United States         Air and Radiation        EPA420-R-03-001
           Environmental Protection                  January 2003
           Agency                        M6.PM.001
&EPA    MOBILES.1  Particulate
           Emission Factor Model
           Technical Description
           Final Report
                                     > Printed on Recycled Paper

-------
                                                          EPA420-R-03-001
                                                              January 2003
    MOBILE6.1 Participate Emission Factor Model
                    Technical Description

                            Final Report

                            M6.PM.001
                            Edward L. Glover
                           Mitch Cumberworth
                     Assessment and Standards Division
                    Office of Transportation and Air Quality
                    U.S. Environmental Protection Agency
                                NOTICE

  This technical report does not necessarily represent final EPA decisions or positions.
It is intended to present technical analysis of issues using data that are currently available.
       The purpose in the release of such reports is to facilitate the exchange of
     technical information and to inform the public of technical developments which
      may form the basis for a final EPA decision, position, or regulatory action.

-------
1.     Introduction

       Since 1995, EPA has made available the PARTS model, a Fortran program that estimates
particulate  air pollution emissions of in-use gasoline-fueled and diesel-fueled highway motor
vehicles. It calculates particle emission factors in grams per mile (g/mi) for on-road automobiles,
trucks, and motorcycles, for particle sizes of 1-10 microns.  The particulate matter (PM) estimates
include emission factors for exhaust particulate, brakewear, and tirewear. The PARTS model is now
outdated.

       MOBILE6 is the most recent EPA emission factor computer model. It calculates in-use fleet
emission factors forthree criteria pollutants: hydrocarbons (HC), carbon monoxide (CO), and oxides
of nitrogen (NOx). These emission estimates are made for gas, diesel and natural gas fueled cars,
trucks, buses and motorcycles for calendar years 1952 through 2050. The model calculates emission
factors under a wide  variety  of conditions  affecting in-use emission levels,  e.g.,  ambient
temperatures, average traffic speeds, etc.

       MOBILE and PARTS have  been  used by EPA and  other organizations in a variety of
applications.  These include evaluations of highway mobile source control strategies by state, local
and regional planning agencies; emission inventories and control strategies for State Implementation
Plans under the Clean Air Act; transportation plans and conformity analyzes by metropolitan
planning organizations and state transportation  departments; environmental impact statements by
industry investigators;  and academic research efforts.

       This document describes the methodology and algorithms used to combine PARTS and
MOBILE6 to produce an integrated MOBILE6.2 model.  This new model produces the same
estimates for HC, CO and NOx  emission as MOBILE6.0, but it also can  estimate particulate
emission factors like the PARTS  model.  The MOBILE6.2 particulate emission estimates differ
somewhat from the PARTS estimates. The principal reasons for these differences are changes in
vehicle registration and technology distributions between PARTS and MOBILE6 and the fact that
some particulate emission rates for future model years have been updated in MOBILE6.2 to reflect
recent rule-makings.
2 .    Overview of MOBILE6.2 Features

       The MOBILE6.2 model offers several advantages relative to the separate MOBILE6.0 and
PARTS models.  First,  the combination eliminates significant duplication of technical material
between the two models. For instance, both models contain many of the same data parameters
relating to vehicle activity and use. Both models also have very similar input requirements and
produce similar output.  Second, combining the two models aids users who are now given a single,
consistent interface for both functions, and allows EPA to support one consistent computer model
product rather than two. Combining PARTS and MOBILE6.0 was a prominent recommendation of
Modeling MOBILE SOURCE Emissions, the National Academy of Science Research  Council's
review of MOBILE. This panel concluded that the process of emission inventory modeling could

-------
be improved by creating a new model or suite of integrated models that could produce emission
factor estimates for a wider range of pollutants and conditions.

       The objective of the MOBILE6.2 update from PARTS was to produce, in the relatively short
term, a combined model that reflects EPA particulate emission modeling done for recent vehicle
emission control rule-makings.  The project takes into account the  fuel sulfur level reductions that
are now mandated, and new vehicle emission standards.

       Originally, it was EPA's intention to update the basic PM emission factors as part of the
MOBILE6.2 development process.  We looked at several test programs that collected particulate
matter data on both light-duty gasoline vehicles and heavy-duty diesel vehicles.  We also looked
carefully at the California vehicle emission factor model - EMFAC7, and we did some limited
analysis of the available data. After careful consideration, EPA decided that the test programs were
not sufficient to allow us to do a comprehensive and scientifically credible update  to the basic
particulate emission factors that could be applied to the nation as a whole.  On the heavy-duty diesel
vehicle side, there were only small data sets collected from numerous sources. Also,  the vehicles
in the test samples tended to be low mileage trucks.  These issues prevented EPA from making a
credible proj ection of in-use  emission function that would be any different from the one made in the
PARTS model, and the 2007 Heavy-Duty Diesel rule-making support documents. On the light-duty
gasoline vehicle side, some of the problems included inconsistent test programs that emphasized new
vehicles, and the inability to accurately quantify the effects and number of vehicles that produce
excessive amount of smoke  and particulates.

       To get the data that  is needed to revise the existing particulate emission factors, EPA is
undertaking maj or test programs in cooperation with CRC, CARB, EIIP, NREL and DOT to update
the particulate matter emission  estimates for heavy-duty diesels and light-duty gasoline vehicles.
Once these data are available, it is our intention to do a rigorous analysis of the data and make the
results available in the new EPA MOVES emission model.

       Until  the new data  have been collected and thoroughly analyzed,  the foundation  of
MOBILE6.2 is  made up of  the basic mobile  source particulate emission rates from the PARTS
model, and from the EPA rule-making modeling sources. These sources are supported by a large
body of engine and vehicle certification test results, and from some limited in-use test programs.

       One of the  new features of MOBILE6.2 is  its  ability to accept alternative basic exhaust
particulate rates into the model as a function of vehicle class, model year, catalyst technology, and
vehicle age. Deterioration estimates as a function of mileage can also be added. This new feature
allows new or alternate  emission factors to be entered into the model without reprogramming.
Changes are merely made to the existing external data files that accompany the model.
       Sections (Technical Description) describes the way PARTS and MOB ILE6.0 were combined
to produce MOBILE6.2 and the new features added.  Here is a brief summary of these updates:
2.1    Base Emission Rates  - The base emission rates for most vehicle classes and model years

-------
       are unchanged from PARTS.  However, the basic emission rates for heavy-duty diesel
       vehicles were updated in MOBILE6.2 to reflect the emission factors modeled in EPA's 2007
       Heavy-Duty Diesel Vehicle Rule-making effort. As a result MOBILE6.2 predicts that 2007
       and later  model year diesel heavy-duty vehicles will meet a 0.01 g/bhp-hr certification
       standard if low sulfur fuel is used. The basic PM emission rates for light-duty and heavy-
       duty gasoline vehicles were updated to assume compliance with EPA's Tier2 vehicle rule-
       making requirements in 2004, and with the 2005 heavy-duty gasoline vehicle rule-making,
       if low sulfur  fuel is used.
2.2    Sulfate Particulate and Gaseous SO2 Emission Factors - PARTS's calculation of sulphate
       particulate and gaseous SO2 exhaust emissions were restructured to account for the sulfur
       levels of gasoline and diesel fuel, while still using the same basic algorithms as PARTS.
       This feature of the program now allows the user to model the effects of different fuels and
       changes in EPA fuel regulations.

2.3    Ammonia Emission Factors  - MOBILE6.2 adds the ability to estimate exhaust emissions
       of ammonia.  These estimates are based on the  emission rates and calculation methods
       described in EPA Report Number EPA/AA/CTAB/PA/81-20, entitled "Determination of a
       Range of Concern for Mobile Source Emissions of Ammonia". While this report dates from
       1981, we are not aware of a better or significantly different basis for such calculations.

2.4    ZEVs  - MOBILE6.2 allows  the user to model  the effects of zero emitting vehicles on
       particulate emissions whereas PARTS did not have this capability.  In MOBILE6.2 the
       exhaust particulate emission factors are assumed  to be zero for ZEVs.  However, their tire
       and brake wear emissions are  assumed to be the same as gasoline-fueled vehicles.

2.5    Natural Gas Vehicles (NGVs)

       PARTS  did not contain exhaust particulate emission  estimates for NGVs. MOBILE6.2
       assumes that the exhaust particulate emissions of NGVs are the same as gasoline-fueled
       vehicles operating on very low sulfur fuel.  This assumption is based on comparisons
       between NGV and gasoline vehicle hydrocarbon emission test results. These test results,
       provided by the NGV industry (See EPA report EPA420-R-01-033) suggest that NGVs
       generally have equivalent or lower emissions than gasoline vehicles. Based on the similarity
       between hydrocarbon and particulate emission formation, the general assumption of rough
       equivalence between these vehicle types was extended to their particulate emission factors.
       The tire and brake wear emissions of NGVs are assumed to be the same as gasoline-fueled
       vehicles.

       Further improvements to the estimation of mobile  source particulate emissions will be made
in the course of the longer term effort to produce a new generation of mobile source air pollution
models (MOVES).  The MOVES model is intended to implement the recommendations of the
National Academy of Science. It will be based on an extensive database of emission measurements
made during actual operation of in-use vehicles and will provide a framework for allocating emission

-------
estimates to much smaller geographic areas and time periods.


3.    Technical Description


3.1    Definitions

      TheMOBILE6.2 model reports separate PM emission factors for twenty-eight vehicle classes
covering model years 1952 through 2050.  The PM and PM-related pollutants are:

OCARBON - The organic carbon portion of diesel exhaust particulate emissions. It was denoted
             as SOF in the PARTS model.

EC ARE ON - The elemental carbon and residual carbon  portion of diesel vehicle exhaust
             particulate. It was denoted as RCP in the PARTS model.

Sulfate -      The sulfate particulate emissions. These are based directly on the sulfur content of
             the fuel.

Lead-        The lead particulate emissions. These are based directly on the quantity of lead in the
             automotive fuel.  Like PARTS, MOBILE6.2 model assumes that post 1975 model
             year vehicles and all calendar years subsequent to 1991 are free from lead PM
             emissions.


Total  Exhaust  Diesel  PM   =     OCARBON +  ECARBON  +  Sulfate  +  Lead
             In MOBILE6.2,  Total Exhaust Diesel PM is calculated by the model and then
             apportioned to the four reported constituents: OCARBON, ECARBON, Sulfate, and
             Lead.
GASPM -     The sum of the organic and elemental carbon portion and any residual carbon portion
             of gasoline vehicle exhaust particulate.


Total  Exhaust Gasoline PM      =      GASPM +  Sulfate  +  Lead

             In MOBILE6.2, Total Exhaust Gasoline PM is the sum  of  three constituents
             GASPM, Sulfate and Lead emissions.


NH3 -        Ammonia emission factors.  These are new to the MOBILE6 and PARTS model
             series. Ammonia is a gaseous pollutant that is converted in the atmosphere to an
             ammonium based particulate emission.  Only the gaseous emissions which are

-------
              directly emitted from a vehicle tailpipe are reported by MOBILE6.2. The model does
              not contain any algorithms pertaining to the conversion of gaseous emissions to
              particulate emissions. These reactions and their effects are calculated in other EPA
              models.
BRAKE -
Particulate emission factors from Brake wear.
TIRE-
Particulate emission factors from Tire wear.
SO2-
Gaseous Sulfur Dioxide Emissions.  These are based directly on the fuel sulfur
content.
       The emission factors listed above are reported by vehicle type. The 28 vehicle types are listed
and described in Table 3.1.  They are the same classifications used in MOBILE6.0.  This is an
expansion from the twelve vehicle classifications that the PARTS model used, but each PARTS
vehicle class corresponds directly to one or to a group of MOBILE6.2 vehicle classes.
Table 3.1
MOBILE6 Vehicle Classifications

Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
MOBILE6
Abbreviation
LDGV
LDGT1
LDGT2
LDGT3
LDGT4
HDGV2b
HDGV3
HDGV4
HDGV5
HDGV6
HDGV7
HDGVSa
HDGVSb
LDDV
LDDT12
HDDV2b
HDDV3
HDDV4
HDDV5
HDDV6
HDDV7

Description
Light-Duty Gasoline Vehicles (Passenger Cars)
Light-Duty Gasoline Trucks 1 (0-6,000 Ibs. GVWR, 0-3,750 Ibs. LVW)
Light-Duty Gasoline Trucks 2 (0-6,001 Ibs. GVWR, 3,751-5750 Ibs. LVW)
Light-Duty Gasoline Trucks 3 (6,001-8500 Ibs. GVWR, 0-5750 Ibs. ALVW)
Light-Duty Gasoline Trucks 4 (6,001-8500 Ibs. GVWR, 5,751 and greater
Ibs. ALVW)
Class 2b Heavy-Duty Gasoline Vehicles (8501-10,000 Ibs. GVWR)
Class 3 Heavy-Duty Gasoline Vehicles (10,001-14,000 Ibs. GVWR)
Class 4 Heavy-Duty Gasoline Vehicles (14,001-16,000 Ibs. GVWR)
Class 5 Heavy-Duty Gasoline Vehicles (16,001-19,500 Ibs. GVWR)
Class 6 Heavy-Duty Gasoline Vehicles (19,501-26,000 Ibs. GVWR)
Class 7 Heavy-Duty Gasoline Vehicles (26,001-33,000 Ibs. GVWR)
Class 8a Heavy-Duty Gasoline Vehicles (33,001-60,000 Ibs. GVWR)
Class 8b Heavy-Duty Gasoline Vehicles (>60,000 Ibs. GVWR)
Light-Duty Diesel Vehicles (Passenger Cars)
Light-Duty Diesel Trucks land 2 (0-6,000 Ibs. GVWR)
Class 2b Heavy-Duty Diesel Vehicles (8501-10,000 Ibs. GVWR)
Class 3 Heavy-Duty Diesel Vehicles (10,001-14,000 Ibs. GVWR)
Class 4 Heavy-Duty Diesel Vehicles (14,001-16,000 Ibs. GVWR)
Class 5 Heavy-Duty Diesel Vehicles (16,001-19,500 Ibs. GVWR)
Class 6 Heavy-Duty Diesel Vehicles (19,501-26,000 Ibs. GVWR)
Class 7 Heavy-Duty Diesel Vehicles (26,001-33,000 Ibs. GVWR)

PARTS
Abbreviation
LDGV
LDGT1
LDGT1
LDGT2
LDGT2
HDGV
HDGV
HDGV
HDGV
HDGV
HDGV
HDGV
HDGV
LDDV
LDDT
2BHDDV
LHDDV
LHDDV
MHDDV
MHDDV
MHDDV

-------
22
23
24
25
26
27
28
HDDVSa
HDDVSb
MC
HDGB
HDDBT
HDDBS
LDDT34
Class 8a Heavy-Duty Diesel Vehicles (33,001-60,000 Ibs. GVWR)
Class 8b Heavy-Duty Diesel Vehicles (>60,000 Ibs. GVWR)
Motorcycles (Gasoline)
Gasoline Buses (School, Transit and Urban)
Diesel Transit and Urban Buses
Diesel School Buses
Light-Duty Diesel Trucks 3 and 4 (6,001-8,500 Ibs. GVWR)
HHDDV
HHDDV
MC
BUSES
BUSES
BUSES
LDDT
3.2    Calculation of Particulate Emission Constituents
3.2.1   Calculation of Organic Carbon (OCARBON) Emissions
       The pollutant type called OCARBON in MOBILE6.2 was formerly called Soluble Organic
Fraction (SOF) in PARTS. This type of particulate emission is generally a complex mixture of
organic chemical matter that is attached to the 'carbon' core of the particle. As the former name
implies, it is soluble in some organic solvents. The name was changed to OCARBON in the model
because it was felt that the former name (soluble organic fraction) was less precise and misleading
(i.e., soluble in which  solvent? and the output is in terms of grams per mile not a fraction or
percentage).

       Other than the name change, no changes from PARTS were made in the definition of the
pollutant, or in the values of OCARBON-related parameters in the associated calculation algorithm.
The algorithm and data parameters presented here are used to model all diesel vehicle classes for all
model years. Due to a lack of consistent and reliable data, gasoline vehicle paniculate emission
factors are not broken out into OCARBON and ECARBON, but are reported as GASPM.

       For diesel vehicles, the organic carbon emissions are calculated by first subtracting the sulfate
and lead  emission factors from the total exhaust PM  emission factor.   The remainder is then
multiplied by the organic carbon fractions (OCFRAC) to produce the OCARBON emission factor.
The values of OCFRAC are  the same  as  in the PARTS model.  The algorithm is shown
mathematically in Equation 3.1.
OCARBON
[Exh PM  -  Sulfate  -  Lead]  *  OCFRAC
    Eqn 3.1
       The values of OCFRAC are a function of the vehicle class. The following values were taken
directly from PARTS.
Vehicle Class Number
                   Vehicle Type
OCFRAC
14
15
28
16
17 and 18
                   LDDV
                   LDDT1, LDDT2
                   LDDT3, LDDT4
                   2b
                   3 and 4
0.18
0.50
0.48
0.51
0.51

-------
19,20,21,26 and 27                      5 through 7, buses                 0.44
22 and 23                               8a and 8b                        0.24
3.2.2   Calculation of Elemental Carbon (ECARBON) Emissions

       The pollutant type called ECARBON or elemental carbon in MOBILE6.2 was formerly
called Remaining Carbon Portion (RCP) in PARTS. As the former name implies it is the 'elemental
carbon' portion of the particulate after all other constituents have  been removed. Other than the
name change no changes were made in the definition of the pollutant. The algorithm presented here
is used  to model all diesel vehicle classes for all model years. Gasoline vehicle particulate emission
factors  are not broken out into OCARBON and ECARBON, but are reported only as the sum
GASPM.   The  elemental carbon emissions  are calculated by subtracting the sulfate, lead and
OCARBON emissions from the total Exhaust Particulate Emission  factor. The algorithm is shown
mathematically in Equation 3.2.


ECARBON  = [total  diesel  exhaust  PM  -  Sulfate  -
                                   OCARBON  -  lead]                     Eqn  3.2
3.2.3   Calculation of LEAD Emissions

       The lead emission factors are based directly on the quantity of lead in the automotive fuel.
The model assumes that all post-1975 model year vehicles that were not tampered with and all
calendar years subsequent to 1991 are free from lead PM emissions. The algorithm and data
coefficients used to calculate LEAD emissions are the same as those used in the PARTS model. The
frequency of leaded fuel  tampering effects (rates of tampering) are the same as those used in
MOBILE6.0.  The PARTS documentation contains a thorough explanation of these calculations.
[DRAFT User's Guide to PARTS: A Program for Calculating Particle Emissions from Motor
Vehicles - EPA-AA-AQAB-94-2, pp 48-52.]


3.2.4   Calculation of BRAKE-WEAR Emissions

       The PM brake wear emission factor was not updated from PARTS. [See PARTS User Guide
page 63.] The brake wear emission factor is assumed to be the same for all vehicle classes in the
model. It is set equal to:

BRAKE      =      0.0128 *  PSBRK                                     Eqn  3.3

where
       PSBRK =     The fraction of particles less than or equal to the particle size cutoff

-------
3.2.5   Calculation of TIRE-WEAR Emissions

       The tire wear emission factor in units of grams per mile was not updated from PARTS. It
is given by equation 3.4.  This equation is used for all vehicle classes and model years.

TIRE  =       0.002 *  PSTIRE *  WHEELS                                  Eqn 3.4

where
       TIRE is the emission factor in grams per mile
       PSTIRE is the fraction of particles less than or equal to the particle size cutoff.
       WHEELs is the number of wheels on a vehicle class.

       The value of 0.002 is the emission rate of airborne particles from tire wear [taken from
Compilation of AirPollutantEmissionFactors, Volume 2,: Stationary Point and Area Sources. EPA
(AP-42, 4th Edition)].

       The tire wear emission factors are the same as those used in PARTS with one exception. In
MOBILE6.2, number of wheels on a School Bus has been increased to 6 from 4 (the analogous
brakewear number does not change because the number of brake disks or drums is not increased by
the addition of two wheels).
3.2.6   Calculation of Sulfate and Gaseous Sulfur Dioxide Emissions

       The methodology for calculating sulfate and gaseous sulfur dioxide emissions (SO2) is based
on PARTS. [See PARTS User Guide - EPA-AA-AQAB-94-2 pp 50 to 60].  PARTS did not have
user inputs for gasoline or diesel fuel sulfur levels.  MOBILE6.2 has user-supplied fuel sulfur levels
and has extended the PARTS  algorithm to use them.

       The overall methodology  for calculating  sulfate  particulate  and SO2 emissions  in
MOBILE6.2 is based on the principal of sulfur conservation and mass balance. This means that the
sulfur contained in the gasoline or diesel fuel must be equal on a mass basis to the sulfur leaving in
the exhaust stream as sulfate and gaseous SO2 emissions.  The proportion of the fuel sulfur that is
converted to either sulfate or gaseous SO2 emissions is discussed below.
3.2.6.1 Calculation of Gasoline Vehicle Sulfate Emissions

       The gasoline vehicle sulfate emissions are a function of catalyst availability, catalyst type,
air inj ection availability, speed and vehicle fuel economy. The calculations require three parameters:
the basic sulfate emission rates (which depend on speed), the technology weighting factors (air
injection type, catalyst type, etc.), and the fuel economy values. The basic sulfate emission factors
(Table 3.4 Sulfate Emission values) were taken from the PARTS model, and are not updated for
MOBILE6.2. The vehicle fleet technology weighting factors were taken from MOBILE6.0 and are
slightly different than those used in PARTS.  The fuel economy values were also taken from the
MOBILE6.0 model, and are slightly different than those used in PARTS.

-------
Basic Sulfate Emission Factors

       The basic gasoline vehicle sulfate emission factors for all model year gasoline vehicles are
shown in  Tables 3.2 and 3.3. All emission factors except the  Sulfate emission slope (sulfate
emissions  versus fuel sulfur level) were taken from PARTS [See PARTS user guide]. The sulfate
emission factors are a function of catalyst type, air inj ection type and average speed bin. Two speed
bins are shown in the table: 19.6 MPH and 34.8 MPH. Sulfate emission levels at intermediate speeds
are calculated by linear interpolation between these two speeds. Speeds below 19.6 MPH are
considered to be 19.6 MPH and speeds above 34.8 are considered to  be 34.8 for this purpose.

       The tables contain two columns of emission values.  The first value is the sulfate emission
factor in grams per mile at a fuel sulfur level of 340 ppm sulfur (0.034 wt%). This value was taken
from PARTS and represents the fuel sulfur level of the underlying emission tests for all gasoline
vehicles. The second 'slope' value is the sulfate emission rate as a function of the fuel sulfur level
in units of [grams/mile ] per ppm Sulfur. These were calculated from a linear interpolation of the
340 ppm sulfur point, and the 0 ppm sulfur point. Logically, the 0 ppm gasoline fuel sulfur level will
produce zero sulfate emissions.

       As a result of the Tier2 rule-making for 2004 and later model years, the 340 ppm fuel sulfur
level is no longer representative of in-use vehicle fuel for these model  year vehicles.  Thus, the base
sulfate emission factors used in pre-Tier2 vehicles are unrepresentative as well. Unfortunately, there
is also no new test data at a lower sulfur fuel level such as 30 ppm in which to develop new sulfate
emission factors. To overcome this lack of representative data, the  pre-2004 model year sulfate
emission factors were ratioed down to the 30 ppm sulfur level using the 'Slopes' in Table 3.2 (also
shown in Table 3.3). These resulting sulfate levels based on 30 ppm fuel sulfur and shown in Table
3.3 then become the basis for the 2004 and later model year gasoline vehicles rather than the sulfate
emission factors shown in Table 3-2.

       In the MOBILE6.2 model the gasoline sulfur effects in Tables 3-2 and 3-3 are extrapolated
linearly to a maximum of 1000 ppm gasoline fuel  sulfur levels (600 ppm maximum sulfur in
gasoline fuel for 2000 and later model years). (The linear sulfur function was used because no data
were available to develop any other functional response.) This approximation has only a minimal
impact on MOBILE6.2's total exhaust PM emission estimates.
                                     Table  3-2
                Gasoline Vehicle  Sulfate Emission  Factors
                              thru  Model  year  2003

-------
Technology
Type
Non Catalyst
Ox Cat/No Air
3W Cat/No Air
Ox Cat / Air
3W Cat/ Air
Non Catalyst
Ox Cat/No Air
3W Cat/No Air
Ox Cat / Air
3W Cat / Air
Speed BIN
< 19.6 MPH
< 19.6 MPH
< 19.6 MPH
< 19.6 MPH
< 19.6 MPH
> 34.8 MPH
> 34.8 MPH
> 34.8 MPH
> 34.8 MPH
> 34.8 MPH
Sulfate
Emission (g/mi)
@340 ppm Sulfur
0.002
0.005
0.005
0.016
0.016
0.001
0.005
0.001
0.020
0.025
Sulfate
Emission
(g/mi*ppm S)
SLOPE
5.882e-6
1.471e-5
1.471e-5
4.7066-5
4.7066-5
2 .941e-6
1.4716-5
2 .941e-6
5.882e-5
7.3536-5

-------
Table 3-3
Gasoline Vehicle Sulfate Emission Factors
Model Years 2004 and Later
Technology
Type
Non Catalyst
Ox Cat/No Air
3W Cat/No Air
Ox Cat / Air
3W Cat / Air
Non Catalyst
Ox Cat/No Air
3W Cat/No Air
Ox Cat / Air
3W Cat / Air
Speed BIN
< 19.6 MPH
< 19.6 MPH
< 19.6 MPH
< 19.6 MPH
< 19.6 MPH
> 34.8 MPH
> 34.8 MPH
> 34.8 MPH
> 34.8 MPH
> 34.8 MPH
Sulfate
Emission (g/mi)
@30 ppm Sulfur
0.0002
0.0004
0.0004
0.0014
0.0014
0.0001
0.0004
0.0001
0.0018
0.0022
Sulfate
Emission
(g/mi*ppm S)
SLOPE
5.882e-6
1.471e-5
1.471e-5
4.7066-5
4.7066-5
2 .941e-6
1.4716-5
2 .941e-6
5.882e-5
7.3536-5
Gasoline Sulfate Emission Technology Weighting Factors

       The gasoline sulfate emission factors shown in Tables 3.2 and 3.3 by technology type are
combined into a composite all technology factor based on the technology weighting factors already
present in the MOBILE6.0 model.  Equation 3.5 is the general equation used to calculate these.
Sulfate
SUM[EF(i)  *  Frac(i)]
Eqn 3.5
Where EF(i) are the sulfate emission factors in Table 3-2 dn 3-3, Frac(i) are the technology fractions,
and indexing by 'i' represents summation over the technology categories and MOBILE6 vehicle
speed bins. The technology fractions are functions of vehicle type and model year that are calculated
in MOBILE6.2 based on vehicle technology distributions already present in MOBILE6.0.  The
technology fraction topic is discussed in detail in EPA report M6.FLT.008A.
3.2.6.2 Calculation of Gasoline Vehicle SO2 Emissions
       The model assumes that all of the sulfur in the fuel is exhausted either as sulfate emissions

-------
or gaseous sulfur dioxide emissions (SO2).  Thus, once the sulfate emissions are calculated, the
remaining sulfur in the fuel is considered to be exhaust SO2.
       The first step in this calculation is to determine the fraction of the gasoline fuel sulfur that
is  converted to sulfate emissions (DCNVRT).   This is  done by using the gasoline fuel  sulfur
relationship from PARTS shown in Equation 3.6.  A value of DCNVRT is calculated for each of the
technology and speed groups.

DCNVRT = Sulfate  *  FE /  [UNITS *  (l.+WATER) *  FDNSTY  *SWGHT ]           Eqn 3.6

Where:

DCNVRT - percent of sulfur in the fuel that is directly converted to sulfate.
Sulfate - is the direct sulfate emission factor of a vehicle in g/mi calculated from Table 3.4a or Table
3.4b.
WATER - is the constant 1.2857 (see PARTS User Guide).
FDNSTY - is the fuel density. It is a constant value of 6.09 Ib/gal.
FE - is the fuel economy in miles per gallon. (These values come from MOBILE6).  They are a
function of model year and vehicle class.
SWGHT - is the weight percent of sulfur in the fuel, (i.e., 0.034 = 340 ppm gasoline fuel sulfur).

UNITS - is the constant 13.6078.  This is the units conversion factor. Calculated by (453.592 *
3)/100. Where 453.592 is the number of grams in a pound, 3 is the weight ratio of SO4 to sulfur,
and the 100 is to correct for the weight percent of sulfur.

       The gaseous SO2 emissions are calculated as in PARTS by plugging the values of DCNVRT
into the SO2 emission equation (Eqn 3.7), and solving for SO2 for each technology and speed group.

S02 =  UNITS#2  *  FDNSTY *  SWGHT  *  (1.  -   DCNVRT)  /  FE         Eqn  3.7

Where:

UNITS#2 = 9.072. This is the units #2 conversion factor. Calculated by (453.592 * 2)/100. Where
453.592 is the number of grams in a pound, 2 is the weight ratio of SO2 to sulfur, and the 100 is to
correct for the weight percent of sulfur.

       The final composite SO2 emission factor is calculated by weighing together the individual
technology and speed SO2 emission factors.  The same weighting factors are used for both Sulfate
and SO2 emissions.

Mathematically, it is shown in Equation 3.8.

Composite  S02   =      SUM[S02(i)  *  Frac(i)]                       Eqn  3.8


Where SO2(i) are the emission factors calculated in the gaseous SO2 Equation X, Frac(i) are the
technology fractions, and indexing by i represents that the summation is  over the technology types
and MOBILE6 speed bins.

-------
3.2.6.3 Gasoline Vehicle Sulfate and SO2 Emission Sample Calculation

      This section provides a sample calculation for the gasoline fueled vehicle sulfate and SO2
emission factors for two technology and speed groups (no weighting factors will be used). It is
provided to give the reader a feel for the relative size of the Sulfate and gaseous SO2 emission
factors.

Sulfate emission conversion:
Speed Bin > 34.8 MPH

DCNVRT =  Sulfate *  FE /  [UNITS *  (l.+WATER) *  FDNSTY *SWGHT  ]

DCNVRT     =     (0.001 g/mi *  25 mile/gal)  /
                  [ 13.6078  *  (1+1.2857) *  6.09  Ib/gal  *  0.034%

DCNVRT     =     0.00384 or 0.39% for  the 3-way catalyst  no air pump
                  group.

DCNVRT     =     0.0970  or  9.70% for  the  3-way catalyst  with  air
                  pump group.

Gaseous S02 Emissions:

S02  = UNITS#2  * FDNSTY  *  SWGHT *  (1. - DCNVRT)  / FE

S02   =     9.072 *  6.09  Ib/gal  * 0.034% * (1-0.00384)  /  25

S02   =     0.0748 g/mi for the  3-way catalyst  no air  pump  group at
            340  ppm  gasoline fuel sulfur.


3.2.6.4 Calculation of Diesel Vehicle Sulfate Emissions

      The diesel vehicle sulfate emissions are a function of the basic user supplied diesel fuel sulfur
level (a required input for PM emission calculation in MOBILE6.2), and the diesel vehicle fuel
economy values. The fuel economy values currently in use for the for diesel vehicles were taken
from the MOBILE6 emission model.  Future versions of the  MOBILE6  model  (Version
MOBILE6.3) may contain updated fuel economy estimates and allow user input of alternative values.
Sulfate emissions are calculated for diesel fueled vehicles in MOBILE6.2 by using Equation 3.9.



Sulfate = UNITS * (1 .+WATER)*DFDNSTY*DWGHT*DCNVRT / FE   Eqn 3.9

      Where:

-------
Sulfate is the direct sulfate emission factor of a vehicle in g/mi.
WATER is the constant 1.2857.
DFDNSTY is the constant 7.11 Ib/gal.
FE is the fuel economy in miles per gallon. (These values are to come from MOBILE6). They are
a function of model year and vehicle class.
DWGHT weight percent of sulfur in the fuel,  (i.e., 0.050 = 500 ppm diesel fuel sulfur).
DCNVRT percent of sulfur in the fuel that is directly converted to sulfate. MOBILE6.2 retrains the
2% value of this parameter from PARTS.
UNITS is the constant 13.6078.
       Sulfate emissions for diesel vehicles are calculated using the assumption from PARTS that
2 percent of the sulfur in the diesel is converted into sulfate compounds, and the remaining sulfur
is converted to SO2 compounds.
3.2.6.5 Calculation of Diesel Vehicle Gaseous SO2 Emissions

       The diesel  vehicle gaseous  SO2  emissions  are calculated  using  equation 3.10.  The
methodology assumes that the 98 percent of the fuel sulfur is converted to gaseous SO2 emissions.
Like the calculation for the diesel vehicle sulfate  emissions, the gaseous  SO2 emissions are a
function of user input fuel sulfur level and the vehicle fuel economies.

S02  = UNITS#2  *  FDNSTY  *  SWGHT  *  (1.  - DCNVRT)  /  FE        Eqn 3.10
3.2.6.6 Diesel Sulfate Emissions on Vehicles with Particulate Trap Technology

       It is anticipated that future technology needed to meet strict paniculate matter standards for
diesel vehicles in model years 2007 and later will include particulate traps.  Such traps may take a
variety of designs; however, the basic principle is for the trap to collect virtually all of the particulate
matter present in the exhaust stream, and to either burn it off at high temperature or to otherwise
remove it from the exhaust stream.  Currently, no data exist as to efficiency of this process nor do
any sulfate emission factor data exist to suggest the magnitude of such emissions. Thus, the model
will  set in calendar years 2007 and later, a very low base diesel fuel sulfur level of 10 ppm as
required in the Heavy-Duty 2007 Rule, and continue to predict that, fleet-wide, 2 percent of this fuel
stream is emitted as sulfate emissions.
3.2.7   Calculation of Total Exhaust PM Emissions

3.2.7.1 Diesel Vehicles

       Total Exh PM Calculation

-------
       The general equation for total exhaust particulate emissions is shown in Eqn 3.10b.  It
includes OCARBON,  ECARBON and  Sulfate  emissions.  Calculation details on these  sub-
components have been previously discussed in Sections 3.2.6.1 through 3.2.6.6.

       Total  Exh  PM    =  OCARBON  + ECARBON  + Sulfate          Eqn  3.1Ob
       The default total exhaust particulate emission rates are represented as a linear function with
respect to mileage. For the light-duty diesel vehicles the rates were taken from the PARTS model
(See EPA report EPA-AA-AQAB-94-2). The default total exhaust particulate parameters for heavy-
duty diesel vehicles are also a linear function, and are shown in the MOBILE6.0 technical support
materials - see EPA reports M6.HDE.OO 1, M6.HDE.002, and M6.HDE.004. They can also be found
in the support materials section of the EPA 2007 heavy-duty rule making docket.

       The total exhaust PM emission rates in MOBILE6.2 are a function of vehicle class (all diesel
classes can have a separate emission factor), model year (1950  - 2020+), and mileage. The mileage
relationship is  linear with a zero mile emission level, two possible slopes and  a user supplied
inflection point between the two slopes (Equation 3.11).


Exh  PM =  ZML +  DETl*mileagel  +  DET2(mileage2-mileagel)    Eqn  3.11a
       The default values of these parameters are provided in the Excel Spreadsheets PMDZML.csv,
PMDDRl.csv  and  PMDDR2.csv.  Examination  of the heavy-duty  emission rates in  these
spreadsheets shows that in virtually all cases the zero mile emission level is assumed to be the
certification standard, and the deterioration rates with respect to mileage are zero.
       Total Exh PM Size Correction Factors

       The total exhaust PM emission factors are computed on the basis of the entire amount of PM
material that is collected on an EPA test filter during the emission tests. This is referred to as PM30.
Exh PM calculated in Eqn 3.1 la is in terms of PM30.

       For use in the MOBILE6.2 model, the particulate emissions must be converted from PM30
terms into particulate  size terms that can range from PM1 to PM10. The general equation for any
size in the range of 1 micron to 10 microns (x) for this transformation is given in Eqn 3.1 Ib.

Exh PM(x)  =      Exh  PM(30)  *  SIZE  CF                          Eqn 3.lib


       This value is not allowed to exceed the certification standard applicable to future years if
future rule-makings are being modeled.

-------
       The values for the SIZE CF used in Eqn 3.1 Ib are shown in Table 3.4. Correction factors
are provided for the range  of  pollutant type, vehicle/fuel classes and particle sizes.  Linear
interpolation should be used to calculate correction factors for particle sizes between those listed in
Table 3.4.

-------
                                  Table  3.4
  Fraction of Particulate Mass Less than or Equal to the Particle Size Cutoff
Vehicle Type/
Particulate Component
Particle Size
Cutoff (PSO
Fraction of Particles less than or
Equal to the Particle Size Cutoff
Gasoline vehicles using
leaded fuel/ Lead, Carbon
      10.0
       2.0
       0.2
             0.64
             0.43
             0.23
Gasoline vehicles with catalyst,
using unleaded fuel/ Lead, Carbon
      10.0
       2.0
       0.2
             0.97
             0.89
             0.87
Gasoline vehicles without a
catalyst, using unleaded fuel/
Lead, Carbon

Diesel vehicles/ Exhaust PM
      10.0
       2.0
       0.2

      10.0
       2.5
       2.0
       1.0
             0.90
             0.66
             0.42

             1.00
             0.92
             0.90
             0.86
All vehicles/ Brake-wear
      10.0
       7.0
       4.7
       1.1
       0.43
             0.98
             0.90
             0.82
             0.16
             0.09
All vehicles/ Tire-wear
      10.0
       0.10
             1.00
             0.01
Reference:

"Size  Specific  Total Particulate Emission  Factors  for  Mobile
Sources",  EPA 460/3-85-005
      Calculation of OCARBON and ECARBON
      The total exhaust particulate emission factor corrected for particulate  size (Exh PM)

-------
calculated in Equation 3.lib is substituted into Equations 3.12 and 3.13 (rewrites of Equation 3.1
and 3.2 where lead is zero for diesel vehicles) to calculate OCARBON and ECARBON emission
factors.  The appropriate sulfate emission factors corrected for particulate size are also substituted
into the two equations to account for these constituents. The sulfate emission factor is the "base"
diesel fuel sulfur level 'Sulfate[b]'. For pre-2007 model years this base level is 500 ppm sulfur. For
2007+ it is 8 ppm sulfur. It is subtracted from the OCARBON and ECARBON emission factors
CJDCARBON =  [C_EXH_PM -  Sulfate[b]]  *  OCFRAC                 Eqn 3.12

C_ECARBON =  [C_EXH_PM -  Sulfate[b]  - OCARBON]                 Eqn 3.13


3.2.7.2 Gasoline Vehicles

       The GASPM emission factors are supplied as a function of vehicle class, catalyst technology,
model year (1950 - 2020+), and mileage.  The mileage relationship is linear with a zero mile
emission level, two possible slopes and a user supplied inflection point between the two slopes
(Equation 3.14).  The default values of these parameters are provided in Excel Spreadsheets
PMGZML.csv, PMGDRl.csv and PMGDR2.csv.

GASPM =  ZML  + DETl*mileagel  + DET2(mileage2-mileagel)      Eqn 3.14

       The default particulate parameters (zero mile and deterioration rates) are taken from PARTS
values for gasoline vehicles (See EPA report EPA-AA-AQAB-94-2). As a result, in all cases the
deterioration rates DET1 and DET2 are assumed to be zero.

       The sulfate emission factors for gasoline vehicles are calculated according to the equations
discussed in Section 3.2.6.1.  The calculated sulfate emission factor is based on the user specified
fuel sulfur level [i] rather than on the "base" level at which original emission factor testing was done.
The lead emission factors are calculated according to the algorithm referenced in Section 3.2.3.
They are a function of technology,  model year, existence of tampering and calendar year.  They
cannot be changed by the user.

       The Exhaust PM emission factor for gasoline vehicles is the sum of the GASPM, sulfate and
lead emission factors, and is shown mathematically in Equation 3.15.

Exhaust  PM  (gas vehicles)  =  GASPM  + sulfate[i]  +  lead      Eqn 3.15

       Like the diesel vehicles, the Exhaust PM emission factor for gasoline vehicles is compared
against the certification standard level and capped at this level if it exceeds it.  This will typically not
happen except in the case of the 2004+ Tier2 emission vehicles which have stringent PM standards.

       The Total Exhaust PM emissions for gasoline vehicles are also adjusted  for particle size
using a particle size distribution function. These particle size correction factors are taken directly
from PARTS and are tabulated in the PARTS User Guide. Mathematically, the calculation is shown
in Equation 3.16.

-------
Total  Exhaust PM     = Exhaust  PM  * Particle  Size Corr      Eqn  3.16



3.3   Ammonia Emission Calculations

3.3.1  Ammonia Emission Factors

      The MOBILE6.2 model calculates a composite, FTP test based (composite running and start
emissions) gaseous ammonia emission factor for all vehicle types and model years.  The base
ammonia emission factors built into the  MOBILE6.2 model were taken from the EPA report
EPA/AA/CTAB/PA/81 -20 "Determination of a Range of Concern for Mobile Source Emissions of
Ammonia" by Robert Garbe, August, 1981.  They can also be found in SAE paper 830987.  They
were selected for use in MOBILE6.2 because  of their established use in EPA's National Trends
modeling for many years, and a lack of new ammonia emission test results. Because the emission
factors are about 20 years old, a literature search was conducted to verify that they are still
representative of current vehicles.  A description of this literature search is contained in Appendix
A.
      The ammonia emission factor values used in the MOBILE6.2 model are shown in Table 3.5.
All units are milligrams per mile.

-------
Table 3.5
Ammonia Emission Factors by Vehicle Class and Catalyst Type
Intercept Values in Regression
(all UNITS are Milligrams per Mile)
MOBILE 6
Vehicle Types
1-5 (LDG)
24 (MC)
6 - 13, 25
(HDG)
14, 15, 28
(LDD)
16 - 23, 26, 27
(HDD)
All

11.265
mg/mi
45.062
mg/mi
6.759
mg/mi
27.037
mg/mi
Non
Catalyst
11.265
mg/mi




Ox
Catalyst
15.128
mg/mi




3 -Way
Catalyst
101.711
mg/mi




LDG are the light-duty gasoline vehicles
MC is the motorcycle class
HDG are the heavy-duty gas vehicles
LDD are the light-duty diesel vehicles
HDD are the heavy-duty diesel vehicles.
       Based on the literature search, EPA concluded that these numbers are in the same general
range as the limited FTP test results, and thus are appropriate for use in MOBILE6.2.  However,
there is substantial variation in ammonia measurements and ammonia is likely a function of sulfur
level, test cycle (FTP versus US06), advancing catalyst technology, and other factors. Additional
research is recommended on this topic.

       The gaseous ammonia emission factors are  reported by the MOBILE6.2 model in the
particulate section because gaseous ammonia reacts with sulfates and/or nitrates to form ammonium
sulfate and ammonium nitrate in the atmosphere.  These ammonium  compounds are classified as
particulate emissions. The MOBILE6.2 model calculates and reports only the gaseous emissions
emitted directly from a vehicle tailpipe.  It makes no attempt to model the atmospheric chemistry of
ammonia conversion to other ammonium based compounds or estimate the direct emissions from
ammonium compounds.  These types of calculation are left to atmospheric chemistry models.
3.4    Indirect Sulfate Emission Calculations
       In addition to the direct sulfate emission  factors  discussed above, the previous model

-------
(PARTS) estimated an indirect sulfate emission factor by assuming that a fraction of the gaseous
sulfur dioxide emissions are later converted in the atmosphere to sulfate material. Based on ambient
sulfur and sulfate measurements in 11 cities, EPA estimated that 12 percent of all gaseous sulfur is
converted to sulfate.

       During the update process for MOBILE6.2 it was decided to drop this calculation from the
model  and not report an estimate for indirect sulfate emission production. The reasoning for this
decision is that the MOBILE6.X series of models are vehicle emission models not atmospheric
models.  They are  best used for estimating emission factors for pollutants directly emitted from
vehicles through pathways such as exhaust, evaporation, brake and tire, and engine draft (PCV),
rather than atmospheric chemical reactions.
3.5    Fugitive Dust Emission Calculations

       MOBILE6.2 does not include estimates of fugitive road dust emissions.  These will be
covered by a simple calculation tool being developed separately by EPA's Office of Air Quality
Planning and Standards (OAQPS). They were removed from the MOBILE6.2 because a new tool
will be available, and because MOBILE6 cannot properly account for the facility / roadway type -
unpaved roads.  Since dust emissions on an unpaved road are usually considerably higher than on
a paved road, the issue of paved versus unpaved roads is critical in any modeling or discussion of
fugitive dust emissions.

       The AP-42 methodology for calculating fugitive road dust does not permit attributing the
emissions to particular vehicle classes in cases where multiple vehicle classes use the road. This also
significantly complicates integrating the road dust calculation into MOBILE6. An EPA  web site
where the current AP-42 methods for calculation fugitive dust are explained is:

http://www.epa.gov/ttn/chief/index.html

The reader is encouraged to browse this web site for the necessary information on fugitive dust.

-------
4.     Results from the MOBILE6.2 Model
       Some limited and preliminary results from the MOBILE6.2 model are shown at the end of
this document,  and are discussed in  this section.  (The notation in the Charts lists the term
MOBILE6.1 at the top of the charts. The term MOBILE6.1 is completely synonymous with the
name MOBILE6.2.)
4.1    Emissions Versus Calendar Year

       These results are shown in a series of Figures (Figures 1 through 15). With the exception of
the Ammonia results,  all the  figures were constructed as comparisons of the MOBILE6.2 and
PARTS emission results.  They are shown in terms of total particulate emissions (TOTEX), total
carbon emissions from gasoline vehicles (GASPM), sulfate emissions and lead emissions.  All of
the results in these figures are shown as a function of calendar year.  The results are also shown for
individual vehicles types: light-duty gasoline vehicles, light-duty truck class 4 vehicles, heavy-duty
gasoline vehicles, light-heavy, medium-heavy, and heavy-heavy duty  diesel vehicles and transit
diesel buses. The emission results in all of the figures are the average emission levels for each
calendar year from 1970 through 2020. A calendar year includes the weighted average emission
result of the current calendar year plus the previous 24 model years. All of the results are shown
in terms of PM10 emissions (i.e., particles less than 10 microns in diameter). Results shown in terms
ofPM2.5 emissions would also be consistent between the MOBILE6.2 model and the PARTS model
since particulate size determination algorithms did not change between the two models.

       Figures 1 through 4 show the results from the light-duty gasoline vehicles. Figure 1 shows
the TOTEX (total exhaust particulate emission) results from both MOBILE6.2 and PARTS. As can
be seen from the figure, only relatively small differences between the two models are observed. The
differences occur mostly in the pre-1980 years and in the post 1996 calendar years. In the early years
they are caused by differences in the underlying methodology of modeling misfueling and tampering
effects on lead particulate emissions. The differences are not due to changes in the basic lead
emission factors.  The differences in the later years are due to  different fuel sulfur levels that create
differences in sulfate emission factors. The PARTS model does  not allow alternate gasoline fuel
sulfur levels to be modeled, and fixes this fuel parameter at 343 ppm. However, the MOBILE6.2
model allows alternative sulfur levels to be modeled. The fuel sulfur level was set at 30 ppm for all
2000 and later calendar years.  Figure 2 shows the carbon particulate emissions comparison  for
light-duty gasoline vehicles without the contribution of lead and sulfate emissions. As can be seen
in Figure  2, once the  different lead and sulfur influences are removed, the  carbon particulate
emissions (GASPM) are shown to be  very similar between MOBILE6.2 and PARTS.   Any
differences between the  two  models is  due to different fleet assumptions such as technology
distributions or mileage accumulations.

       Figure 3 illustrates the impact of different fuel sulfur levels on the sulfate emission factors.
In the later calendar years, the higher PARTS sulfate curve shows the effect of a high fuel sulfur level

-------
(343 ppm), and the lower MOBILE6.2 sulfate curve reflects a reduced sulfur level of 30 ppm after
1999. The rising and/or relatively high sulfate emission levels from the early 1980's until the 1990's
depicted in both models is the result of increasing penetration of air injection systems on vehicles.
Vehicles with air injection systems typically produce more sulfate and less gaseous SO2 than those
without such systems. The very low sulfate emission factors after 2004 show the effect of very low
fuel sulfur and the virtual elimination of air injection  systems that lead to higher sulfate emission
factors.

       Figure 4 compares the lead emission factors between MOBILE6.2 and PARTS. The figure
shows that some differences occur in the pre-1980 model years due to a slightly different
methodology of accounting for tampering and misfueling, and the different technology fractions in
MOBILE6.2 and PARTS.  Both models show zero lead emissions after 1991 due to the complete
phase-out of lead in gasoline.

       Figures 5 and 6 show the average carbon PM10  emission results for light-duty gas truck class
4 and heavy-duty gas trucks for the MOBILE6.2 and PARTS models. As for the light-duty vehicles
(i.e. cars), the results for MOBILE6.2  and  PARTS for the trucks  are relatively close for  most
calendar years. The differences for the light-duty gas trucks can explained in terms of truck size and
different fleet and technology distributions between the two models. For instance, the figure shows
the results for a light-duty gas truck class 4 in MOBILE6.2, but an average result for light-duty gas
truck class 3 and 4 for PARTS (PARTS did not separate class 3 and 4 trucks). Also, sulfate and lead
emission differences between the two models are also present in an analogous fashion as the light-
duty gasoline vehicle graphs on "carbon PM".

       The heavy-duty gas truck result comparison in Figure 6 shows differences that are mostly
technology related (different fleet phase-ins for fuel injected, air injection and catalyst technology)
for emissions in the 1990 through 2005  calendar years. After calendar year 2008, the lower PM
emission factors from MOBILE6.2 are the effect of the EPA Tier2 emission standards. These new
standards are modeled in MOBILE6.2, but not in PARTS.

       Figures 7 through 10 show the TOTEX comparisons for the diesel vehicles. Comparison of
the results in Figure 7 for  Class 2B diesels shows MOBILE6.2 to be slightly higher for calendar years
prior to 2007 and considerably lower for calendar years after 2007.  For medium heavy-duty diesels
shown in Figure 8, the relationship is just the opposite for  the pre-2007 vehicles, but is similar for
the 2007 and later trucks. For heavy-heavy duty vehicles  shown in Figure 9 there are only slight
differences for the pre-2007 calendar years, but similar lower emission factors for calendar years
following 2007. The results for the Buses show the largest differences with MOBILE6.2 predicting
considerably higher PM  emissions for all but the latest calendar years shown in Figure 10.

       The differences in the pre-2007 calendar years arise primarily because the MOBILE6.2
results reflect a more recent analysis done to support the EPA heavy-duty diesel 2007 rule whereas
PARTS reflects an older  analysis. The new analysis had an eight percent compliance margin to the
standards overall, and in a few vehicle class cases (medium heavy-duty vehicles), some very small
amounts of deterioration  versus mileage. However, in  most cases the direct regulated PM emission
factors in units of grams  per brake horsepower-hour (g/bhp-hr) did not change much.  Instead, the

-------
conversion factors changed. This was particularly true in the case of the buses. Also, in the case of
the buses the comparison is more difficult due to the change in category definition. In PARTS, the
buses category included all buses, in MOBILE6.2 they are broken out between urban transit buses
and school buses.

       All of the Figures 7 through 10 show considerably lower PM emissions for 2007 and later
vehicles with MOBILE6.2 being much lower than PARTS. This is due to the implementation of the
new stringent 2007 diesel rule. In general, this rule will lower  PM emissions by an order of
magnitude from 0.1 g/bhp-hr to 0.01 g/bhp-hr.  The effects of this rule were not accounted for in
the PARTS model.

       Figure  11 compares the MOBILE6.2 and PARTS model results for sulfate emissions on
heavy-heavy duty diesel trucks. The PARTS curves are the default emission results that cannot be
modified by the user. They typically are based on very high diesel fuel sulfur levels of 2500 ppm,
and then a lower level of 500 ppm sulfur for all 1993 and later model years.  The MOBILE6.2 results
are based on 500 ppm sulfur for pre-2007  calendar years and 8 ppm diesel fuel sulfur for 2007 and
later calendar years. Note that the MOBILE6.2 and PARTS sulfate curves agree when the fuel sulfur
levels are the same at 500 ppm. If all the calendar-year MOBILE6.2 runs had been done  at the same
fuel sulfur levels as the PARTS runs, the curves would agree for all calendar years. Instead, alternate
fuel sulfur levels were modeled (10 ppm fuel sulfur for 2007 and later) to show the effect of different
diesel fuel sulfur levels on sulfate emissions.

       Figure 12 shows the Ammonia emission factors as a function of calendar year and vehicle
class. As can be observed, the diesel emission factors are not a function of calendar year, but the
gasoline vehicle factors are a function of calendar year.  The gasoline vehicles show a rising and then
a flattening curve of ammonia as calendar year progresses. This rising curve is due to the fact that
modern fuel inj ection and 3 -way catalyst technology has a greater tendency to produce ammonia than
the older non catalyst or oxygenated only catalyst equipped vehicles. The flattening aspect of the
curve reflects the almost complete penetration of fuel injected and 3-way catalyst vehicles into the
fleet by calendar year 2010.

       Figure 13 compares the MOBILE6.2 and PARTS exhaust carbon particulate emissions for
motorcycles. The figure shows close agreement between MOBILE6.2 and PARTS. In both figures
the emissions start out at fairly high levels in the 1970s and drop to considerably lower levels in the
1990 and beyond due to technology improvements.

       Figure 14 compares the MOBILE6.2 and PARTS exhaust carbon emissions for light-duty
diesel vehicles. The figure shows fairly good agreement between the models with similar overall
trends. The models diverge after 2007 because of the incorporation of the effects of 2007 diesel rule
on the MOBILE6.2 emission factors and the lack of such an effect in PARTS. The PARTS graph
shows an unusual  'dip and increase' in emission factors in the 1980 to 1989 calendar years. This
effect is not due to rising general  emission  factors in the model,  but changing registration
distributions between individual model years. For example, in the calendar years where the emission
rate is increasing  the overall LDDV fleet is getting older (new model years are replacing older
vehicles at a slower rate).

-------
       Figure 15 compares the MOBILE6.2 and PARTS exhaust carbon particulate emissions for
light-duty diesel trucks. This figure is analogous to Figure 14 for the LDDVs. It also shows fairly
good agreement between the models with similar overall trends.  The models diverge after 2007
because of the effects of 2007 diesel rule on the MOBILE6.2 emission factors and the lack of such
effect in PARTS.
4.2    Emissions Versus Model Year

       Figure  16 (the only figure based on model year instead of calendar year) presents Total
Exhaust particulate emissions versus model year for the 8B heavy-duty diesel vehicles in calendar
year 2010.  These results show the basic emission factor for 8B diesel vehicles for each individual
model year prior to the application of weighting factors and correction factors.  In comparison, the
results shown in Section 4.1 are by calendar year where each calendar year is a weighted average of
the emission factors from the previous 25 model years.

       The results in Figure 16 show that the 8B and other heavy-duty diesel vehicle basic emission
factors are NOT precisely  the same as those from PARTS.  The differences in Figure 16 occur
because different emission  factors were used to model heavy-duty diesel vehicles in the EPA 2007
Heavy-Duty Rule-making effort than in PARTS. The differences are most notable in model years
1984 through  1989 where the new MOBILE6.2  emission factors now include the effects of
deterioration of particulate emissions versus vehicle odometer. Also, the MOBILE6.2 particulate
emission factors for the 2007 and  later model years are  lower than the corresponding PARTS
emission factors due to the effects of the 2007 rule-making.

-------
                           Figure 1
      Total Exhaust PM10 Emissions from MOBILES.1 and
                      PARTS forLDGVs
0.30
0.00
                                                        •MOBILE6.2
                                                        •PARTS
   1970  1975  1980  1985  1990  1995  2000  2005  2010 2015  2020
                       Calendar Year

-------
                          Figure 2
0.14
0.12
            MOBILE6.1 GASPM Emissions Versus
            PARTS Carbon Emissions  for LDGVs
                                   •••••••••••••••••a
   1970 1975  1980  1985  1990 1995  2000  2005 2010  2015  2020
                       Calendar Year

-------
                          Figure 3

       Comparison of MOBILE6.1 and PARTS SULFATE
                    Emissions for LDGVs
0.012
0.002
0.000
                                                    -0-MOBILE6.2
                                                    -m- PARTS
   1970  1975  1980 1985  1990 1995  2000 2005  2010  2015 2020
                      CALENDAR YEAR

-------
                          Figure 4

    Comparison of MOBILE6.1 and PARTS LEAD Emissions
                         for LDGVs
0.16
                                                      •MOBILE6.2
                                                      • PARTS
   1970 1975  1980  1985 1990  1995  2000 2005  2010  2015 2020
                     CALENDAR YEAR

-------
                          Figure 5

      GASPM Emissions from LDGT4 in MOBILE6.1 and
                      LDGT2 in PARTS
0.25
0.00
   1970 1975  1980 1985  1990  1995  2000  2005 2010  2015 2020
                     CALENDAR YEAR

-------
                           Figure 6

       MOBILE6.1 and PARTS GASPM from Heavy-Duty
                      Gasoline Vehicles
0.25
                                                        •MOBILE6.2
                                                         PARTS
0.00
   1970  1975  1980  1985  1990  1995  2000  2005 2010  2015  2020
                        Calendar Year

-------
                          Figure 7
    MOBILE6.1 and PARTS Total Exhaust PM10 Emissions
             from 2B Heavy-Duty Diesel Vehicles
0.70
0.00
                                                       •MOBILE6.2
                                                       •PARTS
   1970  1975  1980  1985  1990 1995  2000  2005  2010 2015  2020

                      CALENDAR YEAR

-------
                           Figure 8
0.00
    MOBILE6.1 and PARTS Total Exhaust PM10 Emissions
           from Medium Heavy-Duty Diesel Vehicles
                                                       •MOBILE6.2
                                                        PARTS
   1970  1975  1980  1985  1990 1995  2000  2005 2010  2015  2020
                       Calendar Year

-------
                       Figure 9
     Comparison of MOBILE6.1 and PARTS TOTAL
     EXHAUST PM10 for Heavy-Heavy Duty Diesels
                                                -0-MOBILE6.2
                                                -*- PARTS
1970  1975  1980  1985  1990  1995  2000  2005 2010  2015 2020
                  CALENDAR YEAR

-------
                             Figure 10
     MOBILE6.1 and PARTS Total Exhaust PM10 Emissions from
                        Urban Diesel Buses
3.50
3.00
0.50
0.00
                                                             •MOBILE6.2

                                                             • PARTS
   1970  1975  1980  1985  1990  1995  2000  2005  2010  2015  2020
                          Calendar Year

-------
  0.30
  0.25
  0.20
"3)
  0.15
  0.10
  0.05
  0.00
                             Figure 11

          MOBILE6.1 and PARTS SULFATE Emissions for
                             HHDDVs
000000000o°oooooooooooo
                             *•••••
                                                     •MOBILE6.2
                                                     • PARTS
     1970  1975  1980  1985 1990  1995  2000  2005  2010  2015  2020
                        CALENDAR YEAR

-------
                             Figure 12
                Ammonia Emissions from MOBILE6.1
0.12
      DOOOOOOOOOO OA3 O O aO OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
0.00
   1970   1975  1980  1985   1990   1995  2000   2005  2010  2015   2020
                            Calendar Year

-------
                       Figure 13
    MOBILE6.1 and PARTS GASPM from Motorcycles
1970  1975  1980  1985  1990  1995  2000  2005  2010  2015  2020
                      Calendar Year

-------
                             Figure 14
       MOBILE6.1 and PARTS Carbon Emissions from LDDV
  0.80
                                                            •MOBILE6.2
                                                            • PARTS
O 0.20
  0.10
  0.00
     1970  1975  1980  1985  1990  1995  2000  2005  2010  2015 2020
                          Calendar Year

-------
                          Figure 15
     MOBILES.1 and PARTS Carbon Emissions for LDDT3,4
                   and LDDT (Respectively)
0.80
                                                        •MOBILE6.2
                                                        •PARTS
0.00
   1970  1975  1980  1985  1990  1995  2000  2005  2010  2015 2020

                       Calendar Year
                          Figure 16

-------
        PM10 Emission Factors for 8B Diesel Vehicles
                    In Calendar Year 2010
2.25
2.00
                                              MOBILE6.2
                                              PARTS
  1980     1985    1990     1995    2000

                          Model Year
2005
2010
2015

-------
                                     Appendix A

                     Literature Search on Vehicle Ammonia Emissions

       The ammonia emission factors used in the MOBILE6.2 model are based on a 1981 EPA
study which tested only limited numbers of 3-way catalyst vehicles. Thus, as part of the MOBILE6.2
update, EPA did a literature search to determine if other ammonia emission estimates were available,
and to determine if the MOBILE6.2 estimates based on this study were appropriate.

 Recent studies on vehicle ammonia emissions by various researchers have suggested that gaseous
exhaust ammonia emissions may be dependent on catalyst type, vehicle operation and fuel sulfur
levels.  The 1981 study does take different catalyst types into account (although, the 1981 3-way
catalyst may not reflect modern technology). However, it did not address ammonia emissions as a
function of vehicle operation or fuel sulfur levels. As a result, the ammonia emission factors in the
MOBILE6.2 model may be only partially representative of modern vehicles.

Various Studies

1989 Volkswagon Study - Several gasoline and diesel vehicles were studied using the FTP test. The
gasoline sulfur level was 330 ppm.  Non-catalyst gasoline vehicles reported results of 3.52 mg/mi,
diesel vehicles 1.88 mg/mi and 3-way catalyst vehicles 137.4 mg/mi.

Preliminary CE-CERT Work in Calendar  Year 2000 -  Seven vehicles tested so far over three
different fuel sulfur levels (324 ppm and 30 ppm and California reformulated fuel).  The vehicles
were a  1991 Dodge, a 1997 Ford, a 2001 Buick, a 1999 Ford Tierl, a 2001  Suzuki NLEV, a 1999
GM Sonoma TLEV, a 2000 Ford Winstar ULEV. All were 3-way catalyst technology.
Table A-l
CE-CERT Vehicle Test Results of Ammonia Emissions

Vehicle
1991 Dodge
1997 Ford
2001 Buick*
1999 Ford*
2001 Suzuki*
1999 GM*
2000 Ford*
FTP
30 ppm
Sulfur
118 mg/mi
38 mg/mi

70 mg/mi

12 mg/mi
73 mg/mi
324 ppm
Sulfur
86 mg/mi
5 mg/mi





US06
30 ppm
Sulfur
210 mg/mi
237 mg/mi
160 mg/mi
242 mg/mi
415 mg/mi
82 mg/mi
307 mg/mi
324 ppm
Sulfur
161 mg/mi
146 mg/mi






-------
 *  Tested  on California Reformulated  Fuel rather than the  fuels
 with  the  specified sulfur levels  of  30  and  324  ppm.
ORD National Risk Management Research Laboratory Work -One 1993 ChevroletLumina(3-way
catalyst) was tested over various driving conditions (FTP, steady state, hard acceleration, partial and
major enrichment, and some on road data). The FTP ammonia emissions were about 30 mg/mi.
This is lower than other studies. However, the hard acceleration results were 282 mg/mi, and the
major enrichment results were 2,450 mg/mi.
G. Cass Work - California Institute of Technology - These were roadway tunnel studies in Los
Angeles in 1998.  The results were 98 mg/mi ammonia for the fleet as a whole.  116 mg/mi for
LDGV.
A. Kean Work - Lawrence Berkeley Labs - A San Francisco Bay area tunnel study in 1999. Results
79 mg/mi overall fleet result.

M. Baum Work - Oak Crest Institute of Science. CA. - They used remote sensing measurements.
Emissions were measured from vehicles during acceleration in parking lots and freeway ramps.
Results showed very high emissions. The results are available only in ppm (78.6 ppm average). 66%
of the ammonia emissions are emitted by 10% of the fleet. M-85 fueled vehicles had slightly higher
ammonia emissions.

Future Work - EPA Office of Research and Development studies, CE-CERT under EPA cooperative
agreement, and CRC testing project.  Future focus will be on determining fuel effects, and how
ammonia emissions change as NOx emissions are controlled.

-------
                                      Appendix B

                                 Response to Comments

       Appendix B is a  response to specific technical comments submitted by  the Engine
Manufacturer's Association (EMA) on this current document.  For convenience, the comments and
the corresponding responses (shown in italics) are provided here as an appendix.
1.      EMA questions the accuracy of the estimate of PM emissions by component for a 2007 and
       later Class 8b vehicle operating on 15 ppm diesel sulfur fuel. More specifically, the model's
       estimates of elemental carbon emissions of 20.8 mg/mile and organic carbon emissions of
       6.6 mg/mile are erroneously high.  In fact, for filter-equipped diesel engines, which all 2007
       and later engines will be, the elemental carbon and organic carbon portions of the total PM
       essentially are zero. The only PM exhaust emissions will be from sulfates, which are a result
       of the sulfur in the diesel fuel. Indeed, EPA's own analysis with respect to 2007 and later
       PM standards bears this out. EPA estimated that total PM emissions for the 2007 and later
       model years would be 0.005 g/hp-hr,  based on EPA's assumptions of 7 ppm sulfur diesel fuel
       and 70% conversion of the sulfur mass to sulfate. In other words, all the exhaust PM is
       sulfate and, hence, none is elemental  carbon or organic carbon. The model should be revised
       to reflect the absence of carbon emissions from 2007 and later engines.

Response to #1

       While the prospect of essentially zero emissions from diesel trucks is extremely encouraging
      from an environmental perspective, the limited and preliminary emission test data referenced
       in the above comment is undoubtably from engine and vehicle sources which should be
       considered as  engineering prototypes.  As such,  these prototypes have most likely not
       received the necessary field testing in significant  quantities and under a wide variety of
       adverse  conditions such that we can assume that this performance is realistic of fleet-wide
       operation in the future. If such data appear in the next upgrade cycle of this model (MOVES
       emission model), EPA will be pleased to include it. In the meantime, the assumption that
       the 2007 and later model year heavy-duty diesel vehicles will meet and exceed the 2007 rule
       requirements with  an 8 percent compliance  margin over their entire  lifetimes seems
       sufficiently aggressive at this stage.
2.      Diesel fuel sulfur level is a required input to the model to estimate particulate.  However,
       there is no guidance for states and others to use on what sulfur levels should be assumed for
       various years, as there is for gasoline sulfur levels.  EPA should publish guidance in the
       Users Guide on recommended diesel sulfur levels for all calendar years.  Otherwise, users
       may enter the incorrect values, resulting in sulfate PM emissions and SO2 emissions that are
       incorrect.
Response to #2

-------
       EPA recognizes that the burden is now placed on the user to supply the required fuel sulfur
       input.  As such, EPA is working onaMOBILE6.2 Guidance document in which diesel fuel
       sulfur inputs will be discussed.

3.      The diesel fuel sulfur level command does not differentiate between diesel fuel used in 2007
       and later engines and diesel fuel used in pre-2007 engines. EPA's modeling for the 2007 and
       later diesel fuel sulfur rule shows that EPA assumed that for the 2007-2010 phase-in years,
       100% of 2007 and later engines receive low sulfur fuel, but only about 80% of pre-2007
       engines receive low sulfur fuel in the 2007-2009 time period. In order to accurately estimate
       sulfate and SO2  emissions  from the fleet,  EPA should establish two different sulfur
       commands - one for 2007 and later engines, and one for pre-2007 engines - as in the diesel
       fuel sulfur rule.

Response to #3

       EPA regrets to say that the diesel fuel sulfur command is not a function of vehicle model
       year, but only a function of calendar year.  As such, the potential fuel program mentioned
       in the comment cannot be easily modeled with a single MOBILE6.2 run.  Instead, separate
       runs of the model with different fuel sulfur levels will have to made with a weighting of the
       results outside of the model.

4.      The documentation for the 2007  and later PM standards indicates that EPA assumed that
       70% of the sulfur mass for 2007 and later engines is converted to sulfate.  The MOBILE6.1
       model, however, assumes only a 2% conversion of sulfur mass to sulfate,  which is a
       carryover from the pre-2007 engines. The model should be changed to reflect the  70%
       estimate for 2007 and later engines.

Response to Comment #4

       EPA is currently committed to a new test program that obtains additional data from current
       and potentially future vehicles and engines. If such data suggest a higher conversion of fuel
       sulfur to sulfate rather than to gaseous sulfur dioxide, EPA will update the emission factor
       model in the next cycle (MOVES emission factor model).

5.      The MOBILE6.1 model uses particle size cutoffs to estimate the mass of PM  at or below
       certain particle size cutoffs.  The documentation for these particle size  distributions,
       however, indicates that the distributions may have been based on fractions of total particles,
       rather than mass.  This should be carefully reviewed by EPA. If the distributions are based
       on fraction of particles rather than mass, then these distributions cannot be used to estimate
       mass.  If the  documentation is incorrect, then the documentation should be  changed to
       indicate that these are fractions of PM mass at or below each cutoff. It would also be helpful
       for EPA to reference the studies that were used to develop the particle size  cutoffs, as no
       references were provided in either the MOBILE6.1 documentation or the earlier  PARTS
       Users  Guide.  References also were not found in the EPA documentation for  the organic
       carbon fractions; these should be supplied as well.

-------
Response to Comment #5

       As a result of the EMA comment, EPA identified that the labeling in both this technical
       document and the previous PARTS documentation was in error.  The size distributions are
       in terms of mass and their use in the model is in terms of mass. The References Section of
       this document now contains the primary reference to the EPA paniculate size distribution
       work. EPA wishes to thank EMA for pointing out this documentation error which had
       persisted for several years.

6.      The PM emission rates for heavy-duty engines use conversion factors that are combined over
       various vehicle classes, resulting in the same PM emission rates in g/mi for different class
       trucks.  For example, the PM emission rates for class 8a and 8b trucks are identical, although
       the  conversion factors for these classes are not identical.  EPA should revise the PM
       emissions  for each class to utilize the class-specific conversion factors developed for
       MOBILE6.

Response to Comment #6

       Unfortunately, the  current data on paniculate emission is insufficient  to differentiate
       between these classes of heavy heavy-duty dieselin terms of paniculate emission factors in
       units of grams per brake horsepower hour.  When sufficient data becomes available we will
       be pleased to differentiate the emission performance of all classes of heavy-duty diesels, and
       update  the emission factor model in the next cycle (MOVES emission factor model)
7.      EPA added ammoniatothe emission factors reported by MOBILE, based on work performed
       by EPA around 20 years ago. EMA questions whether it is appropriate to include ammonia
       emission rates for diesel vehicles, as EMA is not aware of any diesel exhaust ammonia
       emissions. Moreover, EPA has included no explanation as to why it is appropriate to include
       ammonia emission rates for diesel vehicles, and has included ammonia emission rates based
       on extremely old data. EMA strongly recommends that EPA eliminate ammonia emission
       rates for diesel vehicles based on the lack of data indicating the need for them in EPA's
       emissions inventory model.

Response to Comment #7

       Unfortunately, ammonia emission data  is relatively  scarce on  diesel vehicles.   This
       necessitated the use of older and hopefully representative data. If new ammonia emission
       data becomes available from an upcoming test program, EPA will use it to update the diesel
       ammonia emission in the next emission factor model cycle (MOVES emission factor model).

-------
                                     Appendix C

                                      References
EPA-AA-AQAB-94-2, "Draft User's Guide to PARTS: A Program for Calculating Particle
Emissions from Motor Vehicles",  February, 1995.

EPA 460/3-85-005, "Size Specific Total Paniculate Emission Factors for Mobile Sources",  1985.
Garbe, Robert, EPA/AA/CTAB/PA/81-20 "Determination of a Range of Concern for Mobile Source
      Emissions of Ammonia", August, 1981.
EPA 2007 Heavy-Duty Diesel Rule-Making Docket. See www.epa. gov/otaq/diesel.htm for more
       information.
EPA Tier 2 Gasoline Vehicle and Fuel Rule-Making Docket. See www.epa.gov/otaq/url-fr.htm and
       search for the Tier2 rulemakings.
EPA420-R-01-033, "Modeling Emission Factors for Compressed Natural Gas Vehicles", April,
       2001.
Fugitive Dust Web Site and Reference for AP-42.  See http://www.epa.gov/ttn/chief/index.html
Diesel  and Gasoline Emission Factor Files;  See Attachments PMDZML.csv, PMDDRl.csv,
       PMDDR2.CSV, PMGZML.csv, PMGDR1 .csv, and PMGDR2.csv.  These files contain the
       basic PM emission factors used in the MOBILE6.2 model.

-------