocity
||Risk receptor site parameters
||Upper range of values reported by Pratt, et al (1986) for semivolatile substances. Only use in previous versions of HHRAP-based software. Current HHRAP-based software version uses AERMOD, which calculates deposition.
|Dry vapor depositional velocity
||Upper range of measured values for nitric acid vapor as reported by Pratt, et al (1986). Only use in previous versions of HHRAP-based software. Current HHRAP-based software version uses AERMOD, which calculates deposition.
|Average annual precipitation
||Risk receptor site parameters
||County specific values from the MN Climatology Working Group 2003 at http://climate.umn.edu/img/normals/precip/precip_norm_annual.htm
|Average annual temperature
||Risk receptor site parameters
||County specific values from the MN Climatology Working Group 2003 at http://climate.umn.edu/ .
|Average annual irrigation
||Risk receptor site parameters
||USGS 2000. County specific. Part of the water balance. Data was retrieved for the amount of irrigated land per county (acres) and the total amount of irrigation water used from the USGS, at http://water.usgs.gov/watuse/data/2000/index.html. Based on the number of gallons used each year, acres of farmland, and acres of each county (from 2000 US Census Data).
|Average surface runoff from pervious areas
||Not directly input into HHRAP-based software – calculated from % pervious
||Calculated average surface runoff from pervious areas. Values for surface runoff vary throughout the state. Default values for different regions were provided in Geraghty et al. (1973) – Water Atlas of the United States.
|Water body temperature
||Watershed site parameters
||Estimated from Hondzo and Stefan (1993) study, “Regional Water Temperature Characteristics of Lakes Subjected to Climate Change. Climatic Change. 24:187 211.” Based on the type of water body assessed and the species of fish that might be found in a similar water body.
|Total suspended solids
||Watershed site parameters
||MPCA 2005, calculated Ecoregion values for TSS were taken from the Minnesota Lake Water Quality Assessment Report: Developing Nutrient Criteria (2005). TSS values for rivers are four times the particulate organic carbon content for lakes in the same ecoregion.
|Cover Management Factor (for USLE)
||Watershed site parameters
||MN Agricultural Statistics (2002). County specific.
|USLE rainfall (erosivity) factor
||Watershed site parameters
||Determined by rainfall characteristics of ecoregion. From Wischmeier, W.H. and D.D. Smith. 1978. Predicting Rainfall Erosion Losses – A Guide to Conservation Planning. USDA Handbook 537. Washington, D.C.: U.S. GPO.
||Risk receptor site parameters
||USGS National Water Summary 1987. Calculated by multiplying the total precipitation for a given county by the fraction of precipitation that is evapotranspirated.
If non-recommended values are proposed other than those recommended in this AERA guidance or forms, these values need to be discussed with MPCA staff. The potential effects of other parameter values and calculations used in the assessment need to be explored and explained in the AERA-26 form. This will ensure clarity and transparency of the final risk assessment results.
The equations used to estimate media concentrations are provided in HHRAP Appendix B.
Fish tissue concentrations (mercury)
Some facilities may be requested to assess potential human health risks from mercury exposure for the fish consumption pathway. Monitored fish tissue data are used to estimate non-cancer health effects from exposure to mercury from ingesting fish using specific guidance and tools (MMREM). MPCA risk assessment staff will provide representative Minnesota-specific fish tissue data and/or fish tissue data from USEPA’s National Fish Survey once water bodies are selected.
Due to the uncertainty associated with estimating an accurate average mercury fish tissue concentration, the 95% UCL of the arithmetic mean needs to be used. The USEPA has formulated guidance for calculating the UCL-AM. The guidance has been implemented in the USEPA ProUCL software. This software may be downloaded and run to obtain UCL-AM values from fish tissue data.