MOVES2010
Highway Vehicle Temperature,
Humidity, Air Conditioning, and
Inspection and Maintenance Adjustments
&EPA
United States
Environmental Protection
Agency
-------
MOVES2010
Highway Vehicle Temperature,
Humidity, Air Conditioning, and
Inspection and Maintenance Adjustments
Assessment and Standards Division
Office of Transportation and Air Quality
U.S. Environmental Protection Agency
NOTICE
This technical report does not necessarily represent final EPA decisions or
positions. It is intended to present technical analysis of issues using data
that are currently available. The purpose in the release of such reports is to
facilitate the exchange of technical information and to inform the public of
technical developments.
&EPA
United States
Environmental Protection
Agency
EPA-420-R-10-027
December 2010
-------
Table of Contents
Table of Contents i
List of Tables ii
List of Figures iii
Glossary of Acronyms iv
1. Introduction 1
2. Temperature Adjustments 1
2.1 Data Sources for Temperature Effects 1
2.2 Temperature Adjustment Methodology 2
2.3 Effects of Temperature on Gasoline Fueled Vehicles 3
2.3.1 Temperature Effects on Gasoline Start Emissions 3
2.3.1.1 HC and CO Start Emissions for Gasoline-Fueled Vehicles: 3
2.3.1.2 Temperature Effects on Gasoline NOx Start Emissions 5
2.3.1.3 Temperature Effects on Gasoline PM Start Emissions 7
2.3.2 Temperature Effects on Gasoline Running Emissions 7
2.4 Effects of Temperature on Diesel Fueled Vehicles 9
2.5 Cold Weather Effects 12
2.5.1 Cold Weather CO Requirement 12
2.5.2 Cold Weather HC Requirement 12
2.5.3 Cold Weather PM Effects 14
3. Humidity Adjustments 17
4. Air Conditioning Adjustments 17
4.1 Air Conditioning Effects Data 18
4.2 Mapping Data to VSP Bins 20
4.3 Air Conditioning Effects on Emissions 22
4.3.1 A/C Adjustments for HC, CO and NOx Emissions 22
4.3.2 Full A/C Adjustments for Energy Consumption 22
4.3.3 Uncertainty Analysis 23
4.4 Adjustments to Air Conditioning Effects 25
5. Inspection and Maintenance Programs 28
5.1 Inspection & Maintenance in MOBILE6 28
5.2 Inspection & Maintenance in MOVES 29
5.3 Devel opment of MO VES I/M Factors 30
5.4 Development of MOVES I/M Compliance Inputs 32
6. References 34
Appendix A-Mean Start Emission by Temperature 36
Appendix B - Calculation of Specific Humidity 38
Appendix C - Air Conditioning Analysis Vehicle Sample 39
Appendix D - Toros Topaloglu, Comments 41
Appendix E - ENVIRON International Corporation, Comments 48
Appendix F - Julio Vassallo, Comments 52
Appendix G - Coordinating Research Council Proj ect E-68a Comments 54
-------
List of Tables
Table 2-1 Average NOx Emission Results by Temperature 7
Table 2-2 Diesel Vehicle Emissions by Temperature 9
Table 2-3 Phase-In of Vehicles Meeting Cold Weather HC Standard 13
Table 2-4 Multiplicative Increases of PM at 20° Fahrenheit 16
Table 2-5 Exponential Equation Constant Terms 16
Table 3-1 Humidity Correction Coefficients Used by MOVES 17
Table 4-1 Distribution of Test Vehicles by Model Year 19
Table 4-2 Distribution of Tests by Schedule Type 19
Table 4-3 VSP Bin Definitions 21
Table 4-4 Full Air Conditioning Adjustments for HC, CO and NOx 22
Table 4-5 Full Air Conditioning Adjustments for Energy 23
Table 4-6 Fraction of Vehicles Equipped with Air Conditioning 26
Table 4-7 Fraction of Air Conditioning Units Still Functioning By Age 27
Table 4-8 Effect of Heat Index on Air Conditioning Activity 27
Table 5-1 MOBILE6.2 Runs Used to Populate the MOVES IMF actor 31
Table 5-2 I/M Coverage Table Data Sources 33
Table A-l Change in Mean Start Emissions at Various Temperatures 36
Table A-2 Change in Mean Start Emissions at Various Temperatures 37
Table C-l Vehicle Sample for the Air Conditioning Analysis 39
11
-------
List of Figures
Figure 2-1 Effects of Ambient Temperature on Changes in Cold-Start NOx 6
Figure 2-2 Logarithm of Bag-2 HC Versus Temperature 8
Figure 2-3 Cold-Start HC Emissions (in grams) with Confidence Interval 10
Figure 2-4 Bag-1 minus Bag-3 CO Emissions (in grams) with Confidence Interval 11
Figure 2-5 Bag-1 minus Bag-3 NOx Emissions (grams) with Confidence Intervals 11
Figure 2-6 FTP Bag 1 PM and FTP Bag 1 NMHC for Tier 2 Vehicles 15
in
-------
Glossary of Acronyms
A/C
ACCF
ASM
CO
EPA
F
FID
FTP
g/mi
GVWR
HC
HLDT
I/M
LOT
LDV
LLDT
MDPV
MOVES
MSAT
NMHC
NOx
OBD
PM
RSD
SFTP
THC
VIN
voc
Air Conditioning
Air Conditioning Correction Factor
Acceleration Simulation Mode
Carbon Monoxide
Environmental Protection Agency
Fahrenheit
Flame lonization Detection
Federal Test Procedure
Grams per Mile
Gross Vehicle Weight Rating
HydroCarbons
Heavy Light Duty Truck
Inspection and Maintenance
Light Duty Truck
Light Duty Vehicle
Light Light Duty Truck
Medium Duty Passenger Vehicle
MOtor Vehicle Emission Simulator
Mobile Source Air Toxics
Non-Methane HydroCarbons
Oxides of Nitrogen
On-Board Diagnostic
Paniculate Matter
Remote Sensing Data
Supplemental Federal Test Procedure
Total HydroCarbons (FID detection)
Vehicle Identification Number
Volatile Organic Compounds
IV
-------
1. Introduction
The emission rates in the MOVES model database represent a single (base) scenario of
conditions for temperature, humidity, air conditioning load and fuel properties. MOVES is
designed to adjust these base emission rates to reflect the conditions for the location and time
specified by the user. MOVES also includes a methodology for adjusting the base emission rates
to reflect the effects of local Inspection and Maintenance (I/M) programs. This report describes
how these adjustments for temperature, humidity, I/M and air conditioning were derived.
Adjustments for fuel properties are being addressed in a separate report.
This report describes adjustments that affect running exhaust, start exhaust and extended
idling emissions. The crankcase emission processes are chained to running exhaust, engine start
and extended idling emissions, and thus are similarly affected by the temperature adjustments
describe in this report. The impact of fuels, temperatures and I/M programs on vapor venting,
permeation and liquid leaks is addressed in a separate report on evaporative emissions.
2. Temperature Adjustments
In EPA's previous emissions model (MOBILE6), passenger car and light-duty truck tailpipe
emissions were adjusted relative to its base emission rates at 75 degrees Fahrenheit based on:
1. ambient temperature [1], and
2. for start emissions, an adjustment based on the length of the soak time. [2]
MOVES will take a similar approach, but we will substantially alter the nature of the temperature
adjustments.
2.1 Data Sources for Temperature Effects
For this analysis, we used almost entirely "Bagged" tests. Those data sets consisted of
Federal Test Procedure (FTP) and LA-92 tests for start emissions. For the temperature effects on
running emissions we used the Bag-2 emissions of those FTPs as well as US06 tests (without
engine starts). Some second-by-second test data were also used but only to validate the effects of
temperature on running emissions (HC, CO, and NOx). The data used in these analyses come
from the following four sources:
1. EPA's Mobile Source Observation Database (MSOD) as of April 27, 2005. Over the past
decades, EPA has performed emission tests (usually the FTP) on tens of thousands of
vehicles under various conditions. EPA has stored those test results in its Mobile Source
Observational Database (MSOD). (EPA has supplemented those tests with the results
from many non-EPA testing programs.)
For the MSOD data, we limited our analysis to only tests from the vehicles that were
tested at two or more temperatures. In this analyses, those paired (MSOD) tests covered
-------
the temperature range from 15 to 110 degrees Fahrenheit. Many of those bagged tests
(FTPs) were also used in our earlier MOBILE6 analyses.
Information on EPA's MSOD is available on EPA's website:
http://www.epa.gov/otaq/models.htm
2. A testing program in Kansas City also yielded pairs of test (using LA92s tests rather than
FTPs) from the vehicles that were tested at two or more temperatures. Information on
this study is available at:
http://www.epa.gov/otaq/emission-factors-research/420r08009.pdf
3. EPA's Office of Research and Development (ORD) contracted (through the Clean Air
Vehicle Technology Center, Inc.) the testing of five cars (model years 1987 through
2001). Those vehicles were tested using both the FTP and the EVI240 cycles at
temperatures of: 75, 40, 20, 0 and -20 °F. These five vehicles supplemented the vehicles
from the MSOD and Kansas City . [3]
4. Under a contract with EPA, the Southwest Research Institute (SwRI) tested four Tier 2
vehicles (2005 model year car and light-duty trucks) over the FTP at temperatures of: 75,
20, and 0 °F. These four vehicles also supplemented the vehicles from the MSOD and
Kansas City.
2.2 Temperature Adjustment Methodology
For our analyses of gasoline vehicles, we stratified the paired-test data by the same
parameters that MOVES uses to define the Source Bins, namely: fuel type, regulatory class, and
model year groups.
For this analysis, we started with the model year groups used in MOVES for start emission
rates. By combining several finer model year groups, we consolidated those (MOVES) model
year groups into these six model year groups:
- 1960 to 1980
- 1981 to 1982
- 1983 to 1985
- 1986 to 1989
- 1990 to 2004
- 2005 and later
A preliminary analysis of the test data indicated that vehicles meeting the Tier 0, Tier 1, and
LEV standards all exhibited similar increases in emissions as the ambient temperature drops
from 75° F to 20° F. A single additive adjustment (for each of HC, CO, NOx) can represent this.
Both the Federal FTP and California's Unified Cycle are 3-mode (or 3-bag) tests in which the
first and third modes are identical driving cycles, but the first mode begins with a cold-start (cold
-------
engine and emission control equipment) and the third mode begins with a hot-start (relatively
warm engine and control equipment).. We used the adjusted difference of Bag-1 minus Bag-3
emissions to estimate the cold-start emissions (in grams) for each test.
Similarly, we used the emissions from the FTP Bag-2, IM240, and US06 tests to estimate the
ratios (i.e., multiplicative changes) in the hot-running emission rates.
We combined the test data from the passenger cars and the light-duty trucks. Therefore, the
only stratifying parameter in this analysis was the model year grouping.
Then, within each model year group, we used regression analysis of cold-start and hot-
running emissions versus temperature to find a polynomial fit to describe the change in emissions
as a function of temperature.
For simplicity, we only considered functions that were a multiple of "temperature minus 75°
F" raised the first, second, or third degree. This produced (additive) adjustments that exhibit zero
change at 75 degrees Fahrenheit.
2.3 Effects of Temperature on Gasoline Fueled Vehicles
2.3.1 Temperature Effects on Gasoline Start Emissions
The effects of ambient temperature on HC, CO, and NOx start emissions were modeled using
the following algorithm:
1. Using additive (rather than multiplicative) adjustments.
2. Calculating these adjustments as simple functions of one of the following measures:
— the temperature minus 75° F, or
— the square of the difference of the temperature minus 75° F, or
— the cube of the difference of the temperature minus 75° F.
This approach guarantees a value of zero change for the additive adjustment at 75° F (i.e.,
the nominal temperature of EPA's FTP test). Those coefficients are stored in the
MOVES database table named StartTempAdjustment.
3. Setting the value of the adjustments equal to zero for temperatures higher than 75C
Fahrenheit.
2.3.1.1 HC and CO Start Emissions for Gasoline-Fueled Vehicles:
As described above we used the difference between the Bag-1 emissions and the
corresponding Bag-3 emissions to estimate the cold-start emissions (in grams per start) for each
-------
test. For the gasoline-fueled vehicles, those cold-start emissions were then stratified by model
year group. The mean emissions at 75 °F were subtracted from each of the means to determine
the change in emissions as functions of ambient temperature. (See Appendix A for the resulting
average changes.)
As noted at the beginning of this section, we decided to model the changes in cold-start
emissions as a polynomial (linear, or a quadratic, or a cubic) function of the temperature minus
75° F. Thus, the shape of each adjustment curve at temperature below 75° F would determine
the shape of that curve at temperatures above 75° F. However, the predetermined shape of the
curve at temperatures above 75° F was not always in agreement (directionally) with the test data
above 75° F. Therefore, we set the value of those additive adjustments equal to zero for
temperatures higher than 75° F. Thus, we did not use the changes in emissions from temperature
above the FTP temperature range (68° to 86° F); however, those values are included (if available)
in Appendix A.
We performed a linear, quadratic, and cubic regressions on the data in Appendix A and then
selected the best fit from among those three. The following equations were, thus, chosen as the
"best fit" predictors of the change in cold-start emissions (in grams) as functions of the ambient
temperature:
For the Pre-198Is:
HC temperature Adjustment = tempAdjustTermA * (Temp. - 75)
where: tempAdjustTermA =-0.630705748 R-sqr = 0.99271
CO temperature Adjustment = tempAdjustTermA * (Temp. - 75)
where: tempAdjustTermA = -4.677330289 R-sqr = 0.98973
Each of those linear coefficients is stored in table StartTemp Adjustment, (for the cold-start,
i.e., opModelDof 108)
Forthel981-1982s:
HC temperature Adjustment = tempAdjustTermA * (Temp. - 75)
where: tempAdjustTermA =-0.413584322 R-sqr = 0.98368
CO temperature Adjustment = tempAdjustTermA * (Temp. - 75)
where: tempAdjustTermA =-4.630546442 R-sqr = 0.97761
For the 1983-1985s:
HC temperature Adjustment = tempAdjustTermA * (Temp. - 75)
where: tempAdjustTermA =-0.360706640 R-sqr = 0.88660
CO temperature Adjustment = tempAdjustTermA * (Temp. - 75)
where: tempAdjustTermA = -4.244442967 R-sqr = 0.96367
Forthel986-1989s:
HC* temperature Adjustment = tempAdjustTermB * Sqr_of (Temp. - 75)
where: tempAdjustTermB = 0.002413998 R-sqr = 0.98895
-------
CO temperature Adjustment = tempAdjustTermA * (Temp. - 75)
where: tempAdjustTermA =-1.089740827 R-sqr = 0.99401
Note: HC test data for this model year range were available down to an ambient
temperature of-20° F. However, the "best fit" HC regression curves (linear,
quadratic, and cubic) all exhibited less than ideal fits to those data at temperatures
from zero through 20° F. Deleting the test data at -20° F and rerunning the
regressions produced an improved estimate of the cold-start HC emissions in that
critical temperature range. Therefore, this proposed quadratic regression is based on
the changes in cold-start emissions at only temperatures from zero through 75° F.
For the 1990-2005s:
HC* temperature Adjustment = tempAdjustTermB * Sqr_of (Temp. - 75)
where: tempAdjustTermB = 0.002924240 R-sqr = 0.99409
CO* temperature Adjustment = tempAdjustTermA * (Temp. - 75)
where: tempAdjustTermA =-1.141434345 R-sqr = 0.99017
Note: As with the regressions performed on the test data from the 1986 through 1989
model years, both the HC and CO regressions produced superior estimators of both
HC and CO cold-start emissions (at temperatures above zero degrees F) when the test
data at -20° F was omitted. Therefore, both of these regressions were based on the
changes in cold-start emissions only at temperatures from zero through 75° F.
2.3.1.2 Temperature Effects on Gasoline NOx Start Emissions
For the effects on cold-start NOx emissions associated changes in ambient temperature, we
attempted the same model year stratification that we used for the HC and CO emissions.
However, as is illustrated in the following graph (Figure 2-1), the "by model year" temperature
effects on cold-start NOx emissions did not lend themselves to linear, quadratic, or cubic
regressions (possibly due to insufficient sample size). Also, not unexpectedly, most of the
coefficients produced by those regression analyses were not statistically significant.
-------
Figure 2-1 Effects of Ambient Temperature on Changes in Cold-Start NOx
9.0
t
d
in
d
L.
01
W
X
0
Z
t
d
+rf
n
0
01
01
d
0
6.0
3.0
0.0
-3.0
-20
20 40 60 80 100
Temperature (degrees Fahrenheit)
120
A visual inspection of Figure 2-1 suggests that only three model year groups (1990-1993,
2001, and 2005) exhibited patterns that would result in meaningful regression analyses. We
attempted to group the data into various other model year groups. The only grouping that
produced useful regression analysis results were the ones in which we average together all of the
NOx results (from Appendix A) to obtain the following table:
-------
Table 2-1 Average NOx Emission Results by Temperature
Delta NOx
Temp (grams)
-20.0 1.201
0.0 1.227
1974 0202
20.7 0.089
2274 iQ;i55
Delta NOx
Temp (grams)
31.0 -0.007
40.0 0.876
4878 0127
49.8 0.333
5lTO 0325
Delta NOx
Temp (grams)
54.2 0.438
76.3 0.000
9573 0225
97.1 0.370
10578 0543
Performing regression analyses on these data (again, using only the changes in the NOx cold-
start emissions for temperatures below 86° F as explained in Section 3.2), we found the "best fit"
equation to be:
NOx temperature Adjustment = tempAdjustTermA * (Temp. - 75)
where: tempAdjustTermA =-0.009431682 R-sqr = 0.611349
Although the value of R-squared is not as high as for the HC and CO regression equations,
the coefficient is statistically significant. If we were to evaluate that equation for temperatures
higher than 75° F, it would predict a negative change (i.e., a decrease) in the cold-start NOx
emissions (i.e., a decrease in cold-start NOx emissions), but the actual data indicate that the cold-
start NOx emissions increase as the ambient temperature rises above 90° F. Therefore (as with
the previous adjustments), this additive adjustment is set to zero for temperatures higher than 75°
F.
2.3.1.3 Temperature Effects on Gasoline PM Start Emissions
The temperature effects on particulate matter (PM) start emissions modeled in MOVES
using a multiplicative (not additive) exponential (not polynomial) adjustment. Thus analysis is
included as Chapters 7 and 8 of a separate report ("Analysis of Particulate Matter Emissions from
Light-Duty Gasoline Vehicles in Kansas City"). [4]
2.3.2 Temperature Effects on Gasoline Running Emissions
The test data analyzed to determine the effects of different ambient temperatures on running
emissions consisted of:
1. Bag-2 of the FTP for vehicles tested at multiple temperatures,
2. US06 for vehicles tested at multiple temperatures, and
3. Remote sensing data (RSD) on a random sample of vehicles
tested at Kansas City over a wide range of temperatures.
4. FTP and IM240 tests on a random sample of vehicles tested
at Kansas City
Those test data suggest that there is very little variation in those running emissions of HC,
CO, or NOx. Regression analyses found that the coefficients (slopes) were not statistically
significant (that is, the slopes were not distinguishable from zero). This is consistent with what
-------
we found in our analysis of the Kansas City data. This lack of correlation between running
emissions and ambient temperature is illustrated (as an example) by the following graph of the
HC data:
Figure 2-2 Logarithm of Bag-2 HC Versus Temperature
CM
D)
ro
_a
O
I
t
3 -
2
1
0 -
1C
1
^
-4 -
-5 -
R
«
.0 2C
*
*
^
+
.0 3{
* *
3
^ .
*
^
»
4
<
^ •
»
«? A
.0 ' *4C
* ^
« *
» * <•
* *+ **
<» »*
»*«»
* *«
*
t
» »
, * ^
• » ^
»* «*
»»* . t t
» V *'
^g *»^>5C
^
*v^*» <
:^'*
> ** *
»• * *»
» * ^ •
V
«
»
t *
« * •
«e,»«* 6C
•i>*
* **
»* » *
** •* «
, »*« *
*
* *
* »
* »
^
* »
*0 *^7C
^ *
(**** f*
»* x V
*•*»»*«
»#.*«
• » <
* $«
t *
«» «
• »•
i* »>«?
» ** * « »
^^^^ ^^. *^
***^> ^*± *
< **V**^
* 5 »»*•»
• *»
*
^
> » »
^
,* ^
»0 *»» ,9C
•
? •••
h * *
* •
»•* » »
* *
«
*
,
.« 10
•
D.O
Temp
In this plot, each point represents a single FTP Bag-2 test result from the Kansas City
program. A visual inspection of this plot of the natural log of the FTP Bag-2 HC emissions
suggests no strong relationship between the hot-running HC emissions and the ambient
temperature.
The CO and NOx plots are similar in that they also do not indicate a significant trend.
We looked at the second-by-second test data from IM240 tests run in Chicago (as part of
Chicago's I/M program) to validate this conclusion. To avoid variable preconditioning, we used
only second IM240s when back-to-back IM240s were performed, and for the other EVI240s we
examined the last 120 seconds of full duration EVI240s. We found no evidence of a temperature
effect between 5 and 95 degrees F.
The effect of temperature on hot running HC, CO, and NOx emissions is coded in MOVES
using polynomial functions as multiplicative adjustments. In this version of MOVES, we
propose to set all of those adjustments equal to 1.0, that is, no change in those running emissions
with temperature.
This was not the case for PM emissions. Our data did show a temperature effect for running
emissions of particulate matter. As for start emissions, the temperature effects on particulate
matter (PM) running emissions modeled in MOVES using a multiplicative (not additive)
-------
exponential (not polynomial) adjustment. This analysis is detailed in as Chapters 7 and 8 of the
"Analysis of Particulate Matter Emissions from Light-Duty Gasoline Vehicles in Kansas City").
[4].
2.4 Effects of Temperature on Diesel Fueled Vehicles
We were able to identify only 12 diesel-fueled vehicles with FTPs at multiple temperatures
(nine passenger cars and 3 light-duty trucks). However, only two of those 12 vehicles were
tested at temperatures within the normal FTP range (68° to 86° F). None of these diesel trucks
were equipped with after-treatment devices. The Bag-1 minus Bag-3 emissions for those tests
are shown in Table 2-2. We stratified the test results into four temperature bands which yielded
the following emission values (grams per start) and average temperature value:
Table 2-2 Diesel Vehicle Emissions by Temperature
(grams per start)
Temperature
34.6
43.4
61.5
69.2
Count
6
7
10
2
HC
2.55
2.68
1.69
1.2
CO
2.44
2.03
3
1.91
NOx
2.6
0.32
0.67
0.36
When we plotted the mean HC start emissions (above) versus temperature, we obtained the
following graph (where the vertical lines represent 90 percent confidence intervals and the
"dashed" line represents a linear regression through the data).
-------
Figure 2-3 Cold-Start HC Emissions (in grams) with Confidence Interval
30
40 50 60
Temperature (degrees F)
70
The dashed (blue) line in the figure is a linear regression line having as its equation:
HC = (-0.0420985982* Temperature)+ 4.22477812 R-sqr = 0.9040467
Transforming this equation into an equation that predicts the (additive) change/adjustment in
the cold-start HC emissions from light-duty diesel-fueled vehicles (in the MOVES format), we
obtain:
HC temperature Adjustment = tempAdjustTermA * (Temp. - 75)
where: tempAdjustTermA = -0.0420985982
The coefficient associated with this temperature adjustment term is statistically significant
although its coefficient of variation is relatively large (23.04 percent).
Again, this HC adjustment factor represent the difference of Bag-1 minus Bag-3 and must be
adjusted to estimate the cold-start HC emissions.
It proved more difficult to estimate a diesel light-duty vehicle temperature effect for CO and
NOx. because the cold-start CO and NOx emissions did not exhibit a clear trend relative to the
ambient temperature. Plotting the mean CO and NOx cold-start emissions versus ambient
temperature (with 90 percent confidence intervals) produced the following two graphs:
10
-------
Figure 2-4 Bag-1 minus Bag-3 CO Emissions (in grams) with Confidence Interval
i
30
40 50 60
Temperature (degrees F)
70
Figure 2-5 Bag-1 minus Bag-3 NOx Emissions (grams) with Confidence Intervals
i
I
30
40 50 60
Temperature (degrees F)
70
11
-------
Statistical analyses of both the diesel cold-start CO and NOx emissions failed to produce
coefficients that were significantly different from zero. Therefore, for both cold-start CO and
NOx adjustments from light-duty diesel-fueled vehicles, we propose to set the temperature
adjustment for start emissions to zero:
CO temperature Adjustment = tempAdjustTermA * (Temp. - 75)
where: tempAdjustTermA = 0.0
NOx temperature Adjustment = tempAdjustTermA * (Temp. - 75)
where: tempAdjustTermA = 0.0
Since gasoline adjustments were set to zero, the temperature effects for diesel running
exhaust were also set to zero.
Because temperature effects data was not available for heavy duty trucks, the light duty
results were extrapolated to these vehicles including thethe extended idling emission process for
heavy duty long haul diesel trucks. No attempt has been made to adjust the effects of
temperature on emissions to account for the introduction of after-treatment devices (such as
diesel particulate filters or oxidation catalysts) that will become more common on future diesel
fueled vehicles.
2.5 Cold Weather Effects
There are two sets of regulations that can affect our estimates of emissions at low temperature
(i.e., at 20 degrees Fahrenheit), namely the cold weather CO requirement and the cold weather
HC requirement.
2.5.1 Cold Weather CO Requirement
The cold weather CO requirement for the 1994 and newer model year LDVs and LDTs limits
the composite FTP CO emissions to 10.0 grams per mile at a temperature of 20 degrees
Fahrenheit. However, the FTP test results used for our analysis (for those model years) were
from vehicles that were certified as meeting that cold weather composite CO requirement. Thus,
the temperature adjustments (based on regressions of those FTP results) already incorporated that
cold weather CO requirement into MOVES.
2.5.2 Cold Weather HC Requirement
TheMobile Source Air Toxic (MSAT-2) rule included a limit on low temperature (i.e., at 20
degrees Fahrenheit) non-methane hydrocarbon (NMHC) emissions for light-duty and some
medium-duty gasoline-fueled vehicles. Specifically:
• For passenger cars (LDVs) and for the light light-duty trucks (LLDTs) (i.e., those with
GVWR up to 6,000 pounds), the composite FTP NMHC emissions should not exceed 0.3
grams per mile.
12
-------
• For heavy light-duty trucks (HLDTs) (those with GVWR from 6,001 up to 8,500 pounds)
and for medium-duty passenger vehicles (MDPVs), the composite FTP NMHC emissions
should not exceed 0.5 grams per mile.
These cold weather standards are to be phased-in beginning with the 2010 model year,
specifically:
Table 2-3 Phase-In of Vehicles Meeting Cold Weather HC Standard
Year
2010
2011
2012
2013
2014
2015
/
25%
50%
75%
100%
100%
100%
HLDTs /MDPVs
0%
0%
25%
50%
75%
100%
To incorporate this set of HC requirements into MOVES, we must first determine its impact
on the start emissions (both cold-start and hot-start) as well as on the running emissions for each
class of vehicles.
We already observed that changes in the ambient temperature do not have a significant effect
on the running THC emissions. Therefore, we will assume that the full impact of this
requirement will be on the start emissions.
Our earlier analysis of temperature effects on the emissions of Tier 2 vehicles was based on a
single gasoline-fueled passenger car and three light-duty trucks that were each FTP tested at zero,
20, and 75 degrees Fahrenheit. The average nonmethane HC (NMHC) composite FTP emissions
at 75° F were:
• 0.02 (0.0180) g/mile for the passenger car and
• 0.04 (0.0353) g/mile for the heavy light-duty trucks (over 6,000 GVWR).
Considering the MSAT-2 standards (0.30 and 0.50, respectively), this would mean the
NMHC composite FTP emissions could increase by no more than 0.28 grams per mile (i.e., 0.30
minus 0.02) for LDVs/LLDTs and by no more than 0.46 grams per mile for HLDTs/MDPVs as
the ambient temperature drops from 75° F to 20° F.
Applying those increases in NMHC emission rates to the composite FTP (which simulates a
trip of 7.45 miles in length), those rates convert to total NMHC increases of 2.086 grams (for
LDVs/LLDTs) and 3.427 grams (for HLDTs/MDPVs). Since a composite FTP is composed of a
7.45 mile driving cycle plus a generic engine start (57 percent hot-start and 43 percent cold-start),
those increases represent the increases in the generic start emissions. Using the ratio of hot-start
13
-------
to cold-start from our earlier analysis, the increase in NMHC cold-start emissions (as the ambient
temperature drops from 75° F down to 20° F) are:
• 0.5611592 grams for the LDVs/LLDTs and
• 0.9219045 grams for the HLDTs/MDPVs.
Since the analysis for the MSAT-2 rule assumed that increase in NMHC is linear with
temperature (decreasing 55 degrees from 75 down to 20), those rates convert to decreases in total
NMHC per cold-start of:
• -0.0102029 grams per degree F for the LDVs/LLDTs and
• -0.0167619 grams per degree F for the HLDTs/MDPVs.
These are the rates (slopes) that we propose to use in MOVES for cold-starts (i.e., starts that
follow a 12 hour engine soak). For the seven shorter soak periods (that MOVES uses as
opModes), we will continue to use the ARB soak adjustments for HC emissions for catalyst
equipped vehicles to estimate those HC emissions (following the seven shorter soak periods).
2.5.3 Cold Weather PM Effects
The MSAT-2 rule (signed February 9, 2007) does not explicitly limit cold weather emissions
of paniculate matter (PM). However, the Regulatory Impact Analysis (RIA) document [5] that
accompanied that rule noted there is a strong linear correlation between NMHC and PM2.5
emissions. That correlation is illustrated in the following graph (reproduced from that RIA) of
the logarithm of the Bag-1 PM2.5 versus the logarithm of the Bag-1 NMHC (for various Tier-2
vehicles).
14
-------
Figure 2-6 FTP Bag 1 PM and FTP Bag 1 NMHC for Tier 2 Vehicles
CM _
CD
DQ
CO .
8
%
V
V O A
X
-3
I
-2
i
-1
0
1
Bag1 NHMC-ln(g/mi)
Plot Icons are Vehicle-Specific
Therefore, the limitation on cold weather HC (or NMHC) emissions is expected to result in a
proportional reduction in cold weather PM2.5 emissions. In the MSAT-2 RIA (Table 2.1 .-9),
EPA estimated that this requirement would result in a 30 percent reduction of VOC emissions at
20° F. Applying the same analytical approach that was used in the RIA means that a 30 percent
reduction in VOC emissions would correspond to a 30 percent reduction in PM emissions at 20°
F (for Tier 2 cars and trucks).
EPA's earlier analysis (for MOVES) [4] indicated that ambient temperature affects both start
and running PM emissions, and that effect (for Tier 2 vehicles) is best modeled by (exponential)
multiplicative adjustments of the form:
A*(72-t)
Multiplicative factor = Q , where "t" is the ambient temperature
and where A = 0.0463 for cold-starts and
0.0318 for hot running
(See Table 12 in Reference [4], page 46.)
15
-------
Therefore, for Tier 2 vehicles not affected by the MS AT-2 requirements, as the temperature
decreases from 72° to 20° F, EPA expects PM emissions to increase by factors of:
• 11.10727 for cold-starts and
• 5.22576 for hot running.
Applying the 30 percent reduction for vehicles affected by the MS AT-2 requirements implies
a PM increase as the temperature decreases from 72° to 20° F of:
• 7.77509 for cold-starts and
• 3.65803 for hot running.
Combining this information with the MSAT-2 phase-in schedule from Table 2-3 leads to the
following (multiplicative) increases as the temperature decreases from 72° to 20° F:
Table 2-4 Multiplicative Increase in PM from 72° to 20° Fahrenheit
Model Year
2008
2009
2010
2011
2012
2013
2014
2015
LDVs / LLDTs
Start
11.10727
11.10727
10.27423
9.44118
8.60814
7.77509
7.77509
7.77509
Running
5.22576
5.22576
4.83383
4.44189
4.04996
3.65803
3.65803
3.65803
HLDTs/MDPVs
Start
11.10727
11.10727
11.10727
11.10727
10.27423
9.44118
8.60814
7.77509
Running
5.22576
5.22576
5.22576
5.22576
4.83383
4.44189
4.04996
3. 65803
Solving for the corresponding constant terms so that the preceding exponential equation will
yield these increases, gives us these "A" values:
Table 2-5 Exponential Equation Constant Terms
Model Year
2008
2009
2010
2011
2012
2013
2014
2015
LDVs / LLDTs
Cold-Start
0.046300
0.046300
0.044801
0.043175
0.041398
0.039441
0.039441
0.039441
Running
0.031800
0.031800
0.030301
0.028675
0.026898
0.024941
0.024941
0.024941
HLDTs/MDPVs
Cold-Start
0.046300
0.046300
0.046300
0.046300
0.044801
0.043175
0.041398
0.039441
Running
0.031800
0.031800
0.031800
0.031800
0.030301
0.028675
0.026898
0.024941
16
-------
We assume that the increases in the PM2.5 emissions apply proportionally to the Elemental
Carbon (EC) and Organic Carbon (OC) portions of the PM2.5 emissions. Sulfate PM emissions
are not affected by temperature.
Although the ARB factors that adjust the start emissions based on soak time were not
developed for PM emissions from gasoline-fuel vehicles, the finding that Tier 2 PM emissions
are proportional to HC emissions supports our decision to apply the HC soak adjustments to the
start PM emissions.
3. Humidity Adjustments
In EPA's previous emissions model, MOBILE6, only gasoline vehicle exhaust NOx
emissions were adjusted for humidity. MOVES adjusts both gasoline and diesel vehicle exhaust
NOx emissions. The base exhaust emission rates for NOx in all modes and all processes are
multiplied by a humidity adjustment. This factor is calculated using the following formula:
K = 1.0 - ( (Bounded Specific Humidity - 75.0) * Humidity Correction Coefficient)
The bounded specific humidity is in units of grains of water per pound of dry air. The
specific humidity is not allowed to be lower than 21 grains and is not allowed to be larger than
124 grains. If the specific humidity input exceeds these limits, the value of the limit is used to
calculate the humidity adjustment. Appendix B shows how the hourly relative humidity values
are converted to specific humidity used in this equation using temperature and barometric
pressure.
Table
3-1 Humidity Correction Coefficients Used by MOVES
Fuel Type
Gasoline
Diesel Fuel
Humidity Correction Coefficient
0.0038
0.0026
The diesel humidity correction coefficient is taken directly from the Code of Federal
Regulations[6]. The gasoline humidity correction coefficient is carried over from the coefficient
used in the MOBILE6 model.
4. Air Conditioning Adj ustments
The air conditioning adjustments in MOVES are based on the same data as was used in the
previous MOBILE6 model, but the adjustments themselves were recalculated to be consistent
with the MOVES methodology.
The proposed factors are based on a test procedure meant to simulate air conditioning
emission response under extreme "real world" ambient conditions. These factors predict
emissions which would occur during full loading of the air conditioning system, and will then be
17
-------
scaled down in MOVES according to ambient conditions in a modeling run. The second-by-
second emission data were analyzed using the MOVES methodology of binning the data
according to vehicle characteristics (source bins in MOVES) and vehicle specific power bins
(operating modes in MOVES). The results of the analysis showed statistically significant and
consistent results for three bin combinations (deceleration, idle and cruise/acceleration) and the
three primary exhaust pollutants (hydrocarbon, carbon monoxide and nitrous oxides). This
report shows the results of the analysis for the air conditioning adjustments used in MOVES for
HC, CO, NOx and energy consumption.
MOVES will make adjustments to total energy consumption and exhaust running HC, CO
and NOx emissions separately for each operating mode. The criteria pollutants (HC, CO and
NOx) are only affected for passenger car, passenger truck and commercial light truck source
types. Energy consumption is affected for all source types. The same adjustment values are used
for all source use types affected within a pollutant type.
4.1 Air Conditioning Effects Data
The data for the MOVES A/C Correction Factor (ACCF) was collected in calendar year 1997
and 1998 in specially designed test programs. In the programs the same set of vehicles were
tested at standard FTP test conditions (baseline) and at a nominal temperature of 95 F. Use of
the same set of vehicles and test cycles should eliminate most of the vehicle and test procedure
variability and highlight the difference between a vehicle operating at extreme ambient
conditions and at a baseline condition.
The data used to develop the MOVES ACCF consisted of 54 individual cars and light
trucks tested over a variety of test schedules. Overall the database consisted of a total of 625 test
cycles, and 1,440,571 seconds of emission test and speed / acceleration data. Because of the
need to compute vehicle specific power on a modal basis, only test results which consisted of
second by second data were used in the analysis. All second by second data were time aligned
and quality controlled checked.
The distribution of test vehicles by model year is shown in Table 4-1. Model years 1990
through 1999 were included. The data set consists of 30 cars and 24 light trucks. No test data
were available on other vehicle types (i.e., motorcycles, heavy trucks, etc). The individual test
cycles which the vehicles were run on are shown with the test counts in Table 4-2. The data
shows a nice balance between different test cycles, and cars and trucks. Unfortunately, the study
does not contain any pre-1990 model years. A complete list of the individual vehicles and a
basic description is shown in Appendix C.
Only vehicles which were coded as having an emission test with the A/C system on were
selected. The A/C On tests and the A/C Off (default for most EPA emission tests in general)
were matched by VEST, test schedule and EPA work assignment. The matching ensured that the
same vehicles and test schedules were contained in both the A/C On sample and the A/C Off
sample.
18
-------
Table 4-1 Distribution of Test Vehicles by Model Year
Model Year
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
TOTAL
Count
5
5
6
5
7
5
13
4
O
1
54
Table 4-2 Distribution of Tests by Test Cycle
Schedule Name
ART-AB
ART-CD
ART-EF
F505
FTP
FWY-AC
FWY-D
FWY-E
FWY-F
FWY-G
FWY-HI
LA4
LA92
LOCAL
NONFRW
NYCC
RAMP
ST01
TOTAL
Count
36
36
36
21
21
57
36
36
36
36
36
23
35
36
36
36
36
36
625
19
-------
4.2 Mapping Data to VSP Bins
The overall dataset consisted of a sample of vehicle tests with the A/C system on and a
sample of vehicle tests with the A/C system off. Both samples consisted on the same vehicles
and all tests were modal with a data sampling of 1 hertz (second-by-second data collection).
Prior to analysis the data for each vehicle / test cycle combination was time aligned to insure that
the instantaneous vehicle operating mode was in-sync with the emission collection system.
Following time alignment, the vehicle specific power (VSP) was calculated for each vehicle test /
second combination. This was done using Equation 1.
VSP = 985.5357* Speed* Acoeff/Weight +
440.5729 * SpeedA2 * Bcoeff /Weight +
196.9533 * SpeedA3 * Ccoeff/Weight +
0.19984476 * Speed * Accel + GradeTerm Eq 1
Where
VSP is the vehicle specific power for a given second of operation in units of KW / tonne.
Speed is the instantaneous vehicle speed for a given second in units miles / hour.
Accel is the instantaneous vehicle acceleration for a given second in unit of miles/hr-sec
Weight is the test vehicle weight in pounds.
Acoeff = 0.7457*(0.35/(50*0.447)) * ROAD_HP
Bcoeff = 0.7457*(0.10/(50*50*0.447*0.447))*ROAD_HP
Ccoeff = 0.7457*(0.55/(50*50*50*0.447*0.447*0.447)) * ROAD_HP
Where
ROAD_HP = 4.360117215 + 0.002775927 * WEIGHT (for cars)
ROAD_HP = 5.978016174 + 0.003165941 * WEIGHT (for light
trucks)
GradeTerm (KW/tonne) = 4.3809811 * Speed * Sin(Radians(GradeDeg))
Where
GradeDeg is the road grade in units of degrees. This term is zero for
dynamometer tests.
4.3809811 (mA2*hr/(sA3 * miles) =
9.80665(m/sA2) * 1609.34(m/mile) / 3600(secs/hr)
KW / tonne = mA2 / sA3
9.80665(m/sA2) is the gravitation constant.
20
-------
After computing the VSP for each vehicle test / second combination, we assigned the
individual secondsto the MOVES VSP bins. These VSP bins are defined in Table 4-3. VSP bins
26 and 36 were not defined because bins 27-30 and bins 37-40 overlap them.
Table 4-3 VSP Bin Definitions
VSP Label
0
1
11
12
13
14
15
16
21
22
23
24
25
26
27
28
29
30
33
35
36
37
38
39
40
Definition
Braking
Idling
Low Speed Coasting
Crui se/ Accel erati on;
Crui se/ Accel erati on;
Crui se/ Accel erati on;
Crui se/ Accel erati on;
Crui se/ Accel erati on;
; VSP< 0; K=Speed<25
0<=VSP< 3; 1<= Speed<25
3<=VSP< 6; K=Speed<25
6<=VSP< 9; K=Speed<25
9<=VSP<12; K=Speed<25
12<=VSP; K=Speed<25
Moderate Speed Coasting; VSP< 0; 25<=Speed<50
Crui se/ Accel erati on;
Crui se/ Accel erati on;
Crui se/ Accel erati on;
Crui se/ Accel erati on;
Crui se/ Accel erati on;
Crui se/ Accel erati on;
Crui se/ Accel erati on;
Crui se/ Accel erati on;
Crui se/ Accel erati on;
Crui se/ Accel erati on;
Crui se/ Accel erati on;
Crui se/ Accel erati on;
Crui se/ Accel erati on;
Crui se/ Accel erati on;
Crui se/ Accel erati on;
Crui se/ Accel erati on;
0<=VSP< 3; 25<=Speed<50
3<=VSP< 6; 25<=Speed<50
6<=VSP< 9; 25<=Speed<50
9<=VSP<12; 25<=Speed<50
12<=VSP; 25<=Speed<50
12<=VSP<18; 25<=Speed<50
18<=VSP<24; 25<=Speed<50
24<=VSP<30; 25<=Speed<50
30<=VSP; 25<=Speed<50
VSP< 6; 50<=Speed
6<=VSP<12; 50<=Speed
12 <= VSP; 50<=Speed
12<=VSP<18; 50<=Speed
18<=VSP<24; 50<=Speed
24<=VSP<30; 50<=Speed
30<=VSP; 50<=Speed
An average emission result for each pollutant (HC, CO and NOx) with and without A/C
operation was computed for each VSP Bin. This resulted in 69 (23 VSP bins x 3 pollutants)
pairs of emission averages. However, preliminary analysis of the data grouped into the 23 bins
(defined in Table 4-3) showed unsatisfactory statistical results. In the general, no trends were
evident across VSP bins or within similar subsets of VSP bins. The trends were highly erratic
and the results were generally not statistically significant. In addition, most of the bins labeled
30 or higher had very few data members. An analysis of cars versus trucks was also performed,
and showed no statistical difference between the two.
To produce more consistent results, the individual VSP bins were collapsed down to three
principal bins. These are the Braking / Deceleration bin, the Idle bin and the Cruise /
Acceleration bin. These large bins are quite different in terms of engine operation and emissions
performance. The Braking bin consisted of VSP Bin 0 in Table 4-3, the Idle bin was VSP Bin 1
and the Cruise / Acceleration bin contained the remaining 21 bins.
21
-------
4.3 Air Conditioning Effects on Emissions
4.3.1 A/C Adjustments for HC, CO and NOx Emissions
Full A/C adjustments were generated for each of the nine VSP Bin and pollutant
combinations. This was done by dividing the mean "With A/C" emission factor by the mean
"Without A/C" emission factor for each of the VSP Bin / pollutant combinations. The Full A/C
adjustments are shown in Table 4-4. Measures of statistical uncertainty (coefficient of variation
of the mean) were also computed using the standard error of the mean. They are shown in Table
4-4as"MeanCVofCF."
Table 4-4 Full Air Conditioning Adjustments for HC, CO and NOx
Pollutant
HC
HC
HC
CO
CO
CO
NOx
NOx
NOx
Operating Mode
Braking / Decel
Idle
Cruise / Accel
Braking / Decel
Idle
Cruise / Accel
Braking / Decel
Idle
Cruise / Accel
Full A/C CF
1.0000
1.0796
1.2316
1.0000
1.1337
2.1123
1.0000
6.2601
1.3808
MeanCVofCF
0.48582
0.74105
0.33376
0.31198
0.77090
0.18849
0.19366
0.09108
0.10065
4.3.2 Full A/C Adjustments for Energy Consumption
The use of a vehicle's A/C system will often have a sizeable impact on the vehicle's energy
consumption. This was found statistically by analyzing the available second by second data on
CO2 and other gaseous emissions, and converting them to an energy basis using standard EPA
vehicle fuel economy certification equations. The vehicle emission data were binned by VSPBin
(see above). A mean value was computed for each combination of VSPBin. Separate analysis
was done as a function of sourcebinid (combination of vehicle type, fuel type and model year),
and the results were not statistically different versus sourcebinid given the relatively small
sample sizes. As a result, the A/C adjustments for energy are a function of only VSPBin. The
resulting A/C adjustments are shown in Table 4-5.
22
-------
Table 4-5 Full Air Conditioning Adjustments for Energy
VSPBin
0
1
11
12
13
14
15
16
A/C Factor
1.342
1.365
1.314
1.254
1.187
1.166
1.154
1.128
VSPBin
21
22
23
24
25
26
27
28
29
A/C Factor
1.294
1.223
1.187
1.167
1.157
1.127
1.127
1.127
1.127
VSPBin
30
33
35
37
38
39
40
A/C Factor
1.294
1.205
1.156
1.137
1.137
1.137
1.137
Only very small amounts of data were available for VSPBins 26 through 29 and VSPBins
37 through 40. As a result, the data from these bins was averaged together and binned into two
groups. The resulting group averages were used to fill the individual VSPBins. This averaging
process has the effect of leveling off the effect of A/C at higher power levels for an engine. This
is an environmentally conservative assumption since it is likely that the engine power devoted to
an A/C compressor probably continues to decline as the overall power demand of the engine is
increased. In fact, in some newer vehicle designs the A/C unit will be shut off by an engine
controller if the driver demands a very high level of power from the vehicle. If and when new or
additional data become available on this issue, EPA will re-evaluate the assumption of a constant
A/C factor for the high VSPBins.
4.3.3 Uncertainty Analysis
Measures of statistical uncertainty as indicated by the coefficient of variation of the mean
(mean CV) were calculated using the following formula. The same set of equations were used
for each of the three pollutants (although the equations are shown only once). The values of X
and Y represent second by second emissions from either HC, CO or NOx. The variable "X"
represents emissions with the A/C On and "Y" represents emission with the A/C Off.
Given:
Mean CV
X /Y
SEZ/Z
Where Z is the ratio of A/C On emissions (X) to A/C Off emissions (Y)
SEZ is the standard error of Z
Mean CV is the coefficient of variation of the mean
23
-------
Vz2 = (8Z/8X)2 * Vx2 + (8Z/8Y)2 * Vy2
Where Vz is the variance of Z, Vx is the variance of X and Vy is the variance of Y
8Z/8X is the partial derivative of Z with respect to X
8Z/8Y is the partial derivative of Z with respect to Y
(Vz / Z)2 = ((1/Y2)*VX2) / (X2/Y2) + ((X2/Y4) * Vy2) / (X2/Y2)
This equation reduces to:
(VZ/Z)2 = (Vx / X)2 + (Vy / Y)2
And ultimately to:
SEZ / Z = SQRT [ (SEZ / X)2 + (SEZ / Y)2 ]
The variance term is defined as:
Vz = (1/Y)2 * Sy2x + (-X/Y2) * (-X/Y2) * Sy2y;
Where
X = A/C On emissions
Y = A/C Off emissions
The term Vz represents a contribution from both the X and Y emissions terms (A/C On and A/C
Off). The terms Sy2x and Sy2y also include variance contributions of the "across sample
variance" and the "within a given vehicle test" variance. The "across sample variance" is the
standard variance of the sample and is computed within a given sourcetype (vehicle type such as
car, light truck, heavy truck, etc) and operating mode bin (one of the 23 VSP bin types - See
Table 4-3). The "within a given vehicle test" variance is the additional variance due to the fact
that each vehicle test contributes hundreds or even thousands of test data elements. Because two
data elements may come from the same vehicle, they are not strictly independent of each other.
Sy2x = SA2x/nVeh + SB2x / nCell
Sy2y = SA2y/nVeh + SB2y/nCell
SA2x = (II (nVeh-1) ) * Sumlx
SB2x = ( 1 / (nCell - nVeh) ) * Sum2x
SA2y = (II (nVeh-1) ) * Sumly
SB2y = ( 1 / (nCell - nVeh) ) * Sum2y
24
-------
And
Sumlx = E (YbarVehx - YbarCellx )2
Sum2x = E ( varVehx - ( nMeas - 1) )2
Sumly = E ( YbarVehy - YbarCelly )2
Sum2y = E ( varVehy - ( nMeas - 1) )2
Where
The sums (E ) are across sourcetype and operating mode.
nMeas Count of data elements within a given sourcetype, operating mode and vehicle
test.
nVeh Count of data elements within a given vehicle test
nCell Count of data elements within a given sourcetype and operating mode
varVeh Variance for each vehicle test. Separate values for both X and Y are calculated.
YbarVeh Mean emission rate for each vehicle test. Separate values for both X and Y are
calculated.
YbarCell Mean emission rate for each sourcetype and operating mode. Separate values for
both X and Y are calculated.
For HC, CO and NOx, detailed VSP was not found to be an important variable in regards to
A/C adjustment and A/C usage. However, Full A/C adjustments greater than one were found for
all pollutants for both Idle and Cruise / Acceleration modes. For NOx Idle mode, a fairly large
multiplicative adjustment of 6.2601 was obtained. This large factor reflects the relatively low
levels of NOx emissions during idle operation. A moderately high multiplicative A/C adjustment
of (2.1123) for CO cruise / Accel was also obtained. These adjustments will double CO
emissions under extreme conditions of A/C usage. A/C adjustments of less than or equal to one
were found for the Braking / Deceleration mode for all three pollutants. These were set to one
for use in the MOVES model.
4.4 Adjustments to Air Conditioning Effects
The adjustments for each operating mode are weighted together by the operating mode
distribution calculated from the driving schedules used to represent the driving behavior of
vehicles. Average speed, road type and vehicle type will affect the operating mode distribution.
weightedFull AC Adjustment = SUM( fullACAdjustment*opModeFraction )
25
-------
Since not all vehicles are equipped with air conditioning and air conditioning is normally not
on all of the time, the full air conditioning effect on emissions is adjusted before it is applied to
the emission rate. The SourceTypeModelYear table of the MOVES database contains the
fraction of vehicles in each model year that are equipped with air conditioning [7].
Table 4-6 Fraction of Vehicles Equipped with Air Conditioning
(ACPenetration)
Model Year
1971*
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000**
Passenger Cars
0.592
0.592
0.726
0.616
0.631
0.671
0.720
0.719
0.694
0.624
0.667
0.699
0.737
0.776
0.796
0.800
0.755
0.793
0.762
0.862
0.869
0.882
0.897
0.922
0.934
0.9484
0.9628
0.9772
0.980
0.980
All Trucks and Buses
0.287
0.287
0.287
0.287
0.287
0.311
0.351
0.385
0.366
0.348
0.390
0.449
0.464
0.521
0.532
0.544
0.588
0.640
0.719
0.764
0.771
0.811
0.837
0.848
0.882
0.9056
0.9292
0.950
0.950
0.950
* 1971 model year fractions are applied to all previous model years.
** 2000 model year fractions are applied to all later model years.
Motorcycles are not adjusted for air conditioning.
The fraction of vehicles whose air conditioning is operational varies by age of the vehicle and
is stored in the SourceTypeAge table of the MOVES database.
26
-------
Table 4-7 Fraction of Air Conditioning Units Still Functioning By Age
Age
1
2
3
4
5
6
7
8
9
10
Functioning
1.00
1.00
1.00
0.99
0.99
0.99
0.99
0.98
0.98
0.98
Age
11
12
13
14
15
16
17
18
19
20
Functioning
0.98
0.98
0.96
0.96
0.96
0.96
0.96
0.95
0.95
0.95
Age
21
22
23
24
25
26
27
28
29
30
Functioning
0.95
0.95
0.95
0.95
0.95
0.95
0.95
0.95
0.95
0.95
An equation is used to predict the fraction of those vehicle owners who have air conditioning
available to them that will turn on the air conditioning based on the ambient temperature and
humidity (heat index [7]) of the air outside their vehicles. The heat index values are stored in the
ZoneMonthHour table of the MOVES database.
ACOnFraction = ACActivityTermA
+ heat!ndex*(ACActivityTermB + ACActivityTermC*heatIndex)
Table 4-8 Effect of Heat Index on Air Conditioning Activity
-3.63154
0.072465
-0.000276
ACActivityTermA
ACActivityTermB
ACActivityTermC
Heat Index
67.44
70
75
80
85
90
95
100
105
110
AC On Fraction
0.000
0.089
0.251
0.399
0.534
0.655
0.762
0.855
0.934
1.000
The fraction of vehicles equipped with air conditioning, the fraction of operational air
conditioning and the fraction of air conditioning use are used to adjust the amount of "full" air
conditioning that occurs in each hour of the day.
AC Adjustment = 1+ ( (weightedFullACAdjustment-1)
* ACPenetration*functioningACFraction*ACOnFraction )
27
-------
The air conditioning adjustment is a multiplicative adjustment applied to the emission rate
after it has been adjusted for fuel effects.
Air conditioners are employed for defogging at all temperatures, particularly, at lower
temperatures. This secondary use of the A/C along with associated emission effects is not
addressed in MOVES2010.
5. Inspection and Maintenance Programs
Inspection and Maintenance (I/M) programs are genetically any state-run or locally mandated
inspection of highway motor vehicles intended to identify those vehicles most in need of repair
and requires repairs on those vehicles. Since these programs are locally run, there is great
variability in how these programs are designed and the benefits that they generate in terms of
emission reductions from highway motor vehicles.
5.1 Inspection & Maintenance in MOBILE6
Because MOVES draws heavily on the approaches developed for MOBILE6.2 to represent
the design features of specific I/M programs, it is useful to briefly review these methods.
Readers interested in a more thorough treatment of the topic are encouraged to review the
relevant MOBILE documentation [9].
The MOBILE6.2 model used a methodology that categorized vehicles according to emitter
status (High emitters and Normal emitters), and applied a linear growth model to project the
fraction of the fleet that progresses from the Normal emitter to the High emitter status as a
function of age. Average emission rates of High and Normal emitters were weighted using the
High emitter fraction to produce an overall average emission rate as a function of age, model year
group and vehicle type. The emissions generated represented the emissions of the fleet in the
absence of I/M (the No I/M emission rate).
A similar approach was used to generate I/M emission rates. In this case the initial starting
point for the function (where age=0) was the same as the No I/M case. However, the effects of
I/M programs and associated repairs were represented by reductions in the fraction of high
emitters, which consequently affected the average emission level of the fleet. Balancing these
emissions reductions due to I/M repairs were the re-introduction of high emitters in the fleet due
to deterioration of vehicle emission control systems after repairs. The underlying I/M and non-
I/M deterioration rates were assumed to be the same.
With the passage of time, the non-I/M and I/M emission cases diverged from each other with
the I/M rates being lower. The percentage difference between these two rates is often referred to
as the overall I/M reduction or I/M benefit.
28
-------
5.2 Inspection & Maintenance in MOVES
The MOVES emission rates contain estimates of emission levels as a function of age, model
year group and vehicle type for areas where no I/M program exists (the mean base rate, or the
non-I/M reference rates) and for an area representing the "reference I/M program" (the I/M
reference rates). The I/M reference rates were derived using data from the enhanced I/M
program in Phoenix, Arizona, and represent the design features of that program. The difference
between the non-I/M and I/M reference rates are assumed to represent thel/M benefit of the
Phoenix program design assuming perfect compliance. Equation 1 shows this relationship in a
mathematical form.
Standard I/M difference = EnonM - EM Eq 1
where Enon_M and EM are the non-I/M and I/M reference rates, respectively.
The Phoenix program design was selected as the reference program because virtually all of
the underlying data for MOVES came from this source. The selection does not imply any
judgment on the strengths or weaknesses of this specific program. In MOVES, it is this general
I/M design which is the model, not the actual Arizona I/M program as it is operated.
The object of this process is to generate a general model which can be used to represent all
I/M programs in the United States. This goal was achieved by comparing individual program
designs against the reference program for purposes of developing adjustment to the "standard I/M
difference" representing design features differing from those in the reference program This
concept is shown mathematically in Equation 2,
^p = REjM + (1 ~ R)EnonlM EC1 2
where Ep is the adjusted emission rate for a "target" I/M program, EM is the reference rate, EnonM
is the non-I/M reference rate, and R is an aggregate adjustment representing the difference in
average emission rates between the target program and the reference program. Depending on the
value of R, Ep may be greater than EnonM, fall between EnonM and EM, or be less than EM- Thus
this framework can represent target programs as more effective or less effective than the
reference program. In MOVES, R is referred to as the "IMFactor."
Re-arranging Equation 2 and solving forR gives leads to Equation 3. This equation shows
the I/M adjustment as the ratio of the emission difference between a proposed I/M program
design and the Standard I/M Difference
E -EnonIM
R = — —— Eq 3
29
-------
5.3 Development of MOVES I/M Factors
Early in the MOVES development process it was decided that developing the I/M adjustment
factors based on a completely new analysis would prove infeasible. A major obstacle was a lack
of suitable emissions and I/M program data representing the full range of program designs. Data
sets for certain I/M programs (i.e., transient test based programs) were generally quite complete
and robust. However, mass emission results and random vehicles samples were quite scarce for
other test types such as the Acceleration Simulation Mode (ASM), steady-state, idle tests and
OBD-II scans. This situation was particularly true for old model years at young ages (i.e., a 1985
model year at age five). As a result, EPA decided to develop I/M adjustment factors based on the
information incorporated in MOBILE6.2. Mechanically, this step was achieved by running the
MOBILE6.2 model about 10,000 times over a complete range of pollutant-process
combinations, inspection frequencies, calendar years, vehicle types, test types, test standards, and
model year group / age combinations. The mean emission results for each combination were
extracted from the output and utilized. The IMF actor table includes the following fields:
• Pollutant / Process
• Test Frequency
• Test Type
• Test Standard
• Regulatory Class
• Fuel Type (Only gasoline/ethanol fuels have IMF actors)
• Model Year Group
• Age Group
• IMF actor
The IMF actor value was computed for each combination of the parameters listed in the
IMF actor table. A separate MOBILE6.2 run was done for each parameter combination (Target
design, Ep\ and a second set of runs were done describing the reference program (Reference
design, ER). The IMF actor is the ratio of the mean emission results from these two runs.
Equation 4 illustrates the simple formula.
The Reference program has inputs matching the Phoenix I/M program during the time in
which the data used in the MOVES emission rate development were collected (CY 1995-2005).
The Reference design represents a biennial frequency with an exemption period for the four most
recent model years.. It uses three different I/M test types (basic idle test for MY 1960-1980,
transient tailpipe tests for MY 1981-1995 (EVI240, EVI147), and OBD-II scans for MY 1996 and
late). Each of these test types became the Reference for the respective model year groups.
The specific combinations of MOBILE6.2 runs performed are shown in Table 5-1 below.
Each of these runs represents a particular test type and test standard design which was expressed
as a ratio to the standard reference tests. The first four runs represent the Non I/M reference and
30
-------
the three Phoenix I/M references. A set of these runs were done for each calendar year 1990
through 2030, for cars, light trucks and heavy-duty gasoline vehicles and for pollutants HC, CO
and NOx.
Table 5-1 MOBILE6.2 Runs Used to Populate the MOVES I/M Adjustment Factor
RUN#
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Description
Non I/M Base
IM240 Base (Biennial IM240/147)
OBDBase (Biennial OBD Test)
Basic Base (Loaded - Idle Test)
Biennial - IM240 - Phase-in Outpoints
Annual - IM240 - Phase-in Outpoints
Biennial - IM240 - Final Outpoints
Annual - IM240 - Final Outpoints
Biennial - ASM 2525/50 15 - Phase-in Outpoints
Annual - ASM 2525/50 15 - Phase-in Outpoints
Biennial - ASM 2525/5015 - Final Outpoints
Annual - ASM 2525/50 15 - Final Outpoints
Biennial - ASM 2525 - Phase-in Outpoints
Annual - ASM 2525 - Phase-in Outpoints
Biennial - ASM 2525 - Final Outpoints
Annual - ASM 2525 - Final Outpoints
Biennial - ASM 50 15 - Phase-in Outpoints
Annual - ASM 50 15 - Phase-in Outpoints
Biennial - ASM 50 15 - Final Outpoints
Annual - ASM 50 15 - Final Outpoints
Annual - OBD -
Annual - LOADED/IDLE
Biennial - IDLE
Annual - IDLE
Biennial - 2500/IDLE
Annual - 2500/IDLE
Type
Non I/M Reference
I/M Reference
I/M Reference
I/M Reference
Target I/M Design
Target I/M Design
Target I/M Design
Target I/M Design
Target I/M Design
Target I/M Design
Target I/M Design
Target I/M Design
Target I/M Design
Target I/M Design
Target I/M Design
Target I/M Design
Target I/M Design
Target I/M Design
Target I/M Design
Target I/M Design
Target I/M Design
Target I/M Design
Target I/M Design
Target I/M Design
Target I/M Design
Target I/M Design
The MOBILE6.2 database output option was chosen for all runs. This step produced large
sets of results which were further stratified by facility-cycle / start process and age. This output
format necessitated additional processing of the facility rates into composite running and start
factors (in MOVES the IMF actor is a function of running and start processes).
In addition to the IMF actor, MOVES adjusts rates for particular programs by applying an
additional multiplicative "Compliance Factor" (IMCompliance). The IMF actor (R) represents the
theoretical effectiveness of a specific I/M program design, relative to the reference design, as
described above.
Values of the EVIComplianceFactor (C) are specific to individual programs and represent its
overall operational effectiveness and efficiency, aside from the effectiveness inherent in its
design. Variables which impact the IMCompliance factor include waiver rates, compliance rates
and overall operational efficiency. Default EVIComplianceF actors are provided in the MOVES
database, but alternate values may be entered by the user for specific analyses. The default
factors were taken from the 2005 EPA National Emission Inventory (NEI) [10], and are based on
31
-------
data submitted by individual states in their State Implementation Plan (SIP) processes. The vast
majority of the default EVICompliance factors are greater than 90 percent.
5.4 Development of MOVES I/M Compliance Inputs
The default I/M Compliance inputs are contained in the EVICoverage table in the MOVES
database. The structure of the table is:
Pollutant / Process
State / County
Year
Source Use Type
Fuel Type (only gasoline fuels)
Beginning Model Year of Coverage
Ending Model Year of Coverage
InspectFreq
EVCProgramlD
I/M Test Type
I/M Test Standards
Ignore I/M toggle (user control variable)
Compliance Factor
The EVICoverage table structure shows that the EVI Compliance Factor is a function of
numerous variables that include geography, time, vehicle type / fuel / coverage factors, program
test frequency and specific I/M test / I/M test standards types.
For state SIPs, it is expected that the state will enter their own set of Compliance Factors
which reflect current and expected future program operation. The data in the default MOVES
table is likely out of date (i.e., 2005 NEI), and has not been cross referenced or updated with
recent state I/M program designs / changes.
The underlying data used to construct the default Compliance Factors were taken from
MOBILE6.2 input files used in the NMEVI model to compute the National Emission Inventory of
2005. The data files listed in Table 5-2 were extracted and processed into the various fields in
EVICoverage table.
32
-------
Table 5-2 I/M Coverage Table Data Sources
NMIM Data Source
MOBILE6 Compliance Rate
I/M Outpoints
MOBILE6 Effectiveness Rate
Grace Period
Model Year Range
Test Type
Vehicle Type
MOBILE6 Waiver Rate
MOVES I/M Coverage Parameter
Used in the MOVES Compliance Rate
Calculation
Used to determine MOVES I/M Test Standards
Used in the MOVES Compliance Rate
Calculation
Used in MOVES to Determine Beginning
Model Year of Coverage
Used in MOVES to Determine Ending Model
Year of Coverage
Used to determine MOVES I/M Test Type
Used to determine MOVES Regulatory Class
input
Used in the MOVES Compliance Rate
Calculation
As seen in Table 5-2, MOBILE6.2 and MOVES do not have exactly compatible parameter
definitions. Extraction and processing of the MOBILE6.2 inputs for all of the individual states
was required. The MOBILE6 compliance rate, waiver rate and Effectiveness rate were used to
determine the MOVES Compliance Rate. The new MOVES Compliance Rate is a broader
concept that incorporates three separate MOBILE6.2 inputs. Equation 5 shows the relationship.
C = M6 Compliance Rate x M6 Effectiveness Rate x (1 - M6 Waiver Rate) Eq 5
MOVES does not have separate inputs for the effect of waivers on I/M benefits. Section
3.10.6.2 of the document, "Technical Guidance on the Use of MOVES2010 for Emission
Inventory Preparation in State Implementation Plans and Transportation Conformity" describes
how to calculate the MOVES compliance rate to include the effect of waivers.
In MOVES, it is assumed that any repairs attempted on vehicles receiving waivers are not
effective and do not result in any reduced emissions.
Other fields in the EVICoverage table complete the description of the I/M program in effect in
each county. The MOBILE6.2 I/M Cutpoints data were used only to determine level of
stringency of a state's EVI240 program (if any). The MOBILE6.2 Test Type inputs provided a
description of the specific I/M tests performed by the state and test standards for the ASM and
Basic I/M tests. The MOBILE6.2 inputs of Grace Period and Model Year Range were used to
determine the MOVES Beginning and Ending model year data values for each I/M program. The
MOBILE6.2 Vehicle type input was mapped to the MOVES regulatory class. The Ignore I/M
toggle is a user feature that allows the user to completely disable the effects of I/M for one or
more of the parameter combinations.
33
-------
6. References
1. E. Glover and D. Brzezinski, "Exhaust Emission Temperature Correction Factors for
MOBILE6: Adjustments for Engine Start and Running LA4 Emissions for Gasoline
Vehicles," EPA Report Number EPA420 R 01 029 (M6.STE.004), April 2001.
(Available at: www. epa.gov/otaq/models/mobile6/m6tech.htm)
2. E. Glover and P. Carey, "Determination of Start Emissions as a Function of Mileage and
Soak Time for 1981-1993 Model Year Light-Duty Vehicles," EPA Report Number
EPA420-R-01-058 (M6.STE.003), November 2001.
(Available at: www. epa.gov/otaq/models/mobile6/m6tech.htm)
3. Stump, F. D., D. L. Dropkin, S. B. Tejada, C. Loomis, and C. Pack, "Characterization of
Emissions from Malfunctioning Vehicles Fueled with Oxygenated Gasoline-Ethanol (E-10)
Fuel - Part HI," US EPA's National Exposure Research Laboratory (NERL), U. S.
Environmental Protection Agency, Washington, D.C., EPA Report Number EPA/600/R-
01/053 (NTIS PB2004-106735), July 2002.
(Available at: http://www.epa.gov/nerl/nerlmtbe.htm#mtbe7c)
4. "Analysis of Particulate Matter Emissions from Light-Duty Gasoline Vehicles in Kansas
City" EPA Report Number EPA420-R-08-010, April 2008.
(Available at: http://epa.gov/otaq/emission-factors-research/420r08010.pdf)
5. "Regulatory Impact Analysis for Final Rule: Control of Hazardous Air Pollutants from
Mobile Sources" EPA Report Number EPA420-R-07-002, February 2007, Chapter 2, pages
2-15 to 2-17.
(Available at: http://www.epa.gov/otaq/regs/toxics/fr-ria-sections.htm)
6. Code of Federal Regulations 86.1342-90 (Page 309).
(Available at: http://www.access.gpo.gov/nara/cfr/cfr-table-search.html)
7. "Air Conditioning Activity Effects in MOBILE6", EPA Report Number EPA420-R-01-054
(M6.ACE.001), November 2001.
(Available at: http://www.epa.gov/otaq/models/mobile6/m6tech.htm)
8. "Air Conditioning Correction Factors in MOBILE6", EPA Report Number EPA420-R-01-
055 (M6.ACE.002), November 2001.
(Available at: http://www.epa.gov/otaq/models/mobile6/m6tech.htm)
9. "MOBILE6 Inspection / Maintenance Benefit Methodology for 1981 through 1995 Model
Year Light Vehicles", USEPA Office of Transportation and Air Quality, Assessment and
Standards Division. EPA Report Number EPA420-R-02-014 (M6.EVI.001) March 2002.
(Available at: http://www.epa.gov/otaq/models/mobile6/r02014.pdf)
34
-------
10. "Documentation for the 2005 Mobile National Emissions Inventory (NET) 2005, Version
2", December 2008, and National Mobile Inventory Model (NMEVI) County Database
(NCD) for 2005 V2.
(Available at: http://www.epa.gov/ttn/chief/net/2005inventory.html)
35
-------
Appendix A - Mean Start Emission by Temperature
Table A-l Change in Mean Start Emissions at Various Temperatures
By Model Year Group
Relative to 75° F
Model Yr
Group
Pre-81
Pre-81
Pre-81
Pre-81
Pre-81
Pre-81
Pre-81
Pre-81
Pre-81
Pre-81
Pre-81
Temp
19.75
20.67
2Z63
47.55
4978
52.52
6014
77.31
95736
98.06
105706
HC
(grams)
36.090
33.018
307560
18.569
157252
18.099
11120
0
-27122
-1.755
-4935
CO
(grams)
226.941
254.386
2767341
129.472
126:931
115.776
531517
0
-58:656
-67.555
-86T689
NOx
(grams)
-0.274
-0.925
-T:445
-0.380
-0:634
0.101
1796
0
1640
1.975
3769
Model Yr
Group
81-82
81-82
81-82
81-82
81-82
81-82
81-82
81-82
81-82
81-82
81-82
Temp
19.36
20.69
22733
49.20
5031
51.43
59715
75.73
95722
97.75
105766
HC
(grams)
21.120
23.363
257496
7.782
8:202
9.209
67432"
0
-47659
-5.450
-97958
CO
(grams)
231.180
242.806
2537865
109.851
126:239
132.360
1357663
0
-1447116
-174.532
-343:847
NOx
(grams)
-0.374
-0.252
-0135"
-0.066
0065
0.194
-17416
0
1915
1.814
4568
36
-------
APPENDIX A Continued
Table A-2 Change in Mean Start Emissions at Various Temperatures
By Model Year Group
Relative to 75° F
Model Yr
Group
83-85
83-85
83-85
83-85
83-85
83-85
83-85
83-85
83-85
83-85
83-85
Temp
19.32
21.00
2Z48
28.80
48799
50.33
51730
76.20
95:81
97.19
10579
HC
(grams)
23.299
17.755
14599
20.594
57213
5.946
67490
0
-1044
-1.209
-1124
CO
(grams)
218.857
218.151
2167439
186.549
94414
93.032
95:495
0
-29:275
-35.995
-25T407
NOx
(grams)
0.665
-0.017
-07414
-0.126
0513
0.250
0183
0
0903
0.868
-1:010
Model Yr
Group
86-89
86-89
86-89
86-89
86-89
86-89
86-89
86-89
Temp
-20
0
20
40
75
95.03
96743
106.29
HC
(grams)
27.252
25.087
14011
8.316
0
-0.127
-0139
-0.729
CO
(grams)
178.536
147.714
104604
78.525
o
-4.257
-5:354
-1.017
NOx
(grams)
-2.558
-1.360
-0749
0.312
o
-0.137
-b769i
-0.084
Model Yr
Group
1990-2005
1990-2005
_______
1990-2005
_______
Temp
-20
0
20
40
75
HC
(grams)
38.164
16.540
87154
4.872
0
CO
(grams)
143.260
92.926
567641
33.913
o
NOx
(grams)
1.201
1.227
1082
0.876
o
37
-------
Appendix B - Calculation of Specific Humidity
Equations to convert relative humidity in percent to specific humidity (or humidity ratio) in
units of grains of water per pound of dry air (ref CFR section 86.344-79, humidity calculations).
Inputs:
TF is the temperature in degrees F.
Pb is the barometric pressure.
Hrei is the relative humidity
TQ=647.27-TK
" ratio or seciichumidit ~ 4 3 4 7.0 Py I (/j, Fy
pv=\
\
100
J
db
,
(3.2437+0.00588ro+0.0000000117T03)
^=29.92*218.167*10
,
=6527.557*10
(3.2437+0.00588ro+0.0000000117ro3)
38
-------
Appendix C -Air Conditioning Analysis Vehicle Sample
Table C-l Vehicle Sample for the Air Conditioning Analysis
Model Year
1990
1990
1991
1991
1992
1992
1992
1992
1992
1993
1993
1993
1993
1994
1994
1994
1994
1995
1995
1995
1995
1995
1996
1996
1996
1996
1996
1997
1998
1998
1990
1990
1991
1991
1992
1993
1994
1994
1996
1996
1990
Make
DODGE
NISSAN
CHEVROLET
FORD
CHEVROLET
CHEVROLET
MAZDA
SATURN
TOYOTA
CHEVROLET
EAGLE
HONDA
TOYOTA
CHRYSLER
FORD
HYUNDAI
SATURN
BUICK
BUICK
FORD
SATURN
SATURN
CHEVROLET
HONDA
HONDA
PONTIAC
TOYOTA
FORD
MERCURY
TOYOTA
JEEP
PLYMOUTH
CHEVROLET
PLYMOUTH
CHEVROLET
CHEVROLET
CHEVROLET
PONTIAC
FORD
FORD
CHEVROLET
Model
DYNA
MAXIO
CAVAO
ESCO GT
CAVA
LUMI
PROT
SL
CORO
CORS
SUMMO
ACCOO
CAMRO
LHS
ESCO
ELAN
SL
CENT
REGA LIMI
ESCO
SL
SL
LUMIO
ACCO
CIVI
GRAN PRIX
CAMR
TAUR
GRAN MARQ
CAMR LE
CHER
VOYA
ASTRO
VOYA
LUMI
S10
ASTR
TRAN
EXPL
RANG
SURB
Vehicle Class
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
CAR
LDT1
LDT1
LDT1
LDT1
LDT1
LDT1
LDT1
LDT1
LDT1
LDT1
LDT2
Weight
3625
3375
2750
2625
3000
3375
2750
2625
2500
3000
2500
3250
3250
3750
2875
3000
2750
3995
3658
2849
2610
2581
3625
3500
2750
3625
3625
3650
4250
3628
3750
3375
4250
3750
3875
2875
4750
4250
4500
3750
5250
39
-------
Model Year
1991
1994
1996
1996
1996
1996
1996
1996
1997
1997
1997
1998
1999
Make
FORD
FORD
FORD
DODGE
DODGE
DODGE
DODGE
FORD
DODGE
DODGE
PONTIAC
DODGE
FORD
Model
E1500
F150
F150
DAKO PICK
D250 RAM
GRAN CARA
CARA
F150PICK
GRAN CARA
DAKOT
TRANSSPOR
CARA GRAN
WIND
Vehicle Class
LDT2
LDT2
LDT2
TRUCK
TRUCK
TRUCK
TRUCK
TRUCK
TRUCK
TRUCK
TRUCK
TRUCK
TRUCK
Weight
4000
4500
4500
4339
4715
4199
4102
4473
4318
4382
4175
4303
4500
40
-------
Appendix D - Toros Topaloglu, Comments
Peer Review of US EPA's
'Draft MOVES2009 Highway Vehicle Temperature, Humidity, Air Conditioning, and
Inspection & Maintenance Adjustments"
September 29, 2009
As part ofthe MOVES2010 Peer Review process, EPA solicited comments from Toros
Topaloglu, Ph.D., P.Eng. on the August 2009 draft of report Draft MOVES2009 Highway
Vehicle Temperature, Humidity, Air Conditioning, and Inspection & Maintenance Adjustments.
Dr. Topaloglu is an Environmental Systems Specialist at the Ministry of Transportation Ontario,
Canada.
Dr. Topaloglu's comments are copied below, with EPA response in italics.
1. Introduction
The development of MOVES and its predecessor, MOBILE, represent enormous achievements:
estimating past, present and future emissions of an infinitely diverse and variable vehicle/driver
population under highly variable and ever changing conditions. The US EPA deserves our
sincere gratitude for this unparalleled effort, which continues to deliver ever more powerful and
user-friendly emission simulators.
It has not been easy to think of a few meaningful comments on the above captioned report that
describes various adjustments employed in MOVES. I have limited myself to "constructive
criticism", assuming that this is what you expect from me and that this will be viewed in a
positive vein coming from someone who has a direct and genuine interest in making MOVES as
useful as possible. Where I am silent, I fully concur with the adopted approach and its
presentation. This happens to be the case for over 99% of the report.
In this review, I relied primarily on my personal experience and knowledge but consulted also the
relevant MOBILE6 documentation and a few specific publications listed under Section 5
(references).
2. General Comments
2.1. I agree with the empirical/statistical approach adopted in the derivation of the
adjustments - given the imprecise nature of cars and the near infinite variability in their
population. Some scientist and engineers may feel more comfortable with relations that
have a theoretical basis; however, even with the "best" data "the multitude of
41
-------
mechanisms involved in each adjustment make a mechanistic approach very difficult to
implement.
2.2. Adjustments for greenhouse gas emission factors may not have been uniformly
addressed. The vehicle emissions certification process does not automatically yield
adjustments for CC>2, CH4 and N2O emissions. Given the urgency to address Climate
Change, MOVES will be called upon frequently to derive more accurate GHG emission
factors.
2.3. Adjustments for individual air toxic emission factors may not have been fully addressed.
It is conceivable that adjustments for NMHC may not apply equally to each and every air
toxic, since they are not formed by identical physical and chemical mechanisms.
2.4. It is not clear that the US EPA adjustments deal fully with up-and-coming technologies
such as hybrid, plug-in hybrid and battery-powered electric vehicles. Emissions of these
vehicles, where they exist, are less sensitive to variations in ambient conditions, air-
conditioning (A/C), and inspection and maintenance (I/M). In fact, they are generally
exempt from I/M. The number of electric hybrids in the US fleet exceeds one million
already and is expanding rapidly. Hence, it will become progressively more important to
account for their characteristics.
2.5. In future updates of MOVES, it may be worthwhile to try and correlate adjustments with
major vehicle technologies and fuel types - beyond what is in place. This may improve
the ability of the model to simulate future emissions. Vehicle manufacturers often have
this information and might share it with the US EPA.
2.6. The accuracy of the adjustments depends, in part, on the representativeness of the test
vehicle sample. It is obvious that the US EPA has spent enormous effort to achieve a
high degree of representativeness. However, limitations with the test data and, to a
lesser extent, unexpected but deliberate efforts to alter the vehicle population such as the
recent "cash for clunkers" program of the US government may have somewhat thwarted
this effort. Given these factors, it is rather difficult for a regular MOVES user to judge
the adequacy of the proposed adjustments.
3. Specific Comments
3.1 Ambient Temperature
3.1.1. I concur with the observation that the principal influence of ambient temperature
(Tamb) on emissions is during the warm-up phase of cold-starts. Its influence on
warmed-up vehicle running emissions is relatively small - albeit not nil,
particularly, under extremely cold conditions when steady-state temperatures of
vehicle components (lubricants, tires, etc.) may stay below their "normal" values.
3.1.2. It is not certain that the difference in emissions between Bags 1 and 3 of the FTP
cycle can fully account for cold-start emissions under extremely cold conditions
42
-------
(below 32°F) when it takes extended periods of time to reach stead-state. Hence,
adjustments based on these data will probably result in underestimates.
EPA has seen increased emphasis by manufacturers on decreasing the amount of time it
takes to light off the catalytic convenor in order to address tighter emission standards,
and since the 1990 's vehicles have had to meet emission standards even at low
temperatures. Once the catalytic convenor is fully operational, any small effects from
the ambient temperature on emission formation in the engine are easily lost in the
catalyst. EPA believes that any temperature effects not captured in the first bag (505
seconds) of the FTP are negligible and existing data bears that out.
3.1.3. The decision to neglect adjustment for ambient temperatures above 75°F is a
reasonable but not a perfect one. Reference (1) provides some evidence for less fuel
consumption and CC>2 emissions at higher ambient temperatures - perhaps due to
less throttling (higher volumetric efficiency). Other emissions are probably also
affected, but the test data do not seem to allow for these smaller effects - as noted on
page 7 of the report, in the discussion of the Tamb adjustment for NOX emissions.
3.1.4. Part of the difficulty with adjusting for Tamb in the general fleet may be due to the
many vehicle parking options: outdoors, unheated indoors, heated indoors or with a
plugged-in block heater. If a vehicle is parked outdoors, the wind chill factor might
also influence cold-start emissions. The test data do not seem to account for all of
these factors.
The temperature adjustments in MOVES are intended to represent the effects on vehicle
emissions when the ambient temperature to which the vehicle is subjected is known.
There may be factors that cause difficulty in determining the appropriate temperature to
apply to the fleet, such as the variation of ambient temperature over the area you wish to
model. However, these are issues for guidance on how best to use the model for specific
scenarios.
3.2. Humidity
3.2.1. One would expect a weak dependence of carbon dioxide emissions on ambient
humidity, as reported forNOx. The EPA certification humidity adjustments
should, however, account for this effect.
3.2.2. I expect that the EPA certification humidity adjustments are sufficient for
inventory work.
3.3. Air Conditioning
3.3.1. The US EPA report indicates that all emission tests with the A/C on were carried
out at 95°F only. This implies that the A/C adjustments are not based on emission
data obtained at the same temperature, with A/C on and A/C off. If so, according to
Reference 1, a significant "error" may have been incurred by not accounting for the
co-existing effect of Tamb on some emissions such as those of CO2.
43
-------
On the contrary, because the MOVES model does not apply temperature adjustments to
running emissions (except for particulate matter), the comparison ofA/C on emissions at
high temperature andA/C off emissions at low temperature allows the A/C correction to
incorporate any necessary temperature effect. However, as discussed in Section 2.3.2.
our data suggests such effects are not significant.
3.3.2. A/Cs are employed for defogging at all temperatures - particularly, at lower
temperatures. This secondary use of the A/C along with associated emission effects
do not seem to have been accounted for (according to Ref 1, defogging costs a 1.5 -
7% in CC>2 emissions at 55°F - depending on driving cycle).
MOVES does not account for the A/C effects at low temperatures from the use of A/C for
defogging. The text has been updated to describe this omission.
3.3.3. Many modern vehicles are equipped with climate control systems, which are
usually set by drivers to maintain automatically a preset optimum compartment
temperature. The A/C systems of these vehicles switch on when this temperature
set-point is exceeded - irrespective of humidity (as is the case with house
thermostats). The conditioned air is cold and dry (often reheated with engine
coolant). Hence, in these modern vehicles the compressor usage may be largely
independent of humidity. The compressor load and hence energy use and some
emissions are however very dependent on ambient air humidity.
Modeling the behavior of modern A/C systems can be very tricky. As a first cut, MOVES
simply addresses the need for A/C based on how comfortable humans will be based on
the combination of temperature and humidity. This should adequately capture the need
for A/C and the extra loads that result for inventory estimates. A better A/C load model
may be developed as our understanding of these systems improves.
3.4. Inspection and Maintenance
3.4.1. The repeated application of MOBILE 6 to predict the relative emission
consequences of various I/M program design features appears to involve certain
assumptions; viz., all vehicles at a given age have the same odometer reading, are
subject to the same deterioration rates, and, if repaired, experience the same
emission improvements. It may be worthwhile to test the benefits of replacing these
point assumptions with appropriate distributions or variables.
Even with the use of distributions, the average impact of 1/M programs on fleet emissions
would be the same. We believe the added complexity of using distributions would only
add to the complexity of our already complex modeling and provide very limited insight
into the benefits of I/M repairs.
3.4.2. Future failure rates will likely be smaller than current ones - largely due to
incremental improvements in vehicle technology but also due to the observed shift
to inherently low emission vehicles.
44
-------
All modeling of future model years is fraught with uncertainty. EPA has taken the
position that improvements in emission performance will only occur if there is an
incentive to improve, such as new emission standards. As such, it is reasonable that the
existing failure rate, which is already very small, will continue into the future unless
there is some regulatory reason why manufacturers would take the time and money to
develop solutions that would significantly reduce their failure rates. Even without
reductions in failure rates, the benefits of I/M programs will decrease as the emission
impact of failure grows smaller on vehicles with new (lower) emission standards.
3.4.3. MOBILE 6 apparently assumed that waived vehicle emission rates are invariably
20% lower than those of failed vehicle emissions (see Ref, 2). Is this assumption
carried through in MOVES? If so, it may be worthwhile to re-examine it.
In MOVES, vehicles which receive waivers are assumed not to have been repaired at all.
The text has been updated to include this information. Waived vehicles are typically a
small fraction of the fleet and are difficult to study. Given the limited impact that these
vehicles will have on total fleet emissions, determining a more precise impact from
waiver vehicles will not be a high priority.
3.4.4. The National Research Council (Ref. 2) raised a number of additional I/M related
concerns with MOBILE 6: (a) assumption that vehicles with and without I/M
deteriorate at the same rate; (b) no explicit allowance for those vehicles that are
repaired before or after inspection but rapidly revert to high-emitter status; and (c)
no I/M credit for high emitters that are scrapped or shipped outside of the region. It
would be helpful to explain how these concerns were addressed in MOVES.
The description of the I/M program effects in the report has been revised to more
explicitly address the concerns of the National Research Council.
3.4.5. Another concern in the I/M community, namely the effectiveness of OBD systems
and OBD based I/M programs, deserves also a fuller discussion.
4. Editorial Comments
EPA has made many changes to the text of the report to address the following editorial
comments.
4.1. The term "adjustment", as used in the title of the report, expresses the goal of the effort
clearly and concisely. The terms "correction factor" and "adjustment factor", as used in
the body of the report, are less clear. First, a correction is not an adjustment. Second,
the word "factor" implies a multiplication whereas most of the proposed adjustments are
additive. I recommend that the report stick to the term "adjustment" throughout the
report.
4.2. This is not a free-standing report. Its contents cannot be fully understood without
referring to a series of other reports (at least, the documentation of MOBILE 6). It would
45
-------
have been preferable to have a free-standing report - not for the sake of peer reviewers
but for younger MOVES users who haven't witnessed the evolution of MOBILE.
4.3. A Table presenting the principal assumptions made and the resulting effectiveness
estimates for major I/M program types would add to the value of the report.
4.4. I am assuming that the final report will include lists of acronyms (with explanations),
tables, figures as well as equation numbers, etc. - all the usual pieces that make a report a
bit more accessible.
4.5. Minor notes:
• Last sentence on page 8 refers to Section 4.1.3, which does not exist.
• Section 4.1 apologizes for lack of data with A/C on MC. Do you really mean
motorcycles?
• Section 4.1, third paragraph, refers to Appendix A for a list of vehicles and their
description. It should instead refer to Table 4-1. Also, I don't see any description of
the vehicles.
• The title of Section 4.3.2 reads "Energy Emissions". It should probably read "Energy
Consumption".
• On page 30, "OBD" is spelled "OBC".
5. Response to Specific Questions Posed in the US EPA Letter to Me
5.1 The Clarity of the Presentation
• The report is well written and very clear to individuals with a technical background in
the subject area. It may however require some editing to make it more easily
accessible to a wider audience - if this were necessary.
5.2 The Integration of Information from Multiple Areas
• The report is based on an enormous volume of previous work and the resulting
information. Given the difficulty of condensing this vast volume of information into
a relatively compact report, the author(s) have done very well. The information is
well integrated. However, as noted in my general comments (Section 1 above), the
report is not a stand-alone document. It cannot be fully understood without reading
its references.
5.3 The Appropriateness and Completeness of the Literature Discussed
• The literature referenced in the report is highly appropriate and sufficiently complete.
5.4 Appropriateness of the Resulting Adjustments
46
-------
• In spite of the inherent complexity of the subject and the limitations of the available
data, the author(s) have succeeded in:
o Identifying those effects that call for adjustments
o Eliminating those effects (variables) that are too insignificant to adjust for
o Deriving robust adjustments that reflect the totality of the empirical evidence
available and also conform to theory
• In my opinion, the adjustments are highly appropriate. The few comments provided
in this review are intended to contribute to any future effort to update MOVES and
make it as useful as possible to all potential users.
6. References
(1) Weilenmann, M.F.; Vasic A-M; Stettler P.; and Novak, P. Influence of Mobile Air-
Conditioning on Vehicle Emissions and Fuel Consumption: A Model Approach for Modern
Gasoline Cars Used in Europe. Environ. Sci. Technol. 2005, 39, 9601-9610.
(2) National Research Council. Evaluating Vehicle Emissions Inspection and Maintenance
Programs. The National Academies Press, Washington, DC. 2001.
(3) Eisinger D.S. and Wathern, P. Policy Evolution and Clean Air: The Case of US Motor
Vehicle Inspection and Maintenance. Transportation Research Part D. 2008, 13, 359-368.
Toros Topaloglu, Ph.D., P.Eng.
Thank you.
47
-------
Appendix E - ENVIRON International Corporation, Comments
ENVIRON Review of EPA Draft Report:
"MOVES Temperature, Humidity, Air Conditioner, and Inspection and Maintenance
Adjustments"
As part of the MOVES2010 Peer Review process, EPA solicited comments from Christian E.
Lindhjem of ENVIRON International Corporation on the August 2009 draft of report Draft
MOVES2009 Highway Vehicle Temperature, Humidity, Air Conditioning, and Inspection &
Maintenance Adjustments.
Chris Lindhjem has a PhD. in Chemical Engineering from Rensselaer Polytechnic Institute and
has more than 15 years of experience in automotive issues with particular focus on emissions
from highway and non-road vehicles, engines, and engine fuels.
Dr. Lindhjem's comments are copied below, with EPA response in italics.
Christian Lindhjem
ENVIRON International Corporation
773 San Marin Drive, Suite 2115
Novato, California 94998
415.899.0700
30 September 2009
Introduction
This report appears to gather all of the adjustments to the MOVES basic emission rates into one
document despite the seeming unrelated topics discussed. Temperature and humidity are ambient
conditions that affect the engine and after-treatment control effectiveness. Air condition loads are
influenced by ambient conditions, but in fact are only one of many potential loads. New engine
standards and inspection and maintenance programs are primarily emission reduction program
credit assessments.
Yet despite the varied types of adjustments, it is reasonable to include all adjustments if indeed
all adjustments to the model are included in this document. However, as new data or new
approaches are required, updating the document could be more difficult or confusing to the
reader because the document addresses so many issues. And if this document does not include all
such adjustments, it might be confusing to understand all such adjustments split over many
documents.
48
-------
Gasoline Vehicle Temperature Adjustments
To the extent that temperature adjustments presented here differ from the official test procedures,
EPA should make sure that the original data is free of temperature adjustments that the original
researchers might have used on the reported data. Often researchers will follow the official test
procedures to the letter and adjust the reported values to account for unique test conditions.
Overall the method for temperature adjustments seem sound with a few comments noted here
that might help the explanation for the reader. EPA correctly separated the start and running
temperature adjustments to account for the likely different effects when the engine and after-
treatment are at operating temperature.
For gasoline vehicle start emissions, hydrocarbon and carbon monoxide effects are presented
sufficiently to understand the results. The NOx results were considerably more complex, and
there may be reasons for the observed start NOx emission with regard to technology by model
year grouping. However, given that the effect is more predominant with older model years, the
NOx effect may be less important for most uses of MOVES. For all cases, it would be helpful to
put the adjustment estimates in perspective of the base emission rates to demonstrate the relative
importance of the temperature adjustment effect, such as inclusion of a description of the
percentage effect.
For gasoline vehicle running emissions, I agree with the assessment that ambient temperature has
little effect on emissions. It might be helpful to note several same-vehicle tests at different
ambient temperatures in Figure 2-2, such as by symbol and/or lines, to demonstrate that a
temperature effect is not observed with the same equipment. Mixing vehicle tests and
temperature conditions tests may mask an effect that could be observed in same vehicle tests.
Diesel Vehicle Temperature Adjustments
The diesel vehicle start emissions are presented, and I have no dispute with the results for start
emissions determined. But this discussion could use some context in terms of vehicle types and
applicability. For instance, based on the use of the FTP test cycles to determine the start
emissions, I suspect that these 12 vehicles were pickup trucks, light heavy-duty vehicles, so EPA
should discuss the relevance of using these vehicles to represent all diesel vehicles.
The text of this document describing the temperature adjustments to diesel start emissions has
been updated to better address the types of vehicles in the samples used.
EPA makes no claim about particulate matter or running emissions temperature adjustments for
diesel vehicles, so the report approach is inconsistent to that for gasoline vehicles. In addition,
there was no discussion of whether these vehicles used after-treatment devices (either diesel
particulate filters (DPF) or oxidation catalysts (OC) or future systems expected for 2010 model
years and beyond) and how that might affect the results as was done to incorporate the cold
weather CO and HC requirements for gasoline vehicle temperature adjustments.
49
-------
Less is known about the effects of temperatures on diesel particulate matter emissions. The text
of this document describing the temperature adjustments to diesel particulate matter emissions
has been updated to better address the technologies in the samples used.
Cold Weather CO and HC Requirements
The methodology to estimate the benefits credited to the light-duty cold weather regulations
appears to be a reasonable approach as presented. However, it is questionable if the cold weather
regulation adjustments should be applied to high emitters given that the control systems might
not be functioning. To the extent that MOVES identifies high emitters, independent temperature
adjustments should be applied to high emitters.
Since MOVES does not identify high emitting vehicles during calculations, independent
temperature adjustments for high emitters cannot be applied.
Humidity
Without performing additional testing, it is reasonable to use the Federal Register humidity
corrections. Because these adjustments would be multiplicative, they would be applicable to the
lower emission rates of later model years.
Air Conditioning
The air conditioning adjustment approach appears to be counterintuitive to approach of MOVES
defining power bins to reflect the engine loads. There may be some reasons for this approach
given that idle and coasting\braking bins would not otherwise incorporate the auxiliary air
conditioner loads. Another reason could be that air conditioner loads would oscillate between
VSP bins when the compressor is engaged and disengaged unrelated to the driving demands.
The approach presented is easy to follow in concept, but there should be more description of the
overall air conditioning effect for sample vehicle types. To help the reader understand how
important the air conditioning adjustment is, EPA should plot of the effect with respect to the
humidity index, noting the heat index below which there is no air conditioning adjustment. The
"ActivityTerm" coefficients for the ' ACOnFraction' estimates should be presented in the
document.
The text of this document describing the air conditioning adjustments has been updated to better
display the effects of the activity adjustments versus the humidity index.
Inspection and Maintenance (I/M)
Using the I/M benefits from the MOBILE6 analysis is a reasonable approach without an
extensive reanalysis of the benefits under the MOVES modeling framework. As with the
assessment of new vehicle emission standards, the emission credits estimated for various
programs may not be entirely based on a quantitative assessment of each program. Therefore,
because the credits assigned have been well vetted under the MOBILE6 plan, it becomes a more
accepted approach to use for MOVES as well.
50
-------
Because the MOBILE6.2 benefits only include HC, CO, and NOx and the Figure 2-6 was given
as evidence of a relationship between PM and HC emission, PM benefits for I/M programs
should also be considered. It would stand to reason, even without direct evidence, that emission
reductions of the primary pollutions evaluation would also lead to PM emission reductions when
malfunctioning vehicles are repaired.
EPA does not yet have sufficient data to estimate PM emission reductions for I/M programs
without further evidence that repairs that reduce HC, CO and NOx emissions will significantly
affect PM emissions.
Errata
Numerous changes to the text have been made to address these minor edits.
Page 12 above Figure 2-4; "adjustments" has an extra "s"
Page 25 below Table ?4-5? (label missing), just above section 4.3.3, "If and when ..."
Page 27 above Section 4.4: "A/C correction factors of less than unity or unity were found for.
Some Tables have headings and some table headings are missing and references for those tables
in the text are not clear.
Section 5, Eq. 1, 2, 3 and variable description of nonEVI emissions should read the same, such as
"EnonEVI" in all equations and descriptions.
Equation 6 or should it be Equation 5? label on last page is incomplete
51
-------
Appendix F - Julio Vassallo, Comments
Review of US EPA's
"Draft MOVES2009 Highway Vehicle Temperature, Humidity, Air Conditioning, and
Inspection & Maintenance Adjustments"
September 25, 2009
Additional comments, not part ofthe formal MOVES2010 Peer Review process, were submitted
by Julio E. Vassallo on the August 2009 draft of report Draft MOVES2009 Highway Vehicle
Temperature, Humidity, Air Conditioning, and Inspection & Maintenance Adjustments.
Julio Vassallo is a Chemical Engineer and the Technical Manager of Area new vehicles
Approval and Certification in the the Laboratory of Vehicle Gaseous Emission Control (LCEGV)
of the Ministry of Environment and Sustainable Development (SAyDS) in Buenos Aires,
Argentina.
Julio Vassallo's comments are copied below, with EPA response in italics.
Page 6: The behavior presented for vehicles without (or deactivated) catalyst (that might be
included in the group pre 1981) is different with respect to the NOx emission, those with catalyst.
The vehicle can be considered as two reactors in series, combustion reactor in homogeneous
phase (cycle Otto engine) and oxidation-reduction catalytic reactor (catalytic converter). The
generation of NOx in the engine is principally a function of temperature in the cylinder and the
partial pressures of nitrogen and oxygen. Therefore when the engine is cold the issue (without
catalyst) is the lowest and increases as the engine warms up (example in doc "Start Emission"
vehicles without catalytic converters, emissions EVI240 consecutive test series). In contrast to an
engine with catalyst but also emissions start to increase with increasing temperature once you
reached the temperature catalyst "light off" decreases again (example in doc "Cold Emission"
vehicles with catalyst) Moreover, the emission of NOx is also heavily dependent on power (VSP)
Then, depending on which portion of the emission is correlated is provable that the temperature
hasn't the same effect (function) for vehicles with catalyst and without catalyst.
I think that the start NOx emission (FTP NOx emission Bag3 minus Bagl) of the vehicles
without catalyst are highest than those with catalyst and has different start temperature
dependence.
With CO an HC start emission is different, because both reactors (engine and catalyst) the
emission decreases with temperature (example in doc "Start Emission" vehicles with and without
catalytic converters, emissions EVI240 consecutive test series).
The behavior of catalyst and non-catalyst vehicles is handled in MOVES by having separate
temperature adjustments by model year group.
52
-------
Page 7:1 think that is provably if you correlate taking account VSP and the vehicles with catalyst
in other group that those without catalyst the R-square coefficient will be better.
In MOVES, since the temperature adjustments are grouped by model year, some model year
groupings will include catalyst and some non-catalyst vehicles in the correlations. Unless
MOVES is redesigned to accommodate separate technologies in addition to model years, a
separate correlation for each technology is not possible.
Page 8: I agree that to reach working temperature (running) both the emission generated in the
engine and removed by the catalyst that should not be so sensitive to the test temperature such as
the start
While the supply air temperature should influence the reactions of improving combustion
efficiency at higher temperatures of income, is provable that the high working temperatures of
engine determine less sensibility for that purpose, on the other hand those vehicles with catalyst
in the regime temperature will have to be less sensibly since over 90% of the pollutants are
converted and that masks any engine inefficiency specially to low exhaust flow (low rpm / VSP).
Page 11: The analysis of diesel engine emissions are different from that of Otto cycle, in the case
of CO are not as significant and therefore may be more affected by the measurement uncertainty
when it comes to a small population, such as that of the reporter. For NOx, in this case normally
pre 2007 alone technologies are oxidation catalysts (remove only CO and HC) therefore in this
case has only effect the engine and NOx emissions should increase, ie emissions Bag 1 Bag
minus 3 should be negative. For example the mean value obtained to 34.6 ° F will have to be
negative -2.6? I haven't studies with a diesel emission test series to different ambient temperature
in the start, but you have studies for example that about humidity and temperature effects how I
adjunct (in page 7 HUMIDITY AND TEMPERATURE CORRECTION FACTORS FOR NOx
EMISSIONS FROM DIESEL ENGINES SwRI Project No. 03.30.10.06599).
Page 13: as you get this value? This increasing of cold start emission (value 0,5611592) will be
in grams per mile?. My doubt is because I think that if you have a total increase of 2.086 g
NMHC = 0,43 (M Bagl+MBag2) + 0,57 (MBag2 + MBag3); then the value in grams of the cold
start (M NMHC Bag 1) should be higher than 0,5611592.
The effects of the MSATrule on the cold temperature adjustment for HC emissions of engine
starts will need to be revisited once vehicles compliant with these standards are available for
testing. The current adjustment is based solely on the emission standard values.
53
-------
Appendix G- Coordinating Research Council Project E-68a
Comments
December 3, 2009
Additional comments regarding the adjustments described in the report, "DraftMOVES2009
Highway Vehicle Temperature, Humidity, Air Conditioning, and Inspection & Maintenance
Adjustments, that were not part of the formal MOVES2010 Peer Review process, were submitted
as part of the Coordinating Research Council (CRC) Project E-68a.
Comments from the report that are relevant to the topics covered in the EPA report are copied
below, with EPA response in italics. Readers are encouraged to obtain the entire CRC Project
E-68a report in order to fully understand the comments in their full context.
Correction Factors (Fuels and Temperature)
Regarding temperature correction factors, EPA examined recent data and found that cold start
HC, CO, and NOx emissions should be adjusted for temperature, but there is no ambient
temperature effect on running exhaust emissions. EPA developed additive cold start increments
for HC, CO, and NOx that increase with lower temperatures.
One concern with the temperature increments is that there is no analysis of how these may
change as vehicles age, and the available data seemed to omit the CRC E-74b testing program,
which was completed in May of 2009. EPA could utilize the Kansas City temperature data to
determine whether the temperature relationships change with vehicle age. Also, the CRC E-74b
testing program data could be used to further check the MOVES cold start correction factors.
EPA believes that studies, such as the Kansas City study and CRC E-74, which include vehicles
of different ages, but do not follow the vehicle fleet over time, are inadequate to conclude that the
effects of temperature vary by vehicle age. EPA in cooperation with others, is planning a study
specifically designed to follow the vehicle fleet over time and should produce the type of
information needed to determine the effects of vehicle age on temperature effects.
A second concern is that the method used to develop HC temperature increments for the MSAT
rule (which requires lower HC standards at cold temperatures) assumes a compliance margin
with respect to the HC standards at 75° F, but no compliance margin with respect to the HC
standards at 20° F. As a result, the HC increments for vehicles meeting the MSAT requirements
are over-estimated. The method should be revised to include a compliance margin at 20° F to be
consistent with the margin currently being utilized at 75° F.
EPA believes that any compliance margin at 20 degrees Fahrenheit will likely differ significantly
from the margin observed at 75 degrees. Further testing will be needed on vehicles compliant
with the MSAT standards to determine the appropriate margin.
54
-------
A third concern is that vehicles subject to the lower MS AT HC standards will very likely have
much lower CO emissions as well. Once vehicles are certified to the MSAT cold HC standards,
an analysis should be conducted of certification or other data to determine how much the CO
increments change for these vehicles as well.
CO emission rates already assume the impact of explicit standards for CO emissions at low
temperatures. EPA believes that strategies to reduce HC emissions at low temperatures will
likely have minimal further impacts on CO emissions at low temperatures.
Particulate Matter Emissions for Gasoline Vehicles
Temperature correction factors were estimated from the matched vehicle pairs. Unlike HC, CO,
and NOx emissions, where the temperature correction factors were only for cold start emissions,
EPA found an increase in running PM emissions with decreasing ambient temperatures, albeit
lower than for the cold start.
The first concern is that the combined MSAT and Kansas City data on matched pairs does not
appear to support a cold temperature adjustment for running emissions. Results from other
studies such as NFRAQS should be included in the analysis, with special regard to high PM
emitters.
It is true that data from the matched pairs in the combined MSAT and Kansas City was
inconclusive in determine the temperature effect on running emissions. However, using other
analysis techniques, EPA was satisfied that a significant temperature effect could be determined.
A third concern is that in the draft model, vehicles meeting lower HC standards in response to the
MSAT rule currently are not assumed to have lower PM emissions. Since HC and PM emissions
seem to correlate well, we believe there will be lower PM emissions with a lower HC standard at
cold temperature. In the section on correction factors, we recommend evaluating certification
data or other data to examine the effect of cold HC standards on HC and CO emissions. This
should be extended to PM as well if possible.
EPA has not been able to establish a clear correlation between HC emissions andPM
measurements that would support assuming that PM emissions at low temperatures would be
significantly affected by changes in the HC standard. EPA will be updating the emission
estimates in future versions of MOVES as new data on vehicles certified to the new standards are
tested.
Summary of Recommendations
EPA should utilize the Kansas City data to determine whether temperature correction factors
change with vehicle age. Also, the CRC E-74b testing program data could be used to further
check the MOVES cold start correction factors.
As stated above, we believe the Kansas City data is inadequate for this purpose, but we hope to
collect appropriate data to do this analysis in the future..
55
-------
The Tier 2 cold temperature response should be lower than for Tier 1 vehicles. In addition, the
MSAT rules should reduce CO emissions as well as HC emissions.
The method used to develop HC temperature correction factors for the MSAT rule should be
revised to include margin at 20° F to be consistent with the margin currently being utilized at 75°
F.
We don't believe these changes are justified based on currently available data. Now that MSAT
vehicles are entering the fleet, we hope to gather in-use data on vehicles meeting these
standards.
The combined MSAT and Kansas City data on matched pairs does not support a cold
temperature adjustment for running emissions. Results from other studies such as NFRAQS
should be included in the analysis, with special regard to high PM emitters.
56
------- |