United States
Environmental Protection
Agency
Office of Transportation EPA420-P-04-016
and Air Quality November 2004
Update of Methane and
Nitrous Oxide Emission
Factors for On-Highway
Vehicles
-------
EPA420-P-04-016
November 2004
Update of Methane and Nitrous Oxide
Emission Factors for On-Highway Vehicles
Assessment and Standards Division
Office of Transportation and Air Quality
U.S. Environmental Protection Agency
Prepared for EPA by
ICF Consulting
EPA Contract No. 68-W-99-054
Work Assignment No. WA4-37
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles Draft Report
CONTENTS
1 INTRODUCTION 1
1.1 Purpose 1
1.2 Previous Emission Factors 1
1.3 Definitions of Emission Control Technologies and Standards 1
2 METHODOLOGY 4
3 RESULTS 6
3.1 Nitrous Oxide Emission Factors 6
3.1.1 N2O Federal Test Procedure Results 6
3.1.2 N2O Running and Start Emissions 7
3.1.3 N2O HR505 Comparisons 8
3.2 Methane Emission Factors 9
3.2.1 CH4 Federal Test Procedure Results 9
3.2.2 CH4 Running and Start Emissions 11
3.2.3 CH4HR505 Comparisons 12
4 EMISSION ESTIMATES FOR OTHER CATERGORIES 15
4.1 Nitrous Oxide Emission Factor Estimates 15
4.1.1 N2O Estimates for Light-Duty Gasoline Vehicles 15
4.1.2 N2O Estimates for Light-Duty Gasoline Trucks 15
4.1.3 N2O Estimates for Heavy-Duty Gasoline Trucks 17
4.1.4 N2O Estimates for Motorcycles 17
4.1.5 N2O Estimates for Heavy-Duty Diesel Vehicles 18
4.1.6 N2O Estimates for Light-Duty Diesel Cars and Trucks 18
4.2 Methane Emission Factor Estimates 18
4.2.1 CH4 Estimates for Light-Duty Gasoline Vehicles 19
4.2.2 CH4 Estimates for Light-Duty Gasoline Trucks 19
4.2.3 CH4 Estimates for Heavy-Duty Gasoline Trucks 19
4.2.4 CH4 Estimates for Motorcycles 20
4.2.5 CH4 Estimates for Heavy-Duty Diesel Vehicles 20
4.2.6 CH4 Estimates for Light-Duty Diesel Cars and Trucks 20
5 RECOMMENDED EMISSION FACTORS FOR ON-HIGHWAY VEHICLES .... 22
APPENDICES 23
A Description of Test Data 23
B Response To Peer Review Comments From Thomas Durbin 24
Contract 68-W-99-054, WA 4-37 i ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles
Draft Report
LIST OF TABLES
Table 1. Emission Factors for N2O and CH4 for Highway Vehicles 2
Table 2. Driving Cycles 4
Table 3. N2O Emission Factors for the Federal Test Procedure 6
Table 4. N2O Emission Factors Comparisons against previous IPCC values 6
Table 5. N2O Running and Start Emission Factors using Equations 2 and 3 7
Table 6. N2O Running and Start Emission Factors using Equations 5 and 6 7
Table 7. HR505 Comparison with Bag 2 Running emissions in grams per mile 8
Table 8. HR505 Comparison with Bag 2 Running emissions in grams per hour 8
Table 9. CH4 Emission Factors for the Federal Test Procedure 10
Table 10. CH4 Emission Factors Comparisons against previous IPCC values 10
Table 11. CH4 Running and Start Emission Factors using Equations 2 and 3 11
Table 12. CH4 Running and Start Emission Factors using Equations 5 and 6 11
Table 13. HR505 Comparison with Bag 2 Running emissions in grams per mile 13
Table 14. HR505 Comparison with Bag 2 Running emissions in grams per hour 13
Table 15. CO2 Emissions in grams per mile for Vehicle Types and Emission Control
Technologies 16
Table 16. Estimated N2O Emission Factors for Light-Duty Gasoline Vehicles 17
Table 17. Estimated N2O Emission Factors for Light-Duty Gasoline Trucks 17
Table 18. Estimated N2O Emission Factors for Heavy-Duty Gasoline Trucks 17
Table 19. Estimated N2O Emission Factors for Motorcycles 18
Table 20. Estimated N2O Emission Factors for Heavy-Duty Diesel Vehicles 18
Table 21. Estimated N2O Emission Factors for Light-Duty Diesel Vehicles and Trucks 18
Table 22. Estimated CH4 Emission Factors for Light-Duty Gasoline Vehicles 19
Table 23. Estimated CH4 Emission Factors for Light-Duty Gasoline Trucks 19
Table 24. Estimated CH4 Emission Factors for Heavy-Duty Gasoline Trucks 20
Table 25. Estimated CH4 Emission Factors for Motorcycles 20
Table 26. Estimated CH4 Emission Factors for Heavy-Duty Diesel Vehicles 20
Table 27. Estimated CH4 Emission Factors for Light-Duty Diesel Vehicles and Trucks 21
Table 28. Recommended Values for N2O and CH4 Emission Factors 22
LIST OF FIGURES
Figure 1. Comparison of Start Emissions using two methods 8
Figure 2. Comparison of Running Emissions in grams per mile 9
Figure 3. Comparison of Running Emissions in grams per hour 9
Figure 4. Comparison of Start Emissions for light-duty vehicles using two methods 12
Figure 5. Comparison of Start Emissions for light-duty vehicles using two methods 12
Figure 6. Comparison of Running Emissions in grams per mile 14
Figure 7. Comparison of Running Emissions in grams per hour 14
Contract 68-W-99-054, WA 4-37
ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles Draft Report
1 INTRODUCTION
1.1 Purpose
The U.S. Environmental Protection Agency Office of Transportation and Air Quality (OTAQ) are
currently developing a new mobile source emissions factor model called MOVES. This new model will
estimate greenhouse gas (GHG) emissions for highway vehicles and will be incorporated into
transportation GHG inventory development. Besides other improvements in the methodology, the model
will use updated emission factors for nitrous oxide (N2O) and methane (CH^. While MOVES is
somewhat behind schedule, data to update N2O and QrU emission factors are available for this year's
inventory. These revised emission factors will be incorporated into the model itself.
1.2 Previous Emission Factors
Emission factors used in the US EPA Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-
20011 are listed in Table 1 and are taken from Annex E of that document. It states "The EPA does not
systematically track emissions of QrU and N2O; therefore, estimates of these gases were developed using
a methodology similar to that outlined in the Revisedl996 IPCC Guidelines2" Many of these values will
be updated with new information detailed in this report. In addition, MOVES specifies separate running
and start emissions, which are combined in the emission factors shown in Table 1 .
1.3 Definitions of Emission Control Technologies and Standards
The N2O and QrU emission factors used depend on the emission standards in place and the corresponding
level of control technology for each vehicle type. The definitions of these control technologies are listed
in Annex E of the EPA Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-20011 and
reproduced here:
Uncontrolled (Unc) — Vehicles manufactured prior to the implementation of pollution control
technologies are designated as uncontrolled. Gasoline light-duty cars and trucks (pre-1973), gasoline
heavy-duty vehicles (pre-1984), diesel vehicles (pre-1983), and motorcycles (pre-1996) are assumed to
not have significant control technologies in place.
Non-catalyst (Neat) — These emission controls were common in gasoline passenger cars and light-duty
gasoline trucks during model years (1973-1974) but phased out thereafter, in heavy-duty gasoline vehicles
beginning in the mid-1980s, and in motorcycles beginning in 1996. This technology reduces hydrocarbon
(HC) and carbon monoxide (CO) emissions through adjustments to ignition timing and air-fuel ratio, air
injection into the exhaust manifold, and exhaust gas recirculation (EGR) valves, which also helps meet
vehicle NOx standards.
Oxidation catalyst (Ocat) — This control technology designation represents the introduction of the
catalytic converter, and was the most common technology in gasoline passenger cars and light-duty
gasoline trucks made from 1975 to 1980 (cars) and 1975 to 1985 (trucks). This technology was also used
1 EPA. "Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2001," Report EPA 430-R-03-004, April 2003.
2 Revised 1996 IPCC Guidelines for National Greenhouse Gas Inventories, Paris: Intergovernmental Panel on
Climate Change, United Nations Environment Programme, Organization for Economic Co-Operation and
Development, International Energy Agency.
Contract 68-W-99-054, WA 4-37 1 ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles
Draft Report
Table 1. Previous Emission Factors for N2O and CH4 for Highway Vehicles
Vehicle Type/Control Technology
Gasoline Passenger Cars (LDGV)
Low Emission Vehicles
EPA Tier 1 a
EPA Tier Oa
Oxidation Catalysts
Non-Catalyst
Uncontrolled
Gasoline Light-Duty Trucks (LDGT)
Low Emission Vehicles
EPA Tier 1 a
EPA Tier Oa
Oxidation Catalysts
Non-Catalyst
Uncontrolled
Gasoline Heavy-Duty Vehicles (HDG
Low Emission Vehicles
EPA Tier 1 a
EPA Tier Oa
Oxidation Catalysts b
Non-Catalyst
Uncontrolled
Diesel Passenger Cars (LDDV)
Advanced
Moderate
Uncontrolled
Diesel Light Duty Trucks (LDDT)
Advanced
Moderate
Uncontrolled
Diesel Heavy Duty Vehicles (HDDV)
Advanced
Moderate
Uncontrolled
Motorcycles (Mot)
Non-catalysts Control
Uncontrolled
N2O
(g/mi)
0.0283
0.0463
0.0816
0.0518
0.0166
0.0166
0.0354
0.0581
0.1022
0.0649
0.0208
0.0208
V)
0.1133
0.1394
0.1746
0.1109
0.0354
0.0354
0.0161
0.0161
0.0161
0.0322
0.0322
0.0322
0.0483
0.0483
0.0483
0.0071
0.0071
CH4
(g/mi)
0.0402
0.0483
0.0644
0.1126
0.1931
0.2173
0.0483
0.0563
0.1126
0.1448
0.2253
0.2173
0.0708
0.0966
0.1207
0.1448
0.2012
0.4345
0.0161
0.0161
0.0161
0.0161
0.0161
0.0161
0.0644
0.0805
0.0966
0.2092
0.4184
Sources: IPCC/UNEP/OECD/IEA (1997), EPA (1998)
a The categories "EPA Tier 0" and "EPA Tier 1" were substituted for the early three-way catalyst and advanced
three-way catalyst categories, respectively, as defined in the Revised 1996 IPCC Guidelines. Detailed descriptions
of emissions control technologies are provided at the end of this annex.
b The methane emission factor was assumed based on the oxidation catalyst value for gasoline light-duty trucks.
in some heavy-duty gasoline vehicles between 1982 and 1997. The two-way catalytic converter oxidizes
HC and CO, significantly reducing emissions over 80 percent beyond non-catalyst-system capacity. One
reason unleaded gasoline was introduced in 1975 was due to the fact that oxidation catalysts cannot
function properly with leaded gasoline.
Contract 68-W-99-054, WA 4-37
ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles Draft Report
EPA Tier 0 (TO) —This emission standard from the Clean Air Act was met through the implementation of
early "three-way" catalysts, therefore this technology was used in gasoline passenger cars and light-duty
gasoline trucks sold beginning in the early 1980s, and remained common until 1994. This more
sophisticated emission control system improves the efficiency of the catalyst by converting CO and HC to
CO2 and H2O, reducing NOx to nitrogen and oxygen, and using an on-board diagnostic computer and
oxygen sensor. In addition, this type of catalyst includes a fuel metering system (carburetor or fuel
injection) with electronic "trim" (also known as a "closed-loop system"). New cars with three-way
catalysts met the Clean Air Act's amended standards (enacted in 1977) of reducing HC to 0.41 g/mile by
1980, CO to 3.4 g/mile by 1981 and NOx to 1.0 g/mile by 1981.
EPA Tier 1 (Tl) — This emission standard created through the 1990 amendments to the Clean Air Act
limited passenger car NOx emissions to 0.4 g/mile, and HC emissions to 0.25 g/mile. These bounds
represent a 60 and 40 percent reduction, respectively, from the EPA Tier 0 standard set in 1981. For
light-duty trucks, this standard set emissions at 0.4 to 1.1 g/mile for NOx and 0.25 to 0.39 g/mile for HCs,
depending upon the weight of the truck. Emission reductions were met through the use of more advanced
emission control systems, and applied to light-duty gasoline vehicles beginning in 1994. This advanced
emission control systems included advanced three-way catalysts, electronically controlled fuel injection
and ignition timing, EGR, and air injection.
Low Emission Vehicles (LEV) — This emission standard requires a much higher emission control level
than the Tier 1 standard. Applied to light-duty gasoline passenger cars and trucks beginning in small
numbers in the mid-1990's, LEV includes multi-port fuel injection with adaptive learning, an advanced
computer diagnostics systems and advanced and close coupled catalysts with secondary air injection.
LEVs as defined here include transitional low-emission vehicles (TLEVs), low emission vehicles, ultra-
low emission vehicles (ULEVs) and super ultra-low emission vehicles (SULEVs). In this analysis, all
categories of LEVs are treated the same due to the fact that there are very limited CIL, or N2O emission
factor data for LEVs to distinguish among the different types of vehicles. Zero emission vehicles (ZEVs)
are incorporated into the alternative fuel and advanced technology vehicle assessments.
Moderate control (Mod) — Improved injection timing technology and combustion system design for
light- and heavy-duty diesel vehicles (generally in place in model years 1983 to 1995) are considered
moderate control technologies. These controls were implemented to meet emission standards for diesel
trucks and buses adopted by the EPA in 1985 to be met in 1991 and 1994.
Advanced control (Adv) — EGR and modern electronic control of the fuel injection system are designated
as advanced control technologies. These technologies provide diesel vehicles with the level of emission
control necessary to comply with standards in place from 1996 through 2003.
Contract 68-W-99-054, WA 4-37 3 ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles
Draft Report
2 METHODOLOGY
Data obtained from EPA3 included testing on many of the vehicle type/control technology categories
listed in Table 1. This data has been used to develop new overall emission factors as well as running and
start emission profiles for MOVES.
Overall emissions which compare directly to those listed in Table 1 were determined using the U.S.
Federal Test Procedure (FTP). The FTP incorporates three driving segments in which the vehicle's
exhaust is captured in separate "bags," one for each driving segment. Each bag is analyzed separately and
then combined to calculate composite emissions. The formula to calculate composite emissions for the
FTP emissions test is given below:4
Composite =
Bag1 * 0.43 + Bag2 + Bag3 * 0.57
FTP distance
Where FTP distance is approximately 7.44 miles
Equation 1
The federal test procedure includes both starts and running emissions. The bag 1 segment starts from a 12
hour soak at approximately 75°F and is driven over a transient driving cycle for 505 seconds with an
average speed of 25.55 mph. It contains cold start emissions and running emissions. The bag 2 segment
has no start and represents running emissions. Its length is 867 seconds with an average speed of 16.02
mph. The third bag segment is a repeat of the bag 1 segment, but after only a 10 minute soak. This
contains both hot start emissions and running emissions. In addition, some of the data included a hot
running 505 second (HR505) driving cycle. This cycle contains only running emissions (no starts) during
the same cycle used for Bag 1 and Bag 3. It can be used to calculate cold and hot start emissions from the
Bag 1 and Bag 3 segments of the FTP. Since the HR505 cycle has an average speed of 25.55 mph and the
Bag 2 driving cycle has an average speed of 16.02 mph, the two cycles could be used to determine speed
factors at low speed. In this report, however, they are used to determine whether grams per mile or grams
per hour are more constant over the low speed range. Details of the cycles are given in Table 2.
Table 2. Driving Cycles
Cycle
FTP3
Bag 1
Bag 2
Bag 3
HR505
Length
Time
(seconds)
1372
505
867
505
505
Distance
(miles)
7.44
3.58
3.86
3.58
3.58
Average
Speed
(mph)
19.53
25.55
16.02
25.55
25.55
Start
Cold/Hot
Cold
No
Hot
No
While the FTP actually lasts 1877 seconds, the bag 1 and bag 3
results are multiplied by 43% and 57% respectively to represent cold
start activity 43% of the time and hot start activity 57% of the time.
The datasets received from EPA represented 13,277 FTP tests on 6,950 vehicles for methane emissions and 95
FTP tests on 64 vehicles for nitrous oxide emissions. It also included 14,636 non-FTP tests on 2,963 vehicles for
methane emissions and 232 non-FTP tests on 74 vehicles for nitrous oxide. The non-FTP tests included a hot
running 505 as well as several other driving cycles not utilized in this report. Methane tests were performed in
various U.S. locations during the period between April 1982 and June 2000. Nitrous oxide tests were performed in
various U.S. locations during the period between January 2000 and June 1998.
' Code of Federal Regulations Title 40: Protection of the Environment, Chapter 1, Part 600, Section 1134-78.
Contract 68-W-99-054, WA 4-37
ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles Draft Report
Since there are many more FTP tests than HR505 tests, FTP bag 2 emissions in grams per mile are being
used in this report to calculate running emissions. To calculate start emissions, the running emissions
were subtracted from the FTP emissions in grams per mile and multiplied by the length in miles of the
FTP (approximately 7.44 miles). This provided average start emissions which combined both cold and
hot start emissions. These are shown in Equations 2 and 3 below. Running emissions were then
compared against the HR505 emission rate for vehicles in which both an FTP and HR505 were run.
Running Emissions (g/mi) = Bag2 Emissions (g/mi) Equation 2
Start Emissions (g/start) = (FTP Emissions - Bag 2 Emissions) x Actual FTP Distance Equation 3
Another approach to calculate running emissions is to calculate them in grams per hour using the average
speed of each cycle. Start emissions can then be calculated from the FTP emissions in grams per hour
and the running emissions in grams per hour as shown in Equations 4 through 6.
FTP Emissions (g/hr) = FTP Emissions (g/mi)
x Actual FTP distance x 3600 second/hr /1372 seconds Equation 4
Running Emissions (g/hr) = Bag2 Emissions (g/mi)
x Bag 2 distance x 3600 sec/hr / 867 seconds Equation 5
Start Emissions (g/start) = (FTP Emissions (g/hr) - Running Emissions (g/hr))
x (1372/3600) hrs Equation 6
Because the distribution of each set of tests varied, an arithmetic mean was used to determine the average
of all tests. In addition to the arithmetic mean (Average), a standard deviation (SD) and a 95%
confidence interval (95% CI) were also calculated for each set of data. Only data taken at the FTP
temperature range (68°F to 86°F) were used in this analysis. Temperature correction factors using
additional data at higher temperatures might be part of a later report.
Contract 68-W-99-054, WA 4-37 5 ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles
Draft Report
3 RESULTS
Emission factor results for nitrous oxide and methane for on-highway vehicles are discussed in this
section. Only some of the vehicle type/emission tier categories produced statistically significant results
within the 95% confidence interval. Those data are discussed here. Those categories that did not have
enough data to produce statistically significant results within the 95% confidence interval are discussed in
Section 4. Recommended emission factors for all categories are given in Section 5.
3.1 Nitrous Oxide Emission Factors
Emission factors for nitrous oxide are presented in this subsection. Emission results for the federal test
procedure are discussed in Section 3.1.1, running and start emissions are discussed in Section 3.1.2, and
comparisons of running emissions from Bag 2 with HR505 emissions are discussed in Section 3.1.3.
3.1.1
Federal Test Procedure Results
Emission factors for N2O for on-highway vehicles are given in Table 3 for the federal test procedure.
FTP emissions include both start and running emissions. Emission factors are displayed in both grams
per mile and grams per hour.
Table 3. N2O Emission Factors for the Federal Test Procedure
Vehicle
Type
LDGV
LDGT
HDDV
Emission
Tier
LEV
T1
TO
LEV
T1
Adv
No of
Test Pts
7
12
12
5
16
6
FTP Emissions (g/mi)
Average
0.012
0.030
0.054
0.009
0.067
0.005
SD
0.009
0.012
0.050
0.007
0.061
0.001
95% Cl
0.007
0.007
0.028
0.006
0.030
0.001
FTP Emissions (g/hr)
Average
0.245
0.582
1.057
0.178
1.321
0.096
SD
0.179
0.243
0.987
0.135
1.193
0.016
95% Cl
0.133
0.138
0.559
0.118
0.584
0.012
FTP Dist
(mi)
7.491
7.472
7.494
7.489
7.466
7.470
The gram per mile emission factors are compared against Table 1IPCC emission factors in Table 4. As
can be seen from this table, the newly calculated emission factors are in most cases lower than previous
values. This is most likely because newer technologies are represented in the dataset versus those used to
derive the IPCC factors. The newer values better represent the current vehicle fleet.
Table 4. N2O Emission Factors Comparisons against previous IPCC values
Vehicle
Type
LDGV
LDGT
HDDV
Emission
Tier
LEV
T1
TO
LEV
T1
Adv
Emission Factors (g/mi)
IPCC
0.028
0.046
0.082
0.035
0.058
0.048
This Study
0.012
0.030
0.054
0.009
0.067
0.005
Contract 68-W-99-054, WA 4-37
ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles
Draft Report
3.1.2
Running and Start Emissions
Running emissions in grams per mile and start emissions in grams per start, calculated using Equations 2
and 3, are provided in Table 5 along with standard deviations (SD) and 95% confidence intervals (95%
CI). Running emissions in grams per hour and start emissions in grams per start, calculated using
Equations 5 and 6, are provided in Table 6.
Table 5. N2O Running and Start Emission Factors using Equations 2 and 3
Vehicle
Type
LDGV
LDGT
HDDV
Emission
Tier
LEV
T1
TO
LEV
T1
Adv
Running Emissions (g/mi)
Average
0.000
0.015
0.042
0.001
0.041
0.005
SD
0.001
0.014
0.044
0.002
0.052
0.001
95% CI
0.001
0.008
0.025
0.002
0.025
0.001
Start Emissions (g/start)
Average
0.090
0.113
0.092
0.059
0.200
-0.002
SD
0.063
0.056
0.107
0.036
0.154
0.003
95% CI
0.046
0.032
0.060
0.032
0.076
0.003
Table 6. N2O Running and Start Emission Factors using Equations 5 and 6
Vehicle
Type
LDGV
LDGT
HDDV
Emission
Tier
LEV
T1
TO
LEV
T1
Adv
Running Emissions (g/hr)
Average
0.007
0.235
0.671
0.019
0.652
0.083
SD
0.018
0.218
0.716
0.035
0.835
0.019
95% CI
0.013
0.123
0.405
0.031
0.409
0.015
Start Emissions (g/start)
Average
0.091
0.132
0.147
0.061
0.255
0.005
SD
0.064
0.051
0.134
0.039
0.186
0.002
95% CI
0.047
0.029
0.076
0.034
0.091
0.001
Comparisons of start emissions using the two methods (g/mi and g/hr) are shown in Figure 1. As can be
seen from this figure, the start emissions calculated using the two methods are statistically similar within
the 95% confidence interval, except for the heavy-duty diesel vehicle. In that case assuming no start
emissions is a good assumption.
Contract 68-W-99-054, WA 4-37
ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles
Draft Report
Figure 1. Comparison of Start Emissions using two methods
0.40
I
0.30
0.20
1
$
V)
0.10
0.00
LEV T1 TO LEV T1 Adv
LDGV LDGT HDDV
3.1.3 N2O HR505 Comparisons
Several of the vehicles tested in the dataset provided by EPA also included hot running 505 test emissions
along with the FTP emissions tests. Since not all vehicles in the dataset also included HR505 tests,
comparisons were made for those vehicles that did have both tests. Since the HR505 test contains no
starts, it is equivalent to a running emission. The HR505 emissions are compared in terms of grams per
mile in Table 7 and grams per hour in Table 8 to the FTP Bag 2 running emissions presented in Section
3.1.2.
Table 7. HR505 Comparison with Bag 2 Running emissions in grams per mile
Vehicle
Type
LDGV
LDGT
Emission
Tier
T1
T1
No of
Test Pts
9
12
FTP Bag 2 Emissions (g/mi)
Average
0.018
0.052
SD
0.014
0.056
95% Cl
0.009
0.032
HR505 Emissions (g/mi)
Average
0.022
0.059
SD
0.016
0.054
95% Cl
0.010
0.031
Table 8. HR505 Comparison with Bag 2 Running emissions in grams per hour
Vehicle
Type
LDGV
LDGT
Emission
Tier
T1
T1
No of
Test Pts
9
12
FTP Bag 2 Emissions (g/hr)
Average
0.295
0.827
SD
0.219
0.898
95% Cl
0.143
0.508
HR505 Emissions (g/hr)
Average
0.574
1.508
SD
0.400
1.397
95% Cl
0.261
0.791
Running emissions are compared on a grams per mile basis in Figure 2 and a grams per hour basis in
Figure 3. As can be seen in those figures the Bag 2 running emissions in either grams per mile or grams
per hour are statistically similar to the HR505 running emissions. In absolute value, however, the bag 2
emission levels in grams per mile are closer to the HR505 emission levels in grams per mile than the
comparison in grams per hour. Thus, it is suggested that emission rates in grams per mile be used for the
low speed case. Conversion to grams per hour should be done at FTP speed, which is somewhere
between the Bag 2 and HR505 speeds.
Contract 68-W-99-054, WA 4-37
ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles
Draft Report
Figure 2. Comparison of Running Emissions in grams per mile
0.10
&
o O.C
D.08
in 0
.04
0)
0.02
0.00
f
LDGV
LDGT
Figure 3. Comparison of Running Emissions in grams per hour
2.50
•c- 2.00
5
in
n
in
1.50
1.00
0.50
0.00
LDGV
LDGT
3.2 Methane Emission Factors
Emission factors for methane are presented in this section. Emission results for the federal test procedure
are discussed in Section 3.2.1, running and start emissions are discussed in Section 3.2.2, and
comparisons of running emissions from Bag 2 against HR505 emissions are discussed in Section 3.2.3.
3.2.1 CH4 Federal Test Procedure Results
Emission factors for CH4 for on-highway vehicles are given in Table 9 for the federal test procedure. The
federal test procedure combines both start and running emissions. Emission factors are displayed in both
grams per mile and grams per hour.
Contract 68-W-99-054, WA 4-37
ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles
Draft Report
Table 9. CH4 Emission Factors for the Federal Test Procedure
Vehicle
Type
LDGV
LDGT
HDGV
HDDV
Emission
Tier
LEV
T1
TO
Ocat
Neat
LEV
T1
TO
Ocat
T1
TO
Ocat
Adv
No of
Test Pts
7
131
9504
690
20
10
80
1666
455
36
101
90
8
FTP Emissions (g/mi)
Average
0.013
0.020
0.066
0.133
0.162
0.017
0.034
0.071
0.143
0.047
0.218
0.209
0.004
SD
0.006
0.010
0.087
0.129
0.130
0.016
0.018
0.067
0.112
0.018
0.115
0.076
0.003
95% Cl
0.005
0.002
0.002
0.010
0.057
0.010
0.004
0.003
0.010
0.006
0.022
0.016
0.002
FTP Emissions (g/hr)
Average
0.254
0.383
1.294
2.609
3.199
0.327
0.672
1.396
2.804
0.904
4.230
4.018
0.081
SD
0.121
0.193
1.711
2.536
2.552
0.306
0.358
1.308
2.185
0.361
2.213
1.453
0.049
95% Cl
0.089
0.033
0.034
0.189
1.118
0.190
0.078
0.063
0.201
0.118
0.432
0.300
0.034
FTP Dist
(mi)
7.434
7.453
7.457
7.475
7.509
7.443
7.441
7.449
7.455
7.401
7.393
7.354
7.471
The gram per mile emission factors are compared against the Table 1 IPCC emission factors in Table 10.
As can be seen from this table, the newly calculated emission factors are in most cases lower than
previous values. This is most likely because newer technologies are represented in the dataset versus
those used to derive the IPCC factors. The newer values better represent the current vehicle fleet. In the
HDGV case, however, the results from the data analysis were higher than those developed by IPCC,
except for the EPA Tier 1 case. Since the IPCC HDGV values were estimated from the light-duty
gasoline vehicle values based upon fuel economy, it is suggested that the newer values be used as these
represent real test data.
Table 10. CH4 Emission Factors Comparisons against previous IPCC values
Vehicle
Type
LDGV
LDGT
HDGV
HDDV
Emission
Tier
LEV
T1
TO
Ocat
Neat
LEV
T1
TO
Ocat
T1
TO
Ocat
Adv
Emission Factors (g/mi)
IPCC
0.040
0.048
0.064
0.113
0.193
0.048
0.056
0.113
0.145
0.097
0.121
0.145
0.064
This Study
0.013
0.020
0.066
0.133
0.162
0.017
0.034
0.071
0.143
0.047
0.218
0.209
0.004
Contract 68-W-99-054, WA 4-37
10
ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles
Draft Report
3.2.2 CELt Running and Start Emissions
Running emissions in grams per mile and start emissions in grams per start, calculated using Equations 2
and 3, are provided in Table 11 along with standard deviations (SD) and 95% confidence intervals (95%
CI). Running emissions in grams per hour and start emissions in grams per start, calculated using
Equations 5 and 6, are provided in Table 12.
Table 11. CH4 Running and Start Emission Factors using Equations 2 and 3
Vehicle
Type
LDGV
LDGT
HDGV
HDDV
Emission
Tier
LEV
T1
TO
Ocat
Neat
LEV
T1
TO
Ocat
T1
TO
Ocat
Adv
Running Emissions (g/mi)
Average
0.009
0.012
0.062
0.132
0.155
0.011
0.023
0.062
0.130
0.024
0.194
0.179
0.006
SD
0.006
0.011
0.102
0.155
0.151
0.017
0.019
0.073
0.125
0.020
0.105
0.066
0.004
95% CI
0.004
0.002
0.002
0.012
0.066
0.011
0.004
0.003
0.012
0.007
0.020
0.014
0.003
Start Emissions (g/start)
Average
0.032
0.055
0.034
0.009
0.059
0.046
0.082
0.072
0.099
0.163
0.183
0.215
-0.011
SD
0.024
0.034
0.192
0.300
0.298
0.015
0.040
0.148
0.250
0.060
0.263
0.178
0.011
95% CI
0.018
0.006
0.004
0.022
0.131
0.009
0.009
0.007
0.023
0.020
0.051
0.037
0.008
Table 12. CH4 Running and Start Emission Factors using Equations 5 and 6
Vehicle
Type
LDGV
LDGT
HDGV
HDDV
Emission
Tier
LEV
T1
TO
Ocat
Neat
LEV
T1
TO
Ocat
T1
TO
Ocat
Adv
Running Emissions (g/hr)
Average
0.139
0.196
0.989
2.125
2.500
0.169
0.374
0.989
2.090
0.390
3.074
2.831
0.090
SD
0.096
0.173
1.648
2.500
2.443
0.271
0.303
1.164
2.015
0.320
1.652
1.026
0.062
95% CI
0.018
0.030
0.033
0.187
1.071
0.168
0.066
0.056
0.185
0.105
0.322
0.212
0.043
Start Emissions (g/start)
Average
0.044
0.071
0.116
0.184
0.266
0.060
0.114
0.155
0.272
0.196
0.441
0.452
-0.003
SD
0.024
0.029
0.143
0.207
0.231
0.017
0.042
0.140
0.216
0.056
0.313
0.218
0.006
95% CI
0.008
0.005
0.003
0.015
0.101
0.011
0.009
0.007
0.020
0.018
0.061
0.045
0.004
Contract 68-W-99-054, WA 4-37
11
ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles
Draft Report
Comparisons of start emissions using the two methods (g/mi and g/hr) are shown in Figure 4 for light-
duty vehicles and Figure 5 for heavy-duty vehicles. As can be seen from these figures, the start emissions
calculated using the two methods do not result in statistically the same value within the 95% confidence
interval, except for a few cases. As described in Section 3.2.3 below, the grams per mile method
produced running emission values that are closer to the HR505 values, so it is recommended to use start
emissions based upon the grams per mile method.
Figure 4. Comparison of Start Emissions for light-duty vehicles using two methods
0.30
-. 0.25
t
I
"01
0.20
1
UJ 0.10
3
w o.os
0.00
r
i
T\
a
I
LEV T1 TO Ocat Neat LEV T1 TO Ocat
LDGV LDGT
Figure 5. Comparison of Start Emissions for heavy-duty vehicles using two methods
0.50
-. 0.40
1
5
c
I
E
CO
0.30
0.20
0.10
0.00
-0.10
T1
TO
HDGV
Ocat
Adv
HDDV
3.2.3 CH4 HR505 Comparisons
Several of the vehicles tested in the dataset provided by EPA also included hot running 505 test emissions
along with the FTP emissions tests. Since not all vehicles in the dataset also included HR505 tests,
Contract 68-W-99-054, WA 4-37
12
ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles
Draft Report
comparisons were made for those vehicles that did have both tests. Since the HR505 test contains no
starts, it is equivalent to a running emission. The HR505 emissions are compared in terms of grams per
mile in Table 13 and grams per hour in Table 14 to the FTP Bag 2 running emissions presented in Section
3.2.2.
Table 13. HR505 Comparison with Bag 2 Running emissions in grams per mile
Vehicle
Type
LDGV
LDGT
HDGV
Emission
Tier
T1
TO
Ocat
LEV
T1
TO
Ocat
T1
TO
Ocat
No of
Test Pts
31
51
7
2
43
33
6
22
76
74
FTP Bag 2 Emissions (g/mi)
Average
0.015
0.085
0.097
0.040
0.023
0.130
0.430
0.042
0.196
0.189
SD
0.013
0.119
0.105
0.019
0.018
0.109
0.426
0.046
0.100
0.067
95% Cl
0.005
0.033
0.077
0.026
0.005
0.037
0.341
0.019
0.022
0.015
HR505 Emissions (g/mi)
Average
0.014
0.062
0.069
0.033
0.022
0.096
0.302
0.050
0.156
0.175
SD
0.009
0.074
0.050
0.022
0.015
0.076
0.267
0.049
0.078
0.085
95% Cl
0.003
0.020
0.037
0.030
0.005
0.026
0.214
0.021
0.017
0.019
Table 14. HR505 Comparison with Bag 2 Running emissions in grams per hour
Vehicle
Type
LDGV
LDGT
HDGV
Emission
Tier
T1
TO
Ocat
LEV
T1
TO
Ocat
T1
TO
Ocat
No of
Test Pts
31
51
7
2
43
33
6
22
76
74
FTP Bag 2 Emissions (g/hr)
Average
0.248
1.366
1.556
0.630
0.376
2.070
6.886
0.643
3.108
2.984
SD
0.212
1.918
1.671
0.304
0.295
1.738
6.855
0.758
1.565
1.044
95% Cl
0.075
0.526
1.238
0.421
0.088
0.593
5.485
0.324
0.352
0.238
HR505 Emissions (g/hr)
Average
0.354
1.598
1.758
0.829
0.572
2.450
7.708
1.266
3.965
4.407
SD
0.221
1.891
1.278
0.558
0.393
1.934
6.862
1.247
1.959
2.115
95% Cl
0.078
0.519
0.947
0.773
0.117
0.660
5.490
0.521
0.440
0.482
Running emissions are compared on a grams per mile basis in Figure 6 and a grams per hour basis in
Figure 7. As can be seen in those figures the Bag 2 running emissions in either grams per mile or grams
per hour are statistically similar to the HR505 running emissions within the 95% confidence interval. In
absolute value, however, the bag 2 emission levels in grams per mile are closer to the HR505 emission
levels in grams per mile than the comparison in grams per hour. Thus, it is suggested that emission rates
in grams per mile be used for the low speed case. Conversion to grams per hour should be done at FTP
speed, which is somewhere between the Bag 2 and HR505 speeds.
Contract 68-W-99-054, WA 4-37
13
ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles
Draft Report
Figure 6. Comparison of Running Emissions in grams per mile
0.90
0.80 h - n Bag 2
E 0.70 DHR505
Bi
^ 0.60
c
•§ 0.50
i* 0.30
'E
§ 0.20
0.10 . ._.
0 00 r^r^ i i i i i i—i i i i-H-Li rr-n i i i i i i Prn
T1 TO Ocat LEV T1 TO Ocat T1 TO Ocat
LDGV LDGT I-DGV
Figure 7. Comparison of Running Emissions in grams per hour
14
DBag2
12 J--DHR505
2 10
(A
O
'55
LU 6
O)
= 4
th
T1 TO Ocat LEV T1 TO Ocat T1 TO Ocat
LDGV LDGT HDGV
Contract 68-W-99-054, WA 4-37
14
ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles Draft Report
4 EMISSION ESTIMATES FOR OTHER CATERGORIES
Several of the vehicle types/control technology categories either contained no data or statistically
insignificant data within the 95 percent confidence interval. For these data, estimation of emission factors
is needed. This section discusses how the IPCC data was derived and makes recommendations for newer
estimates.
4.1 Nitrous Oxide Emission Factor Estimates
Nitrous oxide emissions data was either non-existent or produced statistically insignificant results in the
case of most of the older control technologies for gasoline light-duty vehicles. In addition, there were
either no or statistically insignificant data for heavy-duty gasoline vehicles and light duty diesel vehicles
and trucks. For heavy-duty diesel vehicles, only the most advanced control technology had any data and
that was only for sanitation trucks. There were no motorcycle data.
The data shown in Table 4 indicates that the new data produces statistically significant values that are
lower than the previous IPCC values. Most of the values used currently in the Inventory of U.S.
Greenhouse Gas Emissions and Sinks: 1990-2001 come from an EPA report by Harvey Michaels.5 In
this report, Michaels examined limited data for gasoline passenger cars and developed new N2O emission
factors based upon that data. For other vehicle types, he used the ratio of fuel economies to produce new
values. For diesel vehicles, he suggested using the European values listed in the IPCC guidelines.
A similar method is suggested here. To estimate the emission factors of N2O for other vehicle types, but
the same emission tier, it is suggested that the ratio of CO2 emissions be used. CO2 emissions per mile
were extracted from the Harvey Michaels report and converted to grams per mile. These values are
shown on Table 15. Estimates of CO2 emissions for LEV and Tier 1 heavy-duty gasoline vehicles were
estimated from the CO2 emissions for Tier 0 heavy-duty gasoline vehicles and the ratio of CO2 emissions
between a Tier 1 and LEV light-duty truck and a Tier 0 light-duty truck.
4.1.1 NiO Estimates for Light-Duty Gasoline Vehicles
The datasets provided by EPA produced statistically significant emission factors for LEV, Tier 1 and Tier
0 vehicles. There was no data for any of the earlier technologies such as oxidation catalyst, non-catalyst,
and uncontrolled. To estimate emissions for the oxidation catalyst category, the Tier 0 emission levels
were multiplied by the ratio of CO2 emissions for the Tier 0 light-duty gasoline vehicle divided by the
oxidation catalyst CO2 emissions for the same vehicle type. This ratio was applied to the FTP, running
and start emissions. For the non-catalyst and uncontrolled, previous IPCC values were used for the FTP
values, and the ratio of FTP to running and start emissions for the oxidation catalyst category was used to
determine the running and start emissions for the non-catalyst and uncontrolled levels. Estimated values
for these three control technologies are shown in Table 16.
4.1.2 N2O Estimates for Light-Duty Gasoline Trucks
The datasets provided by EPA produced statistically significant emission factors for LEV and Tier 1
vehicles only. Estimated emission factors for the other control technologies for light-duty trucks were
estimated from the emission factors for light-duty gasoline cars based upon the ratio of the light-duty
truck CO2 emission rate versus the light-duty gasoline car CO2 emission rate. Estimated values for the
5 EPA, "Emissions of Nitrous Oxide from Highway Mobile Sources: Comments on the Draft Inventory of U. S.
Greenhouse Gas Emissions and Sinks, 1990-1996 (March 1998),"Report No. EPA420-R-98-009, August 1998.
Contract 68-W-99-054, WA 4-37 15 ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles
Draft Report
Table 15. CO2 Emissions in grams per mile for Vehicle Types and Emission Control Technologies
Vehicle/Control Technology
CO2 (g/mi)
Gasoline Passenger Cars
Low Emission Vehicles
Tierl
TierO
Oxidation Catalyst
Non-Catalyst
Uncontrolled
451
459
480
616
855
814
Gasoline Light-Duty Trucks
Low Emission Vehicles
Tierl
TierO
Oxidation Catalyst
Non-Catalyst
Uncontrolled
637
637
801
801
967
932
Gasoline Heavy-Duty Vehicles
Low Emission Vehicles
Tierl
TierO
Oxidation Catalyst
Non-Catalyst Control
Uncontrolled
1,301
1,301
1,637
1,667
2,124
2,124
Diesel Passenger Cars
Advanced
Moderate
Uncontrolled
381
399
513
Diesel Light Trucks
Advanced
Moderate
Uncontrolled
531
533
668
Diesel Heavy-Duty Vehicles
Advanced
Moderate
Uncontrolled
1,588
1,627
1,765
Motorcycles
Non-Catalyst Control
Uncontrolled
352
428
various emission control categories where there were not statistically significant data are shown in Table
17.
Contract 68-W-99-054, WA 4-37
16
ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles
Draft Report
Table 16. Estimated N2O Emission Factors for Light-Duty Gasoline Vehicles
Emission Control
Technology
Oxidation Catalyst
Non-Catalyst
Uncontrolled
FTP
(g/mi)
0.042
0.017
0.017
Running
(g/mi)
0.032
0.013
0.013
Start
(g/start)
0.072
0.028
0.028
Table 17. Estimated N2O Emission Factors for Light-Duty Gasoline Trucks
Emission Control
Technology
TierO
Oxidation Catalyst
Non-Catalyst
Uncontrolled
FTP
(g/mi)
0.090
0.054
0.019
0.019
Running
(g/mi)
0.069
0.042
0.015
0.015
Start
(g/start)
0.153
0.093
0.032
0.032
4.1.3 NiO Estimates for Heavy-Duty Gasoline Trucks
The datasets provided by EPA did not produce any statistically significant emission factors for heavy-
duty trucks [because N2O emissions were measured for only a single heavy-duty gasoline truck].
Emission factors for heavy-duty gasoline trucks were estimated from light-duty gasoline trucks based
upon the ratio of CO2 emissions for each control technology. Estimated values for the various emission
control categories where there were not statistically significant data are shown in Table 18.
Table 18. Estimated N2O Emission Factors for Heavy-Duty Gasoline Trucks
Emission Control
Technology
LEV
Tierl
TierO
Oxidation Catalyst
Non-Catalyst
Uncontrolled
FTP
(g/mi)
0.019
0.138
0.183
0.113
0.041
0.043
Running
(g/mi)
0.002
0.083
0.142
0.088
0.032
0.033
Start
(g/start)
0.120
0.409
0.313
0.194
0.070
0.074
4.1.4 NiO Estimates for Motorcycles
The datasets provided by EPA did not contain any test data for motorcycles. Emission factors for
motorcycles were estimated from light-duty gasoline cars based upon the ratio of CO2 emissions for the
two control technologies. Estimated values for motorcycles are shown in Table 19.
Contract 68-W-99-054, WA 4-37
17
ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles
Draft Report
Table 19. Estimated N2O Emission Factors for Motorcycles
Emission Control
Technology
Non-Catalyst
Uncontrolled
FTP
(g/mi)
0.007
0.009
Running
(g/mi)
0.005
0.007
Start
(g/start)
0.012
0.015
4.1.5 NiO Estimates for Heavy-Duty Diesel Vehicles
The datasets provided by EPA only produced statistically significant emission factors for the advanced
control technology. While the data only represented the light heavy-duty trucks (i.e., with GVWR
between 8,500 and 10,000 pounds), the same engine technology would apply to other heavy-duty truck
types. Also since the advanced technology does not include any aftertreatment devices, it is assumed that
the N2O emissions from the other categories are the same as those for the advanced technology. Estimated
values for the various emission control categories where there were not statistically significant data are
shown in Table 20.
Table 20. Estimated N2O Emission Factors for Heavy-Duty Diesel Vehicles
Emission Control
Technology
Moderate
Uncontrolled
FTP
(g/mi)
0.005
0.005
Running
(g/mi)
0.005
0.005
Start
(g/start)
-0.002
-0.002
4.1.6 NiO Estimates for Light-Duty Diesel Cars and Trucks
The datasets provided by EPA did not produce any statistically significant emission factors for either
light-duty diesel cars or trucks. Emission factors for light-duty cars and trucks were estimated from
heavy-duty diesel vehicles based upon the ratio of CO2 emissions for the various control technologies.
Estimated values for the light-duty diesel cars and trucks are shown in Table 21.
Table 21. Estimated N2O Emission Factors for Light-Duty Diesel Vehicles and Trucks
Vehicle
Type
Light-Duty
Diesel
Vehicles
Light-Duty
Diesel Trucks
Emission Control
Technology
Advanced
Moderate
Uncontrolled
Advanced
Moderate
Uncontrolled
FTP
(g/mi)
0.001
0.001
0.001
0.002
0.002
0.002
Running
(g/mi)
0.001
0.001
0.002
0.002
0.002
0.002
Start
(g/start)
0.000
0.000
-0.001
-0.001
-0.001
-0.001
4.2 Methane Emission Factor Estimates
While significantly more data were available for the methane emission factor analysis, there were still
several categories of vehicle/control technology combinations for which there was either no or not enough
data to produce statistically significant results within the 95% confidence interval. This included most
diesel vehicles, motorcycles and earlier technologies of gasoline vehicles (non-catalyst and uncontrolled).
Contract 68-W-99-054, WA 4-37
18
ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles
Draft Report
The data in Table 10 indicate that the IPCC LEV and Tl emission factors for light duty gasoline vehicles
and trucks were higher than the new data suggests. This is most likely because the newer data represents
advancements in emission control above that used to determine the IPCC values. The IPCC values were
determined from EPA's earlier mobile source emission factor model MOBILES.
In order to estimate methane emissions for other vehicles, the ratio of CO2 emissions were used to
extrapolate values similar to that for N2O emission factors.
4.2.1 CH4 Estimates for Light-Duty Gasoline Vehicles
Statistically significant data were available for all but the uncontrolled light-duty gasoline vehicle
category. This was estimated from the non-catalyst emission factors based upon the ratio of CO2
emission rates. Estimated methane emission factors for uncontrolled light-duty gasoline cars are given in
Table 22.
Table 22. Estimated CH4 Emission Factors for Light-Duty Gasoline Vehicles
Emission Control
Technology
Uncontrolled
FTP
(g/mi)
0.171
Running
(g/mi)
0.162
Start
(g/start)
0.062
4.2.2 CH4 Estimates for Light-Duty Gasoline Trucks
The datasets provided by EPA produced statistically significant emission factors for all but the non-
catalyst and uncontrolled categories. Estimated emission factors for these control technologies were
estimated from the emission factors for light-duty gasoline cars based upon the ratio of the light-duty
truck CO2 emission rate versus the light-duty gasoline car CO2 emission rate for the given control
technology. Estimated values for the various emission control categories where there were not
statistically significant data are shown in Table 23.
Table 23. Estimated CH4 Emission Factors for Light-Duty Gasoline Trucks
Emission Control
Technology
Non-Catalyst
Uncontrolled
FTP
(g/mi)
0.184
0.195
Running
(g/mi)
0.175
0.186
Start
(g/start)
0.067
0.071
4.2.3 CH4 Estimates for Heavy-Duty Gasoline Trucks
The datasets provided by EPA produced statistically significant emission factors for Tier 1, Tier 0, and
oxidation catalyst heavy-duty gasoline trucks. Emission factors for the other categories were estimated
from light-duty gasoline trucks based upon the ratio of CO2 emissions for each control technology.
Estimated values for the various emission control categories where there were not statistically significant
data are shown in Table 24.
Contract 68-W-99-054, WA 4-37
19
ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles
Draft Report
Table 24. Estimated CH4 Emission Factors for Heavy-Duty Gasoline Trucks
Emission Control
Technology
LEV
Non-Catalyst
Uncontrolled
FTP
(g/mi)
0.034
0.403
0.445
Running
(g/mi)
0.022
0.384
0.423
Start
(g/start)
0.094
0.147
0.162
4.2.4 CH4 Estimates for Motorcycles
The datasets provided by EPA did not contain any test data for motorcycles. Emission factors for
motorcycles were estimated from light-duty gasoline cars based upon the ratio of CO2 emissions for the
two control technologies. Estimated values for motorcycles are shown in Table 25.
Table 25. Estimated CH4 Emission Factors for Motorcycles
Emission Control
Technology
Non-Catalyst
Uncontrolled
FTP
(g/mi)
0.067
0.090
Running
(g/mi)
0.064
0.085
Start
(g/start)
0.024
0.033
4.2.5 CH4 Estimates for Heavy-Duty Diesel Vehicles
The datasets provided by EPA only produced statistically significant emission factors for the advanced
control technology. Since the advanced technology does not include any aftertreatment devices, it is
assumed that the QrU emissions from the other categories are the same as those for the advanced
technology. Estimated values for the various emission control categories where there were not statistically
significant data are shown in Table 26.
Table 26. Estimated CH4 Emission Factors for Heavy-Duty Diesel Vehicles
Emission Control
Technology
Moderate
Uncontrolled
FTP
(g/mi)
0.004
0.004
Running
(g/mi)
0.006
0.006
Start
(g/start)
-0.011
-0.011
4.2.6 CELt Estimates for Light-Duty Diesel Cars and Trucks
The datasets provided by EPA did not produce any statistically significant emission factors for either
light-duty diesel cars or trucks. Emission factors for light-duty cars and trucks were estimated from
heavy-duty diesel vehicles based upon the ratio of CO2 emissions for the various control technologies.
Estimated values for the various emission control categories for light-duty diesel cars and trucks are
shown in Table 27.
Contract 68-W-99-054, WA 4-37
20
ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles
Draft Report
Table 27. Estimated CH4 Emission Factors for Light-Duty Diesel Vehicles and Trucks
Vehicle
Type
Light-Duty
Diesel
Vehicles
Light-Duty
Diesel Trucks
Emission Control
Technology
Advanced
Moderate
Uncontrolled
Advanced
Moderate
Uncontrolled
FTP
(g/mi)
0.001
0.001
0.001
0.001
0.001
0.002
Running
(g/mi)
0.001
0.001
0.002
0.002
0.002
0.002
Start
(g/start)
-0.003
-0.003
-0.003
-0.004
-0.004
-0.004
Contract 68-W-99-054, WA 4-37
21
ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles
Draft Report
5 RECOMMENDED EMISSION FACTORS FOR ON-HIGHWAY
VEHICLES
Table 28 presents previous IPCC values used in the 2001 US Inventory and newly recommended
emission factors for methane and nitrous oxide emissions for all categories. Values in bold represent
those derived using statistically significant data within the 95% confidence interval.
Table 28. Recommended Values for N2O and CH4 Emission Factors
Vehicle Type
Control Technology
Gasoline Passenger Cars
Low Emission Vehicles
TieM
TierO
Oxidation Catalyst
Non-Catalyst
Uncontrolled
Nitrous Oxide
IPCC
g/mi
0.028
0.046
0.082
0.052
0.017
0.017
FTP
g/mi
0.012
0.030
0.054
0.042
0.017
0.017
Run
g/mi
0.000
0.015
0.042
0.032
0.013
0.013
Start
g/start
0.090
0.113
0.092
0.072
0.028
0.028
Methane
IPCC
g/mi
0.040
0.048
0.064
0.113
0.193
0.217
FTP
g/mi
0.013
0.020
0.066
0.133
0.162
0.171
Run
g/mi
0.009
0.012
0.062
0.132
0.155
0.162
Start
g/start
0.032
0.055
0.034
0.009
0.059
0.062
Gasoline Light-Duty Trucks
Low Emission Vehicles
TieM
TierO
Oxidation Catalyst
Non-Catalyst
Uncontrolled
0.035
0.058
0.102
0.065
0.021
0.021
0.009
0.067
0.090
0.054
0.019
0.019
0.001
0.041
0.069
0.042
0.015
0.015
0.059
0.200
0.153
0.093
0.032
0.032
0.048
0.056
0.113
0.145
0.225
0.217
0.017
0.034
0.071
0.143
0.184
0.195
0.011
0.023
0.062
0.130
0.175
0.186
0.046
0.082
0.072
0.099
0.067
0.071
Gasoline Heavy-Duty Vehicles
Low Emission Vehicles
TieM
TierO
Oxidation Catalyst
Non-Catalyst Control
Uncontrolled
Diesel Passenger Cars
Advanced
Moderate
Uncontrolled
Diesel Light Trucks
Advanced
Moderate
Uncontrolled
0.113
0.139
0.175
0.111
0.035
0.035
0.016
0.016
0.016
0.032
0.032
0.032
0.019
0.138
0.183
0.113
0.041
0.043
0.001
0.001
0.001
0.002
0.002
0.002
0.002
0.083
0.142
0.088
0.032
0.033
0.001
0.001
0.002
0.002
0.002
0.002
0.120
0.409
0.313
0.194
0.070
0.074
0.000
0.000
-0.001
-0.001
-0.001
-0.001
0.071
0.097
0.121
0.145
0.201
0.435
0.016
0.016
0.016
0.016
0.016
0.016
0.034
0.047
0.218
0.208
0.403
0.445
0.001
0.001
0.001
0.001
0.001
0.002
0.022
0.024
0.194
0.179
0.384
0.423
0.001
0.001
0.002
0.002
0.002
0.002
0.094
0.163
0.183
0.215
0.147
0.162
-0.003
-0.003
-0.003
-0.004
-0.004
-0.004
Diesel Heavy-Duty Vehicles
Advanced
Moderate
Uncontrolled
Motorcycles
Non-Catalyst Control
Uncontrolled
0.048
0.048
0.048
0.007
0.007
0.005
0.005
0.005
0.007
0.009
0.005
0.005
0.005
0.005
0.007
-0.002
-0.002
-0.002
0.012
0.015
0.064
0.081
0.097
0.209
0.418
0.004
0.004
0.004
0.067
0.090
0.006
0.006
0.006
0.064
0.085
-0.011
-0.011
-0.011
0.024
0.033
Contract 68-W-99-054, WA 4-37
22
ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles Draft Report
Appendix A
Description of Test Data
EPA provided its contractor (ICF) with datasets with test results containing methane measurements:
13,277 FTP tests on 6,950 vehicles and
14,636 non-FTP tests on 2,963 vehicles,
and with datasets with test results containing nitrous oxide measurements:
95 FTP tests on 64 vehicles and
232 non-FTP tests on 74 vehicles.
The FTP tests that measured nitrous oxide emissions were primarily those performed by EPA,
supplemented by tests performed by the University of California at Riverside CE-CERT, Southwest
Research Institute (SwRI), and CARB.
The non-FTP tests included a hot running 505 which were used in this study to validate the approach used
to separate the start and running emissions as well as several other driving cycles not utilized in this
report. Methane tests were performed in various U.S. locations during the period between April 1982 and
June 2000. Nitrous oxide tests were performed in various U.S. locations during the period between June
1998 and May 2002.
The analyses performed by ICF were limited to the FTP tests that were performed within the temperature
range of 68 degrees to 86 degrees Fahrenheit (i.e., at a nominal temperature of 75° F).
Since the goal of ICF's analyses was to develop separate emission rates for both the running operation and
engine starts, the analyses focused on the FTP tests since they contained both of those two types of
vehicle operation.
Contract 68-W-99-054, WA 4-37 23 ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles Draft Report
Appendix B
Response to Peer Review Comments from Thomas Durbin
This report was formally peer reviewed by Thomas Durbin, Ph.D., Associate Research
Engineer with the College of Engineering-Center for Environmental Research and
Technology (CE-CERT) at the University of California-Riverside. In this appendix,
comments from Thomas Durbin are reproduced in plain text, and EPA's responses to those
comments are interspersed in indented italics.
************************************
October 10, 2004
The following is a review of the IFC Consulting document "Update of Methane and Nitrous Oxide
Emission Factors for On-Highway Vehicles" and the underlying datasets used in developing this
document. This document is being used by the United States Environmental Protection Agency (US EPA)
in the development of emissions factors for the EPA MOVES model. This review covers several relevant
areas including the dataset completeness, methodology, and report clarity. The suggestions given in this
review are to provide EPA guidance in moving forward and improving the emission factors for methane
and nitrous oxide (N2O) from vehicles.
Overall, the report appears to be satisfactory in characterizing CFI4 and N2O emission factors based on the
information provided in the EPA database. The primary concern with the updated emission factors is that
there are still some gaps in the EPA database and there is also a need to develop emission factors for some
categories by extrapolating data from more broadly tested categories using comparisons of CO2
emissions. In reviewing the EPA datasets for CFLj and N2O, it was found that a number of studies with
CFLt and/or N2O emissions measurements have not been included. It is suggested that the next step in
improving the EPA emission factors for CFLt and N2O is to augment the current database with additional
information from the literature, especially in under populated categories such as diesel vehicles. In the
larger context of greenhouse gases (GHG), the contribution of CH4 and N2O emissions is still less than
5% of the total GHG contribution from mobiles sources; therefore, improvements in CO2 estimates from
vehicles should probably remain a higher priority.
RESPONSE: The analyses for the MOVES2006 version will make use of all the available data.
For the report itself, some description of how the emission factors will be implemented in the MOVES
model would be useful. To provide additional detail to the report, it would be useful to include number of
test points available in each vehicle category and a brief discussion of the datasets. A discussion of the
criteria used in judging the statistical significance of the available data in particular categories could also
be added. Finally, it is suggested that as the emissions factors are improved through the years that the
potential effects of other parameters on CFLt and N2O emissions be considered. These could include fuel
sulfur level, different driving cycles, vehicle mileage/age, and ambient temperature, with fuel sulfur level
being one of the most important of these parameters.
RESPONSE: The implementation of these emission factors is discussed in more detail in the
report entitled "MOVES2004 Energy and Emissions Inputs.
Contract 68-W-99-054, WA 4-37 24 ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles Draft Report
A more detailed commentary on different aspects of the report/underlying datasets is provided below. To
address the EPA's primary areas of interest, this review is broken down into four main topics:
completeness of data sources, overall methodology and analysis, additional factors to be considered in
developing emission factors, and presentation and report clarity.
Completeness of datasets selected. The robustness of the underlying datasets in developing emission
factors is essential in the overall accuracy of emissions inventories. The datasets used appear to be
obtained from a larger database of emissions testing results maintained by EPA. While this database is
sufficient CIL, emissions for some of the larger categories, some additional data sources need to be
considered in other N2O and CF^ categories. This is one of the most significant weaknesses of the
methodology. Emissions results for CIL, and N2O from vehicles are reviewed by Lipman and Delucchi
(2002) and in the discussion below.
The available emissions data for CH4 emissions appears to be large enough for some of the more
important categories (i.e., Tl, TO, Ocat). The CFL, data for LEV vehicles is relatively limited, however (7
LDGV and 10 LDGT). Is it possible that information on CIL, emissions for late model vehicles can be
obtained from certification data using the difference between THC and NMHC emissions? Other studies
of fuel properties for LEV certified vehicles may also provide information on THC and NMHC emissions
for late model vehicles (AAM/AIAM, 2001; Durbin et al., 2003), again using differences in THC and
NMHC emissions to get CFL,. In the motorcycle category, The California Air Resources Board (CARB)
has done testing on a series of 100 1966-1999 motorcycles (Jones, 2000). This report does not include
CH4 directly, but the CH^ contribution to THC for motorcycles could be estimated from data of other
sources to provide a better emission factor CFU for motorcycles.
For N2O emissions, the database appears to be limited to tests conducted directly by EPA, some of the
earlier work from the University of California at Riverside CE-CERT, and smaller number of tests from
the Southwest Research Institute (SwRI) and CARB. Several more recent studies should be considered
for inclusion, including those by Durbin et al. (2003) and Huai et al. (2002, 2003, 2004) that include
approximately 20 LEV LDGVs and 10 LEV LDGTs. The limitations of the LEV N2O emissions
estimates are evident in comparing the emission factors for LDGV and LDGT. Specifically, on the basis
of 5 test points, the emission factor for the LDGT is found to be less than that of the LDGVs, contrary to
what is found with a more complete review of the literature (Huai et al., 2003). CARB has also
characterized N2O for a fleet of in-use vehicles (Behrentz et al. 2004). The individual vehicle results are
not presented in this study, but may be available through CARB. Environment Canada has also collected
N2O emissions for a fleet of 21 1978-1996 vehicles (Graham, 1999). Becker et al. (1999) and Baronick et
al. (2000) have conducted studies of 1996 and newer vehicles, although specific vehicle information is
not included in the work by Becker et al. Michaels et al. (1998) also reviewed some earlier data sets that
are not included in the current N2O dataset used for this study, although in some cases the previous IPCC
values based on these data are still used. Huai et al. (2003b) conducted a more recent review including
these data sets as well as some more current information.
RESPONSE: The analyses for the MOVES2006 version will make use of all the available data.
The data on CFL, and N2O for diesel vehicles is limited to a small number of tests on medium -duty diesel
trucks conducted at SwRI. There are some additional sources of data that should be considered for diesel
vehicles, although even a more comprehensive literature review yields only a limited number of diesel
test records for these specific emissions. Merritt (2003) of SwRI conducted a comprehensive literature
review of diesel emissions data as part of Coordinating Research Council's (CRC) project No. AVFL-
10A. They identified approximately 10 studies that include either CFL, and/or N2O diesel emissions data
from vehicles or engines. For CH^ emissions, several organizations have made measurements on heavy-
duty diesel vehicles including West Virginia University (Gautam et al., 2003; Gautam et al., 1996), SwRI
Contract 68-W-99-054, WA 4-37 25 ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles Draft Report
(Ullman et al., 2003) and Ecotraffic of Sweden (Ahlvik and Brandberg, 2000). For light- to medium-duty
vehicles, Schaurer et al. (1999) and researchers from Ford (Siegl et al., 1999) and CE-CERT (Durbin et
al., 1999) have all made speciation hydrocarbon measurements, although methane emissions are not
reported in all of these publications. For light-duty diesel vehicles, N2O emissions have been measured by
Ahlvik (2002), Fanick et al. (2001) of SwRI [which may be in the database], and Oyama and Kakegawa
(2000).
RESPONSE: The analyses for the MOVES2006 version will make use of all the available data.
In examining the datasets, it is also useful to consider the stratification of the vehicle technology binning
structure. The current stratification appears to be sufficient at the present time. As newer LEV II vehicles
are introduced into the fleet, these categories should be reflected in the stratification structure. It may also
be worthwhile in conducting some sensitivity studies on the older Tier 0 vehicles, since there was
considerable evolution of vehicle technology over this time period (early 1980s to mid-1990s). While
these vehicles will compose a progressively smaller fraction of the fleet over time, they can still represent
a large fraction of the total emissions inventory. It would be useful to further break the Tier 0 group down
into roughly 5 year periods based on model year (1980-1985, 1986-1990, and 1990 and newer) and
compare the results with the composite emission factor.
RESPONSE: As more data become available, we shall revisit the selection of model year
groupings.
Analysis and Overall Methodology - The methodologies used in this document are reasonable given the
limitations of the datasets provided. As discussed above, there are areas where the EPA database is
incomplete and additional data are available. This additional data should be used instead of extrapolating
data from CO2 for certain categories.
RESPONSE: As more data become available, we shall use those data rather than relying on
extrapolations.
Separating the FTP emissions into start and running emissions. The separation of emissions into
start and running emissions is a good idea since each represents an important segment of the emissions
inventory. The comparisons of hot running 505s (HR505) and the bag 2 emissions indicate that the
subtraction of bag 2 emissions is reasonable for determining start emissions, at least for CFL, and N2O.
Presently, the methodology characterizes start emissions as a combination of cold and hot starts. Some
analysis should be performed to evaluate the differences between cold and hot start emissions for these
pollutants to better understand whether the contribution of these emissions should be considered
separately in the model. A quick review of the N2O dataset by this reviewer indicated that the bag 1
emissions were very similar to those for bag 3 averaged over the entire fleet, but varied considerably from
vehicle to vehicle. On this basis, using combined hot and cold start emissions for N2O is probably
adequate for fleet wide emissions. Similar analysis was not done for CFL, emissions.
RESPONSE: In estimating the HC, CO, and NOx emissions in the MOVES2006 version, we
shall distinguish between hot-start and cold-start emissions. We shall use that
opportunity to revisit our estimates ofCH4 and N2O start emissions to determine
whether to also distinguish between hot-start and cold-start for these emissions
as well.
Contract 68-W-99-054, WA 4-37 26 ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles Draft Report
Extrapolation of emission estimates from untested strata using CO2 emissions. While CO2 and fuel
use can be successfully used in some situations to predict emission rates of various pollutants, this
relationship depends on a variety of factors such as the vehicle and emissions control technology, the
operating conditions, and the emissions standard to which the vehicle is certified to. While relationships
between fuel economy and particular emissions may be found under aggressive or off-cycle conditions,
under FTP conditions emissions would primarily be related to the control technology required to meet the
applicable emissions standards (which are not directly related to fuel economy). Nevertheless, there are
some real differences in emissions control technology and emissions standards between passenger cars
and trucks and vehicles in different weight classifications that have trends directionally consistent with
increasing CO2 emissions. It is uncertain, however, if these differences would best be obtained from
straight comparisons of emissions standards or using CO2 emissions.
RESPONSE: In developing the MOVES2006 version of the model, we shall estimate the HC,
CO, and NOx emissions. We shall use that opportunity to revisit our estimates of
CH4 and N2O start emissions to determine whether to base those extrapolations
on those estimates ofHC, CO, or NOx emissions.
N2O emissions for a specific vehicle are not expected to correlate with CO2 over a range of operating
conditions. In fact, N2O emissions tend to have higher formation rates at intermediate catalyst temperature
ranges [250-450°C] that occur when the catalyst is warming up to its operation temperature. Under higher
speeds or more aggressive driving conditions, where fuel use would be at a maximum, catalyst
temperature would also be at a maximum and N2O emissions would be low. This has been observed in
several studies (Pringent and DeSeote, 1989; Hirano et al., 1992; Odaka et al., 1998; Koike et al., 1999,
Huai et al., 2003).
For heavy-duty gasoline vehicles, only one N2O test record for a Tier 0 truck was found in the database.
In the almost complete absence of emissions test data, it is reasonable to expect that these vehicles would
have higher N2O emissions than lighter vehicles due to differences in emissions controls. As such, it
seems reasonable N2O emissions would increase in some proportionally to CO2 emissions. It is worth
noting that the single Tier 0 test record is lower than the emission factor given in Table 18 [55 vs. 183
mg/mi]. A limited number of tests on fairly old technology vehicles are also provided in Dietzmann et al.
(1981). These data should probably be considered for comparison or possible inclusion.
For the non-catalyst technologies, it is agreed that previous IPCC N2O emission factors should be used
instead of extrapolating from CO2 emissions for catalyst-equipped vehicles, since the formation of N2O is
more directly related to the catalyst than combustion conditions. Overall, the values for non-catalyst
vehicles may be a little high, since the primary mechanism for forming N2O is over the catalyst. In
deriving the previous IPCC values for non-catalyst vehicles, Michaels et al. (1998) used results from three
primary studies (Pringent and De Soete 1989; Dasch, 1992; Urban and Garbe, 1979). Of these studies,
Pringent and De Soete (1989) reported FTP emission rates of approximately 50 mg/mi, considerably
higher than the results observed in the other studies that were below 5 mg/mi (Dasch, 1992; Urban and
Garbe, 1979). Perhaps the Pringent and De Soete (1989) data are outliers. Other studies have reported
N2O emission rates of 15-20 mg/mi for non-catalyst vehicles (Warner-Selph and Harvey, 1990; Robinson,
1991), while Huai et al. (2003) found an emission rate below 10 mg/mi for non-catalyst light-duty truck.
The oxidation catalyst N2O emission factors are extrapolated from the Tier 0 results using CO2 emission
ratios since the data available in the EPA database was not statistically significant. Earlier estimates by
Michaels et al. (1998) in the category, however, indicate that 11 vehicles records were available in the
oxidation catalyst category (mostly LDGVs). It seems like using the direct N2O measurements from
oxidation catalysts for at least the LDGVs would be more appropriate than extrapolating the results from
only 12 Tier 1 test points using CO2.
Contract 68-W-99-054, WA 4-37 27 ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles Draft Report
The diesel section is somewhat confusing when the database is cross-referenced with the applicable text
on N2O emissions estimates. The database itself appears to only have 6 records for light-heavy-duty pick-
up trucks. The text in section 4.1.5 mentions sanitation truck data that did not appear to be present in the
database examined by this reviewer. Since diesel engines are typically not put in pick-up trucks smaller
than those in the database, having a separate category for light-duty diesel trucks may not be needed. For
light-duty diesel vehicles, Oyama and Kakegawa (2000) found emission rates of about 4-8 mg/mi. Ahlvik
(2002) found much higher rates [on the order of 40 mg/mi], but expressed skepticism that these results
were too high and could not rule out a mix-up in samples. The experimental values for light-duty diesel
vehicles are higher than those obtained from extrapolating CO2 measurements. For a complete inventory
standpoint, however, N2O emissions from light-duty diesel passenger cars are expected to make a small
contribution.
RESPONSE: The reviewer is correct about the sanitation trucks. There were none in that
database. We revised the text by replacing "sanitation" truck with "light heavy-
duty" truck.
For CFi4 emissions, there is statistically significant data for all but the oldest light-duty vehicles, light-
duty truck non-catalyst and uncontrolled trucks, and heavy-duty gasoline trucks. These vehicles probably
make relatively small contribution to the inventory so estimates based on CO2 emissions should be
sufficient to provide factors in these categories. As discussed above, CARB has conducted some testing
on motorcycles (Jones, 2000). It is suggested the THC emissions from this study be extrapolated to obtain
the CFLt estimates for motorcycles as opposed to estimates based on CO2 emissions from light-duty
gasoline cars.
For diesel vehicles, CFI4 emissions generally comprise a small portion of the overall THC emissions.
Studies at SwRI on light-heavy-duty and heavy-heavy-duty diesel vehicles have shown CFI4 emissions to
generally be below 10 mg/mi, and near background levels compared with the THC (Fanick et al., 2001;
Ullman et al., 2003). Gautam et al. (2003) reported higher emission rates of ~40 to 140 mg/mi for a small
set of tests on the UDDS in the E-55/59 program, with emissions going up to over 2 grams/mi for the
"creep" portion of the CARB heavy-duty cycle. For light- and medium-duty vehicles, values reported by
Siegl et al. (1999), Oyama and Kakegawa (2000), and Durbin et al. (1999) range from 1-20 mg/mi.
Statistical Significance - It is mentioned throughout section 4 that statistically significant emission factors
could not be obtained for particular emission/technology categories. An examination of the N2O database
by this reviewer, however, indicated that in some of these categories, there was actually either no
available data or only a single data point (i.e., N2O emissions for HDGVs). A better explanation why data
were classified as not statistically significant should be given (i.e., number of data points, variability of
the data, etc.).
RESPONSE: We revised the text to include that the reason for the lack of statistical
significance was that only a single gasoline-fuel heavy-duty truck produced N2O
FTP emissions.
Parameters used to characterize emissions.
The main parameters used in the IFC Consulting document for characterizing the emission factors for
N2O and CH4 include the vehicle technology stratification, running emissions, and start emissions. There
are several additional parameters that could be important to characterize for N2O and CH4 emission
factors in future efforts.
Contract 68-W-99-054, WA 4-37 28 ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles Draft Report
Fuel S Effects
As fuel sulfur levels continue to be reduced nationally, it is important that these effects be evaluated or
included in the estimates. A number of studies have shown that decreasing fuel sulfur level leads to
significant reductions in N2O emissions (Baronick et al. 2000; Huai et al, 2002; Michaels et al. 1998;
Durbin et al. 2003). In reducing fuel sulfur levels from levels near 300 to levels closer to 30, reductions in
N2O emissions of more than 50% are often found. These changes should be considered in characterizing
N2O emissions in going forward. If possible, it would be useful to add the fuel sulfur level as a parameter
in the N2O database. These fuel sulfur effects would likely be stronger and more important when
considering older data sets where higher sulfur fuels were more likely used. For example, measurements
made by Ballantyne et al. (1994) on a fleet of Canadian vehicles were performed with a fuel with a 700
ppm sulfur level.
A number of studies have shown that for gasoline vehicles hydrocarbons can increase with fuel sulfur in
the range of 5-800 ppm (AAM/AIAM, 2001; Korotney et al., 1995; Rutherford et al., 1995; Benson et al.,
1991; Durbin et al., 2003). It is not known, however, if the effects of fuel sulfur on CH4 have specifically
been characterized. It is possible that CH4 emissions may also increase with fuel sulfur. Since CH4
emissions are more difficult to oxidize over the catalyst than other hydrocarbons, catalysts are generally
not as effective in controlling CH4 emissions compared to other hydrocarbons. As such, it is anticipated
that the impact of fuel sulfur on CH4 emissions would be smaller than its impact on THC emissions.
RESPONSE: In estimating the HC, CO, and NOx emissions in the MOVES2006 version, we
shall consider the effects on those emissions of the sulfur content of the fuel. We
shall use that opportunity to revisit our estimates of CH4 and N2O emissions to
determine if they are also sensitive to the sulfur content of the fuel.
Driving cycles
There is no discussion about how emissions factors for different or off-cycle operating conditions would
be implemented. For N2O, emissions are expected to be a stronger function of catalyst temperature as
opposed to operating condition. As such, start emissions that are already included are the most critical
operating condition in characterizing N2O emissions. Measurements of N2O emissions for higher speed or
more aggressive driving have generally shown that N2O emission rates are low under these conditions
(Dasch, 1992; Sasaki and Kameoka, 1992; Huai et al., 2002, 2003, 2004). Since there are not significant
increases in N2O emissions under aggressive operating conditions, the influence of operating conditions
can be a lower priority parameter. The effects of different driving cycles would most likely track the effects
of driving cycles on THC emissions, hence, these effects can probably be estimates based on THC
emissions.
RESPONSE: 777/5 analysis (by ICF) was limited to emissions produced over the standard FTP
/ LA-4 driving cycle. Speed / cycle adjustment factors are being studied for the
MOVES2006 version of the model.
Vehicle Mileage/Catalyst Age
Several studies have shown that vehicles with older catalysts can have higher emission rates of N2O.
These include studies where direct comparisons between older and newer catalysts were made on the
same vehicle or under the same operating conditions and other studies where the comparisons were
made between sets of newer vs. older vehicles. Jobson et al. (1994) and Odaka et al. (1998, 2000) both
showed in laboratory studies that aged catalysts can result in increased N2O emissions. Odaka et al.
(2000) suggested that this could be attributed more to a decline in the ability of the catalyst to decompose
N2O than a reduction in the generation of N2O. Odaka et al. (1998) found that the effect of catalyst age on
Contract 68-W-99-054, WA 4-37 29 ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles Draft Report
N2O emission levels depended on the catalyst composition, with Pt/Rh and Pd catalysts with high metal
contents showing little differences with catalyst age while Pt/Rh catalysts with low metal contents showed
dramatic increases with catalyst age. Some studies have shown that N2O formation also occurs at higher
temperatures with aged compared to new catalysts (Odaka et al., 1998). It has been suggested that this
could lead to higher actual N2O vehicle emissions because a greater proportion of the driving cycle may
occur in the higher temperature "window" of formation (Lipman and Delucchi, 2002).
In actual vehicle applications, the effects of catalyst age on N2O emissions have been mixed. Odaka et al.
(1998) found that N2O emissions for a low Pt/Rh content passenger car increased under stabilized
operating conditions after 30,000 miles of operation, but decreased under start conditions. De Soete
(1993) found that N2O emissions increased comparing a catalyst aged to 15,000 miles with a new
catalyst. Durbin et al. (2003), on the other hand, found that catalyst age did not have a statistically
significant effect on N2O emissions in comparing 12 vehicles operated with new and aged catalysts.
These vehicles were all late model and in the LEV category. In other studies, vehicles with higher
mileage/older catalysts also represented older technologies so these data are more difficult to interpret
(Ballantyne et al., 1994; Laurikko and Aakko, 1995).
It is expected that CH4 emissions would also show deterioration with age. Lipman and Delucchi (2002)
characterized CH* emissions from a variety of sources and observed that most data showed an increase in
CH4 emissions with catalyst age. They found that modern vehicles and fuels (e.g., 1990s vintage vehicles
operating on reformulation fuels) showed emissions levels of 50 mg/mi under new conditions, rising to
150 mg/mi when the catalyst was significantly aged. For older three-way catalyst vehicles, they found
these vehicles in a "new" condition had emission rates of approximately 100 mg/mi, increasing to 300
mg/mi with higher age.
RESPONSE: In estimating the HC, CO, and NOx emissions in the MOVES2006 version, we
shall consider the effects on those emissions of vehicle age and/or mileage
accumulation. We shall use that opportunity to revisit our estimates ofCH4 and
N2O emissions to determine if they are also sensitive to vehicle age and/or
mileage accumulation.
5.1 Ambient Temperature Effects
Ambient temperature is known to have impacts on regulated emissions, with emissions increasing at
colder temperatures. Few studies have directly looked at ambient temperature effects on CFL, and N2O
emissions. Ahlvik (2002) looked at the effects of temperature between -7 and 22°C on N2O emissions for
2 light-duty gasoline and 2 light-duty diesel vehicles. The lower temperature results only showed a large
increase for one of the gasoline vehicles, with slight changes for two other vehicles. Stump et al. (1989,
1990) looked at temperature and oxygenated fuel effects on CFLj emissions. THC emissions decreased
slightly as the temperature was increased from 40 to 90°F, with CH4 emissions proportionally changing
with THC. These results suggest the effects of temperature on CFL, emissions can be estimated from the
effects of temperature on THC.
RESPONSE: This analysis (by ICF) was limited to emissions produced within the standard
FTP temperature range (68 to 86 degrees Fahrenheit). Temperature adjustment
factors will be developed for the MOVES2006 version of the model.
Presentation and Report Clarity. Some additional details and information would help to clarify some
of the steps of the methodology and how the emission factors will be implemented.
Contract 68-W-99-054, WA 4-37 30 ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles Draft Report
The level of detail used in describing the data sources is very limited. It would be useful to have some
description of the data sources to give the reader an idea of what data may have been included/excluded.
In the EPA's "Inventory of U.S. Greenhouse Gas Emissions and Sinks, 1990-2001" document (US EPA,
2003), a short discussion of data sets is provided in appendix E. Something similar could be added here.
For the CFLt data, it would be useful to include descriptions of some of the larger data sources, since
considerably more records are available for CFLj. For N2O, since data is limited, the specific sources could
even be listed as done in the US EPA (2003) document. It would also be useful to have a table of the
number of vehicles used in each technology for each of the pollutants, instead of just the information
provided in footnote 3 on page 4.
RESPONSE: Appendix A has been added to this report to provide a brief description of the test
data used by ICF to produces these estimated emission rates.
It would be useful to provide a brief one or two paragraph description of the how the emission factors will
be implemented in to the EPA MOVES model. If not, a reference to where more details on the model can
be found. How are running and start emissions going to be implemented into the model? How will the
running emissions be implemented in terms of the MOVES modal binning structure? Finally, emissions
are broken down into g/mi and g/hour, but there is no discussion on how the g/hour would be
implemented into the model.
RESPONSE: The discussion of how these emission rates are incorporated into the MOVES
model appears in the MOVES technical report entitled "MOVES2004 Energy
and Emissions Inputs. "
A description of the technology categories is provided, but no information is provided on the different
weight categories used (i.e., light-duty vs. heavy-duty).
Contract 68-W-99-054, WA 4-37 31 ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles Draft Report
GRAMMAR
i.
p 4 - paragraph below equation 1. line 9. There is an extra period after - segments of the FTP.
2. p. 19 - Table 23. Is the first entry under the emission control technology supposed to be non-
catalyst instead of moderate?
RESPONSE: The reviewer is correct on both points. The text has been corrected.
References
AAM/AIAM (2001) "Industry Low-Sulfur Test Program." Presented to the California Air Resources
Board by the Alliance of Automobile Manufacturers (AAM) and the Association of International
Automobile Manufacturers (AIAM), www.arb.ca.gov/cbg/meeting/200 l/mtg2001 .htm.
September.
Ahlvik, P. and Brandberg,A.R.L. (2000) "Relative Impact on Environment and Health from the
Introduction of Low Emission City Buses in Sweden." SAE Technical Paper No. 2000-01-1882.
Ahlvik, P. (2002) "Environmental and Health Impact from Modern Cars - A Comparison between two
Petrol and Two diesel Cars with Varying Emission Control Tehnology," Vagverket: Swedish
National Road Administration, Publication No. 2002:62, May.
Ballantyne, V. F., Howes, P., and Stephanson, L. (1994). "Nitrous Oxide Emissions from Light Duty
Vehicles." SAE Technical Paper No. 940304.
Baronick, J., Heller, B., Lach, G., and Ramacher, B. (2000). "Impact of Sulfur in Gasoline on Nitrous
Oxide and Other Exhaust Gas Components." SAE Technical Paper No. 2000-01-0857.
Becker, K. H., Lorzer, J. C., Kurtenbach, R., Wiesen, P., Jensen, T. E., and Wallington, T. J. (1999).
"Nitrous oxide (N2O) emissions from vehicles." Environmental Science & Technology, 33(22),
4134-4139.
Behrentz, E., Ling, R., Rieger, P., and Winer, A. M. (2004). "Measurements of Nitrous Oxide Emissions
from Light-Duty Motor Vehicles: A Pilot Study." Atmospheric Environment, 38(26), 4291-4303.
Benson, J.D., Burns, V., Koehl, W.J., Gorse, R.A., Painter, L.J., Hochhauser, A.M., and Reuter, R.M.
(1991) "Effects of Gasoline Sulfur Level on Mass Exhaust Emissions - Auto/Oil Air Quality
Improvement Research Program." SAE Technical Paper No. 912323.
Dasch, J. M. (1992). "Nitrous-Oxide Emissions from Vehicles." Journal of the Air & Waste Management
Association, 42(1), 63-67.
De Soete, G. (1993) "Nitrous Oxide from Combustion and Industry: Chemistry, Emissions, and Control."
Rev de I'lnstitut Francois du Petrole, vol 48, pp. 413-451.
Dietzmann, H.E., Parness, M.A., and Bradown, R.L. (1981) "Emissions from Gasoline and Diesel
Delivery Trucks by Chassis Transient Cycles." American Society of Mechanical Engineers, New
York.
Contract 68-W-99-054, WA 4-37 32 ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles Draft Report
Durbin, T.D., Collins, J.R., Norbeck, J.M., and Smith, M.R. (1999) "Evaluation of the Effects of
Alternative Diesel Fuel Formulations on Exhaust Emissions Rates and Reactivity," Final report
by CE-CERT to the South Coast Air Quality Management District under Contract No. 98102,
April.
Durbin, T. D., Miller, J. W., Pisano, J. T., Younglove, T., Sauer, C. G., Rhee, S. H., Huai, T., and
MacKay, G. I. (2003). "The Effect of Fuel Sulfur on NH3 and Other Emissions from 2000-2001
Model Year Vehicles." Final report for the Coordinating Research Council, under CRC Project
No. E-60, by the Bourns College of Engineering-Center for Environmental Research and
Technology at the University of California at Riverside, CE-CERT Technical Report No. 02-VE-
59971-E60-02, May.
Fanick, E.R., Schubert, P.P., Russel, B.J., and Freerks, R.L. (2001) "Comparison of Emission
Characteristics of Conventional, Hydrotrated, and Fischer-Tropsch Diesel Fuels in a Heavy-Duty
Diesel Engine." SAE Technical Paper No. 2001-01-3519.
Graham, L. (1999) "Gaseous and Particulate Matter Emissions from In-Use Light-Duty Gasoline Motor
Vehicles." ERMD Report #99-67 by the Emissions Research and Measurement Division of
Environment Canada.
Gautam, M, Clark, N.N., Wayne, W.S., Thompson, G., Lyons, D.W., riddle, W.C., Nine, R.D., Staggs,
B., Williams, V.A., Hall, T., and Thiagarajan, S. (2003) "Heavy-Duty Vehicle Chassis
Dynamometer Testing for Emissions Inventory, Air Quality Modeling, Source Apportionment
and Air Toxics Emissions Inventory" Final report for CRC Project No. E-55/E-59, April.
Guatam, M., Gupta, D., El-Gazzar, L., Lyons, D.W. and Popuri, S. (1996) "Speciation of Heavy-Duty
Diesel Exhaust Emissions under Steady-State Operating Conditions" SAE Technical Paper No.
962159.
Huai, T., Durbin, T. D., Rhee, S. H., Miller, J. W., and Norbeck, J. M. (2002). "The Impact of Gasoline
Fuel Sulfur Levels on NH3 and N2O Emissions." Proceedings of the 12th CRC On-Road Vehicle
Emissions Workshop, San Diego, CA, April.
Huai. T., Durbin, T. D., and Norbeck, J. M. (2003). "Analysis of Nitrous Oxide and Ammonia Emissions
from Motor Vehicles." Final report for the California Air Resources Board, by the Bourns
College of Engineering-Center for Environmental Research and Technology at the University of
California at Riverside, CE-CERT Technical Report No. 03-VE-18599-01, October.
Huai, T., Durbin, T. D., Miller, J. W., and Norbeck, J. M. (2004). "Estimates of the Emission Rates of
Nitrous Oxide from Light-Duty Vehicles using Different Chassis Dynamometer Test Cycles."
Atmospheric Environment, in press.
Jobson, E., Smedler, G., Malmberg, P., Bernler, H., Hjortsberg, O., Gotterg, I., and Rosen, A. (1994).
"Nitrous Oxide Formation over Three-way Catalysts." SAE Technical Paper No. 940926.
Jones, J. (2000) Final Report for Surveillance Testing of Motorcycles for Emissions. Final Report
prepared by California Environmental Engineering for the California Air Resources Board under
contract #97-600, February.
Korotney, D.J., Rao, V., Lindhjem, C.E., and Sklar, M.S. (1995) "Reformulated Gasoline Effects on
Exhaust Emissions: Phase III; Investigation on the Effects of Sulfur, Olefins, Volatility, and
Aromatics and the Interactions Between Oldfins and Volatility or Sulfur." SAE Technical Paper
No. 950782.
Contract 68-W-99-054, WA 4-37 33 ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles Draft Report
Laurikko, J., and Aakko, P. (1995). "The Effect of Ambient Temperature on the Emissions of Some
Nitrogen Compounds: A Comparative Study on Low-, Medium- and High-Mileage Three-Way
Catalyst Vehicles." SAE Technical Paper No. 950933.
Lipman, T. E., and Delucchi, M. A. (2002). "Emissions of nitrous oxide and methane from conventional
and alternative fuel motor vehicles." Climatic Change, 53(4), 477-516.
Merritt, P.M. (2003) "Diesel Exhaust Standard Phase 1: CRC Project No. AVFL-10A" Final Report for
the Coordinating Research Council, August.
Michaels, H., Fulper, C., and Kolowich, B. (1998). "Nitrous Oxide Emission Factors for Mobile Sources."
Presented at the A WMA Emission Inventory Conference, New Orleans, LA, December.
Odaka, M., Koike, N., and Suzuki, H. (1998). "Deterioration Effect of Three-way Catalyst on Nitrous
Oxide Emission." SAE Technical Paper No. 980676.
Odaka, M., Koike, N., and Suzuki, H. (2000). "Influence of Catalyst Deactivation on N2O Emissions from
Automobiles." Chemosphere-Global Change Science, vol. 2, pp. 413-423.
Oyama, K. and Kakegawa, T. (2000) "Japan Clean Air Program - Step I Study of Diesel Vehicle and Fuel
Influence on Emissions" SAE Technical Paper No. 2000-01-1973.
Prigent, M., and De Soete, G. (1989). "Nitrous Oxide (N2O) in Engine Exhaust Gases - A First Appraisal
of Catalyst Impact." SAE Technical Paper No. 890492.
Robertson, K. (1991) "Emissions of N2O in Sweden - Natural and Agricultural Sources." Ambio, vol. 20,
pp.151-154.
Rutherford, J.A., Burns, V.R., Leppard, W.R., Rippon, B., Koehl, W.J., Hochhauser, A.M., Painter, L.J.,
Reuter, R.M., Benson, J.D., Knepper, J.C., and Rapp, L.A. (1995) "Effects of Gasoline Properties
on Emissions of Current and Future Vehicles - T50, T90, and Sulfur Effects - Auto/Oil Air
Quality Improvement Research Program." SAE Technical Paper No. 952510.
Sasaki, S. and Kameoka, A. (1992) "Nitrous Oxide Emissions from Automobiles" 5th International
Workshop on Nitrous Oxide Emissions, Research Institute of Innovative Technology for the
Earth, Tsukuba, Japan.
Schauer, J.J., Kleeman, M.J., Cass, G.R., and Simoneit, B.R.T. (1999) Environ. Sci. Technol., vol. 33, pp.
1578-1587.
Siegl, W.O., Hammerle, R.H., Herrmann, H.M., Wenclawaik, B.W., and Luers-Jongen, B. (1999) Atmos.
Environ., vol. 33, pp 797-805.
Stump, F.D., Knapp, K.T., and Ray, W.D. (1990) "Seasonal Impact of Blending Oxygenated Organics
with Gasoline on Motor Vehicle Tailpipe and Evaporative Emissions," J. Air Waste Manage.
Assoc., vol. 40, pp. 872-880.
Stump, F.D., Knapp, K.T., Ray, W.D., Burton, C., and Snow, R. (1990) "The Seasonal Impact of
Blending Oxygenated Organics with Gasoline on Motor Vehicle Tailpipe and Evaporative
Emissions - Part II" SAE Technical Paper No. 902129.
United States Environmental Protection Agency. (2003). "Inventory of U.S. Greenhouse Gas Emissions
and Sinks, 1990-2001." Office of Mobile Sources, Assessment and Modeling Division, U.S.
Environmental Protection Agency, May.
Ullman, T.L., Smith, L.R, Anthony, J.W., Slodowske, W.J., Trestrail, B., Cook, A.L., Bunn, W.B.,
Lapin, C.A., Wright, K.J., and Clark, C.R. (2003) "Comparison of Exhaust Emissions, including
Contract 68-W-99-054, WA 4-37 34 ICF Consulting
-------
Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles Draft Report
Toxic Air Contaminants, from Schools Buses in Compressed Natural Gas, Low Emitting Diesel,
and Conventional Diesel Engine Configuations." SAE Technical Paper No. 2003-01-1381.
Urban, C.H., and Garbe, R.J. (1979) "Regulated and Unregulated Exhaust Emissions from
Malfunctioning Automobiles." SAE Technical Paper No. 790696.
Warner-Selph, M.A. and Harvey, C.G. (1990) "Assessment of Unregulated Emissions from Gasoline
Oxygenated Blends." SAE Technical Paper No. 902131.
Contract 68-W-99-054, WA 4-37 35 ICF Consulting
------- |