EPA-AA-SDSB-85-03
Technical Report
Motor Vehicle NOx Inventories
By
Amy Brochu and Dale Rothman
November 1984
Standards Development and Support Branch
Emission Control Technology Division
Office of Mobile Sources
Office of Air and Radiation
U. S. Environmental Protection Agency
NOTICE
Technical Reports do not necessarily represent final EPA
decisions or positions. They are intended to present
technical analysis of issues using data which are
currently available. The purpose in the release of such
reports is to facilitate the exchange of technical
information and to inform the public of technical
developments which may form the basis for a final EPA
decision, position or regulatory action.
-------
Table of Contents
Page
Background 1
NEDS Approach . " 3
Nationwide Approach 4
Localized Approach 6
Comparison of Baseline Inventories 10
Comparison of Future NOx Inventories 14
Conclusions 23
References 24
APPENDIX
-------
Motor Vehicle NOx Inventories
Background
In support of the Notice of Proposed Rulemaking (NPRM) on
Light-duty Truck/Heavy-duty Engine NOx and Particulate
Standards, published on October 15, 1984, (49 FR 40258) air
quality projections were made using a model based on the
"rollback" theory. This rollback model uses the change in
emissions inventory between two given years to predict the
accompanying change in air quality. The future emissions used
in this model are estimated by applying travel growth rates and
emission factor ratios to a base-year emissions inventory.[1]
Since the mid-1970's, EPA's NOx projections have been made
using a baseline NOx inventory taken from the National
Emissions Data System (NEDS), compiled by the Office of Air
Quality and Planning Standards (OAQPS).[2] The motor vehicle
portion of this base-year NEDS inventory is developed using
estimates of annual vehicle miles travelled (VMT) within
individual counties across the U.S.
Because NEDS does not include estimates of diesel
particulate emissions, the NPRM particulate modelling used a
different approach. This approach differs from NEDS in that no
county or city-specific figures are determined; instead,
emissions are modelled on a nationwide urban basis, using
annual VMT estimates from Energy and Environmental Analysis,
Inc.(EEA).[3]
In an effort to confirm the NOx and particulate modelling
contained in the NPRM, the motor vehicle portion of the
base-year (1981) NEDS NOx inventory was compared to an
analogous NOx inventory developed using the nationwide urban
approach. This comparison revealed that the two approaches
differed significantly in their predicted breakdown of
nationwide urban VMT by vehicle class, particularly with
respect to the heavy-duty diesel vehicle (HDDV) fraction of
travel. As shown in later sections, NEDS attributes a VMT
fraction to HDDVs that is over twice that calculated with the
nationwide method. Because the HDDV class has the largest
projected VMT growth and has the largest per mile emission
factor, this two-fold difference has a significant impact on
future NOx inventory projections.
The inconsistency between the NEDS and nationwide
approaches led to a search for local VMT data for the 11 cities
being modelled for NOx emissions (listed in Table 1). Local
and state agencies were contacted, and total 1981 VMT figures
were obtained for all 11 cities; and, most importantly,
breakdowns of local VMT by vehicle class were obtained for a
majority of the cities.
-------
Table 1
Eleven Urban Areas Currently Being Modelled for NOx*
Low Altitude
Boston, MA
Chicago, IL
Nashville, TN
New York City, NY
Newark, NJ
Philadelphia, PA
Seattle, WA
Tucson, AZ
Washington, D.C.
High Altitude
Denver, CO
Reno, NV
These 11 cities were within 20 percent of the NOx NAAQS of
0.053 PPM in 1981.
-------
-3-
This report briefly describes these three approaches to
NOx inventory development (NEDS, nationwide, and localized) and
presents comparisons of the findings under each with respect to
both baseline and future NOx inventories. Detailed
descriptions of the NEDS and nationwide methodologies appear in
the Appendix, along with the various sources of input data used
in each approach. In addition, the Appendix outlines the
methods used by each individual city to determine local VMT
characteristics, and presents the results for each locality.
NEDS Approach
As mentioned earlier, the NEDS emissions inventory is
constructed on a county-specific basis; therefore, its primary
task is to estimate VMT by county for each vehicle class. This
is generally accomplished in three major steps: 1) nationwide
VMT figures from the Federal Highway Administration (FHwA) are
split between the four pertinent vehicle classes, 2)
class-specific nationwide VMTs are apportioned to each state,
then 3) statewide VMTs (by class) are distributed to individual
counties. (Detailed equations outlining the precise NEDS
methodology appear in the Appendix.)
The first major step begins with FHwA's nationwide VMT
figures.[4] Passenger car, or light-duty vehicle (LDV), VMT is
reported as a separate category by FHwA; therefore further
manipulation is not necessary. However, the remaining portion
of FHwA's nationwide VMT total (trucks plus buses) must be
divided between the light-duty truck (LOT), heavy-duty gasoline
(HDGV) and HDDV classes for use in emissions modelling. To do
this, a VMT breakdown (by the three truck classes) is first
calculated using model-year sales and scrappage rates from
MVMA[5], and mileage-accumulation rates from EPA's MOBILE
computer program[6]. This breakdown is then applied to the
FHwA figure for total truck/bus VMT to yield nationwide annual
VMT for each of the three pertinent truck types.
The second step involves the distribution of the
class-specific nationwide VMT figures (calculated above)
between individual states based on FHwA's state fuel
consumption totals.[4] Nationwide HDDV VMT is apportioned
using relative state diesel fuel consumption figures; VMTs for
the remaining classes (LDV, LOT, and HDGV) are basically
divided between states based on total gasoline consumption.
Finally, these statewide VMT totals for each of the four
vehicle classes are broken down by county. This is done
separately for each class, using each county's fraction of
state vehicle registrations (from R. L. Polk).[7]
-------
-4-
Several of the assumptions used in the NEDS procedure lead
to some possible weaknesses in the percentage breakdown of
county-specific VMT (and thus emissions) by vehicle class.
Probably the strongest weakness is the use of HDDV
registrations by county as a surrogate for HDDV travel in that
particular county. Because the majority of HDDV travel can be
characterized as long-range, there is no certain relationship
between registrations and usage. For example, HDDVs appear to
be usually registered in urban areas, while their actual travel
consists mainly of trips between these urban areas. On the
other hand, the use of county registrations to apportion county
VMT for the remaining classes (LDV, LOT, and HDGV) is probably
a reasonable approach, since these vehicles are primarily used
in local and short-range applications. Thus, as will be seen
later, the overall impact of the use of county registrations as
a surrogate for county VMT appears to be an overestimation of
the HDDV fraction of VMT in urban areas.*
As mentioned earlier, details of the NEDS methodology
(including equations) are provided in the Appendix, along with
a list of the sources used in the NEDS program. Also, because
NEDS is based upon FHwA's figures for nationwide VMT and state
fuel consumption, a description of FHwA's Highway Performance
Monitoring System (HPMS) and details on the construction of
FHwA's Table VM-1 (in Highway Statistics[4]) are provided in
the Appendix, as well.
Nationwide Approach
Unlike NEDS, the approach taken in the NPRM diesel
particulate analysis[8] examines the nation as a whole and
applies the results equally to all urban areas; no
county-specific distributions of VMT are derived. Annual
nationwide urban emissions for any given year are estimated
simply by multiplying calculated nationwide urban VMT figures
by the appropriate emission factors.
The nationwide approach is based upon the annual VMT
figures estimated by the Energy and Environmental Analysis
(EEA) fuel consumption model.[3] The VMT figures are already
divided between the appropriate vehicle classes; however,
separation into gasoline and diesel portions, and conversion to
urban VMT, is necessary before the urban emission inventories
This problem with overestimation of the HDDV fraction of
VMT appears only at the county level; on a statewide or
nationwide basis, the HDDV fraction would be expected to
be much more accurate.
-------
-5-
can be calculated. (Details of the methodologies and sources
used in these calculations are provided in the Appendix, along
with a brief description of EEA's estimation of total
nationwide VMT figures.)
The method used to separate EEA's nationwide LDV and LOT
VMTs into gasoline and diesel fractions involves much of the
fleet characterization data developed for MOBILES.[9] The
diesel VMT portion is estimated using a 20-model-year summation
of the registration fraction distribution (by age), mileage
accumulation rates for diesel vehicles (by age), and diesel
sales fractions (by model year). This diesel summation is then
divided by the total of the diesel and analogous gas summations
to yield the diesel fraction of total VMT for each class.
Finally, because the emphasis is on emissions in urbanized
areas, it is necessary to convert these nationwide VMT figures
into urban miles. This is accomplished using FHwA's estimates
of the urban fractions of LDV and LOT VMT (0.59 and 0.49*,
respectively). Because the fractions are assumed to be
constant over all model years, they can simply be multiplied by
the nationwide VMT totals to yield urban VMT for each class.
The approach used to calculate the gasoline and diesel
fractions of heavy-duty truck VMT is just slightly different
from the LDV/LDT method; instead of using model-year diesel
sales fractions, EEA's total calendar-year registrations of
HDDVs and HDGVs are substituted. As before, MOBILES mileage
accumulation rates and distributions of calendar-year
registrations (by age) are used in the gasoline and diesel
summations. However, unlike those for light-duty, urban
fractions of heavy-duty VMT change slightly by model year, due
to the ongoing conversion of gasoline applications (local and
short-range) into diesel usage — a result of
dieselization.[10] Because the urban fraction is not constant,
it is necessary to include this term inside the 20-year
summations. These urban fractions of VMT, derived for the
MOBILES conversion factor analysis[10] and based on figures
from the 1977 Truck Inventory and Use Survey (TIUSHll], are
lower for HDDVs because of their orientation toward long-range
travel. Overall, the urban fractions for HDDVs are roughly
one-third of those for the gasoline trucks (1981 fleet
composites of 0.23 versus 0.68 for HDDVs and HDGVs,
respectively).[10] The calculation of gas and diesel urban VMT
is carried out separately for each of the heavy-duty weight
divisions (Classes 2B, 3-5, 6, and 7-8) and then summed.
From 1982 Highway Statistics, Table VM-1, FHwA; passenger
car figures used for LDVs, and single-unit trucks used to
represent LDTs.[4]
-------
-6-
The limitation associated with this nationwide method is
that inventories are not developed on a county-specific basis,
but rather for the urban portion of the nation as a whole.
However, because the inaccuracies of county-specific
distribution are avoided, the nationwide urban estimates should
be fairly accurate. As shown in later sections, information
developed with this approach can be translated to the local
level; the nationwide average urban VMT breakdown can be
applied to a locally-generated total VMT figure to yield local
VMT by class and, eventually, emissions. Of course, in using
this approach any distinctions in traffic characteristics from
one city to another are lost, since the same nationwide urban
VMT breakdown is used for all urban areas.
Table 2 presents a comparison of the 1981 (base-year) VMT
breakdowns by vehicle class computed using the NEDS and the
nationwide approaches; the findings add support for the
proposed weakness associated with the NEDS approach. The most
significant discrepancy between the two analyses appears in the
urban VMT breakdown, where NEDS estimates a HDDV fraction more
than twice that found with the nationwide method (5.8 compared
to 2.0 percent). Because HDDVs are expected to experience the
largest VMT growth, and have the highest per-vehicle NOx
emission rates, the impact of this two-fold difference in VMT
on the HDDV contribution to future NOx emissions is significant.
It is this inconsistency in the projected HDDV fraction of
urban VMT (and emissions) that necessitated the search for more
information. Because the VMT breakdown developed with the
nationwide approach represents an average for all urban areas
across the nation, and not for specific cities, local VMT
information from each of the 11 areas being modelled for NOx
was sought.* The following section provides a discussion of
the various methods used by localities to determine total VMT
and its breakdown by vehicle class.
Localized Approach
As mentioned earlier, 12 local areas (the 11 NOx cities
plus Detroit) were contacted in an attempt to obtain local
estimates of total VMT and its breakdown by vehicle class. The
primary sources for most of this local data were state
departments of transportation, local councils of government,and
regional transportation and planning commissions. Estimates of
total VMT were obtained for all 12 of the local areas; however,
Also, because of its geographical proximity to the Ann
Arbor EPA office, Detroit's VMT data were also gathered
(although Detroit is not currently modelled for NOx).
-------
Table 2
NEDS vs. Nationwide Approach: Base-year (1981) U.S. VMT
NEDS Approach
Vehicle
Type
LDV
LOT
HDGV
HDDV
Total
Vehicle
Type
LDV
LOT
HDGV
HDDV
Total
Total U.S. VMT Urban
Billions (%) Fraction
1103.1
293.6
49.6
86.5
1532.8
Total
Bil
1166.
242.
43.
73.
1525.
(72.0)
(19.2)
(3.2)
(5.6)
(100.0)
Nationwide
U.S. VMT
lions (%)
1 (76.4)
5 (15.9)
5 (2.9)
3 (4.8)
4 (100.0)
0.63
0.53
0.58
0.62
—
Approach
Urban
Fraction
0.59
0.49
0.68
0.23
—
Urban VMT
Billions (%)
695.0
155.6
28.8
53.6
933.0 (
(74.5)
(16.7)
(3.1)
(5.8)
100.0)
Urban VMT
Billions (%)
688.0
118.8
29.6
16.9
853.3
(80.8)
(13.8)
(3.4)
(2.0)
(100.0)
-------
-8-
VMT breakdowns were available for only 8 of the 12 cities.*
The total VMT estimates, being much easier to determine than
actual breakdowns by class, have a higher degree of confidence
associated with them. To date, the analysis has focused on the
eight urban areas with VMT breakdown estimates. A brief
description of the methods used to estimate both total VMT and
its breakdown by vehicle class is given below, followed by an
analysis of the methods' strengths and weaknesses. A detailed
description of each area's methodology is contained in the
Appendix.
For the eight localities that were explored in detail, two
basic approaches were used to estimate total local VMT— 1)
network models coupled with traffic counts, and 2) traffic
counts alone. With the exception of Tucson, every city uses a
network modelling system (which is described below).
Local traffic counts are obtained using automatic counters
on selected roads throughout the study area. The number of
counting sites varies between localities, but is generally
substantial. Nashville (Davidson County), for example, has 381
traffic count stations,[12] and there are 5,000-8,000 sites
throughout New York State.[13] The data are generally sampled
and summed by functional classification of roadway (e.g.
interstate, arterial). Often, local roads are estimated as
they represent only a small proportion of travel.
The network modelling systems utilized by the local areas
all operate along the same general principles. The area being
modelled is subdivided into zones, and the transportation
network consists of the links connecting these zones. Using
population and employment data by zone, trip origins and
destinations are estimated, and trips are routed through the
network, either via mass transit (i.e., by train or bus) or
personal vehicle. Truck trips must be generated separately as
they are not generally related to population and employment.
After determining the number of trips and trip lengths, total
VMT can then be calculated. Traffic counts are, in turn, used
to either normalize total VMT in the model (in which case the
estimates of total VMT are actually independent from the
network model), or to calibrate parts of the modelling system
which then determines total VMT (in which case the model has
some effect on the total VMT estimate).
* Chicago, New York, Tucson, Reno, Seattle, Nashville,
Detroit, and Philadelphia.
-------
-9-
Unlike the fairly standard methods used to estimate total
VMT, the determination of VMT breakdown by vehicle class is
much less uniform across the eight local areas that provided
such data. To estimate VMT breakdown, visual counts, vehicle
registration data, mileage accumulation estimates, and survey
data were used in varying degrees by each local area. The
important aspects of each area's procedures include: 1) the
number and nature of the visual counts, 2) the types of
vehicles classified (and how they relate to the four vehicle
classes desired here), and 3) the determination of gas/diesel
splits.
For those localities that use visual counts to determine
VMT breakdown (Nashville, New York, Philadelphia, Detroit,
Reno, Tucson, and Chicago), the number and "statistical
quality" of the sites vary greatly. For example, Nashville's
counting program consists of nine total sites on five different
classes of roadway; these sites were randomly chosen several
years ago and have since remained the same. The sites are
monitored once a year, at no particular time of year, for a
continuous 6-hour period; one of the nine sites is no longer
monitored.[14]
At the other end of the spectrum is Philadelphia, where
the Delaware Valley Regional Planning Commission performed a
statistically-derived sampling of 112 locations, stratified by
functional classification of roadway and area type (e.g. urban,
rural). The survey was conducted from mid-March to late-June
of 1980, with the day for each site chosen at random. At each
site, samples were taken over an 8-hour period from 7:00 am to
3:00 pm, Monday through Friday, with an automatic machine count
of total VMT also being taken at the same time. To obtain
regional totals, a weighted sum across roadway and area type
was used. The confidence level associated with this survey is
reported to be 95 percent for the regional LOT and HDT
estimates.[15]
For the other localities, the particulars of the surveys
differ, and, in turn, so does the confidence with which their
estimates can be used. In general, they fall somewhere between
Nashville and Philadelphia with respect to their level of
confidence.
The vehicle classification scheme used by the different
local areas also varied somewhat, especially in the detail with
which trucks were classified. A general classification scheme
for the visual counts consisted of: automobiles, light trucks,
heavy trucks, buses, and motorcycles. The only concerns
expressed by the local organizations involved the
differentation between the LDTi and LDT2 classes, and the
diesel/gas splits for all classes. Because NEDS combines the
-------
-10-
two LDT categories, only the second of these concerns is
pertinent to this comparison. However, based on the visual
similarity of heavy LDTs and light HDVs, there is also a
concern about a person's ability to accurately place these
vehicles in their proper category.
Diesel/gas splits for heavy-duty vehicles were estimated
by the local areas using a variety of methods. Several of the
localities used vehicle registration data collected at either
the local or state level. Detroit used state registration data
directly[16], whereas Philadelphia and Seattle VMT-weighted the
state registrations with average mileage accumulation rates
[17,18]. Nashville simply used MOBILE2 default values to
obtain the gas/diesel splits.[12] To varying degrees, these
methods share the HDDV problem identified above with the NEDS
methodology. Other (possibly better) methods include Reno's
classification of all 2-axle, single-unit, dual-tire trucks and
2-axle buses as HDGVs, and all other heavy-duty vehicles as
diesel.[19] In Detroit, initial attempts have been made at
visually classifying HDGVs and HDDVs by the positioning of the
exhaust pipe (horizontal versus vertical).[16] Before using
any of these local estimates, confidence in the method used to
derive the HDGV/HDDV VMT split will have to be established.
The strengths of the local data are centered upon the fact
that the data are collected locally and generally involve some
amount of actual visual monitoring. Seven of the eight cities
utilized visual monitoring of some sort to obtain VMT breakdown
by class, and most reduced their data statistically. However,
the VMT breakdown for at least one city was that for the entire
state.[18] Theoretically, a well-defined, statistically valid
survey of this type would provide the best possible estimates,
at least in differentiating between light-duty vehicles,
light-duty trucks, and heavy-duty trucks. Most of these
surveys appear to have been designed to attain a specified
statistical confidence, though a detailed evaluation of each
area's methodology has not yet been completed. In general,
however, most of the local approaches appear to have fewer
weaknesses than the NEDS methodology, and provide local
distinction that the nationwide method does not.
Comparison of Baseline Inventories
Using the VMT estimates from each of the three
approaches — NEDS, nationwide, and localized — 1981 NOx
emissions inventories were calculated for each of the 12 cities
being examined. Because the nationwide approach only produces
a VMT breakdown and no estimate of total VMT for the local
area, the total city VMT figure from the localized approach was
used here as well. For the four cities that provide no local
-------
-11-
VMT breakdown (Washington, D. C. , Boston, Newark, and Denver),
the local VMT estimates are based on the local VMT total,
broken down using the average localized VMT breakdowns from the
other eight cities. In all three analyses, MOBILES emission
factors (without speed or temperature correction) were used to
develop the NOx inventories.[6] (For further details on the
individual calculations and city-specific results, see the
Appendix.) Estimates of non-motor vehicle emissions
(designated "other"), which included emissions from off-highway
vehicles and stationary area sources (e.g., home heating), are
taken from NEDS under all three approaches.
A comparison of the 1981 VMTs and NOx emissions
inventories developed with the three methods is presented in
Table 3; the figures shown are the combined totals for the 12
cities in the analysis.* As indicated in the table, the
fraction of VMT attributed to HDDVs under the localized
approach (2.6 percent) is nearer to that under the nationwide
approach (2.0 percent) than that of the NEDS approach (4.3
percent). The VMT fractions (by vehicle class) estimated for
each city using each approach are summarized in Table 4. As
shown, for every one of the cities, NEDS attributed a
significantly higher share of VMT to HDDVs than did either of
the other two methods. Thus, the various local approaches
appear to avoid the HDDV problem associated with the NEDS
methodology.
With respect to the LDT fraction (Tables 3 and 4), the
differences are not quite as great in magnitude, however, again
the NEDS method results in a higher LDT fraction than the other
two methods (average values of 15.8, 12.1, and 13.8 for the
NEDS, local and nationwide approaches, respectively). This
pattern occurs for each city, except for Nashville where NEDS
estimates 18.8 percent for LDT VMT, compared to 20.2 percent
with the local method. For Newark, the LDT fractions are equal.
These differences in VMT breakdown translate into larger
inconsistencies in breakdown of NOx emissions by vehicle class,
as the bottom portion of Table 3 indicates. Although the
figures vary from city to city, NEDS consistently estimates a
higher HDDV contribution to emissions; overall, the
combined-city figures show a HDDV fraction of 27.2 percent with
NEDS, compared to only 18.4 and 14.1 percent with the local and
nationwide approaches, respectively. Also, consistent with
trends in the VMT breakdown, NEDS reports a higher LDT fraction
of emissions than does the localized approach (12.2 versus 9.9
percent).
The area modelled for each locality is actually the
Standard Metropolitan Statistical Area (SMSA).
-------
Table 3
Combined Base-year Motor Vehicle Inventories: Three Approaches*
LDV
LDT
HDGV
HDDV
Total
VMT, Millions (%)
Localized Approach Nationwide Approach
171065.7 (85.2)
19007.8 (9.5)
5464.5
5307.5
(2.7)
(2.6)
200845.5 (100.0)
162283.2 (80.8)
27716.7 (13.8)
6828.7 (3.4)
4016.9 (2.0)
200845.5 (100.0)
NEDS Approach
156531.4 (81.0)
24446.4 (12.7)
3903.4 (2.0)
8280.8 (4.3)
193162.0 (100.0)
NOx Emissions, 1000 tons (%)
Localized Approach Nationwide Approach
454.32 (64.3)
107.93 (15.3)
44.68 (6.3)
99.78 (14.1)
LDV 476.23 (66.7)
LDT 71.03 (9.9)
HDGV 35.77 (5.0)
HDDV 131.51 (18.4)
Total 714.54 (100.0)
706.71 (100.0)
NEDS Approach
435.85 (57.2)
93.20 (12.2)
25.50 (3.3)
207.19 (27.2)
761.74 (99.9)
Eleven urban areas shown in Table 1, plus Detroit.
-------
Table 4
Effect of Three Approaches on Class-specific
Fractions of Total 1981 VMT
LDV Fraction (%)
City
Nashville
Seattle
Reno
I*i ila
Tucson
NYC
Chicago
Boston
Newark
Wash, DC
Denver
Detroit
Average
Local
75.3
79.3
78.2
82.5
71.3
88.0
90.2
85.2
85.2
85.2
85.2
83.7
82.4
Nationwide
80.8
80.8
80.8
80.8
80.8
80.8
80.8
80.8
80.8
80.8
80.8
80.8
80.8
NEDS
68.9
73.5
60.7
81.0
70.9
87.5
81.2
84.4
85.3
83.8
73.2
80.1
77.5
LOT Fraction (%)
Local
20.2
13.1
16.'6
10.6
23.6
6.9
4.7
9.4
9.4
9.4
9.4
12.2
12.1
Nationwide
13.8
13.8
13.8
13.8
13.8
13.8
13.8
13.8
13.8
13.8
13.8
13.8
13.8
NEDS
18.8
21.3
29.4
11.5
24.2
7.4
11.1
10.1
9.4
11.3
20.8
14.4
15.8
HDGV Fraction (%)
Local
2.5
4.3
2.1
2.4
3.9
3.6
2.7
2.7
2.7
2.7
2.7
1.5
2.8
Nationwide
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
NEDS
2.9
1.0
1.5
2.3
1.3
2.2
2.0
2.0
1.8
2.0
1.8
2.2
1.9
HDDV Fraction (%)
Local
2.0
3.3
3.1
4.5
1.2
1.5
2.4
2.7
2.7
2.7
2.7
2.6
2.6
Nationwide
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
NEDS
9.4
4.2
8.4
5.2
3.7
2.9
5.7
3.5
3.5
2.9
4.2
3.4
4.8
-------
-14-
Comparison of Future NOx Inventories
Using the baseline NOx inventories discussed in the
previous section, future NOx inventories were projected for
each of the eleven cities assuming five different emission
control strategies:
Model Year LDTt/LDT2 HDE
Applicable (g/mi) (g/BHP-hr)
(1) 1987 2.3/2.3 10.7
(2) 1987 2.3/2.3 6.0
(3) 1987 1.2/1.7 6.0
(4) 1987 2.3/2.3 6.0
1990 2.3/2.3 4.0
(5) 1987 1.2/1.7 6.0
1990 1.2/1.7 4.0
Because of indications that the local data (where available)
are the most accurate, the following analysis focuses on the
future projections based on the 1981 inventories developed with
the localized approach.
The degree of growth in NOx emissions between the baseline
year (1981, in this case) and the future year of projection is
an important consideration in assessing the environmental
impact of various control scenarios. Tables 5 and 6 quantify
these growths by dividing future emissions by baseline (1981)
emissions; in other words, any values in the tables greater
than one indicate growth over baseline emissions. All five
control scenarios are examined and three years of projection
(1990, 1995, and 2000) are included. Although city-specific
growths are shown only for the local approach (i.e., 1981
inventories based on local data), totals for all eleven cities
combined are shown for all three approaches for comparison.
The growth in both motor vehicle and total discounted NOx
emission inventories are presented in Tables 5 and 6,
respectively.* Total discounted emissions are also depicted
graphically in Figures l through 6.
Total discounted emissions differ from total emissions in
that stationary point sources (e.g., power plants) are not
included due to their minor impact on annual ambient NO2
levels measured by fixed site monitors.
-------
Table 5
MOTOR VEHICLE NQx EMISSIONS (Ratio Relative to 1981 Baseline)
Local Approach
2.3/10.7 Stds.
Cities
Boston
Chicago
Nshvll.
NYC
Newark
Phila.
Seattle
Tucson
Wash DC
Denver
Reno
Total
1990
0.91
0.88
0.93
0.84
0.91
0.99
0.96
0.91
0.91
1.13
1.21
0.92
1995
1.06
1.01
1.07
0.94
1.06
1.20
1.14
1.01
1.06
1.38
1.46
1.07
2000
1.30
1.23
1.28
1.11
1.29
1.52
1.41
1.19
1.30
1.70
1.85
1.31
2.3/6.0 Stds.
1990
0.84
0.81
0.88
0.80
0.83
0.87
0.87
0.88
0.83
1.04
1.10
0.83
1995
0.88
0.84
0.93
0.83
0.88
0.94
0.93
0.93
0.88
1.15
1.23
0.89
2000
1.03
0.98
1.09
0.95
1.03
1.11
1.09
1.07
1.03
1.35
1.44
1.03
1.2/1.
1990
0.83
0.80
0.86
0.79
0.82
0.87
0.86
0.87
0.82
1.02
1.03
0.83
7/6.0
1995
0.86
0.83
0.88
0.81
0.86
0.91
0.90
0.88
0.86
1.13
1.21
0.87
Stds.
2000
1.00
0.96
1.01
0.93
0.99
1.08
1.05
0.98
0.99
1.31
1.36
1.01
2.3/4.0 Stds
1990
0.84
0.81
0.88
0.80
0.83
0.87
0.87
0.88
0.83
1.04
1.10
0.84
1995
0.82
0.78
0.88
0.78
0.81
0.84
0.85
0.89
0.82
1.07
1.13
0.82
2000
0.93
0.88
1.00
0.88
0.92
0.97
0.97
1.01
0.93
1.22
1.31
0.93
Total 0.91 1.04 1.25
Total 1.00 1.22 1.53
Nationwide Approach
0.85 0.91 1.05 0.84 0.87 1.00
NEDS Approach
0.89 0.96 1.14 0.88 0.93 1.10
1.2/1.7/4.0 Stds.
1990 1995 2000
0.83 0.80 0.89
0.8d 0.77 0.86
0.86 0.83 0.93
0.79 0.76 0.85
0.82 0.79 0.89
0.87 0.82 0.94
0.86 0.82 0.93
0.85 0.83 0.91
0.83 0.79 0.89
1.02 1.04 1.17
1.03 1.10 1.23
0.83 0.80 0.90
0.85 0.85 0.96 0.84 0.82 0.91
0.89 0.87 1.00 0.88 0.84 0.96
-------
Table 6
TOTAL DISCOUNTED NOx Emissions (Ratio Relative to 1981 Baseline)
Local Approach
2.3/10.7 Stds.
Cities
Boston
Chicago
Nshvll.
NYC
Newark
Phila.
Seattle
Tucson
Wash DC
Denver
Reno
Total
1990
0.99
0.97
1.01
0.94
0.98
1.04
1.04
1.00
0.98
1.15
1.18
1.00
1995
1.10
1.10
1.14
1.03
1.11
1.21
1.20
1.11
1.12
1.32
1.35
1.13
2000
1.30
1.29
1.33
1.19
1.30
1.46
1.41
1.27
1.33
1.53
1.60
1.33
2.3/6.0 Stds.
1990
0.94
0.93
0.97
0.91
0.92
0.97
0.98
0.98
0.93
1.11
1.13
0.95
1995
0.99
0.99
1.05
0.96
0.98
1.04
1.06
1.06
0.99
1.22
1.23
1.02
2000
1.12
1.12
1.19
1.08
1.11
1.19
1.21
1.19
1.13
1.37
1.40
1.15
1.2/1
1990
0.93
0.92
0.96
0.90
0.92
0.96
0.97
0.97
0.92
1.10
1.12
0.94
Nationwide
Total
1.00
1.16
1.29
0.96
1.03
1.16
.7/6.0 Stds.
1995
0.98
0.98
1.01
0.95
0.92
1.03
1.04
1.02
0.98
1.21
1.22
1.00
Approach
0.95 1.01
2000
1.10
1.11
1.14
1.06
1.09
1.17
1.18
1.13
1.11
1.35
1.36
1.13
1.13
2.3/4.0 Stds
1990
0.94
0.93
0.97
0.91
0.92
0.97
0.98
0.98
0.93
1.11
1.13
0.95
0.96
1995
0.95
0.94
1.01
0.93
0.93
0.98
1.10
1.03
0.94
1.18
1.18
0.97
0.99
2000
1.05
1.05
1.13
1.03
1.04
1.10
1.13
1.15
1.06
1.31
1.33
1.08
1.10
1.2/1.
1990
0.93
0.92
0.96
0.90
0.92
0.96
0.97
0.97
0.92
1.10
1.12
0.94
0.95
7/4.0 Stds.
1995 2000
0.94 1.03
0.94 1.04
0.97 1.08
0.91 1.01
0.92 1.02
0.97 1.08
0.99 1.11
1.00 1.09
0.93 1.03
1.17 1.29
1.17 1.29
0.96 1.06
0.97 1.07
NEDS Approach
Total
1.05 1.23 1.49 0.97 1.04 1.22 0.96 1.01 1.18
0.97 0.98 1.11 0.96 0.96 1.08
-------
Figure 1
Discounted NOx Emissions for All Cities Combined
2000
NEDS Inventory
1981
1990 1995
Year
2000
Legend
tza LDV
•i LOT
HDGV
HDDV
OTHER
-------
Figure 2
Discounted NOx Emissions for All Cities Combined
2000
NEDS Inventory
1981
1990 1995
Year
2000
Legend
tza LDV
•I LOT
ED HDGV
CD HDDV
OTHER
-------
Figure 3
Discounted NOx Emissions for All Cities Combined
Local Inventory
J-H
cd
o>
CO
O
O
O
O
^-4,
CO
w
CO
-0
0)
•s
§
O
CO
•r-t
Q
1500
1000-
500-
1981
1990 1995
Year
2000
Legend
CZ3 LDV
§• LOT
ED HDGV
CD HDDV
OTHER
-------
Figure 4
Discounted NOx Emissions for All Cities Combined
1500
Local Inventory
1981
1990 1995
Year
2000
Legend
cza LDV
mm LOT
ED HDGV
CD HDDV
OTHER
-------
Figure 5
Discounted NOx Emissions for All Cities Combined
1500
Nationwide Inventory
1981
1990 1995
Year
2000
Legend
tza LDV
•H LOT
EJ HDGV
CZ1 HDDV
ESS OTHER
-------
Figure 6
Discounted NOx Emissions for All Cities Combined
cti
0)
CO
a
o
-*->
o
o
o
CO
g
•r-t
co
CO
0)
O
O
CO
•i-H
Q
1500
1000-
500-
Nationwide Inventory
1981
1990 1995
Year
2000
Legend
E2 LDV
•I LOT
KS3 HDGV
CD HDDV
OTHER
-------
-23-
Overall, the effect of using the local approach to develop
the baseline inventories is a decrease in the amount of future
growth under all scenarios compared to the NEDS-based
projections. As expected, the effect of local versus NEDS data
is more pronounced in Table 5, where only the growth in motor
vehicle emissions is examined. For example, in the year 2000,
with no further control of emissions (first scenario)
NEDS-based projections show a 53-percent growth in vehicle
emissions, compared to the lower 31-percent growth estimated
using the localized approach (a 22-percent difference).
However, with respect to total discounted NOx (Table 6), the
magnitude of the difference is slightly less than before, with
a NEDS-based growth of 49 percent, compared to the localized
figure of 33 percent (only a 16-percent discrepancy). This
difference decreases as LOT and HDE NOx controls become more
stringent until, under the fifth scenario, the difference in
the year 2000 is only 2 percent.
Conclusions .
Based on the findings presented above, some basic
conclusions can be made regarding the suitability of the NEDS
approach to creating base-year motor vehicle NOx inventories
for specific cities. First, NEDS appears to consistently
predict a significantly higher HDDV fraction of total urban VMT
(and thus emissions) than that estimated using either the
localized or the nationwide approaches. This difference
appears to be due to the use of registrations to apportion
statewide VMT, which would not be expected to be accurate for
HDDVs. Second, for every city but one, the NEDS fraction for
LDT VMT is also higher than that estimated by each individual
city; however, the magnitude of the LDT difference is smaller
than that for HDDVs and the cause is not as clear. Thus, there
appears to be sufficient reason to search for more, accurate VMT
estimates.
The nationwide approach appears to avoid the HDDV problem,
but treats all cities the same. The local VMT estimates appear
to also avoid the HDDV problem, but not completely. As the
evaluation of the detailed methodology of each local area has
has been only partially completed to date, no final judgments
can be made at this time concerning the most accurate source of
the motor vehicle NOx inventory for each city for which NOx
emissions are a concern.
-------
-24-
References
1. Discussions on Rollback Modelling in "Methodology to
Conduct Air Quality Assessments of National Mobile Source
Emission Control Strategies," EPA-450/4-80-026.
2. "1981 National Emissions Report," National Emissions
Data System of the Aerometric and Emissions Reporting System,
U. S. EPA, OAWM/OAQPS/NADB/MDAD, 1984.
3. "The Highway Fuel Consumption Model: Tenth Quarterly
Report," Energy and Environmental Analysis, Inc., for U. S.
Department of Energy, November 1983.
- 4. "Highway Statistics 1981," Federal Highway
Administration, U. S. Department of Transportation,
FHwA-HP-HS-81.
5. "MVMA Motor Vehicle Facts and Figures '82," Motor
Vehicle Manufacturers Association of the U. S., Inc., Public
Affairs Division, 1982.
6. "User's Guide to MOBILES (Mobile Source Emissions
Model)," U. S. EPA/OAR/OMS/ECTD/TEB, EPA-460/3-84-02, June 1984.
7. Registration data from R. L. Polk & Company.
8. Notice of Proposed Rulemaking (NPRM): "Gaseous
Emission Regulations for 1987 and Later Model-Year Light-duty
Vehicles, Light-duty trucks, and Heavy-duty Engines;
Particulate Emission Regulations for 1987 and Later Model-Year
Heavy-duty Diesel Engines," published in Federal Register,
Volume 49, October 15, 1984, p. 40258.
9. "Fleet Characterization Data Used for MOBILES,"
U. S. EPA/OAR/ECTD/TEB, EPA-AA-TEB-EF-84-2, August 1984.
10. "Heavy-duty Vehicle Emission Conversion Factors,
1962-1997," Technical Report, Mahlon C. Smith, IV,
EPA-AA-SDSB-84-1, August 1984.
11. "Truck Inventory and Use Survey," U. S. Department
of Commerce, Bureau of the Census, 1977.
12. Bucky Crowell: Memorandum to Paul Bontrager, "1981
Nashville-Davidson County, Tennessee Mobile Source Emissions
Inventory and Documentation of Methodology," Nashville:
Metropolitan Planning Commission, 24 June 1982.
13. David Fifield, New York Department of
Transportation, phone conversation, 28 September 1984.
-------
-25-
References (cont'd)
14. Rich Walpole, Tennessee Department of
Transportation, Transportation Mapping Division, phone
conversation, 17 September 1984.
15. Delaware Valley Regional Planning Commission,
Automobile Occupancy and Truck Travel Monitoring for
Transportation Air Quality Planning, Philadelphia, February
1981.
16. Chuck Hersey and Adeile Nwanko, Southeastern
Michigan Council of Governments, meeting, 25 October 1984.
17. Delaware Valley Regional Planning Commission,
Delaware Valley Highway Emissions Inventory, Philadelphia,
September 1982.
18. Dave Kircher, Puget Sound Air Pollution
Administration, phone conversations, October 1984.
-------
APPENDIX
I. NEDS Approach
A. Description of Methodology
B. Sources of Data
C. Basis for FHwA's VMT Estimates
II. Nationwide Approach
\
A. Description of Methodology
B. Sources of Data
C. Basis for EEA's VMT Estimates
D. Urban VMT Growth Rates (based on EEA)
III. Localized Approach
A. Description of Local Methods
B. Details of Calculations
IV. MOBILE3 Emission Factors Used in NOx Inventory Development
-------
NEDS EMISSION INVENTORY METHODOLOGY
Step I; Split FHwA's nationwide VMT into LDV, LOT, HDGV, and
HDDV classes.
Step II: Calculate county gasoline and diesel fuel
consumption for four vehicle classes: HDDV, HDGV,
LDT and LDV.
Step III: Calculate nationwide fuel economies for four vehicle
classes, using FHwA VMT and nationwide fuel
consumptions.
Step IV: Multiply county fuel consumption by nationwide fuel
economy (mpg) to yield county VMT for each vehicle
class.
Step V: Multiply class-specific county VMTs by appropriate
emission factors to yield county inventories, broken
down by vehicle class.
-------
Step I: Split FHwA' s Nationwide VMT Between Classes
A. LDV; FHwA1 s nationwide VMT data can be used
directly because passenger cars are reported as a
separate category. Diesel VMT is calculated using
model-year sales, survival rates, and mileage
accumulation rates, and then subtracted from total
LDV VMT.
B. LPT and HDV; FHwA1 s nationwide VMT data are not
divided into the proper weight categories nor
separted between gas and diesel; therefore, above
approach cannot be used directly.
0 For each vehicle class (LDTj_, LDT2, HDGV, HDDV),
the following equation is used:
Estimated
Nationwide
VMT
by Class
20
Model-Year.
Sales
Survival
Rate
% v
)mx
m=l
Mileage
.Accumulation Rate
' by Model Year
NOTE: m = Model years in the 1981 in-use fleet; model-year
sales for LDT categories include only gasoline vehicles.
0 The calculated VMTs for all four classes (LDT]_,
LDT2, HDGV, and HDDV) are summed, and the fraction
of total truck VMT by class is computed.
° These fractions are then applied to FHwA1 s estimate
of total nationwide VMT for trucks and buses, to
yield VMT figures for each class.
m
-------
Step II; Calculate Fuel Consumption by County for HDDV, HDGV,
and LD Classes
A. HP Diesel Fuel Consumption by County:
HDDV Fuel Usage = ( (County HDDV Registrations } x
State HDDV Registrations
. State diesel State diesel > > .County pop. x State diesel.
( fuel ~ bus fuel ;; l State pop bus fuel '
Assumptions ;
1. All diesel fuel is consumed by heavy-duty
vehicles .
2. HDDV registrations include all diesel trucks
over 6,000 Ibs . GVW (inclusion of LDT2
6,000- 8,500 Ibs. — adjusted for in later
steps) .
3. All commercial buses run on diesel fuel.
B . HP Gasoline Consumption by County;
3 , County HDGV . , average annual .
HDGV "V * registrations X mileage accumulation rates
fuel =^ < - - (MPGl - -
usage w=1 w
... County population. „ ,_. . , i • \ \
+(( State population) X (State bus gasoline))
Assumptions ;
1. HDGV registrations are divided into three GVW
classes: 6,001-10,000 Ibs., 10,001- 19,500
Ibs., and 19,501-26,000 Ibs.; adjustment
for LDT2 made with MPG.
2. All institutional buses (school bus, etc.) run
on gasoline.
NOTE: The figures used for MPG here are "fudge
factors," computed internally by NEDS; for w = 1 to
3, MPGs are 99.9, 6.3, and 2.7, respectively. The
first MPG is highly overestimated to "remove" LDT2
fuel consumption from HD gasoline usage.
-------
C. LD Gasoline Consumption by County:
0 For states that report VMT for each county (not
broken down by class):
Gasoline =( County VMT x Total state . _ , Heavy-Duty .
State VMT Gasoline Usage Gasoline Usage
0 For states that do not report VMT by county:
/County LDV + LDTi , _, , _ , .
LD ( _ i , L ) X (Index)
_ , . , Registrations »
Gasoline =( - =7- — ; - =-5 - ~ - z - )
v Statewide Sum of
//County LDV + LOT.. . v , _ , ..
( ( T, . . . • 1) X (Index))
Registrations v '
X (Total State Gasoline) - (Heavy-Duty Gasoline)
Assumptions :
1. "Index" refers to a relative mileage accumulation
index for each county; ranges from 0.67 for a
predominently urban county, to 1.15 for a
predominently rural county.
2. "Heavy-duty gasoline" is that calculated in part B.
D. Split County LD Gasoline Consumption Between LDV and LPT
Classes
LDV
Fraction _ (LDV Registrations) X (LDV Factor) _
of LD ~ LDV LDV LOT LOT LOT, LOT
l y \ + ( V ) + ( Y
Gasoline v Reg Factor' vReg Factor' vReg. Factor
0 Factors for each class take into account relative
fuel economies and mileage accumulation rates:
LDV factor = 8.1
LDTi factor = 8.7
LDT2 factor =12.3
0 LOT fraction of LD gasoline = 1 - (LDV faction of LD
gasoline)
0 Fuel consumption (by county):
LDV usage = (LDV fraction) X (total LD gasoline)
LOT usage = (LOT fraction) X (total LD gasoline)
NOTE: LDT]^ and LDT2 fuel consumption is combined into
one category.
-------
Step III; Calculate Nationwide Average Fuel Economies by Class
For each vehicle class (LDV, LOT, HDGV, HDDV):
Nationwide (FHwA-based) VMT
MPG =
Nationwide Sum of County Fuel Consumption
The NEDS class-specific fuel economies for 1981 are:
LDV = 16.85 MPG
*LDT = 12.81 MPG
HDGV = 4.89 MPG
HDDV = 5.78 MPG
and LDT2 VMTs are combined and divided by LOT
fuel consumption figures to yield an overall MPG for LDTs .
Step IV: Calculate County VMT by Vehicle Class
For each class (LDV, LOT, HDGV, HDDV):
County VMT _ ^Calculated County. , Calculated .
by Class ~ Fuel Consumption Nationwide MPG
0 Some states (approximately 15) report estimated
total VMT by county; however, no breakdowns by
vehicle class are provided. Therefore, the
calculated class-specific VMTs (above) for each of
these counties are used to determine a fractional
breakdown of VMT by class, which is then applied to
the measured county VMT total. Then, county VMTs
are adjusted so that their sum is equal to the state
VAT total reported by FHwA.
0 For the other 35 states, where no county data are
available, calculated VMTs and breakdowns are used
(already consistent with FHwA1 s VMT totals for these
states).
-------
Step V: Calculate NOx Emissions Inventory for Each County
0 Calculated county VMT figures are multiplied by the
appropriate NOx emission factors to yield tons of
pollutant for each county, broken down by vehicle
class.
0 The 1981 NEDS inventories are based on MOBILE2.5
emission factors, taking "vehicle speed correction"
into account. These speed factors are chosen on the
basis of urban versus rural VMT within each county;
for urban travel, an average speed of 19.6 mph is
assumed, while the rural speed is estimated at 45
mph. Emission factors are adjusted for each
condition.
0 The states that report VMT by county already specify
urban and rural miles. For the other states, county
VMTs are split up using the rural/urban fractions of
county populations from the 1980 U. S. Census, which
defines urban areas as those with populations
greater then 2,500 and other "non-incorporated
urbanized areas".
0 Because the splits are population-based, the same
urban fractions are assumed for all vehicle classes
within each county; therefore, use of a speed
correction factor has no effect on the breakdown of
VMT by vehicle class within that county. In
addition, there is also no effect on county-by-
county allocation of VMT.
0 For purposes of comparison within this report, the
NEDS inventories were updated using MOBILES NOx
emission factors. Because the emission factor
ratios used in the rollback model are based on the
average vehicle speed of the urban driving cycle
(19.6 mph, correction factor equal to 1.0), these
MOBILES emission factors were also used to calculate
the NEDS baseline inventory.
-------
Sources Used in NEDS Approach
Factors
Sources
State fuel consumption figures
County vehicle registrations
County population counts
Bus fuel consumption figures
HDGV fleet average annual
mileage accumulation rates
(miles/vehicle/year)
Index for mileage accumulation
(urban/rural)
Nationwide annual VMT
Model-year sales & survival
rates (1981 calendar year)
Model-year mileage accumulation
rates (1981 calendar year)
1982 Highway Statistics
(FHwA)
1979 R.L. Polk
1980 U. S. Census
1982 Highway Statistics
(FHwA)
1977 Truck Inventory and
Use Survey (TIUS)
1974 Walden Research Corp.
report
1982 Highway Statistics
(FHwA)
1983 MVMA Facts and
Figures
MOBILE2.5
-------
Basis for FHwA's Nationwide VMT Estimates
I. Highway Performance Monitoring System (HPMS)
Three major types of data submitted by each state:
A. Universe road mileage data: complete inventory of
road mileage classified by functional system,
jurisdiction, and selected operational
characteristics.
B. Sample road mileage data: specific inventory,
condition, and operational data for the sample
panels of roadways used to obtain samples of VMT
counts. These data are used to obtain expansion
factors so that sample VMT data can be adjusted to
represent the universe of road mileage.
C. VMT data (sample statistics)
- Stratified by area type.
1. Rural
2. Small Urban (population 5000-49,999)
3. Individual Urbanized Areas (population 50,000
and over)
- Stratified by functional class of roadways:
1. Principal Arterial - Interstate
2. Principal Arterial - Other Freeways and
Expressways (Urbanized and Small Urban Only)
3. Other Principal Arterial
4. Minor Arterial
5. Collector (Major and Minor for Rural Only)
6. Local (this data is estimated)
Further stratified within functional class of
roadway by daily traffic volume.
- Local road data is usually estimated.
- No breakdown by vehicle type.
- Provides data for construction of Table VM-2 and VMT
figures in the extreme right hand column of Table
VM-1.
- Areawide data expanded from samples to the universe
using expansion factors from sample mileage data.
Sample data for interstates is complete. No
expansion necessary.
-------
II. FHwA's Construction of Table VM-1
Vehicle Miles Travelled
A. Totals (column 11, rows 1-8): from HPMS data
B. Breakdown by vehicle class (columns 1-10, rows 1-8):
- Based on samples of class-specific VMT breakdowns by
class obtained at 600-700 truck weight sites
throughout the nation (mostly rural with heavy truck
travel) and at 300-400 other sites (mostly urban) in
only six states: Maine, Georgia, Tennessee,
Minnesota, Oklahoma, Oregon, with supplementary data
from Texas and Nebraska.
- Above data segregated into five regions (based on
accessibility of information) and each region
stratified into 26 groups by area type (urban/rural),
functional road classification, and volume of
traffic.
- Twelve vehicle classes recorded at the 300-400 other
sites; more at truck weight sites, no differentation
by fuel type (cars, motorcycles, pickups and vans,
buses, 8 truck types: single-axle and up).
These 12 vehicle classes aggregated over the nation
and combined into the six shown in Table VM-1.
Number of Motor Vehicles Registered (row 9):
- FHwA estimates taken from state registration
information, summed over entire year; some
duplication possible due to sales of used vehicles.
Total Motor Fuel Consumed (row 11, column 11):
Obtained from monthly data from state motor fuel tax
agencies.
- Off-highway-use fuel removed.
Can be some problem with diesel fuel as home heating
oil may be used as a substitute.
Average Miles Travelled Per Gallon of Fuel Consumed
(row 13):
Iterative process.
- For the first iteration, the previous year's
estimates are used along with estimated increase.
This is adjusted in later iterations as seen below.
-------
Remainder of Table (Average Miles Travelled Per Vehicle, Fuel
Consumed by Vehicle Type, Average Fuel Consumption per Vehicle):
- Iterative process.
First iteration: compute directly from other
figures.
Check for consistency of results: i.e., does fuel
consumed by individual vehicle classes sum to total
fuel consumed? Are average miles travelled
consistent with National Personal Transportation
Survey (NPTS), Truck Inventory and Use Survey (TIUS)?
Second and further iterations: adjust the MPG and
other figures for passenger cars and combination
trucks as they are, respectively, the prime
consumers of gasoline and diesel fuel. NOTE: The
figures for VMT by class, motor vehicle registra-
tions by class, and total motor fuel consumed are
held constant and are not adjusted in these
iterations.
Other Comments
- A vehicle classification study was performed from
late summer 1980 to early fall of 1981 at a total of
139 sites by five local agencies for FHwA (Delaware
Valley Regional Planning Commission, States of
Arkansas, Iowa, Minnesota, and Washington). This
information is used as a check on the VMT breakdown
by class obtained from the site data.
Beginning with this year (1984) states are being
required to report travel activity by vehicle type
in each functional class. However, statistically
sound measurements are not required as of yet.
Estimates based on available data are allowed. The
vehicle types are: motorcycles (optional),
passenger cars, buses, single-unit trucks,
single-trailer trucks, multi-trailer trucks, other.
This is part of HPMS, so it will be divided into
rural, small urban and urbanized areas.
-------
Nationwide Approach
Step I: Split EEA's nationwide annual VMT — already divided
by class (LDV, LDTt/ LDT2, and HD classes 2B,
3-5, 6, and 7-8) — into gasoline and diesel
fractions.
Step II: Convert total nationwide gasoline and diesel VMT (by
class) into urban fractions.
Step III: Calculate annual nationwide urban NOx inventories by
vehicle class, using nationwide urban VMT and
MOBILES emission factors.
-------
Nationwide Approach
Calculation of Light-Duty Urban VMT:
0 Same calculation carried out separately for LDV,
, and LDT2 classes.
Division between gasoline and diesel fractions:
VMTD = VMT X
VMTG = VMT - VMTD
(RF) (DMA) (DSF)
_
RF) (GMA) (1-DSF) +
(DMA) (DSF)
Where:
VMTD = nationwide diesel VMT (by class)
VMTG = nationwide gasoline VMT (by class)
VMT = EEA1 s class-specific VMT (total nationwide)
RF = light-duty registration distribution by age (gas and
diesel same)
DMA = diesel mileage-accumulation rate (by age)
GMA = gasoline mileage-accumulation rate (by age)
DSF = diesel fraction of total sales (by model year)
0 Conversion to Urban VMT:
UVMTD =
UVMTG =
Where:
UVMTD =
UVMTG =
UF =
VMTD * UF
VMTG * UF
class-specific urban diesel VMT (nationwide)
class-specific urban gasoline VMT
(nationwide)
urban fraction of VMT (gas and diesel same);
constant 0.594, 0.488, and 0.488 for LDV,
and LDT2, respectively.
-------
Calculation of Heavy-Duty Urban VMT;
Same calculation carried out separately
Classes IIB, III-V, VI, and VII-VIII.
for HD
Division between gasoline and diesel fractions, and
conversion to urban VMT accomplished in the same
step. (Because HD urban fractions change with
model year, it is necessary to include these urban
terms within the summations, in the numerator only.)
UVMTD = VMT X
(DRF) (DMA) (TDR) (UFD)
UVMTG
Where:
UVMTD
UVMTG
VMT
DRF
GRF
TDR
TGR
DMA
GMA
UFD
UFG
= VMT X-
(GMA) (TGR) + ^(DRF) (DMA) (TDR)
y(GRF) (GMA) (TGR) (UFG)
(GMA) (TGR) +
DRF) ( DMA) ( TDR)
Class-sepcific urban diesel VMT (nationwide)
Class-specific urban gasoline VMT (nationwide)
EEA1 s class-specific VMT (total nationwide)
HDDV registration distribution (by age)
HDGV registration distribution (by age)
Calendar-year HDDV registrations (by class)
Calendar-year HDGV registrations (by class)
Diesel mileage-accumulation rate (by age)
Gasoline mileage-accumulation rate (by age)
Urban fraction of diesel VMT (by model year)
Urban fraction of gasoline VMT (by model year)
-------
Sources Used in Nationwide Approach
Vehicle Class
LDV
Factors Used
Sources
LDT
HDV
Total nationwide VMT
(by calendar year)
Registration dis-
tribution (by age)
Mileage accumulation
rates (by age)
Diesel sales fractions
(by model year)
Urban fraction of VMT
(all model years)
Total nationwide VMT
(by calendar year)
Registration dis-
tribution (by age)
Mileage accumulation
rates (by age)
Diesel sales fractions
& Urban fraction of VMT
Total nationwide VMT
(by calendar year)
Registration dis-
tribution (by age)
Mileage accumulation
rates (by age)
Total vehicle reg-
istrations (by calendar Report
year)
EEA's 10th Quarterly
Report
MOBILES
(1977 & 1981 R.L. Polk)
MOBILES (General Motors
Research (GMR), derived
from 1979 National
Purchase Dairy Research,
Inc. (NPD))
MOBILE3; Diesel
Particulate Study (DPS)
(Jack Faucett; Data
Resources)
FHwA (1982)
EEA's 10th Quarterly
Report
MOBILES (1978 & 1981
R. L. Polk)
MOBILES (1977 Truck
Inventory and Use Survey
(TIUS))
Same sources as LDV
EEA's 10th Quarterly
Report
MOBILES (1972 & 1977
TIUS)
MOBILES (1977 TIUS)
EEA's 10th Quarterly
Urban fraction of VMT
(by model year)
Heavy-Duty Conversion
Factor Analysis (1977
TIUS)
-------
Basis for EEA's Estimates of Historic Nationwide VMT
1. Obtain total cars and total trucks in use from R.L.
Polk* registration data.
2. Obtain gas/diesel, domestic/imported, light/heavy
truck splits by applying scrappage rates (found in
the literature) to actual model year sales.
3. Obtain vehicle mileage-accumulation rates in a base
year from NPD** for light-duty vehicles and trucks,
and TIUS (77) for heavy-duty trucks.
4. Use Federal Highway data in Table VM-1 to get change
in average annual mileage-accumulation rates. No
other data from Table VM-1 used.
* R.L. Polk registration data: vehicles in use as of July
1; adjustments made for scrappage and duplicates.
** NPD: National Panel Diary by NPD Research, Inc., Port
Jefferson, New York.
- a diary panel survey of 5000 families.
- data collected contains information on 1) fuel
purchases: date, gallons, type of fuel, total costs;
2) vehicle type; and 3) miles travelled.
- average length of participation in survey is 10
months.
-------
Annual Urban VMT Growth Rates
Used in NOx Projections
The table below lists the estimated percentage growth in
VMT from base-year (1981) to year of projection, for each
of the vehicle classes.
These growth rates are based on EEA's nationwide VMT
estimates (Tenth Quarterly Fuel Consumption Model),
separated into gasoline and diesel fractions and converted
to urban VMT using the Nationwide Approach (explained in
detail in this Appendix).
The following growth rates were used in future NOx
projections with all three approaches (NEDS, nationwide,
and local); they were also used to convert some 1980 local
VMT data into 1981 figures.
Annual Fleet Urban VMT Growth Rates
(Percent Change from 1981 Base Year)
Year
LDGV
LDOV
LDV
LDGT
LOOT
LOT
HDGV
HDDV
1982 -1.76
1983 -1.01
1984 -0.38
1985 0.06
1986 0.36
1987 0.60
1988 0.79
1989 0.95
1990 1.05
1991 1.15
1992
1993
1994
1995
1996
1997
1998
1999
2000
1.23
1.30
1-35
1.39
1.42
1.43
1.43
1.43
1.42
29.16
25-67
24.19
22.96
21.93
21.05
20.36
19-81
19.36
18.76
18.06
17.31
16.57
15.85
15.14
14.45
13.80
13.19
12.62
-1.24
-0.50
0.14
0.60
0.91
1.17
1.38
1.56
1.69
1.81
1.89
1.96
2.00
2.03
2.04
2.03
2.01
1.99
1.96
-0.41
0.38
1.27
1.69
1.91
2.04
2.10
2.09
2.03
1-99
1.96
1.92
1.90
1.87
1.85
1.83
1.82
1.80
1.79
56.41
51-25
48.74
46.26
43.42
40.95
38.83
36.90
35.08
33-28
31.56
29-95
28.48
27.14
25.86
24.67
23-54
22.49
21.51
0.46 0.33
1.35 0.52
2.38 1
2.95 1
3.28 1
3.53 l
3-72 i
3-83
3.89
3.94
3.96
3-96 I
1.14
1.45
1.57
1.64
1.75
1.83
1.93
1.94
1.98
J.03
3.96 2.04
3-95 :
2.07
3-92 2.09
3.89 2.09
3.84 2.08
3.78 2.07
3.72 2.06
5-07
4.08
4.92
5.66
6.18
6.56
6.79
6.88
6.90
6.87
6.83
6.79
6.73
6.68
6.61
6.53
6.43
6.32
6.21
Source: "Rollback II With Variable Mobile Source Growth Rates,"
Memorandum from Mark Wolcott, TEB, to Charles L. Gray,
ECTD, October 29, 1984.
-------
Local Methodologies and Calculations
Chicago, Illinois
The estimate of total VMT for the Chicago SMSA was
determined as follows. The Illinois Department of
Transportation (IDOT) provided data on vehicle miles travelled
in Northeastern Illinois based on characteristics of highways in
this area and in Illinois as a whole. The values for 1980 and
1981 are:
Average Weekday
VMT (IDOT)
1980 96014225
1981 97848249
The Chicago Area Transportation Study (CATS), as part of
its network modelling, made adjustments to the 1980 value based
upon traffic counts made by themselves and IDOT, and to account
for non-network traffic. This resulted in a value for 1980 of
113,199,890.[2] The same adjustment factor was then applied
here, as well as a factor of 365 days per year, to derive an
adjusted total VMT for 1981 of 42.1 billion miles.
The sources for the above data were continuous machine
counts taken at 165 non-expressway sites and an undetermined
number of expressway sites. Visual vehicle classification
counts were taken at all of the 165 non-expressway sites at one
time. At selected expressway sites, approximately ten at a
time, visual vehicle classification counts have been made more
frequently, although not at the same sites each time. [3] The
following four vehicle classes were defined at each site and
then broken into the eight vehicle classes used by the MOBILE
model:[4]
(1) passenger cars;
(2) light trucks - 4 tires;
(3) medium trucks - 3 axle or 6 tires; and
(4) heavy trucks - all others.
[1] Illinois Environmental Protection Agency, Division of Air
Pollution Control, Proposed Revision to the State
Implementation Plan for Ozone, Chicago, 23 December 1983.
[2] Chicago Air Transportation Study 1982 State Implementation
Plan Submittal for Northeastern Illinois, Chicago, 24 June
1982.
[3] Chicago Area Transportation Study, Conversation with Roy
Veil, 30 October 1984.
[4] Ibid.
-------
The local VMT breakdown used here was that for 1982 as
provided by the Illinois EPA. Combined with the 1981 total VMT
estimates, the following estimate of VMT by vehicle class was
obtained:
1981 VMT
VMT %[6] (millions)
LDGV 86.6 36,465
LDGT, 02.8 1,179
LDGT2 01.6 674
HDGT 02.7 1,137
LDDV 02.7 1,137
LDDT 00.2 84
HDDV 02.4 1,011
MC 01.0 421
Using the MOBILES emission factors, NOx emissions for each
vehicle class were calculated. The LDV and LDT classes were
created by summing their subclasses, and the MC VMT and
emissions were redistributed based upon VMT share. The final
inventories are presented in Table A-l.
[5] Op Cit, 1982 SIP Submittal.
[6] Illinois EPA, Division of Air Pollution Control, The
Effect of U.S. EPA Proposals to Relax Vehicle Emission
Standards on Chicago Air Quality, Springfield, Illinois,
October 1981.
-------
Table A-l
1981 Annual Comparisons
Chicago
VMT, Millions (%)
Localized Approach Nationwide Approach*
LDV 37984.0 (90.2)
LDT 1979.2 (4.7)
HDGV 1137.0 (2.7)
HDDV 1010.6 (2.4)
Total 42110.8 (100.0)**
34025.5 (80.8)
5811.3 (13.8)
1431.8 (3.4)
842.2 (2.0)
42110.8 (100.0)
NEDS Approach
28989.4 (81.2)
3959.5 (11.1)
720.9 (2.0)
2043.9 (5.7)
35713.7 (100.0
NOx Emissions, 1000 Tons (%)
Localized Approach Nationwide Approach NEDS Approach***
101.64 (64.4)
24.15 (15.3)
10.01 (6.3)
22.00 (13.9)
157.80 (99.9)
LDV 111.81 (72.6)
LDT 7.86 (5.1)
HDGV 7.95 (5.2)
HDDV 26.41 (17.1)
Total 154.03 (100.0)
86.60 (53.6)
16.45 (10.2)
5.04 (3.1)
53.39 (33.1)
161.48 (100.0)
* Nationwide VMT calculated using local VMT total, and
breakdown of urban VMT (by vehicle class) from the
nationwide approach.
** Before adjustment for unaccounted traffic was made by
Chicaco Area Transportation Study, this value was 35715.0,
or essentially the same figure estimated by the NEDS
approach.
*** NEDS emissions inventory updated with MOBILES NOx emission
factors (original NEDS based on MOBILE2.5).
-------
Detroit, Michigan
Local total VMT data were only available for five of the
six counties that comprise the Detroit SMSA: Livingston,
Macomb, Oakland, St. Glair, and Wayne. No data was available
for Lapeer County, so the 1981 VMT from NEDS was used for this
county instead. The data for the other counties was obtained
from the Southeastern Michigan Council of Governments
(SEMCOG).[1]
SEMCOG obtained their 1981 total VMT estimates by taking a
base year (1980) estimate of travel and applying growth rates,
by county, derived from the U.T.P.S. network model. The base
year data was obtained from historical vehicle counts along
many of the 7500 coded links in SEMCOG's jurisdiction, and
another series of vehicle counts made to improve traffic
estimation for air quality purposes.[2]
The estimate of total VMT for the combined counties of
Livingston, Macomb, Oakland, St. Clair, and Wayne equalled
81,361,000 miles/day.[3] A factor of 365 was applied here to
yield an annual VMT of 29.7 billion miles. To this, the NEDS
value for Lapeer County of 0.54 billion miles[4] was added to
arrive at a total 1981 VMT for the Detroit SMSA of 30.2 billion
miles.
The local estimates for VMT breakdown were also obtained
from SEMCOG. The primary source for these estimates was the
1980 Regional Vehicle Classification and Occupancy Study.
One-hundred and fifteen sites were intended to be surveyed, but
only 43 of these were used in determining a breakdown by
vehicle type.[5]
At each site the following vehicles were classified[6]:
(1) passenger car;
(2) passenger/commercial van;
(3) passenger/commercial light-duty truck (i.e.,
Ford Bronco);
[1]
[2]
[3]
[4]
[5]
[6]
SEMCOG, meeting with Chuck Hersey and Adeile Nwanko, 25
October 1984.
SEMCOG, Procedures Used in Generating the Mobile Source
Emission Inventory, Detroit, MI, June 1984, p. 27-8.
Ibid, p. 42.
NEDS Area Source Reports.
Op Cit. SEMCOG Meeting
SEMCOG, Regional Vehicle Classification and Occupancy
Study, Detroit, MI, November 1982, revised January 1983,
p. 18
-------
(4)
(5)
(6)
The sites
roadway:[7]
heavy-duty trucks: large single-unit trucks;
semi-trucks: large tractor-trailer combination
vehicles;
motorcycles
were located along 5 different types of
(1)
(2)
(3)
(4)
(5)
urban
urban
urban
rural
rural
freeway;
arterial
arterial
freeway;
arterial
, major
, minor
and
.
The sites were surveyed during three different time
periods:[8]
(1) 7:00 a.m. to 9:00 a.m.;
(2) 11:00 a.m. to l:00p.m.; and
(3) 2:00 p.m. to 6:00 p.m.
In conjunction with the network traffic modelling, and a
few assumptions, regional VMT percentages were derived for the
eight vehicle classes used in the MOBILE models. A few of the
key assumptions are stated below.
(1) Diesel penetration occurs within
i.e., an LDDV will replace a LDGV
LDGT; [ 9 ]
(2) VMT is proportional to registration
purposes of splitting gas and
vehicles;[10]
(3) VMT percentage by trip type remains constant
over time.[11]
classes;
and not a
for the
diesel
[7]
[8]
[9]
[10]
[11]
[12]
Op Cit.,
Op Cit. ,
Op Cit. ,
Ibid, p
Ibid, p,
Ibid, p.
"SEMCOG Meeting
Regional Vehicle.
Procedures,
60
62
63
12
17
-------
These assumptions yield the following breakdown of
VMT by vehicle type for the Detroit SMSA.
1981
LDGV
LDGT2
HDGV
LDDV
LDDT
HDDV "
MC
VMT %[12]
81.9
11.8
0.2
1.5
1.4
0. 1
2.6
0. 5
VMT (millions)
24,767.0
3,568.4
60.5
453.6
423.4
30.2
786.3
151.2
This VMT was then used with MOBILE3 emission factors to
determine the 1981 NOx inventory. The LDVs and LDTs were
summed across their subclasses, and the MC emissions and VMT
were redistributed based upon VMT percentage. The final
inventories are presented in Table A-2.
-------
Table A-2
1981 Annual Comparisons
Detroit
VMT, Millions (%)
Localized Approach Nationwide Approach*
LDV 25317.0 (83.7)
LOT 3677.4 (12.2)
HDGV 455.9 (1.5)
HDDV 790.2 (2.6)
Total 30240.5 (100.0)
24434.3 (80.8)
4173.2 (13.8)
1028.2 (3.4)
604.8 (2.0)
30240.5 (100.0)
NEDS Approach
20159.2 (80.1)
3619.3 (14.4)
548.2 (2.2)
855.4 (3.4)
25182.1 (100.1)
NOx Emissions, 1000 Tons (%)
Localized Approach Nationwide Approach NEDS Approach**
72.99 (64.4)
17.34 (15.3)
7.18 (6.3)
15.80 (13.9)
113.31 (99.9)
LDV 75.29 (67.1)
LDT 13.28 (11.8)
HDGV 3.17 (2.8)
HDDV 20.54 (18.3)
Total 112.28 (100.0)
60.20 (59.4)
15.04 (14.8)
3.80 (3.8)
22.30 (22.0)
101.34 (100.0)
Nationwide VMT calculated using local VMT total, and
breakdown of urban VMT (by vehicle class) from the
nationwide approach.
NEDS emissions inventory updated with MOBILE3 NOx emission
factors (original NEDS based on MOBILE2.5).
-------
Nashville, Tennessee
The VMT breakdown by vehicle classification in
Nashville-Davidson County was determined by visual counts at
eight locations throughout the county. These locations include
one interstate, five primary arterials, one minor arterial, and
one airport entrance. The vehicles observed were put into the
following classifications:
(1) passenger cars (standard and small),
(2) motorcycles,
(3) buses (commercial and school),
(4) single-unit trucks (single unit pickup or
panel, single rear-tire, dual rear-tire), and
(5) tractor-trailers (3-axle, 4-axle, semi,
etc. ).[l]
The sample counts are performed by the Tennessee
Department of Transportation (TDOT). They are performed once a
year for a continuous six hour period. The sites were
originally picked randomly and they remain the same every
year. This is not a statistically determined sample, and TDOT
does nothing beyond providing the data for each site.[2]
TDOT also makes counts of total VMT at 381 traffic count
stations in Davidson County. These 381 stations cover five
functional classifications: interstate, expressway, primary
arterial, minor arterial, and collector. Local roads, not
counted, are assumed to account for 8 percent of total traffic
in Davidson County.[3]
These traffic count data are then applied by the
Comprehensive Planning Division of the Metropolitan Government
of Nashville and Davidson County. The total VMT data are first
summed to provide totals by functional road classification.[4]
The VMT breakdown by class is then determined for each
functional classification from the TDOT data using the
following assumptions:
[1] Conversations with Rich Walpole, TDOT Transportation
Mapping Division (9/17/84) and Bucky Crowell, Nashville
Metro Government, Comprehensive Planning Division
(10/22/84).
[2] Ibid.
[3] Metropolitan Planning Commission Memo, 1981
Nashville-Davidson County, Tennessee Mobile Source
Emissions Inventory and Documentation of Methodology,
Bucky Crowell (Comprehensive Planning Division) to Paul
Bontrager (Metropolitan Health Department), June 24, 1982.
[4] Ibid.
-------
(1) single-unit trucks comprise all of the
light-duty trucks,
(2) buses and tractor-trailers comprise all of the
heavy-duty trucks, and
(3) the VMT breakdown for collector roads is the
same as that for minor arterials.
The VMT breakdowns for freeways and local roads came from
older TDOT surveys that are not updated annually. [5] This
makes these figures less certain, but these two functional
classifications account for only 12.8 percent of the total
VMT. Finally) the splits for gas vs. diesel and light-duty
truck 1 and 2 were determined from MOBILE2 default estimates
for vehicle mix. [6] For the year 1981, the 1980 VMT breakdown
for minor arterials, collectors and local streets was adjusted
to reflect the changes in MOBILE2 default values from 1980 to
1981.[7] The results are shown below.
1981 Nashville VMT Breakdown[8]
Road/ LDGV LDGT LDGT HDGV LDDV LDDT HDDV MC
Vehicle Type 1 2
866
854
854
.044
.040
.040
.043
.039
.039
.002
.014
.014
.009
.009
.009
.028
.025
.025
.002
.014
.014
. 006
.005
.005
Local
Collector
Minor
Arterial
Primary .750 .141 .079 .010 .008 .001 .008 .003
Arterial
Expressway .792 .117 .066 .006 .009 .001 .005 .004
Interstate .680 .142 .080 .048 .008 .002 .038 .002
1981 Nashville Daily VMT[9]
Local 750,831
Collector 372,423
Min. Arterial 501,178
Prim. Arterial 3,526,835
Expressway 447,699
Interstate 3,786,401
Total 9,385,365
[5] Op Cit. Phone Conversations.
[6] Op Cit. Metropolitan Planning Commission Memo.
[7] Ibid.
[8] Ibid.
[9] Ibid.
-------
From the tables of 1981 Vehicle Mix by Functional
Classification and Vehicle Type, and Daily Vehicle Miles of
Travel by Functional Roadway Class (above), vehicle miles of
travel by vehicle type were computed. The only assumption made
was that Daily Vehicle Miles of Travel X 365 would equal Annual
Vehicle Miles of Travel. These are the figures, therefore,
that are used in all subsequent calculations involving
Nashville-Davidson County. The final inventories are presented
in Table A-3.
-------
Table A-3
1981 Annual Comparisons
Nashville
VMT, Millions (%)
Localized Approach Nationwide Approach*
LDV 4827.3 (75.3)
LDT 1295.0 (20.2)
HDGV 160.3 (2.5)
HDDV 128.2 (2.0)
Total 6410.8 (100.0)
5179.9 (80.8)
884.7 (13.8)
218.0 (3.4)
128.2 (2.0)
6410.8 (100.0)
NEDS Approach
4476.1 (68.9)
1219.7 (18.8)
186.9 (2.9)
612.1 (9.4)
6494.8 (100.0)
NOx Emissions, 1000 Tons (%)
Local
LDV
LDT
HDGV
HDDV
Total
14
5
1
3
24
ized Approach
.42
.38
. 12
.35
.27
(59
(22
(4
(13
(100
.4)
.2)
.6)
.8)
.0)
Nationwide Approach
15
3
1
3
24
.47
.68
.52
.35
.02
(64
(15
(6
(13
(99
.4)
.3)
.3)
.9)
.9)
* *
13.37 (37.4)
5.07 (14.2)
1.33 (3.7)
16.00 (44.7)
35.77 (100.0)
Nationwide VMT calculated using local VMT total, and
breakdown of urban VMT (by vehicle class) from the
nationwide approach.
NEDS emissions inventory updated with MOBILES NOx emission
factors (original NEDS based on MOBILE2.5).
-------
New York City
The New York City SMSA consists of eight counties in two
states. Localized VMT estimates for each county had to be
compiled from a number of sources, which are described below.
The following estimates of 1980 daily VMT by vehicle class
for the counties of Bronx, New York, Queens, Rockland,
Richmond, and Westchester were obtained from the New York
Metropolitan Transportation Council (MTC).[1]
1980 Daily VMT
1980 Annual VMT (millions)
LDV 52133900 19029.1 (88.5%)
LDGT 3786500 1382.1 ( 6.4%)
HDGV 2073200 756.7 ( 3.5%)
HDDV 885200 323.1 ( 1.5%)
The total VMT for these counties was derived using the
Highway Evaluation Model, a network model. This model,
originally calibrated in 1963 based upon a tri-state home
interview and a truck survey, has been updated using 1970
Census Data, 1978 Taxi Survey data, and updated VMT estimates
in 1975. These 1975 updates involved at least one 48-hour
count on each link in the arterial and expressway networks.[2]
The breakdown by vehicle class was obtained from various
surveys. For the two suburban counties, Westchester and
Rockland, the mix was based on a 1975 traffic count survey on
the Long Island Expressway. [3] For Manhattan and the rest of
New York City, the mix was based upon 1973 and 1979 bridge and
tunnel counts, respectively.[4] The classes were further
subdivided using 1981 registration data from the New York State
Department of Motor Vehicles.[5]
An estimate of 16,927,000 for the 1980 daily VMT in Bergen
County, New Jersey, was obtained from the New Jersey DOT.[6]
This was converted to an annual VMT by multiplying by 365. The
1981 annual VMT estimate for Putnam County, New York, of 749
million was taken from NEDS.[7]
[1] New York Metropolitan Transportation Council, letter from
David Jordan, Assoc. Transp. Analyst, 28 September 1984.
[2] Tri-State Regional Planning Commission, 1982 SIP Revision,
New York, March 1982, pp. 1,7.
[3] Ibid, p. 19.
[4] Ibid, p. 35.
[5] Ibid.
[6] Ibid, p. 30
[7] NEDS area source reports.
-------
To convert these estimates into a 1981 SMSA estimate of
VMT by vehicle class, several steps were taken:
(1) The 1980 total for Bergen County was broken down by
class using the breakdown for the six New York
counties.
(2) The 1980 VMTs for each class were updated to 1981
using growth factors, by vehicle class, derived from
the Nationwide Method (shown earlier in the
Appendix).
(3) A new 1981 VMT breakdown was established.
(4) The 1981 total for Putnam County was split between
classes using the new 1981 breakdown for the other
counties.
(5) The class-specific VMTs for all of the counties were
totalled.
This resulted in the following 1981 VMT breakdown by
vehicle class:
Total 1981 VMT
LDV 25155.6 (87.9)
LDGT 1985.7 ( 6.9)
HDGV 1050.0 ( 3.6)
HDDV 444.2 ( 1.5)
These results, with the MOBILES emission factors, were
used to determine the mobile source NOx inventory. The final
inventories are presented in Table A-4.
-------
Table A-4
1981 Annual Comparisons
New York City
VMT, Millions (%)
Localized Approach Nationwide Approach*
LDV 25155.6 (88.0)
LDT 1985.7 (6.9)
HDGV 1050.0 (3.6)
HDDV 444.2 (1.5)
Total 28635.5 (100.0)
23137.5 (80.8)
3951.7 (13.8)
973.6 (3.4)
572.7 (2.0)
28635.5 (100.0)
NEDS Approach
25523.3 (87.5)
2157.2 (7.4)
633.1 (2.2)
854.1 (2.9)
29167.7 (100.0)
* *
NOx Emissions, 1000 Tons (%)
NEDS Approach**
76.24 (68,1)
8.96 (8.0)
4.42 (4.0)
22.31 (19.9)
111.93 (100.0)
Nationwide VMT calculated using local VMT total, and
breakdown of urban VMT (by vehicle class) from the
nationwide approach.
NEDS emissions inventory updated with MOBILES NOx emission
factors (original NEDS based on MOBILE2.5).
LDV
LDT
HDGV
HDDV
Total
Localized Approach
75.15 (73.4)
8.25 (8.1)
7.34 (7.2)
11.60 (11.3)
102.34 (100.0)
Nationwide Approach
69. 12 (64 .4)
16.43 (15.3)
6.81 (6.3)
14 .96 (13 . 9)
107.32 (99.9)
-------
Philadelphia
The estimates of total VMT and the VMT breakdown by
vehicle class for the Philadelphia SMSA were provided by the
Delaware Valley- Regional Planning Commission (DVRPC). The
DVRPC has used these estimates in the 1982 SIPs for
Pennsylvania and New Jersey.
Total VMT by county is obtained from a transportation
modelling system to estimate current and future travel. This
system is equivalent to the network models described before.
The model estimates for 1979, 1980, and 1987 are as follow:[1]
Year
Daily VMT
1979 59,250,000
1980 61,821,000
1987 65,951,000
In order to obtain an estimate for 1981 total WIT, a
linear extrapolation between 1980 and 1987 could have been
used. This method was rejected as being inaccurate, however,
based upon the economic conditions known to exist. The
following method was used instead.
1981
VMT Phil.
Where:
198°
Phil.
.1981 VMT Nation.
*1980 VMT Nation'
1987 VMT Phil.
1980 VMT Phil. ,
1987 VMT Nation'
1980 VMT Nation
(X VMT) Phil. = total VMT in year X for the Phil. SMSA.
(X VMT) Nation = total VMT in year X for nationwide urban
from the nationwide method.
(1980 VMT) nation = 840,640,000,000
(1981 VMT) nation = 857,560,000,000
(1987 VMT) nation = 945,930,000,000
The result is a 1981 daily VMT for the Philadelphia SMSA
of 59,789,773 and a 1981 total VMT of 21.8 billion miles.
The VMT breakdown by vehicle class was obtained from a
1980 vehicle occupancy and vehicle mix monitoring program
performed by the DVRPC. The intent of the study was to
establish regional and more disaggregate estimates for light
[1]Delaware Valley Regional Planning Commission, Delaware
Valley Highway Emissions Inventory, Philadelphia,PA,
September 1982.
-------
and heavy truck percentages with a confidence level of 95
percent, and average regional passenger vehicle occupancy with
a confidence level of 65 percent.[2]
The final survey consisted of 112 monitoring sites
stratified by functional classification of roadway (freeways
and expressways, principal arterials, minor arterials, and
collectors), and area type (central business district, urban,
suburban, and rural). The individual sites were surveyed on
days from 18 March until 30 June 1980, chosen at random,
excluding holidays and weekends. At most stations, automatic
machine counts were taken in conjunction with the visual
counts. The visual counts were taken from 7:00 a.m. to 3:00
p.m. the day of the count.
The vehicles classified were as follow:
(1) passenger vehicles;
(2) light trucks - pickup or panel, single
rear-tires;
(3) heavy trucks - dual rear-tires, 3 axles, 4 or
more axles, all semi or full trailers;
(4) buses; and
(5) motorcycles.
The results from the vehicle classification survey were
used by the DVRPC in the determination of their highway
emissions inventory. HDGV and HDDV VMT was split using the
product of state vehicle registrations and mileage accumulation
rates from the 1977 TIUS for Pennsylvania and New Jersey
combined. This resulted in the diesel share of HDV VMT being
65 percent. The light-duty trucks were split into the two
weight classes based upon national sales statistics by model
year. This resulted in a two-to-one split for LDTi over
LDT2.[3]
The final VMTs and the breakdown by class were as follow:
VMT %[4] VMT (millions)
LDV 82.0 17895.0
LDT, 7.2 1571.3
LDT2 3.6 785.6
HDGV 2.5 545.6
HDDV 4.7 1025.7
[2] Delaware Valley Regional Planning Commission, Automobile
Occupancy and Truck Travel Monitoring for Transportation
Air Quality Planning, Philadelphia, PA, February 1981.
[3] Op Cit. Highway Emissions Inventory
[4] Ibid, and conversation with Ron Roggenburk, DVRPC, October
1984.
-------
These values, combined with the properly weighted MOBILE3
emission factors were used to generate the mobile source NOx
emission inventory for the Philadelphia SMSA. The final
inventories are presented in Table A-5.
-------
Table A-5
1981 Annual Comparisons
Philadelphia
VMT, Millions (%)
Localized Approach Nationwide Approach*
17631.0 (80.8)
3011.2 (13.8)
741.9 (3.4)
436.4 (2.0)
LDV 18002.0 (82.5)
LOT 2312.9 (10.6)
HDGV 523.7 (2.4)
HDDV 981.9 (4.5)
Total 21820.5 (100.0)
21820.5 (100.0)
NEDS Approach
19033.0 (81.0)
2707.6 (11.5)
547.9 (2.3)
1220.1 (5.2)
23508.6 (100.0)
* *
NOx Emissions, 1000 Tons (%)
NEDS Approach**
56.86 (54.8)
11.25 (10.8)
3.83 (3.7)
31.87 (30.7)
103.81 (100.0)
Nationwide VMT calculated using local VMT total, and
breakdown of urban VMT (by vehicle class) from the
nationwide approach.
NEDS emissions inventory updated with MOBILE3 NOx emission
factors (original NEDS based on MOBILE2.5).
Localized Approach
LDV
LOT
HDGV
HDDV
Total
53
9
3
25
92
.78
.37
.66
.66
.47
(58.
(10.
(4.
(27.
(100.
2)
1)
0)
7)
0)
Nationwide Approach
52
12
5
11
81
.67
. 51
. 18
. 40
. 76
(64
(15
(6
(13
(99
• 4)
.3)
.3)
.9)
.9)
-------
Reno, Nevada
The estimates for daily VMT for the Reno SMSA (Washoe
County) were obtained from the Reno Regional Transportation
Commission. The breakdown of VMT by vehicle class was provided
by the Nevada Department of Transportation to the Reno Regional
Transportation Council.
The Reno Regional Transportation Commission uses the
U.T.P.S. network modelling system to estimate daily VMT for the
Washoe County area. The models have been calibrated and
validated twice by traffic counts, but the current estimates
are direct outputs of the model.[1]
The 1983 estimates for daily VMT by functional
classification of roadway as provided are as follow:
Roadway 1983 Daily VMT[2]
Local 260,710
Collector 188,710
Minor Arterial 1,550,560
Major Arterial 396,220
Freeway 1,183,780
The Nevada Department of Transportation (NDOT) does
periodic vehicle classification surveys from which a vehicle
mix can be determined.[3] The urban survey is for statewide
urban, which also includes Las Vegas, so the resulting
estimates are not specific to Reno.
The surveys were performed in 1978 on the five functional
classifications of roadways, with the larger roadways being
surveyed more frequently. The number of sites surveyed has
varied from year to year, and the results are reported only by
functional class and not by site. Freeway sites are monitored
for a minimum of twenty-four hours and the other road classes
for a minimum of eighteen hours. Some adjustment is made for
the day of the week. The vehicles classified at each site are:
1) Cars
2) Small Cars
3) Pickups
4-31) Trucks in detail
[1] Reno Regional Transportation Council, Conversations with
Tom Brinkman, October 1984.
[2] Ibid.
[3] Nevada Department of Transportation, conversations with
Dan Gross, October 1984.
-------
These 31 vehicle types were reassigned into the eight
vehicle classes used by the MOBILE models under the following
assumptions:[4]
LDGV = .97* (cars + small cars)
LDGT!= .73* (pickups)
LDGT 2 = .25* (pickups)
HDGT = all 2-axle single unit dual
tire trucks + all 2 axle
buses
LDDV = .02* (cars + small cars)
LDDT = .02* (pickups)
HDDT = all 3-axle single unit dual
tire and larger trucks + all
3-axle buses
MC = .01* (cars + small cars)
The factors for diesel percentage and light duty truck
size were obtained from many methods, including parking lot
surveys of parked vehicles and are updated periodically.[5]
The results (for 1978, with diesel updates) are summarized
in the matrix below:[6]
Road Type LDGV LDGT LDGT HDGV LDDV LDDT HDDV
Vehicle Type 1 2
Local .846 .092 .031 .011 .017 .003 .000
Collector .780 .133 .045 .017 .061 .004 .005
Minor Arterial .778 .133 .045 .018 .016 .004 .006
Major Arterial .779 .132 .044 .022 .016 .004 .012
Freeway .704 .121 .040 .028 .014 .003 .090
Total VMT for each vehicle type was calculated for this
report by multiplying the percentage of each vehicle type in
each functional road classification by the total VMT for that
functional classification and summarizing by vehicle type.
This yields 1983 daily VMT by vehicle type as:
1983 Daily VMT
LDGV 2712561
LDGT! 450847
LDGT2 151134
HDGV 75849
LDDV 55173
LDDT 12875
HDDV 121542
[4,5] Ibid.
[6] Op Cit, Tom Brinkman
-------
In order to convert these into 1981 annual estimates,
growth rates for urban VMT for each vehicle class, as
determined from the nationwide method are used.[7] These
results are multiplied by 365 days/year and yield:
1981 Annual VMT
LDGV
LDGT,
LDGT2
HDGV
LDDV
LDDT
HDGV
1,010,411,300
158,075,330
57,354,634
27,395,594
12,747,560
2,044,748
40,957,548
Using the above estimates of 1981 annual VMT and MOBILES
high altitude emission factors, the NOx emission inventory was
calculated. The LDV and LDT estimates were then calculated
simply by summing up their subcategories. These final results
are presented in Table A-6.
[7] The nationwide method is described earlier in this
Appendix.
-------
Table A-6
1981 Annual Comparisons
Reno
VMT, Millions (%)
Localized Approach Nationwide Approach*
LDV 1023.2 (78.2)
LDT 217.5 (16.6)
HDGV 27.4 (2.1)
HDDV 41.0 (3.1)
Total 1309.1 (100.0)
1057.8 (80.8)
180.7 (13.8)
44. 5 (3.4)
26.2 (2.0)
1309.2 (100.0)
NEDS Approach
887.9 (60.7)
429.7 (29.4)
22.1 (1.5)
123.3 (8.4)
1463.0 (100.0)
NOx Emissions, 1000 Tons (%)
NEDS Approach**
1.85 (29.4)
1. 12 (17.8)
0.11 (1.7)
3.22 (51.1)
6.30 (100.0)
Nationwide VMT calculated using local VMT total, and
breakdown of urban VMT (by vehicle class) from the
nationwide approach.
NEDS emissions inventory updated with MOBILE3 NOx emission
factors (original NEDS based on MOBILE2.5).
Localized Approach
LDV
LDT
HDGV
HDDV
Total
2
0
0
1
3
. 14
.61
. 13
.07
.95
(54
(15
(3
(27
(100
.2)
.4)
.3)
.1).
.0)
Nationwide Approach
2
0
0
0
3
.21
.52
.21
.68
.62
(61
(14
(5
(18
(100
.0)
.4)
.8)
.8)
.0)
-------
Seattle, Washinqton[1]
The estimates for total VMT and VMT breakdown by vehicle
class for Seattle, Washington, were obtained from the
Washington State Department of Transportation.
Total VMT for the two counties that make up the
Seattle-Everett SMSA, King and Snohomish, are estimated using
the Urban Transportation Planning System (U.T.P.S.) network
modelling system. Some calibration is done with traffic counts.
The VMT 'breakdown by class was obtained from the State
Energy Office, which was only able to provide a VMT breakdown
for the entire state of Washington. The vehicle mix was
determined from statewide vehicle registration and applying
mileage accumulations by model year from AP-42 (March 1981).
No visual counts were known to have been made. The figures
that were provided are as follow:
1981 Calendar Year VMT by County:
King: 8,871,343,000
Snohomish: 2,080,452,000
Statewide Vehicle Mix:
Vehicle Type VMT % VMT (millions)
LDGV 72.9% 7983.9
LDGT1 7.7% 843.3
LDGT2 4.6% 503.8
HDGV 4.1% 460.0
LDDV 5.7% 624.3
LDDT . 0.7% 76.7
HDDV 3.3% 361.4
MC 0.9% 98.6
The county VMTs were summed and the statewide vehicle mix
was used to generate VMT by vehicle class. MOBILES emission
factors were then applied to obtain NOx emissions. LDGV and
LDDV were combined to form LDV, and the same was done with
LDGTi , LDGT2, and LDDT to form LDT. MC VMT and emissions
were redistributed to the other vehicle classes based upon VMT
share. The final inventories are presented in Table A-7.
[1] Phone conversations with Dave Kircher, Project
Administrator, Puget Sound Air Pollution Administration,
October 1984.
-------
Table A-7
1981 Annual Comparison
Seattle
VMT, Millions (%)
Localized Approach Nationwide Approach*
LDV 8613.5. (79.3)
LDT 1424.6 (13.1)
HDGV 460.3 (4.3)
HDDV 361.6 (3.3)
Total 10860.0 (100.0)
8774.9 (80.8)
1498.7 (13.8)
369.2 (3.4)
217.2 (2.0)
10860.0 (100.0)
NEDS Approach
8412.6 (73.5)
2431.8 (21.3)
117.9 (1.0)
478.5 (4.2)
11440.8 (100.0)
NOx Emissions, 1000 Tons (%)
NEDS Approach**
25.13 (51.8)
10.11 (20.8)
0.82 (1.7)
12.50 (25.7)
48.56 (100.0)
Nationwide VMT calculated using local VMT total, and
breakdown of urban VMT (by vehicle class) from the
nationwide approach.
NEDS emissions inventory updated with MOBILES NOx emission
factors (original NEDS based on MOBILE2.5).
Localized Approach
LDV
LDT
HDGV
HDDV
Total
24
5
3
9
43
.87
.68
.19
.36
.10
(57.
(13.
(7.
(21.
(100
7)
2)
4)
7)
.0)
Nationwide Approach
26
6
2
5
40
.21
.23
.58
.67
.69
(64.
(15.
(6.
(13.
(99.
4)
3)
3)
9)
9)
-------
Tucson, Arizona
Estimates of daily VMT and vehicle mix for Tucson, Arizona
were obtained from the Arizona DOT via the Pima County
Association of Governments.[1] The data that was provided is
as follows:
1981 Daily VMT: 10,156,000 miles
Vehicle Type Freeways Arterial
LDV 72.6 69.9
LDT 18.0 26.0
HDG 7.5 2.2
HDD 1.3 1.2
MC 0.6 0.6
In order to obtain a regional breakdown, the freeway and
arterial breakdown were combined using a weighting factor of
.31 for freeways and .69 for arterials, derived from data in
the 1982-83 Statistical Abstract on nationwide urban VMT.[2]
This regional breakdown was then used to determine 1981 daily
VMT by vehicle class, and then multiplied by 365 to yield 1981
VMT.
VMT % 1981 Daily (millions) 1981 Annual (millions)
LDV 70.9 7.201 2628.2
LDT 23.5 2.387 871.1
HDV 3.9 0.396 144.6
HDV 1.2 0.122 44.5
MC 0.6 0.061 22.2
These VMT values were then used to determine 1981 NOx
emissions using MOBILES emission factors. The emissions and
VMT by motorcycles were redistributed by VMT share. The final
inventories are presented in Table A-8.
[1] Pima County Association of Governments, conversation with
Nick Buccholz, October, 1984.
[2] U.S. Department of Commerce, Bureau of the Census
Statistical Abstract of the United States 1982-83.
-------
Table A-8
1981 Annual Comparisons
Tucson
VMT, Millions (%)
Localized Approach Nationwide Approach*
LDV 2643.0. (71.3)
LDT 874.8 (23.6)
HDGV 144.6 (3.9)
HDDV 44.5 (1.2)
Total 3706.9 (100.0)
2995.2 (80.8)
511.6 (13.8)
126.0 (3.4)
74. 1 (2.0)
3706.9 (100.0)
NEDS Approach
2547.0 (70.9)
870.0 (24.2)
46.1 (1.3)
131.0 (3.7)
3594.1 (100.1)
NOx Emissions, 1000 Tons (%)
Local
LDV
LDT
HDGV
HDDV
Total
7.
3.
1.
1 .
13.
ized Approach
86
62
01
16
65
(57
(26
(7
(8
(100
.6)
.5)
.4)
.5)
.0)
Nationwide Approach
8.
2.
0.
1.
13.
95
13
88
94
90
(64
(15
(6
(13
(99 .
.4)
.3)
.3)
.9)
9)
7.61 (50.8)
3.62 (24.2)
0.32 (2.2)
3.42 (22.9)
14 .97 (100.1)
Nationwide VMT calculated using local VMT total, and
breakdown of urban VMT (by vehicle class) from the
nationwide approach.
NEDS emissions inventory updated with MOBILES NOx emission
factors (original NEDS based on MOBILE2.5).
-------
Other Cities
For several of the cities, estimates of the VMT breakdown
by vehicle class were unavailable. An estimate of total VMT
for the SMSA was generally available, however. In order to
estimate the mobile source NOx inventory for these cities, the
local estimate of total VMT and the average VMT breakdown for
the eight cities which did provide data were used. The
resulting inventories are presented in Tables A-9 through A-12.
A summary of how the total VMT for each of these cities
was obtained follows.
Boston
The estimates of total VMT for the Boston SMSA were
obtained from the Central Planning Division of the
Massachusetts DOT.[1] The 1981 value provided was
44,480,000/day, which corresponds to 16.2 billion miles per
year.
Washington, DC
The estimates of total VMT for the Washington DC SMSA were
obtained by county for the year 1980 from the Transportation
Planning Board of the Washington Council of Governments
(WASHCOG). One county, Charles County, MD, was not included,
and the 1981 estimate from the NEDS area source report for this
county was used.
The estimate for all of the counties from WASHCOG totalled
45,451,000 miles/day in 1980. [2] This was multiplied by 365
and then by a growth factor derived from the nationwide urban
method to yield a 1981 annual estimate of 16.9 billion miles.
Adding in the value from NEDS for Charles County, [3] this
yielded a total of 1981 VMT of 17.6 billion for the Washington,
D.C. SMSA.
[1] Massachusetts Department of Transportation, Central
Planning Division, conversation with Robert Siebert,
October 1984.
[2] Washington Council of Governments, Transportation Planning
Board, conversation with Ron Sarros, September 21, 1984.
[3] NEDS area source reports.
-------
Table A-9
1981 Annual Comparisons
Boston
Localized Approach*
LDV 13802.4 (85.2)
LDT 1522.8 (9.4)
HDGV 437.4 (2.7)
HDDV 437.4 (2.7)
Total 16200.0 (100.0)
VMT, Millions (%)
Nationwide Approach**
13089.6 (80.8)
2235.6 (13.8)
550.8 (3.4)
324.0 (2.0)
16200.0 (100.0)
NEDS Approach
14252.0 (84.4)
1705.5 (10.1)
342.2 (2.0)
585.3 (3.5)
16885.0 (100.0)
NOx Emissions, 1QQQ Tons (%)
Localized Approach Nationwide Approach NEDS Approach***
LDV
LDT
HDGV
HDDV
Total
33.97 (66.5)
5.21 (10.2)
2.52 (4.9)
9.40 (18.4)
51.10 (100.0)
32.18 (64.4)
7.65 (15.3)
3.17 (6.3)
6.97 (13.9)
49.97 (99.9)
35.03 (63.2)
5.83 (10.5)
1.97 (3.6)
12.58 (22.7)
55.41 (100.0)
* No local breakdown available; average local breakdown from
other 8 cities applied to the VMT figure provided by
locality.
** Nationwide VMT calculated using local VMT total, and
breakdown of urban VMT (by vehicle class) from the
nationwide approach.
*** NEDS emissions inventory updated with MOBILE3 NOx emission
factors (original NEDS based on MOBILE2.5).
-------
Table A-10
1981 Annual Comparisons
Washington, D.C.
VMT, Millions (%)
Localized Approach* Nationwide Approach**
LDV 14897.2 (85.2)
LDT 1643.6 (9.4)
HDGV 472.1 (2.7)
HDDV 472.1 (2.7)
Total 17485.0 (100.0)
14127.9 (80.8)
2412.9 (13.8)
594.5 (3.4)
349.7 (2.0)
17485.0 (100.0)
NOx Emissions, 1000 Tons (%)
Localized Approach
LDV 36.67 (66.5)
LDT 5.62 (10.2)
HDGV 2.72 (4.9)
HDDV 10.15 (18.4)
Total 55.16 (100.0)
Nationwide Approach
34.73 (64.4)
8.25 (15.3)
3.42 (6.3)
7.52 (13.9)
53.92 (99.9)
NEDS Approach
13002.9 (83.8)
1755.2 (11.3)
306.2 (2.0)
446.3 (2.9)
15510.6 (100.0)
NEDS Approach***
31.96 (64.8)
6.00 (12.2)
1.76 (3.6)
9.59 (19.4)
49.31 (100.0)
* No local breakdown available; average local breakdown from
other 8 cities applied to VMT figure provided by locality.
** Nationwide VMT calculated using local VMT total, and
breakdown of urban VMT (by vehicle class) from the
nationwide approach.
*** NEDS emissions inventory updated with MOBILES NOx emission
factors (original NEDS based on MOBILE2.5).
-------
Denver, Colorado
The estimate of 25,900,000 miles per day (9.45 billion
miles per year) for 1981 VMT for the Denver SMSA was obtained
from the Colorado State Highway Department.[4] In the Denver
area, approximately 25 traffic counters are used, and some
seasonal adjustments are made in the estimates. The local
(city streets) VMT is estimated.
Newark, NJ
The estimates of VMT for the Newark SMSA were obtained by
county for the year 1980 from the New Jersey Department of
Transportation. [ 5] The total for the four counties was 35.3
million miles per day, which translates into a 1980 annual
total of 12.8 billion miles. This value was converted into a
1981 estimate using a growth rate determined . from the
nationwide urban method, resulting in a total VMT for 1981 of
13.1 billion miles.
[4] Colorado State Highway Department, conversation with Dick
Mango, October 3, 1984.
[5] New Jersey, DOT, Documentation of New Jersey Department of
Transportation Methodology Used in Developing Emissions
Inventories for Hydrocarbon, Carbon Monoxide, and Nitrogen
Oxide for Mobile Sources, December 1981, p. 15.
-------
Table A-ll
1981 Annual Comparisons
Denver
VMT, Millions
Localized Approach*
Approach
LDV 8054.8 (85.2)
LDT 888.7 (9.4)
HDGV 255.3 (2.7)
HDDV 255.3 (2.7)
Total 9454.0 (100.0)
Nationwide Approach**
7638.8 (80.8)
1304.7 (13.8)
321.4 (3.4)
189.1 (2.0)
9454.0 (100.0)
NEDS
8488.6 (73.2)
2410.5 (20.8)
206.2 (1.8)
483.9 (4.2)
11589.2 (100.0)
NOx Emissions, 1000 Tons ('
Localized Approach Nationwide Approach
LDV
LDT
HDGV
HDDV
Total
13.83 (61.7)
2.10 (9.4)
1.00 (4.5)
5.49 (24.5)
22.42 (100.1)
13.10 (60.9)
3.09 (14.4)
1.25 (5.8)
4.07 (18.9)
21.51 (100.0)
NEDS Approach***
14.55 (46.2)
5.71 (18.2)
0.80 (2.5)
10.40 (33.1)
31.46 (100.0)
* No local breakdown available; average local breakdown from
other 8 cities applied to VMT figure provided by locality.
** Nationwide VMT calculated using local VMT total, and
breakdown of urban VMT (by vehicle class) from the
nationwide approach.
*** NEDS emissions inventory updated with MOBILE3 NOx emission
factors (original NEDS based on MOBILE2.5).
-------
Table A-12
1981 Annual Comparisons
Newark
VMT, Millions (%)
Localized Approach* Nationwide Approach**
LDV 10745.7 (85.2)
LDT 1185.6 (9.4)
HDGV 340.5 (2.7)
HDDV
340.5 (2.7)
Total 12612.3 (100.0)
10190.8 (80.8)
1740.5 (13.8)
428.8 (3.4)
252.2 (2.0)
12612.3 (100.0)
NEDS Approach
10759.4 (85.3)
1180.4 (9.4)
225.7 (1.8)
446.9 (3.5)
12612.4 (100.0)
NOx Emissions, 1000 Tons (%)
Localized Approach Nationwide Approach NEDS Approach***
LDV 26.44 (66.5)
LDT 4.05 (10.2)
HDGV 1.96 (4.9)
HDDV 7.32 (18.4)
Total 39.77 (100.0)
25.05 (64.4)
5.95 (15.3)
2.47 (6.3)
5.42 (13.9)
38.89 (99.9)
26.45 (63.9)
4.04 (9.8)
1.30 (3.1)
9.61 (23.2)
41 . 40 (100 . 0)
* No local breakdown available; average local breakdown from
other 8 cities applied to the VMT figure provided by
locality.
** Nationwide VMT calculated using local VMT total, and
breakdown of urban VMT (by vehicle class) from the
nationwide approach.
*** NEDS emissions inventory updated with MOBILES NOx emission
factors (original NEDS based on MOBILE2.5).
-------
MOBILE3 Emission Factors
Calendar-year NOx emission factors for each individual
vehicle class' are presented in tables on the following two
pages (low and high-altitude).
These emission factors were calculated using EPA's MOBILES
computer program, assuming six different emission control
scenarios (described in tables).
By multiplying the 1981 annual VMT figures for each vehicle
class by these class-specific NOx emission factors (in
grams per mile), base-year inventories can be calculated.
Future inventories can be estimated by applying the
appropriate emission factor ratios (future year over
baseline) and the VMT growth rates (shown in an earlier
section) to the 1981 NOx inventory calculated above.
Shown below are vehicle mix figures from MOBILES, which
were used in combining some of the local data into the
appropriate vehicle classes.
Vehicle Mix Figures
Vehicle Default
Type Vehicle Mix
LDGV 0.662
LDDV 0.008
LDGTi 0.133
LDGT2 0.088
LDDT 0.002
HDGV 0.040
HDDV 0.060
MC 0.007
-------
5 NOVEMBER 19H4
TABLE i
LOW 'NuriTUUE f-mx EMISSION RESULTS FROM MOBILE ASSUMING THAT
AMBIENT TtMh-ER/uiwE = 7b F: COLD/HOT START = 20.6/27.3/20.6
VtHlCLE SOEED = 19.!, MPh rnP l_n f. MC- 20.0 MPH FOR HO: MOBILE3 CALCULATED VMT MIX.
_CI_ _LDC'V_ _LDQI1
1981 ?.73 1.29
1990 1.57 2.89
1995 1.43 2.46
2000 1.42 2.13
__ . LQDi _L(_ __
4. SO 6.36 1.45 1.94 23. /O 0.62
3.22 5.47 1.10 1.44 ]b.l4 O.MS
2.77 4. 48 1.03 1 . 3« 9.a2 O.H5
2.47 4.^1 j.(ij 1 . ?V 8.46 O.MS
4.34
2.6b
2.12
2.73 .1.29
1.57 3.1"
1.43 3.09
1.4? 2.98
LDDI
4.50 -S.36 1.45 1.94 23.70
3.39 5.47 1.10 1.63 15.14
3.1? 4.48 1.03 1.65 9.82
3.00 4.21 1.03 1.68 8.46
0.62
O.B5
0.85
0.85
4.34
2.70
?.Z3
2.10
1981 2.73 3.29
1990 1.S7 3.18
199S 1.43 3.09
2000 1.42 2.98
4.bO 6.36 1.4S 1.94
3.39 5.41 1.10 1.63
3.17 5.33 1.03 1.6b
3.UO S'.cJO 1.03 1.68
;3./0 0.6?
IV.U'j O.HS
19.21 O.H5
19. ^0 0.-
-------
TABLE 2
HIGH ALTlTUOt M0« EMISSION RESULTS FROM MORILF3 ASSUMING THAT
AMBIENT TEMPERATURE = 75 F: COLO/HOT START = 3n.6/37.3/20.6
VEHICLE 9PF.ED = 19. h MPH Fnt> I n *. MC. 20.0 MPH FOR HO: MORILE3 CALCULATED VMT MIX.
5 NOVEMBER 1984
1981
1990
1995
2000
.LI?GY LQ&TI
1.90 2.36
1 .45
1.42
1.43
2.68
2.39
2.12
3.00
2.89
2.68
2.4b
_ LDUV. LDDI
4.33 1.45 1.90
4.16 1.10 1.43
3.67 1.03 1.38
3.55 1.03 1.39
23. /O
15.14
9.H2
8.46
MC__
0.41
0.57
0.57
0.57
2.47
2.07
1.94
1.90
1.45
1.42
1.43
LDGII
2.36
2.97
3.03
2.97
3.00
3.07
3.07
3.00
LQDI
4.33 1.45 1.90
4.16 1.10 1.62
3.67 1.03 1.65
3.55 1.03 1.68
23.70
15.14
9.82
8.46
0.41
0.57
0.57
0.57
3.45
2.5?
2.17
2.o«
_CI_ _LQOy LDGIi LDQ12
1981
1990
1995
2000
1.90
1 .45
1.42
1 .43
2.36
2.97
3.03
2.97
3.00
3.07
3.07
3.00
LQDI
4.33 1.45 1.90
3.75 1.10 1.62
3.74 1.03 1.65
3.67 1.03 1.68
2J. /O
IV. U5
19.21
19.20
N LEYE.LS. ______
0.41
0.57
0.57
0.57
3.4b
2.69
2.60
2.56
_LDfiV_
1.90
1.45
1.42
1.43
2.36
2.97
3.03
2.97
3.00
3.07
3.07
3.00
_ LDQY LDDI
4.33 1.45 l.QO
3.75 1.10 1.62
3.74 1.03 1.65
3.67 1.03 1.68
bDDV
23.70
20.68
21.74
21.93
0.41
0.57
0.57
0.57
3.45
2.77
2.72
2.68
1981
1990
1995
2000
._ _LQ&I1
1.90 2.36
1.45 2.68
1.42 2.39
1.43 2.12
LQ£I2 _t)DQV_ LQD¥ LDDI
3.00 4.33 1.45 1.90
2.89 4.16 1.10 1.43
2.68 4.53 1.03 1.38
2.46 4. SB 1.03 1.39
23. /O
1^.14
12.65
11.94
MC__
0.41
0.57
0.57
0.57
3.45
2.47
2.23
2.13
1.90
1.45
1.42
1.43
LDGI1
2.36
2.97
3.03
2.97
LDGI2
3.00
3.07
3.07
3.00
.
_ LDDM LDDI
4.33 1.45 1.90 23.70
4.16 1.10 1.62 15.14
4.53 1.03 1.65 12.65
4.58 1.03 1.68 11.94
__ ALL_V£H
0.41 3.45
0.57 2.52
0.57 2.34
0.57 2.27
------- |