Guidance Page Menu:

• Submittal Requirements for Modeling Analyses for Permit Applications

• AERMOD

• Regulatory Defaults

• Urban Dispersion Option  

• Building Downwash

• Averaging Times

• Receptor Grid

• Flagpole Receptor Heights

• Elevations

• Conversion of NOx to NO2  

• Shoreline Fumigation and Inversion Break-Up  

• Meteorological Data  

• Surface Characteristics  

• Background Concentrations  

• Variable Emissions   

• Resources 

Submittal Requirements for Modeling Analyses for Permit Applications

[Return to Top]

AERMOD

As of December 9, 2006, the United States Environmental Protection Agency promulgated AERMOD as a replacement for ISCST3 as the recommended air quality dispersion model. All modeling analyses should be conducted using the most recent version of AERMOD available at the time of submittal, unless approval is received prior to submittal to use a previous version of AERMOD. Please see the U.S. EPA’s SCRAM webpage for the latest version of AERMOD and the associated pre-processors. South Coast AQMD’s required modeling options for AERMOD are discussed below.

Regulatory Defaults

AERMOD should be executed with the U.S. EPA regulatory default option, which is required by South Coast AQMD. If the default option is not utilized, the report should contain a discussion to justify this change and include all supporting data and information.

[Return to Top]

Urban Dispersion Option

For most projects and sources within South Coast AQMD’s jurisdiction, AERMOD should be executed using the urban modeling option due to the high degree of land use development in most areas. Projects and sources in urban areas should be modeled with urban effects using the population of the County where the project is located. Table A below lists the various County populations within South Coast AQMD’s jurisdiction. For projects and sources in rural areas or if the user is not clear on which option to use, please refer to the procedure outlined by U.S. EPA in Section 7.2.1.1.(b) of 40 CFR Part 51, Appendix W  (October 2023).

Table A: County Population to Use in AERMOD

Note: Population data based on County Population Estimates of July 1, 2023 data from the U.S. Census Bureau.

Building Downwash

Building downwash effects must be included and should be modeled using BPIP-PRIME (BPIPPRM).

Averaging Times

When using AERMOD to determine the maximum annual concentration of criteria pollutants such as NO2, PM10, PM2.5, and SO2, South Coast AQMD requires that each calendar year is run separately. Some software GUI’s have the functionality to output maximum annual concentrations by individual year through the use of POST files. 

1-hour NO2, 24-hour PM10, 24-hour PM2.5, and 1-hour SO2 National Ambient Air Quality Standards (NAAQS) have special processing requirements and specific keywords need to be used in AERMOD to generate the correct concentrations. Please refer to the AERMOD User’s Guide for the correct keywords to use in AERMOD. 

When using AERMOD to determine concentrations to calculate cancer risks or chronic health indexes, it is appropriate to estimate the maximum annual concentration for the entire 5-years of meteorological data provided. Please use the averaging time keyword PERIOD for calculating long-term health risks.

[Return to Top]

Receptor Grid

To identify the maximum impacted receptors, a uniform Cartesian grid with a spacing of 100 meters or less must be used for all distances less than 1 kilometer. All receptors should be identified in UTM coordinates. The size of the modeling domain and placement of the receptors should be at such a distance and resolution to identify the maximum impacts to the surrounding areas and the significant concentration gradients from the source under review. Table B below can be used as an example for receptor spacing within the modeling domain. Note that some modeling analyses will require remodeling with a finer grid resolution if additional areas of concern are identified in the coarse grid modeling.

Table B: Example of Nested Grid Receptor Spacing*

*This table is an example of a nested table grid typically used in a SIL or NAAQS/CAAQS analysis.

Discrete Cartesian receptors on the ambient air boundary (which can be considered the property or fenceline boundary) must be placed along the boundary following the maximum spacing requirements shown in Table C below. Receptors inside the facility’s ambient air boundary should be removed.  Sensitive receptors must be identified by exact UTM coordinates using discrete Cartesian receptors. Elevations must be provided for all receptors within the modeling domain.

Table C: Maximum Receptor Spacing Requirements for Ambient Air Boundary Receptors

WARNING:   According to the section 4.1 of the AERMOD Implementation Guide (revised August 3, 2015), for cases in which receptor elevations are lower than the base elevation of the source, AERMOD will predict concentrations that are less than what would be estimated from an otherwise identical flat terrain situation. While this is appropriate and realistic in most cases, for cases of down-sloping terrain where the plume is terrain-following, AERMOD will tend to underestimate concentrations when terrain effects are taken into account. In order to avoid underestimating concentrations in such situations, South Coast AQMD recommends the following:

1. If all receptor elevations are lower than the base elevation of the source, the non-default option within AERMOD should be applied to assume flat, level terrain.

2. If some receptors are lower and some receptors are higher than the base elevation of the source, AERMOD should be run twice – once using the default option and the second time using the non-default option. The maximum ground-level concentration from both runs should be reported.

[Return to Top]

Flagpole Receptor Heights

All receptors should be set to the elevation, so that only ground-level concentrations are analyzed. Please use the Control pathway keyword ELEV within AERMOD. Flagpole receptors are only necessary for analyses that have instances where sensitive receptors are located on patios/decks at nearby high-rise apartment buildings.

Elevations

The AERMOD modeling system includes AERMAP, which is a terrain data pre-processor. Terrain data, available from the United States Geological Survey (USGS), is used by AERMAP to produce terrain base elevations for each receptor and source and a hill height scale value for each receptor. AERMAP must be used to develop the terrain data required for AERMOD. It is highly recommended that National Elevation Dataset (NED) data in GeoTIFF format be used as input into AERMAP instead of Digital Elevation Model (DEM) data, as is recommended in the U.S. EPA’s AERMOD Implementation Guide due to the DEM data set being static and not updated by USGS for a number of years.  A resolution of 1 arc-second (approx. 30 meters) or 1/3 arc-second (approx. 10 meters) are acceptable. Elevations should be imported for all modeled sources, receptors, and buildings in the modeling analysis.

[Return to Top]

Conversion of NOx to NO2

Methodologies for modeling the conversion of NOx to NO2 should follow the Tiers laid out in U.S. EPA’s Appendix W, Section 4.2.3.4. They are as follows:

• Tier 1: Assume 100% conversion of NOx to NO2, no additional steps or data needed.

• Tier 2: ARM2 ratio. The minimum and maximum NO2/NOx ratios should use U.S. EPA default values (0.5 and 0.9, respectively). Note that the previous ARM methodology (0.80 for 1-hour and 0.75 for annual averaging times) will no longer be accepted as of October 1, 2017.

• Tier 3: OLM and PVMRM. The U.S EPA default values for the NO2/NOx In-Stack Ratio (ISR) and NO2/NOx Equilibrium Ratio (0.5 and 0.9, respectively) should be used. Non-default ISR values can be used if technical justification is provided, such as stack test results, manufacturer test data, or U.S. EPA’s ISR database. As discussed in U.S. EPA’s Appendix W, PVMRM is most appropriate for analyses with relatively isolated and elevated sources. OLM is more appropriate for analyses with area sources, near-surface releases, or where plume overlap from multiple sources will occur. When using OLM, use the option OLMGROUP ALL in AERMOD.

Site- specific hourly ozone data files, concurrent with the site's meteorological data files are available for download through the "Hourly ozone data" layer in the interactive map. Please contact the senior meteorologist for questions regarding the most recent and updated methodologies for NO2 modeling, including which options are most appropriate for your source. 

Shoreline Fumigation and Inversion Break-Up

Sources with tall stacks located on the Pacific Coast shoreline and subject to the PSD program (South Coast AQMD Reg. XVII) should evaluate impacts resulting from shoreline fumigation and inversion break-up. Currently, AERMOD cannot be used to evaluate these effects.  AERSCREEN must be used to evaluate shoreline fumigation and inversion break-up for the short-term averaging periods.

[Return to Top]

Meteorological Data

Meteorological data files for AERMOD have been processed by South Coast AQMD staff and are available at https://www.aqmd.gov/assets/aermet/AERMET_files_And_HRA_Tool.html . Modeling for criteria pollutants and HRA’s in AERMOD should use the most recently available and meteorologically-appropriate 5-year data set, as is required by U.S. EPA’s Appendix W, Section 8.4. Considerations for choosing a meteorological station includes the source’s meteorological conditions (such as prevailing winds, mixing heights, etc.), terrain, surrounding land use and surface characteristics, and proximity. This means that the closest meteorological station to the source under review is not always the most representative meteorologically. Modeling protocols and reports should include the technical justification on the most meteorologically-appropriate station for your source. Technical justification includes comparisons of the surrounding land use, surface characteristics (such as surface roughness), wind roses, significant terrain features, etc.

Surface Characteristics

When using AERMET to process meteorological data for AERMOD, values for three surface characteristics (surface roughness, albedo, and Bowen ratio) must be determined. When using the South Coast AQMD provided AERMOD-ready meteorological data, this information is not necessary as the surface characteristics have already been incorporated into the AERMET processed meteorological data. However, if AERMET will be used to develop a project-specific meteorological data set for use in AERMOD, then AERSURFACE should be used to determine the surface characteristics to be used in AERMET. Please consult South Coast AQMD modeling staff before processing any project-specific meteorological data.

[Return to Top]

Background Concentration

When conducting air dispersion modeling for demonstrating compliance against the National Ambient Air Quality Standards (NAAQS) and/or California Ambient Air Quality Standards (CAAQS), background concentrations are required to be included in the analysis for all attainment pollutants. Background concentrations, as described in the U.S. EPA’s Appendix W, Section 8.3.1(ii), refers to the “…portion of the background attributable to natural sources, other unidentified sources in the vicinity of the project, and regional transport contributions from more distant sources” where “…the ambient concentrations from these sources are typically accounted for through use of ambient monitoring data…”. Monitoring data for the South Coast Air Basin and Coachella Valley can be obtained via the “Background Concs” layer in the interactive map (identical to South Coast AQMD’s Historical Data by Year), CARB’s Air Quality Data, or U.S. EPA’s Air Data webpages. When selecting a monitoring station to represent the background concentrations for the source under consideration, please refer to section 8.3.2 of the U.S. EPA’s Appendix W for guidance. Generally, the closest upwind monitor should be selected, with preference to the monitor that has the most similar characteristics to the area for the source under consideration.

The exceptions for including background concentrations are the PM10 CAAQS and both the PM2.5 NAAQS and CAAQS, due to the nonattainment status. Modeled maximum 24-hour and annual concentrations should be compared directly to the significant change thresholds of 2.5 and 1.0 micrograms/cubic meter, respectively.

The most recent 3 years of available monitoring data should be used to develop background concentration values. For all CAAQS and most NAAQS pollutants, the maximum value from the most recent 3 years should be used. The exceptions to this are for the NAAQS pollutants with special design values. This includes 1-hour NO2 (3 year average of the 98th percentile of the daily maximum 1-hour average) and 1-hour SO2 (3 year average of the 99th percentile of the daily maximum 1-hour average). The "Background Concs" layer in the interactive map also provides these design values. 

[Return to Top]

Variable Emissions

All applicants conducting dispersion modeling and HRAs for permitting purposes should model the sources under consideration as a continuous operation (24 hours/day, 7 days/week, and 52 weeks/year), unless there are existing permit conditions that restrict the operating hours or the applicant is willing to take conditions in their permit that restrict the hours of operation.

For Rule 1401 compliance demonstrations, non-continuous operations should be accounted for through the use of variable emission factors (AERMOD keyword EMISFACT) or emission scalars, as they are sometimes also referred to. See section 3.3.11 of the AERMOD User’s Guide for more information on variable emission factors.  Emission scalars should be applied as specified in the Variable Emissions section of the HARP2 User’s Guide. This will require two runs within AERMOD.  The model run for the long-term health risks (chronic and cancer) should be run with adjusted emission scalars (to sum the annual emissions up to 31,536 kg/yr). The emission rate (Qs) specified for the source should be 1 g/s. However, the emission scalar will take into account the adjustment. For example, a facility operating 16 hours/day, 7 days/week, 52 weeks/year would specify an emission factor of 1.5 (EMISFACT HROFDY 1.5) for hours when the source is operating, and an emission factor of 0 (EMISFACT HROFDY 0) for hours when the source is not operating. The model run for the short-term health risks (acute) should be run with a value of 1 g/s when the source is operating, and 0 g/s when the source is not operating.

[Return to Top]

Resources

All applicants conducting air quality dispersion modeling and HRAs for permit applications should familiarize themselves with the following resources:

• U.S. EPA's Support Center for Regulatory Atmospheric Modeling (SCRAM) website

• 40 CFR Part 51, Appendix W: Guidelines on Air Quality Models (commonly referred to as U.S. EPA's Appendix W)

• AERMOD Modeling System, including the AERMOD User's Guide and AERMOD Implementation Guide

• AERSCREEN

• Office of the Environmental Health Hazard Assessment (OEHHA)

• OEHHA's Air Toxics Hot Spots Program Guidance Manual for the Preparation of Risk Assessments

• California Air Resources Board (CARB)

• HARP2

• South Coast AQMD Rules 1303, 1401, 1703, and 2005

• South Coast AQMD Risk Assessment Procedures for Rules 1401 and 212

• National Ambient Air Quality Standards (NAAQS)

• California Ambient Air Quality Standards (CAAQS)

[Return to Top]