ArticlePDF Available

Emission Projections for the EPA Section 812 Second Prospective Clean Air Act Cost/Benefit Analysis

Authors:
  • SC&A, Inc.

Abstract and Figures

Section 812 of the Clean Air Act Amendments of 1990 requires the U.S. EPA to perform periodic, comprehensive analyses of the total costs and total benefits of program implemented pursuant to the CAA. The first prospective analysis was completed in 1999. The second prospective analysis was initiated during 2005. The first step in the second prospective analysis was the development of base and projection year emission estimates, which will be used to generate benefit estimates of CAAA programs. This paper describes the analysis methods and results of the recently completed emission projections. There are several unique features of this analysis. One is the use of consistent economic assumptions from the Department of Energy's "Annual Energy Outlook 2005" projections as the basis for estimating 2010 and 2020 emissions for all sectors. Another is the analysis of the different emissions paths for both with and without CAAA scenarios. Other features of this analysis include being the first EPA analysis that uses the 2002 National Emission Inventory files as the basis for making 48 state emission projections, incorporating control factor files from the regional planning organizations that had completed emission projections at the time the analysis was performed, and modeling the emission benefits of the expected adoption of measures to meet the 8 hour ozone NAAQS, the Clean Air Visibility Rule, and the PM 2.5 NAAQS. The results of this study have been reviewed by EPA's Science Advisory Board.
Content may be subject to copyright.
1
Emission Projections for the EPA Section 812 Second Prospective Clean Air Act
Cost/Benefit Analysis
James H. Wilson Jr., Maureen A. Mullen, Andrew D. Bollman, Kirstin Thesing, Manish Salhotra
E.H. Pechan & Associates, Inc., 5528-B Hempstead Way, Springfield, VA 22151
James Neumann, Jason Price
Industrial Economics, Inc.
James DeMocker
U.S. Environmental Protection Agency, Office of Policy Analysis and Review, Washington, DC.
Corresponding Author: Jim Wilson
Email: jim.wilson@pechan.com
Phone: 703-813-6700 ext 102
Fax: 703-813-6729
ABSTRACT
Section 812 of the Clean Air Act Amendments of 1990 requires the U.S. EPA to perform periodic,
comprehensive analyses of the total costs and total benefits of program implemented pursuant to the
CAA. The first prospective analysis was completed in 1999. The second prospective analysis was
initiated during 2005. The first step in the second prospective analysis was the development of base and
projection year emission estimates, which will be used to generate benefit estimates of CAAA programs.
This paper describes the analysis methods and results of the recently completed emission projections.
There are several unique features of this analysis. One is the use of consistent economic assumptions
from the Department of Energy’s “Annual Energy Outlook 2005” projections as the basis for estimating
2010 and 2020 emissions for all sectors. Another is the analysis of the different emissions paths for both
with and without CAAA scenarios. Other features of this analysis include being the first EPA analysis
that uses the 2002 National Emission Inventory files as the basis for making 48 state emission
projections, incorporating control factor files from the regional planning organizations that had
completed emission projections at the time the analysis was performed, and modeling the emission
benefits of the expected adoption of measures to meet the 8 hour ozone NAAQS, the Clean Air
Visibility Rule, and the PM2.5 NAAQS. The results of this study have been reviewed by EPA’s Science
Advisory Board.
INTRODUCTION
Section 812 of the Clean Air Act Amendments of 1990 (CAAA) requires the U.S. Environmental
Protection Agency (EPA) to perform periodic, comprehensive analyses of the total costs and total
benefits of programs implemented pursuant to the Clean Air Act (CAA). The first analysis required was
a retrospective analysis, addressing the original CAA and covering the period 1970 to 1990. The
retrospective was completed in 1997. Section 812 also requires performance of prospective cost-benefit
analyses, the first of which was completed in 1999. The prospective analyses address the incremental
costs and benefits of the CAAA. The first prospective covered implementation of the CAAA over the
period 1990 to 2010.
EPA’s Office of Air and Radiation (OAR) began work on the second prospective with the drafting of an
analytical plan for the study. This analytical plan was reviewed by a statutorily-mandated outside peer
review group, the Advisory Council for Clean Air Compliance Analysis (Council), and the Council
provided comments, which have been incorporated into the technical analysis planning. This paper
2
describes the development of base and projection year emission estimates for the second prospective
section 812 analysis.
The scope of this analysis is to estimate future emissions of all criteria pollutants except lead: volatile
organic compounds (VOCs), oxides of nitrogen (NOx), carbon monoxide (CO), sulfur dioxide (SO2),
particulate matter of 10 microns or less (PM10), and particulate matter with an aerodynamic diameter of
2.5 microns or less (PM2.5). Estimates of current and future year ammonia (NH3) emissions are also
included in this study because of their importance in the atmospheric formation of secondary particles.
The study years are 1990, 2000, 2010, and 2020. This paper presents the results of EPA’s analysis of
the future effects of implementation of the CAAA’s programs on air emissions from the following
emission sectors: electricity generating units (EGUs), non-electricity generating unit point sources,
nonroad engines/vehicles, on-road vehicles, and nonpoint sources. The purpose of this paper is to
present the methods used to generate emissions projections under the two different control scenarios,
and to provide emission summaries for each. Examples of programs modeled under this analysis
include:
Title I VOC and NOx reasonably available control technology (RACT) requirements in ozone
nonattainment areas (NAAs);
Title II on-road vehicle and nonroad engine/vehicle provisions; and
Title III National Emission Standards for Hazardous Air Pollutants (NESHAPs).
The effects of Title IV CAAA programs on emissions from EGUs are analyzed using EPA’s Integrated
Planning Model (IPM). The results of the EGU modeling are reported here, along with a summary of
the approach. The results of this analysis provide the input for the air quality modeling and benefits
estimation stages of the second prospective analyses. The emission inputs to the modeling are more
detailed than the summaries provided in this paper, but will be available online at
www.epa.gov/oar/sect812.
METHODS
Selection of Base Year Inventory
Table 1 summarizes the key databases that were used in this study to estimate emissions for historic
years 1990 and 2000. These two years are the respective base years for preparing emission projections
for the with- and without-CAAA scenarios for 2010 and 2020. The without-CAAA scenario emission
projections are made from a 1990 base year. For EGU and non-EGU point sources, 1990 emissions are
estimated using the 1990 EPA National Emission Inventory (NEI) point source file. This file is
consistent with the emission estimates used for the First Section 812 Prospective and is thought to be the
most comprehensive and complete representation of point source emissions and associated activity in
that year. Similarly, the 1990 EPA NEI nonpoint source file (known at the time as the area source file) –
with a few notable exceptions – is used to estimate 1990 nonpoint source sector emissions. The
exceptions are where 1990 emissions were re-computed using updated methods developed for the 2002
NEI for selected source categories with the largest criteria pollutant emissions and most significant
methods changes.
The 1990 onroad and nonroad vehicle/engine sector emissions were estimated independently for this
project using consistent modeling approaches and activity estimates across the scenarios and years of
interest. For example, MOBILE6.2 emission factors and 1990 and 2000 NEI vehicle miles traveled
(VMT) databases were used to estimate onroad vehicle emissions for 1990 and 2000. Similarly, EPA’s
NONROAD 2004 model was used to estimate 1990 and 2000 emissions for nonroad vehicles/engines.
3
For calendar year 2000, with-CAAA scenario non-EGU point source emissions were estimated using the
2002 EPA NEI point source file (final version 2.0). For nonpoint sources, with-CAAA scenario
emissions in calendar year 2000 were estimated using the 2002 EPA NEI nonpoint source file (final
version 1). We selected the year 2002 NEI to represent the year 2000 emissions for two reasons:
(1) because the 2002 NEI incorporates a number of emission methods refinements over the 1999 NEI,
improving the accuracy of the base year estimate; and (2) because we believe that emissions for the year
2000 for this sector are not significantly different from emissions for the year 2002. For nonpoint
sources, with-CAAA scenario emissions in calendar year 2000 also were estimated using the 2002 EPA
NEI nonpoint source file (final) for the same reasons.
The logic for these base year inventory choices relates to the specific definitions of the scenarios
themselves. The with-CAAA scenario tracks compliance with CAAA requirements over time; as a
result, the best current basis for projecting the with-CAAA scenario incorporates decisions made since
1990 to comply with the Act. The 2002 NEI provides the best current understanding of the technologies
applied to meet emission reductions mandated under the CAAA. Over the next several decades,
however, we would expect that the mix of economic activity across polluting sectors will change. In
addition, we would expect that continued technological progress could improve the effectiveness and/or
reduce the cost of applying these technologies. Pollution prevention and changes in production methods
could also lead to reductions in air pollution. The change in the mix of economic activity is addressed
directly by our choice of activity drivers for the projections, as discussed below. Addressing the pace of
technological progress is more difficult; in many cases, we have only limited ability to forecast
technological advancements and their effect on air pollution emissions. In other cases, we can use the
pace of technological progress to date to project the pace of future improvements. To address this
factor, the overall analytical plan includes an assessment of the effects of learning-by-doing on costs, in
a sector-specific fashion. This is consistent with our assessment that, for most of the federal measures
assessed as part of the with-CAAA scenario, which require specific emission reductions, technologies,
or caps, emission outcomes will not be affected by technological progress, but the costs of those
reductions will be affected. It is also consistent with the trend in emissions just prior to 1990 as
documented in the First Prospective analysis. Just prior to the passage of the CAAA, the steep
downward emissions trend that had been seen in the 1970s and early 1980s for many pollutants were
starting to be reversed—that is, emissions were starting to move upward as economic activity continued,
but the stringency of standards remained largely fixed.
Criteria pollutant emissions were projected to 2010 and 2020 to determine the impact of CAA controls
on future year emission levels. Emissions were projected under two scenarios:
Without-CAAA – applies expected increases in activity levels with no additional controls implemented
beyond those that were in place when the CAAA were passed.
With-CAAA – applies expected increases in activity levels and incorporates the effects of controls
mandated under the 1990 Amendments to the CAA.
The general procedure used to project emissions is as follows:
Grow base year emissions or activity levels to the future year; and
Apply future year control efficiencies or emission factors.
Table 2 details the modeling approach used to project emissions for each of the major sectors. Table 3
summarizes the two projection scenarios.
One of the major study objectives was to provide the maximum feasible internal consistency in the use
of projection methods. We expect that energy demand, energy prices, and diffusion rates of
4
technologies are closely tied to the rate of growth of future air pollutant emissions and are closely linked
to expectations of the future growth path of the U.S. economy. Economic growth projections enter the
emissions analysis of the second prospective in three places:
The electricity demand forecast included in IPM (this forecast has in the recent past been based on the
reference case economic growth assumptions included in the Department of Energy’s AEO 2005);
The fuel consumption forecast for non-utility sectors that serves as the activity driver for major fuel-
consuming sectors (this forecast is also based on the reference case economic growth assumptions
included in AEO); and
The economic growth projections that serve as activity drivers for several other sources of air pollutants
(see Table 2).
In addition, the AirControlNET model that we use to assess compliance options for meeting the new
NAAQS, and which also calculates associated emissions implications, has recently been re-designed to
accept energy prices and labor rates as global inputs.
For this analysis, the Agency chose to use fully integrated economic growth, energy demand, and fuel
price projections for central case economic growth scenarios. The primary advantage of this approach is
that it allows the project team to conduct an internally consistent analysis of economic growth across all
emitting sectors. The system chosen was the Department of Energy’s National Energy Modeling
System (NEMS). Our central case emissions estimates rely on the DOE Annual Energy Outlook (AEO)
2005 reference case scenarios. A major strength of this approach is the integrated nature of the key
scenario driver data.
Figure 1 illustrates how the control requirements at the Federal, regional, local, and source levels are
considered in order to determine the most stringent (or binding) requirement by source category for
application in the core scenarios analysis. The core scenarios analysis is what is described in this paper,
and it includes the measures that have been adopted by areas to meet attainment requirements for the
1-hour ozone NAAQS, and the PM10 NAAQS.
The analysis of local controls to meet attainment requirements for 8-hour ozone and PM2.5 ambient
standards is described separately. Key components of this analysis include estimating the incremental
emission reductions expected to be associated with attaining the 8-hour ozone NAAQS, the PM2.5
NAAQS, and the Federal BART rule. These emission reductions are measured from the core scenarios
analysis baseline. The bottom part of Figure 1 depicts the key parts of this local controls analysis.
Economic Factors
In keeping with past EPA practice, this study relies on energy data from the U.S. Department of Energy
(DOE)’s Energy Information Administration (EIA) to backcast/forecast energy consumption and energy
production emission source categories. To reflect the 1990 to 2000 trend in energy consumption for
source categories, the project team generally relied on historical time-series energy data for each State
from an EIA energy consumption database1. For Crude Oil and Natural Gas Production source
categories, we obtained relevant 1990 and 2000 State-level activity data from an EIA source that
provides the number of operating oil well days (used for Crude Oil Production) and the number of
operating gas well days (used for Natural Gas Production)2. For source categories that describe railroad
and marine distillate fuel consumption emission processes, we obtained State-level 1990 and 2000
consumption estimates from an EIA distillate fuel data resource3.
Each year, the EIA produces energy projections for the United States. These projections, which forecast
U.S. energy supply, demand, and prices through 2025, are published in an EIA document entitled
Annual Energy Outlook 2005 (AEO 2005)4. For most energy sectors/fuel types, AEO 2005 reports
5
energy forecasts by Census division. These divisions are defined by State boundaries (e.g., Texas is
included in the West South Central region). When AEO 2005 produces Census division forecasts, these
regional data were used to project changes in the emissions activity for each State in the division. For
example, Stage II (Gasoline Vehicle Refueling) emission activity in Texas is projected using AEO 2005
projections of West South Central region transportation sector motor gasoline consumption. This study
relies on national energy forecasts whenever AEO 2005 only produces national projections for the
energy growth indicator of interest. Figure 2 displays forecast data for 3 of the approximate 50 energy
sectors for which AEO 2005 only produces national projections.
Because population growth and the performance of the U.S. economy are two of the main determinants
of energy demand, the EIA also prepares socioeconomic projections. These projections feed into energy
demand models incorporated into the EIA’s National Energy Modeling System (NEMS). NEMS
incorporates population projections and economic output forecasts for most industry sectors by Census
division. For non-energy intensive economic sectors (e.g., Wholesale Trade), EIA prepares national-
level output forecasts. This study relies on AEO 2005 historical and forecast socioeconomic data as
surrogates for emission activity level changes for most non-energy source categories. When AEO 2005
reported Census division forecasts, each emission source’s State identifier was used to link to the
appropriate AEO 2005 regional projections. National AEO 2005 data were used whenever NEMS only
produces national forecasts for the growth surrogate of interest. Table 6 presents key national AEO
2005 assumptions over the 2003 to 2025 forecast period. (As noted earlier, year 2000 emission activity
data were only needed in preparing the without CAA case emission estimates from 1990 base year
emissions. For these years, we relied on historical energy data.)
Non-EGU Point Sources
The non-EGU point source emission projection approach for the with-CAAA scenario uses the 2002
version 2.0 NEI point source emission file (October 26, 2006 release date) as the base year, applied the
growth factors described above to estimate activity changes between the base year and the 2010 and
2020 projection years, and applies control factors or emission caps to simulate the effect of air pollution
control programs in each forecast year.
One of the important components of the emission projections is identifying and quantifying the effect of
federal, state and local air pollution control strategies on post-2002 emission rates. Because of the
recent and ongoing activity of the five RPOs in developing emission projections for their own modeling
domains, each of the RPOs was queried, and any available control factor files were obtained. The
common projection year for the RPOs is 2018. All RPOs have either developed, or are working toward
developing, 2018 emission forecasts. Some are also developing emission forecasts for 2009 or 2010
because these are expected 8 hour ozone attainment years. For this section 812 analysis, control factors
for the different projection years were reviewed, and adjusted where necessary to account for the timing
of regulation implementation. Table 7 lists the RPOs, the geographic areas that they include, their
projection years, and the information that was received from each to support this analysis.
In order to estimate the 2010 and 2020 emission benefits of air pollution emission regulations in
California, California ARB staff provided a control factor file that was used in the Central California
Ozone Study modeling effort. The California file included control factors by district, air basin, and
county, with source categories designated by California’s Emission Inventory Codes. The California
file has both rule-specific and composite (with all rules applied) control factors. The composite control
factors were used in this analysis.
The base year for evaluation of the without-CAAA scenario is the 1990 EPA NEI. For point sources,
this database was used along with the activity growth indicators described above to estimate without0-
CAAA emissions in 2000, 2010, and 2020. Because the 1990 NEI was developed before information
6
was available to state, local and tribal emission inventory preparers about how to estimate PM_filterable
and PM-condensable fractions, it was necessary to augment the 1990 NEI point source file to have it
include estimates of all of the PM components. In addition to providing a complete reporting of
filterable and condensable PM emissions, this augmentation step also fills in PM10 or PM2.5 primary
emission estimates where they appear to be missing, and converts any situation where PM2.5 emissions
are greater than PM10 emissions. The net result of applying the PM augmentation procedures to the
1990 NEI point source file (non-EGU portion) was a 99 thousand ton increase in PM10 primary
emissions and a 60 thousand ton decrease in PM2.5 primary emissions.
Electricity Generating Units
To assess CAAA-related emission impacts for NOx, SO2, and mercury, the project team used the
Integrated Planning Model (IPM) developed by ICF Resources, Inc. IPM is a dynamic, linear
programming model of the electric power sector that represents several key components of energy
markets (i.e., markets for fuels, emissions allowances, and electricity) and the linkages among them.
The model determines the utility sector’s least cost strategy for meeting energy and peak demand
requirements over a specified time period, accounting for a number of regulatory and non-regulatory
constraints (e.g., emission caps and transmission constraints).
Nonroad Engines/Vehicles
Nonroad engine/vehicle emission estimates were developed using EPA’s Office of Transportation and
Air Quality’s (OTAQ) NONROAD 2004 model for the source categories that this model covers. This
version of the model incorporates all Federal engine exhaust standards, and includes updates to the base
year diesel engine populations.
The NONROAD sector emissions modeling approach involved (1) revising existing model inputs to
better reflect region-specific growth rates, (2) preparing State and county-specific input files to model
local fuel programs for the with-CAAA scenario runs, and (3) modifying fleet emission rate inputs to
remove the effect of CAAA-related standards for the without-CAAA scenario simulations.
Appropriate temperature and fuel data inputs were compiled for each of the years of interest (1990,
2000, 2010, and 2020). Seasonal, state, or county-specific NONROAD model option files were
prepared to generate nationwide emissions for each scenario. The project team performed NONROAD
model runs to generate seasonal emissions for each inventory year. Seasonal emissions were summed to
develop annual emission estimates at the county and SCC level for each scenario year.
On-Road Vehicles
The general method for computing historical and projection year on-road vehicle emissions is to
multiply activity in the form of VMT by pollutant-specific emission factors. Emission factors for these
pollutants were generated using MOBILE6.2—EPA’s latest mobile source emission factor model.
Because California’s emission standards differ from those for the rest of the nation, and cannot be
accurately estimated with MOBILE6.2, emission factor estimates generated by the ARB were used for
the California emission calculations in 2000, 2010, and 2020. (Some non-California states have elected
to adopt the California motor vehicle emission standards in accordance with section 177 of the Clean Air
Act. The emission effects of these State adoptions have not been incorporated in this analysis.) Control
program inputs such as inspection and maintenance programs and fuel programs are specified at the
county-level for the MOBILE6.2 simulations. Temporally, emissions are computed by month and
summed to develop annual emission estimates.
7
RESULTS
Table 4 summarizes the national emission estimates by sector for each of the scenario years evaluated in
this study. Table 5 provides all sectors combined emission results for the same set of scenario years.
Non-EGU Point Sources
The VOC emission projections for non-EGU point sources show an overall 3 percent increase in VOC
emissions from 2002 to 2010 and a 14 percent increase from 2010 to 2020. For VOC emissions, there is
no dominant source category. This is a sector where many of the sources added controls in the 1990 to
2000 period in response to EPA NESHAPs. Between 2000 and 2010, there are additional NESHAP
requirements for certain source categories like petroleum refineries that produce lower emissions in
2010 than in 2000. However, for most source categories, VOC emissions are estimated to increase from
2000 to 2010. Then, because no additional emission control requirements are imposed after 2010, VOC
emissions in the 2010 to 2020 period increase in proportion to expected activity growth in this period.
Non-EGU point source NOx emissions are a product of fuel combustion. In the eastern United States,
many of the large fuel combustion sources are subject to the requirements of the NOx SIP Call, and these
requirements affect industrial boiler, gas turbine, RICE engine, and cement kiln emissions starting after
2002. Outside the NOx SIP Call area, there are stringent NOx rules affecting NOx sources in eastern
Texas, the Baton Rouge area in Louisiana, and in many Air Districts in California. Sources and
geographic areas affected by these requirements contribute to the expected emission reductions between
2000 and 2010. After 2010, some NOx emission increases are anticipated as fuel consumption by the
industrial sector continues to grow. Uncertainties in the NOx emission projections include whether NOx
SIP Call States include their affected non-EGU boilers and gas turbines as trading program sources,
whose NOx emissions are effectively capped, and whether sources affected by a 5 month ozone season
control program install controls that also reduce NOx emissions during the 7 month winter season.
Non-EGU SO2 emissions are expected to stay relatively stable over the forecast period. Industrial fuel
combustion SO2 emissions from boilers decline slightly from 2000 to 2010, and then increase to near
2000 levels by 2020. The slight upward trend in non-EGU SO2 emissions over the forecast period is a
result of string expected activity growth in the chemical industry and other industrial processes. Some
industries, such a copper smelting, that have historically been major SO2 emission contributors, are now
modest SO2 emission contributors to non-EGU SO2 emissions, and have little influence on future
national SO2 emissions in this sector. Refinery settlements produce SO2 emission reductions in the
forecast period for the petroleum industry.
Electricity Generating Units
For EGUs, under the without-CAAA scenario, NOx and SO2 emissions grow significantly between 1990
and 2010, but emissions of both pollutants remain relatively flat during the 2010 to 2020 period.
Without-CAAA NOx emissions from EGUs increase by about 4 percent during this period, while SO2
emissions fall by about 1 percent. This reflects the confluence of a number of factors during the 2010-
2020 period, including increased reliance on coal-fired plants in compliance with the NSPS and a
change in relative prices of different types of coal. Based on AEO2005 projections of coal mine
productivity, IPM estimates that the price of low-sulfur sub-bituminous coal will decline relative to
other types of coal during this period.
Under the with-CAAA scenario, EGU emissions of both NOx and SO2 decline significantly between
1990 and 2020. As shown in Table 4, EGU NOx emissions fell from 6.4 million tons in 1990 to
4.5 million tons in 2001. It is estimated that these emissions will continue falling to 2.4 million tons in
2010 and 2.0 million tons in 2020. Relative to the without-CAAA scenario, this represents a 71 percent
8
reduction in EGU NOx emissions for 2010 and a 77 percent reduction for 2020. Similarly, SO2
emissions from EGUs fell from 15.8 million tons in 1990 to 10.8 million tons in 2001. These emissions
are expected to continue to decline under the CAAA scenario to 6.4 million tons in 2010 and 4.3 million
tons in 2020. For 2001, this represents a 40 percent reduction in SO2 emissions relative to the without-
CAAA scenario and a 77 percent reduction for 2020.
In addition to NOx and SO2, we estimate that the CAAA also lead to reduced emissions of PM10 and
PM2.5 from EGUs, although these reductions are not as significant. For 2010, we estimate that primary
PM2.5 and PM10 emissions will be 25 percent and 21 percent less, respectively under the with CAAA
scenario than under the without CAAA scenario. By 2020, these differences will change to 34 and 29
percent for PM2.5 and PM10, respectively.
Nonroad Engines/Vehicles
For nonroad engines/vehicles, for the with-CAAA scenarios, overall VOC and NOx emissions
Decrease between 1990 and 2000, with further declines in 2010 and 2020. In some cases, the effects of
growth outweigh the impact of VOC and CO emission standards (for example, gasoline lawn and garden
and light commercial between 2010 and 2020). Overall NOx emissions for this sector generally decrease
with time as well, and are lower in the with-CAAA scenario. This results primarily from the large
reductions in NOx emissions from diesel engine emission standards. For gasoline-powered nonroad
equipment, however, NOx emissions increase relative to the without-CAAA scenario each year. This
occurs because of the use of HC and CO-reducing technologies that control the air-fuel mixture in the
cylinder, but produce higher NOx emissions due to higher combustion temperatures and increased
supply of oxygen.
PM2.5 and PM10 emissions from nonroad engines/vehicles are expected to decline with time in the with-
CAAA scenario. This occurs because Federal emission standards serve to reduce PM emission rates
with time, and reduced fuel sulfur contents reduce PM sulfate and lead to lower PM emissions as well.
These reductions in fuel sulfur levels also produce lower SO2 emissions for this sector during the
forecast period.
On-road Vehicles
For on-road vehicle emissions, in all cases—with the exception of ammonia—the total on-road
emissions in 2020 with the CAAA control measures in place are below the 1990 emission levels, despite
significant increases in VMT during this time period. By contrast, the 2020 without-CAAA emissions
are greater than the 1990 total on-road emissions for NOx, SO2, and NH3, and only modest emission
decreases occur for VOC, CO, PM10, and PM2.5.
For VOC, CO, and NOx, the on-road vehicle emissions from 1990 to each of the with-CAAA projection
scenarios show steady declines over time, while the emissions in the without-CAAA projections initially
decrease from 1990 levels, but then begin to increase. Several control programs in place in 1990
account for these initial declines including the Federal Motor Vehicle Control Program, Phase I RVP
requirements, and I/M programs already in-place in 1990. By 2000, several CAAAA programs begin to
reduce on-road emissions. These include: Phase II RVP requirements, the Tier 1 emission standards,
evaporative control requirements, federal reformulated gasoline, oxygenated gasoline, more stringent
I/M requirements, and California LEV standards in California. After 2000, the national LEV emission
standards, Phase II of the Federal reformulated gasoline program, local low RVP gasoline programs, the
Tier 2 emission standards, low sulfur gasoline, heavy-duty vehicle emission standards, and low sulfur
diesel fuel all contribute to lowering emissions.
9
Nonpoint Sources
National nonpoint source VOC emissions are dominated by evaporative emissions from solvent
utilization. While there is some additional regulation of these emissions after 2002 in areas with
continuing ozone nonattainment, in most areas of the country, solvent utilization emissions grow after
2002 in proportion to activity indicators like population and employment. Another prominent VOC
emitting category is Fuel Combustion-Other, which is mostly residential fireplace and woodstove
emissions. (Most highly efficient fuel combustors are low VOC emitters.) Fireplace and woodstove
emissions are projected to decline after 2002 as NSPS-certified woodstoves replace non-certified stoves.
Another prominent VOC-emitting source category with expected emissions declines in 2010 and 2020 is
fuel storage and transport. Control programs that contribute to these emission reductions include
onboard VRS on gasoline-powered vehicles and more stringent State and local programs to reduce
emissions at various points in the gasoline distribution system.
National NOx emissions for the nonpoint source sector are dominated by off-highway sources (marine,
aircraft and railroad). NOx emission reductions between 2002 and 2010 are a result of Federal emission
standards for some commercial marine vessel engines and locomotive engines. Besides off-highway
engines, the other nonpoint source NOx emitters with more than 10 percent of total emissions for this
sector are: industrial and other fuel combustion and petroleum and related processes. These are all small
fuel combustors that are exempt from regulations like the NOx SIP Call because of their size. Their NOx
emissions are expected to increase slightly during the study time horizon.
SO2 emissions for this sector are expected to stay relatively stable from 2002 to 2020. The dominant
source type is industrial fuel combustion and these emissions represent coal and fuel oil combustion that
occurs in sources that are not included in the 2002 NEI point source file. The off-highway sector SO2
contribution is small because most of the off-highway source emissions are from diesel engines
(Commercial marine vessels and locomotives) or jet aircraft engines.
CONCLUSIONS
This analysis shows that the 1990 CAAA have produced significant reductions in most criteria pollutant
emissions since 1990 and that these emission reductions are expected to continue through 2020.
REFERENCES
1. Energy Information Administration, “State Energy Consumption, Price, and Expenditure
Estimates (SEDS),” U.S. Department of Energy, Washington, DC, Texas data retrieved on
July 20, 2005 from
http://www.eia.doe.gov/emeu/states/state.html?q_state_a=tx&q_state=TEXAS.
2. Energy Information Administration, “Distribution and Production of Oil and Gas Wells by
State,” U.S. Department of Energy, Washington, DC, Texas data retrieved on July 22, 2005 from
http://www.eia.doe.gov/pub/oil_gas/petrosystem/petrosysog.html.
3. Energy Information Administration, “Sales of Distillate Fuel Oil by End Use,” U.S. Department
of Energy, Washington, DC, Texas data retrieved on July 20, 2005 from
http://tonto.eia.doe.gov/dnav/pet/pet_cons_821dst_dcu_STX_a.htm.
4. Energy Information Administration, Annual Energy Outlook 2005 With Projections to 2025,
DOE/EIA-0383(2005), U.S. Department of Energy, Washington, DC, February 2005.
10
5. MACTEC, Inc., “Documentation of the Revised 2002 Base Year, Revised 2018, and Initial 2009
Emission Inventories for VISTAS,” prepared for Visibility Improvement State and Tribal
Association of the Southeast, Asheville, NC, 2005.
6. E.H. Pechan & Associates, Inc., “Development of Growth and Control Factors for Lake
Michigan Air Directors Consortium (LADCO),” Springfield, VA, prepared for Lake Michigan
Air Directors Consortium, Des Plaines, IL, December 15, 2004.
7. E.H. Pechan & Associates, Inc., “Development of Growth and Control Inputs for CENRAP 2018
Emissions – Technical Support Document,” Springfield, VA, prepared for Central Regional Air
Planning Association, Oklahoma City, OK, May 2005.
11
KEY WORDS
Motor vehicle pollution
Clean Air Act
Electric power
Air pollution control
Off-road vehicles
12
Table 1. Base year emission data sources for the With- and Without CAAA scenarios.
Sectors Without-CAAA Scenario – 1990 With-CAAA Scenario – 2000
EGU 1990 EPA Point Source NEI Estimated by the EPA Integrated
Planning Model for 2001
Non-EGU Point 1990 EPA Point Source NEI 2002 EPA Point Source NEI (Final
Version 2.0)
Nonpoint 1990 EPA Nonpoint Source NEI with
Adjustments for Priority Source Categories 2002 EPA Nonpoint Source NEI
(Final Version 1)
On-Road MOBILE6.2 Emission Factors and 1990 NEI
VMT Database MOBILE6.2 Emission Factors and
2000 NEI VMT Database. The
California Air Resources Board
(ARB) supplied estimates for
California
Off-Road/Nonroad NONROAD 2004 Model Simulation for
Calendar Year 1990 NONROAD 2004 Model Simulation
for Calendar Year 2000
Table 2. Modeling approach by major sector.
Sector Growth Forecast Controls Modeling Approach
Non-EGU Point U.S. Department of Energy (DOE) Annual Energy Outlook
2005 forecasts Based on control factors developed by
the five Regional Planning
Organizations (RPOs), and California
information from the ARB
EGU DOE Annual Energy Outlook 2005 forecasts Integrated Planning Model (IPM)
Nonroad EPA NONROAD Model growth forecasts are largely based
on historical trends in national engine populations by
category/sub-category of engine
EPA NONROAD Model
Onroad National VMT Forecast from Annual Energy Outlook 2005
(AEO 2005) MOBILE6.2 emission factors
Nonpoint DOE AEO 2005 forecasts Based on control factors developed by
the five RPOs, and California
information from the ARB
13
Table 3. Projection scenario summary by major sector in the second prospective.
Sector Without-CAAA With-CAAA*
Non-Electricity
Generating Unit
Point
RACT held at 1990 levels
NOx:
VOC/HAP:
SOx:
NOx/VOC:
RACT for all NAAs (except NOx waivers),
Ozone Transport Commission (OTC) small NOx source model rule (where adopted),
Cases and settlements,
NOx measures included in ozone State Implementation Plans (SIPs) and SIP Call post-2000,
Additional measures to meet PM and ozone National Ambient Air Quality Standards
(NAAQS).
RACT for all NAAs,
VOC measures included in ozone SIPs,
2-, 4-, 7-, and 10-year maximum achievable control technology (MACT) standards,
New control technique guidelines (CTGs).
Cases and settlements,
Additional measures to meet revised PM NAAQS.
Rate-of-Progress (3 percent per year) requirements (further reductions in VOC),
Early action compacts.
Electricity
Generating Unit
RACT and New Source
Review (NSR) held at 1990
levels.
250 ton Prevention of
Significant Deterioration
(PSD) and New Source
Performance Standards
(NSPS) held at 1990 levels.
NOx:
SOx:
RACT and NSR for all non-waived (NOx waiver) NAAs,
SIP Call post -2000,
Phase II of the OTC NOx memorandum of understanding,
Title IV Phase I and Phase II limits for all boiler types,
250 ton PSD and NSPS,
Clean Air Interstate Rule (CAIR),
Clean Air Mercury Rule,
Cases and settlements,
Additional measures to meet PM and ozone NAAQS.
Title IV emission allowance program,
CAIR,
Clean Air Mercury Rule,
Cases and settlements,
Additional measures to meet revised PM NAAQS.
14
Table 3 (continued).
Sector Without-CAAA With-CAAA*
Non-road Engines/
Vehicles**
Controls (engine standards)
held at 1990 levels.
NOx:
VOC/HAP:
CO:
PM:
SOx:
Federal Phase I and II compression ignition (CI) and spark-ignition (S-I) engine standards,
Federal locomotive standards,
Federal commercial marine vessel standards,
Federal recreational marine vessel standards,
NOx measures included in ozone SIPs,
Nonroad Diesel Rule.
Federal Phase I and II S-I engine standards,
Federal recreational marine vessel standards,
Federal large SI/recreational vehicle engine standards,
Federal large SI/evaporative standards,
VOC measures included in ozone SIPs.
Federal large S-I evaporative standards,
Federal Phase I and II S-I engine standards.
Federal Phase I and II CI engine standards,
Federal Phase I and II S-I engine standards,
Federal locomotive standards,
Federal commercial marine vessel standards,
Nonroad Diesel Rule.
Nonroad Diesel Rule,
Gasoline fuel sulfur limits.
15
Table 3 (continued).
Sector Without-CAAA With-CAAA*
On-road Motor
Vehicles***
Federal Motor Vehicle Control
Program - engine standards set
prior to 1990.
Phase 1 Reid vapor pressure
(RVP) limits.
Inspection and maintenance
(I/M) programs in place by
1990.
NOx :
VOC/HAP:
CO:
PM:
SOx:
Tier 1 tailpipe standards (Title II), Tier 2 tailpipe standards,
49-State low-emission vehicle (LEV) program (Title I), I/M programs for ozone and CO NAAs
(Title I),
Federal reformulated gasoline for ozone NAAs (Title I),
California LEV (California only) (Title I),
California reformulated gasoline (California only) (Title I),
NOx measures included in ozone SIPs, heavy-duty diesel vehicle (HDDV) standards,
HDDV defeat device settlements
Additional measures to meet PM and ozone NAAQS.
Tier 1 tailpipe standards (Title II), Tier 2 tailpipe standards,
49-State LEV program (Title I), I/M programs for ozone and CO NAAs (Title I),
Phase 2 RVP limits (Title II), Federal reformulated gasoline for ozone NAAs (Title I),
California LEV (California only) (Title I),
California reformulated gasoline (California only) (Title I),
VOC measures included in ozone SIPs, HDDV standards,
Enhanced evaporative test procedures,
Additional measures to meet PM and ozone NAAQS.
49-State LEV program (Title I), I/M programs for CO NAAs (Title I),
Tier 2 tailpipe standards, California LEV (California only) (Title I),
California reformulated gasoline (California only) (Title I),
Oxygenated fuel in CO NAAs (Title I), HDDV standards.
HDDV standards, diesel fuel sulfur content limits (Title II) (1993).
Diesel fuel sulfur content limits (Title II) (1993),
HDDV standards and associated diesel fuel sulfur content limits, Gasoline fuel sulfur limits,
Tier 2 tailpipe standards, Additional measures to meet new PM NAAQS.
16
Table 3 (continued).
Sector Without-CAAA With-CAAA*
Area/Nonpoint
Controls held at 1990 levels
NOx:
VOC/HAP:
PM:
NOx/VOC:
RACT requirements,
NOx measures included in ozone SIPs,
Additional measures to meet PM and ozone NAAQS.
RACT requirements,
New CTGs, 2-, 4-, 7-, and 10-year MACT Standards,
Onboard vapor recovery (vehicle refueling),
Stage II vapor recovery systems (VRS),
Federal VOC rules for architectural and industrial maintenance (AIM) coatings, autobody
refinishing, and consumer products,
Additional measures to meet PM and ozone NAAQS.
PM2.5 and PM10 NAA controls,
VOC measures included in ozone SIPs.
Rate-of-Progress (3% per year) requirements (further reductions in VOC),
Model rules in OTC States,
Early action compacts.
NOTES: *Also includes all Without-CAAA measures.
**The nonroad mobile source standards included in the With-CAAA scenario are based on the standards found within the NONROAD2004 emissions inventory model. Three other nonroad mobile
standards, not captured by the NONROAD2004 model, are also included in the With-CAAA scenario: the locomotive standards, commercial marine engine standards, and the large SI/evaporative
standards.
***The motor vehicle mobile source standards included in the With-CAAA scenario are based on the standards found within the MOBILE6.2 emissions inventory model. Note that emissions
associated with the Final Rule for Cleaner Highway Motorcycles (promulgated in 2004) are not accounted for in the MOBILE6.2 model, and are not included in the With-CAAA scenario.
17
Table 4. Summary of national (48 state) emission estimates by scenario year.
Pollutant
Sector
1990 2000 without-
CAAA 2000 with-
CAAA 2010 without-
CAAA 2010 with-
CAAA 2020 without-
CAAA 2020 with-
CAAA
VOC EGU 34,558 40,238 40,882 43,333 42,661 48,001 46,992
Non-EGU Point 2,609,368 3,077,597 1,402,343 3,462,797 1,435,475 3,999,199 1,647,551
Nonpoint 11,152,804 12,268,609 8,544,345 13,425,477 8,872,248 15,702,681 9,715,546
Nonroad 2,665,710 3,217,810 2,564,790 4,076,796 1,874,723 4,753,500 1,489,644
On-Road Vehicle 9,327,660 5,872,983 5,245,756 5,734,012 2,614,007 6,784,539 1,670,617
NOx EGU 6,410,533 7,734,000 4,493,981 8,349,482 2,437,219 8,686,216 1,986,463
Non-EGU Point 3,133,450 3,331,308 2,292,311 3,555,874 2,246,621 3,997,276 2,509,040
Nonpoint 4,768,841 4,650,355 3,885,707 4,840,735 3,688,289 5,198,279 3,725,010
Nonroad 2,067,745 2,190,711 2,091,459 2,664,838 1,643,413 3,162,409 998,918
On-Road Vehicle 9,535,993 8,782,108 8,073,738 9,105,919 4,349,062 10,695,419 1,915,842
CO EGU 303,713 496,430 503,306 602,048 617,860 750,538 771,654
Non-EGU Point 5,667,404 6,466,855 3,112,631 6,808,250 3,290,804 7,381,679 3,677,434
Nonpoint 16,799,105 15,634,196 14,613,968 14,707,662 14,605,108 15,088,612 15,451,487
Nonroad 22,176,262 25,458,930 22,330,110 31,541,817 26,229,083 37,199,473 28,999,459
On-Road Vehicle 109,566,997 79,037,081 67,130,866 80,491,386 42,387,967 95,549,545 36,239,508
SO2 EGU 15,831,702 18,146,659 10,819,399 18,867,532 6,365,458 18,738,860 4,270,125
Non-EGU Point 4,293,268 4,099,586 2,193,213 4,487,265 2,177,099 4,871,531 2,387,367
Nonpoint 2,354,778 2,071,308 1,875,282 2,453,986 1,877,630 3,044,248 1,941,752
Nonroad 163,254 178,247 177,095 225,300 16,930 270,252 2,750
On-Road Vehicle 500,064 632,766 253,592 797,345 29,954 986,882 36,457
PM10 EGU 530,663 751,696 728,719 834,655 658,151 896,790 637,311
Non-EGU Point 1,734,810 2,013,691 597,875 2,201,812 582,635 2,491,106 681,858
Nonpoint 22,495,048 23,118,860 19,329,848 22,816,379 18,844,942 24,255,816 19,015,260
Nonroad 308,562 286,623 265,778 323,187 202,507 367,252 131,185
On-Road Vehicle 384,733 247,056 220,854 229,246 154,216 268,733 135,559
PM2.5 EGU 357,674 634,287 610,638 704,443 529,163 762,326 506,512
Non-EGU Point 1,299,259 1,515,932 365,260 1,651,644 393,943 1,871,552 451,169
Nonpoint 5,255,977 5,416,570 4,103,247 5,366,784 4,060,026 5,732,422 4,166,546
Nonroad 283,960 263,798 244,620 297,466 186,440 338,036 120,854
On-Road Vehicle 321,852 191,723 165,515 169,690 96,356 199,153 70,899
NH3 EGU 0 3,217 3,162 1,023 822 612 559
Non-EGU Point 243,615 236,126 153,944 237,459 173,946 255,636 201,670
Nonpoint 3,257,139 3,621,848 3,551,567 3,828,468 3,713,161 4,130,614 3,986,783
Nonroad 1,530 1,789 1,715 2,248 2,042 2,665 2,399
On-Road Vehicle 154,103 272,569 272,464 336,083 334,417 397,618 395,319
18
Table 5. Emission totals by pollutant - all sectors (thousand tons per year).
2000 without- 2000 with- 2010 without- 2010 with- 2020 without- 2020 with-
Pollutant 1990 CAAA CAAA CAAA CAAA CAAA CAAA
VOC 25,790 24,477 17,798 26,742 14,839 31,288 14,570
NOx 25,917 26,688 20,837 28,517 14,365 31,740 11,135
CO 154,513 127,093 107,691 134,151 87,131 155,970 85,140
SO2 23,143 25,129 15,319 26,831 10,467 27,912 8,638
PM10 24,841 25,688 21,143 25,636 20,442 27,411 20,601
PM2.5 6,906 7,292 5,489 7,420 5,266 8,035 5,316
NH3 3,656 4,136 3,983 4,405 4,224 4,787 4,587
19
Table 6. Key national assumptions reflected in AEO 2005.
Variable 2003 to 2025 Annual Growth Rate (%)
Population 0.8
Real Gross Domestic Product 3.1
GDP Chain-Type Price Index 2.5
Nonfarm Business Labor Productivity 2.2
Total Industrial Output 2.3
Manufacturing Output 2.6
Energy Intensive Manufacturing Output 1.5
Nonenergy Intensive Manufacturing Output 2.9
Services Sector Output 3.3
Energy Use Per Capita 0.5
Energy Use Per $ of Real Gross Domestic Product -1.6
Table 7. Regional Planning Organization Criteria Pollutant Control Factors for Regions/States –
Base Case 2010 and 2020
Regional Planning
Organization
Geographic Area Covered Analysis
Year(s)
Notes
1. Mid-Atlantic/
Northeast
Visibility Union
(MANE-VU)
Northeast and Mid-Atlantic 2018 MANE-VU provided a matrix that
summarized their on-the-books rules
by State and sub-state area. Their
control factors were not available
during this study period.
2. Visibility
Improvement -
State and Tribal
Association of
the Southeast
(VISTAS)
Southeast 2009
2018 Source: MACTEC, 20055
3. Lake Michigan
Air Directors
Consortium
(LADCO)
Great Lakes area 2007
2009
2012
2018
Source: Pechan, 20046
4. Central Regional
Air Planning
Association
(CENRAP)
Midwest 2018 Source: Pechan, 20057
5. Western
Regional Air
Partnership
(WRAP)
Western 2018 Control factors were not available
during this study period.
5a. California 2010
2020 California projection year control
factors were provided by the Air
Resources Board.
20
Figure 1. Control applications in the core scenario and local controls for NAAQS compliance analysis.
Federal Measures
Select Most
Stringent
Regulation by
Source
Regional Measures
Local Measures
Source-Specific
Requirements
Core Scenario
Examples:
MACT standards
Onroad and nonroad standards
Examples:
NOxSIP Call
OTC model rules
Examples:
1-hour ozone SIP measures
RFG
Examples:
Refinery settlements
Core Scenario Control Application:
Local Controls for NAAQS Compliance Application:
8-hour ozone
attainment simulation
Model PM2.5 SIPs
BART Rule
application
Attainment
Scenario
21
Figure 2. Sample national AEO 2005 energy sector forecasts.
300
350
400
450
500
550
600
650
700
2002 2004 2006 2008 2010 2012 2014 2016 2018 2020
Year
Energy Consumption (Trillion Btu)
Iron and Steel Industry,
Metallurgical C oal
Pa
p
er Industr
y
, Natural Gas
Freight Rail, Distillate Fuel
ResearchGate has not been able to resolve any citations for this publication.
State Energy Consumption, Price, and Expenditure Estimates (SEDS)
1. Energy Information Administration, " State Energy Consumption, Price, and Expenditure Estimates (SEDS), " U.S. Department of Energy, Washington, DC, Texas data retrieved on July 20, 2005 from http://www.eia.doe.gov/emeu/states/state.html?q_state_a=tx&q_state=TEXAS.
Sales of Distillate Fuel Oil by End Use Texas data retrieved on
Energy Information Administration, " Sales of Distillate Fuel Oil by End Use, " U.S. Department of Energy, Washington, DC, Texas data retrieved on July 20, 2005 from http://tonto.eia.doe.gov/dnav/pet/pet_cons_821dst_dcu_STX_a.htm. 4. Energy Information Administration, Annual Energy Outlook 2005 With Projections to 2025, DOE/EIA-0383(2005), U.S. Department of Energy, Washington, DC, February 2005.
Documentation of the Revised
  • Inc Mactec
MACTEC, Inc., "Documentation of the Revised 2002 Base Year, Revised 2018, and Initial 2009