Scope Study for Expanding the Great Lakes Toxic Emission Regional Inventory to include Estimated Emissions from Mobile Sources

Chapter 3 Level of Detail in the Estimation of Mobile Source Air Toxic Emissions

• 3-1. Survey of Existing Inventories of Mobile Source Emissions
• 3-2. Level of Detail in the Estimation of Mobile Source Emissions
  • 3-2-1. Source Categorization
  • 3-2-2. Pollutants
  • 3-2-3. Resolution
  • 3-2-4. Update Frequency
  • 3-2-5. Summary
• 3-3. References

In this chapter, the appropriate level of detail required to estimate mobile source toxic air emissions in the Great Lakes Region is determined. First, the current status of mobile source emission inventories was analyzed through a survey of the eight Great Lakes States, Ontario Province, and the Lake Michigan Air Directors Consortium (LADCO). Then, based on survey results, and also considering the potential uses of the inventory, characteristics of mobile source emissions, and resource availability, the appropriate level of detail for each of the following items was determined:

• Source categorization
• Pollutants included
• Spatial resolution
• Temporal resolution
• Update frequency

3-1. Survey of Existing Inventories of Mobile Source Emissions

In February 1996, MPCA staff conducted a survey of the current status of the eight Great Lakes States and Ontario, Canada, with regard to estimating criteria and toxic pollutant emissions from mobile sources. The purpose of this survey was to determine a consistent and feasible approach to the regional toxic air emission inventory for mobile sources.

The survey consists of two parts. Part I contains general questions about mobile source emission inventories. Part II contains specific questions about vehicle miles traveled (VMT) data. The original survey form and cover letter are included in Appendix A-3-1. The following are noteworthy summary results:

1. Compiling mobile source emission inventory data is a continuing effort for the Great Lakes States and Ontario. Four states update the inventory every three years, four states update it as needed, while Ontario prepares its inventory annually.
2. All Great Lakes States have mobile source emission inventories for nonattainment areas. The pollutants included in the inventories are dependent on the classification of the nonattainment area. These inventories represent emissions during a typical day in the pollutant season and may extrapolate from daily emissions to annual emissions. Three states also have statewide emission inventories. The nonattainment designation is not applicable in Canada. Ontario has a province-wide emission inventory which represents annual emissions.
3. Toxic air pollutants have not been included in these emission inventories.
4. For highway vehicles, MOBILE5a is the model used by all Great Lakes States in the preparation of their emission inventories for VOC, NO_x, and CO. Ontario uses MOBILE5c to estimate VOC, NOx, and CO. MOBILE5c, a modified version of MOBILE5a, uses the base emission rate and deterioration rate of Canadian vehicles. New York uses the PART5 model to estimate PM10 emissions. Ontario also has emission estimates for SOx and PM based on data provided from Environment Canada.
5. The Great Lakes States follow EPA’s guidance (EPA, 1992) when preparing an emission inventory. The temporal resolution of the inventories is seasonal. The inventories represent emissions during a typical day in the pollutant season, most commonly summer for ozone and winter for CO. Ontario’s inventory is also seasonal. Illinois has data available for hourly emissions for certain areas. New York and Ontario can also derive daily and hourly emissions depending on their needs.
6. The spatial resolution of the inventories is on a county basis for the Great Lakes States. Illinois has link (i.e., road) based emission data available for some areas. Ontario’s inventory is based on variable Universal Transverse Mercator (UTM) grids (5 x 5 km for urban areas, 50 x 50 km for rural areas, and 100 x 100 km for remote areas).
7. The mobile source emission inventories include emissions from onroad, nonroad, and aircraft for the Great Lakes States and Ontario.
8. In general, the Department of Transportation in each state or province is responsible for developing the VMT data, while the state or province environmental agencies prepare the emission inventory for mobile sources. However, the Michigan Department of Transportation also prepares the emission inventory for highway vehicles.
9. VMT data and MOBILE5 input parameters with reasonable default values are available or could be available for each county in the Great Lakes States and Ontario. Link-based VMT data could also be available for all states and the province.

Table 3-1 summarizes the level of detail in the existing mobile source emission inventories for the Great Lakes States and Ontario. A detailed survey report for the eight Great Lakes States and Ontario province is provided in Appendix A-3-2.

The survey form was also sent to LADCO. LADCO has prepared two mobile source emission inventories. One is for the Lake Michigan Ozone Study (LMOS) project. This inventory contains VOC (or hydrocarbon), NOx, and CO emissions on a gridded area basis (4 km grid) for the summer days (June - August) 1990, and on an hourly/temperature specific basis. It covers the entire states of Wisconsin and Michigan and parts of Illinois and Indiana (including most of the metropolitan areas within IL and IN). Although the LMOS is based on the primary arterials and highways, it accounts for over 80% of the VMT in the study area.

The other LADCO emission inventory is prepared for the Ozone Transport Assessment Group (OTAG) which includes EPA, states, and other organizations. The OTAG emission inventory provided VOC (or hydrocarbon), NOx, and CO emissions on a county basis for 38 states in the eastern United States and three provinces in Canada, including the eight Great Lakes States and Ontario province. Most of the VMT data were obtained from the Federal Highway Administration’s (FHWA) Highway Performance Monitoring System (HPMS) database, some states also supplied their own specific data when available. The emissions were estimated for each individual day in July 1990 on an hourly/temperature specific basis. Individual day estimates also are available for July 13 -21, 1991. Both emission inventories contain the emissions from onroad and nonroad vehicles and aircraft, and MOBILE5a was used in both to estimate highway vehicle emissions.

3-2. Level of Detail in the Estimation of Mobile Source Emissions

3-2-1. Source Categorization

EPA has a guidance document for emission inventory preparation (EPA, 1992) which presents an overview of the mobile source category and can be used in the Great Lakes regional toxic air emission inventory for mobile sources. Figure 3-1 shows a flow chart for source category and subcategory. Of the four categories of mobile sources, highway vehicles (i.e., onroad mobile sources) is the most complex. Although nonroad mobile sources, aircraft, and locomotives have been treated as area sources in some emission inventories, the Steering Committee for the Great Lakes Emission Inventory of Toxic Air Contaminants has agreed to consider them as mobile source categories to be consistent with EPA’s guidance (SC, 1996a).

Figure 3-1. Overview of mobile source categorization

3-2-2. Pollutants

In Chapter 2, it was determined that seventy two pollutants emitted from mobile sources are of concern. For these pollutants, we have discussed their chemical and physical properties, human health impact, and environmental transfer and fate. Besides the toxicity and environmental fate information that has been discussed, it is also important to know the possible contribution of mobile sources to the total emissions of each pollutant. Table 3-2 contains estimates of the mass of emissions contributed by mobile sources. This table is expanded from Table 2-3, Secondary pollutant list. Three new columns are added: carcinogenicity, mass of emissions, and a reference for the mass of emissions. The mass of emissions for specific pollutants that are contributed by mobile sources is obtained by the following methods:

• When data are available for the percentage of mobile source contribution in the total emissions of the United States, the percentage is used. The emission year when the emission percentage was estimated should be referred to the reference.
• When numerical data are not available for the mobile source contribution in the total emissions, three levels are defined as shown below:
  • Small: when the reference does not specify emissions from mobile sources or from fuel combustion as an emission source of the pollutant
  • Certain: when the reference specifies emissions from mobile sources or from fuel combustion as an emission source of the pollutant
  • Large: when the reference specifies major emissions from mobile sources or from fuel combustion
• When no reference is available to obtain the emission sources for the pollutant, a blank space is used.

There are numerous ways to categorize or prioritize the pollutants. According to the decision of the Steering Committee (SC, 1996b), if possible, emissions of all pollutants listed in Table 3-2 should be estimated to provide data to various users.

3-2-3. Resolution

The resolution of an emission inventory is determined by the needs of users of the inventory, characteristics of the emission sources, and resource availability. The regional air toxic emission inventory will be used by researchers and other interested parties. As defined in the CAA, emission inventories are key tools in "providing for attainment and maintenance of health protective national ambient air quality standards" (CAA, 1990). The potential uses of the emission inventory have been identified in the Air Toxic Emissions Inventory Protocol for the Great Lakes Commission (Radian, 1994). Table 3-3 shows the potential uses of the mobile source emission inventory. Different users have deferent requirements. Air quality analyses involving use of a photochemical model, such as ozone transport assessment, require hourly emission data of gridded areas (e.g., 4 km grid) for episode days. Other analyses, such as identifying source categories of interest and estimating the relative contribution of source categories, do not need location-specific emissions and hour-specific emissions and require less detailed emission data. The required data resolutions for each potential use are also presented in Table 3-3.

The following sub-sections analyze the advantages and disadvantages of different spatial and temporal resolutions, the characteristics of mobile source emissions, and the availability of resources.

3-2-3-1. Spatial Resolution

The emission inventory can be prepared on a county basis or on a gridded area basis. Each resolution has advantages and disadvantages which are shown in the following table.

Comparison of a county-based inventory versus a gridded area-basis inventory

The current version of RAPIDS is not designed to handle data based on gridded areas, and the hardware systems in most states do not have the capacity to deal with the vast amount of data required for this approach. Although some uses need spatially-refined data, an approximation can be made by allocating county-based data to the gridded areas. Therefore, based on staffing constraints, hardware requirements, and existing RAPIDS features, a county-based emission inventory is a reasonable choice. When resources become available, a motor vehicle source emission estimator RAPIDS client application should be developed to calculate highway vehicle emission estimates on a gridded area-basis by using link-specific travel data.

3-2-3-2. Temporal Resolution

The temporal resolution of the regional air toxics emission inventory for mobile sources can be annual, seasonal, daily, or hourly. More precise temporal resolution providing more detailed information requires that more resources, including personnel and computing hardware, must be dedicated to this task. Since motor vehicle emissions are strongly sensitive to the ambient temperature, the most accurate approach is an hour-by-hour estimation. However, an hourly or even a daily inventory are too detailed for the existing hardware systems in the Great Lakes States. Also, the current RAPIDS client application does not temporally resolve emission data and apportion estimates into hour-by-hour (or daily) values. On the other hand, an annual inventory will not reflect the relationship between vehicle emissions and temperature. A good compromise is a seasonal inventory which has the same temporal resolution as all existing emission inventories for mobile sources.

The four seasons are defined in three-month periods as follows:

• Spring -- March through May
• Summer -- June through August
• Fall -- September through November
• Winter -- December through February

In each season, an average day should be selected. In addition, a hot Summer day and a cold Winter day should also be considered to represent emissions in the extreme scenarios. Therefore, at least six "typical days" should be defined. More "typical days" defined in each season throughout the year improves emission estimates. With each day of the year defined as one of the "typical days", emissions are calculated for each of these "typical days" and annual emissions are obtained by summing each day’s emissions.

In the future, when new emission model client applications are developed in RAPIDS and more powerful hardware systems are equipped in the states and province, the emission estimates should be modified to an hourly basis.

3-2-4. Update Frequency

The update frequency of existing emission inventories for mobile sources varies from annual to "as needed" in the Great Lakes States and Ontario. However, the regional mobile source emission inventory should be consistent with the regional emission inventory for point and area sources. As decided by the Steering Committee (SC, 1996b), the regional emission inventory for point and area sources is to be, at a minimum, updated annually. The mobile source update can be accomplished in several ways. First, the states and province may actually update their corresponding mobile source emission inventory annually, redoing all the work. Second, the states and province may use some growth factors to project the emissions for the year in which emissions are to be updated. Third, the states and province may use the previous year’s inventory if there have been no significant changes in VMT. Since a lot of work is required to prepare a mobile source emission inventory, the suggestion is to prepare an actual emission inventory every three years and use growth factors to project the emissions for the years when emission data are not collected.

3-2-5. Summary

The Great Lakes regional toxic air emission inventory for mobile sources should consider source categories that are consistent with EPA’s guidance (EPA, 1992). Emissions of all seventy two pollutants identified in Table 3-2 should be estimated if possible. A county-based, season-by-season emission inventory is a reasonable choice. When resources become available, a gridded area-based hour-by-hour emission inventory should be developed. The emission inventory should, at a minimum, be updated annually. The states and province may actually update their corresponding emission inventory on the availability of resources. A suggestion is to prepare an actual emission inventory every three years and use growth factors to project the emissions for the years when emission data are not collected.

3-3. References

ATSDR 23, Fact sheet: Antimony, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/tfacts23.html, September 1995.

ATSDR 24, Fact sheet: Barium, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/tfacts24.html, September 1995.

ATSDR 29, Fact sheet: 2-Butanone, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/tfacts29.html, September 1995.

ATSDR 33, Fact sheet: Cobalt, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/tfacts33.html, September, 1995.

ATSDR 58, Fact sheet: Vanadium, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/tfacts58.html, September, 1995.

ATSDR 60, Fact sheet: Zinc, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/tfacts60.html, September, 1995.

ATSDR 8802, Public Health Statement: Xylene, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/ToxProfiles/phs8802.html, March 1989.

ATSDR 8805, Public Health Statement: Benzo(a)pyrene, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/ToxProfiles/phs8805.html, May 1990.

ATSDR 8810, Public Health Statement: Chromium, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/ToxProfiles/phs8810.html, July 1989.

ATSDR 8811, Public Health Statement: Chrysene, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/ToxProfiles/phs8811.html, March 1990.

ATSDR 8812, Public Health Statement: Cyanide, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/ToxProfiles/phs8812.html, December, 1989.

ATSDR 8814, Public Health Statement: 1,4-Dichlorobenzene, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/ToxProfiles/phs8814.html, January 1989.

ATSDR 8819, Public Health Statement: Nickel, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/ToxProfiles/phs8819.html, December, 1988.

ATSDR 8916, Public Health Statement: Mercury, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/ToxProfiles/phs8916.html, December, 1990.

ATSDR 8920, Public Health Statement: Phenol, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/ToxProfiles/phs8920.html, December 1989.

ATSDR 8923, Public Health Statement: Toluene, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/ToxProfiles/phs8923.html, December 1989.

ATSDR 9003, Public Health Statement: Ammonia, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/ToxProfiles/phs9003.html, December 1990.

ATSDR 9006, Public Health Statement: Chlorobenzene, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/ToxProfiles/phs9006.html, December 1990.

ATSDR 9008, Public Health Statement: Copper, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/ToxProfiles/phs9008.html, December, 1990.

ATSDR 9015, Public Health Statement: Ethylbenzene, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/ToxProfiles/phs9015.html, December 1990.

ATSDR 9018, Public Health Statement: Naphthalene, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/ToxProfiles/phs9018.html, December, 1990.

ATSDR 9024, Public Health Statement: Silver, Agency for Toxic Substances and Disease Registry, http://atsdr1.atsdr.cdc.gov:8080/ToxProfiles/phs9024.html, March, 1992.

CAA, Clean Air Act Amendments, Public Law #101-549, Section 301, November 15, 1990.

CARB, Toxic Air Contaminant Identification List: Compound Summaries, Draft, State of California Air Resource Board, January 1996.

EPA, Emission Inventory of Section 112(c)(6) Pollutants: Policyclic Organic Matter (POM), 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)/2,3,7,8-tetrachlorodizenzofuran (TCDF), and Polychlorinated Biphenyl Compounds (PCBs), Final Report and background data, Radian Corporation, Research Triangle Park, NC. March, 1995b.

EPA, Motor Vehicle-Related Air Toxics Study, EPA 420-R-93-005, Office of Mobile Sources, Emission Planning and strategies Division, Ann Arbor, MI, April, 1993.

EPA, National Air Pollutant Emissions Trends, 1900-1994, EPA-454/R-95-011, p 3-13, Office of Air Quality Planning and Standards, U.S. Environmental Protection Agency, Research Triangle Park, NC. October 1995.

EPA, National Screening Inventory of Sources and Emissions of Candidate 112(k) Pollutants to Support the Urban Area Source Program, Benzene, 1,3-Butadiene, Perchloroethylene, Trichloroethylele, Methylene Chloride, carbon Tetrachloride, and Formaldehyde, Review Draft Report and background data, U.S. Environmental Protection Agency, Research Triangle Park, NC. April, 1995a.

EPA, Procedures for Emission Inventory Preparation: Volume IV: Mobile Sources, U.S. Environmental Protection Agency, Office of Mobile Sources, Ann Arbor, MI and Office of Air Quality Planning and Standards, Research Triangle Park., NC, EPA-450/4-81-026d (Revised), 1992.

Radian, Air Toxics Emissions Inventory Protocol for the Great Lakes Commission, Radian Corporation, Sacramento, CA, June, 1994.

SC, Steering Committee for the Great Lakes Emission Inventory of Toxic Air Contaminants Conference Call, May 22, 1996a.

SC, Steering Committeefor the Great Lakes Emission Inventory of Toxic Air Contaminants Meeting, May 1-2, 1996b.

Table 3-3. Potential uses and their required resolutions for the regional mobile source emission inventory

* The size of gridded areas may vary but is always smaller than the size of a county.