J^PP/V United States
bl Environmental Protection Agency
Office of Chemical Safety and
Pollution Prevention
User's Guide:
Integrated Indoor-Outdoor Air Calculator
March 2019
-------
TABLE OF CONTENTS
LIST OF TABLES iv
LIST OF FIGURES v
1 Introduction 1
1.1 Overview and Purpose 1
1.2 Acknowledgements and Disclaimer 2
2 Available Air Modeling Tools 2
2.1 ISC3, AERSCREEN, and SCREEN3 2
2.2 AERMOD 3
2.3 Air Modeling Applications 3
3 General Description of IIOAC 7
3.1 General Description 7
3.2 User Inputs 8
3.2.1 Emission Parameters 8
3.2.2 Chemical and System Parameters 9
3.2.3 Location Parameters 10
3.3 IIOAC Outputs 10
4 Using IIOAC 13
4.1 Downloading and Operating IIOAC 13
4.2 Hardware and Software Requirements for IIOAC 13
4.3 Introduction Tab 13
4.4 Chemical Tab 15
4.5 Source Inputs Tab 16
4.6 Output Tab 19
4.7 Export and Reset Features 20
5 Selection of AERMOD Inputs 20
5.1 Overview 20
5.2 Source Characterization 20
5.2.1 Point Sources 20
5.2.2 Fugitive Sources 22
5.2.3 Area Sources 22
5.3 Emission Characterization 23
5.4 Meteorology and Land Cover 23
5.5 Urban/Rural 31
5.6 Receptors 31
5.7 Recommendation on Default Selections in IIOAC 32
5.7.1 Default Source Scenarios 32
5.7.2 Default Selection of Vapor or Particle 33
ii
-------
5.7.3 Default Selection of Urban or Rural 33
5.7.4 Default Selection of Meteorology 33
5.8 Summary of AERMOD Runs for Point Sources 33
5.9 Summary of AERMOD Runs for Fugitive and Area Sources 35
6 Modeling Approach for Facility Sources 39
6.1 Overview and Assumptions 39
6.2 Post-Processing of AERMOD Hourly Air Concentrations 39
6.3 Calculating Outdoor Air Concentration and Particle Deposition Estimates 41
6.4 Aggregation of Stack and Incinerator Sources into Single Point Source 42
7 Modeling Approach for Area Soil Sources 42
7.1 Overview and Assumptions 42
7.2 Equations to Calculate Daily-Averaged Air Concentrations 43
8 Modeling Approach for Area Water Sources 48
8.1 Overview and Assumptions 48
8.2 Equations to Calculate Daily-Averaged Air Concentrations 49
9 Scaling Factors for Fugitive and Area Water/Soil Sources 53
9.1 Overview 53
9.2 Scaling Factor for Different Area Sizes 53
9.3 Scaling Factor for Different Emissions 55
9.4 Overall Calculation of Air Concentration 56
9.5 Illustrative Example to Calculate Scaling Factors 56
10 Indoor: Outdoor Ratio 57
10.1 Overview 57
10.2 Literature Search and Screening 58
10.3 Data Analysis 64
10.4 Illustrative Example to Calculate Indoor Air Concentration 67
11 Dose Calculations 68
12 Example Application of IIOAC 71
13 Remaining Uncertainties and Potential Future Updates 72
14 References 72
Appendix A Regression Coefficients for Air Concentration versus Area Size 75
Appendix B Comparison of AERMOD Results for Selected Point Sources 76
Appendix C Illustrative Example for Facility Sources 77
Appendix D Illustrative Example for Area Soil Sources 79
Appendix E Illustrative Example for Area Water Sources 82
iii
-------
LIST OF TABLES
Table 1. Overview of existing air modeling applications 5
Table 2. Example of multiple emission scenarios entered by user for each site 9
Table 3. Chemical and system-specific parameters required for IIOAC 9
Table 4. Location parameters required for IIOAC 10
Table 5. Example IIOAC output. Outputs for stack and incinerators are aggregated into point
source. High-end values are defined as the 95th percentile 12
Table 6. Point source configurations used in the pre-run AERMOD scenarios 20
Table 7. Fugitive-source configurations used in the pre-run AERMOD scenarios 22
Table 8. Area-source configurations used in the pre-run AERMOD scenarios 22
Table 9. Specifications of the meteorology stations used in the AERMOD runs 26
Table 10. Season assignments (defined by vegetation and snow) for the meteorological stations
used in the AERMOD runs 30
Table 11. Number of receptor points modeled by source type and source size 32
Table 12. AERMOD scenarios for point sources 34
Table 13. AERMOD scenarios for fugitive and area (water and soil) sources 37
Table 14. Example of hourly concentrations set to zero when there is no emission for a 1, 4, and
8 hour release duration 39
Table 15. Example lookup table for one AERMOD emission scenario and receptor group 40
Table 16. Summary of mass flux and air concentrations due to emissions from soil at a
hypothetical site with three releases 47
Table 17. Summary of mass volatilized and air concentrations due to emissions from water at a
hypothetical site with three releases 52
Table 18. Example data of air concentration as a function of area size 54
Table 19. Use of indoor-outdoor ratios to calculate indoor air concentration 57
Table 20. Search strategy used to identify potential indoor-outdoor ratio papers 58
Table 21. Acceptance criteria applied in title/abstract screening 59
Table 22. Usability criterion applied in full text screening 62
Table 23. Summary of means and standard deviations of indoor-outdoor ratios by chemical
type 66
Table 24. Final summary of indoor-outdoor ratios 67
Table 25. Mean body weights by age group (taken from Table 8-1 in US EPA, 2011) 68
Table 26. Average inhalation rates for light intensity by age group (taken from Tables 6-1 and 6-
2 in US EPA, 2011) 69
Table 27. Exposure duration for acute and chronic exposures by age group (taken from Table 3-
2 in Versar, Inc. 2007). Exposure duration for adults was modified from 30 to 33 years.69
Table 28. Recommended values for activity patterns by age group (taken from Tablel6-1 in US
EPA, 2011) 70
Table 29. Emission rates corresponding to target maximum air concentrations 72
IV
-------
LIST OF FIGURES
Figure 1. Schematic of the different components of IIOAC 7
Figure 2. Introduction Tab with the import file feature 14
Figure 3. Options for point source 15
Figure 4. Chemical-specific properties required by IIOAC for area water sources 16
Figure 5. Example Source Inputs Tab for point sources when an input file is not imported 17
Figure 6. Information button for area water sources specifying that the user can define a batch
system by entering a flowrate of zero 17
Figure 7. Release duration options for point and fugitive sources 18
Figure 8. IIOAC output metrics of outdoor air concentration, indoor air concentration, and total
particle deposition 19
Figure 9. IIOAC output metrics of acute exposure dose by age groups 19
Figure 10. Map of the meteorology stations used in the AERMOD runs 29
Figure 11. Wind roses for the meteorology stations used in the AERMOD runs 30
Figure 12. Schematic of area soil source, modeled as a batch system 43
Figure 13. Schematic of area water source, modeled as (a) batch and (b) continuous flow-
through system 48
Figure 14. Outdoor air concentration as a function of area size for inner ring receptors 54
Figure 15. Summary of title/abstract and full text screening 59
Figure 16. Ensemble text analytics method for prioritizing studies for screening 61
Figure 17. Diagram illustrating indoor-outdoor ratios with different fractional indoor source
contributions 63
Figure 18. Central tendencies of indoor-outdoor ratios grouped by vapor pressure,
microenvironment, and season 65
Figure 19. Summary of indoor-outdoor ratio means and standard deviations for different
chemical types. Number labels are the number of measurements (number of studies);
error bars are the standard deviations 67
v
-------
1 Introduction
1.1 Overview and Purpose
The EPA Office of Pollution Prevention and Toxics (OPPT) assesses a wide variety of chemical
substances that are released to air from facility (stack, incinerator, and fugitive), area soil, and
area water sources. In addition to existing chemicals, OPPT must also assess air emissions for
new chemical submissions. Site-specific information is often not known when assessing new
chemical submissions. For example, location, size, number of stacks, and/or incinerator
characteristics may all be unknowns when modeling air concentrations and exposures
stemming from facility releases. Therefore, a versatile approach was developed to estimate
outdoor and indoor air concentrations, as well as particle deposition, resulting from air releases
by distance from the source.
Thus, OPPT designed the Integrated Indoor-Outdoor Air Calculator (IIOAC), a user-friendly
Excel-based tool that estimates indoor and outdoor air concentrations, as well as particle
deposition, by distance, from chemical releases to air. IIOAC allows for different meteorological
stations and local land cover, release durations, particle/vapor scenarios, urban/rural settings,
and types of sources. Releases may occur through facility (stack, incinerator, and fugitive), area
soil, and area water sources. Daily-averaged and annual-averaged air concentrations are used
to estimate chemical exposure doses. IIOAC was developed to process multiple scenarios from
multiple sources at once; the tool allows for intermittent releases and variation in
meteorological conditions to account for potential variability in exposure conditions. OPPT
reviewed available air modeling applications and determined that a tool meeting these needs is
not currently available (see Section 2).
IIOAC is able to quickly process new and existing chemicals from multiple sources and multiple
releases for release and exposure potential. The tool uses pre-run results from a suite of
AERMOD dispersion scenarios run in a variety of meteorological and land-use settings.
AERMOD is a modeling system comprised of several modeling routines that work together to
estimate time-average air concentrations and deposition rates around emissions sources.
AERMOD is fully promulgated as a replacement to the Industrial Source Complex (ISC3)
Dispersion Models, in accordance with the Revisions to the Guideline on Air Quality Models (US
EPA, 2017d).
This user's guide describes OPPT's development of IIOAC. The guide is intended to allow a user
to:
1. Learn how to use IIOAC, and
2. Thoroughly understand the modeling approaches and input parameters to aid with
output interpretation.
Sections 2 and 3 provide an overview of existing model applications and a description of IIOAC.
Section 4 uses screen shots and instructions to teach the user how to simulate various
1
-------
combinations of exposure scenarios. Sections 5 through 11 document modeling inputs and
approaches, including how the AERMOD model was configured, how ambient air
concentrations are estimated for different release types, how indoor concentrations are
estimated from ambient concentrations, and how doses are calculated. Section 12 provides an
example application of IIOAC. Finally, Section 13 discusses remaining uncertainties in and
potential updates to the model.
1.2 Acknowledgements and Disclaimer
This work was conducted under work assignment 4-53 of EPA's contract EP-W-12-010 with ICF.
The views expressed in this report do not represent the policies of the U.S. Environmental
Protection Agency. Mention of trade names of commercial products should not be
interpreted as an endorsement by the U.S. Environmental Protection Agency
2 Available Air Modeling Tools
The primary component of IIOAC is the use of AERMOD to simulate the transport of a pollutant
to receptors. As part of the overall design effort, available EPA air modeling tools were
reviewed. A brief summary of each is provided in this section along with a discussion of other
applications of these models for exposure assessment.
2.1 ISC3, AERSCREEN, and SCREEN3
ISC3 is a steady-state Gaussian plume model that evaluates pollutant concentrations from a
range of different industrial sources while accounting for: settling and dry deposition of
particles; downwash; volume, area, point, and line sources; plume rise; point source separation;
and limited terrain adjustment. ISC3 operates in short-term (ISCST3) and long-term (ISCLT3)
modes, and both operate under the same assumptions but require different meteorological
data. ISCST3 tends to be overly conservative in stable conditions, but performs somewhat
better under neutral conditions (US EPA, 1995b).
SCREEN3 and AERSCREEN are simplistic models that quickly estimate worst-case air
concentrations using a limited set of inputs. SCREEN3 is a screening version of ISC3 and is
incorporated into E-FAST 2.0. AERSCREEN is a screening version of AERMOD. They create
random combinations of meteorological parameters that are not site-specific and that are used
in order to identify the combinations of parameters that lead to worst-case air concentrations
in the modeling. SCREEN3 and AERSCREEN do not calculate deposition, and their outputs do not
include a variety of averaging times that the user might be interested in (US EPA 1995a; US EPA
2016a).
2
-------
2.2 AERMOD
OPPT's recommended modeling system for steady-state air quality dispersion is AERMOD, a
Gaussian plume dispersion model based on planetary boundary layer principles. The AERMOD
modeling system is comprised of several modeling routines that work together to estimate
time-average air concentrations and deposition rates around emission sources. AERMOD can
incorporate a variety of emission source characteristics, chemical deposition properties,
complex terrain, and site-specific hourly meteorology to estimate air concentrations and
deposition amounts at user-specified receptors and at a variety of averaging times. AERMOD is
fully promulgated as a replacement to ISC3, in accordance with the Revisions to the Guideline
on Air Quality Models (US EPA, 2017c).
2.3 Air Modeling Applications
The air models described above have been used in various applications to estimate chemical
releases to various media. A brief description of three applications is provided below, with a
comparison of the applications in Table 1. Two additional tools, the Volatilization algorithm in
the Pesticide in Water Calculator (PWC) and European Union System for the Evaluation of
Substances (EUSES), were also included as they estimate air concentration due to chemical
release; however, these two tools do not use the air models described above.
• Risk-Screening Environmental Indicators (RSEI) (US EPA, 2017a): EPA's RSEI tool is a
screening tool that incorporates risk-related perspective to quickly assess the potential
health and environmental impacts from industrial chemical releases. RSEI uses AERMOD
to estimate air concentrations from stack and fugitive releases to air, given source
parameters and meteorological input data. These air concentrations are used to
calculate doses, which can be translated into a risk-related score when population data
and toxicity weights are accounted for. RSEI scores are used for comparison purposes to
rank and prioritize chemicals and industry sectors. Values calculated from RSEI are only
meaningful when compared to other values produced from RSEI.
• Exposure and Fate Assessment Screening Tool (E-FAST) (Versar, Inc., 2007): EPA's E-
FAST is a screening-level tool that estimates industrial and household chemical releases
to air, water, and land, and uses these values to calculate inhalation and ingestion
exposure. E-FAST uses SCREEN3 to model air concentrations from stack and fugitive
sources, given release information and meteorological parameters. The estimated air
concentrations are then used to calculate inhalation acute and chronic exposure doses
for individuals who breathe the air containing the chemical.
• Volatilization Screening Tool (VST) (US EPA, 2014a): EPA's Volatilization Screening Tool
estimates air concentrations downwind of fields treated with semi-volatile pesticides
using pre-run AERSCREEN results. The inputs of the tool are application rates of
chemical onto soil and meteorological and land surface input parameters to estimate
downwind air concentrations. The Volatilization Screening Tool provides a fast estimate
of values previously calculated by the resource-intensive Probabilistic Exposure and Risk
model for FUMigants (PERFUM).
3
-------
• Volatilization algorithm in the Pesticide in Water Calculator (PWC) (US EPA, 2015b):
Within the EPA's PWC, the volatilization algorithm calculates the chemical release of
land-applied pesticide to air and water. None of the air models described above are
used; rather, the algorithm uses chemical parameters such as Henry's Law constant to
calculate daily mass flux into air over a specified time period and is used for bare soil
and pre-emergent applications of fumigant and conventional pesticides. The algorithm
should not be used for foliar applications or for semi-volatile chemicals with Henry's law
constant below 10~7 atm-m3/mol. Inhalation exposure due to vapor-phase
concentrations cannot be evaluated as the daily average flux is not precise enough to
capture spikes over short time scales.
• European Union System for the Evaluation of Substances (EUSES) (European Chemicals
Agency, 2016): Developed by the European Commission, the EUSES tool is used to
conduct environmental exposure assessments to industrial chemicals and biocides.
EUSES follows Registration, Evaluation, Authorisation and Restriction of Chemicals
(REACH) requirements, a European Union regulation. Air concentration is estimated
using the Gaussian plume model Operational Priority Substance (OPS), along with
chemical parameters such as vapor pressure and Henry's Law constant. EUSES uses
multiple interactive forms to facilitate data entry to parameterize the chemical being
released, the process releasing it, environmental parameters helping govern chemical
fate and transport, and exposure parameters leading to estimates of chemical exposure
and risk.
4
-------
Table 1. Overview of existing air modeling applications.
Air Modeling A
pplication or Tool
IIOAC
RSEI
E-FAST
VST
PWC
EUSES
Features
Air Model
AERMOD
AERMOD
SCREEN3
AERSCREEN
n/a
OPSa
Screening tool
y
y
y
y
Inputs
Source types
(Releasing to air)
Point
y
y
y
y
Fugitive
y
y
y
y
Area soil
y
y
y
Area water
y
y
y
Source origins
Consumer products
y
y
Industrial activities
y
y
y
y
Urban conditions considered
y
y
y
Particle phase considered
y
N/A
N/A
y
Intermittent releases possible
y
y
Atmospheric transformation considered
y
y
Terrain considered
y
y
y
Only simple,
uniform
terrain
y
Choice of meteorology conditions
y
y
y
y
Choice of land cover conditions
y
By selecting
different
meteorology
sites
y
y
Building downwash considered
y
y
Exposure to
runoff and
leaching
5
-------
Air Modeling A
pplication or Tool
IIOAC
RSEI
E-FAST
VST
PWC
EUSES
Multiple releases processed at once
y
y
Outputs
Conservative estimates
y
y
y
y
Deposition modeled
y
y
Outputs at different distances from
source
y
y
y
y
Exposure doses calculated
y
y
Indoor air concentrations calculated
y
Averaging periods
used
1-hr
y
y
Daily
y
y
Annual
y
y
y
y
a: OPS = Operational Priority Substances
6
-------
3 General Description of IIOAC
While there are a variety of tools to estimate air concentrations resulting from chemical
releases, the comparison outlined in Table 1 shows the need for a tool that is able to quickly
and simultaneously process multiple emission scenarios for multiple sources, while allowing for
intermittent releases and variation in meteorological conditions. To meet this need, OPPT
designed the IIOAC tool.
3.1 General Description
IIOAC is a tool that assesses the release to air and exposure potential for new and existing
chemicals. A key feature of the tool is the grouping of inputs to define emission scenarios. An
emission scenario is a collection of releases featuring one or multiple source types, each with
different temporal patterns and emission rates. For each emission scenario, the tool provides
output summarizing air concentrations, particle deposition, and exposure doses at different
receptor distances for each source type. A general overview of the Excel-based tool is
i lustrated in Figure 1.
User Inputs
Emission release parameters
Chemical parameters
Source and location parameters
AERMOD
Emissions
results
Excel-based
IIOAC
AERMOD
Point Source
Particle
Deposition by
Distance
Outdoor Air
Concentration
Emissions
Emissions
by Distance
Fugitive
Source
Land/water
Facility sources
Area soil sources
Area water sources
Indoor Air
Concentration
by Distance
Receptor
Figure 1. Schematic of the different components of IIOAC.
IIOAC considers releases from the following emission source types:
• Facility sources (point and fugitive) - point sources are defined as stack and incineration
releases,
• Area soil sources, and
• Area water sources (batch and continuous-flow systems).
7
-------
For each source type, a suite of generic AERMOD (version 16216r) runs have been designed and
pre-run; the resulting air concentrations and particle depositions were post-processed in R and
packaged into lookup tables as part of IIOAC (see Section 4.1 for downloading and operating
IIOAC. Note, the zip file needs to be located in the same folder as IIOAC). Running the Excel-
based IIOAC involves specifying emission scenario inputs so that the applicable AERMOD run is
selected and the associated concentrations and depositions are adjusted to account for the
particular emission rate and if applicable, area size. Three types of user inputs are required to
characterize each emission scenario:
• Emission parameters: source type, emission rate, and number of releases per year,
• Chemical and system parameters: area source size or chemical-specific parameters, and
• Location parameters: facility parameters, climate region, urban/rural, and particle sizes.
Each of these types are explained in greater detail in Section 3.2.
IIOAC allows these inputs to be either imported via an input file or manually entered. IIOAC is
currently designed to allow for up to a maximum of 100 release profiles (i.e., the combination
of number of releases per year and the emission rate) per source type. Based on the user
inputs, the tool will automatically calculate and display, for each emission scenario and at pre-
defined receptor distances, the resulting outdoor and indoor air concentrations (by applying an
indoor-outdoor ratio, see Section 10); particle deposition to surfaces; and acute and chronic
dose at pre-defined life stages. An export feature is available that allows the user inputs and
associated outputs to be saved into a separate Excel workbook.
3.2 User Inputs
3.2.1 Emission Parameters
For each site of interest, users have the option to import an Excel data file or manually input
information on the emission source type, number of emission scenarios, number of releases
per scenario, and for each release, the mass released per day and the number of release days.
Table 2 provides an example of user-defined emission scenarios and release profiles. In the
example in Table 2, three types of emission scenarios occur and are given the following names:
manufacturing, processing, and use. Multiple source types with multiple releases can occur for
each emission scenario. For example, in the Use scenario, there are four different releases from
both fugitive and area land sources. IIOAC can process all source types and emission scenarios
at once and provide a summary of results as described in Section 3.3. Note the default release
duration is 24 hours for all source types. However, the user has the option of selecting release
durations of 1, 4, and 8 hours for point (stack, incinerators) and fugitive sources.
8
-------
Table 2. Example of multiple emission scenarios entered by user for each site.
Emission
Scenario #
Emission
Scenario
Source Type
Release #
Mass Released
per Day (kg/day)
# of Release
Days per Year
1
Manufacturing
Incineration
1
100
5
1
Manufacturing
Fugitive
1
10
100
2
Processing
Incineration
1
50
12
2
Processing
Incineration
2
1
100
2
Processing
Fugitive
1
100
5
2
Processing
Fugitive
2
10
50
2
Processing
Area Water
1
1
250
2
Processing
Area Water
2
1
100
2
Processing
Area Water
3
0.1
250
2
Processing
Area Water
4
0.01
365
3
Use
Fugitive
1
1
5
3
Use
Fugitive
2
0.5
12
3
Use
Fugitive
3
0.25
100
3
Use
Fugitive
4
0.001
365
3
Use
Area Soil
1
100
1
3
Use
Area Soil
2
10
5
3
Use
Area Soil
3
1
12
3
Use
Area Soil
4
0.1
30
3.2.2 Chemical and System Parameters
Depending on the source type selected, the user will also be asked to enter chemical-specific
and/or system-specific parameters. All emission scenarios with a given source type use the
same system parameters. Table 3 outlines the required user inputs:
Table 3. Chemical and system-specific parameters required for IIOAC.
User Input
Symbol
Source Type
Point
Fugitive
Soil
Water -
Batch3
Water -
Continuous
flowb
System-specific parameters
(Surface) Area (m2)
A
~
~
~
~
Depth of water (m)
D
~
~
Flowrate (m3/day)
Q
~
Chemical-specific parameters
Vapor pressure (Torr)
VP
~
~
~
Solubility (mg/L)
Sol
~
Organic carbon sorption
Koc
~
9
-------
coefficient (mL/g)
Volatilization half-life (1/day)
tl/2
s
s
Molecular weight (g/mol)
s
s
s
a: batch water sources are considered to be area water sources with no flow in or out of the system, e.g., lake,
surface impoundment for wastewater, open tanks
b: continuous flow water sources have a constant flowrate in and out of the system, e.g., river, aeration tank in
wastewater treatment process
3.2.3 Location Parameters
For each source type selected, Table 4 lists the location and deposition parameters that must
be provided by the user. Further information on each of the inputs is described in Sections 5.3-
5.5.
Table 4. Location parameters required for IIOAC.
User Input
Source Type
Point
Fugitive
Soil
Water -
Batch3
Water -
Continuous
flow"
Urban or rural setting
~
~
~
~
~
Particle size or vapor
~
~
Climate region
~
~
~
~
~
a: batch water sources are considered to be area water sources with no flow in or out of the system, e.g., lake,
surface impoundment for wastewater, open tanks
b: continuous flow water sources have a constant flowrate in and out of the system, e.g., river, aeration tank in
wastewater treatment process
3.3 IIOAC Outputs
The meteorology data used in IIOAC varies hourly throughout the year and results in a wide
range of air concentrations for a given set of emission inputs. As a result, for each emission
scenario defined by the user, IIOAC will provide output metrics for two groups of receptors:
inner ring or fenceline ring receptors, and near-facility community receptors. A description of
these receptor groups, along with the number of receptors for each AERMOD run, are provided
in Section 5.6.
For each group of receptors, the meteorology data is used to calculate the following
parameters:
• Daily-averaged air concentration (i.e., hourly concentrations averaged over one day),
• Annual-averaged air concentration values (i.e., hourly concentrations averaged over one
year), and
• Annual-averaged total annual particle deposition (wet and dry) (i.e., hourly deposition
averaged over one year).
IIOAC then calculates and reports the central-tendency and high-end values, defined as the
average (mean) and 95th percentile, respectively, of all values of the above three parameters.
10
-------
While total annual particle deposition (total, wet, dry) is not used further in IIOAC, these results
can be used as inputs to models that estimate soil and surface water concentrations.
In addition to air concentrations and particle deposition, the mean and high-end acute and
chronic exposure doses are also calculated using the mean and high-end daily- and annual-
averaged air concentrations (see Section 11). Exposure doses are provided for the following age
groups:
• Young toddler (1-<2 years),
• Adult (16-<78 years), and
• Lifetime (0- <78 years) - calculated for chronic exposure doses only.
Table 5 provides an example IIOAC output for one run. Output metrics are calculated for each
emission scenario. The stack and incinerator sources are aggregated into one source called
point source. For fugitive and area sources, IIOAC outputs are calculated based on the user-
specified area size. However, these outputs can be scaled to a different area size if needed,
using regression coefficients in Appendix A.
11
-------
Table 5. Example IIOAC output. Outputs for stack and incinerators are aggregated into point source. High-end values are defined as
the 95th percentile.
Source
Type
Emission
Scenario
Statistic
Location
Outdoor Air
Concentration
(Hg/m3)
Indoor Air
Concentration
(Hg/m3)
Deposition
(g/m2)
Acute Dose
(mg/kg/day)
Chronic Dose (mg/kg/day)
Daily
Annual
Daily
Annual
Tot
Wet
Dry
Young
Toddler
Adult
Young
Toddler
Adult
Lifetime
Point
Manufacturing
High-End
& Mean
Fenceline
Community
Processing
High-End
& Mean
Fenceline
Community
Use
High-End
& Mean
Fenceline
Community
Fugitive
Manufacturing
High-End
& Mean
Fenceline
Community
Processing
High-End
& Mean
Fenceline
Community
Use
High-End
& Mean
Fenceline
Community
Area
Water
Manufacturing
High-End
& Mean
Fenceline
Community
Processing
High-End
& Mean
Fenceline
Community
Use
High-End
& Mean
Fenceline
Community
Area
Soil
Manufacturing
High-End
& Mean
Fenceline
Community
Processing
High-End
& Mean
Fenceline
Community
Use
High-End
& Mean
Fenceline
Community
12
-------
4 Using IIOAC
4.1 Downloading and Operating IIOAC
To use IIOAC, two files must be downloaded and saved to the same folder. These are an Excel
file containing the main program file (IIOAC_Locked_vl.0.xlsm), and a "zip" file (i.e., a file
format commonly used for compression and transmission of large computer files) of pre-run
AERMOD results (IIOAC_RunFiles.zip). Files from the zip file should not be extracted and
should remain within the zip file. The user will not directly access the zip file; rather, the main
program file will call on files within the zip file.
4.2 Hardware and Software Requirements for IIOAC
The hardware and software requirements to run IIOAC are listed below. Note that higher
specifications will lead to increased performance and decreased runtime.
Hardware (Windows Vista Business Requirements):
• 1-gigahertz (GHz) 32-bit (x86) processor or 1-GHz 64-bit (x64) processor,
• 1 GB of system memory, and
• 128 MB of graphic memory (minimum).
Software:
• Windows Operating System and
• MS Excel 2010 or greater.
4.3 Introduction Tab
In IIOAC, the Introduction Tab provides a general description of IIOAC and directs the user to
choose a source type from the drop-down menu. The available options are: point source (stack
or incineration), fugitive source, area soil source, area water source, and all sources (i.e., more
than one type of source). After clicking Begin, the user has the option to import an input file or
to manually enter scenario and release data (Figure 2).
13
-------
Integrated Indoor-Outdoor Air Calculator
oEPA \l/
2
3
4 ]
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
United Stales
Environmental Protection
Agency
'ICF
The Integrated Indoor-Outdoor Air Calculator (IIOAC) was developed by the
US Environmental Protection Agency's Office of Pollution Prevention and
Toxics and ICF. It is a tool intended to quickly process new and existing
chemicals for release and exposure potential.
Users can import an input file or build multiple emission scenarios, each with
multiple releases, while choosing from different meteorological stations,
emission durations, particle/vapor scenarios, urban/rural settings, and
source type.
IIOAC estimates outdoor and indoor air concentrations, as well as wet and
dry particle deposition, by distance, from chemical releases to air through
facility (stack, incinerator, and fugitive), area soil, and area water sources.
Daily-averaged and annual-averaged air concentrations are used to estimate
exposure doses.
Information on the equations and assumptions used to develop IIOAC can
be found in the User's Guide.
Choose Source Type
Point Source
Microsoft Excel
Would you like to import an inputs file?
Ves
No
Figure 2, Introduction Tab with the import file feature.
If the user chooses to import an input file, the input file must have column headers and the
columns must be in the following order starting in column A;
• Scenario number: must be a whole number,
• Emission scenario: name or description of emission scenario,
• Source type: must be stack, fugitive, incineration, area soil, or area water (e.g.,
incinerator will not be recognized during the file import process),
• Release number: must be a whole number,
• Mass released per day, in kg/day: must be a number greater than zero, and
• Number of release days per year: must be a whole number between 1 and 365,
inclusive.
For the source type of point source, if the user selects the import inputs file option, an
additional window appears (Figure 3), asking the user to specify if the point source is a stack, or
one of two possible incinerator options (see Section 5.2.1 for parameters corresponding to
these three point sources). Once selected, data from the inputs file is auto-populated into the
Source Inputs Tab (see Section 4.5) and the user is automatically directed to the Chemical Tab.
Note that when importing a file, the source type 'incineration' must be used in the inputs file
instead of'incinerator'.
14
-------
2
3_
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
The Integrated Indoor-Outdoor Air Calculator (IIOAC) was developed by the
US Environmental Protection Agency's Office of Pollution Prevention and
Toxics and ICF. It is a tool intended to quickly process new and existing
chemicals for release and exposure potential.
Choose Source Type
Point Source
Users can import an input file or build multiple emission scenarios, each with
multiple releases, while choosing from different meteorological stations,
emission durations, particle/vapor scenarios, urban/rural settings, and
source type.
Select A Source Type
IIOAC estimates outdoor and indoor air concentrations, as well as wet and
dry particle deposition, by distance, from chemical releases to air through
facility (stack, incinerator, and fugitive), area soil, and area water sources.
Daily-averaged and annual-averaged air concentrations are used to estimate
exposure doses.
Information on the equations and assumptions used to develop IIOAC can
be found in the User's Guide.
Stack
Tnrinpratnr 1
Begin
Please choose the Point Source type you
would like to run.
Integrated Indoor-Outdoor Air Calculator
oEPA
United States ^
Environmental Protection #1 E
Agency ¦ W ¦
Figure 3. Options for point source.
4.4
Chemical Tab
Incinerator 2
Depending on the source type selected, the user will be required to enter chemical-specific
properties, In Figure 4, the boxes greyed out are not applicable to the source type and do not
need to be filled in. For example, the source type selected in Figure 4 is for area water sources
and therefore information on solubility and the organic carbon sorption coefficient are not
needed. For the required information in blue boxes, IIOAC has built-in error messages if the
user enters a value that is not valid (e.g., negative number for volatilization half-life). Question
marks next to a chemical property provides additional information for the user. For example,
the question mark next to vapor pressure provides the unit conversion from Torr to Pascal (Pa)
or standard atmospheres (atm).
For volatilization half-life, the user can click on the question mark which leads to a pop-up
window that provides a link to EPA's EPI Suite (US EPA, 2017b), a parameter estimation
program. EPI Suite is a screening-level tool and should not be used if acceptable measured
values are available. EPI Suite provides the following default values to estimate volatilization
half-life:
• Water depth = 1 m (for both river and lake),
• Wind velocity = 5 m/s (river); 0.5 m/s (lake), and
• Current velocity = 1 m/s (river); 0.05 m/s (lake).
Volatilization half-life values are used in flux calculations for area water sources and should
differ between batch and continuous-flow sources by entering a flowrate value of zero for
batch sources.
15
-------
A
B
C
D
E
F
G
H
Chemical Information and Properties
3 Step 1: Enter Chemical Information
4
Step 2: Enter Chemical Properties
5
Chemical Information
Chemical Properties
8
9
10
6 Chemical Name
7 CAS Number
Vapor Pressure (Torr) |?|
Solubility (mg/L)
Org. Carbon Sorption Coeff (KoC) (mL/g) |T]
Volatilization Half-Life (hrs) [?]
Molecular Weight (g/mol)
11
12
13
14
15
16
17
18
19
20
21
Figure 4. Chemical-specific properties required by IIOAC for area water sources.
4.5 Source Inputs Tab
IIOAC has a Source Inputs Tab for each source type, which varies slightly depending on the
source type. In general, the Source Inputs Tab consists of three steps: (1) entering source
parameters; (2) selecting location and deposition settings; and (3) defining emission scenarios.
See Table 12 and Table 13 in Section 5 for full details on source parameters and location and
deposition settings.
In step 1, the user is required to enter source parameters. Depending on the source type, these
source parameters include area of source, and for area water sources, the surface area, depth
of water, and flowrate. For point sources, if the user did not import an inputs file, the user must
select the point source type (stack, incinerator 1, incinerator 2) from the drop-down menu,
which will auto-populate the source parameters (see Figure 5).
For area water sources, the user can differentiate between batch and continuous-flow sources
by specifying a flowrate value of zero for batch sources and a non-zero flowrate value for
continuous-flow sources. The question mark next to the flowrate in IIOAC reminds the user of
this differentiation (see Figure 6).
16
-------
Point Source
Step 1: Select Point Source Type
Source Parameters
Release Height (m)
Stack Inside Diameter (m)
Exit Gas Temperature (K)
Exit Gas Velocity (m/s)
Stack
Incinerator 1
Incinerator 2
ep 2: Select Location and Deposition Settings
Location and Deposition Settings
ITT
300
9
10
11
12
13
14
15 j
16 Step 3: Define Emission Scenarios
17
Select Urban or Rural
Population
Select Particle Size
Mean Aerodynamic Diameter (\im)
Density (g/cm3)
Select Climate Region
Surface Station
Upper-air Station
Select Cyclical or Consecutive
18 Scenario #
1 9
20
21
22 I
23 I
119
120
121
Scenario Name
Release #
Release Amount
(kg/site/day)
Release Duration
(hours/day)
Release Frequency
(days/year)
Add Another Scenario
Next Page
Figure 5. Example Source Inputs Tab for point sources when an input file is not imported.
10
11
12 j
13
14
15
16
17 i
18
19
20
21
22
23
24
25
26
Area Water Source
Step 1: Enter Source Parameters
Source Parameters
Step 2: Select Location Settings
Location Settings
Surface Area (m )
Depth of Water (m)
Flowrate (m3/day) [?]
Select Urban or Rural
Population
Select Climate Region
Rural
N/A for Rural
Step 3: Define Emission Scenarios
Scenario # Scenario Name
Microsoft Excel
For batch systems, please enter a flowrate of 0.
Release #
(kg/site/day)
Release Duration
Release Frequency
24 hr/day (continuous)
24 hr/day (continuous)
24 hr/day (continuous)
24 hr/day (continuous)
24 hr/day (continuous)
Add Another Scenario
Next Page
Figure 6, Information button for area water sources specifying that the user can define a batch
system by entering a flowrate of zero,
17
-------
In step 2, the user defines the location and deposition settings through a series of drop-down
menus for population scenario (urban or rural), particle size (fine, coarse, or no particles), and
climate region (14 possibilities). The choices offered in IIOAC for location and deposition
settings are explained in further detail in Section 5. Blue boxes define what the user must select
and the grey boxes are auto-populated based on the user's selection. Depending on what the
user selects in this step, the tool will access the corresponding pre-run AERMOD results of air
concentration and particle deposition from the zip file and import these results into the tool.
In step 3, defining emission scenarios, if the user imported an input file, the table in step 3 will
already be auto-populated with a default release duration of 24 hours/day. The user can also
manually change the release duration for point and fugitive sources (see Figure 7). If an input
file was not imported, the user must manually enter the information. Additional scenarios can
be added using the Add Another Scenario button, up to a total of 100 scenarios for each source
type. For point and fugitive sources, step 3 has an additional feature of asking the user to select
whether the releases in a scenario are cyclical (i.e., evenly spaced out over 365 days) or
consecutive releases (i.e., consecutive days of release).
Note the default release duration is 24 hours for all source types. However, the user has the
option of selecting release durations of 1, 4, and 8 hours for point (stack, incinerators) and
fugitive sources.
A
B
C
D
E
F
G
1
Fugitive Source
Reset Tool
3
4
Step 1: Enter Source Parameters
I Step 2: Select Location and Deposition Settings
5
Source Parameters
Location and Deposition Settings \
6
Area of Source (m2)
200
Select Urban or Rural
Urban
7
Release Height (m)
3.05
Population
1,000,000
8
I
9
Select Particle Size
No particles (vapor only)
10
Mean Aerodynamic Diameter (am)
N/A for Vapor
11
Density (g/cm3)
N/A for Vapor
12
I
13
Select Climate Region
Northeast (Coastal)
14
Surface Station
Camp Springs, MD
15
Upper-air Station
Sterling, VA
16
Step 3: Define Emission Scenarios
I
17
Select Cyclical or Consecutive
I Cyclical I
18
Scenario #
Scenario Name
Release #
Release Amount
(kg/site/day)
Release Duration
(hours/day)
Release Frequency
(days/year)
19
1
Processing
1
10
I 4 hr/day (12-4 pm)
F
12
20
2
Use
1
100
1 hr/day
6
21
Use
4 hr/day 112-4 pml
52
z
2
8 hr/day (8-4 pm)
24 hr/day fcontinuous]
22
2
Use
3
25 L
365
23
3
Disposal
1
12
8 hr/day (8-4 pm)
30
24
3
Disposal
2
5
24 hr/day (continuous)
180
25
3
Disposal
3
200
4 hr/day (12-4 pm)
95
26
3
Disposal
4
62
1 hr/day (12-1 pm)
62
27
3
Disposal
5
125
8 hr/day (8-4 pm)
250
120
•lOi I
Add Another Scenario
Next Page
Figure 7. Release duration options for point and fugitive sources.
18
-------
For users that do riot have specific source and/or location and deposition settings in mind, the
default settings recommended to provide the conservative estimates for a given scenario (i.e.,
generally the largest values of air concentration or deposition amounts) are point (specifically
stack) sources, urban settings, coarse particles, and climate region corresponding to Idaho Falls
(East North Central). Further detail of default settings are provided in Section 5.7.
4.6 Output Tab
The Output Tab provides the outdoor, indoor, total annual particle deposition, and acute and
chronic exposure doses (see Figure 8 and Figure 9) for each of the emission scenarios provided
by the user. High-end and mean results are provided by receptor group (inner ring or fenceline
receptors and near-facility community receptors). For area soil and area water sources, the
particle deposition columns will be empty as these sources do not emit fine or coarse particles.
A
B
c
D
E
F
G
H
I
J
K
1
Export ^
Reset
2
Source Type
Emission Scenario
Statistic
Location
Outdoor Air
Concentration (ua/m3)
Indoor Air
Concentration (ua/m3)
Total Annual Particle Deposition
(q/m2)
3 ,
Daily
Annual
Daily
Annual
Total
Wet
Dry
4
High-End
Fenceline Avg
3.50E+01
2.76E+00
3.50E+01
2.76E+00
3.59E-06
3.59E-06
2.93E-09
6
Manufacturing
Community Avg
5.98E+00
1.97E-01
5.98E+00
1.97E-01
2.37E-07
5.38E-10
2.37E-07
7
Mean
Fenceline Avg
3.50E+01
2.27E+00
2.28E+01
1.47E+00
2.95E-06
2.95E-06
1.18E-09
9
Community Avg
4.95E+00
1.63E-01
3.22E+00
1.06E-01
1.98E-07
2.17E-10
1.97E-07
10
High-End
Fenceline Avg
8.92E+01
4.33E+01
8.92E+01
4.33E+01
5.57E-05
5.57E-05
2.75E-08
12
Fugitive Source
Use
Community Avg
1.13E+01
3.11E+00
1.13E+01
3.11E+00
3.75E-06
5.11E-09
3.75E-06
13
Mean
Fenceline Avg
8.72E+01
3.84E+01
5.67E+01
2.49E+01
4.99E-05
4.99E-05
1.99E-08
15
Community Avg
9.90E+00
2.76E+00
6.44E+00
1.79E+00
3.35E-06
3.67E-09
3.34E-06
16
High-End
Fenceline Avg
1.18E+02
8.10E+01
1.18E+02
8.10E+01
1.54E-04
1.54E-04
7.43E-08
18
Processing
Community Avg
2.33E+01
8.79E+00
2.33E+01
8.79E+00
1.05E-05
1.38E-08
1.05E-05
19
Mean
Fenceline Avg
1.17E+02
7.96E+01
7.59E+01
5.18E+01
1.36E-04
1.36E-04
5.41 E-08
21
Community Avg
2.00E+01
7.52E+00
1.30E+01
4.89E+00
9.12E-06
9.98E-09
9.11E-06
22 '
Max
1.18E+02
8.10E+01
1.18E+02
8.10E+01
1.54E-04
1.54E-04
1.05E-05
Figure 8. IIOAC output metrics of outdoor air concentration, indoor air concentration, and total
particle deposition.
A BCD MRTY Z
1
Export |
Reset
Acute Dose
(mg/kg/day)
Chronic Dose (mg/kg/day)
Young
Young
2
Source Type
Emission Scenario
Statistic
Location
Toddler
Adult
Toddler
Adult
Lifetime
3
1 - <2 years
16 - <78 years
1 - <2 years
16 - <78 years
4
High-End
Fenceline Avg
5.31 E-02
7.83E-03
1.93E-03
5.31 E-04
2.43E-04
6
Manufacturing
Community Avg
9.07E-03
1.34E-03
1.38E-04
3.79E-05
1.74E-05
7
Mean
Fenceline Avg
3.50E-02
5.61 E-03
1.05E-03
3.13E-04
1.42E-04
9
Community Avg
4.95E-03
7.93E-04
7.53E-05
2.25E-05
1.02E-05
10
High-End
Fenceline Avg
1.35E-01
1.99E-02
3.04E-02
8.35E-03
3.82E-03
12
Fugitive Source
Use
Community Avg
1.71 E-02
2.53E-03
2.19E-03
6.00E-04
2.75E-04
13
Mean
Fenceline Avg
8.71 E-02
1.40E-02
1.78E-02
5.30E-03
2.40E-03
15
Community Avg
9.89E-03
1.59E-03
1.28E-03
3.80E-04
1.72E-04
16
High-End
Fenceline Avg
1.80E-01
2.65E-02
5.69E-02
1.56E-02
7.15E-03
18
Processing
Community Avg
3.52E-02
5.20E-03
6.16E-03
1.69E-03
7.75E-04
19
Mean
Fenceline Avg
1.17E-01
1.87E-02
3.68E-02
1.10E-02
4.98E-03
21 I
Community Avg
2.00E-02
3.21 E-03
3.48E-03
1.04E-03
4.70E-04
22
Max
1.80E-01
2.65E-02
5.69E-02
1.56E-02
7.15E-03
Figure 9. IIOAC output metrics of acute exposure dose by age groups.
19
-------
4.7 Export and Reset Features
IIOAC offers the user the ability to export the Inputs file, Source Inputs Tab and Output Tab
into a separate Excel workbook. Throughout the use of the tool, the user can also click on Reset
to clear all data entered into the tool and restart the tool.
5 Selection of AERMOD Inputs
5.1 Overview
IIOAC uses pre-run AERMOD results to calculate the output metrics listed in Section 4.6. Within
IIOAC, the pre-run AERMOD results are scaled according to the user-specified release amount
and, for fugitive and area water/soil sources, area size (see Sections 5.9 and 9).
Point sources release emissions from a one-dimensional point location, while fugitive and area
sources have emission releases spread out across a two-dimensional area. A classic example of
a point source is a chimney or other pipe/stack releasing emissions from a combustion activity.
Fugitive sources may include loading docks, bag houses, and areas where a facility building is
leaking or venting. Area sources in IIOAC are open-air soil or water sources where a chemical is
first applied to the soil or water, and then the chemical volatilizes off the surface and into the
air. Examples of area water sources are surface impoundments, lakes, and clarifiers in
wastewater treatment processes, while area soil sources may include waste applied to fields.
This section describes the selected AERMOD scenarios that were pre-run in AERMOD version
16216r and the rationale for selection.
5.2 Source Characterization
5.2.1 Point Sources
Three point source scenarios, shown in Table 6, were developed to cover a range of release-
point parameters that may exist at U.S. facilities.
Table 6. Point source configurations used in the pre-run AERMOD scenarios.
Exit Gas
Release
Exit Gas
Inside
Velocity
Point Source Configuration
Height (m)
Temperature (K)
Diameter (m)
(m/s)
Stack
10
300
2
5
Average Incinerator
25
500
1
15
High-temperature Incinerator
50
1,200
2
15
20
-------
Stack Heights. The three release heights used (10, 25, and 50 m above ground) were
considered representative of most U.S. point sources. According to the 2011 version of the U.S.
EPA National-scale Air Toxics Assessment (NATA; this was the latest version available at the
time of IIOAC development), about 80 percent of U.S. point sources have release heights
between 5 and 75 m, with the majority being 5 to 15 m tall (EPA 2015a).
As a supplementary measure of representativeness, about 88 percent of individual types of
point sources have default release heights within the range of 5-75 m, again with the majority
being 5-15 m. These default point-source release parameters came from version 4.0 of the U.S.
EPA SMOKE emissions model (UNC 2016; specifically, the PSTK file of point-source replacement
stack parameters), which was used in the 2011 NATA for populating missing point-source
parameters at individual facilities, based on Source Classification Code (SCC) (EPA 2015c). Note
that these statistics using default SCC parameters weight each SCC equally, not accounting for
the number of U.S. point sources that correspond to a given SCC.
Exit Gas Temperatures. The three exit gas temperatures used (300, 500, and 1,200 K) were
considered broadly representative of most U.S. point sources. According to the 2011 NATA (EPA
2015a), about 78 percent of U.S. point sources have exit gas temperatures between 230 and
630 K (with the majority being below about 400 K), and the modeled 1,200 K value covers most
of the higher temperatures as well. In addition, about 96 percent of point-source SCCs have
default exit gas temperatures in the range of 230-630 K (UNC 2016).
Inside Diameter. The two inside diameters used (1 and 2 m) were considered representative of
the majority of U.S. point sources. About 52 percent of U.S. point sources in the 2011 NATA had
inside diameters between 0.5 and 2.5 m (especially near 1 m) (EPA 2015c). As a supplementary
statistic, about 65 percent of point-source SCCs have default inside diameters in this range
(UNC 2016).
Exit Gas Velocities. The two exit gas velocities (5 and 15 m/s) were considered representative
of an approximate majority of U.S. point sources. About 48 percent of U.S. point sources in the
2011 NATA had exit gas velocities between 2.5 and 17.5 m/s, somewhat preferring smaller
values (EPA 2015c). In addition, about 64 percent of point-source SCCs had default exit gas
velocities in this range (UNC 2016).
Combined Parameters. The three combinations of these values (creating the stack, incinerator
1, and incinerator 2 point sources) were created so that together they would result in a wide
range of air concentrations. As shown in Table B1 of Appendix B, the stack point source should
generally result in the highest air-concentration and deposition values relative to the other
point sources, due to its lower height and lower plume rise (due to lower buoyancy and
momentum). The two incinerator point sources should generally result in much smaller air-
concentration and deposition values relative to the stack source, due to their higher heights
and plume rise values—the incinerator 2 in particular should often show reductions in air-
concentration and deposition values of more than 90 percent relative to the stack source, at
receptor locations both close to (100 m) and farther away (1,000 m) from the emission source.
21
-------
5.2.2 Fugitive Sources
For fugitive sources, five scenarios for horizontal dimensions (shown in Table 7) were modeled
in AERMOD to cover a wide range of values that OPPT may consider. Each of these are not
available as source-configuration choices in IIOAC—rather, as discussed further in Section 5.9,
they were used in the pre-run AERMOD scenarios to derive regressions associating changes in
horizontal dimension with changes in air concentrations and deposition. These regressions use
the 100 m2 fugitive source as the baseline source upon which those regression associations are
applied in IIOAC for a user-defined size of fugitive source. Iowa City, IA was the baseline
meteorology station, as discussed in Section 5.4. The release height (3.05 m) is the default value
used for fugitive sources in the 2011 NATA (EPA 2015c).
Table 7. Fugitive-source configurations used in the pre-run AERMOD scenarios.
Area (m2; equal length and height)
Release Height (m)
25
3.05
50
100 (default)
200
500
5.2.3 Area Sources
Similar to fugitive sources (see Section 5.2.2), five scenarios for horizontal dimensions (shown in
Table 8) were modeled in AERMOD to cover a wide range of values that OPPT may consider.
Each of these are not available as source-configuration choices in IIOAC—rather, as discussed
further in Section 5.9, they were used in the pre-run AERMOD scenarios to derive regressions
associating changes in horizontal dimension with changes in air concentrations. These
regressions use the 50 acre area source as the baseline source upon which those regression
associations are applied in IIOAC for a user-defined size of area source. Iowa City, IA was the
baseline meteorology station, as discussed in Section 5.4. Because these area sources are used
in the IIOAC as releases from soil or water surfaces, the release heights are at ground level.
Table 8. Area-source configurations used in the pre-run AERMOD scenarios.
Area (acres; equal length and height)
Release Height (m)
10
0
20
50 (default)
200
500
22
-------
5.3 Emission Characterization
For the pre-run AERMOD scenarios, each source emitted at a constant rate of 1 g/s. For sources
modeled as areas (the fugitive and area sources described respectively in Sections 5.2.2 and
5.2.3), the 1 g emitted mass was spread evenly across the area. For example, a 25 m2 fugitive
area emitted 0.04 g/s/m2.
These emission rates were specific to the phase of the emitted pollutant. Vapor phase
pollutants will disperse more widely than particle phase pollutants. Smaller diameter particles
will disperse more widely than larger diameter particles. Separate AERMOD runs, with separate
outputs, allowed for the modeling of 1 g/s of particle phase pollutant with larger diameters, 1
g/s of particle phase pollutant with smaller diameters, and 1 g/s of vapor phase pollutant. All
source types (point, fugitive, and area) were modeled to emit vapor phase pollutants (with
deposition assumed to be negligible in the near-field), and point and fugitive sources
additionally emitted particle phase pollutants (with deposition processes modeled).
Smaller diameter particles were modeled with mass-mean aerodynamic diameters of 2.5 |am
(the upper limit of the typical definition of "fine" particles) and an assumed density of 1 g/cm3.
The same density was used for larger particles, which had mass-mean aerodynamic diameters
of 10 |am (the upper limit of the typical definition of "coarse" particles). Larger particles will
deposit closer to the emission source relative to smaller particles.
5.4 Meteorology and Land Cover
With the goal of providing stations that broadly represent meteorology conditions encountered
throughout the U.S., 14 meteorological stations were used for surface meteorology in the pre-
run AERMOD scenarios and were selected based on previous analyses for OPPT (US EPA,
2014b). These surface stations, and the upper-air stations they were paired with, are listed in
Table 9 and shown in a map in Figure 10. Table 9 also contains information on the elevation
above sea level of each surface station as well as a qualitative description of the land cover
within 1 km and 10 km of the surface station (according to year-1992 land-cover data, which
are the vintage of data used by EPA's land-cover processor [AERSURFACE] for AERMOD; MRLC
2001).
In that previous analysis for OPPT, one station was initially selected for each of nine U.S. climate
regions. The representativeness of a station relative to its region was determined using the
ventilation factor, which is the product of wind speed and mixing height. The ventilation factor
represents a measurement of the dispersion flux through an idealized box around a source of
pollution, where larger ventilation factors represent larger mixing volumes and lower average
concentrations. The representative station was selected whose distribution of hourly
ventilation factors was most similar to that of the aggregate of all stations in the region.
For regions with coastlines, if the selected station (using the ventilation-factor method) was on
the coast, then an inland station was added for the region; if the selected station was inland,
23
-------
then a coastal station was added. Relative to the stations initially selected above using the
ventilation factor, these supplementary stations experienced significantly different wind
patterns and were in different states in the region.
Meteorology data were used from the five most recent years available at the beginning of the
project development (years 2011-2015), including one-minute processing of wind data at all
but one station (the Camp Springs, MD station did not have one-minute wind data available).
Corresponding period-average wind roses are shown in Figure 11. Meteorological processors
and pre-processors included: AERMINUTE vl5272, AERSURFACE vl3016, and AERMET vl6216.
Hourly surface data, one-minute surface wind data, and twice-daily upper-air data were
obtained from NOAA (2017b, 2017a, and 2017c, respectively), and land-cover data were
obtained from MRLC (2001). Upper-air stations were selected based on proximity to the surface
station (primary factor) and on having similar geography and meteorological characteristics
relative to the surface station (secondary factor).
For the purposes of deriving micrometeorological parameters using AERSURFACE, several
recent years of remotely-sensed vegetation data (specifically the Normalized Difference
Vegetation Index as a measure of greenness; NASA 2017) were analyzed in the vicinity of each
station. Qualitative assumptions were made about monthly vegetative-season assignments
based on the greenness index (Table 10). Historical snow-cover data (NOAA 2012) were further
used to identify whether the site typically has more than 30 days of snow on the ground, and in
such cases the months identified as "after harvest or first frost" were identified as also having
continuous snow cover. AERSURFACE was run assuming average surface-moisture conditions
relative to climatology, with surface-roughness calculations conducted within a 1-km radius and
within 12 30-degree sectors around the surface meteorological station.
All 14 meteorology datasets were modeled for all point-source AERMOD runs. For fugitive and
area sources, where pre-run AERMOD scenarios were used to derive regressions associating
changes in horizontal dimension with changes in air concentrations and deposition, the Iowa
City, IA location was used as the baseline meteorology scenario to calculate those regressions.
It was assumed that those regression relationships (relating source size and AERMOD outputs)
derived using Iowa City, IA meteorology data would be roughly applicable to other
meteorological conditions from the other 13 meteorology stations; that is, the user may select
any of the 14 meteorological stations in IIOAC, but for fugitive and area sources the regression
applied to the user-entered source size comes from modeling using the Iowa City station.
Iowa City was selected for the fugitive- and area-source regressions because it is reasonably
representative of U.S. meteorological stations as a whole. The Iowa City station is located near
the center of the country and does not experience substantial terrain or water-body/coastal
influences. In order to gauge the representativeness of the Iowa City's wind speeds, mixing
heights, and precipitation amounts relative to overall U.S. conditions, a small comparison
exercise was conducted using year-2016 meteorological data from over 800 stations across the
U.S. Note that these are not the same years of data used in the modeling to support IIOAC—
EPA OAQPS had already run these year-2016 data through AERMOD's meteorology processor,
24
-------
and the pre-run meteorology data are available from EPA OAQPS's Human Exposure Model
website (EPA 2017c). Using those year-2016 data, a comparison was made between the typical
conditions at the non-coastal stations used in this tool and the typical conditions from all 800+
stations (which included coastal stations). According to that comparison, from among the non-
coastal stations used in IIOAC, the Iowa City station ranked 3rd, 5th, and 3rd most representative
of all U.S. stations for wind speed, mixing height, and precipitation, respectively; all other
stations had at least one ranking of 6th or larger. Therefore, the Iowa City station was judged to
be a reasonably representative station.
25
-------
Table 9. Specifications of the meteorology stations used in the AERMOD runs.
Surface Station
Upper-air Station
Coastal or Inland
WBAN
(with approx.
Lat.
Climate
(Call
distances to water
Long.
Qualitative Land-cover
Lat.
Region
Sign)
Location
for coastal)
Elev.
Description (1992)a
WBAN
Location
Long.
1
14937
Iowa City,
Inland
41.633
1 km: Mostly developed,
94982
Davenport,
41.6
(East
(KIOW)
IA
-91.543
some farmland
IA
-90.57
North
10km: Half developed, half
Central)
198 m
farmland, some forest
2
13705
Camp
Coastal (31 km
38.811
1 km: Developed
93734
Sterling, VA
38.98
(North-
(KADW)
Springs,
from Chesapeake
-76.867
10 km: Mostly developed,
-77.47
east)
MD
Bay, 60 km from
Atl. Ocean)
86 m
some forest and farmland
14762
Pittsburgh,
Inland
40.355
1 km: Mostly developed,
94823
Township,
40.53
(KAGC)
PA
-79.922
380 m
some forest
10 km: Mostly developed,
some forest
PA
-80.23
3
24222
Everett,
Coastal (4 km from
47.908
1 km: Developed
24232
Salem, OR
44.92
(North-
(KPAE)
WA
Puget Sound, 50
-122.28
10 km: Developed, some
-123.02
west)
km from Salish
Sea, 180 km from
Pac. Ocean)
184 m
open water
24145
Idaho Falls,
Inland
43.516
1 km: Developed
24061
Riverton,
43.06
(KIDA)
ID
-112.06
1,441 m
10 km: Mostly farmland,
some development
WY
-108.47
4
13920
Topeka, KS
Inland
38.95
1 km: Developed
13996
Topeka, KS
39.07
(South)
(KFOE)
-95.664
325 m
10 km: Mostly farmland,
some development
-95.62
26
-------
Climate
Region
Surface Station
Upper-air Station
WBAN
(Call
Sign)
Location
Coastal or Inland
(with approx.
distances to water
for coastal)
Lat.
Long.
Elev.
Qualitative Land-cover
Description (1992)a
WBAN
Location
Lat.
Long.
03937
(KLCH)
Lake
Charles, LA
Coastal (41 km
from Gulf of
Mexico)
30.125
-93.228
3 m
1 km: Half developed, half
farmland
10 km: Mostly farmland,
some development and
open water
03937
Lake
Charles, LA
30.12
-93.22
5
(South-
east)
93727
(KNCA)
New River,
NC
Coastal (20 km
from Atl. Ocean)
34.7
-77.433
8 m
1 km: Mostly developed,
some wetlands
10 km: Mix of wetlands,
forest, farmland, open
water, and development
93768
Morehead
City, NC
34.7
-76.8
13874
(KATL)
Atlanta, GA
Inland
33.64
-84.427
308 m
Mostly developed
53819
Peachtree
City, GA
33.35
-84.56
6
(South-
west)
23066
(KGJT)
Grand
Junction,
CO
Inland
39.134
-108.538
1,481 m
1 km: Mix ofshrubland
(rough terrain) and
development
10 km: Mostly shrubland
(rough terrain) with some
development and farmland
23062
Denver, CO
39.77
-104.88
7
(West)
93111
(KNTD)
Point
Mugu, CA
Coastal (2 km from
Pac. Ocean)
34.117
-119.110
4 m
1 km: Developed
10 km: Mix of shrubland,
farmland, and open water,
with some development
93214
Vandenberg
AFB, CA
34.75
-120.57
27
-------
Climate
Region
Surface Station
Upper-air Station
WBAN
(Call
Sign)
Location
Coastal or Inland
(with approx.
distances to water
for coastal)
Lat.
Long.
Elev.
Qualitative Land-cover
Description (1992)a
WBAN
Location
Lat.
Long.
23169
(KLAS)
Las Vegas,
NV
Inland
36.079
-115.155
665 m
1 km: Developed
10 km: Mostly developed,
some shrubland (desert)
53103
Flagstaff, AZ
35.23
-111.82
8
(West
North
Central)
14944
(KFSD)
Sioux Falls,
SD
Inland
43.577
-96.754
435 m
1 km: Mostly developed
10 km: Mix of development
and farmland
94980
Omaha, NE
41.32
-96.37
9
(Central)
94822
(KRFD)
Rockford,
IL
Inland
42.196
-89.093
223 m
1 km: Developed
10 km: Mostly farmland,
with some development
and forest
94982
Davenport,
IA
41.6
-90.57
a MRLC (2001)
28
-------
NOAA Climate Regions Representative Surface Stations Upper-air Stations Paired with Surface Stations
| 1 (East North Central)
2 (Northeast)
J 3 (Northwest)
| | 4 (South)
] 5 (Southeast)
6 (Southwest)
7
8 (West North Central)
3 9 (Central)
~
1 (East North Central)
£
1 (East North Central)
~
2 (Northeast)
2 (Northeast)
+
3 (Northwest)
~
3 (Northwest)
4 (South)
~
*
5 (Southeast)
4 (South)
*
6 (Southwest)
~
5 (Southeast)
+
7 (West)
~
8 (West North Central)
6 (Southwest)
£
9 (Central)
it
7 (West)
~
8 (West North Central)
~
9 (Central)
Figure 10. Map of the meteorology stations used in the AERMOD runs.
29
-------
9 Rockford, IL
Figure 11. Wind roses for the meteorology stations used in the AERMOD runs.
Table 10. Season assignments (defined by vegetation and snow) for the meteorological stations
used in the AERMOD runs.
Climate Region
Location of Surface
Station
Month and Season Assi
gnment3
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
1
(East North Central)
Iowa City, IA
Is
Is
2
2
2
3
3
3
4
4
4
Is
2
(Northeast)
Camp Springs, MD
1
1
2
2
2
3
3
3
4
4
4
1
Pittsburgh, PA
Is
Is
Is
2
2
3
3
3
4
4
4
Is
3
(Northwest)
Everett, WA
1
1
1
2
2
3
3
3
3
4
1
1
Idaho Falls, ID
Is
Is
Is
2
2
3
3
4
4
4
Is
Is
4
(South)
Topeka, KS
1
1
1
2
2
3
3
3
4
4
4
1
Lake Charles, LA
1
1
2
2
2
3
3
3
3
4
4
1
5
(Southeast)
New River, NC
1
1
1
2
2
3
3
3
3
4
4
1
Atlanta, GA
1
1
1
2
3
3
3
3
3
4
4
1
6
(Southwest)
Grand Junction, CO
1
1
2
2
2
3
3
3
4
4
4
1
7
(West)
Point Mugu, CA
3
3
3
3
3
3
3
3
3
3
3
3
Las Vegas, NV
3
3
3
3
3
3
3
3
3
3
3
3
30
-------
Climate Region
Location of Surface
Station
Month and Season Assi
gnment3
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
8
(West North Central)
Sioux Falls, SD
Is
Is
Is
2
2
2
3
3
4
4
4
Is
9
(Central)
Rockford, IL
Is
Is
2
2
2
3
3
3
4
4
4
Is
a 1 = after harvest or first ¦
rost, Is = 1 with continuous snow cover, 2 = partia
green coverage or s
iort
annuals, 3 = lush vegetation, 4 = autumn prior to harvest or first frost
5.5 Urban/Rural
Each point and fugitive source in the pre-run AERMOD scenarios was modeled once with the
urban-effects setting turned on in AERMOD and once without it turned on. The urban setting in
AERMOD modifies turbulence parameters to better account for the effect of the urban heat
island on the nocturnal boundary layer and subsequent transition to the daytime boundary
layer. AERMOD uses a population-count value to tailor its urban-heat-island calculations, and a
value of 1 million people was used in the urban runs.
It was assumed that soil and water area sources would not be in urban areas. As such, area
sources were not modeled with the urban setting.
5.6 Receptors
Output metrics are determined for two groups of receptors (fenceline and community) as
indicated below:
• 16 inner ring or fenceline receptor-points: a polar-grid ring of receptor points 100 m
from the point source (or approximately 100 m from the corners of the fugitive- or area-
source square), spaced every 22.5 degrees (oriented north, north-northeast, northeast,
east-northeast, etc. of the center of the source)
• 228-658 near-facility community receptor points: a Cartesian grid of receptor points
filling the space between the fenceline receptors and an outer-ring receptors 1000 m
from the source, with 100-m spacing between receptors. The number of community
receptors varies by source type and area size (the area between the two rings of
receptors grows as the source becomes bigger), as shown in Table 11.
Air-concentration metrics were obtained at a 1.8-m height above ground to coincide with
typical breathing height (for all source types), and deposition metrics were obtained at ground
level (only for point and fugitive sources).
31
-------
Table 11. Number of receptor points modeled by source type and source size.
Source Type
Number of Receptor Points
Fenceline
Community
Total
Point
16
228
244
Fugitive (25 m2)
16
236
252
Fugitive (50 m2)
16
236
252
Fugitive (100 m2)
16
234
250
Fugitive (200 m2)
16
236
252
Fugitive (400 m2)
16
240
256
Area (10 acres)
16
296
312
Area (20 acres)
16
316
332
Area (50 acres)
16
366
382
Area (200 acres)
16
502
518
Area (500 acres)
16
658
674
5.7 Recommendation on Default Selections in IIOAC
The default scenarios recommended generally characterize central-tendency (mean) and high-
end (95th percentile) exposures for a range of exposure scenarios based on the pre-run
AERMOD results. Using these selections provides a central-tendency and high-end conservative
estimate for air concentrations and exposure doses. In some cases, one scenario may lead to
the highest air concentrations and deposition at the fenceline while not at the community
receptors, or at one single point but not at another, and one scenario may lead to the highest
air concentrations at a location but not the highest deposition amounts. The ranking of
scenarios was determined based on the average air concentration and deposition values from
five years of meteorological data (2011 through 2015) across all fenceline and community
receptors within each scenario. For area and fugitive sources, only the base scenarios (i.e.,
those with an area of 50 acres and 100 m2, respectively) were considered in the analysis. The
default scenarios described below are recommended for first-tier assessment, and follow-up air
modeling may be needed based on chemical and site specific conditions.
5.7.1 Default Source Scenarios
The fugitive source is more conservative relative to the point sources and and area sources.
For point sources, the stack source is considerably more conservative than the incinerator
sources. For fugitive sources, smaller fugitive sources will generally provide the most
conservative air concentrations and deposition amounts. For area sources, smaller area sources
will generally provide the most conservative air concentrations (deposition amounts are not
modeled for area sources).
32
-------
5.7.2 Default Selection of Vapor or Particle
Vapor phase chemicals are the only option available for area sources.
For point sources, modeling with coarse particles will generally be the most conservative option
overall. For fugitive sources, coarse particles will generally be the most conservative option for
deposition amounts, while vapor-phase chemical will generally be the most conservative option
for air concentrations.
5.7.3 Default Selection of Urban or Rural
Rural is the only option available for area sources.
For point sources, the urban selection will tend to lead to higher air concentrations and
deposition amounts, while for fugitive sources the rural selection will. However, the magnitude
of the effect of the urban setting is dependent on the selected meteorology station and the
distance between the source and the receptor.
5.7.4 Default Selection of Meteorology
The Lake Charles, LA station (South, Coastal) is a good choice for conservative air concentration
results as it has the highest average air concentration from among the 14 station options;
however it has the 8th highest deposition rank. The Pittsburgh, PA (Northeast, Inland) station is
a good choice for conservative particle deposition results, as it has the highest deposition
values from among the 14 station options; however it has the lowest air concentration values. If
both air concentration and particle deposition are considered, the Iowa City, IA station (East
North Central) is a good choice for conservative results, as it has the 3rd and 5th highest air
concentration and deposition ranks, respectively, among 14 station options.
For central tendency results, the Sioux Falls, SD station (West North Central) would be a good
selection when considering either air concentration or particle deposition, as it exhibits the 6th
and 9th highest average air concentration and deposition values, respectively, out of the 14
station options.
5.8 Summary of AERMOD Runs for Point Sources
The pre-run point source scenarios are outlined in Table 12. These scenarios were defined to
represent a range of possible site conditions across the U.S. In total, there were 252 pre-run
AERMOD scenarios for point sources: 3 source type scenarios x 2 population scenarios x 3
particle/vapor scenarios x 14 meteorology scenarios.
Point sources include stack and incinerator sources that release emissions directly to the air,
with source specifications (heights, diameters, temperatures, and velocities) that differ among
the three source types. AERMOD runs for point sources were performed using 1 g/s emission
rates that were constant in time, with 1 g/s of vapor phase emissions modeled separately from
33
-------
1 g/s fine particle phase emissions and 1 g/s coarse particle phase emissions. User-specified
variations on emission rates are accounted for in IIOAC itself.
Point source runs included 14 meteorological stations (years 2011-2015) covering a wide range
of climatological, geographic, and land-cover characteristics. Each point source run included an
urban option that modified turbulence parameters to better represent the urban heat island (as
well as runs without the urban option). Each run included 16 fenceline polar-grid receptors and
228-658 community Cartesian-grid receptors, with air concentrations estimated at 1.8-m above
ground and deposition estimated at ground level.
Table 12. AERMOD scenarios for point sources.
Parameter
Options
Comments
Source Type
Stack
h=10 m, d=2 m, t=300 K, v=5 m/s
Concentrations from stack emissions
should generally be larger than those
from the other point sources, and
concentrations from the high-
temperature incinerator's emissions
should generally be lower than those
from the other point sources
Incinerator 1
h=20 m, d=l m, t=500 K, v=15 m/s
Incinerator 2
h=50 m, d=2 m, t=1,200 K, v=15 m/s
Population
Not urban
Urban (1 million people)
Constant
Emission
Duration
1 hour once per day (12-l)pm
Run AERMOD under constant emission,
extract relevant data for each emission
duration using post-processing code
4 continuous h/d (12-4pm)
8 continuous h/d (8-4pm)
Constant
Particle/Vapor
Vapor (no deposition)
Vapor phase pollutants and finer-sized
particles will disperse more widely than
larger-sized particles, which deposit
closer to the source
Fine particles (PM2.5, 2.5 pim)
Coarse particles (PM10, 10 pim)
Meteorology and
Land Cover
Iowa City, IA
Use meteorological data from years
2011-2015, land-cover data from year
1992, average wetness for Bowen ratio
calculations, and local determinations of
arid/non-arid and vegetative seasons
Camp Springs, MD
Pittsburgh, PA
Everett, WA
Idaho Falls, ID
Topeka, KS
Lake Charles, LA
New River, NC
Atlanta, GA
Grand Junction, CO
Point Mugu, CA
Las Vegas, NV
Sioux Falls, SD
Rockford, IL
Note: h = height, d = inside diameter, t = exit gas temperature, v = exit gas velocity, m = meters, K = Kelvin, m/s =
meters per second, h/d = hours per day
34
-------
5.9 Summary of AERMOD Runs for Fugitive and Area Sources
The pre-run fugitive and area source scenarios are outlined in Table 13. Like point sources,
fugitive and area sources release emissions to the air; unlike point sources, fugitive and area
releases are spread out evenly across a two-dimensional area. All were modeled using 1 g/s
emission rates that were constant in time (identical to point sources). Area sources emitted
only vapor phase pollutant, as emissions are due to volatilization of chemicals to the air from
the soil and water. Fugitive sources had 1 g/s of vapor phase emissions modeled separately
from 1 g/s fine particle phase emissions and 1 g/s coarse particle phase emissions. These
emission rates were then spread evenly around the two-dimensional area, so that a 25 m2
fugitive source emitted at a rate of 0.04 g/s/m2, for example. User-specified variations on
emission rates are accounted for in IIOAC itself. All fugitive sources emitted at 3.05 m above
ground (the default value used for fugitive sources in the 2011 NATA), while all area sources
emitted at ground level.
The sizes modeled in AERMOD for the two-dimensional areas were selected to be
representative of the range of sources OPPT is likely to consider for each type of source.
However, IIOAC is designed to estimate pollutant outputs for user-customized sizes. Test
modeling showed an approximate linear relationship between the sizes of larger sources and
the AERMOD results. A number of smaller sources were also included because the test
modeling showed more variability in their AERMOD results, introduced by the surface
geometry. In total, 30 AERMOD scenarios were run to derive the regression relationship
between the size of a fugitive source and outputs of air concentrations and deposition—2
population scenarios x 3 particle/vapor scenarios x 5 source sizes (using the Iowa City, IA
meteorology scenario). Similarly, five AERMOD scenarios were run to derive the regression
relationship between the size of an area source and outputs of air concentrations,
corresponding to five source sizes (using the Iowa City, IA meteorology scenario). A baseline
fugitive-source size of 100 m2 is used in IIOAC, upon which the regression is applied to relate
the outputs of the baseline source size to those of the user-entered source size—84 fugitive-
source AERMOD scenarios were run with the baseline size (2 population scenarios x 3
particle/vapor scenarios x 14 meteorology scenarios). A baseline area source size of 50 acres is
used in IIOAC, upon which the regression is applied to relate the outputs of the baseline source
size to the user-entered source size—14 area source AERMOD scenarios were run with the
baseline size (corresponding to the 14 meteorology stations). See Section 1 on scaling factors
for fugitive and area sources, and see Section 5.4 for a discussion on using the Iowa City, IA
station as a representative U.S. station.
Fugitive and area source runs included 14 meteorological stations (years 2011-2015) covering a
wide range of climatological, geographic, and land-cover characteristics. Area sources were run
without the urban option in AERMOD (assuming soil and water sources are not located in
urbanized areas), while fugitive sources, like point sources, included runs with and without the
urban option (the urban option modifies turbulence parameters to better represent the urban
heat island). As with the point-source runs, each fugitive and area source run included 16
fenceline polar-grid receptors and hundreds of community Cartesian-grid receptors at a spacing
35
-------
of 100 m. Air concentrations were estimated at 1.8-m above ground and, for fugitive sources
only, deposition values were estimated at ground level.
36
-------
Table 13. AERMOD scenarios for fugitive and area (water and soil) sources.
Parameter
Fugitive Source Options
Area Source Options
Comments
Area Size
and Release
Height
25 m2
1=5 m, w=5 m, h=3.05 m
10 acres
1=201 m, w=201 m, h=0 m
50 m2
1=7.1 m, w=7.1 m, h=3.05 m
20 acres
1=284 m, w=284 m, h=0 m
100 m2 (default)
1=10 m, w=10 m, h=3.05 m
50 acres (default)
1=450 m, w=450 m, h=0 m
200 m2
1=14.1 m, w=14.1 m, h=3.05 m
200 acres
1=900 m, w=900 m, h=0 m
400 m2
1=20 m, w=20 m, h=3.05 m
500 acres
1=1,422 m, w= 1,422 m, h=0 m
Population
Not urban
Not urban
Urban (1 million people)
Constant
Emission
Duration
1 hour once per day (12-l)pm
Constant
For Fugitive sources run AERMOD under
constant emission, extract relevant data for
each emission duration using post-processing
code
4 continuous h/d (12-4pm)
8 continuous h/d (8-4pm)
Constant
Particle/
Vapor
Vapor (no deposition)
Vapor (no deposition)
Vapor phase pollutants and finer-sized particles
will disperse more widely than larger-sized
particles, which deposit closer to the source
Fine particles (PM2.5, 2.5 |am)
Coarse particles (PM10, 10 |am)
Meteor-
ology and
Land Cover
Iowa City, IA
Iowa City, IA
Use meteorological data from years 2011-2015,
land-cover data from year 1992, average
wetness for Bowen ratio calculations, and local
determinations of arid/non-arid and vegetative
seasons
Camp Springs, MD
Camp Springs, MD
Pittsburgh, PA
Pittsburgh, PA
Everett, WA
Everett, WA
Idaho Falls, ID
Idaho Falls, ID
Topeka, KS
Topeka, KS
Lake Charles, LA
Lake Charles, LA
New River, NC
New River, NC
Atlanta, GA
Atlanta, GA
37
-------
Parameter
Fugitive Source Options
Area Source Options
Comments
Grand Junction, CO
Grand Junction, CO
Point Mugu, CA
Point Mugu, CA
Las Vegas, NV
Las Vegas, NV
Sioux Falls, SD
Sioux Falls, SD
Rockford, IL
Rockford, IL
Note: I = length, w = width, h = height, m = meters, h/d = hours per day
38
-------
6 Modeling Approach for Facility Sources
6.1 Overview and Assumptions
IIOAC calculates outdoor air concentration based on post-processed AERMOD results for unit
emission and adjusts these values by the user-specified release duration and days of release per
year. Indoor air concentrations are calculated by multiplying the outdoor air concentration by
the indoor-outdoor ratio.
Releases from facility sources (i.e., point or fugitive) emit directly to outdoor air and are
assumed to follow either a pattern of consecutive or cyclical (evenly spaced) release days. Once
an emission stops, air concentrations fall to zero instantaneously, which is a property of
AERMOD. Because AERMOD calculates dispersion, air concentrations, and deposition in hourly
time steps, IIOAC uses hourly emissions and hourly meteorological data. For the IIOAC tool,
AERMOD was run with a unit emission rate of 1 g/s for all scenarios. The hourly model outputs
were then post-processed to calculate daily-and annual-averaged outdoor air concentrations.
For fugitive sources, AERMOD runs were set up using an area size of 100 m2.
6.2 Post-Processing of AERMOD Hourly Air Concentrations
All AERMOD emission scenarios were run with constant emission for five years, based on EPA
regulatory guidance (US EPA, 2017d). However, within IIOAC, users can select from release
durations of 1, 4, 8, or 24 hours per day, which correspond to emission times of 12-lpm, 12-
4pm, 8am-4pm, and all day, respectively. For release durations that are 1, 4, and 8 hours,
AERMOD data for hours when the emission is not occurring were set to zero, as shown in the
table (Table 14) for one day.
Table 14. Example of hourly concentrations set to zero when there is no emission for a 1, 4, and
8 hour release duration.
Pre-processed Hourly
AERMOD Outputs
Post-processed Hourly AERMOD Outputs
Continuous Emission, 1 g/s
Air Concentration (|j,g/m3) -
Set to Zero When No Emission
Day
Time
Air
Concentration
(Mg/m3)
1 hr Duration
4 hrs Duration
8 hrs Duration
1-Jan
0:00
0.06
0
0
0
1-Jan
1:00
0.51
0
0
0
1-Jan
2:00
0.52
0
0
0
1-Jan
3:00
0.00
0
0
0
1-Jan
4:00
0.60
0
0
0
1-Jan
5:00
0.37
0
0
0
39
-------
1-Jan
6:00
0.59
0
0
0
1-Jan
7:00
0.60
0
0
0
1-Jan
8:00
0.79
0
0
0
1-Jan
9:00
0.51
0
0
0.51
1-Jan
10:00
0.43
0
0
0.43
1-Jan
11:00
0.49
0
0
0.49
1-Jan
12:00
0.31
0
0
0.31
1-Jan
13:00
0.71
0.71
0.71
0.71
1-Jan
14:00
0.04
0
0.04
0.04
1-Jan
15:00
0.86
0
0.86
0.86
1-Jan
16:00
0.16
0
0.16
0.16
1-Jan
17:00
0.34
0
0
0
1-Jan
18:00
0.15
0
0
0
1-Jan
19:00
0.60
0
0
0
1-Jan
20:00
0.87
0
0
0
1-Jan
21:00
0.05
0
0
0
1-Jan
22:00
0.24
0
0
0
1-Jan
23:00
0.96
0
0
0
For each receptor group and AERMOD emission scenario (e.g., fugitive, urban setting, fine
particles, Northeast climate region), all five years of the AERMOD hourly outputs were post-
processed to determine the mean and high-end (defined as the 95th percentile) daily-averaged
and annual-averaged concentrations. This was done for all release days (i.e., 1 through 365) per
year. For the first four years of data, the annual-averaged concentrations start with the first day
of release and extends for a one year period. For year 5, the annual-averaged concentration is
simply the year 5 average in order to have 365 days to average over. The post-processing
results are organized in Excel lookup tables like the example shown in Table 15.
Table 15. Example lookup table for one AERMOD emission scenario and receptor group.
Number of Release Days
Release
Exposure Metric
Duration
1 2 3 4 5 6 7 8 9 10 ... 365
(hrs/day)
Mean Daily Average
1
4
8
24
Filled in with post-processed
1
AERMOD results in |Jg/m3
High-End Daily Average
4
8
40
-------
24
1
Mean Annual Average
4
8
24
1
4
High-End Annual Average
8
24
6.3 Calculating Outdoor Air Concentration and Particle
Deposition Estimates
IIOAC calculates outdoor air concentration and particle deposition based on the release
duration and number of days of release per year entered by the user (e.g., release occurs 4
hrs/day for 52 days in a year). An adjusted emission rate is first calculated, as shown in
Equation 1, to take into account the release duration and convert the user-defined mass
released per day into g/s.
ERadl = ™ ' °'2778 (1)
where ERadj = adjusted emission rate [g/s]
ER = user-defined mass released per day [kg/day]
h = emission duration [hrs/day]
0.2778 = conversion factor from kg/hr to g/s
Air concentrations are calculated in Equation 2 by scaling the post-processed AERMOD result,
obtained based on an emission of 1 g/s, by the adjusted emission rate. For fugitive sources,
scaling by just the adjusted emission rate gives an air concentration corresponding to an area
size of 100 m2, the same as that used in the AERMOD runs. To account for a different area size,
an area size scaling factor, SFj, is applied. Further details on the area size scaling factor is
described in Section 9.2.
ER H'
Coutdoor — —~' SFj ' Postprocessed AERMOD result (2)
1 g/s
where Coutdoor = outdoor air concentration [|Jg/m3]
ERadj = adjusted emission rate [g/s]
SFj = scaling factor for fugitive area size j [-]; set to 1 for point sources
For point and fugitive sources, three particle size scenarios are available:
41
-------
• Fine particles (with a mass-mean aerodynamic diameter of 2.5 |am),
• Coarse particles (with a mass-mean aerodynamic diameter of 10 |am), and
• Vapor (no particles).
All calculated air concentrations of fine and coarse particles are capped by an upper limit equal
to the National Ambient Air Quality Standards (NAAQS) for particulate matter (PM) (US EPA
2016b). These limits are 35 and 150 |am/m3 for fine and coarse particles (i.e., the NAAQS for
PM2.5 and PM10), respectively. For vapors, the chemical is released in gaseous form and
therefore there is no transfer from one phase to another. IIOAC currently does not set an upper
limit for point and fugitive sources in vapor form.
When there are multiple releases at a single site, the air concentrations estimates are totaled
for the site. Concentrations of particulates are capped by the upper-limit before totaling.
6.4 Aggregation of Stack and Incinerator Sources into Single
Point Source
In the case where an input file is imported into IIOAC and all sources are selected, the tool
aggregates the outputs of stack and incinerator sources into a single point source output. As an
example, if an emission scenario has both a stack and incinerator source, the aggregated point
source output at a specific receptor is calculated as:
fig
High-end daily-averaged air concstack = 4 —7
ms
High-end daily-averaged air concincinerator = 2.9 —7
ms
fig
High-end daily-averaged air concvoint = 4 + 2.9 = 6.9 —-
ms
Appendix C provides an example calculation using a hypothetical site with three releases of
varying release amounts, frequencies, and durations.
7 Modeling Approach for Area Soil Sources
7.1 Overview and Assumptions
Area soil sources are modeled as a batch system, as shown in Figure 12, using the following
assumptions:
(i) The chemical only leaves the soil through volatilization.
(ii) Releases are evenly applied across the entire surface area of the soil.
(iii) Releases are applied immediately at the start of each day.
(iv) The first day of all releases is January 1st.
42
-------
(v) Releases are evenly timed throughout the year.
(vi) The time scale of emissions from area soil sources is much longer (hours to days)
than the time scale of emissions directly to air (i.e., point and fugitive sources),
allowing short-term impacts of meteorology to be excluded.
Figure 12. Schematic of area soil source, modeled as a batch system.
For area sources, AERMOD outputs are based on an emission rate of 1 g/s and an area of 50
acres. To calculate the chemical concentration in air as a result of volatilization from the soil,
IIOAC must first calculate the mass flux from the soil to air. The mass flux is dependent on both
the physicochemical properties, the area of the source, and the mass of chemical on the soil.
Depending on the chemical, the mass flux can differ greatly, for example, a 100 kg/m2 release
over an area of 200,000 m2 results in a flux of 1.48xl0~4 and 0.29 kg/m2/day for Aldicarb and
hexachloroethane, respectively. As such, a mass balance must be performed each day to
determine the mass on the soil at the start of each day, which is then used to calculate the
corresponding emission rate.
7.2 Equations to Calculate Daily-Averaged Air
Concentrations
All daily-averaged air concentrations are calculated within IIOAC using the equations described
in this section.
IIOAC uses a modified version of the Wood row and Sieber equation (1997) to calculate the
mass flux,/, due to volatilization from soil. The original study developed empirical natural
logarithm correlations between flux for pesticides and known physicochemical properties. EPA
later modified and used the equation in the Volatilization Screening Tool (US EPA, 2014a)
developed by EPA to estimate screening-level air concentrations downwind of fields treated
with semi-volatile pesticides. The modified equation used in Volatilization Screening Tool is
given by Equation 3, with R defined in Equation 4.
Release/Input
Volatilization
U
exp(0.8688 -fl+21.535)
3600
¦ (8.64 X 1(T5)
(3)
and
43
-------
R = In
\Sol ¦ Koc)
(4)
where / = mass flux from soil to air [kg/m2/day]
VP = vapor pressure [Pa]
AR = mass of chemical per area [kg/ha]
Sol = solubility in water [mg/L]
Koc = organic carbon sorption coefficient [mL/g]
0.8688 = regression coefficient (US EPA, 2014a)
21.535 = regression coefficient (US EPA, 2014a)
8.64xl0"5 = conversion factor from |j,g/m2/s to kg/m2/day
Note that the equation above was developed using data points for —16 < R < 0. Users that
enter chemical properties resulting in R values outside this range will receive a warning
message within IIOAC indicating that the calculated results are potentially outside the scope of
applicability.
To determine the outdoor air concentration, the AERMOD results are scaled up as shown in
Equation 5. For each flux value calculated, an emission rate in kg/day is determined by
multiplying the flux with the user-specified area size. The AERMOD results are then scaled up by
the calculated emission rate and by an area size scaling factor, SFj. The area size scaling factor,
SFj, accounts for different area sizes and is further described in Section 9.2. As short-term
impacts of meteorology can be excluded (see assumptions above), an hourly concentration
averaged over the day was used as the AERMOD result to be scaled up.
^outdoor,i = ' 0-01157 ¦ SFj • Postprocessed AERMOD resultt (5)
where Coutdoor i = air concentration for day/'[|j,g/m3]
Ji = mass flux from soil to air for day/'[kg/m2/day]
A = area of the source, as defined by the user [m2]
0.01157 = conversion factor from [kg/day to g/s]
SFj = scaling factor for area sizey [-]
All calculated air concentrations are compared to the saturation air concentration, which is the
concentration at which exchange between the gas and liquid phases of the chemical are at
equilibrium, and is calculated as shown in Equation 6.
r _ (1.33X105)-VP-MW .
^sat
where Csat = saturation air concentration [|j,g/m3]
VP = vapor pressure [Torr]
MW = molecular weight [g/mol]
R = universal gas constant, 8.314 J/mol-K
44
-------
T = absolute temperature, set to 298 [K]
1.33xl05 = conversion factor from Torr to Pa and from g to |j,g
Calculated air concentrations may not be greater than the saturation air concentration. In the
case when a calculated mass flux results in an air concentration above the saturated air
concentration, IIOAC replaces the calculated air concentration for that day with the saturation
air concentration. IIOAC returns any chemical mass above the saturation air concentration to
the area soil source to be used in the calculation for the next day. IIOAC uses concentrations at
the inner receptors to determine the excess concentration and corresponding mass to be
returned to the soil.
For each day, to calculate Jit the AR value is calculated as shown in Equation 7.
ARt = —
1 A
where ARi = mass of chemical per area [kg/ha]
Mi = mass of chemical on the soil at the start of day/'[kg]
A = area of the source [m2]
To determine the mass of chemical on the soil at the start of day /', a mass balance must be
performed as follows:
AM = Mj — Mi_1 = TRaddi ¦ At — MV0j j_x (8)
where AM = change in mass between the start of day /' and the start of day i-1 [kg]
Mi = mass of chemical on the soil at the start of day/'[kg]
Mj_x = mass of chemical on the soil at the start of day i-1 [kg]
TRadd.i = total mass added to soil at the start of day /' from all releases [kg/day]
At = change in time [days]
Mvoi,i-i = mass volatilized in day i-1 [kg]
The mass volatilized is calculated in Equation 9 as the mass flux multiplied by the surface area
of the source.
Mvoi,i-1 = /i-i -A-At (9)
where Mvon_1 = mass volatilized in day i-1 [kg]
/j_x = mass flux from soil to air in day i-1 [kg/m2/day]
A = area of the source [m2]
At = change in time [days]
For day /' and At of 1 day, the mass on the soil at the start of the day is given by Equation 10.
Mi = Mj_! + (TRaddii -Ji_x ¦ A) ¦ (1 day) (10)
45
-------
where M,-
= mass of chemical on the soil at the start of day /' [kg]
= mass of chemical on soil at the start of the previous day /' [kg]
= total mass added to soil at the start of day /' from all releases [kg/day]
= mass flux on day/'-l [kg/m2/day]
= area of the source [m2]
A
Using a hypothetical site with three releases with varying release amounts and release
frequencies, Table 16 shows how daily air concentrations are calculated using the equations
presented above. Step-by-step details of the calculations are presented in Appendix D.
Once daily air concentrations are determined for the 5-year period (2011-2015), the high-end
and mean values, from all the daily-averaged values over the 5-year period, can be calculated. A
high-end and mean annual-averaged concentration can also be determined for each of receptor
group.
46
-------
Table 16. Summary of mass flux and air concentrations due to emissions from soil at a hypothetical site with three releases.
Total Mass Added
Mass on
Mass
AERMOD
Release
Release
Release
to Soil from all
Soil at Start
Mass per
Mass Flux,
Volatilized
Unit
Day
#1
#2
#3
Releases, TRadd
of Day, M
Area, AR
J
from Soil, MVOi
Value3
Air Cone
(kg/day)
(kg/day)
(kg/day)
(kg/day)
(kg)
(kg/ha)
(kg/m2/day)
(kg/ha)
(ug/m3)
(ug/m3)
1
100
75
25
200.00
200.00
10.00
2.70E-04
5.43
3.41b
2
0
0
25
25.00
212.88
10.64
2.85E-04
19.1
12.7b
3
0
0
25
25.00
234.45
11.72
3.10E-04
34.6
24.9b
4
0
0
25
25.00
257.55
12.88
3.36E-04
16.5
12.9b
5
0
0
25
25.00
278.55
13.93
3.60E-04
12.5
10.4b
6
100
0
25
125.00
398.27
19.91
4.91E-04
30.9
35.3b
7
0
0
25
25.00
421.14
21.06
5.15E-04
29.1
34.9b
8
0
75
25
100.00
518.88
25.94
6.17E-04
16.9
24.3b
9
0
0
25
25.00
540.00
27.00
6.39E-04
3.96
5.90b
10
0
0
25
25.00
548.40
27.42
6.48E-04
6.79
10.3b
continue until M or J falls below a threshold level of 10"7
a AERMOD Unit Value based on 1 g/s of emission
b Indicates calculated air concentration exceeds saturation air concentration
47
-------
8 Modeling Approach for Area Water Sources
8.1 Overview and Assumptions
The approach for area water sources estimates chemical volatilization from water into air, and
is can be used to model releases from wastewater treatment plants or standing bodies of water
such as bays, estuaries, or impoundments, or holding ponds. Area water sources are modeled
as either a batch or continuous flow-through system, as shown below in Figure 13. The
difference between the two systems is the extra chemical loss term for a flow-through system,
due to the continuous flow of water into and out of the system.
Release/In
(a) Batch (b) Continuous flow-through
Figure 13. Schematic of area water source, modeled as (a) batch and (b) continuous flow-
through system.
The following assumptions are made for area water sources:
(i) The chemical only leaves the water through volatilization and/or continuous flow
out of the system (chemical removal through biodegradation and adsorption to
sludge are not considered).
(ii) Instantaneous mixing occurs.
(iii) Steady state conditions reached immediately.
(iv) Releases are applied immediately at the start of each day.
(v) The first day of all releases is January 1st.
(vi) Releases are evenly timed throughout the year.
(vii) The time scale of emissions from area water sources is much longer (hours to days)
than the time scale of emissions directly to air (i.e., point and fugitive sources),
allowing short-term impacts of meteorology to be excluded.
Like the approach for area soil sources (Section 7), the approach for area water sources uses air
concentration results from AERMOD with an emission rate of 1 g/s and an area of 50 acres. To
calculate the chemical concentration in air from the area water source, IIOAC uses mass
balance calculations to estimate the mass volatilized each day.
48
-------
8.2 Equations to Calculate Daily-Averaged Air
Concentrations
All daily-averaged air concentrations are calculated within IIOAC using the equations described
in this section. The equations in this section were developed for a flow-through system with a
flowrate in and out of the system. For batch systems, the same equations apply; however, the
flowrate would be set to zero.
As shown in Equation 11, outdoor air concentrations from area water sources are calculated by
multiplying the mass volatilized each day by the unit emission AERMOD result (i.e., in |J,g/m3)
and an area size scaling factor, SFj. The area size scaling factor, SFj, accounts for the size of
the area water source and is described in detail in Section 9.2. As short-term meteorological
effects can be excluded (see assumptions above), the area water source calculations uses daily-
averaged AERMOD results.
Coutdoor.i = Mv°lX^y " 0.01157 ¦ SFj ¦ Postprocessed AERMOD resultt (11)
where Coutdoor i = air concentration for day/'[|j,g/m3]
Mvol i = mass volatilized on day /' [kg]
0.01157 = conversion factor from kg/day to g/s
SFj = scaling factor for area size j [-]
All calculated air concentrations are capped at chemical specific saturation air concentrations
(Equation 6). If a calculated air concentration is above the saturated air concentration, IIOAC
replaces the calculated air concentration for that day with the saturation air concentration. For
mass balance purposes, IIOAC returns the chemical mass in excess of the saturation air
concentration to the area water source. IIOAC uses concentrations at the inner receptors to
determine the excess concentration and corresponding mass to be returned to the water.
For each day, the mass volatilized on day /' is calculated with Equation 12.
Mvoi,i — kVoi ¦ V ¦ (1 day) (12)
where Mvon = mass volatilized on day /' [kg]
kvoi = volatilization rate constant [1/day]
Cj = chemical concentration in water on day/'[kg/m3]
V = volume of the water source [m3]
The volatilization rate constant, kvoi, is calculated as shown in Equation 13.
i ln(2) , x
Koi = —^4- (is)
1 /2 24
where kvoi = volatilization rate constant [1/day]
49
-------
tjy2 = half-life of volatilization from water [hrs]
1
— = conversion factor from hrs to days
24 *
The volume of the water source is calculated using Equation 14.
V = A ¦ d (14)
where V = volume of the water source [m3]
A = area of water source [m2]
d = depth of the water source [m]
The half-life of volatilization from water, t]y2, is calculated in EPA's Estimation Programs
Interface (EPI) Suite screening-level tool (US EPA, 2017b) based on depth of water, water
velocity, and wind velocity. EPI Suite uses a two-film concept described in Thomas (1990) to
estimate mass flux across the air-water interface, and corresponding volatilization half-lives. In
addition to estimating parameters for chemicals with existing CAS numbers, this program also
estimates parameters of new chemicals based on their structure.
The mass volatilized on day /' is dependent on the initial mass in water at the start of the day
and is calculated by performing a mass balance on a control volume as shown below (Equations
15 and 16). The concentration of chemical in the water source is assumed to be the same as the
concentration exiting the system.
change in mass
= mass flow in — mass flow out + total mass from releases — mass volatilized (15)
which is calculated as:
AM = Mj — = «? ¦ Cin ¦ At) - ( Q ¦ Cf_! ¦ At) + (TRt ¦ At) - (kvol ¦ ¦ V ¦ At) (16)
where AM = change in mass between the start of day /' and the start of day i-1 [kg]
Mi = mass of chemical in water at the start of day/'[kg]
Mj_x = mass of chemical in water at the start of day i-1 [kg]
Q = water flow rate of the system [m3/day]
Cin = concentration of chemical in the flow rate into the system [kg/m3]
Cj_x = concentration of chemical in the system on day i-1 [kg/m3]
At = change in time [days]
TRi = total mass added at the start of day/'from all releases [kg]
kvoi = volatilization rate constant [1/day]
V = volume of the water area source [m3]
Substituting C for — and setting Cin to zero because there is no continuous mass flow into the
system gives Equation 17.
50
-------
AM — — ^ i ¦ At + Tfy ¦ At - kvol ¦ ¦ At (17)
For day /' and At of 1 day, the mass in water at the start of the day is given by Equation 18.
Using a hypothetical site with three releases with varying release amounts and release
frequencies, Table 17 shows how daily air concentrations are calculated using the equations
presented above. Step-by-step details of the calculations are presented in Appendix E.
Once daily air concentrations are determined for the 5-year period (2011-2015), the high-end
and mean values, from all the daily-averaged values over the 5-year period, can be calculated. A
high-end and mean annual-averaged concentration can also be determined for each of receptor
group.
(18)
51
-------
Table 17. Summary of mass volatilized and air concentrations due to emissions from water at a hypothetical site with three releases.
Total Mass Added to
Mass in Water
Mass Volatilized
Release
Release
Release
Water from All
at Start of Day,
from Water,
AERMOD
Day
#1
#2
#3
Releases, TRadd
M
Mvol
Unit Value3
Air Cone
(kg/day)
(kg/day)
(kg/day)
(kg/day)
(kg/m3)
(kg)
(ug/m3)
(ug/m3)
1
100
75
25
200
200
30.1
3.42
2.67
2
0
0
25
25
194.6
29.4
33.3
25.3
3
0
0
25
25
190.5
28.7
63.8
47.4
4
0
0
25
25
187.4
28.1
37.8
27.6
5
0
0
25
25
184.3
27.7
36.0
25.8
6
100
0
25
125
281.6
42.3
11.5
12.6
7
0
0
25
25
264.3
39.7
16.5
16.9
8
0
75
25
100
324.6
48.8
19.7
24.8
9
0
0
25
25
300.8
45.2
34.2
40.0
10
0
0
25
25
280.6
42.1
7.00
7.63
continue until M or mass fluxb falls below a threshold level of 10"7
a AERMOD Unit Value based on 1 g/s of emission
b For threshold level, a mass flux was calculated for each day
52
-------
9 Scaling Factors for Fugitive and Area Water/Soil Sources
9.1 Overview
For fugitive and area water/soil sources, IIOAC calculates the outdoor air concentration, indoor
air concentration, and acute and chronic doses for a user-specified area size and release
amount. For fugitive sources, particle deposition is also calculated. However, as the tool uses
pre-run AERMOD results with a fixed area size and release amount, scaling factors must be
used to scale the AERMOD results according to the area size and the release amount. The
application of scaling factors is automatically performed in IIOAC and has been incorporated in
the equations in the previous sections to calculate air concentration. This section provides
further detail on the scaling factors calculated and used in IIOAC to adjust for both area size and
release amount.
9.2 Scaling Factor for Different Area Sizes
For fugitive sources and area water/soil sources, AERMOD runs were performed using set area
sizes of 100 m2 and 202,343 m2, respectively, the latter which corresponds to 50 acres.
To determine the scaling factor for various area sizes, five area sizes were run in AERMOD using
meteorological data for the Iowa City station. It is assumed that the remaining 13 met locations
follow the same relationship between area size and air concentration. The rationale for
selecting Iowa City to represent all met stations is provided in Section 5.4. A regression
equation was fit to the AERMOD outputs to determine the relationship between air
concentration and area size for the inner ring and community receptor groups. Using the
regression equation, IIOAC estimate air concentrations for area sizes within the range of 25-500
m2 for Fugitive sources and 40,468-2,023,000 m2 (or 10-500 acres) for area Water/Soil sources.
Table 18 provides example data used to determine the regression coefficients. Note that five
different area sizes need to be run separately for fugitive and for area water/soil sources due to
the difference in release height (3.05 m for fugitive sources and 0 m for area water/soil
sources). For the purposes of the example below, the same AERMOD results are used for both
fugitive and area water/soil sources. All regression coefficients for fugitive and area sources are
provided in Appendix A.
53
-------
Table 18. Example data of air concentration as a function of area size.
Average Concentration in 2011-2015
Area
(M-g/m3)
(acre)
(m2)
Inner Ring
Community
10
40,468.6
47.75
9.37
20
80,937.2
35.13
8.30
50
202,343
21.91
6.25
200
809,372
9.45
3.06
500
2,023,430
5.07
1.84
Using inner ring as an illustrative example, the regression equation is obtained from Figure 14.
60.00
^ 50.00
m
3 40.00
c
o
*2 30.00
+->
c
u
§ 20.00
u
< 10.00
0.00
0 500000 1000000 1500000 2000000 2500000
Area Size (m2)
Figure 14. Outdoor air concentration as a function of area size for inner ring receptors.
y =
23,013.534x0-576
R2 = 0.995
•
•
54
-------
The regression equation has the form:
C = a-Ab (19)
where C = air concentration [|j,g/m3]
A = area size [m2]
a = regression coefficient [-]
b = regression coefficient [-]
Using the example above, for inner ring, the regression coefficients are:
a = 23,013
b = -0.5757
The general equation to calculate a scaling factor for any area,./', is given by Equation 20.
SF> = ¦£- <20'
ubase
where SFj = scaling factor for a re ay [-]
Cj = air concentration for areay [|j,g/m3]
Cbase = air concentration for base area [|J,g/m3]
For area water/soil sources, where the base area used in the AERMOD runs is 202,343 m2, a
scaling factor can be calculated for any area,./', as shown in Equation 21.
a- ib ib
SFj =1Lr = — r (21)
J a-Ab 202,343b
For example, if the user enters an area of 200,000 m2 for an area soil/water source, the scaling
factor would be 1.01.
For fugitive sources, the base area used in the AERMOD runs is 100 m2, which leads to a scaling
factor of 0.013 for an area of 200,000 m2.
9.3 Scaling Factor for Different Emissions
In addition to accounting for different area sizes, a scaling factor is also used to obtain air
concentration values corresponding to the user-defined emission rate. AERMOD runs were
performed using 1 g/s emission over each area size. For area soil/water sources, the unit
emission rate is released over an area of 202,343 m2 (50 acres), while for fugitive sources,
AERMOD runs were performed using an area of 100 m2. When determining the scaling factor to
adjust for different emission rates, the area size is assumed to be that used in the AERMOD
runs.
55
-------
The general equation to calculate a scaling factor for different emissions is given by Equation
22.
SFk = (22)
1 g/s
where SFk = scaling factor for different emissions [-]
ERadj = adjusted user-defined emission rate [g/s/m2]
The corresponding air concentration is then calculated as:
Ck ~ Caermod ' (23)
where Ck = air concentration [|j,g/m3]
Caermod = air concentration from AERMOD output [|j,g/m3]
SFk = scaling factor for different emissions [-]
9.4 Overall Calculation of Air Concentration
Combining the two scaling factors above, the air concentration for fugitive and area water/soil
is calculated using Equation 24.
Cj,k = Caermod ¦ SFj ¦ SFk (24)
where Cj k = air concentration corresponding to user-defined parameters [|j,g/m3]
Caermod = air concentration from AERMOD output [|J,g/m3]
SFj = scaling factor for a re ay [-]
SFk = scaling factor for different emissions [-]
9.5 Illustrative Example to Calculate Scaling Factors
Assume for a fugitive source, the air concentration from the pre-run AERMOD results
(corresponding to 100 m2) is 50 |J,g/m3 for a specific receptor. Using the regression coefficients
obtained from Figure 14, where b = -0.5757, a user defined emission of 0.07 g/s, and an area of
650 m2, the air concentration is calculated as:
ug 6 50 -0,5757 0.07 ug
r . = 50 — = 1 19 LE.
air m3 100"0 5757 1 m3
56
-------
10 Indoor: Outdoor Ratio
10.1 Overview
The indoor-outdoor ratio is a metric that is used to describe the relationship between the
concentration of a chemical indoors to the concentration of that chemical outdoors according
to Equation 25.
Cindoor ~ I® ' Coutdoor (25)
where Cindoor = outdoor air concentration [|jg/m3]
Coutdoor = indoor air concentration [|j,g/m3]
10 = indoor-outdoor ratio [-]
The indoor-outdoor ratio is influenced by many factors including the characteristics of the
building, the meteorological conditions, interior sources or sinks, and the physical form of the
chemical substance (particulate or gas). Within IIOAC, indoor air concentrations are calculated
by multiplying the modeled ambient air concentrations by an indoor-outdoor ratio.
IIOAC calculates a mean and high-end indoor air concentration based on the outdoor/ambient
air concentration and mean and high-end indoor-outdoor ratios. In many screening models, the
indoor-outdoor ratio is set to a value of 1, which represents the upper bound of this ratio if
there are no indoor sources. In IIOAC, indoor-outdoor ratios of 0.65 and 1 are used for the
mean and high-end ratios, respectively. The indoor-outdoor ratio of 0.65 is used to calculate
indoor air concentrations corresponding to the mean outdoor air concentration for each
receptor group. The indoor-outdoor ratio of 1 is used to calculate the indoor air concentration
corresponding to the 95th percentile of outdoor air concentration of each receptor group as
shown below in Table 19.
Table 19. Use of indoor-outdoor ratios to calculate indoor air concentration.
Outdoor Air
Indoor Air
Source
Type
Emission
Scenario
Statistic
Location
Concentration
(Hg/m3)
Concentration
(Hg/m3)
Daily
Annual
Daily
Annual
High-End
Fenceline
9.27
0.31
9.27
0.31
Point
Manufacturing
Community
0.23
0.0075
0.23
0.0075
Mean
Fenceline
6.93
0.23
4.50
0.15
Community
0.15
0.0050
0.099
0.0033
The process used to define the indoor-outdoor ratios used in IIOAC are described in the
following sections and include a literature search to identify potential data sources of indoor-
outdoor ratios, followed by analysis of these data. As IIOAC focuses on chemicals with exterior
sources only, studies where interior sources dominated (and indoor-outdoor ratios are above 1)
57
-------
were not the focus of the search and were eliminated to the extent possible, as explained
below.
10.2 Literature Search and Screening
To determine the appropriate values for use in the tool, an open literature search was
conducted. The goal of the search was to identify potential data sources for all chemicals; then,
based on the scope of the information found, a subset of those sources were used to determine
refined indoor-outdoor ratios; the search and screening was not intended to collect data from
all possible sources in a comprehensive fashion. The search was performed using Web of
Science with the search strategy shown in Table 20.
Table 20. Search strategy used to identify potential indoor-outdoor ratio papers.
Database
Keywords and Web of Science Categories
Date
of
Search
Number
of
Returned
Studies
Web of
Science
TOPIC: (air OR concentration OR concentrations)
AND TOPIC:
((indoor OR interior OR indoors) AND (outdoor OR outdoors
OR ambient OR exterior))
Refined by: WEB OF SCIENCE CATEGORIES: (
ENVIRONMENTAL SCIENCES OR PUBLIC ENVIRONMENTAL
OCCUPATIONAL HEALTH OR ENGINEERING
ENVIRONMENTAL )
Timespan: 2000-2017. Indexes: SCI-EXPANDED, SSCI.
May
26,
2017
4,047
From the 4,047 results returned using the search strategy outlined above, title/abstract
screening and full text screening were used to identify data for inclusion in the analysis. A
diagram summarizing the overall screening process is shown in Figure 15 and is discussed
below. A subset of titles/abstracts were screened for relevance, where relevance was
determined using the acceptance criteria shown in Table 21. An initial batch of 1,525 titles and
abstracts were reviewed to find a set of relevant articles that spanned multiple chemicals and
conditions. Of these, 526 were marked as relevant using the acceptance criteria. This was
termed the "Round 1" screening.
58
-------
Title/Abstract Screening with Acceptance Criteria |
Full Text Screening with
Usability Criteria
W
Total
Usable
80
Figure 15. Summary of title/abstract and full text screening.
Table 21. Acceptance criteria applied in title/abstract screening.
Quality Criterion
Acceptance Specification
Currency
• 2000 to present
Geographic Scope
• North America and Europe
Accuracy/Reliability
• Publication in peer-reviewed journal and is a primary source
• Source has documented qualifications/ credentials to discuss
particular topic
• The chemical measured is clearly noted in the text
Unbiased
• Objective of the information is clear
• Methodology is designed to answer a specific question and is
clearly described
Comparability
• Range of ratios is comparable to other studies of a similar chemical
Representativeness
• Sample size (in terms of number of buildings sampled or number of
time points sampled) is greater than or equal to five
Full review of all 4,047 references was not feasible with available resources and was not
necessary to meet the project objectives, so prioritization techniques were used to select the
full set of studies to be screened. For prioritization, text analytic algorithms (K-means and Non-
negative matrix factorization, Varghese et al., 2017) were used to find studies whose titles and
abstracts were similar to the relevant studies in the initial search, as shown in Figure 16. These
algorithms create a user-defined number of study clusters based on keyword similarities in the
title and abstract, and each algorithm is broadly-accepted in the text analytics scientific field.
For this analysis, each of the algorithms was used to bin the studies into 10, 20, or 30 clusters,
for a total of six different cluster analyses (six large circles in the figure). A random sample of
sixty of the studies identified as relevant during Round 1 was included in the full body of
literature that was clustered and served as "tracer" studies (pink circles in the figure). The
tracer method involves following these relevant studies and determining the clusters the
majority occur in; these clusters are then deemed more likely to contain other (as-yet
unidentified) relevant studies.
59
-------
To identify these high-priority clusters, the clusters containing up to 75% of the Round 1
relevant tracer studies were collected for each of the six analyses. These were termed the "high
concentration of tracer" clusters. All non-tracer studies in those high-concentration clusters
were then marked as "positive" for that analysis. Across the six analyses, studies were retained
as sufficiently similar to the Round 1 relevant studies if they were "positive" in two or more
analyses; studies that clustered with the Round 1 relevant studies in zero or one analysis were
set aside from further screening. This ensemble method is used to increase confidence in the
selection of prioritized studies by mitigating uncertainty from each individual analysis. After the
prioritization, of the remaining 2,522 studies, 1,727 were identified as similar to the relevant
tracer studies and their titles and abstracts were screened in Round 2.
60
-------
e©©@®
GOO0
NMF
Algorithm
10 20 30
clusters clusters clusters
Figure 16. Ensemble text analytics method for prioritizing studies for screening.
K-Means
Algorithm
/©©OOOO
OOQOOO
. ©©©©©
\©0©OO
V©©©^
Q Algorithm/number of
clusters combination
^ Cluster identified by
algorithm
q Tracer study (identified
as relevant during
Round 1 screening)
Studies to be
prioritized
O Prioritized clusters for
each analysis (have up
to 75% of tracers)
61
-------
After the completion of Round 1 and 2 title/abstract screening, 467 studies were identified as
potentially relevant based on the title and abstract. Full text screening was then implemented
for those studies with additional usability criteria (Table 22). These criteria focused on the
practicality of extracting information from the study and on limiting the impact of indoor
sources.
With regard to limiting indoor source effects, note that IIOAC is intended for use to estimate
exposure from ambient sources for a wide variety of chemicals. The indoor sources of each
chemical is likely unknown and is being excluded from consideration at this time. Thus, the
indoor-outdoor ratio should not reflect ratios where indoor sources played a large role in the
overall indoor concentration. Figure 17 illustrates how indoor sources might contribute to the
overall indoor-outdoor ratio, where strong indoor sources might lead to ratios either above one
(left) or less than one (middle). Ideally, we want to capture the indoor-outdoor ratio when no
indoor sources are present (right). However, practically speaking, it is impossible to exclude
studies with strong indoor sources without explicit information about the relative contribution
from the study authors. Because this information is not typically available, the screening
focused on studies where the indoor-outdoor ratio mean is less than one; this method provides
a simple way to limit the impact of indoor sources (excludes studies where the mean
measurement resembles the left side of Figure 17), although it does not completely eliminate
the impact (includes studies where the mean measurement may resemble the center of Figure
17). Studies with means greater than one were handled differently in two different cases:
• Study provided individual measurements: if the study provided individual measurements,
the measurements greater than one were excluded and the mean was recalculated by the
screener; this occurred in 23 of the 404 total measurements used in the final analysis. This
recalculation was intended to refocus the calculation on the study buildings without strong
indoor sources.
• Study did not provide individual measurements: if the study did not provide individual
measurements, the study was excluded as not usable.
In this round, 80 studies were identified as both relevant and usable.
Table 22. Usability criterion applied in full text screening.
Quality Criterion
Usability Specification
Usability
• PDF of article is available
• At least one measurement of indoor-outdoor ratio in the study is
less than one and the mean indoor-outdoor ratio is not greater
than one. If measurements greater than one from the original
study can be excluded and a new mean calculated, this calculation
was performed and the resulting mean was flagged.
• Ratios are summarized in tables (no figures were digitized)
62
-------
Outdoor
Sources
Indoor
Sources
Outdoor
Sources
Outdoor
Sources
Indoor
Sources
Indoor sourcss Indoor sourc©s Indoor sources do not
strongly contribute, strongly contribute, strongly contribute,
ratio greater than 1 ratio less thanl ratio less thanl
Figure 17. Diagram illustrating indoor-outdoor ratios with different fractional indoor source
contributions.
As full text screening identified relevant and usable studies, key study attributes for those
studies were extracted into an Excel spreadsheet. Attributes were selected based on the
knowledge of what might affect the indoor-outdoor ratio. These included:
• Chemical/Particle type
o Particulate matter
¦ Particulate matter only
¦ Organics on particulate matter
¦ Metals on particulate matter
o Vapor Phase
¦ Free organics
• Semi volatile organic compounds (SVOCs)
• Volatile organic compounds (VOCs)
¦ Free Metals
• Chemical Name, with CAS number and vapor pressure
• Particle size range (for particulate only)
• Microenvironment type
• Location
• Season
• Number of sites/measurements
• Indoor-outdoor ratio minimum
• Indoor-outdoor ratio maximum
• Indoor-outdoor ratio mean or median
• Indication of whether ratio was calculated by screener
63
-------
In cases where paired concentrations (indoor and outdoor) at the same location were provided
but the actual indoor-ratio was not calculated in the paper, these ratios were calculated and
flagged as "ratio calculated by screener".
10.3 Data Analysis
Next the data were examined by the different attributes collected to determine trends by
chemical/particle type, microenvironment, and season. Figure 18 below shows the graph of
central tendency (mean or median depending on the individual study) indoor-outdoor ratio by
vapor pressure, by microenvironment, and by season. In general, the figures do not indicate a
trend by any of these variables; instead, the values are spread between zero and one
somewhat uniformly across the different variable values. This suggests the variation in
individual building parameters and local conditions may affect the indoor-outdoor ratio more
strongly than the chemical properties, microenvironment type, or season.
Vapor Pressure
1.00- #
0.25-
1000 1500
Vapor pressure (Pa)
Organics Only
• vocs
~ SVOCs
64
-------
Microenvironment
Aerosol Type
• PM Only
• Metals
• VOCs
• SVOCs
School/Office
Microenvironment
Multiple/Other
Season
Aerosol Type
• PM Only
• Metals
• VOCs
• SVOCs
Cooling
Heating
Season
Figure 18. Central tendencies of indoor-outdoor ratios grouped by vapor pressure,
microenvironment, and season.
Because numerical trends were not observed in the individual means, the data were compiled
by finding the mean across usable data for a variety of different chemical types as shown in
Table 23 and Figure 19. Table 23 has a) the mean calculation restricted to studies that provided
65
-------
means and b) the mean calculation including studies that provided either a mean or median. A
comparison between the two sets of columns indicates inclusion of the medians does not
strongly affect the overall mean for each chemical category. Table 23 also stratifies the data in
different ways to determine any overall trend for different particulate matter (PM) types or
different aggregation of organics (e.g., SVOCs versus VOCs, free SVOCs versus SVOCs on
particulate matter, etc.). Aside from SVOCs on PM (which uses only two studies), all the
different categories indicate mean ratios between 0.55 and 0.7 with standard deviations of
approximately 0.2. In addition, the SVOCs on PM differences from the overall mean is not
statistically significant. The consistency of mean values across different chemical types and in a
variety of different study designs suggests this value of 0.65 with a standard deviation of 0.2
(the "overall" value) is representative of an expected value for an indoor-outdoor ratio for a
variety of chemicals/microenvironments/seasons. A corresponding "high" value could be the
central tendency plus two standard deviations, but in this case 0.65 + 0.4 is greater than one;
thus, an upper value of one is used. These final values are shown in Table 24.
Table 23. Summary of means and standard deviations of indoor-outdoor ratios by chemical
type.
Category
Aerosol
No. of central
tendencies
(Unique
studies)
Means only
All central
tendencies
Mean
Std
Dev
Mean
Std
Dev
Overall
367 (80)
0.65
0.22
0.65
0.21
PM
PM (All)3
305 (66)
0.65
0.20
0.66
0.20
PM Only
164 (59)
0.69
0.17
0.69
0.18
Organics (SVOCs) on
PM
3(2)
0.81
0.18
0.81
0.18
Metals on PM
138 (15)
0.61
0.22
0.62
0.22
Organics
Organics (All)b
64 (17)
0.59
0.23
0.61
0.23
Free VOCs and
SVOCs
61 (15)
0.57
0.23
0.60
0.23
Free VOCs
45 (11)
0.58
0.23
0.61
0.24
Free SVOCs
16(4)
0.57
0.21
0.57
0.21
SVOCs on PM
3(2)
0.81
0.18
0.81
0.18
SVOCs (All)c
19(6)
0.60
0.22
0.60
0.22
a : Organics on PM and Free SVOCs
b: Organics on PM, Free SVOCs, and Free VOCs
c: PM Only, Organics on PM, and Metals on PM
66
-------
1
Overall PM (All) PMOnly Organics Metals on Organics Free Free Free SVOCs on SVOCs
(SVOCs) PM (All) VOCsand VOCs SVOCs PM (All)
on PM SVOCs
All PM Organics
Figure 19. Summary of indoor-outdoor ratio means and standard deviations for different
chemical types. Number labels are the number of measurements (number of studies); error
bars are the standard deviations.
Table 24. Final summary of indoor-outdoor ratios.
Central Tendency
High
Indoor/outdoor ratio
0.65
1.00
10.4 Illustrative Example to Calculate Indoor Air
Concentration
Assume IIOAC calculates an outdoor air concentration of 75 |Jg/m3 for a specific receptor. The
corresponding high-end and mean indoor air concentrations are calculated as follows:
uq
r — r — 7^
^indoor,hiqh ~ uoutdoor — ' ^ q
m3
fig
^indoor,mean ~ 0.65 X Cout(ioor — 48.75 —r
m3
67
-------
11 Dose Calculations
Acute and chronic dose rates by age group are calculated for the two receptor categories as
given by Equations 26 and 27 (Versar, Inc., 2007).
a r\ n ^^24 " IfthR " ED ' CF-\
ADRpnT = —22 26
^UI BW-AT
t , T-* ACvr ¦ InhR ¦ ED ¦ CF±
LADDP0T = —^ (27)
^UI BW-AT
where ADRP0T = potential acute dose rate [mg/kg/day]
LADDpot = potential chronic average daily dose [mg/kg/day]
AC24 = weighted daily-averaged air concentration [|ag/m3]
ACyr = weighted annual-averaged average air concentration [|-ig/m3]
InhR = inhalation rate, in m3/day for LADDpot, and in m3/hr for ADRpot
ED = exposure duration, in years for LADDpot, and in days for ADRpot
BW = body weight [kg]
AT = averaging time, in years for LADDpot, and in days for ADRpot
CFX = conversion factor from mgto |ag, and is equal to 10"3 mg/|ag
The weighted daily-averaged and annual-averaged air concentrations are calculated using the
outdoor and indoor air concentrations, and the daily activity patterns that specify the time
spent outdoors and indoors:
AC = . Coutioor + (-^-) ¦ Cindoor (28)
^Lin~t~Lout' ^Lin~t~Lout'
where AC = weighted daily-averaged or annual-averaged air concentration [|-ig/m3]
tout = total time spent outdoors in one day [min]
tin = total time spent indoors in one day [min]
Coutdoor = outdoor air concentration calculated in IIOAC [|-ig/m3]
Cindoor = indoor air concentration calculated using 10 ratio [|-ig/m3]
Parameters for inhalation rates, body weights, exposure durations, and activity patterns by age
group are obtained from the Exposure Factors Handbook (US EPA, 2011) and the E-FAST
documentation manual (Versar Inc., 2007) and are presented in Table 25-Table 28 below.
Table 25. Mean body weights by age group (taken from Table 8-1 in US EPA, 2011).
Age Range
Mean Body Weight (kg)
Birth to 1 month
4.8
1 to <3 months
5.9
3 to <6 months
7.4
6 to <12 months
9.2
68
-------
1 to <2 years
11.4
2 to <3 years
13.8
3 to <6 years
18.6
6 to <11 years
31.8
11 to <16 years
56.8
16 to <21 years
71.6
Adults
80.0
Table 26. Average inhalation rates for light intensity by age group (taken from Tables 6-1 and 6-
2 in US EPA, 2011).
Age Range
Average Inhalation Rate
(m3/day) (chronic)
Average Inhalation Rate
(m3/hr) (acute)
Birth to 1 month
3.6
1 to <3 months
3.5
3 to <6 months
4.1
6 to <12 months
5.4
Birth to <1 year
5.4
0.456
1 to <2 years
8.0
0.72
2 to <3 years
8.9
0.72
3 to <6 years
10.1
0.66
6 to <11 years
12.0
0.66
11 to <16 years
15.2
0.78
16 to <21 years
16.3
0.72
21 to <31 years
15.7
0.72
31 to < 41 years
16.0
0.72
41 to <51 years
16.0
0.78
51 to <61 years
15.7
0.78
61 to <71 years
14.2
0.72
71 to <81 years
12.9
0.72
> 81 years
12.1
0.72
Table 27. Exposure duration for acute and chronic exposures by age group (taken from Table 3-
2 in Versar, Inc. 2007). Exposure duration for adults was modified from 30 to 33 years.
Age range
Exposure Duration
Birth to <1 year
1 day (acute)
1 to <2 years
1 day (acute)
2 to 5 years
1 day (acute)
6 to 12 years
1 day (acute)
13 to 19 years
1 day (acute)
Adult
1 day (acute)
Adult
33 years (chronic)
69
-------
Table 28. Recommended values for activity patterns by age group (taken from Tablel6-1 in US
EPA, 2011).
Age Range
Total Time Indoors (min)
Total Time Outdoors (min)
Birth to <1 month
1,440
0
1 to <3 months
1,432
8
3 to <6 months
1,414
26
6 to <12 months
1,301
139
1 to <2 years
1,353
36
2 to <3 years
1,316
76
3 to <6 years
1,278
107
6 to <11 years
1,244
132
11 to <16 years
1,260
100
16 to <21 years
1,248
102
18 to <65 years
1,159
281
>65 years
1,142
298
IIOAC calculates acute and chronic exposure doses for the three age groups listed below using
an age of 78 years to define the upper bound of adults. If necessary, a weighted body weight
and inhalation rate is calculated for these age groups using the values listed in Table 25 and
Table 26:
• Young toddler (1-<2 years),
• Adult (16-<78 years), and
• Lifetime (0- <78 years) - calculated for chronic exposure doses only.
For acute dose, IIOAC outputs only the exposure doses for the Young toddler and Adult age
groups. For chronic dose, the Young Toddler, Adult, and Lifetime groups are displayed. Only
selected age groups are output because preliminary analysis of IIOAC results indicated the
maximum exposure doses always occurred in the Young toddler group. Exposure doses for the
additional age groups listed below can be calculated using equations 26-28:
• Infant (<1 year),
• Young toddler (1-<2 years),
• Toddler (2-<3 years),
• Small child (3-<6 years),
• Child (6-<11 years), and
• Teen (11-<16 years).
70
-------
12 Example Application of 110AC
In addition to calculating air concentration and particle deposition for various scenarios, the
IIOAC tool or its results can be used in various applications. One potential application of IIOAC is
to determine the emission rate for each source type that would result in a specific air
concentration. For example, if the user is interested in an upper limit of 10 ug/m3 for a daily-
averaged air concentration or an upper limit of 0.1 ug/m3 for the annual-averaged air
concentration, IIOAC can be used to determine the corresponding emission rate.
The approach for determining the emission rate depends on the source type. For point and
fugitive sources, AERMOD results were post-processed and placed in lookup tables. As a result,
the user can select a scenario and use the values in the lookup table to calculate the emission
rate corresponding to 10 ug/m3 for a daily-averaged air concentration or 0.1 ug/m3 for the
annual-averaged air concentration as follows:
For area soil/water sources, calculations are performed in the tool itself and therefore, back-
calculation of the emission rate from a target air concentration is not possible. The user needs
to use a trial-and-error approach to determine the emission rate.
As an illustrative example, the emission rate for each source type was calculated below for
Idaho Falls, ID station. To obtain the most conservative value, it is assumed that the release
occurs on one day out of the year (and therefore the selection of cyclical versus consecutive
release days does not affect the results). For point and fugitive sources, the release duration is
1 hr and as such, the emission rate is equal to the adjusted emission rate. Additional
parameters selected are shown in the table below. For area soil/water sources, benzene is used
as the example chemical (vapor pressure = 75 Torr; solubility = 1790 mg/L; Koc = 66.1;
volatilization half-life = 1 hr; molecular weight = 78.1 g/mol). Note that area soil/water sources
only allow for 24 hour release duration.
ER
Coutdoor"1 9 Is
(29)
SFj-Postprocessed AERMOD result
(30)
71
-------
Table 29. Emission rates corresponding to target maximum air concentrations.
Emission Rate
Source Type
Parameters
Maximum daily-
averaged air
concentration of
10 ug/m3
Maximum annual-
averaged air
concentration of
0.1 ug/m3
Point
Stack, Urban, Vapor, Consecutive,
1 hr duration, 1 day/yr, Idaho Falls
71 kg/day
257 kg/day
Fugitive
Urban, Vapor, 100 m2,
Consecutive, 1 hr duration, 1
day/yr, Idaho Falls
22 kg/day
82 kg/day
Area Soil
Urban, Vapor, 10 acres, 1 day/yr,
Idaho Falls
0.05 mg/day
0.25 mg/day
Area Water
Urban, Vapor, 10 acres, 1 day/yr,
Idaho Falls, 100 m3/day
800 kg/day
3000 kg/day
13 Remaining Uncertainties and Potential Future Updates
In the development of IIOAC, several assumptions were made, leading to uncertainties in the
outputs of IIOAC. Potential future updates can address the following issues:
• Use of generic parameters rather than facility specific parameters.
• Use of regional meteorological and land cover data as representative of site-specific
meteorological and land cover.
• No chemical-specific properties were accounted for in the point and fugitive source
outputs, thus atmospheric chemistry and degradation were not factored in and vapor
deposition was not calculated as this requires chemical-specific properties.
• Assumption that the regression relationships (relating source size and AERMOD outputs)
derived using Iowa City, IA meteorology data would be roughly applicable to other
meteorological conditions from the other 13 meteorology stations.
14 References
European Chemicals Agency. (2016) Guidance on Information Requirements and Chemical
Safety Assessment, Chapter R.16: Environmental exposure assessment.
Multi-resolution Land Characteristics Consortium. (2001). National Land Cover Dataset 1992.
https://www.mrlc.gov/nlcdl992.php.
National Aeronautics and Space Administration. (2017). MODIS Vegetation Index Products
(NDVI and EVI). https://modis.gsfc.nasa.gov/data/dataprod/modl3.php.
72
-------
National Oceanic and Atmospheric Administration. (2012). 1981-2010 30-year Normals of
Precipitation, ftp://ftp.ncdc.noaa.gov/pub/data/normals/1981-2010/products/precipitation/.
National Oceanic and Atmospheric Administration. (2017a). Automated Surface Observing
System one-minute wind data, ftp://ftp.ncdc.noaa.gov/pub/data/asos-onemin/.
National Oceanic and Atmospheric Administration. (2017b). Integrated Surface Database.
https://www.ncdc.noaa.gov/isd.
National Oceanic and Atmospheric Administration. (2017c). Radiosonde Database.
https://ruc.noaa.gov/raobs/.
The University of North Carolina at Chapel Hill, Institute for the Environment. (2016). Sparse
Matrix Operator Kernel Emissions (SMOKE) User's Manual, Version 4.0. 9/30/2016.
https://www.cmascenter.Org/smoke/documentation/4.0/manual smokev40.pdf.
US Environmental Protection Agency. (1995a). SCREEN3 Model User's Guide.
https://www3.epa.gov/ttn/scram/userg/screen/screen3d.pdf.
US Environmental Protection Agency. (1995b). User's Guide for the Industrial Source Complex
(ISC3) Dispersion Models, https://www3.epa.gov/scram001/userg/regmod/isc3v2.pdf.
US Environmental Protection Agency. (2011). Exposure Factors Handbook.
US Environmental Protection Agency. (2014a) Appendix A: Volatilization Screening Tool
Guidance Document Draft dated 3/1/2014.
US Environmental Protection Agency. (2014b). Approach for Estimating Exposures and
Incremental Health Effects from Lead Due to Renovation Repair and Painting Activities in Public
and Commercial Buildings. August, 2014. https://www.regulations.gov/document?D=EPA-HQ-
OPPT-2010-0173-0234.
US Environmental Protection Agency. (2015a). 2011 NATA Emissions Facility Release data.
11/23/2015. https://www.epa.gov/national-air-toxics-assessment/2011-nata-assessment-
results.
US Environmental Protection Agency. (2015b). Guidance for Using the Volatilization Algorithm
in the Pesticide in Water Calculator and Water Exposure Models.
https://www.epa.gov/sites/production/files/2016-
03/documents/volatilization algorithm guidance.pdf.
US Environmental Protection Agency. (2015c). Technical Support Document, EPA's 2011
National-scale Air Toxics Assessment. 12/2015.
https://www.epa.gov/sites/production/files/2015-12/documents/2011-nata-tsd.pdf.
73
-------
US Environmental Protection Agency. (2016a). AERSCREEN User's Guide.
https://www3.epa.gov/ttn/scram/models/screen/aerscreen userguide.pdf.
US Environmental Protection Agency. (2016b). NAAQS Table, https://www.epa.gov/criteria-air-
pollutants/naaqs-table
US Environmental Protection Agency. (2017a) EPA's Risk-Screening Environmental Indicators
(RSEI) Methodology, https://www.epa.gov/sites/production/files/2017-
01/documents/rsei methodology v2.3.5 O.pdf.
US Environmental Protection Agency. (2017b). Estimation Programs Interface Suite.
https://www.epa.gov/tsca-screening-tools/epi-suitetm-estimation-program-interface.
US Environmental Protection Agency. (2017c). Human Exposure Model (HEM).
https://www.epa.gov/fera/download-human-exposure-model-hem.
US Environmental Protection Agency. (2017d) Revisions to the Guideline on Air Quality Models:
Enhancements to the AERMOD Dispersion Modeling System and Incorporation of Approaches
To Address Ozone and Fine Particulate Matter.
https://www3.epa.gov/ttn/scram/appendix w/2016/AppendixW 2017.pdf
Thomas, R.G. (1990). Volatilization from Water. In: Handbook of Chemical Property Estimation
Methods. Lyman, W.J. et al. (eds), Washington, DC: American Chemical Society, Chapter 15.
Varghese, A., Cawley, M., and Hong, T. (2017). Supervised Clustering for Automated Document
Classification and Prioritization: A Case Study Using Toxicological Abstracts. Environment
Systems and Decisions. https://doi.org/10.1007/slQ669-017-9670-5.
Versar, Inc. (2007). Exposure and Fate Assessment Screening Tool (E-FAST) Version 2.0
Documentation Manual, https://www.epa.gov/sites/production/files/2015-
04/documents/efa st2man.pdf.
Wood row, J.E. and Sieber, J.N. (1997). Correlation Techniques for Estimating Pesticide
Volatilization Flux and Downwind Concentrations. Environmental Science & Technology 31, 523-
529.
74
-------
Appendix A Regression Coefficients for Air Concentration
versus Area Size
Regression coefficients for fugitive and area sources are provided below where the regression
equation has the form:
C = a ¦ Ab
where C = air concentration [|Jg/m3]
A = area size [m2]
a = regression coefficient [-]
b = regression coefficient [-]
Table Al. Regression coefficients for fugitive and area sources.
Inner Ring
Community
Fugitive, Coarse, Rural
a =
154.47
9.12
b =
-0.05298
-0.0205
Fugitive, Coarse, Urban
a =
109.95
5.29
b =
-0.06118
-0.01674
Fugitive, Fine, Rural
a =
74.73
6.77
b =
-0.04418
-0.01582
Fugitive, Fine, Urban
a =
43.97
2.58
b =
-0.05751
-0.01669
Fugitive, Vapor, Rural
a =
142.91
13.21
b =
-0.04507
-0.01585
Fugitive, Vapor, Urban
a =
83.76
4.95
b =
-0.05675
-0.01659
Area
a =
7.18
0.0716
b =
0.2092
0.3923
75
-------
Appendix B Comparison of AERMOD Results for Selected
Point Sources
The three options for point source offered in IIOAC (the stack, incinerator 1, and incinerator 2
point sources) were created so that together they would result in a wide range of air
concentrations. As shown below, the stack point source should generally result in the highest
air-concentration and deposition values relative to the other point sources, due to its lower
height and lower plume rise (due to lower buoyancy and momentum). The two incinerator
point sources should generally result in much smaller air-concentration and deposition values
relative to the stack source, due to their higher heights and plume rise values—the incinerator
2 in particular should often show reductions in air-concentration and deposition values of more
than 90 percent relative to the stack source, at receptor locations both close to and farther
away from the emission source. Comparisons of air concentrations were made at the inner ring
or fenceline receptors and an outer ring, set to 1000 m from the source.
Table Bl. Comparison of air concentration for the three point sources at the inner and outer
ring from a test run.
Stack
Incinerator 1
Incinerator 2
Height (m)
10
25
50
Temperature (K)
300
500
1200
Diameter (m)
2
1
2
Velocity (m/s)
5
15
15
Example Run - Annual Average Unit Concentration
at Inner Ring (0,100 meters) (|ag/m3)
Value
5.830E+00
5.150E-02
1.839E-03
Reduction vs. Stack
--
99.12%
99.97%
Example Run - Annual Average Unit Concentration
at Outer Ring (0,1000 meters) (|ag/m3)
Value
4.240E-01
1.250E-01
3.367E-02
Reduction vs. Stack
--
70.52%
92.06%
76
-------
Appendix C Illustrative Example for Facility Sources
Assume the user inputs three point source releases that occur at the same site, selects fine
particles, and enters release data according to the following:
Table CI. User inputs for example releases for a point source.
User Input
kg/day/site
# days/year
# hours/day
Release #1
100
73
24
Release #2
75
52
4
Release #3
25
365
1
Using Equation 1, the adjusted emission rate for the three releases are calculated as follows:
100 g
ERadjil=—-0.2778 = 1.157^
ERadj,2 = 1 0.2778 = 5.208^-
2 ej
Oady,3=y 0.2778 = 6.944^
Based on the number of days per year and the hours of release per day, IIOAC looks up the
corresponding AERMOD post-processed result and outputs the mean and high-end daily-
averaged and annual-averaged hourly concentrations. In Table C2 below, hypothetical data
have been used to fill in the lookup table.
Table C2. Example AERMOD post-processed air concentration results corresponding to the
number of release days and release duration for releases 1, 2, and 3.
Air Concentration
Release
Duration
(hrs/day)
Number of Release Days
(Hg/m3)
1
2
51
52 ...
72
73 ...
365
1
1.1
Mean Daily
4
0.8
Average
8
24
0.7
1
6.1
High-End Daily
4
1.4
Average
8
24
1.2
Mean Annual
1
0.6
Average
4
0.3
77
-------
8
24
1
4
8
24
0.4
High-End Annual
Average
1.0
0.6
1.1
For point source releases 1, 2, and 3, the high-end daily-averaged air concentration is calculated
as:
1.2 fig
High-end daily-averaged air concER1 = 1.157 ¦ — = 1.39 —r
1 ms
1.4 fig
High-end daily-averaged air concER2 = 5.208 ¦ — = 7.29 —r
1 ms
6.1 fig
High-end daily-averaged air concER3 = 6.944 ¦ — = 42.4 —-
1 ms
For fine particles, an upper limit of 35 |J,g/m3 is applied to each of the individual point source
releases. In the example above, only the third point source release results in an air
concentration greater than the upper limit. For the three releases, the high-end daily-averaged
air concentration is then equal to the sum of the high-end daily-averaged air concentrations for
each of the individual releases, or 43.7 |J,g/m3
78
-------
Appendix D Illustrative Example for Area Soil Sources
Assume the user inputs three releases and the physicochemical properties of a chemical
according to the following table:
Table Dl. User inputs for example releases for an area soil source.
User Input
Tool Conversion
kg/day/site
# days/year
Release occurs
Release #1
100
73
Every 5 days
Release #2
75
52
Every 7 days
Release #3
25
365
Every day
Chemical =
Aldicarb
MW =
190.26 g/mol
VP =
0.01 Pa
Sol =
6030 mg/L
II
u
O
21 mL/g
Area =
20 ha
which gives SF, = 1.01
200000 m2
Using equations 3-10, the mass flux and air concentration for day 1 and 2 at the inner ring are
calculated as follows, with results from days 1-10 shown in the table below:
• Day 1:
Total mass released on day 1 = TE= 100 + 75 + 25 = 200 kg
Mass at start of day 1 = Mi = M0 + (TERi — J0 ¦ A) ¦ (1 day)
= 0 + (200 + 0) ¦ (1 day) = 200 kg
Mx 200 kg kg
Mass per area for day 1 = AR-, = —- = ——— = 10-—
H 1 y 1 A 20 ha ha
Mass flux in day 1 = J1
expf0.8688 • In (l'P.' ^ ) + 21.535 )
= ^ (8.64 x 10-=)
3600
_ exp (0.8688 • In + 21.535)
kg
= 2.70 x 10~4 , * ,—
m/ x day
79
-------
Coutdoor for day 1
A ¦ 200,000
= ¦ 0.01157 ¦ 1.01 ¦ Postprocessed AERMOD result1
If the post-processed AERMOD result is 5.43 |J,g/m3, this gives Coutdoor = 3.41 |j,g/m3.
For Aldicarb, the saturation air concentration is calculated using equation 6 to be 0.766
|j,g/m3. As such, Coutdoor is set to 0.766 |j,g/m3.
The mass flux that corresponds to an air concentration of 0.766 |J,g/m3 is calculated as:
Jlsat" 200,000
0.766 = 5.43 ¦ ' , ¦ 0.01157 ¦ 1.01
Jl.sat — 6.04 X 10
-5 *9
m2xday
• Day 2:
Total mass released on day 2 = TER2 = 25 kg
Mass at start of day 2 = M2 = + (TER2 — J1 ¦ A) ¦ (1 day)
= 200 + [25 - (6.04 x 10-5) ¦ 200000] ¦ (1 day)
= 213 kg
M2 171 kg kg
Mass per area for day 2 = AR? = —- = ——— = 10.6 -—
H 1 y 2 a 20 ha ha
Mass flux in day 2 = J2
exp (0.8688- In + 21.535)
3600
exp (0.8688 ¦ In (gQgQ 12i) + 21.535)
= 2.84 x 10-4
3600
kg
(8.64 x 10-5)
¦ (8.64 x 10-5)
m2 x day
Coutdoor f or day 2
J2 ¦ 200,000
= 0.01157 ¦ 1.01 ¦ Postprocessed AERMOD result2
The calculated air concentration is then compared to the saturation air
concentration, similar to the previous day.
• Days 1-10:
80
-------
Table D2. Summary of mass flux and air concentrations due to emissions from soil at a hypothetical site with three releases.
Total Mass Added
Mass on
Mass
AERMOD
Release
Release
Release
to Soil from all
Soil at Start
Mass per
Mass Flux,
Volatilized
Unit
Day
#1
#2
#3
Releases, TRadd
of Day, M
Area, AR
J
from Soil, MVOi
Value3
Air Cone
(kg/day)
(kg/day)
(kg/day)
(kg/day)
(kg)
(kg/ha)
(kg/m2/day)
(kg/ha)
(ug/m3)
(ug/m3)
1
100
75
25
200.00
200.00
10.00
2.70E-04
5.43
3.41b
2
0
0
25
25.00
212.88
10.64
2.85E-04
19.1
12.7b
3
0
0
25
25.00
234.45
11.72
3.10E-04
34.6
24.9b
4
0
0
25
25.00
257.55
12.88
3.36E-04
16.5
12.9b
5
0
0
25
25.00
278.55
13.93
3.60E-04
12.5
10.4b
6
100
0
25
125.00
398.27
19.91
4.91E-04
30.9
35.3b
7
0
0
25
25.00
421.14
21.06
5.15E-04
29.1
34.9b
8
0
75
25
100.00
518.88
25.94
6.17E-04
16.9
24.3b
9
0
0
25
25.00
540.00
27.00
6.39E-04
3.96
5.90b
10
0
0
25
25.00
548.40
27.42
6.48E-04
6.79
10.3b
continue until M or J falls below a threshold level of 10"7
a AERMOD Unit Value based on 1 g/s of emission
b Indicates calculated air concentration exceeds saturation air concentration
81
-------
Appendix E Illustrative Example for Area Water Sources
Assume the user inputs three releases with the following profiles. The user obtains the
volatilization half-life from EPI Suite, which requires that the user enter a water depth, water
velocity, and wind velocity.
Table El. User inputs for example releases for an area water source. Note the value for water
depth was selected for illustrative purposes and is not the default value (1 m for both river and
lake) used in EPI Suite to estimate volatilization half-life.
User Input
Tool Conversion
kg/day/site
# days/year
Release occurs
Release #1
100
73
Every 5 days
Release #2
75
52
Every 7 days
Release #3
25
365
Every day
Chemical =
Naphthalene
MW =
128.2 g/mol
VP =
1140 Pa
Depth of water =
5 m
Water velocity =
1 m/s
Wind velocity =
1 m/s
tl/2 =
110 hrs
Surface area =
50,000 m2 which gives SFj = 2.24
Flowrate =
50 m3/day
Using equations 11-18, the mass flux and air concentration for day 1 and 2 are calculated as
follows, with results from days 1-20 shown in the table below:
• Day 1:
7. _ ln(2) _ ln(2) _ n r 1
1 ^ I V/ ¦ 1 O
^ (110hrS)"^ ^
v = A ¦ d = 50,000 ¦ 5 = 250,000 m3
Total mass released on day 1 = TE= 100 + 75 + 25 = 200 kg
Mass at start of day 1 = M1 = M0 + |'tER1 — (kvoi + ^ ¦ M0^ ¦ 1
/ 50 \
= 0 + 200 - 0.15 + ¦ 0 = 200 kg
V 250,000/ 3
Mass volatilized on day 1 = Ml voi = kvoi ¦ M1- (1 day) = 0.15 ¦ 200
82
-------
= 30 kg
Mvoi i/day
Cgutdoori = ~n 0.01157 ¦ SFj ¦ Postprocessed AERMOD result1
1 —
s
30
= — ¦ 0.01157 ¦ 2.24 ¦ Postprocessed AERMOD result1
If the AERMOD Unit Value is 3.42 |J,g/m3, this gives Coutdoor = 2.66 |J,g/m3, which is
below the saturation air concentration calculated using Equation 6.
• Day 2:
Total mass released on day 2 = TER2 = 25 kg
Mass at start of day 2 = M2 = + (tER2 — (kvoi + ¦ M^j ¦ 1
/ 50 \
= 200 + 25 - 0.15 + ¦ 200
V 250,000/
= 195 kg
Mass volatilized on day 2 = M2iVOi = kvoi ¦ M2 ¦ (1 day)
= 0.15- 195 = 29.2 kg
Mvoi 2,1day
Cgutdoor 2 = ~n 0.01157 ¦ SFj ¦ Postprocessed AERMOD result2
1y
s
29.2
= —— ¦ 0.01157 ¦ 2.24 ¦ Postprocessed AERMOD result2
The calculated air concentration is then compared to the saturation air
concentration, similar to the previous day.
• Days 1-10:
83
-------
Table E2. Summary of mass volatilized and air concentrations due to emissions from water at a hypothetical site with three releases.
Total Mass Added to
Mass in Water
Mass Volatilized
Release
Release
Release
Water from All
at Start of Day,
from Water,
AERMOD
Day
#1
#2
#3
Releases, TRadd
M
Mvol
Unit Value3
Air Cone
(kg/day)
(kg/day)
(kg/day)
(kg/day)
(kg/m3)
(kg)
(ug/m3)
(ug/m3)
1
100
75
25
200
200
30.1
3.42
2.67
2
0
0
25
25
194.6
29.4
33.3
25.3
3
0
0
25
25
190.5
28.7
63.8
47.4
4
0
0
25
25
187.4
28.1
37.8
27.6
5
0
0
25
25
184.3
27.7
36.0
25.8
6
100
0
25
125
281.6
42.3
11.5
12.6
7
0
0
25
25
264.3
39.7
16.5
16.9
8
0
75
25
100
324.6
48.8
19.7
24.8
9
0
0
25
25
300.8
45.2
34.2
40.0
10
0
0
25
25
280.6
42.1
7.00
7.63
continue until M or mass fluxb falls below a threshold level of 10"7
a AERMOD Unit Value based on 1 g/s of emission
b For threshold level, a mass flux was calculated for each day
84
------- |