EPA420-R-01-038
                                           August 2001
The Projection of Mobile Source Air Toxics from 1996
      to 2007: Emissions and Concentrations

                      DRAFT

-------
                   Limitations and Uncertainties

This analysis has a number of significant limitations and uncertainties
which must be considered in interpreting results. First, although on-
highway inventory estimates in 1996 and 2007 for benzene, 1,3-butadiene,
formaldehyde, and acetaldehyde were estimated using local data on fuel
properties, inspection/maintenance programs, and other input parameters,
for other parameters such as average speed, defaults were used. Use of
default parameters can result in significant underestimates or overestimates
of emissions at the local scale.  Also, MOBTOX modeling for these HAPs
was done for a limited number of areas (10 urban areas and sixteen
geographic regions) and the rest of the country was "mapped" to these
modeled areas or regions based on I/M programs, fuel parameters, and
temperature regimes. In addition, for some vehicle classes, in particular
heavy duty diesel vehicles, toxic emission estimates are based on very
limited speciation data. Furthermore, there are considerable uncertainties
associated with emission factors, activity, allocation surrogates and
speciation data used to develop the nonroad inventories.  The EPA has
several efforts underway to address these uncertainties.  Finally, projection
of emissions using surrogates such as VOC and activity introduces
significant additional uncertainty into the 2007 inventory.

Spatial allocation of emissions from the county to the census tract level also
introduces uncertainty, particularly for nonroad sources. Although there
are many types of nonroad equipment, all emissions were grouped into six
source categories - 2-stroke gasoline, 4-stroke gasoline, diesel, locomotives,
commercial marine vessels, and aircraft - prior to spatial allocation.

There are also uncertainties associated with the application of the ASPEN
dispersion model. Gaussian dispersion models such as ASPEN are designed
to work with inert pollutants and deal with reactive decay and secondary
transformation in a less sophisticated manner than an atmospheric
chemistry model.  Analyses done as part of the NSATA (EPA, 2001a)
suggest ASPEN may be underestimating concentrations for more reactive
species.  Model to monitor comparisons also suggest ASPEN may
significantly underestimate concentrations of metals. Finally, it should be
emphasized that this model does not reliably capture localized impacts.

-------
The Projection of Mobile Source Air Toxics from 1996 to 2007:
                  Emissions and Concentrations
                            Draft Report



                            July 10, 2001



                            Prepared for:

                     Dr. Madeleine Strum, MD-14

                 U.S. Environmental Protection Agency
              Research Triangle Park, North Carolina 27711
                    EPA Contract No. 68-D-98-006
                      Work Assignment No. 4-01
      RTF Area Office
     2327 Englert Drive
     Durham, NC 27713
       (919) 484-0222
                            Prepared by:



                           William Battye

                         EC/R Incorporated
   Chapel Hill Office
1129 Weaver Dairy Road
 Chapel Hill, NC 27514
    (919)933-9501

-------
                        Disclaimer

Although the research described in this document has been funded
wholly or in part by the United States Environmental Protection
Agency contract 68-D-98-006 to EC/R Incorporated, it has not been
subject to the Agency's review and therefore does not necessarily
reflect the views of the Agency, and no official endorsement should be
inferred.

-------
                                       Contents

                                                                                     Pag<

1.  Introduction	1

2.  Methodologies	5
   2.1 Emission Inventories for 1996	5
       2.1.1  1996 National Toxics Inventory (NTI)	5
       2.1.2  1996 National Emission Trends (NET) inventory, speciated for particular VOC . . 5
   2.2 Projected Emissions for 2007 	7
       2.2.1  Onroad emissions of gaseous HAPs	7
          2.2.1.1  Benzene, acetaldehyde, formaldehyde, butadiene, andMTBE  	7
          2.2.1.2  Other gaseous organic HAPs	8
       2.2.2  Nonroad emissions of gaseous HAPs	9
       2.2.3  Onroad emissions of particulate HAPs 	11
          2.2.3.1  Polycyclic organic matter	11
          2.2.3.2  Trace metals and dioxins	12
       2.2.4  Non-road emissions of particulate HAPs 	13
          2.2.4.1  Polycyclic organic matter	13
          2.2.4.2  Trace metals and dioxins	14
       2.2.5  Emissions of HAP precursors	15
   2.3 Spatial and Temporal Processing of Emissions 	16
   2.4 Ambient Concentration Modeling	17

3.  Results	21
   4.1 Projected Emissions in 2007	21
   4.2 Projected Ambient Impacts	32

4.  References 	48
                                           in

-------
                                        Figures

                                                                                     Pag<

Figure 1. Contribution of Mobile Source Emissions to Total Emissions of Gaseous HAPs  .... 2
Figure 2. Contribution of Mobile Source Emissions to Total Emissions of HAPs in the
       Particulate Phase  	2
Figure 3. Overview of EMS-HAP Processing of Mobile Sources  	18
Figure 4. Processing of Airports in EMS-HAP  	19
Figure 5. Processing of Other Mobile Sources in EMS-HAP	20
Figure 6. Projected county-level mobile source emission reductions for acetaldehyde  	33
Figure 7. Projected county-level mobile source emission reductions for acrolein	33
Figure 8. Projected county-level mobile source emission reductions for benzene	34
Figure 9. Projected county-level mobile source emission reductions for 1,3-butadiene	34
Figure 10. Projected county-level mobile source emission reductions for formaldehyde	35
Figure 11. Projected increase in total vehicle miles traveled	35
Figure 12. Projected mobile source ambient impact reductions for acetaldehyde	43
Figure 13. Projected mobile source ambient impacts for acetaldehyde in 2007	43
Figure 14. Projected mobile source ambient impact reductions for acrolein	44
Figure 15. Projected mobile source ambient impacts for acrolein in 2007  	44
Figure 16. Projected mobile source ambient impact reductions for benzene	45
Figure 17. Projected mobile source ambient impacts for benzene in 2007  	45
Figure 18. Projected mobile source ambient impact reductions for 1,3-butadiene	46
Figure 19. Projected mobile source ambient impacts for 1,3-butadiene in 2007	46
Figure 20. Projected mobile source ambient impact reductions for formaldehyde	47
Figure 21. Projected mobile source ambient impacts for formaldehyde in 2007	47
                                           IV

-------
                                       Tables

                                                                                   Pag<

Table 1. HAPs Included in Mobile Source Projections  	4
Table 2. Species Used for Modeling Secondary HAP Formation	6
Table 3. Growth Factors for Particulate HAP Emissions from Nonroad Engines	14
Table 4. Summary of Mobile Source Emission Projection Results for the Contiguous
       48 States and the District of Columbia	22
Table 5. Gaseous HAP Emission Projections for Different Onroad Vehicle Categories 	23
Table 6. Parti culate HAP Emission Projections for Different Onroad Vehicle Categories	24
Table 7. HAP Precursor Emission Projections for Different Onroad Vehicle Categories 	25
Table 8. Gaseous HAP Emission Projections for Different Nonroad Engine Categories	26
Table 9. Parti culate HAP Emission Projections for Different Nonroad Engine Categories  .... 27
Table 10. HAP  Precursor Emission Projections for Different Nonroad Engine Categories  .... 28
Table 11. Projected State-Level Changes in Total Mobile Source Emissions of Gaseous
       HAPs	29
Table 12. Projected State-Level Changes in Total Mobile Source Emissions of Paniculate
       HAPs	30
Table 13. Projected State-Level Changes in Total Mobile Source Emissions of HAP
       Precursors	31
Table 14. Summary of Projected Nationwide Ambient Impacts from Mobile Sources 	36
Table 15. Projected Average Ambient Impacts from Mobile Sources in Urban and Rural
       Counties	38
Table 16. Projected Changes in Ambient Concentrations - Nationwide Estimates  	39
Table 17. Projected Changes in Ambient Concentrations - Urban and Rural Counties	39
Table 18. Projected State-Level Changes in Total Mobile Source Ambient Impacts for
       Organic HAPs  	40
Table 19. Projected State-Level Changes in Total Mobile Source Ambient Impacts for
       Paniculate HAPs	41

-------
                             Definition of Acronyms
ASPEN      Assessment System for Population Exposure Nationwide
EFIG        EPA OAQPS Emission Factors and Inventories Group
EPA         U.S. Environmental Protection Agency
EMS-HAP    Emission Modeling System for Hazardous Air Pollutants
HAP         Hazardous Air Pollutant
HDDVHeavy duty diesel vehicles
HDGVHeavy duty gasoline vehicles
LDDT       Light duty diesel trucks
LDDV       Light duty diesel vehicles
LDGT       Light duty gasoline trucks, both categories 1 and 2
LDGT1      Light duty gasoline truck category 1
LDGT2      Light duty gasoline truck category 2
LDGV       Light duty gasoline vehicles
MC          Motorcycles
MEK        Methyl ethyl ketone
MTBE       Methyl tertiary butyl ether
NATA       National Air Toxics Assessment
NET         EPA National Emissions Trends inventory
NTI          EPA 1996 National Toxics Inventory
OAQPS      EPA Office of Air Quality Planning and Standards
OTAQ       EPA Office of Transportation and Air Quality
PAH         Polynuclear aromatic hydrocarbons (a subset of POM)
7-PAH       The seven specific PAH that nave been named as animal carcinogens by the
             International Agency for Research on Cancer (IARC)
POM        Polycyclic Organic Matter
ppm          parts per million
VMT        Vehicle miles traveled
VOC         Reactive volatile organic compounds (excluding methane and other nonreactive
             compounds)
TOG         Total organic gas
                                         VI

-------
                                   1.  Introduction
       The Office of Air Quality Planning and Standards' (OAQPS) has conducted an Initial
National-Scale Assessment of Air Toxics as part of the National Air Toxics Assessment (NATA)
activities in the Urban Air Toxics Program. This effort uses the 1996 National Toxics Inventory
(NTI)1 and the Assessment System for Population Exposure Nationwide (ASPEN)2 dispersion
model to predict nationwide ambient concentrations of toxic air pollutants for the year  1996.
EPA has also used the 1996 NTI to predict emissions and ambient concentrations of toxic air
pollutants in 2007. This report details the methodologies used to project future emissions of toxic
air pollutants from mobile sources, which consist of highway vehicles and other nonroad mobile
sources such as lawn mowers and other small engines, nonroad equipment, airports, marine
vessels, and railroads.

       The Initial National-Scale Assessment effort focuses on 33 toxic air pollutants, also known
as hazardous air pollutants, or HAPs, that are judged to present the greatest threat to public
health in urban areas. Mobile sources emit 15 of these 33 HAPs.  Five of these 15 are gaseous
HAPs: acetaldehyde, acrolein, benzene, 1,3-butadiene, formaldehyde. As shown in Figure 1,
mobile sources contribute a substantial share of total nationwide emissions of each of these
gaseous HAPs.

       The remaining ten HAPs emitted by mobile sources (from the list of 33) are trace metals
and compounds associated primarily with the particulate phase: arsenic, beryllium," cadmium,3
chromium, dioxins/furans,b lead, manganese, mercury, nickel, and polycyclic organic matter
(POM).C  Figure 2 shows that mobile source emissions of metals, and POM are small in
comparison with stationary source emissions.  In the case of POM, separate analyses were
performed for total POM and for seven polynuclear aromatic hydrocarbons (7-PAH) that have
been named as animal carcinogens by the International Agency for Research on Cancer (IARC).
       a Beryllium and cadmium were not included in EPA's recently finalized list of mobile
source air toxics, because they were all from marine vessels using residual fuel oil, and a profile
for stationary industrial and commercial boilers was used to develop this list, rather than actual
measurements from a commercial marine engine.

       b National estimates are not yet available for dioxins and furans from stationary sources.

       c Estimates of total  POM for mobile sources represent the sum of 16 polynuclear aromatic
hydrocarbons (16-PAH). The following is a list of compounds included in the 16-PAH category,
with asterisks denoting compounds  that are also members of 7-PAH: acenaphthene,
acenaphthylene, anthracene, benz(a)anthracene,* benzo(a)pyrene,* benzo(b)fluoranthene,*
benzo(g,h,i)perylene, benzo(k)fluoranthene,* chrysene,* dibenz(a,h)anthracene,* fluoranthene,
fluorene, indeno(l,2,3-c,d)pyrene,*  naphthalene, phenanthrene, and pyrene. Mobile source
emissions estimates for POM account only for the  particulate phase, although some significant
POM are emitted in the gaseous phase. The 1999 National Toxicx Inventory will include mobile
source gas-phase PAH.

                                            1

-------
 Acetaldehyde

       Acrolein
      Benzene
 1,3-Butadiene

 Formaldehyde
                                  100               200
                                   Annual emissions (thousand tons)
                            300
                         400
                             Onroad Mobile
                             Nonroad Mobile
         Major stationary sources
         Stationary area sources
Figure 1. Contribution of Mobile Source Emissions to Total Emissions of Gaseous HAPs
             Contiguous 48 States and the District of Columbia (excludes Alaska and Hawaii)
     Arsenic
   Beryllium
   Cadmium
   Chromium
 Manganese
    Mercury
       Nickel
  Total POM*
       7-PAH
                          500
                                     1,000         1,500
                                          Annual emissions
                                                              2,000
                                                                          2,500
                                                                                      3,000
  ' Total national emissions of POM are
   estimated at 17,300 tons per year.
Onroad Mobile
Nonroad Mobile
Major stationary sources
Stationary area sources
         Figure 2.  Contribution of Mobile Source Emissions to Total Emissions
                             of HAPs in the Particulate Phase
             Contiguous 48 States and the District of Columbia (excludes Alaska and Hawaii)

-------
These seven compounds represent a subset of total POM. Diesel PM was not included in this
projection, but it is one of the HAPs of greatest concern.  The EPA projected large reductions in
exposure to diesel PM over time in its recent 2007 heavy duty rulemaking.3

       We projected 2007 emissions from mobile sources for the 15 HAPs (with separate
estimates for POM and 7-PAH). We also made projections for seven gaseous HAPs which are
not on the list of 33, but were of interest to EPA's Office of Transportation and Air Quality
(OTAQ). These seven are ethylbenzene, hexane,  methyl tert-butyl ether (MTBE),
propionaldehyde, styrene, toluene, and xylenes.

       In addition to the HAPs that are emitted directly, many gaseous compounds emitted by
mobile sources can be converted to HAPs by reactions in the ambient air. For instance, 1-octene,
which is present in gasoline, can react in the atmosphere to produce formaldehyde.  Some HAP
precursors are HAPs themselves (for example, acetaldehyde and MTBE are precursors to
formaldehdye) but most are non-HAP volatile organic compounds (VOC). We projected mobile
source emissions of precursors pollutants that react to produce formaldehyde, acetaldehyde and
propionaldehyde, as these have been identified as HAPs for which secondary formation may
account for a significant portion of ambient concentrations.4  Table 1 gives a list of the HAPs that
are projected, including both HAPs that are directly emitted and HAPs that formed indirectly in
the atmosphere from mobile source VOC emissions.

       The mobile source HAP projections take into account all regulatory programs which are
projected to impact on HAP emissions in 2007, including EPA's reformulated gasoline program,
the national low emission vehicle program, the Tier 2 motor vehicle emissions standards, and
gasoline sulfur control requirements.  The recently proposed heavy-duty engine and vehicle
standards have not been included, since their impacts on HAP emissions will  by realized after
2007. These mobile source standards were put in place primarily to reduce emissions of criteria
pollutants;d however, they have reduced and will continue to reduce emissions of HAPs
significantly.
       dCriteria pollutants are regulated under National Ambient Air Quality Standards
(NAAQS).

-------
                 Table 1. HAPs Included in Mobile Source Projections
                                            Direct emissions are     Emissions of precursor
                  HAP                           projected          pollutants are projected
Gaseous HAPs on the NATA list of 33
       Acetaldehyde                                  S                      S
       Acrolein                                      S                      S
       Benzene                                      S
       1,3-Butadiene                                  S
       Formaldehyde                                  S                      S
Particulate HAPs on the NATA list of 33
       Arsenic                                       S
       Beryllium                                     S
       Cadmium                                     S
       Chromium                                    S
       Dioxins/furans                                 S
       Lead                                         S
       Manganese                                    S
       Mercury                                      S
       Nickel                                       /
       POM, total                                    /
       7-PAH                                       /
Gaseous HAPs not on the NATA list of 33
       Ethylbenzene                                  S
       Hexane                                       S
       Methyl tert-butyl ether (MTBE)                  /
       Propionaldehyde                               S                      S
       Styrene                                       S
       Toluene                                       S
       Xylenes                                       S

-------
                                 2. Methodologies
       We prepared inventories of projected mobile source emissions in 2007 for input into the
Emission Modeling System for Hazardous Air Pollutants (EMS-HAP)5 and the ASPEN dispersion
model. Two separate inventories were prepared: the first for HAPs that are directly emitted from
mobile sources, and the second for volatile organic compounds (VOC) pollutants that react to
form HAPs in the atmosphere (HAP precursors).  In both of these cases, we began with a 1996
base year inventory and projected the emissions to 2007. The base year emission data for directly
emitted HAPs were obtained from the February 2000 version of the 1996 National Toxics
Inventory (NTI).1  HAP precursor emission data for the base year were obtained from two
separate sources:  (1) non-HAP VOC came from Version 3 of 1996 National Emissions Trends
(NET)6 inventory,  speciated for specific organic compounds; (2) data for HAPs that are
precursors to other HAPs came from the 1996 NTI (same version as  specified above). The next
subsections provide more details on the sources of 1996 emissions data, how we projected the
1996 emissions to  2007, and how we prepared the data for EMS-HAP and ASPEN.

2.1  Emission Inventories for 1996

       2.1.1 1996 National Toxics Inventory (NTI)

       We obtained the mobile source inventory as a flat text file containing information on all 50
States and the District of Columbia. However, emissions were projected only for the continental
U.S. (not Alaska and Hawaii) and the District of Columbia. The  1996 NTI mobile source
documentation is in four volumes:7

       •      Documentation for the 1996 Base Year National Toxics Inventory for Aircraft
              Sources
       •      Documentation for the 1996 Base Year National Toxics Inventory for Commercial
             Marine Vessel and Locomotive Mobile Sources
       •      Documentation for the 1996 Base Year National Toxics Inventory for Nonroad
             Vehicle and Equipment Mobile Sources
       •      Documentation for the 1996 Base Year National Toxics Inventory for Onroad
              Sources

These can be accessed on the EPA web site at  http://www.epa.gov/ttn/chief/ei_guide.html#toxic.

       2.1.2 1996 National Emission Trends (NET) inventory, speciated for particular VOC

       Table 2 provides a list of the precursors for the HAPs that are secondarily formed.  The
majority of these precursors (33 out of 36) are non-HAP species, and are therefore not in the
1996 NTI. We extracted mobile source emission data for the non-HAP VOC species resulting
from a speciation of the Version 3 1996 NET inventory. We received this  data for the continental
U.S. and the District of Columbia. The methods we used to speciate the VOC data from the NET

-------
into the needed non-HAP precursor species are detailed in Appendix D, Section D. 1.2 of the
EMS-HAP User's Guide.52
             Table 2. Species Used for Modeling Secondary HAP Formation

                                      HAP formed from precursor species
Precursor
ethene
propene
1-butene
1-pentene
1-hexene
1-heptene
1-octene
1-nonene
1-decene
isobutene (2methylpropene)
2-methyl- 1 -butene
3 -methyl- 1 -butene
3 -methyl- 1 -pentene
2,3 -dimethyl- 1 -butene
isoprene
2-ethy 1-1 -butene
2-methyl- 1 -pentene
4-methyl- 1 -pentene
2,4,4-trimethyl- 1 -pentene
2-butene
2-pentene
2-hexene
2-heptene
2-octene
2-nonene
2-methyl-2-butene
3 -methyl-2 -pentene
4-methyl-2 -pentene
ethanol
3-hexene
1,3 -butadiene *
acetaldehyde *
MTBE*
methanol *
formaldehyde
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X











X
X
X
X
acetaldehyde

X

















X
X
X
X
X
X
X
X
X
X





propionaldehyde


X

















X








X




          * Precursors denoted with asterisks are also HAPs themselves.

-------
2.2  Projected Emissions for 2007

       2.2.1 Onroad emissions of gaseous HAPs

             2.2.1.1 Benzene,  acetaldehyde, formaldehyde, butadiene, andMTBE

       We estimated emissions of benzene, acetaldehyde, formaldehyde,  1,3-butadiene, and
MTBE in 2007 using projected emissions developed by the EPA OTAQ, along with projections
of future vehicle miles traveled in the EPA NET. OTAQ projected county-level emissions in 2007
for the above HAPs from eight vehicle types: light duty gasoline vehicles (LDGV), light duty
gasoline truck (LDGT) categories 1 and 2, heavy duty gasoline vehicles (HDGV), motorcycles
(MC), light duty diesel vehicles (LDDV), light duty diesel trucks (LDDT), and heavy duty diesel
vehicles (HDDV). This projection assumed a 30 ppm cap on sulfur in gasoline nationwide, and
implementation of exhaust emission standards for light duty vehicles recently finalized in the
Tier 2 mobile source rulemaking. It did not take into account potential phase-out of MTBE usage
in the reformulated gasoline program.8

       Emission factors were calculated using a toxic emission factor model, MOBTOXSb. This
model is based on a modified version of MOBILESb, which estimates emissions of regulated
pollutants, and essentially applies toxic fractions to total organic gas (TOG) estimates.  The TOG
basic emission rates used in this modeling incorporated available elements from MOBILE6 used
to develop the VOC  inventory  for the Tier 2 final rule.  The model accounted for differences in
toxic fractions between technology groups, driving cycles, and normal versus high emitting
vehicles and engines ("high emitters"). Impacts of fuel formulations were also addressed in the
modeling.

       OTAQ modeled toxic emissions for 10 urban areas and 16 geographic regions.  These
areas were selected to encompass a broad range of inspection and maintenance (I/M) programs,
fuel parameters, and temperature regimes.  The intent of the selection process was to best
characterize the different combinations needed to perform accurate nationwide toxic emissions
estimates.  Every county in the country was then "mapped" to one of these modeled areas or
regions (i.e., the emission factor  for the modeled area was also used for the area "mapped" to it).
OTAQ then multiplied the resulting county level emission factors by county-level estimates of
total VMT from EPA's Emission Trends Database.  VMT were apportioned among the 8 vehicle
types using national  average fractions (the same in every county).  The following VMT fractions
were used in the OTAQ projections:

       LDGV-      0.395
       LDGT1 -    0.383
       LDGT2-    0.127
       HDGV-     0.023
       LDDV-      0.000
       LDDT -      0.002
       HDDV-     0.065
       MC -        0.005

-------
       We used the county-level VMT fractions developed for the recently finalized 2007 heavy
duty rule with two changes.  First, we consolidated the two LDGT categories into a single vehicle
category in order to maintain consistency with the 1996 NTI.  Second, we used county-specific
estimates for the distribution of VMT among different vehicle categories, in place of the national
average values used in the OTAQ projection.  These county-specific VMT distributions were
obtained from the NET. We used the following equation:

                 _                           V M11 cnty i; veh
       cnty i, veh. type j      OTAQ, cnty i, veh. type j
                                           VM11 QTAQ-natl, veh. type j
where:
             E Cnty;, veh. type j =  projected 2007 onroad emissions of a given pollutant for vehicle
                            type j in county i

        EOTAQ, cnty;, veh. type j =  OTAQ's 2007 projected onroad emissions of the given pollutant
                            for vehicle type j in county i

        VMTf cntyijVeh t^ j =  fraction of VMT for vehicle type j in county i (based on county-
                            level VMT data in the NET)

     VMTf oxAQ-nati, veh. type j =  national fraction of VMT for vehicle type j used in OTAQ
                            estimates (see list above)

Light duty gasoline trucks were combined prior to this projection, so the calculation of LDGT
emissions can be expressed as follows:

                                                       VMTf cnty i, LDGT1 + VMTf cnty i, LDGT2
 F        = F F              +F              Ix  	
 ^ cnty i, LDGT  I ^ OTAQ, cnty i, LDGT1   ^ OTAQ, cnty i, LDGT2 J
                                                    VM 11 OTAQ natl, LDGT1 + VM 11 OTAQ natl, LDGT2

       The three inventories used in the above equations (NTI, OTAQ, and NET for VMT) did
not always mesh at the county-level because of differences in the allocation of national and state-
level emissions to counties. In particular, NET VMT fractions could not be calculated for some
independent cities in Virginia and for some counties in Arizona and New Mexico.  In the Virginia
cities, VMT fractions from the surrounding county were used. In cases where county-level  VMT
fractions could not be computed, default state-level VMT fractions were used. The NTI also
reported no HAP emissions in 1996 from motorcycles in five counties where OTAQ estimated a
small amount of emissions in 2007 (0.005 tons total for all five counties).  Emissions from
motorcycles in 2007 were set at zero in these five cases.

              2.2.1.2 Other gaseous organic HAPs

       We used county-level predictions of VOC emissions to project future emissions of
gaseous organic HAPs that are produced by onroad vehicles but are not explicitly addressed by
MOBTOXSb. This approach was used for acrolein, ethyl benzene, hexane, propionaldehyde,

-------
styrene, toluene, and xylenes. The EPA Emission Factor and Inventory Group (EFIG) projected
onroad VOC emissions for 2007 under the NET emission inventory program.9  These projections
are based on the regulatory impact analysis for the heavy duty engine rulemaking,10 and include all
final and proposed mobile source rules except heavy duty vehicle standards that will begin to take
effect in the 2007 model year. (The impact of these standards in 2007 will be small.)

       For the gaseous HAPs not covered by MOBTOXSb, we have assumed that emissions are
proportional to total VOC emissions for a given vehicle class and set of operating conditions.
Therefore these HAPs are assumed to change by the same ratio as VOC between 1996 and 2007.
However, this ratio will depend  on the vehicle class and the county, since operating conditions
change from county to county.  The following equation was used:

                                             p
                 _                           -'-'2007 VOC, cntyi, veh. type j
     •'-' cnty i, veh. type j ~~  ^1996 NTT, cnty i, veh. type j       ~
                                             E 1996 VOC, cnty i, veh. type j

where:
            E cntyi, veh. type j =  projected 2007  onroad emissions of a given pollutant for vehicle
                           type j in county i

     E1996 NTT, cntyi, veh. type j =  1996 NTI onroad emission estimate of the given pollutant for
                           vehicle type j in county i

     E 2007 voc, cnty i, veh. type j =  VOC onroad emissions estimate for vehicle type j in county i from
                           the 2007 projected VOC inventory

     E1996 voc, cnty i, veh. type j =  VOC onroad emissions estimate for vehicle type j in county i from
                           the 1996 VOC inventory

Light duty gasoline trucks (LDGT1 and LDGT2) were combined in this calculation, as follows:

                                     F                + F
 -_          _                      ^ 2007 VOC, cnty i, LDGT1 ^ ^ 2007 VOC, cnty i, LDGT2
 rH        — L                 V  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 J-' cnty i, LDGT  J-' 1996 NTT, cnty i, veh. type j
                                     E 1996 VOC, cnty i, LDGT1  E i996 VOC, cnty i, LDGT2

       NET VOC scaling factors could not be calculated for some  independent cities in Virginia
and for some counties in Arizona and New Mexico. In the Virginia cities, VOC scaling factors
from the surrounding county were used. In cases where county-level VOC factors could not be
computed, default state-level VOC factors were used.

       2.2.2 Nonroad emissions of gaseous HAPs

       We projected nonroad emissions of gaseous HAPs based on NET predictions of VOC
emission reductions, using a similar approach to the one described for other gaseous HAPs from
onroad vehicles (section 2.2.1.2). The 1996 NTI contains HAP emissions estimates for six broad

-------
nonroad mobile source categories, which correspond roughly to 4-digit area and mobile source
(AMS) emissions inventory codes:

       Off-highway 2-stroke gasoline engines  -    AMS 2260000000
       Off-highway 4-stroke gasoline engines -    AMS 2265000000
       Off-highway diesel engines -               AMS 2270000000
       Aircraft -                                AMS 2275000000
       Commercial marine vessels -               AMS 2280000000
       Railroad diesel engines -                  AMS 2285002000

       We have assumed that, within a given county, HAP emissions in each of the above broad
categories will change by the same ratio as VOC between 1996 and 2007.  EPA/EFIG projected
county-level nonroad VOC emissions in 2007 under the NET emission inventory program.9
These projections were developed using an updated version of EPA/OTAQ's NONROAD
emission model,11 which was developed to support the RIA for the recent heavy duty engine
rules.10 The 2007 VOC projections also include OTAQ instructions regarding commercial marine,
locomotives and commercial aircraft. The 2007 VOC inventory divides emissions into a more
detailed list of emission source categories than the above list (about 60 categories in all). We
linked the detailed NET emission categories to the broader NTI categories, excluding those VOC
emission categories which do not emit significant quantities of HAPs (such as natural gas
combustion engines).

       The following equation was used to project county -level emissions of specific HAPs for
each of the broad nonroad engine categories listed above:
                _                            -"^ 2007 VOC, cnty i, category j
       cnty i, category j       1996 NTI, cnty i, categor
                                  'goryj        T-,
                                             •'-' 1996 VOC, cnty i, category j

where:

            E Cnty;, category j =   projected 2007 nonroad emissions of a given pollutant from
                           category j in county i

     E1996 NTI, cnty i, category j =   1996 NTI nonroad emission estimate of the given pollutant for
                           category j in county i

     E 2007 voc, cnty i, category j =   VOC nonroad emissions estimate for category j in county i from
                           the 2007 projected VOC inventory

     E1996 voc, cnty i, category j =   VOC nonroad emissions estimate for category j in county i from
                           the 1996 VOC inventory

County-specific growth factors could not be estimated for marine vessel emissions in nine
counties.  In these cases, a national average growth factor (1.097) was used.
                                           10

-------
       2.2.3  Onroad emissions of particulate HAPs

       Particulate HAP emissions from onroad vehicles consist of POM, dioxins and furans, and
various HAP metals, including arsenic, chromium, lead, manganese, mercury, and nickel.
Different methodologies were used for POM than for the other HAPs, as discussed in the
following two subsections.

             2.2.3.1 Poly cyclic organic matter

       A number of measures are being implemented between 1996 and 2007 to control fine
particulate matter (PM2 5) emissions from onroad vehicles.  EPA/EFIG has projected county-level
onroad PM2 5 emissions in 2007 under the NET emission inventory program.9  These projections
were developed to support the RIA for the recent heavy duty engine rules.10 There is generally a
reasonable correlation between POM and total carbon emissions.12 Therefore, emissions of POM
were assumed to be reduced by the same ratio as PM2 5. The following equation was used:

                _                          PM2. J 2007, cnty i, veh. type j
    •'-' cnty i, veh. type j ~~  ^1996 NTT, cnty i, veh. type j       __ ^_
                                            PM2.51996]CntyiiV6

where:

           E cnty;, veh. type j =   projected 2007 onroad emissions of a given pollutant for vehicle
                           type j in county i

     E1996 NTT, cnty;, veh. type j =   1996 NTI onroad emission estimate of the given pollutant  for
                           vehicle type j in county i

   PM2.5 2007, cnty i, veh. type j =   ?M2 5 estimate for vehicle type j in county  i from the 2007
                           projected VOC inventory
   PM2.5 1996, cnty;, veh. type j =   PM2 5 estimate for vehicle typej in county i from the 1996 VOC
                           inventory

       In some cases, PM25 emissions in 1996 were too small to allow calculation of PM25
emission ratios. This occurred for motorcycles in 376 counties, for LDDT in 90 counties, and for
other vehicle categories in fewer than 10 counties.  In most of these cases, we used PM10 emission
ratios in place of PM2 5 ratios. In about  145 counties  for motorcycles and 8 counties for LDDT,
PM10 emissions in 1996 were also too small to compute emission ratios. Emissions of POM in
2007 were assumed to be the same as emissions in 1996 in these situations.

       Neither PM2 5 emissions nor PM10 emissions were available for some independent cities in
Virginia and for some counties in Arizona and New Mexico.  In the Virginia cities, PM2 5 factors
from the  surrounding county were used. In cases where county-level PM2 5 and PM10 emissions
were not available, default state-level VMT factors were used.
                                            11

-------
              2. 2. 3. 2  Trace metals and dioxins

       The emission controls coming online for onroad vehicles between 1996 and 2007 are not
expected to have a substantial impact on emissions of HAP metals, which are primarily due to
their presence in fuels and lubricants, or from engine wear. Diesel emissions of dioxins and furans
may be impacted by new diesel standards, but these impacts are uncertain. Therefore, we have
assumed that all HAP emissions from onroad vehicles will increase or decrease in proportion to
changes in vehicle usage between 1996 and 2007.

       HAP emissions from all gasoline vehicles and light duty diesel s were projected based on
county -level predictions of future vehicle miles traveled (VMT).  The following equation was
used:
                 _                                   o7; cnty j; veh. type j
     •'-' cnty i, veh. type j ~~  ^1996 NTT, cnty i, veh. type j
                                              VMT1996 cnty i veh type j

where:

            E Cnty;, veh. type j =  projected 2007 onroad emissions of a given pollutant for vehicle
                           type j in county i

     E1996 NTT, cnty i, veh. type j =  1996 NTI onroad emission estimate of the given pollutant for
                           vehicle type j in county i

     VMT 2007 cnty; veh typg j =  VMT estimate for vehicle type j in county i from the 2007
                           projected VOC inventory

     VMT 1996 cnty; veh jypg j =  VMT estimate for vehicle type j in county i from the 1996 VOC
                           inventory

       NET VMT factors could not be calculated for some independent cities in Virginia and for
some counties in Arizona and New Mexico.  In the Virginia cities, VMT factors from the
surrounding county were used.  In cases where county-level VMT factors could not be computed,
default state-level VMT factors were used.

       The above equation was not used for heavy duty diesel vehicles (HDDV). Instead, future
HAP emissions from HDDV were estimated from the predicted increase total energy output (in
horse-power-hours) at a national level. Total energy output for HDDV is projected to increase by
34% between 1996 and 2007. Thus, the following equation was used to project future emissions
of paniculate HAPs from HDDV:

      E Cnty i, HDDV ~~  E i996 NTI, cnty i, HDDV  X 1 • -> 4
                                            12

-------
where:
             E cnty; JJJ3JJV =  proj ected 2007 onroad emissions for a given pollutant for HDDV
                           in county i

         1996 NTI, cnty i, HDov =  1996 NTI onroad emission estimate for a given pollutant for
                           HDDV in county i

                   1.34=  Predicted increase in total horse-power-hours for HDDV between
                           1996 and 2007
       2.2.4 Non-road emissions of particulate HAPs

              2.2.4.1 Polycyclic organic matter

       Emissions of POM from nonroad sources are expected to be reduced by the same ratio as
PM2 5. EPA/EFIG has projected county-level onroad PM2 5 emissions in 2007 under the NET
emission inventory program.9  We used the PM25 emissions estimates to project emissions of
7-PAH and 16-PAH at the county level for the six broad categories of nonroad emissions in the
NTI (listed in section 2.2.2). The following equation was used:
                                                 > <;
     _              _                           '•-' 2007, cnty i, category j
     H           ==  H                 X   ^^^^^^^^^^^^^™^^^^^^^^^^^^^™
     J-^ cnty i, category j    J-^ 1996 NTI, cnty i, category j
                                             P\/| / S
                                             ± iv±z,. j 1996] cnty ^ category j
where:
            E cnty ; veh type j =  proj ected 2007 onroad emissions of a given pollutant for nonroad
                           category j in county i
     E 1996 NTT, cnty ;, veh. type j =   1996 NTI onroad emission estimate of the given pollutant for
                            nonroad category j in county i

   PM2. 5 2007 cnty ; veh ,ype j =   PM2 5 estimate for nonroad category j in county i from the 2007
                            projected VOC inventory
   PM2.5 1996, cnty ;, veh. type j =  PM2 5 estimate for nonroad category j in county i from the 1996
                           VOC inventory

       Some aircraft and marine vessel emissions estimates in the NTI could not be matched to
PM25 emissions in the NEI because of differences in the methodologies used for allocating
emissions from the state level to the county level in the two inventories.   These mismatches
occurred for for aircraft in about 500 counties and marine vessels in 10 counties. In these cases
default state-level PM2 5 ratios were used to project PAH emissions.
                                            13

-------
              2.2.4.2 Trace metals and dioxins

       EPA has not estimated the potential impacts on particulate HAP emissions for emission
controls coming online for nonroad engines between 1996 and 2007. Therefore, our projections
are based on the conservative assumption that particulate HAPs will increase or decrease in
proportion to changes in nonroad engine usage between 1996 and 2007. Projections were made
based on national average growth factors, using the following equation:
     F           =  F                  x GF
     ^cntyi, category j    J-' 1996 NTI, cnty i, category j    ^-Ji cati
                                             tegoryj
where:
             ' cnty i, veh. type j
      ' 1996 NTT, cnty i, veh. type j
              GFcat,
                   tegoryj
projected 2007 nonroad emissions of a given pollutant for category
j in county i

1996 NTI nonroad emission estimate of the given pollutant for
category j in county i

VMT estimate for vehicle type j in county i from the 2007
projected VOC inventory
       Table 3 gives the estimated national growth factors for the various nonroad emission
categories. Growth factors for the 2-stroke, 4-stroke, off-road diesel, and commercial marine
categories were calculated based on the predicted change in total energy usage (in horse-power-
hours) from the February 2000 version of the NONROAD model.11  For locomotives, fuel
consumption has remained static for many years and OTAQ projects it will continue to do so.

       Separate growth factors were used for lead emissions from aircraft. Lead emissions
emanate mainly from general aviation, therefore the growth factor for lead is based on the
projected increase in fuel usage in general aviation.
      Table 3. Growth Factors for Particulate HAP Emissions from Nonroad Engines
Emission category
2-stroke gasoline engines
4-stroke gasoline engines
Off highway diesel engines
Aircraft
Commercial marine vessels
Railroad diesel engines
AMS code HAP
2260000000
2265000000
2270000000
2275000000 Lead
2280000000
2285002000
Growth
factor
1.09
1.15
1.38
1.11
1.13
1.00
Basis
Energy output
Energy output
Energy output
Fuel consumption for general
Energy output




aviation

Fuel consumption, assumed constant
                                            14

-------
       2.2.5 Emissions of HAP precursors

       HAP precursors which are HAPs themselves were projected as described in Sections 2.2.1
and 2.2.2. Onroad and nonroad emissions of non-HAP precursors were also projected with an
approach similar to the approaches used for "other" gaseous HAPs.  We began with the 1996
NET inventory for VOC, speciated for individual VOC emissions. As discussed above,
EPA/EFIG has projected county-level onroad and nonroad VOC emissions in 2007 under the
NET emission inventory program.9 We assumed that, within a given county and for a given
emission source category, emissions of each precursor pollutant will change by the same ratio as
VOC between 1996 and 2007. Thus, we assume that engine design and emission control
technology do not substantially change the ratio of HAP to VOC for these "other" gaseous HAPs.
The following equation was used:
     F           =  F
     ^ cnty i, category j    J-^ 1996, cnty i, category j
                                             •p
                                             J-' 2007 VOC, cnty i, category j
                                             -^ 1996 VOC, cnty i, category j

where:

            E Cnty;, category j =  Pr°J ected 2007 onroad or nonroad emissions of a given pollutant
                           from category j in county i

        E1996, cnty;, category j =  1996 onroad or nonroad emission estimate of the given pollutant
                           for category j in county i

     E 2007 voc, cnty i, category j =  VOC onroad  or nonroad emissions estimate for category j in
                           county i from the 2007 projected VOC inventory

     E1996 voc, cnty i, category j =  VOC onroad  or nonroad emissions estimate for category j in
                           county i from the 1996 VOC inventory

       One difference between the projection of VOC precursors and the HAP projections
described above is that there was no need to aggregate VOC emissions into the categories used in
the NTI for HAPs. Rather, each emission source category included in the NET was projected
individually.  Thus, onroad emissions were projected separately for LDGT1 and LDGT2.  In
addition, nonroad emissions were generally projected at the 7-digit AMS level, allowing the
differentiation treatment for a much wider array  of engines (for instance, 2-stroke recreational
vehicles are distinguished from 2-stroke construction equipment).  VOC emissions for 2-stroke
and 4-stroke recreational vehicles in  five Alabama counties were too low for the calculation of
detailed scaling factors (at the 7-digit AMS level).  In these cases, average VOC factors were
applied based on the total emissions  from all 2-stroke engines and all  4-stroke engines (the 4-digit
AMS level).
                                            15

-------
2.3 Spatial and Temporal Processing of Emissions

       We used EMS-HAP to prepare the projected emissions inventories for dispersion
modeling. EMS-HAP performs the following operations for mobile sources:

       •      Allocates the county-level estimates of aircraft emissions (AMS codes beginning
              with 227500 or 227505) from the mobile source inventory to the actual locations
              of commercial and civil airports in each county, and assigns point source variables
              such as geographic coordinates and stack parameters to each allocated aircraft
              emission record.  (See Chapter 2 of the EMS-HAP User's Guide.5) The
              parameters we used for airports were: IVENT=1 (nonstacked emission point
              which means that the ASPEN model performs no plume rise calculations), stack
              height = 5 meters, stack diameter = 1 meter, stack temperature = 395 K, stack
              velocity = 0.5 meters/seconds (EMS-HAP guide, Section 3.1.2).

       •      Appends the aircraft emissions estimates to  the point sources inventory or
              (depending on user options) creates a separate point source file containing
              allocated aircraft emissions. We ran the allocated aircraft emissions from the NTI
              as a separate point source file through EMS-HAP. We ran the allocated precursor
              emissions together with the other point source precursor emissions through
              EMS-HAP.

       •      Spatially allocates all mobile source county level emissions (which includes all
              mobile sources other than the air craft emissions that were allocated to point
              sources) to Census tracts (a typical Census  tract covers about 4000 people) using
              spatial surrogates (See EMS-HAP guide, Section 10.1.2 and Appendix D.8.5)

       •      Temporally allocates annual average emissions to eight 3-hour periods in a typical
              day (midnight to 3 AM, 3 AM to 6 AM, etc.). Each day is assumed to have the
              same total emissions. (No seasonal or day of week variation is taken  into
              account.) (Chapter 5 of the EMS-HAP guide discusses temporal allocation
              methods for aircraft modeled as point sources, and Section  10.1.3 gives
              methodologies other mobile sources. Appendix D.7 of the EMS-HAP guide
              presents the temporal allocation factors used for mobile sources.5)

       Figure 3 gives an overview of EMS-HAP processing for mobile sources, including the
processing of airports as point sources. Figure 4 provides more detail on the allocation of county-
level airport emissions to specific locations, and Figure 5 provides more detail on the processing
of other mobile source emissions.  The User's Guide for EMS-HAP discusses these operations in
detail.5  The options used in processing the 2007 projection inventory are the same as those used
in carrying out the  1996 national assessment, as documented in Appendix D of the EMS-HAP
User's Guide.
                                           16

-------
2.4 Ambient Concentration Modeling

       Ambient impacts of mobile source emissions were predicted using the ASPEN dispersion
model.2'4 This model is based on the EPA's Industrial Source Complex Long Term model
(ISCLT2) which simulates the behavior of the pollutants after they are emitted into the
atmosphere.  ASPEN uses emissions data and meteorological data to predict the ambient
concentrations of HAPs for each Census tract in the contiguous U.S. (including the District of
Columbia), and the Virgin Islands and Puerto Rico.  For the 2007 mobile source projections,
ambient impacts were estimated only for the contiguous U.S.
                                          17

-------
  Point Source Emissions
          File
Mobile Source Emissions
          File
AirportProc


            I
          1
   Point Source Emissions
  File (including allocated
   aircraft emissions data)
Mobile Source Emissions
File (excluding allocated
 aircraft emissions data)
       PtDataProc
     MobilePreproc
      PtAspenProc
        AMProc
       PtTemporal
                                ASPEN Mobile Source
                                   Emissions Files
       PtGrowCntl
      PtFinalFormat
   ASPEN Point Source
     Emissions Files
Figure 3. Overview of EMS-ELj^P Processing of Mobile Sources

-------
  Batch File: Program Control
    Options and Keywords
! Mobile Source Emissions File
                                               Reads Program Control Options and
                                                           Keywords
                                               Reads Mobile Source Inventory and
                                                Extract Aircraft Emission Records
       Mobile Source Inventory File
        without Aircraft Emissions
                                                   Allocates County-Level Aircraft
                                               Emissions to Specific Locations within
                                                               County
Unallocated Aircraft
Emissions Records

Allocated Aircraft
Emissions Records
      Appends Unallocated Aircraft
   Emission Records to Mobile Source
  Inventory File or Create Separate File
     (depending on program control
                option)
                                            Creates Variables Required for Processing
                                               Aircraft Emissions as Point Source
                                                  Emissions through EMS-HAP
r
              OR
                                             1
Mobile Source Inventory File with
 Unallocated Aircraft Emissions
Unallocated Aircraft
  Emissions File
  Point Source Inventory File
                                              Appends Allocated Aircraft Emission
                                            Records to Point Source Inventory File or
                                               Create Separate File (depending on
                                                    program control option)
                                                                   OR
                 Point Source Inventory File with
                   Allocated Aircraft Emissions
                                                               Allocated Aircraft
                                                                 Emissions File
                     Figure 4. Processing of Airports in EMS-HAP
                                           19

-------
  Batch File Containing Keywords
   e.g. File Names and Locations,   •	*\              Reads Keywords
         Program Options



;   Emissions Inventory File   j	H      Reads and Summarize Emissions

     r	
     !    Emissions Summaries    f*	


         HAP Table        I	*            Pollutant Processing
                                            (selection, grouping, and partitioning)


j  ASPEN Source Group File  j	*•    Assignment of ASPEN Source Groups


,	
  AMS-to-Spatial Surrogate
                                           Spatial Allocation of County Emissions to
                                                       Census Tracts
  Spatial Allocation Factors



     Temporal Allocation    i	  Temporal Allocation of Annual Emissions to
          Factors                                     3-Hour Periods



          Emissions Summaries    t*	       Produces Emissions Summaries


                                 ,       ^^^_^^^_
         ASPEN Emissions Files   1*	       Writes ASPEN Emissions Files
            Figure 5. Processing of Other Mobile Sources in EMS-HAP
                                        20

-------
                                      3.  Results
       The following sections present the results of HAP emission projections for mobile sources
in 2007, and the results of dispersion modeling using the 2007 emission projections.

4.1  Projected Emissions in 2007

       Table 4 summarizes the results of mobile source HAP emission projections.  Results are
provided for gaseous HAPs, particulate HAPs, and HAP precursors. The table gives estimated
emissions in the 1996 NTI, projected emissions in 2007, and the resulting change in emissions.
Separate estimates are given for onroad and nonroad emissions. Separate estimates are also given
for each of the HAPs that are directly emitted, but precursor emissions are lumped together into
an aggregate estimate for each HAP that is produced. The lumped precursor emissions shown in
the table are a straight sum of each of the precursors (nonHAP and HAP) that form that particular
HAP; these sums do not account for the different reactivities or molar yields of the individual
precursors, and therefore do not reflect the reactivity and molar yield-weighted precursor mass
that is input into the ASPEN model. It should also be noted that not all of the precursor
emissions will be converted to HAPs in the atmosphere.

       As Table 4 shows, reductions of 24 to 50% are projected for all of the gaseous HAPs,
both for onroad vehicles  and nonroad engines. Metal emissions are projected to increase by 5 to
36% from 1996, but these emissions are three orders of magnitude lower than the projected
emissions of gaseous HAPs. Emissions of HAP precursors are projected to decrease substantially
for both onroad and nonroad sources.

       Tables 5 through  7 give national projections of onroad emissions from the different vehicle
types for gaseous HAPs,  particulate HAPs, and HAP precursors, respectively.  Each table gives
estimated emissions in 1996, projected emissions in 2007, and the resulting change in emissions.
For comparison, the projected changes in VMT  are also shown for the different vehicle
categories.

       Table 5 shows that emissions of gaseous HAPs are projected to decrease substantially
despite increases in overall VMT. The projected reductions in gaseous HAP emissions are driven
mainly by the projected reductions for LDGV and LDGT, although emission reductions are
projected for all of the categories except MC. LDGT emissions of gaseous HAPs are projected to
decrease by 12 to 60% despite an increase of over 100% in VMT. LDDV usage is expected to be
phased out by 2007.

       As shown in Table 6, the projected changes in particulate HAP emissions match projected
changes in VMT, with the exception of heavy duty categories.  In these categories, particulate
HAP emissions are expected to increase in proportion to total power usage (in horse-power-
hours).  The phaseout of LDDV results in reductions in overall emissions for arsenic and mercury.
However, this result stems mainly from differences between the speciation factors used for LDDV
                                           21

-------
  Table 4. Summary of Mobile Source Emission Projection Results for the Contiguous 48 States and the District of Columbia
                                                      (excludes Alaska and Hawaii)
Estimated 1996 emissions

Gaseous HAPs
Acetaldehyde
Acrolein
Benzene
1,3 -Butadiene
Ethyl benzene
Formaldehyde
Hexane
MTBE
Propionaldehyde
Styrene
Toluene
Xylenes
Total gaseous HAPs
Paniculate HAPs
Total POM
7-PAH
Arsenic
Beryllium
Cadmium
Chromium
Dioxins/Furans
Lead
Manganese
Mercury
Nickel
Total particulate HAPs
Precursors
Acetaldehyde precursors a
Acrolein precursors a
Formaldehyde precursors a
Propionaldehyde precursors a
Onroad

28,567
4,932
167,488
23,393
80,313
82,663
62,854
65,106
5,478
16,227
546,536
309,117
1,392,674

90.1
41.5
0.3
0
0
13.4
0.00013
18.9
5.8
0.2
10.6
139

308,307
23,393
768,247
77,489
Nonroad

40,126
7,226
91,330
9,242
56,493
85,030
39,129
53,936
6,819
3,230
228,611
234,043
855,215

40.4
17.0
1.9
0.02
0.26
34.0
0.0
527.2
34.9
6.6
87.9
733

152,458
9,242
535,136
38,420
(tons/year)
All mobile

68,692
12,159
258,818
32,635
136,805
167,693
101,983
119,043
12,297
19,457
775,147
543,160
2,247,889

130.5
58.5
2.2
0.02
0.26
47.4
0.00013
546.1
40.7
6.8
98.5
872

460,765
32,635
1,303,383
115,909
Proiected 2007 emissions (tons/year)
Onroad

16,540
3,008
83,545
11,203
44,046
41,718
34,921
25,092
3,423
8,995
299,656
169,481
741,628

59.5
27.4
0.1
0.0
0.0
16.2
0.00017
22.0
7.0
0.1
13.0
100

191,847
11,203
465,635
48,157
Nonroad

25,000
5,019
63,515
6,832
40,931
54,483
28,126
31,846
4,481
2,464
165,908
171,076
599,681

44.1
18.1
2.1
0.02
0.3
43.6
0.0
585.2
44.7
9.0
101.5
831

103,679
6,832
382,155
27,156
All mobile

41,539
8,026
147,060
18,035
84,977
96,201
63,048
56,939
7,904
11,459
465,564
340,557
1,341,309

103.6
45.5
2.3
0.0
0.3
59.8
0.00017
607.2
51.7
9.1
114.5
930

295,526
18,035
847,791
75,313
Change in emissions (%)
Onroad

-42
-39
-50
-52
-45
-50
-44
-61
-38
-45
-45
-45
-47

-54
-54
-55
na
na
21
36
17
21
-21
23
-28

-38
-52
-39
-38
Nonroad

-38
-31
-30
-26
-28
-36
-28
-41
-34
-24
-27
-27
-30

9
6
13
13
13
28
na
11
28
36
15
13

-32
-26
-29
-29
All mobile

-40
-34
-43
-45
-38
-43
-38
-52
-36
-41
-40
-37
-40

-34
-41
5
13
13
26
36
11
27
34
16
7

-36
-45
-35
-35
aHAP precursor emissions estimates include some precursors that are themselves HAPs. Not all of the mass of precursor emissions is converted to HAPs.
na = not applicable, emissions were 0 or not reported in 1996

-------
   Table 5.  Gaseous HAP Emission Projections for Different Onroad Vehicle Categories
              Contiguous 48 States and the District of Columbia (excludes Alaska and Hawaii)
HAP
Estimated 1996 emissions
Acetaldehyde
Acrolein
Benzene
1,3 -Butadiene
Ethyl benzene
Formaldehyde
Hexane
MTBE
Propionaldehyde
Styrene
Toluene
Xylenes
Projected 2007 emissions
Acetaldehyde
Acrolein
Benzene
1,3 -Butadiene
Ethyl benzene
Formaldehyde
Hexane
MTBE
Propionaldehyde
Styrene
Toluene
Xylenes
Projected change (%)
Acetaldehyde
Acrolein
Benzene
1,3 -Butadiene
Ethyl benzene
Formaldehyde
Hexane
MTBE
Propionaldehyde
Styrene
Toluene
Xylenes
Change in VMT (%)
LDGV
(tons)
10,011
1,991
87,768
10,682
51,640
26,426
40,166
36,665
1,955
10,428
352,924
199,049
(tons)
2,631
738
27,187
3,044
19,662
5,394
15,609
9,127
728
3,957
134,498
75,733

-74
-63
-69
-72
-62
-80
-61
-75
-63
-62
-62
-62
-14
LDGT

8,534
962
66,010
8,822
24,492
24,816
17,683
24,502
952
5,100
168,785
95,150

4,643
832
45,201
5,444
21,582
9,961
15,790
12,902
825
4,486
148,910
83,860

-46
-14
-32
-38
-12
-60
-11
-47
-13
-12
-12
-12
104
HDGV

1,912
906
9,462
1,756
3,541
9,069
3,301
3,077
122
50
23,663
13,347

1,180
577
7,122
687
2,279
4,224
2,150
2,310
78
30
15,245
8,587

-38
-36
-25
-61
-36
-53
-35
-25
-36
-40
-36
-36
44
MC

201
1
936
281
27
736
19
862
1
6
184
105

280
2
1,161
363
32
933
22
754
1
7
217
124

39
18
24
29
18
27
18
-13
18
18
18
18
44
LDDV

224
58
364
164
33
702
91
0
304
35
53
79

0
0
0
0
0
0
0
0
0
0
0
0

-100
-100
-100
-100
-100
-100
-100
na
-100
-100
-100
-100
-100
LDDT

115
19
187
84
11
362
30
0
101
12
18
26

21
10
35
16
6
67
16
0
53
6
9
14

-82
-48
-82
-82
-48
-82
-48
na
-48
-48
-48
-48
-6
HDDV

7,570
995
2,760
1,604
569
20,554
1,564
0
2,043
597
910
1,362

7,785
850
2,839
1,649
485
21,140
1,335
0
1,738
510
111
1,163

3
-15
3
3
-15
3
-15
na
-15
-15
-15
-15
44
Total
onroad

28,567
4,932
167,488
23,393
80,313
82,663
62,854
65,106
5,478
16,227
546,536
309,117

16,540
3,008
83,545
11,203
44,046
41,718
34,921
25,092
3,423
8,995
299,656
169,481

-42
-39
-50
-52
-45
-50
-44
-61
-38
-45
-45
-45
27
aLDDV usage is expected to be phased out by 2007.
na = not applicable, emissions were 0 or not reported in 1996
Onroad vehicle categories:
   LDGV - Light duty gasoline vehicles               LDDV - Light duty diesel vehicles
   LDGT - Light duty gasoline trucks                 LDDT - Light duty diesel trucks
   HDGV - Heavy duty gasoline vehicles              HDDV - Heavy duty diesel vehicles
   MC - Motorcycles
                                                23

-------
 Table 6. Particulate HAP Emission Projections for Different Onroad Vehicle Categories
              Contiguous 48 States and the District of Columbia (excludes Alaska and Hawaii)
HAP LDGV
Estimated 1996 emissions
Total POM
7-PAH
Arsenic
Chromium
Dioxins/Furans
Lead
Manganese
Mercury
Nickel
Projected 2007 emissions
Total POM
7-PAH
Arsenic
Chromium
Dioxins/Furans
Lead
Manganese
Mercury
Nickel
Projected change (%)
Total POM
7-PAH
Arsenic
Chromium
Dioxins/Furans
Lead
Manganese
Mercury
Nickel
Change in VMT (%)
(tons)
54.2
24.6
0.0
8.5
0.0
13.9
2.8
0.0
6.2
(tons)
28.8
13.1
0.0
7.3
0.0
11.9
2.4
0.0
5.3
-47
-47
na
-14
na
-14
-14
na
-14
-14
LDGT
23.8
10.9
0.0
3.6
0.0
5.0
1.2
0.0
2.6
23.2
10.6
0.0
7.4
0.0
10.1
2.5
0.0
5.4
-2
-2
na
104
na
104
104
na
104
104
HDGV
3.2
1.6
0.0
0.5
0.0
0.0
1.1
0.1
0.6
2.1
0.9
0.0
0.8
0.0
0.0
1.6
0.1
0.9
-35
-39
na
44
na
na
44
44
44
44
MC
0.62
0.28
0.00
0.00
0.00
0.00
0.00
0.00
0.00
1.56
0.70
0.00
0.00
0.00
0.00
0.00
0.00
0.00
152
149
na
na
na
na
na
na
na
44
LDDVa
0.33
0.15
0.15
0.15
0.00
0.00
0.23
0.08
0.08
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
-100
-100
-100
-100
-100
na
-100
-100
-100
-100
LDDT
0.36
0.10
0.04
0.04
0.00
0.00
0.06
0.02
0.02
0.11
0.03
0.04
0.04
0.00
0.00
0.06
0.02
0.02
-70
-70
-6
-6
na
na
-6
-6
-6
-6
HDDV
7.6
4.1
0.1
0.5
0.00012
0.0
0.3
0.0
1.1
3.7
2.0
0.1
0.7
0.00017
0.0
0.4
0.0
1.4
-51
-50
34
34
38
na
34
na
34
44
Total
onroad
90.1
41.6
0.3
13.4
0.0
18.9
5.8
0.2
10.6
59.5
27.4
0.1
16.2
0.0
22.0
7.0
0.1
13.0
-34
-34
-55
21
na
17
21
-21
23
27
aLDDV usage is expected to be phased out by 2007.
na = not applicable, emissions were 0 or not reported in 1996
                                              24

-------
  Table 7. HAP Precursor Emission Projections for Different Onroad Vehicle Categories
              Contiguous 48 States and the District of Columbia (excludes Alaska and Hawaii)
HAP produced
LDGV
LDGT
HDGV
MC
LDDVa
LDDT
HDDV
Total
onroad
Estimated 1996 emissions (tons)
Acetaldehyde
Acrolein
Formaldehyde
Propionaldehyde
156,959
10,682
372,220
39,589
111,274
8,822
274,481
28,001
17,558
1,756
36,985
4,463
2,306
281
4,820
590
924
164
3,587
222
411
84
1,623
99
18,874
1,604
74,530
4,527
308,307
23,393
768,247
77,489
Projected 2007 emissions (tons)
Acetaldehyde
Acrolein
Formaldehyde
Propionaldehyde
Projected change (%)
Acetaldehyde
Acrolein
Formaldehyde
Propionaldehyde
Change in VMT (%)
62,080
3,044
139,631
15,656

-60
-72
-62
-60
-14
33,879
5,444
93,524
8,527

-70
-38
-66
-70
104
11,459
687
23,793
2,914

-35
-61
-36
-35
44
2,993
363
5,868
766

30
29
22
30
44
0
0
0
0

-100
-100
-100
-100
-100
218
16
793
52

-47
-82
-51
-47
-6
16,145
1,649
65,341
3,872

-14
3
-12
-14
44
191,847
11,203
465,635
48,157

-38
-52
-39
-38
27
aLDDV usage is expected to be phased out by 2007.

and HDDV, which may be result from uncertainties in the original tests and the fact that the data
on HDDV were collected in a much more recent study.

       The results shown in Table 7 for HAP precursors are similar to the results for gaseous
HAPs. Again, emissions are projected to decrease despite increases in overall VMT,  and
reductions are seen for most vehicle categories (except MC).

       Tables 8 through 10 give national projections of emissions from the different nonroad
categories for gaseous HAPs, particulate HAPs, and HAP precursors, respectively. Table 8
shows significant reductions (24 to 41%) in gaseous HAP emissions, driven mainly by the
2-stroke gasoline engine, 4-stroke gasoline engine, and nonroad diesel engine categories.  For
particulate HAPs (Table 9), emissions increases ranging from 11 to 36% are projected, driven
primarily by the nonroad diesel engine category. HAP precursor emissions (Table 10) are
projected to decrease significantly, driven by the 2-stroke, 4-stroke, and nonroad diesel
categories.

       Tables 11 through  13 give state-level projections of mobile source emission changes for
gaseous HAPs, particulate HAPs, and HAP precursors.  For each pollutant and each  state, the
tables give the predicted change in total mobile source emissions between 1996 and 2007.
Table 11 shows reductions in almost all of the gaseous HAPs for all states. The one exception is
MTBE, where emissions in states that did not use the  additive in reformulated gasoline were
already low in 1996. It should be noted that our projections did not take into account potential
phase-out of MTBE usage in the reformulated gasoline program. The magnitudes of percentage
reductions for gaseous HAPs vary from state to state,  because of differences in the  relative
importance of the various onroad vehicle categories and nonroad engine categories.

                                            25

-------
  Table 8.  Gaseous HAP Emission Projections for Different Nonroad Engine Categories
              Contiguous 48 States and the District of Columbia (excludes Alaska and Hawaii)
HAP
2-Stroke
gasoline
4-Stroke
gasoline
Nonroad Marine
diesel diesel Railroad
Airports
Total
nonroad
Estimated 1996 emissions (tons)
Acetaldehyde
Acrolein
Benzene
1,3 -Butadiene
Ethyl benzene
Formaldehyde
Hexane
MTBE
Propionaldehyde
Styrene
Toluene
Xylenes
2,900
511
43,951
3,016
40,539
4,541
26,552
26,286
420
2,147
166,494
180,516
2,818
470
37,439
4,554
14,308
8,576
11,463
27,651
1,262
482
54,571
48,038
32,278
4,996
8,839
809
1,347
64,994
691
0
4,281
258
6,502
4,608
-j
j
77
0
0
44
23
121
0
134
46
70
105
0
167
0
0
96
0
263
0
291
100
153
229
2,127
1,006
1,101
863
159
6,896
40
0
431
196
821
546
40,126
7,226
91,330
9,242
56,493
85,030
39,129
53,936
6,819
3,230
228,611
234,043
Projected 2007 emissions (tons)
Acetaldehyde
Acrolein
Benzene
1,3 -Butadiene
Ethyl benzene
Formaldehyde
Hexane
MTBE
Propionaldehyde
Styrene
Toluene
Xylenes
Projected change (%)
Acetaldehyde
Acrolein
Benzene
1,3 -Butadiene
Ethyl benzene
Formaldehyde
Hexane
MTBE
Propionaldehyde
Styrene
Toluene
Xylenes
2,191
385
33,137
2,350
30,684
3,399
19,816
15,045
317
1,629
125,925
136,767

-24
-25
-25
-22
-24
-25
-25
-43
-25
-24
-24
-24
1,781
297
23,962
2,927
9,154
5,406
7,518
16,801
798
306
35,046
30,694

-37
-37
-36
-36
-36
-37
-34
-39
-37
-37
-36
-36
18,323
2,836
5,017
459
765
36,895
392
0
2,430
147
3,691
2,616

-43
-43
-43
-43
-43
-43
-43
na
-43
-43
-43
-43
4
84
0
0
48
25
132
0
147
50
77
115

10
10
10
na
10
10
10
na
10
10
10
10
0
139
0
0
79
0
218
0
242
83
127
190

na
-17
na
na
-17
na
-17
na
-17
-17
-17
-17
2,701
1,277
1,399
1,096
202
8,758
50
0
547
249
1,042
694

27
27
27
27
27
27
27
na
27
27
27
27
25,000
5,019
63,515
6,832
40,931
54,483
28,126
31,846
4,481
2,464
165,908
171,076

-38
-31
-30
-26
-28
-36
-28
-41
-34
-24
-27
-27
na = not applicable, emissions were 0 or not reported in 1996
                                             26

-------
 Table 9. Particulate HAP Emission Projections for Different Nonroad Engine Categories
              Contiguous 48 States and the District of Columbia (excludes Alaska and Hawaii)
2-Stroke
HAP gasoline
Estimated 1996 emissions
Total POM
7-PAH
Arsenic
Beryllium
Cadmium
Chromium
Lead
Manganese
Mercury
Nickel
Projected 2007 emissions
Total POM
7-PAH
Arsenic
Beryllium
Cadmium
Chromium
Lead
Manganese
Mercury
Nickel
Projected change (%)
Total POM
7-PAH
Arsenic
Beryllium
Cadmium
Chromium
Lead
Manganese
Mercury
Nickel
(tons)
23.8
12.4
0.0
0.0
0.0
1.6
0.0
3.2
0.3
1.9
(tons)
28.7
14.9
0.0
0.0
0.0
1.7
0.0
3.4
0.3
2.0

21
20
na
na
na
9
na
9
9
9
4-Stroke
gasoline

7.2
3.8
0.0
0.0
0.0
1.5
0.0
2.9
0.2
1.7

4.7
2.5
0.0
0.0
0.0
1.7
0.0
3.4
0.3
2.0

-35
-35
na
na
na
15
na
15
15
15
Nonroad
diesel

2.7
0.8
0.0
0.0
0.0
21.1
0.0
21.1
6.0
9.0

2.1
0.6
0.0
0.0
0.0
29.1
0.0
29.1
8.3
12.5

-22
-22
na
na
na
38
na
38
38
38
Marine
diesel Railroad

0.8
0.008
1.9
0.0
0.3
9.8
1.0
7.7
0.1
75.1

0.9
0.009
2.1
0.0
0.3
11.0
1.2
8.7
0.1
84.9

8
6
13
13
13
13
13
13
13
13

0.00
0.00
0.01
0.00
0.00
0.09
0.00
0.05
0.00
0.17

0.00
0.00
0.01
0.00
0.00
0.09
0.00
0.05
0.00
0.17

na
na
0
na
na
0
na
0
na
0
Airports

5.8
0.1
0.0
0.0
0.0
0.0
526.1
0.0
0.0
0.0

7.7
0.1
0.0
0.0
0.0
0.0
584.0
0.0
0.0
0.0

34
34
na
na
na
na
11
na
na
na
Total
nonroad

40.3
17.0
1.9
0.0
0.3
34.0
527.2
34.9
6.6
87.9

44.1
18.1
2.1
0.0
0.3
43.6
585.2
44.7
9.0
101.5

18
12
13
13
13
28
11
28
36
15
na = not applicable, emissions were 0 or not reported in 1996

-------
 Table 10.  HAP Precursor Emission Projections for Different Nonroad Engine Categories
              Contiguous 48 States and the District of Columbia (excludes Alaska and Hawaii)
2-Stroke
HAP produced gasoline
Estimated 1996 emissions
Acetaldehyde
Acrolein
Formaldehyde
Propionaldehyde
Projected 2007 emissions
Acetaldehyde
Acrolein
Formaldehyde
Propionaldehyde
Projected change (%)
Acetaldehyde
Acrolein
Formaldehyde
Propionaldehyde
(tons)
37,737
3,016
70,119
12,549
(tons)
19,721
2,350
39,401
6,558

-48
-22
-44
-48
4-Stroke
gasoline

50,688
4,554
188,082
7,707

30,761
2,927
114,398
4,677

-39
-36
-39
-39
Nonroad
diesel

32,331
809
145,045
7,754

17,962
459
80,981
4,308

-44
-43
-44
-44
Marine
diesel

2,562
0
8,814
655

2,810
0
9,671
718

10
na
10
10
Railroad

3,451
0
11,951
828

3,455
0
11,964
829

0
na
0
0
Airports

25,688
863
111,124
8,927

28,969
1,096
125,739
10,067

13
13
13
13
Total
nonroad

152,458
9,242
535,136
38,420

103,679
6,832
382,155
27,156

-32
-26
-29
-29
na = not applicable, emissions were 0 or not reported in 1996
                                             28

-------
Table 11. Projected State-Level Changes in Total Mobile Source Emissions of Gaseous HAPs (%)
State
Alabama
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Contiguous 48
i-s
§f
i N
§£
-33
-39
-31
-58
-34
-66
-46
-60
-31
-47
-27
-47
-41
-29
-36
-38
-31
-26
-51
-52
-24
-17
-31
-42
-23
-36
-39
-24
-63
-33
-43
-34
-36
-44
-33
-42
-43
-61
-31
-30
-33
-41
-30
-18
-41
-36
-39
-22
i23
-38
^ 
S T3
II
PH 13
-34
-33
-32
-41
-29
-51
-40
-46
-30
-37
-30
-38
-37
-34
-32
-35
-28
-31
-37
-38
-33
-28
-34
-32
-26
-30
-29
-33
-47
-29
-35
-32
-30
-39
-31
-38
-39
-42
-34
-32
-34
-38
-28
-27
-34
-32
-37
-36
i22
-36
a
%
£>
CO
-30
-35
-29
-59
-35
-71
-44
-60
-26
-47
-28
-50
-41
-34
-34
-36
-31
-34
-53
-55
-30
-25
-29
-41
-28
-35
-35
-33
-67
-29
-47
-31
-38
-46
-30
-43
-46
-63
-28
-36
-30
-41
-32
-29
-39
-33
-37
-33
i30
-41
Toluene
-32
-38
-31
-60
-35
-69
-45
-61
-29
-48
-28
-49
-42
-32
-36
-38
-32
-29
-53
-55
-26
-20
-30
-43
-26
-37
-39
-27
-66
-32
-45
-33
-38
-46
-32
-44
-45
-63
-30
-33
-32
-41
-32
-22
-41
-35
-39
-26
i27
-40
Xylenes
-33
-39
-31
-58
-33
-67
-45
-60
-31
-48
-26
-47
-40
-28
-37
-38
-31
-26
-51
-52
-23
-16
-31
-42
-22
-36
-39
-23
-64
-33
-42
-34
-36
-44
-33
-42
-42
-61
-31
-29
-33
-41
-30
-17
-41
-36
-39
-21
i22
-37
                                          29

-------
Table 12. Projected State-Level Changes in Total Mobile Source Emissions of Particulate HAPs (%)
State
Alabama
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Contiguous 48
Total
POM PAH-7
-33
-28
-30
-23
-19
-34
-22
-20
-23
-28
-4
-20
-31
-25
-33
-29
-24
o
J
-18
-21
-7
-4
-33
-28
-9
-31
-17
1
-26
-28
-17
-23
-24
-31
-33
-18
-23
-30
-30
-19
-32
-21
-14
12
-25
-16
-36
-4
-12
-21
-33
-30
-32
-25
-24
-35
-43
-20
-26
-32
-4
-24
-32
-25
-34
-33
-27
9
-21
-16
-5
o
-J
-33
-32
-8
-32
-26
1
-27
-33
-18
-25
-25
-33
-35
-17
-27
-33
-31
-20
-33
-25
-17
10
-26
-16
-37
-2
-11
-22
Ar- Beryl-
senic Hum
6
-51
-26
1
-48
0
9
-51
5
-10
-49
-7
3
-50
-47
-11
12
8
5
0
5
8
7
1
-43
-41
-50
-5
0
-51
7
-15
-45
4
-32
6
6
6
-5
-53
-2
10
-51
-53
6
9
8
-24
-38
5
13
na
13
13
na
13
13
13
13
13
na
13
13
na
na
13
13
13
13
13
13
13
13
13
na
na
na
13
13
na
13
13
na
13
13
13
13
13
13
na
13
13
na
na
13
13
13
13
na
13
Cad-
mium
13
na
13
13
na
13
13
13
13
13
na
13
13
na
na
13
13
13
13
13
13
13
13
13
na
na
na
13
13
na
13
13
na
13
13
13
13
13
13
na
13
13
na
na
13
13
13
13
na
13
Chro-
mium
24
34
28
29
32
27
22
32
30
30
31
28
25
27
29
27
17
24
27
27
24
24
23
26
30
28
36
28
28
31
25
30
27
25
28
27
25
26
28
29
28
24
34
30
25
26
23
28
26
26
Di-
oxins
27
27
27
27
27
-4
23
27
27
23
27
27
27
27
27
27
27
14
27
16
27
27
27
27
27
27
27
13
-19
27
24
27
27
27
27
27
27
-16
27
27
27
27
27
26
27
27
27
27
27
25
Man-
Lead ganese
11
27
15
21
23
14
17
21
26
21
20
14
12
10
14
13
13
14
17
14
10
15
12
14
19
14
35
16
13
22
8
20
10
11
14
19
11
12
18
16
17
19
28
17
17
22
10
14
ii
16
25
32
29
29
31
28
23
31
29
30
31
29
26
29
30
28
17
25
27
28
26
25
24
27
30
29
33
29
28
31
27
30
29
26
29
27
26
27
29
30
28
24
32
30
26
26
24
29
28
27
Mer-
cury Nickel
33
35
34
34
34
34
33
35
34
34
34
34
34
34
34
34
32
34
34
34
34
34
34
34
34
34
35
34
34
34
34
34
34
34
34
34
34
34
34
34
34
34
35
34
34
34
34
34
33
34
15
32
23
19
29
18
14
29
18
21
28
20
16
24
25
19
13
15
16
18
16
15
15
17
26
24
35
19
18
29
16
22
23
16
23
16
16
16
20
27
18
15
32
27
16
15
15
22
23
16
na = not applicable, emissions were 0 or not reported in 1996
                                              30

-------
  Table 13. Projected State-Level Changes in Total Mobile Source Emissions
                           of HAP Precursors (%)
State
Alabama
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Contiguous 48
Acetaldehyde
-26
-28
-27
-49
-29
-55
-37
-33
-24
-42
-26
-38
-36
-35
-31
-33
-29
-31
-44
-45
-35
-31
-25
-35
-29
-32
-29
-32
-48
-22
-46
-28
-35
-43
-27
-40
-42
-46
-26
-36
-27
-34
-30
-38
-34
-30
-32
-41
-29
-36
Acrolein
-48
-46
-45
-41
-62
-48
-63
-50
-41
-47
-54
-40
-48
-45
-49
-50
-45
-26
-47
-46
-33
-25
-47
-46
-53
-48
-54
-31
-44
-67
-37
-45
-44
-50
-62
-45
-37
-47
-45
-40
-47
-56
-61
-21
-38
-42
-52
-29
-54
-45
Formaldehyde
-25
-26
-26
-50
-28
-48
-39
-26
-21
-38
-26
-35
-35
-35
-31
-35
-25
-28
-40
-41
-35
-30
-23
-34
-28
-31
-24
-31
-42
-23
-40
-27
-32
-41
-27
-38
-41
-43
-25
-34
-28
-32
-29
-35
-33
-28
-32
-40
-29
-35
Propion-
aldehyde
-25
-27
-26
-48
-28
-54
-37
-27
-22
-41
-24
-37
-35
-34
-30
-33
-29
-30
-43
-44
-35
-29
-23
-35
-28
-31
-28
-31
-48
-21
-45
-27
-34
-43
-26
-40
-42
-46
-25
-36
-27
-34
-29
-37
-33
-29
-31
-41
-28
-35
na = not applicable, emissions were 0 or not reported in 1996

-------
       Particulate HAP emissions (Table 12) are projected to increase for most compounds in
most states.  One exception is arsenic, where reductions as a result of the phaseout of LDDV
offset increases from other emission categories in some states.  If only onroad emissions were
considered, a similar situation would be seen with mercury, but in this case the reductions in
LDDV emissions are entirely offset by emissions increases from the nonroad categories.  In the
case of dioxin, increases of about 27% are projected from most states, largely as a result of
increases in HDDV traffic. However, the dioxin emissions increases from HDDV are offset by
reductions from LDDV in some of the more urban states, especially Rhode Island, New Jersey,
and Connecticut. Projected reductions in HAP precursors in Table 13 are similar to the projected
reductions for gaseous HAPs.

       Figures 6 through 10 show the county-level distributions of projected emission reductions
for acetaldehyde, acrolein, benzene, 1,3-butadiene, and formaldehyde, respectively.  These are the
HAPs from the NAT A list of 33 for which mobile sources contribute the largest share of national
emissions. Each figure shows the percentage change in emissions between 1996 and 2007,
accounting for projected  emission controls and changes in VMT and nonroad activity. Overall
emission reductions vary from state to state and from county to county, depending on the
projected increases in VMT and nonroad activity  and on the county-specific mix of onroad vehicle
categories and nonroad engine categories.  As Figures 6 through 10 show, emission controls more
than offset projected increases VMT and nonroad activity for all  of the above pollutants.  For
comparison, Figure  11 shows projected increases in VMT at the county level.

4.2 Projected Ambient Impacts

       This section  gives modeling results for five organic HAPs - acetaldehyde, acrolein,
benzene, 1,3-butadiene, and formaldehyde - and eight HAP metals - arsenic, beryllium, cadmium,
chromium, lead, manganese, mercury, and nickel.  Although future emissions have been projected
for an additional seven organic HAPs that are of interest for mobile sources, these HAPs are not
currently included in the  NATA modeling effort.  Modeling results also are not yet available for
POM (16-PAH and 7-PAH) and dioxins/furans.

       Table 14 summarizes nationwide (excluding Alaska and Hawaii) ambient HAP
concentration impacts for mobile source emissions in 1996 and 2007. The table gives average
ambient concentrations in 1996 and 2007 for onroad mobile sources, nonroad mobile sources, and
the sum of onroad and nonroad mobile sources. The averages  shown in the table are the average
of all the census tract ambient concentrations in the 48 States and the district of Columbia (60,803
census tracts).  The median mobile source census tract concentrations are also given for each
HAP for 1996 and 2007. Median ambient impacts are also given for all mobile sources in
1996and 2007. For  comparison, Table 14 also shows the average and median ambient impacts of
all anthropogenic sources - stationary and mobile - in the 1996 base year.  The ambient impacts
are based on the ASPEN dispersion model, which takes into account the dispersion and decay of
HAP emissions, as well as HAPs formed from secondary reactions of precursor pollutants. The
concentrations given in Table 14 represent the total impacts of both directly emitted HAPs and
HAPs formed from precursor pollutants.
                                           32

-------
     Projected emission reduction
         0-20%
         20-40%
         40-50%
     	50-60%
     ^m 60-75%
    Figure 6. Projected county-level mobile source emission reductions for
                                   acetaldehyde
     Projected emission reduction
         0-20%
         20-40%
         40-50%
     	50-60%
     ^m 60-75%
Figure 7.  Projected county-level mobile source emission reductions for acrolein

                                        33

-------
     Projected emission reduction
         0-20%
         20-40%
         40-50%
       | 50-60%
     ^1 60-75%
Figure 8.  Projected county-level mobile source emission reductions for benzene
     Projected emission reduction
         0-20%
         20-40%
         40-50%
       | 50-60%
     ^| 60-75%
    Figure 9. Projected county-level mobile source emission reductions for
                                  1,3-butadiene
                                        34

-------
 Projected emission reduction
     0-20%
     20-40%
     40-50%
   | 50-60%
 ^1 60-75%
Figure 10.  Projected county-level mobile source emission reductions for
                              formaldehyde
    Projected growth in VMT
        1-2%/yr
        2-2.5%/yr
        2.5-3%/yr
    ^B 3-3.5%/yr
    ^| 3.5-4%/yr
      Figure 11. Projected increase in total vehicle miles traveled

                                    35

-------
              Table 14. Summary of Projected Nationwide Ambient Impacts from Mobile Sources
                                             (excluding Alaska and Hawaii)"
Average modeled ambient impacts
for mobile sources in 1996 (us/m3)
HAP
Gaseous HAPs
Acetaldehyde
Acrolein
Benzene
Butadiene
Formaldehyde
Particulate HAPs
Arsenic
Beryllium
Cadmium
Chromium
Lead
Manganese
Mercury
Nickel
Onroad

4.0e-01
4.5e-02
5.6e-01
5.0e-02
3.8e-01

7.0e-07
O.Oe+00
O.Oe+00
4.2e-05
5.9e-05
1.7e-05
6.0e-07
3.3e-05
Nonroad

2.7e-01
4.0e-02
2.3e-01
1.7e-02
4.8e-01

7.6e-06
l.Oe-07
9.0e-07
1.4e-04
3.5e-03
1.4e-04
2.9e-05
3.2e-04
All mobile

6.7e-01
8.5e-02
7.9e-01
6.7e-02
8.6e-01

8.3e-06
l.Oe-07
9.0e-07
1.8e-04
3.5e-03
1.6e-04
3.0e-05
3.5e-04
Average modeled impacts for mobile
sources in 2007 (us/m3)
Onroad

2.2e-01
2.2e-02
2.8e-01
2.4e-02
1.8e-01

3.0e-07
O.Oe+00
O.Oe+00
5.0e-05
6.8e-05
2.1e-05
5.0e-07
4.0e-05
Nonroad

1.9e-01
3.1e-02
1.5e-01
1.2e-02
3.4e-01

8.6e-06
l.Oe-07
l.le-06
1.8e-04
3.7e-03
1.8e-04
4.0e-05
3.7e-04
All mobile

4.1e-01
5.2e-02
4.3e-01
3.6e-02
5.2e-01

8.9e-06
l.Oe-07
l.le-06
2.3e-04
3.8e-03
2.0e-04
4.0e-05
4.1e-04
Median modeled
impacts for all mobile
sources (us/m3)
1996

4.5e-01
5.4e-02
6.4e-01
5.2e-02
5.1e-01

8.0e-07
O.Oe+00
O.Oe+00
7.8e-05
4.8e-04
6.3e-05
l.le-05
6.2e-05
2007

2.9e-01
3.1e-02
3.4e-01
2.7e-02
2.9e-01

4.0e-07
O.Oe+00
O.Oe+00
9.9e-05
5.3e-04
8.1e-05
1.4e-05
7.7e-05
Total modeled impacts of
all sources (stationary
plus mobile sources) for
1996 (us/m3)
Averase

7.4e-01
l.le-01
9.1e-01
8.0e-02
l.Oe+00

1.6e-04
3.1e-05
1.9e-04
3.4e-03
5.8e-03
4.4e-03
2.0e-04
2.3e-03
Median

5.2e-01
7.9e-02
7.3e-01
6.0e-02
6.5e-01

6.8e-05
l.le-05
4.1e-05
8.5e-04
1.8e-03
2.2e-03
7.6e-05
9.8e-04
a Ambient impacts are expressed in scientific notation: l.Oe-03 = 1.0 x 10"3 = 0.0010
                                                         36

-------
       As Table 14 shows, average and median ambient impacts from both onroad and nonroad
sources are projected to decline substantially between 1996 and 2007 for organic HAPs.  For
metal HAPs, the combined impact of onroad and nonroad emissions is projected to increase in all
cases.  However, contribution of mobile sources to the total manmade ambient impact is much
smaller for HAP metals than for organic HAPs.

       Table 15 gives the same information as Table 14, but broken out between urban and rural
counties. This table shows that the substantial nationwide decreases in gaseous HAP impacts are
spread across both urban and rural areas. Likewise, small net increases in metal HAP impacts
concentrations are predicted in both urban and rural areas. Table  15 also shows that ambient
HAP concentrations are considerably higher in urban areas than in rural areas, because of the
increased density of both mobile and stationary emission sources in an urban setting.

       Table 16 summarizes the projected changes in ambient impact, in terms of percentage
reductions on an overall national basis.  For comparison, the table also gives the projected
changes in nationwide direct HAP emissions.  As the table shows, the projected changes in
modeled ambient concentrations are similar to the projected changes in direct HAP emissions, but
they are not identical, especially for the organic HAPs.  A number of factors can cause the
projected change in the ambient impact of a given HAP to differ from the corresponding change in
direct emissions. These include decay reactions for organic HAPs, the formation of organic HAPs
from precursor pollutants, and differences in the spatial  distributions of ambient impacts and
emission reductions.

       Table 17 summarizes the projected changes in ambient HAP impacts for urban and rural
counties. The projected changes are similar for urban and rural counties. The one exception is
arsenic, where the average ambient impact is projected to increase by 5% in urban counties, and
decrease by 4% in rural counties.  This is because the decreases in arsenic from LDDV generally
offset increases in from HDDV and diesel marine vessels. Diesel marine vessels, which account
for the greatest increases in arsenic emissions, are concentrated in urban port areas.

       Tables 18 and 19 give state-level projections of changes in mobile source ambient impacts
for gaseous HAPs and particulate HAPs, respectively.  For each pollutant and each state, the
tables give the predicted change in total  mobile source ambient impact between 1996 and 2007.
Table 18 shows reductions in all of the five gaseous HAPs for all states.  The magnitudes of
percentage reductions in ambient impacts for gaseous HAPs range from  17% less (for
acetaldehyde in Colorado) to 26% more (for acrolein in Connecticut) than the corresponding state
level reduction in direct HAP emissions (From Table 11). As noted earlier, a number of factors
can cause these differences,  including decay reactions for organic  HAPs, the formation of organic
HAPs from precursor pollutants, and differences  in the spatial distributions of ambient impacts
and emission reductions.

       Ambient levels of HAP metals (Table 19) are projected to increase for most compounds in
most states.  These increases are similar to the projected increases in emissions (Table 12).  The
one exception is arsenic, where reductions as a result of the phaseout of LDDV offset increases
from other emission categories in some states.

                                           37

-------
Table 15. Projected Average Ambient Impacts from Mobile Sources in Urban and Rural Counties
                Contiguous 48 states and the District of Columbia (excludes Alaska and Hawaii)
Average modeled ambient impacts
for mobile sources in 1996 (ug/m3)
HAP
Urban counties
Gaseous HAPs
Acetaldehyde
Acrolein
Benzene
Butadiene
Formaldehyde
Paniculate HAPs
Arsenic
Beryllium
Cadmium
Chromium
Lead
Manganese
Mercury
Nickel
Rural counties
Gaseous HAPs
Acetaldehyde
Acrolein
Benzene
Butadiene
Formaldehyde
Paniculate HAPs
Arsenic
Beryllium
Cadmium
Chromium
Lead
Manganese
Mercury
Nickel
Onroad


4.8e-01
5.5e-02
6.7e-01
6.1e-02
4.6e-01

7.7e-07
O.Oe+00
O.Oe+00
5.0e-05
7.0e-05
2.1e-05
6.7e-07
3.9e-05


7.9e-02
9.9e-03
1.2e-01
l.le-02
7.9e-02

1.8e-07
O.Oe+00
O.Oe+00
1.2e-56
1.6e-05
4.7e-06
1.6e-07
9.0e-06
Nonroad


3.3e-01
4.9e-02
2.8e-01
2.0e-02
5.9e-01

9.5e-06
8.3e-08
1.2e-06
1.7e-04
4.3e-03
1.7e-04
3.6e-05
4.0e-04


3.3e-02
5.0e-03
5.1e-02
3.6e-03
5.5e-02

4.9e-07
4.1e-09
5.7e-08
1.4e-05
1.7e-04
1.4e-05
3.3e-06
2.2e-05
All mobile


S.le-01
l.Oe-01
9.5e-01
8.1e-02
l.le+00

l.Oe-05
8.3e+08
1.2e-06
2.2e-04
4.4e-03
2.0e-04
3.7e-05
4.4e-04


l.le-01
1.5e-02
1.8e-01
1.5e-02
1.3e-01

6.7e-07
4.1e-09
5.7e-08
2.5e-05
1.8e-04
1.9e-05
3.4e-06
3.1e-05
Average modeled ambient impacts
for mobile sources in 2007 (ug/m3)
Onroad


2.6e-01
2.6e-02
3.3e-01
2.9e-02
2.2e-01

3.6e-07
O.Oe+00
O.Oe+00
6.0e-05
8.0e-05
2.5e-05
5.5e-07
4.7e-05


5.4e-02
5.4e-03
6.5e-02
5.8e-03
4.3e-02

7.7e-08
O.Oe+00
O.Oe+00
1.4e-05
1.9e-05
5.7e-06
1.3e-07
l.le-05
Nonroad


2.3e-01
3.8e-02
1.8e-01
1.4e-02
4.2e-01

l.le-05
9.4e+08
1.3e-06
2.2e-04
4.6e-03
2.3e-04
4.9e-05
4.6e-04


2.2e-02
3.4e-03
4.0e-02
2.8e-03
3.5e-02

5.6e-07
4.6e-09
6.4e-08
1.8e-05
1.7e-04
1.9e-05
4.4e-06
2.6e-05
All mobile


5.0e-01
6.4e-02
5.1e-01
4.3e-02
6.4e-01

l.le-05
9.4e-08
1.3e-06
2.8e-04
4.7e-03
2.5e-04
5.0e-05
5.1e-04


7.6e-02
8.8e-03
l.Oe-01
8.6e-03
7.8e-02

6.3e-07
4.6e-09
6.4e-08
3.2e-05
1.9e-04
2.4e-05
4.6e-06
3.7e-05
Median modeled
impacts for all mobile
sources (ug/m3)
1996


6.1e-01
7.3e-02
8.2e-01
6.8e-02
7.0e-01

1.6e-06
2.8e-09
3.9e-08
l.le-04
l.Oe-03
9.3e-05
1.5e-05
l.le-04


8.7e-02
l.le-02
1.2e-01
8.6e-03
l.Oe-01

1.3e-07
O.Oe+00
O.Oe+00
1.7e-05
1.9e-05
1.2e-05
2.3e-06
l.le-05
2007


3.7e-01
4.1e-02
4.3e-01
3.5e-02
4.0e-01

1.2e-06
3.1e-09
4.4e-08
1.4e-04
l.le-03
1.2e-04
2.0e-05
1.3e-04


5.8e-02
6.4e-03
6.4e-02
4.7e-03
5.8e-02

6.5e-08
O.Oe+00
O.Oe+00
2.1e-05
2.2e-05
1.6e-05
3.0e-06
1.4e-05
Total modeled impacts of
all sources (stationary and
mobile) for 1996 (ug/m3)
Average


9.0e-01
1.3e-02
l.le+00
9.3e-02
1.2e+00

1.9e-04
3.7e-05
2.2e-04
4.1e-03
7.0e-03
4.9e-03
2.3e-04
2.7e-03


1.6e-01
4.2e-02
2.5e-01
2.8e-02
3.9e-01

6.1e-05
8.7e-06
7.1e-05
6.7e-04
9.9e-04
2.3e-03
6.3e-05
5.8e-04
Median


6.8e-01
9.6e-02
9.3e-01
7.3e-02
8.2e-01

8.7e-05
1.4e-05
6.0e-05
1.3e-03
2.9e-03
2.7e-03
l.le-04
1.4e-03


1.2e-01
3.4e-02
1.8e-01
1.6e-02
2.1e-01

1.4e-05
1.3e-06
4.7e-06
7.5e-05
1.6e-04
4.2e-04
l.Oe-05
l.Oe-04
                                             38

-------
     Table 16. Projected Changes in Ambient Concentrations - Nationwide Estimates
             Contiguous 48 States and the District of Columbia (excludes Alaska and Hawaii)




Change in modeled average ambient
impact from 1996 to 2007 (%)
HAP
Gaseous HAPs
Acetaldehyde
Acrolein
Benzene
1,3 -Butadiene
Formaldehyde
Particulate HAPs
Arsenic
Beryllium
Cadmium
Chromium
Lead
Manganese
Mercury
Nickel
Onroad

-45
-52
-50
-52
-52
-54
na
na
19
14
19
-17
20
Nonroad

-29
-24
-36
-28
-30
13
13
13
29
6
29
36
16
All mobile

-39
-39
-46
-46
-40
8
13
13
27
6
28
35
16
Change in
median
impact - all
mobile
(%)

-36
-43
-46
-48
-43
-44
13
13
27
10
28
33
25

Change in
Onroad

-42
-39
-50
-52
-45
-55
na
na
21
17
21
-21
23


emissions from 1996 to
2007 (%)
Nonroad

-38
-31
-30
-26
-28
13
13
13
28
11
28
36
15
All mobile

-40
-34
-43
-45
-38
5
13
13
26
11
27
34
16
na - not applicable, impact was 0 or insignificant in 1996
   Table 17. Projected Changes in Ambient Concentrations - Urban and Rural Counties
             Contiguous 48 States and the District of Columbia (excludes Alaska and Hawaii)
HAP
Gaseous HAPs
Acetaldehyde
Acrolein
Benzene
1,3 -Butadiene
Formaldehyde
Particulate HAPs
Arsenic
Beryllium
Cadmium
Chromium
Lead
Manganese
Mercury
Nickel
Change in
in urban
Average
onroad

-39
-50
-51
-52
-50
-55
na
na
21
16
21
-16
22
modeled average ambient impact
counties from 1996 to 2007 (%)
Average
nonroad

-31
-27
-35
-32
-32
13
13
13
27
6
27
36
15
Average
all mobile

-36
-42
-47
-47
-42
7
13
13
25
6
26
34
16
Median
all mobile

-37
-43
-47
-48
-42
5
13
13
25
9
26
34
16
Change in modeled average ambient impact
in rural counties from 1996 to 2007 (%)
Average
onroad

-31
-47
-49
-51
-46
-56
na
na
20
15
20
-16
21
Average
nonroad

-37
-34
-23
-23
-38
13
13
13
31
2
31
36
17
Average
all mobile

-33
-43
-42
-45
-43
-4
13
13
26
4
28
34
18
Median
all mobile

-32
-44
-42
-44
-42
-4
13
13
26
10
28
34
18
na - not applicable, impact was 0 or insignificant in 1996
                                            39

-------
Table 18. Projected State-Level Changes in Total Mobile Source Ambient Impacts for
                             Organic HAPs (%)
State
Alabama
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
National
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Contiguous 48
Acetaldehyde
-29
-26
-26
-42
-44
-51
-43
-43
-21
-46
-32
-36
-37
-34
-31
-35
-28
-28
-43
-45
-37
-33
-26
-36
-34
-39
-33
-26
-35
-43
-30
-42
-29
-33
-43
-27
-41
-44
-44
-25
-32
-29
-37
-42
-34
-36
-28
-35
-43
i42
-39
Acrolein
-44
-29
-40
-37
-56
-51
-51
-42
-35
-48
-48
-32
-45
-41
-43
-43
-36
-25
-40
-45
-42
-35
-43
-35
-44
-39
-41
-32
-39
-39
-50
-35
-41
-36
-47
-50
-39
-41
-46
-41
-37
-43
-43
-50
-29
-36
-26
-46
-41
;48
-39
Benzene
-47
-59
-46
-37
-65
-49
-52
-53
-44
-49
-57
-44
-49
-46
-48
-50
-45
-29
-54
-54
-44
-31
-46
-48
-53
-46
-48
-58
-42
-46
-65
-44
-48
-43
-51
-48
-48
-49
-54
-45
-42
-48
-48
-61
-27
-50
-46
-51
-39
i59
-46
Butadiene
-48
-44
-46
-43
-69
-48
-64
-48
-43
-49
-59
-38
-49
-48
-50
-51
-46
-25
-49
-49
-45
-35
-47
-47
-55
-46
-49
-51
-40
-43
-68
-41
-47
-44
-51
-64
-46
-44
-49
-45
-44
-48
-55
-66
-26
-42
-41
-52
-39
i62
-46
Formaldehyde
-45
-29
-41
-40
-47
-46
-49
-44
-35
-47
-46
-31
-46
-46
-46
-46
-41
-30
-42
-45
-43
-39
-43
-40
-46
-40
-47
-30
-40
-38
-45
-36
-42
-44
-48
-44
-41
-44
-44
-41
-43
-44
-40
-47
-34
-39
-31
-47
-41
i5_5
-40
                                    40

-------
       Table 19. Projected State-Level Changes in
                                    Particulate
 Total Mobile Source Ambient Impacts for
HAPs (%)
State
Alabama
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
National
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Contiguous 48
Arsenic
9
-55
-9
4
-47
o
J
11
-31
10
-7
-53
6
7
-51
-29
8
12
10
11
7
6
-1
11
10
-49
8
-37
-55
o
-J
5
-54
-17
-20
-56
8
-16
11
11
6
2
-60
o
J
10
-53
-55
10
11
11
-6
-50
8
Beryl-
lium
13
na
13
13
na
13
13
13
13
13
na
13
13
na
13
13
13
13
13
13
13
13
13
13
na
13
na
na
13
13
na
13
13
na
13
13
13
13
13
13
na
13
13
na
13
13
13
13
13
na
13
Cadmium
13
na
13
13
na
13
13
13
13
13
na
13
13
na
13
13
13
13
13
13
13
13
13
13
na
13
na
na
13
13
na
13
13
na
13
13
13
13
13
13
na
13
13
na
13
13
13
13
13
na
13
Chro-
mium
22
32
27
28
30
26
19
34
26
30
31
25
24
25
28
23
16
23
22
26
25
30
20
23
27
27
27
36
27
28
29
32
30
22
25
27
22
21
24
27
27
27
23
34
29
23
20
19
27
21
27
Lead
8
o
J
o
J
6
8
8
10
11
8
11
o
J
6
6
o
J
9
o
J
5
-0
7
10
7
8
4
6
1
6
6
4
o
J
8
2
6
5
o
J
7
6
7
7
4
o
J
-2
5
7
9
o
-J
1
6
o
J
5
^
6
Man-
ganese
24
30
28
28
29
27
21
34
25
30
30
25
25
27
29
25
17
24
22
27
27
30
21
24
26
28
28
30
28
29
28
33
30
23
26
29
23
23
25
28
26
28
23
32
31
24
20
20
28
22
28
Mercury
34
33
34
35
33
34
33
36
34
34
34
34
34
33
34
34
31
34
34
35
35
35
33
34
31
35
34
30
34
35
30
36
34
30
35
34
34
34
34
34
30
34
34
35
34
34
30
33
34
28
35
Nickel
15
31
20
18
27
17
13
30
15
21
29
16
15
22
23
15
13
15
14
16
16
20
14
14
24
16
23
34
19
17
27
26
24
19
15
21
14
14
16
18
24
17
15
32
27
15
14
14
19
18
16
na - not applicable, impact was 0 or insignificant in 1996
                                             41

-------
       Figure 12 shows the county-level distribution for the percentage change between 1996 and
2007 of the average ambient impacts for acetaldehyde from mobile sources. Figure 13 shows
estimated acetaldehyde impacts (in terms of ambient concentration) from mobile sources in 2007.
Figures 14 and 15 give similar distributions for acrolein, Figures 16 and 17 for benzene,
Figures 18 and 19 for 1,3-butadiene, and Figures 20 and 21 for formaldehyde, respectively.
                                           42

-------
   Reduction in average concentration
     0-20%
     20-40%
     140-50%
     50-60%
     60-75%
Figure 12.  Projected mobile source ambient impact reductions for acetaldehyde
    Concentration in 2007 (ug/m3)
        0-0.1
      | 0.1 -0.2
•        0.2-0.5
        0.5-1.2
        1.2-3
 Figure 13. Projected mobile source ambient impacts for acetaldehyde in 2007
                                        43

-------
Reduction in average concentration
   0-20%
   20-40%
h   40-50%
   50-60%
   60-75%
Figure 14. Projected mobile source ambient impact reductions for acrolein
  Concentration in 2007 (ugfrn3)
     0-0.015
     0.015 - 0.05
     0.05 - 0.1
  ^B0.1 -0.2
  ^•0.2 -0.5
 Figure 15.  Projected mobile source ambient impacts for acrolein in 2007

                                        44

-------
Reduction in average concentration
   0-20%
   20-40%
   140-50%
   50-60%
   60-75%
Figure 16.  Projected mobile source ambient impact reductions for benzene
  Concentration in 2007 (ug/m3)
     0-0.1
     0.1 -0.2
B     0.2-0.5
     0.5-1.2
     1.2-3
 Figure 17. Projected mobile source ambient impacts for benzene in 2007

                                      45

-------
   Reduction in average concentration
      0-20%
      20-40%
h      40-50%
      50-60%
      60-75%
Figure 18. Projected mobile source ambient impact reductions for 1,3-butadiene
     Concentration in 2007 (ug/m3)
        0 • 0.01
        0.01  . 0.02
        50.02  - 0.05
        0.05  • 0.1
       • 0.1 - 0.2
 Figure 19. Projected mobile source ambient impacts for 1,3-butadiene in 2007

                                         46

-------
   Reduction in average concentration
      0-20%
      20-40%
h      40-50%
      50-60%
      60-75%
Figure 20.  Projected mobile source ambient impact reductions for formaldehyde
    Concentration in 2007 (ugfrn3)
        0-0.1
        0.1 - 0.2
        0.2 - 0.5
    |^B0.5 -1.2
    ^B 1.2 - 5
 Figure 21. Projected mobile source ambient impacts for formaldehyde in 2007

                                         47

-------
                                   4.  References
1.      Driver, L.; Pope, A,; Billings, R.; Wilson, D. "The 1996 National Toxics Inventory and
       Its Role in Evaluating the EPA's Progress in Reducing Hazardous Air Polluants in
       Ambient Air," Presented at the 92nd Annual Meeting of the Air & Waste Management
       Association, St. Louis, Missouri, June 1999; paper 91-501.

2.      "User's Guide: Assessment System for Population Exposure Nationwide Model,
       Version 1.1. EPA-454/R-00-017, U.S. Environmental Protection Agency.  April 2000.

3.      "Regulatory Impact Analysis: Heavy-Duty Engine and Vehicle Standards and Highway
       Diesel Fuel Sulfur Control Requirements." EPA-420-R-00-026, U.S. EPA Office of
       Transportation and Air Quality, Ann Arbor, Michigan. December 2000.

4.      Rosenbaum, Arlene, Mary Ligoki, and Yi Hua Wei. "Modeling Cumulative Outdoor
       Concentrations of Hazardous Air Pollutants.  U.S. Environmental Protection Agency,
       Washington, D.C. February 1999. www.epa.gov/oppecumm/resource/report.htm

5.      "User's Guide for the Emissions Modeling System for Hazardous Air Pollutants,
       Version 1.1. EPA-454/R-00-018, U.S. Environmental Protection Agency, October 2000.

6.      "National Air Pollutant Emission Trends Procedures Document, 1900-1996," EPA-450/R-
       98-008, U.S. Environmental Protection Agency. May 1998.

7.      U.S. Environmental Protection Agency. Emission Inventory Guidance.
       http://www.epa.gov/ttn.chief/ei_guide.html#toxic.

8.      "Analysis of the Impacts of Control Programs on Motor Vehicle Toxic Emissions and
       Exposure Nationwide, Volume I."  EPA420-R-99-029/030, U.S. Environmental
       Protection Agency, Ann Arbor, Michigan. 1999.

9.      2007 VOC inventory.

10.    Regulatory Impact Analysis: Control of Emissions of Air Pollution from Highway Heavy-
       Duty Engines. EPA-420-00-010. U.S. Environmental Protection Agency Office of
       Transportation and Air Quality.  July 2000.
       http://www.epa.gov/orcdizux/regs/hd-hwy/2000frm/r00010.pdf

11.    Updated Draft NONROAD Model in Support of 2007 HD Vehicle Rule. U.S.
       Environmental Protection Agency Office of Transportation and Air Quality, Ann Arbor,
       Michigan. June 8, 2000. http://www.epa.gov/otaq/nonrdmdl.htm

12.    Watson, J.D., E. Fjujita, J.C. Chow, and B. Zielinska.  1998.  Northern Front Range Air
       Quality Study. Desert Research Institute,  pg. 4-41.
                                          48

-------