OCR error (C:\Conversion\JobRoot\0000062M\tiff\2000MDXE.tif): Unspecified error
-------
EPA-450/3-78-038
Air Quality Assessment of Particulate
Emissions from Diesel-Powered Vehicles
by
Terrence Briggs, Jim Throgmorton, and Mark Karaffa
PEDCo Environmental, Inc.
Chester Towers
11499 Chester Road
Cincinnati, Ohio 45246
Contract No. 68-02-2515
EPA Project Officer: Justice A. Manning
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
March 1978 __n r,_
*'-'"' ' . " - ,- V^oQti
-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers. Copies are
available free of charge to Federal employees, current contractors and
grantees, and nonprofit organizations - in limited quantities - from the
Library Services Office (MD-35), U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina 27711; or, for a fee, from the
National Technical Information Service, 5285 Port Royal Road, Sprinqfield
Virginia 22161.
This report was furnished to the Environmental Protection Agency by
PEDCo Environmental, Inc., Chester Towers, 11499 Chester Road,
Cincinnati, Ohio 45246, in fulfillment of Contract No. 68-02-2515. 'The
contents of this report are reproduced herein as received from PEDCo
Environmental, Inc. The opinions, findings, and conclusions expressed
are those of the author and not necessarily those of the Environmental
Protection Agency. Mention of company or product names is not to be
considered as an endorsement by the Environmental Protection Agency.
Publication No. EPA-450/3-78-038
-------
CONTENTS
Page
LIST OF FIGURES V
LIST OF TABLES
LIST OF ABBREVIATIONS AND SYMBOLS
ACKNOWLEDGMENT
1.0 SUMMARY 1-1
2.0 INTRODUCTION 2-1
3.0 CHARACTERIZATION AND HEALTH EFFECTS ASSESSMENT OF 3-1
DIESEL PARTICULATE EMISSIONS
3.1 Particulates 3-2
3.2 Polycyclic Organic Matter 3-17
3.3 Sulfates 3-27
3.4 Minor Components 3-30
3.5 Current Research Status 3-30
4.0 TEST CITY METHODOLOGY AND PROJECTIONS 4-1
4.1 Nonmotor Vehicle Emissions 4-6
4.2 Motor Vehicle Exhaust Emissions 4-11
4.3 Projected Impact of Diesel Emissions on Air 4-34
Quality
4.4 Assessing Population Exposure 4-37
5.0 ESTIMATES OF POPULATION EXPOSURE TO TSP AND BaP 5-1
ill
-------
CONTENTS (continued)
Page
5.1 National Population Impact 5-1
5.2 Projected Maximum Impact of Diesel Emissions 5-20
on Air Quality
5.3 Discussing Exposure Data 5-28
APPENDIX A A_1
APPENDIX B _
IV
-------
FIGURES
Number Page
3-1 Particle Size Distribution 3-10
3-2 Impact of Vehicle Exhaust on Ambient 3-13
Particulate Size Distribution Data
for Gasoline-Powered Vehicles
3-3 Particle Size Deposition Probablities 3-15
4-1 The Test City Study Area - Kansas City, 4-3
Missouri
4-2 Projected Heavy-duty Vehicle Sales 4-14
4-3 Procedures for calculating projection 4-17
Years' Grid VMT by Vehicle Category for
Two Diesel Introduction Rate Assumptions
4-4 Dosage Spectrum Distribution in the 4-42
Tri-State Region
5-1 Correlation of Average NASN Annual 5-4
Geometric Mean TSP Levels for Selected
SMSA's Versus SMSA Population and Popula-
tion Density
5-2 Information Flow Diagram for the Develop- 5-9
ment of the Population TSP Dose Relation-
ship for Each Projection Case
-------
TABLES
Number
Pacre
1-1 Exhaust Emissions of TSP and BaP, Gasoline- 1-2
Versus Diesel-Powered Vehicles
1-2 Diesel Share of New Sales by Model Year 1-2
1-3 Projected Motor Vehicle Particulate Exhaust 1-8
Emissions for Kansas City
1-4 Predicted Peak Levels of Diesel-Generated 1-9
TSP for Kansas City
1-5 Estimated Population Exposure to More Than l-ll
the Federal Standard for TSP
3-1 Particulate Emissions from Diesel Versus 3-3
Gasoline Passenger Cars
3-2 Elemental and Trace Metal Composition of 3-5
Diesel Exhaust Particulates
3-3 Aerodynamic Diameters of Diesel Exhaust 3-9
Particulates Collected in Various Stages of
an Anderson Sampler
3-4 Diesel Exhaust Particle Size 3-11
3-5 Frequency of Occurrence of Formulas for 3-18
Carcinogenic Compounds in Diesel Exhaust
Particulates
3-6 Compounds Detected in Various Atmospheric 3-20
Pollutants (G), (D), and (A) Samples
3-7 Benzo{a]pyrene Emissions from Diesel Versus 3-22
Gasoline Cars
3-8 Sulfate Emissions from Diesel Versus Gaso- 3-29
line Passenger Cars
VI
-------
Number
TABLES (continued)
Page
4-1 Particulate Emissions in Test City 4-7
4-2 Comparison of Measured Versus Predicted TSP 4-9
Concentrations, Kansas City
4-3 Base-Year Fractions of Total Vehicles in 4-19
Use Nationwide
4-4 Base-Year Fractions of Total Heavy-Duty 4-21
Trucks in Use Nationwide (and Diesel Frac-
tions Thereof)
4-5 Classification of Trucks by GVM 4-22
4-6 Truck Sales and Diesel Penetration Fractions 4-24
4-7 Diesel Vehicle Introduction Rates 4-26
4-8 Fraction of Urban VMT by Mobile Source Gate- 4-27
gory in Projection Years
4-9 Exhaust Emission Factors 4-29
4-10 Weighted Emission Factors for Gasoline- 4-31
Powered Vehicles
4-11 Projected Motor Vehicle Exhaust Emissions 4-32
(Particulate)
4-12 Projected Motor Vehicle Exhaust Emissions 4-33
(BaP)
4-13 Ratio of BaP to Particulate Emissions 4-35
(Diesels Only)
4-14 Projected Regional Annual Average Concentra- 4-36
tions of TSP from Diesel Exhaust for Test
City
4-15 Projected Regional Annual Average Concentra- 4-38
tions of BaP from Diesel Exhaust for Test
City
vii
-------
TABLES (continued)
Number
Page
5-1 Distribution of U.S. Population by SMSA 5-5
Population Range and Population Density for
Projection Years - National SMSA Populations
5-2 Summary of TSP Data from Test City Monitor- 5-7
ing Stations
5-3 Estimated Annual Exposure Concentrations of 5-10
TSP From Diesel Vehicle Exhaust
5-4 Percent of Population Exposure to TSP 5-11
Attributable to Diesel Exhaust Emissions
5-5 Estimated Population Exposed to More Than 5-13
the Federal Standard for TSP
5-6 Annual Average Exposure Concentratios of 5-16
BaP Emitted by Coke Ovens
5-7 Annual Average Ambient BaP Concentrations at 5-17
NASN Urban Stations Without Coke Oven Impact
5-8 Population Exposure to BaP in Urban Areas 5-18
Without Coke Oven Impact
5-9 Estimated Total Population Dosage of BaP in 5-21
19 76
5-10 Annual Average Exposure Concentrations of 5-22
BaP From Diesel Exhaust Emissions
5-11 Projected Concentrations of TSP from Diesel 5-24
Exhaust
5-12 Projected Maximum Concentrations of BaP From 5-29
Diesels
A-l Fractions of Light-Duty Vehicle VMT in Pro- A-2
jection Years
A-2 Fractions of Light-Duty Trucks VMT in Proiec- A-I
tion Years
A-3 Fractions of Heavy-Duty Truck VMT for Pro- A-4
jections Years, Urban Only
B-l Average SMSA TSP Levels Correlated with A-5
Population Parameters
viii
-------
LIST OF ABBREVIATIONS AND SYMBOLS
The following is an alphabetical list of terms used in
the report.
Ao - land area used in estimating population dose.
ADT - annual daily traffic volume.
AQDM - Air Quality Dispersion Model.
BaP - benzo[a]pyrene.
Best est. - best estimate of light and heavy diesel vehicle
growth trends.
C - average contribution of paved roads to measured TSP
level, yg/m3.
C max. - maximum pollutant concentration expected for a
time period of concern.
5 - dosage threshold used to estimate a dosage spectrum
S (D).
FHWA - Federal Highway Administration.
FTP - Federal Test Procedure.
GVW - gross vehicle weight.
HDD - heavy-duty trucks, diesel.
HOT - heavy-duty trucks.
HDG - heavy-duty trucks, gasoline.
HDV - heavy-duty vehicles.
LOT - total light-duty trucks.
IX
-------
LDV - total light-duty vehicles.
LDVD - light-duty diesel vehicles.
LDVG - light-duty gasoline vehicles.
MARC - Mid-American Regional Council.
Max. - maximum estimate of light- and heavy-diesel vehicle
growth trends.
M - annual geometric mean pollutant concentration.
MMD - mass median diameter.
N(r,D) - threshold function used in dosage estimation.
NAAQS - National Ambient Air Quality Standard.
NADB - National Air Data Branch.
NASN - National Air Surveillance Network.
NEDS - National Emissions Data System.
9
ng - nanogram, 10 grams.
OXY - oxygenated hydrocarbon fraction of POM.
PNA - polynuclear aromatic hydrocarbon.
PPOM - particulate polycyclic organic matter.
r - slant (or direct) distance between pollutant monitor
and roadway, ft.
S(D) - dosage spectrum.
SET - Sulfate Emission Test.
Sg - standard geometric deviation.
SMSA - Standard Metropolitan Statistical Area.
T - average daily traffic volume.
TRN - transitional hydrocarbon fraction of POM.
-------
TSP - total suspended particrulate.
VMT - vehicle miles traveled.
VPOM - vapor phase of polycyclic organic matter.
x - horizontal distance between roadway and pollutant
monitor.
z - sampler height.
0 - arctan («/x).
yg - micrograms, 10 gram.
XI
-------
ACKNOWLEDGMENT
This report was furnished to the U.S. Environmental
Protection Agency by PEDCo Environmental, Inc., Cincinnati,
Ohio. Terrence Briggs was the PEDCo Project Manager and
George Jutze functioned as Service Director. Principal
authors of the report were Terrence Briggs, Jim Throgmorton,
and Mark Karaffa.
Justice Manning was the Task Officer for the U.S.
Environmental Protection Agency. The authors appreciate the
contributions made to this study by Mr. Manning and other
EPA personnel.
Xll
-------
1.0 SUMMARY
Sales of both light- and heavy-duty diesel-powered
vehicles are projected to increase markedly in the next
several years. This prediction and new toxicity data have
caused attention to be focused on the potential of the
resulting increased particulate exhaust emissions from this
source having an impact on public health. In evaluating
this impact, issues of major concern are the higher particu-
late emission rates (versus those from comparable gasoline-
powered vehicles), the high fraction of the particulate
matter in the respirable size range, and the potential
toxicity of this particulate matter.
The report presents estimates of the impact diesel-
powered emissions will have on the levels of total suspended
particulates (TSP) and benzolalpyrene (BaP) to which the
population is exposed. iLevels of BaP are generally used as
an index of total polynuclear aromatic hydrocarbon (PNA)
content, primarily because of its potent carcinogenicity.]
The values in Table 1-1 show that both TSP and BaP are
emitted at a significantly higher rate from the exhausts of
diesel-powered vehicles than from comparable, catalytically
equipped, gasoline-powered vehicles.
1-1
-------
Table 1-1. EXHAUST EMISSIONS OF TSP AND BaP
GASOLINE- VERSUS DIESEL-POWERED VEHICLES '
Vehicle category
Light-duty gasoline (catalyst)
(noncatalyst)
Light-duty diesel
Heavy-duty gasoline (catalyst)
(noncatalyst)
Heavy-duty diesel
Emission factors
Particulates,
g/VMTa
0.006-0.015
0.002-0.25
0.5
0.02-0.05
0.007-0.90
2.0
BaP,
yg/VMT
0.1
1.0 .
1.0-6.0°
0.3
3.0 .
4.6-24.6°
VMT = vehicle miles traveled.
Low and high emission estimates.
The increasing share of the market projected to be
occupied by diesel-powered vehicles will also add these
emissions to the population exposure to TSP and BaP. Table
1-2 shows the predicted increase.
Table 1-2. DIESEL SHARE OF NEW SALES BY MODEL YEAR
Year
1975
1980
1985
1990
Light-duty vehicles,
percent
Best estimate3
0. 5
4.0
10.0
10.0
Max.b
0.5
10.0
25.0
25.0
Heavy-duty trucks,
percent
Best estimate
28.0
31.0
33.0
64.0
Max.
28.0
38.0
78.0
99.0
Indicates best diesel market growth estimate.
Indicates maximum diesel market growth estimate.
The characterization and health effects assessment in
this report focuses specifically on diesel-generated partic-
ulate matter and its components. Particulate emissions from
1-2
-------
diesel-powered automobiles, largely carbonaceous solids, are
about 20 to 50 times higher than those from comparable
automobiles burning unleaded gasoline. Diesel particulates
are small enough to penetrate deeply into the alveolar
region of the respiratory tract. The aerodynamic diameters
of a large proportion of diesel exhaust particulates are
less than 1 ym. Submicronic particles undergo Brownian
motion and are deposited in the lung parenchyma. In some
cases (depending on the material), alveolar clearance of
particulate matter may not occur for some time. Moreover,
experimental evidence suggests that presumably inert carrier
substances (e.g. carbon) can affect pulmonary clearance
mechanisms and, consequently, retention time. Thus, keeping
potentially toxic agents in effective contact with suscep-
tible tissues for prolonged periods increases the likelihood
of chemical induction of biological changes and disease in
critical organs.
A review of the literature suggests that particulate
emissions from diesel engines are not well characterized
chemically, physically, or quantitatively; therefore,
emission factors include a high degree of uncertainty.
Polynuclear aromatic hydrocarbons, one class of organic
compounds known to be emitted in diesel exhaust, are likely
adsorbed on the carbonaceous particulate. The presence of a
1-3
-------
number of PNA's other than BaP (of similar structure and
having carcinogenic potency) and the lack of reliable quan-
titative data concerning BaP in diesel engine exhaust
nevertheless preclude the unqualified use of BaP as an
indicator molecule of total PNA concentration and total
carcinogenic potency from this emission source. Emissions
of BaP from diesel-powered vehicles and older model gaso-
line-powered vehicles (not equipped with catalyst) now
appear to be about the same. Emissions of BaP from cata-
lyst-equipped vehicles, however, are reportedly lower.
Ratios of BaP to total PNA in diesel exhaust and the con-
tribution of BaP content to the total carcinogenic potential
of diesel emissions need to be determined if BaP is to be
used as an index.
Cancer, particularly of the respiratory tract, is the
most significant health problem associated with polycyclic
organic matter (POM). This association is based on epide-
miological evidence of occupational exposures and informa-
tion obtained from animal toxicity studies. A correlation
between atmospheric concentrations of POM and increased
incidence of cancer mortality is suggested, but definitive
evidence of a causal relationship is lacking. Other con-
comitant emission products (e.g. sulfur dioxide, nitrogen
oxides, ozone) are suspected of having a potentiating action
1-4
-------
on the carcinogenic properties of PNA's or possibly re-
sulting in oxygenation of PNA (or POM).
Sulfur compounds are also associated with diesel
particulate emissions. The types of sulfate emitted by
diesels and the sulfuric acid aerosol portion of the sulfur
emission are unknown. Because the health effects of ex-
posure to sulfuric acid differ from those due to exposure to
various sulfates, the public health impact of given amounts
of diesel sulfur cannot be predicted. Sulfate emissions
from diesel-powered cars generally are less than from
gasoline-powered cars with catalyst equipment and air in-
jection, but are higher than those from cars not equipped
with catalysts or cars with three-way catalyst systems.
Sulfate emissions tend to be governed by the sulfur con-
centration in the fuel.
Assessing health effects of diesel particulate emis-
sions as a function of the toxicity of individual chemical
components has obvious limitations. The Ames Salmonella/mi-
crosome mutagenicity assay is being used by the EPA as a
quick, inexpensive method of establishing priorities for
physical and chemical characterization studies and addi-
tional toxicologic tests. Several fractions of diesel
exhaust particulate show significant mutagenic activity in
this bioassay system. Transitional and oxygenated hydro-
1-5
-------
carbon fractions are the most active. It should be empha-
sized that these are very preliminary data. Selected frac-
tions of diesel exhaust particulate must be evaluated by
other confirmatory bioassays and toxicologic methods to
determine the significance of the positive results obtained
in these initial screening tests. The positive results
obtained in the Ames test indicate the utility of such
in vitro bioassays to direct the fractionation of diesel
particulate. This approach should eventually make it pos-
sible to identify the mutagenic components of diesel par-
ticulate. Chemical characterization of the active fractions
is now in progress.
In addition to in vitro bioassays, whole animal studies
are being conducted, in which appropriate test species are
acutely and chronically exposed directly to dilute diesel
exhaust. In this series of experiments, a wide variety of
biological parameters are being measured to determine the
effects of the emission mixture on the respiratory system.
Results obtained from inhalation studies using animals, the
Ames mutagenicity assays, and other confirmatory in vitro
bioassay systems will help define potential health hazards
and estimate the degree of toxicity associated with exposure
to diesel emissions and components thereof.
Diesel exhaust particulate fractions have been found to
be potentially mutagenic and certain POM is known to be
1-6
-------
carcinogenic, and diesel particulates are projected to
increase significantly the population TSP exposure, par-
ticularly in areas near roadways. Total ambient BaP levels
appear to be somewhat less affected by diesel-generated BaP,
but they also were higher near roadways. It is concluded,
therefore, that diesel vehicle exhaust particulates do
represent a health hazard.
The impact of diesel-generated particulates on popula-
tion exposure to TSP and BaP is projected for 1981, 1983,
1985, and 1990. A detailed particulate emission inventory
is developed for a representative test city (Kansas City,
Missouri) for a reference year (1974). Emissions from all
sources except diesel are assumed to remain constant. The
impact of diesel-generated particulates on the population at
165 grid points in this city is determined on the basis of
best estimate and maximum diesel growth cases for each
projection year. A total emissions inventory for Kansas
City in 1974 shows that highway vehicle exhaust emissions
accounted for 1.6 percent of total particulate loading
(1,006 of 64,033 tons/year). Table 1-3 shows projected
motor vehicle exhaust emission rates to be consistently
lower than those in 1974, but diesel vehicles represent a
progressively larger fraction of the total. Trends in BaP
emissions are similar; however, insufficient data are avail-
1-7
-------
I
00
Table 1-3. PROJECTED MOTOR VEHICLE PARTICULATE EXHAUST EMISSIONS FOR KANSAS CITY
(in tons/yr)
Vehicle category
Gaso 1 ine- power ed
Light- duty
Heavy-duty
Diesel- powered
Light- duty
Heavy-duty
Total
1974
733
102
8
163
1006
1981
Best
est.
200
111
29
190
530
Max.
197
301
67
205
573
1983
Best
est.
119
111
65
206
501
Max.
94
94
163
239
589
1985
Best
est.
76
111
106
209
502
Max.
70
76
258
291
695
1990
Best
est.
61
90
190
295
636
Max.
53
36
429
415
933
Increasing diesel-powered vehicle introduction corresponds to decreasino
gasoline-powered vehicle use.
-------
able to make a total BaP emission inventory of Kansas City,
Table 1-4 presents an evaluation of predicted peak
levels of diesel-generated TSP, based on an analysis of
Kansas City data.
Table 1-4. PREDICTED PEAK LEVELS OF
DIESEL-GENERATED TSP FOR KANSAS CITY
TSP, yg/m
Regional annual geometric
mean
Regional 24-hr maximum
Roadside3 annual geometric
mean
Roadside 24-hr maximum
Maximum diesel contribution,
percent
1974
0.35
1.05
3.85
LI. 48
1981
Best
est.
0.45
1.34
4.95
14.76
Max.
0.56
1.66
6.16
18.36
1990
Best
est.
0.96
2.86
10.56
31.48
Max.
1.73
5.16
19.03
56.73
* Typical residential dwelling located adjacent to a major
thoroughfare.
The maximum regional impact of diesel-generated TSP is
projected for 1990. It constitutes 2.3 percent of the pri-
mary national ambient air quality standard (NAAQS) for the
annual mean and 3.4 percent of the secondary NAAQS for the
maximum 24-hour period. Estimated -maximum roadside impact
from diesel-generated TSP in 1990 is 25.3 (annual NAAQS of
75 yg/m3) and 37.8 percent (24-hour standard of 150 yg/m )
of the respective NAAQS. Thus, diesel-generated particulate
emissions represent a potentially significant population
exposure impact.
1-9
-------
Peak diesel BaP concentrations of 0.02 and 0.13 ng/m3
for low and high BaP emission estimates, predicted in a
similar manner, are used for the roadside annual geometric
mean. Corresponding values for the 24-hour maximum roadside
impact are 0.12 and 0.69 ng/m . The corresponding upper
range of population exposures to BaP from coke oven opera-
tions averages 20 to 100 ng/m3 annually. Thus, it appears
that the BaP impact from diesel-powered vehicles is rela-
tively low.
The distribution of population exposure to TSP, devel-
oped for each projection case, includes an estimation of
exposure extremes for each grid area. These data are ex-
trapolated to generate national population exposures, based
on the mean of all ambient monitoring station annual average
TSP levels for Kansas City compared with those from other
selected Standard Metropolitan Statistical Areas (SMSA's).
The SMSA population and urban SMSA population density are
also considered in this analysis. The maximum diesel impact
case is the 1990 maximum diesel growth case, which projects
that 1 million people will be exposed to diesel exhaust TSP
at a level greater than 2.4 yg/m3 annual geometric average.
Table 1-5 summarizes these data for the population exposed
to greater than the primary NAAQS of 75 yg/m3.
1-10
-------
Table 1-5. ESTIMATED POPULATION EXPOSURE TO MORE THAN
THE FEDERAL STANDARD FOR TSP
==========
Projection
year
1981
1983
1985
1990
Millions of people exposed to
more than 75 |ig/mj
Best estimate
Total
exposed
62.7
64.3
66.4
71.1
Diesel
contribution
0.4
0.4
0.4
1.0
Maximun. cjrowtn
Total
expos ed
62.8
64.7
67.3
72.3
Diesel
contribution
0.4
0.8
1.5
2.2
Diesel contribution to overall TSP exposure levels is
relatively greater in areas of high TSP concentration. In
locations with TSP levels of more than 120 yg/m , diesel
vehicle emissions increased the number of people exposed by
3.8 to 10.1 percent; whereas in lower exposure areas, the
diesel impact is frequently less than 2 percent of the
total. Thus, diesel TSP emissions tend to have the greatest
impact in locations where emissions exceed National Ambient
Air Quality Standards. These high diesel exposure locations
generally correspond to maximum roadside diesel TSP impact
locations described above.
A methodology similar to that used to estimate TSP is
used to estimate national BaP exposure attributable to
diesel vehicles. Total BaP exposure relationships are based
on urban ambient monitoring data. Diesel exhaust appears to
have a lower impact on total BaP exposure than on TSP
1-11
-------
exposure. The maximum diesel BaP impact occurs in the 1990
high emission estimate case that has the maximum diesel
growth projection, which results in a diesel contribution of
less than 1 percent of the total concentration for the 5
percent of the population receiving the highest exposure.
Because ambient BaP measurements are quite sparse, these
data really amount to crude estimates.
1-12
-------
2.0 INTRODUCTION
The Clean Air Act amendments require the setting of
particulate emission standards for various classes and
categories of vehicles, beginning with 1981 models. In sup-
port of proposed standards for particulate emissions from
light- and heavy-duty diesel vehicles, this report presents
a preliminary assessment of the impact of diesels on pro-
jected air quality and the potential public health effects
associated with total suspended particulate (TSP) and
particulate polycyclic organic matter (PPOM) in diesel
exhaust.
Because of the very short time allowed to complete this
assessment, some abbreviated procedures were used. A single
compound, benzo[alpyrene (BaP), is used as an indicator of
PPOM because it is the only polycyclic organic substance for
which ambient air quality and diesel emissions data are
currently available. Although normally an indicator of
polycyclic aromatic hydrocarbons (PNA's) in urban environ-
ments, BaP is used here as an indicator of the broader class
of polycylic organics, PPOM.
2-1
-------
An assessment of the effects of diesel exhaust particu-
late on human health is presented first to aid in inter-
preting the potential health impact significance of pro-
jected increases in exposure of the population to diesel-
derived particulates and to provide a basis for the model
used to estimate public risk from exposure. Also presented
are available data on the chemical constituents of diesel
particulate, together with an overview of the health effects
literature regarding significant particulate fractions and
the status of the toxicity assessment of diesel particu-
lates.
In assessing potential exposures of the general popu-
lation to dosages of TSP and BaP attributable to diesel
exhaust, projections are developed for 4 years: 1981, 1983,
1985, 1990. Consideration is given to the introduction of
diesel-powered vehicles into the total automotive market in
each of these years, in terms of a best estimate and a
maximum growth value for light-duty and heavy-duty vehicles
(percent of sales in each class).
To help estimate the potential impact of increasing
diesel vehicle sales on ambient particulate air quality, an
analysis is made of the distribution of population exposures
to TSP and BaP. The analysis indicates both the total
exposures to TSP and BaP and the exposures due to diesel
vehicle exhaust only.
2-2
-------
Based on these exposure estimates, a further estimate
is made of the impact of particulate exhaust emissions from
diesels on exposure of the population to TSP. For each
projection case the dose-distribution for TSP and for the
diesel-derived portion of TSP is determined with respect to
a representative city (Kansas City, Missouri, in this
analysis). Both TSP and the diesel contributions to TSP are
determined at 165 grid locations in the city by use of the
Air Quality Dispersion Model (AQDM). Based on census tract
population data, population versus dose-level relationships
are developed. Then, based on national trends in TSP expo-
sure in a representative sampling of all standardized Metro-
politan Statistical Areas (SMSA1s), distributions are devel-
oped for exposure of the total national population to TSP
and to the diesel-derived portion.
The impact of BaP from diesel-powered vehicles is esti-
mated from a relatively sparse data base in a similar,
though less involved, manner. The dose-distribution of BaP
from diesels is developed for each projection year. Na-
tional exposure relationships are developed, based on pro-
jections of growth in diesel vehicle sales, national popula-
tion distributions, and relationships of national trends to
those predicted for Kansas City. The total national dis-
tribution of exposures to BaP is determined by summing the
2-3
-------
exposures attributable to coke-oven emissions and the ex-
posures attributable to all other sources. These calcula-
tions are based on ambient air quality data from urban
stations of the National Air Surveillance Network (NASN).
The BaP impact attributable to diesels is determined for
each projection year, for each diesel-vehicle growth case,
and also for two diesel-vehicle emission rates (a total of
16 cases).
2-4
-------
3.0 CHARACTERIZATION AND HEALTH EFFECTS ASSESSMENT
OF DIESEL PARTICULATE EMISSIONS
The discussion that follows summarizes available
technical information on the composition and associated
health effects of diesel particulate emissions. It inten-
tionally excludes regulated gaseous components of diesel
exhaust (i.e. carbon monoxide, nitrous oxides, hydrocar-
bons) , some of which may have synergistic or potentiating
biological effects with particulate organic matter. It is
difficult to assess any associated risk to human health from
diesel exhaust because data on health effects are limited,
particle size is variable, and particulate composition
varies both qualitatively and quantitatively. This section
therefore first characterizes particulate emissions from
diesel-powered vehicles by their chemical composition (i.e.
their major and trace elements, PNA's, sulfates, etc.), then
presents an overview of the literature on animal and human
health effects of the major chemical groups identified. It
gives particular attention to the carcinogenicity and
mutagenicity of identified compounds or fractions isolated
from diesel exhausts.
3-1
-------
3.1 PARTICULATES
3.1.1 Emissions
The term "particulate" encompasses a class of emission
products, variable in composition, that exist in the atmo-
sphere in the form of finely dispersed solids or aerosols.
Total particulate emissions from diesel-powered automobiles
are much higher than those from gasoline-powered automo-
biles. Gasoline engines burning leaded fuel emit particu-
lates that are mainly the product of the combustion of lead
and lead scavengers (ethylene dibromide and ethylene di-
chloride). Catalyst-equipped gasoline engines that burn
unleaded fuel produce a sulfuric acid mist over the cata-
lyst, which, in addition to the associated water of hydra-
tion, forms the particulate emissions from such vehicles.
Table 3-1 compares the results of several particulate
emission tests on typical diesel-powered automobiles with
the results of tests on gasoline-powered automobiles. The
mass of particulates emitted during the Federal Test Proce-
dure (FTP) increases with size of the diesel engine, and all
the measurements are much higher than the levels obtained on
gasoline-powered cars, particularly on the two vehicles
using unleaded gasoline.
3-2
-------
Table 3-1. PARTICULATE EMISSIONS FROM DIESEL VERSUS
GASOLINE PASSENGER CARS1
Vehicle type
Diesel vehicles:
VW Rabbit3
4
Peugeot 504
Mercedes 240D
Mercedes 30 OD
Oldsmobile 3503a
Gasoline vehicles:
VW Rabbit
- unleaded
gasoline
Oldsmobile 350
- unleaded
gasoline
Typical leaded
gasoline car
Engine 2
displacement,
CID
90
129
146
183
350
90
350
b
Total
particulates
(FTP) , g/mile
0.291
0.397
0.477
0.490
0.917
0.007
0.011
0.240
a Early prototype model.
Data not available.
The characterization of diesel exhaust particulates has
only recently begun to receive the degree of attention
already given to emissions from gasoline-powered vehicles.
Information on their physical and chemical characteristics
is therefore limited. Data on the elemental composition,
trace metal content, and organic soluble fractions of
3-3
-------
diesel exhaust particulate are given in Table 3-2. Particru-
lates emitted by diesel engines are composed primarily of
carbon and hydrogen, with relatively small amounts of nitro-
gen, sulfur, and oxygen. Their carbon content appears to be
independent of fuel composition. Although trace metals are
present in diesel particulate emissions (those of potential
health concern are mercury, lead, vanadium, and strontium),
the quantities involved limit their danger to health. Sig-
nificant calcium and barium particulate emissions result
from the use of smoke suppressant additives containing these
two metallic elements.
3.1.2 Pulmonary Deposition
Diesel particulates are small enough to penetrate
deeply into the alveolar region of the respiratory tract.
Mentser and Sharkey investigated the composition of
diesel particulates as a function of particle size. They
used seven fraction sizes ranging from <0.2 to X3.0 ym in
diameter. The calculated effective aerodynamic diameters of
the particulate fractions are shown in Table 3-3. Weight
measurements of particulates in the seven stages indicated
that more than 50 percent of the total mass of particulates
in any given experiment were collected on the backup filter
(stage 7). Particle size distributions from a light-duty
diesel engine were also measured by Laresgoiti et al.*4
3-4
-------
Table 3-2. ELEMENTAL AND TRACE METAL COMPOSITION OF DIESEL EXHAUST PARTICULATES'
Reference
6
8
Engine
D.D.A.D 6V-71
Caterpillar 3208
Detroit Diesel
6L-771T
Cummins NTC-290
Fuel
EM-238-F
EM-239-F
EM-240-F
EM-241-F
EM-242-F
EM-238-F
EM-239-F
EM-240-F
EM-241-F
EM-242-F
1-D
2-D
1 1/2-D
1-D
2-D
1 1/2-D
Average weight % by elements
Carbon
80.6
83.9
79.6
86.6
84.9
87.2
84.5
85.2
74.9
79.7.
b 69.7
c 85
b 68.2
c 82
b 74.2
c 80
b 78.2
c 70
b 60.9
c 84
b 80.5
c 78
Hydrogen
10.7
10.9
12.2
10.5
9.8
1.9
2.2
2.1
2.9
1.6
10.2
13
10.2
12
10.8
12
4.7
11
3.3
12
8.1
11
Nitrogen
3.2
1.4
a
a
1.5
0.5
0.4
0.1
0.9
0.8
0.1
0.1
0.4
0.2
0.2
0.4
1.7
3.6
0.2
0.6
0.9
0
Sulfur
1.01
0.79
0.32
0.90
0.83
1.90
1.47
0.46
1.66
1.67
0.6
0.1
2.3
0.8
1.0
0.1
2.8
a
4.3
a
3.3
a
Oxygen
0.2
4.5
a
a
a
a
E
95.5
97.0
92.1
98.0
97.0
91.5
88.6
87.9
80.4
83.8
80.6
98.4
81.1
99.5
86.2
92.5
87.4
84.6
68.7
96.6
92.8
89.0
(jj
Ul
None detected or trace.
Percent of element in total particulates.
Percent of element in organic soluble fraction of particulars.
(continued)
-------
Table 3-2 (continued)
Reference
Engine
Cummin s
NTC-290
Detroit Diesel
6L-71T
Cummins
NTC-290
Fuel
2-D and 1 1/2-D plus 0.25% (vol) smoke suppressant
additive
Ca: 3.6 - 12 g/hr
Ba: 1.7 - 2.1 g/hr
2
Trace metal analysis in vig/cm of collection filter
A. Fuels without additives: ,
Ca: 0 - 3.10 ug/cm.
Cu: 0 - 0.12 ug/cm,
Zn: 0.15 - 4.02 Kg/cm,
Pb: 0 - 0.48 ug/cm2
Sr: 0 - 0.48 ug/cm
Ba: 0
B. Fuels with smoke suppressant additives:
Ca: 1.66 (idle) - 52.57 ug/cm
Cu: 0 ug/cm^
Zn: 0.07 - 1.44 ug/cmf
Pb: 0 - 0.34 ug/cm-
Sr: 0 - 0.15 ug/cm
Ba: trace (idle) - 7.66 ug/cm
V : 0 - 0.42 ug/cm2
A. Fuels without additives:
2
Ca: 0 - 1.61 ug/cm_
Mn: 0 - trace ug/cm2
Cu: 0 - 0.10 ug/cm-
Zn: 0 - 1.89 ug/cm,
Pb: 0 - 0.72 ug/cm,
Sr: 0 - 0.10 ug/cm2
Ba: 0 ug/cm
B. Fuels with smoke suppressant additives:
2
Ca: 3.86 - 58.58 ug/cm-
Mn: 0 - 0.28 ug/cm,
Cu: 0.11 - 0.18 ug/cmf
Zn: (idle)- 0.52 ug/cmf
Pb: - - 0.42 ug/cm
Sr: 0.05 - 0.19 ug/cm2
Ba: 1.35 - 8.94 ug/cm*
2
2
A
(continued)
-------
Table 3-2 (continued)
Reference
9
10
Engine
Detroit Diesel
6L-71T
Detroit Diesel
6V- 71
Nissan
LDMV
Opel
LDMV
Nissan
LDVM
Fuel
2-D and 1 1/2-D plus 0.25% smoke suppressant
additive
Ca: 7.7 - 14 g/hr
1.1 - 3.4 g/hr
5 Fuels Trace elements in particulates
Pb: 5.3 - 6.6 ug/filter
Mn: 4.0 - 4.0 ug/filter
Hg: 3.4 (one fuel) ug/filter
P : 0.6 - 1.6 ug/filter
S s 2.5 - 14 pg/ filter
Na: 0.29 (one fuel) ug/filter
Zn: 1.0 - 1.3 ug/filter
Cu: 1.5 - 11 wg/f ilter
Ca: 1.2 - 2.8 ug/filter
V : 0.44 - 0.73 ug/filter
1 Fuel Weight % of element in exhaust
particulates
C : 70.42 - 72.84
H : 0.43 - 2.22
N : 5.51 - 8.81
Fe: 0.13 - 0.15
Cu: 0 - 0.02
Zn: 0.07 - 0.16
S : 0.51 - 1.12
1 Fuel Weight % of element in exhaust
particulates
C 72.4 - 77.9
H 4.7 - 6.1
N 0.9 - 3.5
Fe 0.13 - 1.08
Cu 0 - 0.01
Zn 0.16 - 0.29
P 0 0.09
S 0.49 - 1.05
3 Fuels Weight % of element in exhaust
particulates
C t 69.2 - 76.6
H : 1.4 - 2.0
. 1 _..~ J \
-------
Table 3-2 (continued).
Reference
11
12
Engine
Single Cylinder
Diesel trucks
(highway)
Fuel
1 Fuel Weight t of element in ashed participates
Si: 0.5 - 0.75
Fe: 0.1 - 0.35
Ca: 0.02 - 0.5
Ba: 0.02 - 0.5
Cr: 0.001
Cu: 0.0005- 0.001
Ti: 0.001
Unknown Ba emission rates from trucks estimated
to be 0.001 to 0.0015 g/mile. Assumes
Ba emitted from diesel fuel and crankcase oil.
I
00
-------
Their results, summarized in Figure 3-1, indicate that about
90 percent of the particles were smaller than 1 ym and 99
percent were smaller than 2 ym. Data from a study conducted
by the Bureau of Mines (Table 3-4) suggest that the mass
median diameter (HMD) of diesel exhaust particulates is
approximately 0.3 ym, and that (in two engines tested) fuel,
operating cycle, and engine type had little effect on
exhaust particulate size.
Table 3-3. AERODYNAMIC DIAMETERS OF DIESEL EXHAUST
PARTICULATES COLLECTED IN VARIOUS STAGES OF AN
ANDERSON SAMPLER133
Sample no.
81, C 91d
82, 92
83, 93
84, 94
85, 95
86, 96
87, 97
Filter stage
1
2
3
4
5
6
7 (backup)
Aerodynamic ^
diameter, ym
>3.0
2.0
1.3
0.8
0.4
0.2
<0.2
Data supplied by A. J. Strazisar of PMSRC.
Effective aerodynamic diameter calculated for the
following sampling conditions: gas flow, 3.0 ftj/min;
gas temperature, 100°F; impaction efficiency, 50
percent.
Engine mode for samples 81-87: 2200 rpm, full load.
Engine mode for samples 91-97: 600 rpm, no load.
3-9
-------
e
a
6.0
5.0
4.0
3.0
2.0
1.0
0.7 -
I 1
I
I
0.001 0.01 0.1 1.0 10.0
NUMBER S OF PARTICLES OF DIAMETER > D
20.0 30.0
Figure 3-1. Particle size distribution.
14
Circles: average of 24 data points for speeds ranging from
800 to 3100 rpm and for loads ranging from 0 to 75% of full
load. Bars: data scatter.
3-10
-------
Table 3-4. DIESEL EXHAUST PARTICLE SIZE
CO
I
Reference
15
Engine type
Caterpillar 6-cyl.
four-cycle 1D1
Cumins 6-cyl
four-cycle Dl
Operating mode
Idle
Intermediate speed
no load
Rated speed 9 half
load
Rated speed t full
load
Intermediate speed 0
full load
Idle
Intermediate speed t
no load
Rated speed 1 half
load
Rated speed 9 full
load
Intermediate speed 9
no load
Fuel
2-D
Heavy 2-D
1-D
2-D
Heavy 2-D
1-D
2-D
Heavy 2-D
1-D
2-D
Heavy 2-D
1-D
2-D
Heavy 2-D
1-D
30 Test averag<
- 30
2-D
Heavy 2-D
1-D
2-D
Heavy 2-D
1-D
2-D
Heavy 2-D
1-D
2-D
Heavy 2-D
1-D
2-D
Heavy 2-D
articulate size,
um HMD
0.25
0.28
0.32
0.27
0.36
0.24
0.35
0.40
0.30
0.29
0.40
0.47
0.25
0.24
0.19
0.44
0.32
0.47
0.39
0.35
0.43
0.20
0.29
0.22
0.20
0.23
0.27
0.29
0.35
-------
Participates indiscriminately adhere to solid surfaces
and to each other. Among the properties of particles that
influence the strength of the adhesive bond are chemical
composition, the presence or absence of moisture or oily
films, electrical charge, and physical characteristics.16
The force of the adhesion of one particle to another cannot
be reliably predicted now, but simple test methods are
available to determine this. It is generally assumed that
airborne particles that contact each other continue to
adhere, i.e., the "collision efficiency" is 100 percent.16
If it is assumed that diesel exhaust particulates behave
similarly, the size distribution of these particulates could
vary depending on the site and time of measurement. Avail-
able evidence indicates that most diesel particulates at
the tailpipe fall within a relatively small size range.
This distribution, however, may not accurately characterize
the size of the diesel exhaust particulates in the ambient
environment. The adherence of diesel particulates to one
another and their interaction with other atmospheric par-
ticles may result in the formation of larger particles.
Coalescence of solid particles results in flocculent, iso-
metrically shaped, or threadlike aggregates. These particle
dynamics are illustrated in Figure 3-2. Although these
data depict the impact of gasoline vehicle emissions, the
3-12
-------
pv '
DUftINQ RUN
tACKONOUMO AFTM HUN
00-4*
0««OSI
JO- I
DO 0 ttt
OOt 0.01
100
Trimodel model particle distribution measured during and after
vehicle proving grounds tests. Note that during the test the
accumulation and coarse particle modes (center and right modes)
have not changed significantly from the background conditions.
On the other hand, practically all of the volume of the nuclei
mode (left mode) is contributed by the cars on the roadway.
Schematic of a trimodal atmospheric aerosol size distribution
showing the principal modes, main sources of mass for each
mode, and the principal processes involved in inserting mass
and removing mass from each mode.
Figure 3-2. Impact of vehicle exhaust on ambient
particulate size distribution data for gasoline-
powered vehicles.
3-13
-------
particle dynamics of diesel vehicle emissions should be
similar. The impact of particle dynamics on pulmonary
deposition and retention and ultimately on health, cannot be
predicted until reliable quantitative models are developed.
Figure 3-3 illustrates particle deposition probabil-
ities as a function of particle size as they relate to
respiratory regions. This size-deposition relationship
depicts average particle deposition probabilities for a man
breathing spontaneously under sedentary conditions. Al-
though deposition probabilities for the submicronic range
are theoretical, particle size-deposition relationships can
be used in risk evaluations of particulate exposure because
they provide useful models for intake or dose estimations
and are helpful in understanding pulmonary clearance pro-
cesses.
As previously mentioned, many diesel exhaust particu-
lates have diameters of less than 1 pm. Submicronic par-
ticles easily penetrate all parts of the respiratory system.
They continually undergo Brownian motion deposition, which
predominates in the alveolar region, although some of them
remain airborne and are expelled.
Particulate clearance from the lung parenchyma (alve-
oli) seems to involve an absorptive mechanism whereby the
particle or its dissolved phase moves into the blood or
3-14
-------
I I
! I
I
(-1
(Jl
1.0
10
,-3
DIFFUSION
SEDIMENTATION
INERTIAL INPACTION
10
-2
10'1 10" 10'
AERODYNAMIC DIAMETER, urn
Figure 3-3. Particle size deposition probabilities
18
-------
lymph. This mechanism appears to depend on permeability
considerations and on endocytosis; the latter is the proc-
ess, including pinocytosis and phagocytosis, whereby foreign
materials are engulfed by migratory cells such as pulmonary
macrophages. Further details of the alveolar clearance
mechanism, especially quantitative data, are lacking. It is
known, however, that clearance of insoluble particles from
the alveoli may vary from hours to years, depending on the
particulate material. Increased deposition and retention in
these susceptible tissues become important when one realizes
that even an inert carrier substance can contain potentially
toxic materials, either in the particle or adsorbed on it.
Such materials include PNA's or sulfates. Once such com-
pounds are deposited in the alveoli, they cannot be re-
suspended easily and may not be cleared or metabolized for
19
a long time, if at all. This increases the likelihood of
chemically inducing disease at critical sites in the body.
The health implications of particulate emissions thus
appear to depend not only on particle size and deposition,
but also on the chemical nature of the particles.
Many of the data on particulates in diesel emissions
are limited to qualitative determinations of organic com-
ponents, or to descriptive analytical procedures that tend
to emphasize technique and instrumentation rather than
3-16
-------
composition. Review of the literature, however, suggests
that diesel engine particulates are not well characterized,
either chemically or physically, and emission factors are
uncertain.
3.2 POLYCYCLIC ORGANIC MATTER
Polycyclic organic matter (POM) is defined as organic
matter that contains two or more ring structures which may
33
or may not be substituted by other chemical groups. Any
combustion process involving fossil fuels or compounds
containing carbon and hydrogen can form POM. The amount
formed in a given combustion process depends on the effi-
ciency of the process. Polycyclic organic matter can be
further separated into the particulate phase (PPOM) or
vaporous phase (VPOM). One group of aromatic compounds of
PPOM, the PNA's, is particularly important because it in-
cludes several carcinogenic materials.
Polynuclear aromatic hydrocarbons have been detected in
fractions of PPOM obtained from diesel exhaust. '
They are believed to result from 1) incomplete combustion of
materials in the fuels, 2) synthesis of aromatic hydro-
carbons of lower molecular weight, and/or 3) pyrolysis of
lubricating oil. The first of these sources is believed to
be the most important.20 Table 3-5 lists PNA's isolated
from diesel exhaust particulates that demonstrate some
3-17
-------
OJ
I
\->
CO
Table 3-5. FREQUENCY OF OCCURRENCE OF FORMULAS FOR CARCINOGENIC
COMPOUNDS IN DIESEL EXHAUST PARTICULATES13
Carcinogenicities are given in Ref. 25, according to the following code:
+ uncertain or weakly carcinogenic
+ carcinogenic
**» +**» +-M-+, strongly carcinogenic.
Formula
C18H12
C20H12
C20H14
C20H16
C21H14
C20H13N
C22H12
C22H14
Carcinogenic coapound with
corresponding formula
Chrysene
Benzo [cjphenanthrene
Benz [a] anthracene
Benzo [ a ] pyrene
Benzo [bj f luoranthene
Benzo [ j ] f luoranthene
Benz [ j ] aceanthry lene
7 , 12-Dimethylbenz [a ) anthracene
Dibenzo [a, g] f luorene
Dibenzo [c, g] carbazole
Indeno (1,2, 3-cd ] pyrene
Dibenz [a , h] anthracene
Dibent [a , j ] anthracene
Dibenz [a , c j anthracene
Carcinogenicitya
+
+"++
+
+++
+4
++
»»
f+4-t-
+
+ »»
f
»+-(
f
f
Molecular
weight
228.0936
252.0936
254.1092
256.1248
266.1092
267.1045
276.0936
278.1092
Frequency of occurrence,
30 samples
28
16
2
1
2
1
3
2
-------
degree of carcinogenic activity. Most of these compounds
have molecular weights from 228 to 302 and frequently exist
in several isometric forms. The frequency with which they
appeared in samples of diesel exhaust particulates is given
in the last column of the table. PNA formulas C18H12 and
C H were by far the most prevalent. The first, corre-
£*\J JL£
spending to chrysene or its isomers, occurred in 28 of 30
exhaust particulates. The second formula, benzo[a]pyrene or
its isomers, occurred in 16 of 30 samples.
o n
Table 3-6, compiled by Lyons, lists PNA compounds
detected in samples of various atmospheric pollutants. The
author noted that several compounds possess the anthracene
stem as part of their structural configuration. Polycyclic
hydrocarbons occurring in highest concentration in the three
soot extracts appeared to have two to five condensed rings.
Primarily because of its potent carcinogenicity and
frequency of occurrence, BaP has typically been measured and
used in vehicular emission research and air pollution moni-
toring as an indicator of total PNA concentration. Con-
sequently, the bulk of available data is in terms of BaP.
As the above examples have indicated, other polycyclic
organic materials of similar structure and carcinogenicity
occur in vehicular exhaust emissions, but reliable quan-
titative data for BaP and other PNA's from diesel engine
3-19
-------
Table 3-6. COMPOUNDS DETECTED IN VARIOUS ATMOSPHERIC
POLLUTANTS (G), (D) , AND (A)3 SAMPLES21
Compound
Naphthalene
Acenaphthylene
Anthracene
Phenanthrene
Anthracene derivatives
Pyrene
Fluoranthene
Alkylpyrene
Benz la] anthracene
Chrysene
Renzo(e]pyrene
Perylene
nenzo[a] pyrene
Ben zo [ ghi) perylene
Benzolb] f luoranthene
7>nthanthrene
Tetracene
Coronene
nibenz I a, hi anthracene
4- Dibenzo [a , 1 ] pyrene
BenzoJK] f luoranthene
Pentaphene
Dibenzo ( a , 1 Inaphthacene
Dibenzo [ a, h) pyrene
Dibenzo [a, e] pyrene
Dibenzo [b , pqr ) perylene
(Dibenzof luorene?)
Tribenzolh.rstlpentaphene
Indeno- 1,2, 3-f luoranthene?
G
4
+
+
-
4
4
+
+
4
+
4
+
+
f
-f
+
+
+
4
4
4
4
+
+
+
+
-
4
-
D
-
4
*
*
4
+
+
-
+
-
+
+
+
4
4
4
-
4
-
4
+
+
-
-
-
-
4
-
f
A
-
4
4
-
4
4
+
-
4
4
+
4
4
4
4
4
-
4
-
-
4
-
-
-
-
-
4
-
-
C: gasoline Boot (ample
D: diesel Boot ample
A: general atmospheric coot sample.
4: detected in cample
-: not detected in sample
Methodology: fluorescence and UV and visible absorption
analysis of particulat* extracts following chromatoqraphic
fractionation.
3-20
-------
exhaust are lacking. Although BaP could be used as an
indicator molecule of urban pollution, its use as an ac-
curate index of total PNA emissions from a single source
such as diesel exhaust is questionable. Nonetheless, total
PNA1s in vehicular exhaust emissions continue to be esti-
mated and are often expressed solely on the basis of BaP.
Polyaromatic hydrocarbon emissions are thought to be
related to fuel and lubricating oil composition and combus-
tion efficiency. Analysis of diesel fuels for PNA compounds
has shown that diesel fuels tend to contain lower concen-
trations (up to 422 ppb of BaP) of PNA compounds than gaso-
line (up to 3000 ppb of BaP).21 Although one might expect
higher concentrations of PNA's in the less volatile diesel
fuel, gasolines actually contain much higher concentrations
of catalytically processed aromatic hydrocarbons.
Table 3-7 shows results of tests for BaP emissions from
a single diesel car compared with results from three
gasoline cars without catalysts. 3f2 Emission levels of
BaP from both combustion sources are about the same (i.e.
1.57 and 1.95 yg/mile). Use of oxidative catalysts and
other pollution control devices has reportedly reduced all
PNA emissions from gasoline-fueled engines by about 99
percent, which would place a catalyst-equipped car well
below a diesel as a source of PNA's. These results should
3-21
-------
be interpreted with caution because the sampling and anal-
ytical procedures used by the different investigators were
not uniform.
Table 3-7. BENZOfA]PYRENE EMISSIONS FROM DIESEL
VERSUS GASOLINE CARS1
Vehicle type
BaP emissions
(FTP) , yg/mile
Peugeot 504 diesel
4
Average of three 1969-72 noncatalyst
gasoline
1.57
1.95
Note: Data on catalyst-equipped gasoline cars were not
available for the same test procedure, but in-
vestigators using a different test procedure20
have observed about 99 percent elimination of
all PNA emissions with catalysts.
It has been clearly established that carcinogenic PNA's
are emitted in the form of particulate matter from gasoline
23 25
and heavy-duty, diesel-powered engines. ' It is xmcer-
tain whether polycyclic organic matter condenses out as
discrete particles after cooling, or condenses on surfaces
of existing particles after formation during combustion.
Light-duty diesel PNA characterization is even less well
defined and is currently part of an important investigation
2 6
by the U.S. Environmental Protection Agency. It is
thought that PNA emissions may be chemically combined, i.e.
adsorbed, with particulate matter simultaneously emitted
from light-duty diesel engines. The significance of this
from the standpoint of health effects has not yet been fully
3-22
-------
elucidated, but it seems likely that carcinogenic PNA
compounds adsorbed on fine particulates could be inhaled and
brought into effective contact with susceptible cells of the
lining of the tracheobronchial tree and parenchyma.
Some of the factors affecting pulmonary deposition have
already been discussed. In certain experimental animals,
presumably inert carrier substances have been shown to have
an important effect on the concentration time determinants
2 8
of toxic inhalants. Experiments by Boren have shown that
carbon functioning as an absorbent greatly increases the
damaging action of nitrogen dioxide on the lung. When
tritiated BaP is incorporated in carbon or asbestos, clear-
29
ance from the lungs of hamsters is slowed. An increase in
the carcinogenic effect of BaP by means of carbon and
-jn oc 32
carrier particles ' and hematite has also been demon-
strated. These experimental results are interesting in view
of the observation that particulate material emitted by
diesel engines is largely carbonaceous and contains car-
cinogenic PNA's as well as other potentially toxic com-
pounds. Although some experimental evidence suggests car-
rier substances can affect pulmonary clearance mechanisms,
27
their exact role can only be surmised.
There is clear evidence that occupational exposure to
airborne particulate organic matter, particularly the poly-
3-23
-------
nuclear aromatic fraction, is responsible for specific
27
adverse biological effects in man. These effects include
cancer of the lungs and skin, nonallergic contact derma-
titis, photosensitization reactions, hyperpigmentation of
the skin, folliculitis, and acne. In concentrations found
in the atmosphere, PPOM does not appear to cause any of
these cutaneous effects; similarly, there is no cleair
evidence that, by themselves, such materials as airborne BaP
directly influence the pathogenesis of nonneoplastic lung
diseases (e.g. bronchitis and emphysema).
Many screening methods have been used to evaluate the
carcinogenicity of PPOM. They have utilized pure samples of
organic compounds of the types found in the ambient environ-
ment, total PPOM and fractions collected from urban atmo-
spheres, as well as organic fractions isolated from combus-
tion sources. The carcinogenic potential of pure PNA's and
extracts of airborne materials has been tested on various
whole animals, tissue cultures, organ cultures, and micro-
organisms. Methods employed on whole animals included skin
painting, subcutaneous injection, systemic inoculation, oral
intake, local implantation (in lung, bladder, or other
organs), intratracheal inoculation, and inhalation. Animal
and bioassay data relating to the toxicity of PPOM are
briefly reviewed in a scientific and technical assessment
3-24
-------
report published by the U.S. Environmental Protection
33
Agency.
Both animal experiments and epidemiologic data indicate
that pulmonary cancer of environmental origin involves a
complex series of factors and events in which PNA's con-
stitute only one of the carcinogenic factors. The possibil-
ity of synergistic or cocarcinogenic effects of other en-
vironmental agents must also be considered. Irritant or
toxic gases (e.g. SO-, NO , and ozone), existing in various
£ jC
concentrations in the atmosphere, are known to have a poten-
tiating action on the carcinogenic properties of PNA's.
This has been demonstrated by the higher incidence in CAF/
Jax mice of pulmonary adenomas produced by simultaneous
exposure to ozone and carcinogens. Work by Laskin et al.
suggests additive or potentiating effects of SC>2 in BaP
carcinogenicity in rats. Individual susceptibility to the
carcinogenic action of PNA's can also be influenced by
smoking habits, occupational exposures, age, and coexistent
viral or other pulmonary diseases.
Examination of epidemiologic studies suggests that
there is an "urban factor" in the pathogenesis of lung
cancer in man. Although a major factor in the causation of
human pulmonary cancer is cigarette smoking, it alone does
not account for the increased incidence of this disease. It
3-25
-------
appears that the incidence of lung cancer among urban
dwellers is twice that of those living in rural areas;
within urban communities, the incidence is even greater
where fossil-fuel emission products are highly concentrated
27
in the air. A strong link between cancer mortality and
nearness to traffic has been reported in a study by Blumer
38
gt al. This epidemiological study, which is based on a
population study of a Swiss mountain town from 1958 through
1970 , found death from cancer nine times more frequent among
those who lived near the local highway than those who lived
440 yards or more away. The level of PNA's was very high in
soil near the highway (300 mg/kg) and less abundant farther
away (4 to 8 mgAg) - The composition of the PNA's in soil
samples resembled that of PNA's in automobile exhaust. Low
values in town and close to industry but remote from the
highway and high PNA values outside of town but near the
highway suggest a correlation between automobile traffic and
PNA content of soils. These results also indirectly suggest
a correlation between automobile traffic and the observed
mortality from cancer in this area. Although mortality data
on lung cancer are not specific and etiologic factors and
PNA emission sources were not conclusively determined., the
public health implications and the need for efforts to
control engine exhaust are considerable.
3-26
-------
Polycyclic hydrocarbons have not been shown to be
teratogenic, although a number of other chemical carcinogens
exhibit this biologic action. The teratogenicity of commun-
ity atmospheric pollutants and defined components thereof
has not yet been tested in mammalian species by inhalation
or by parenteral administration.
No mutagenic effects from PPOM or its PNA components
have been found in animals in vivo, but studies in this area
have not been extensive.33 One might postulate that urban
susceptibility to carcinogens may have been induced by
mutagenic mechanisms over several generations; however, pure
samples of a few selected PNA's of the types found in diesel
exhaust and organic fractions of collected diesel particu-
lates have demonstrated mutagenic activity in in vitro
bioassay systems. Improved and simplified techniques, such
as the Ames mutagenesis bioassay, are expected to yield
significant information on genetic variations and, ulti-
mately, on cellular mechanisms of cancer. Preliminary
results obtained from experiments in which seven diesel
exhaust fractions were tested in the Ames system are de-
scribed in Section 3.5.
3.3 SULFATES
The contribution of diesel-powered passenger cars to
the emission of sulfates, or sulfuric acid, is of interest
3-27
-------
in view of the considerable attention given to this subject
since the advent of catalyst-equipped automobiles. Table
3-8 shows values of sulfate emissions for the same group of
diesel-powered automobiles discussed in the section on
particulates. Sulfate emissions were measured using the
same type of dilution tunnel and filtration system developed
for particulate measurements, but with a different driving
cycle. This was developed especially to represent the
conditions under which sulfate emissions cause the highest
local exposures to people. Comparison with the average
values for gasoline cars with and without catalysts shows
that the diesels fell between the low extreme represented by
the noncatalyst and three-way catalyst cars, and the high
extreme represented by the catalyst cars with air injection.
The values for diesels tended to increase in proportion to
vehicle size, which is reasonable because this is the order
of increasing fuel consumption. Each diesel car apparently
converted about the same fraction of the total sulfur in the
fuel to sulfates (about 2 percent). The fuel used in the
diesel car test work was a typical No. 2 diesel fuel con-
taining 0.228 percent sulfur by weight.
Typical diesel fuel reportedly contains about eight
times the amount of sulfur of typical gasoline.2 Because
sulfate emissions tend to be proportional to fuel sulfur
3-28
-------
level, a reduction in the sulfur level of diesel fuel would
reduce sulfate emissions.
Table 3-8. SULFATE EMISSIONS FROM DIESEL VERSUS
GASOLINE PASSENGER CARS
Vehicle type
Diesel vehicles:
VW Rabbit
4
Peugeot 504
4
Mercedes 240D
Mercedes 300D
Oldsmobile 3503
Gasoline vehicles:
39
Average noncatalyst car
Average catalyst car with
air injection^?
Average catalyst car without
air injection-39
39
Three-way catalyst car
Sulfate
(SET) , g/mile
0.007
0.007
0.014
0.016
0.017
about 0.001
about 0.030
about 0.008
about 0.001
Although the sulfur emitted by gasoline-powered cars is
essentially all sulfuric acid, it is not known what types of
sulfate are emitted by diesels nor how much of it is sul-
furic acid. Because the health effects of exposure to
sulfuric acid differ from those of exposure to other sulfur
compounds, the impact of given amounts of diesel sulfates in
terms of health effects cannot yet be predicted.
3-29
-------
There is no basis as yet for predicting whether in-
creased use of diesel-powered cars would cause a net in-
crease or decrease in sulfate emissions, because it is not
known whether future gasoline-powered cars will use pre-
dominantly three-way catalysts (low sulfate emissions), or
oxidizing catalysts with air injection (high sulfate emis-
sions) .
3.4 MINOR COMPONENTS
Aldehydes and other oxygenates of low molecular weight,
as well as aliphatic, phenolic, and light aromatic hydro-
carbons, are minor volatile components of diesel exhaust
that are most probably emitted in the vapor phase. Since
this assessment is restricted to diesel exhaust particulate
and its major components, potential health effects asso-
ciated with specific vapor constituents are not discussed.
It should be noted, however, that potentiating or cocarcino-
genic effects have been attributed to some of these com-
pounds. Thus, the toxic potential of PPOM in diesel exhaust
could be influenced by other emission factors. The possi-
bility of adsorption and absorption of gaseous and condensed
substances on diesel particulates requires further study.
3.5 CURRENT RESEARCH STATUS
Identification and quantitative analysis of potentially
toxic components of diesel exhaust particulate, and sub-
3-30
-------
sequent testing of these compounds in their pure form under
laboratory conditions, is a common method of estimating the
toxicological impact of an emission source. Chemical
characterization and toxicological testing of all poten-
tially toxic compounds in diesel participates, however,
would be an immense, time-consuming, and probably futile
task. Because of this, faster, less expensive screening
procedures are currently used to establish priorities for
physical and chemical characterization studies and addi-
tional toxicologic tests. The Ames Salmonella/microsome
assay has gained wide use as a quick, economical, and reli-
able screening method to determine if a chemical agent or
mixture is mutagenic or likely to cause cancer. In whole
animal studies, the Ames assay procedure has been shown to
be 85 to 90 percent accurate in detecting substances that
are carcinogenic; and it has about the same accuracy in
identifying substances that are not carcinogenic in animals,
In a joint ESRL/HERL (RTF) project, seven diesel ex-
haust fractions obtained from both a two-stroke and a four-
stroke diesel engine were tested for mutagenicity with the
40
Ames test (by V. Simmon, Stanford Research Institute).
Preliminary data indicate that several fractions of diesel
exhaust particulate are mutagenic. These findings are not
altogether unexpected, because previously reported studies
3-31
-------
had identified chemicals in engine exhaust products that are
known to be mutagenic or carcinogenic.
The fractions were examined with five tester strains of
Salmonella typhimurium (TA 1535, TA 1537, TA 1538, TA 98,
and TA 100) with and without the standard liver microsomal
metabolic activation system. The experiments were conducted
in a dose response fashion (6 to 8 doses/fraction per tester
strain), and each experiment was repeated (except in 3 of
the 14 samples, where the sample size was limiting).
The most mutagenic fraction was the transitional hydro-
carbon (TRN) fraction, which produced a 20-fold increase
over spontaneous mutation rates in TA 1538 at approximately
40 yg/plate. (The transitional hydrocarbon fraction was
described as that portion of the neutral fraction composed
of the middle polar species.)* The oxygenated hydrocarbon
(OXY) fraction produced a 20-fold increase over spontaneous
mutation rates at 100 yg/plate. These results are from the
four-stroke engine samples. Similar data and other experi-
mental details are available for the two-stroke engine
samples. The order of mutagenic activity for the most
active fractions was the same for both engines: TRN>OXY>ACD
(acidic).
Telephone conversation with Prank Black, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina
on December 8, 1977.
3-32
-------
These fractions were mutagenic without metabolic acti-
vation, indicating the predominance of direct-acting itmta-
gens (e.g. the 7, 8-diol-9, 10-epoxide of benzo[a]pyrene)
that do not require enzymatic conversion as provided by
microsomal metabolic activation. In several fractions, the
microsomal metabolic activation increased the mutagenic
activity above that observed in the absence of microsomal
activation, indicating the presence of lesser amounts of
compounds that require metabolic activation (e.g., benzo[a]-
pyrene).
Mutagenic activity was primarily observed with tester
strains that respond to frameshift mutagens (e.g. ICR-191,
benzola]pyrene, aflatoxin B,., and 7,12-dimethylbenzla]-
anthracene).
These results indicate the utility of such in vitrp
bioassays to direct the fractionation of diesel particulate.
If the resources and personnel were available to pursue this
approach, it should be possible to identify the mutagenic
components of diesel particulates. Compounds recently
identified in the TRN fraction, the fuel, or whole exhaust
are now the subject of a literature search pertaining to
microbial mutagenesis. In addition to further fractionation
and bioassay, additional development work is required to
bioassay either the crude particulate itself or a simple
3-33
-------
extract. This would allow evaluation of the relative
mutagenic activity of a variety of engines, fuels, pollution
control devices, etc.
Several selected fractions of the diesel exhaust parti-
culate will be evaluated by other confirmatory bioassays,
such as the mammalian cell mutagenesis and neoplastic trans-
formation methods. Experiments are also in progress to
evaluate the relative toxicity of these fractions in several
in vitro systems. If suitable in vitro procedures can be
developed, comparative studies of various diesel and gaso-
line engines will be consucted.
Because the biological impact of isolated chemical com-
ponents of a mixture is different from that of the mixture
as a whole, inhalation studies are being conducted in which
animals are exposed directly to whole diesel exhaust. In this
series of experiments, a wide variety of biological param-
eters are being measured to determine the effects of the
exhaust on the respiratory systems of several mammalian
species. Acute, subacute, and chronic inhalation exposures
are being conducted using appropriate dilutions of diesel
exhaust emissions. Metabolic effects of the emissions are
examined in terms of their specific biochemical reactions
with lung tissue, alveolar macrophages, and subcellular
organelles. Early biochemical changes that precede the
3-34
-------
appearance of overt symptoms of toxicity or a disease state
may eventually prove to be clinically significant. In
addition, an inherent part of this approach is the con-
sideration of any additive, synergistic, potentiating,
and/or cocarcinogenic effects from exposure to mixtures of
diesel exhaust components and particulates. Results ob-
tained from inhalation studies using animals and from
in vitro bioassays will provide an estimate of the degree of
toxicity associated with diesel exhaust and components
thereof and help define potential health hazards.
A substantial amount of research is now either under
way or being planned to obtain more information on diesel
exhaust particulate emissions from both light- and heavy-
duty diesel-powered vehicles. This work includes a) the
collection of particulates and isolation of organic soluble
components of particulates from a variety of diesel-powered
vehicles under various driving schedules and fuel combina-
tions; b) determination of the biological activity of the
total particulates, the total organic extract of the parti-
culates, and the individual fractions of the extract in
various biological test systems; cl chemical characteriza-
tion of those particulate fractions that show activity in
various biological test systems; d) calculation of emission
rates of these exhaust products for various vehicles,
3-35
-------
driving schedules, and fuel combinations; and e) prediction
of likely concentrations of these biologically active frac-
tions on and near the roadway and in the ambient environ-
ment. In the next 6 to 9 months, appropriate agencies
should have further data from which to draw additional
conclusions.
Scientific evidence, both direct and indirect, indi-
cates that diesel exhaust particulate emissions pose a toxic
hazard to humans. Chemical analysis of diesel exhaust
particulates reveals the presence of a number of scientifi-
cally recognized carcinogens. Diesel exhaust particulates,
on which carcinogenic PNA's may be adsorbed, are well within
the respirable size range. Several diesel exhaust fractions
have demonstrated mutagenic activity in in vitro bioassay
systems. The studies and short-term tests performed thus
far have helped characterize the diesel particulate and have
identified chemical components or fractions thereof with
toxic, carcinogenic, or mutagenic activity. However, these
studies alone do not provide sufficient data to make a
definitive estimate of the public health risk, if any, that
may be associated with emissions from diesel-powered vehi-
cles. Chronic whole animal exposure studies and/or human
epidemiological data are generally required to perfoarm such
health risk assessments. The data that are just emerging
3-36
-------
from ongoing and planned research efforts will permit this
extremely important risk assessment to be performed and
provide a basis for establishing scientific and rational
environmental quality standards for diesel exhaust emis-
sions .
3-37
-------
REFERENCES FOR SECTION 3
1. Kittredge, G. Emissions of Unregulated Pollutants from
Diesel Engines Used in Highway Vehicles. Internal
Report. U.S. Environmental Protection Agency, Washina-
ton, D.C. 1977. y
2. Office of Mobile Source Air Pollution Control (AW-
455). Emission Impacts of Diesel-Powered Light-Duty
Vehicles. Internal Report. U.S. Environmental Protec-
tion Agency, Washington, D.C. September 1977.
3. Unpublished data. EPA contract with Southwest Research
Institute, San Antonio, Texas. iCited in (1)]
4. Braddock, J.N., and P.A. Gabele. Emission Patterns of
Diesel-Powered Passenger Cars - Part II. SAE No.
770168. Society of Automotive Engineers, Inc.,
Warrendale, Pennsylvania. 1977.
5. Springer, K.J., and R.C. Stahman. Diesel Car Emis-
sions - Emphasis on Particulate and Sulfate. SAE No.
770168. Society of Automotive Engineers, Inc., Warren-
dale, Pennsylvania. 1977.
6. Hare, C.T. Characterization of Diesel Gaseous and
Particulate Emissions. Final Report, Contract No.
68-02-1777, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina. 1977.
7. The Health Implications of the Use of Diesel Engines in
Underground Coal Mines. (Unpublished report). Na-
tional Institute for Occupational Safety and Health
Morgantown, West Virginia. 1977.
8. Hare, C.T. Methodology for Determining Fuel Effects on
Diesel Particulate Emissions. EPA-650/2-75-056. U.S.
Environmental Protection Agency. March 1977 [Cited*
in (7)]
9. Characterization of Diesel Gaseous and Particulate
Emissions. Preliminary data from Monthly Progress
3-38
-------
Reports, Contract No. 68-02-1777, U.S. o
Protection Agency, Research Triangle Park, North Caro-
lina. 1977. ICited in (7)]
10 Annual Catalyst Research Program Report. EQA-600/
3-75- 010C. U.S. Environmental Protection Agency.
September 1975. ICited in (7)1
11 Frey, J.W., and M. Corn. Physical and Chemical Charac-
teristics of Participates in a Diesel Exhaust. Am.
Ind. Hyg. Association J. September-October 1967.
12. Pierson, W.R., and W.W. Brochaczek Pa^iculate Matter
Associated with Vehicles on the Road. SAE No. 760039,
Society of Automotive Engineers, Inc., Warrendaie,
Pennsylvania. 1976. ICited in (7)1
13. Mentser, M., and A.G. Sharkey, Jr. Chemical Charac-
terization of Diesel Exhaust Particulates. PERC/
RI-77/5. Pittsburgh Energy Research Center, Pitts-
burgh, Pennsylvania. 1977.
14. Laresgoiti, A., A.C. Loos, and G.S. Springer. Particu-
late and Smoke Emission from a Light-duty Diesel
Engine. Environmental Science Technology, ll:973-/«.
1977.
15 Size Distribution and Mass Output of Particulates from
' Diesel Engine Exhausts. RI 8141, U.S. Bureau of
Mines. 1976. {Cited in (7)1
16 Corn, M. Aerosols and the Primary Air Pollutants -
" Nonviable Particles, Their Occurrence, Properties, and
Effects. In: Air Pollution, Third Edition. Volume I.
Air Pollutants, Their Transportation and Transport.
A.C. Stern, ed., Academic Press, New York. 1976. pp
78-168.
17 Whitley, K.T., et al. Aerosol Size Distributions and
Concentrations Measured During the General Motors
Proving Grounds Sulfate Study. In: The General
Motors/EPA Sulfate Dispersion Experiment, Selected EPA
Research Papers. R.K. Stevens, et al., ed. U.S.
EPA-600/3-76-035. April 1976. pp 29-80.
18. Morrow, P.E. Models for the Study of Particle Reten-
tion and Elimination in the Lung. In: Inhalation
Carcinogenesis. AEC Symposium Series, No. 18. M.G.
3-39
-------
Hanna, P. Nettlesheim, and J.R. Gilbert, eds. U.S.
Atomic Energy Commission, Washington, D.C. 1970
pp. 103-119.
19. Airborne Contaminants. In: Environmental Factors in
Respiratory Disease. D.H.K. Lee, ed. Academic Press,
New York. 1972. pp. 71-90.
20. Gross, G.P. Automotive Emissions of Polynuclear
Aromatic Hydrocarbons. SAE No. 740464. Society of
Automotive Engineers, Inc., Warrendale, Pennsylvania.
1974. iCited in (l)j
21. Lyons, M.J. Comparison of Aromatic Polycyclic Hydro-
carbons from Gasoline Engine and Diesel Engine Ex-
hausts, General Atmospheric Dust, and Cigarette-Smoke
Condensate. National Cancer Institute Monograph, No
9. NCI. 1962. pp. 193-199.
22. Spindt, R.S. First Annual Report on Polynuclear
Aromatic Content of Heavy-duty Diesel Engine Exhaust
Gases. A report submitted to the Coordinating Research
Council by Gulf Research and Development Co. July
1974. ICited in (1)].
23. Jentoft, R.E., and T.H. Gouw. Analysis of Polynuclear
Aromatic Hydrocarbons in Automobile Exhaust by Super-
critical Fluid Chromatography. Anal. Chem., 48:2195-
2200, 1976. [Cited in (1)1
24. Newhall, H.K., et al. The Effect of Unleaded Fuel
Consumption on Polynuclear Aromatic Hydrocarbon Emis-
sions. SAE No. 730834. Society of Automotive Engi-
neers, Inc., Warrendale, Pennsylvania. 1973. [Cited
in (1)]
25. Begeman, C.R. Carcinogenic Aromatic Hydrocarbons in
Automobile Effluents. SAE No. 440C. Society of Auto-
motive Engineers, Inc., Warrendale, Pennsylvania.
1962. [Cited in (4)]
26. Springer, K.J. Investigation of Diesel Powered Vehicle
Emissions - Part VII. Unpublished report to Emission
Control Technology Division, U.S. Environmental Protec-
tion Agency, Contract No. 68-03-2116. August 1976.
[Cited in (4)]
3-40
-------
27. National Academy of Sciences. Participate Polycyclic
Organic Matter. NAS, Washington, D.C. 1972.
28 Boren, H.G. Carbon as a Carrier Mechanism for Irritant
Gases. Arch. Environ. Health, 8:119-24. 1964. {Cited
in (24)]
29. Shabad, L.M., L.N. Pylev, andT.S. Kolesnichenko.
Importance of the Deposition of Carcinogens for Cancer
Induction and Lung Tissue. J. Nat. Cancer Inst.,
33:135-141. 1964. [Cited in (24)1
30. Pylev, L.N. Effect of the Dispersion of Soot in
Deposition of 3,4-Benzpyrene in Lung Tissue of Rats.
Hyg. Sanit., 32:174-79. 1967. {Cited in C24)]
31. Pylev, L.N. Induction of Lung Cancer in Rats by
Intratracheal Insufflation of Carcinogenic Hydrocar-
bons. Acta Un. Int. Cancer, 19:688-91. 1962. {Cited
in (24)]
32. Saffiotti, U., F. Cefis, and L.H. Kolb. A Method for
the Experimental Induction of Bronchogenic Carcinoma.
Cancer Res., 28:104-124. 1968. {Cited in (24)]
33. Scientific and Technical Assessment Report on Particu-
late Polycyclic Organic Matter (PPOM). EPA-600/
6-75-001, U.S. Environmental Protection Agency, Wash-
ington, D.C. 1975.
34. Altshuller, A.P., and J.J. Bufalini. Photochemical
Aspects of Air Pollution: A Review. Environ. Sci.
Technology, 5:39-64. 1971. {Cited in (24)]
35. Ayres, S.M., and M.E. Buehler. The Effects of Urban
Air Pollution on Health. Clin. Pharmacol. Ther.,
11:337-71. 1970. {Cited in (24)]
36. Stokinger, H.E., andD.L. Coffin. Biologic Effects of
Air Pollutants. In: Air Pollution. Volume 1. Air
Pollution and Its Effects. A.C. Stern, ed. Academic
Press, Inc., New York. 1968. pp. 445-546. {Cited in
(24)]
37. Laskin, S. , M. Kerschner, and R.T. Drew. Studies in
Pulmonary Carcinogensis. In: Inhalation Carcino-
gensis. AEC Symposium Series, No. 18. M.G. Hanna, P.
Nettesheim, and J.R. Gilbert, eds. U.S. Atomic Energy
Commission, Washington, D.C. 1970. pp. 321-50.
3-41
-------
38. Blumer, M., W. Blumer, and T. Reich. Polycyclic
Aromatic Hydrocarbons in Soils of a Mountain Valley:
Correlation with Highway Traffic and Cancer Incidence
Environ. Science Technology, 11:1082-84. 1977.
39. Automobile Sulfuric Acid Emission Control - The De-
velopment Status as of December 1975, Report to the
U.S. Environmental Protection Agency. December 1975
[Cited in (1)]
40. Bradow, R., J. Huisingh, and V. Duffield. Facts on
Diesel Particulate Study. (Preliminary data). U.S.
Environmental Protection Agency, Research Triangle
Park, North Carolina. 1977.
3-42
-------
4.0 TEST CITY METHODOLOGY AND PROJECTIONS
This chapter discusses the methodology used to project
the air quality impact of increased diesel-powered motor
vehicle usage in a test city. The impact is first assessed
in terms of ambient air concentrations of TSP and BaP
through the use of appropriate and available diffusion
modeling tools, and then in terms of the number of persons
exposed to varying levels of TSP and BaP concentrations.
The assessment covers a base year (1974) and four projection
years (1981, 1983, 1985, and 1990).
The following paragraphs discuss the rationale used in
selecting a test city and a diffusion modeling approach.
Selecting the Test City
Ideally, selection of a test city for a study of this
type would be based on numerous criteria such as total
population, population density, age of the city, diversity
of industrialization, number of motor vehicles and roadway
miles per capita, and other relevant variables. All of
these would help to identify an average or typical large
(i.e., greater than 200,000 population) metropolitan area.
The selected city would then be modeled, and the resulting
4-1
-------
predicted-versus-measured air pollution concentrations would
be extrapolated to the national data levels of large urban
areas.
Because the time constraints imposed upon the study
precluded any possibility of using such a process, the
criterion for selection becomes simply: "What seemingly
typical large urban area has a usable diffusion model that
is current and quickly accessible to the consultant?"
The Kansas City metropolitan area meets this criterion
quite well. First, it is generally representative of most
large urban areas in the United States. It is situated on
the border of Missouri and Kansas and encompasses 8 counties
and 111 cities. The portion of the area on which this study
focuses covers 255 square miles and has a population of
approximately 760,000, which indicates an average density of
about 2990 people per square mile. Figure 4-1 shows a map
of the Kansas City area, highlighting the portion addressed
in this study.
Second, Kansas City could also provide a recent set of
usable diffusion modeling data. Data used as input in a
1976 modeling effort were readily accessible to the con-
sultant and in a format that could be quickly applied to the
work on this report.
4-2
-------
Figure 4-1. The test city study area - Kansas City, Missouri.
4-3
-------
Selecting the Diffusion Modeling Approach
The decision regarding the approach to use in modeling
the air quality impact of diesel vehicle emissions is
largely a function of the purposes of the overall study.
Reduced to its key ingredients, the purpose of this study is
to assess the impact of two pollutants (TSP and BaP) on
public health, based on two different assumptions concerning
the rate at which new diesel vehicles will be introduced
into the on-road vehicle population. (For TSP, this assess-
ment refers to the impact on annual exposure; for BaP, it
simply refers to the number of people exposed to various
ranges of concentrations.)
Criteria used to select an appropriate model for this
study include the following:
0 The model(s) must yield annual and 24-hour concen-
trations. Even shorter time period predictions
may be useful for BaP.
0 The model(s) must have been validated in a general
sense and, in a more specific sense, calibrated
against measured air quality data for the test
city.
Available population data must be comparable to
the area for which diffusion model results are
obtainable. Thus, if a micro-scale model is to be
used, micro-scale population exposure data should
be available.
0 The model must accept traffic data as an input
variable.
Four separate diffusion models were readily recognized
as candidates for consideration: the Air Quality Display
4-4
-------
Model (AQDM), the Climatological Display Model (CDM), APRAC-
IA, and HIWAY. The first two are regional scale TSP and SO2
models that predict annual concentrations at many different
receptors. The last two are carbon monoxide (CO) models
that predict hourly concentrations at many different recep-
tors. The APRAC-IA model yields regional "background" and
street canyon peak concentrations. The HIWAY model yields
peak concentrations for open terrain, corridors, or inter-
sections. Each CO model can be run so as to produce 8 to
24 hours worth of predicted values, but the cost of prepara-
tion and computation time is high.
Based on the criteria, none of the available models was
entirely suitable for the task at hand. The HIWAY model
will not predict annual concentrations, it has not been
validated for use in modeling particulate emissions, and it
does not generate data that are compatible with regionally
based population data. The first two objections also apply
to the APRAC 1A model. The HIWAY model might possibly be used
to generate maximum predicted 1-hour TSP and BaP concentra-
tions in the vicinity of a typical roadway; however, it
would probably take longer than allowed for the project if
done in conjunction with regional, longer term modeling.
The CDM was objected to for three different reasons:
1) it requires meteorological data that are not available in
4-5
-------
many areas (including Kansas City); 2) it can predict short-
term concentrations only with the aid of statistical assump-
tions, which are not necessarily valid; and 3) it is not
designed to predict the regional impact of motor vehicle
emissions alone.
Although the meteorological data required by AQDM are
normally available in most cities (including Kansas City),
the other two objections to the CDM apply to the AQDM as
well. In addition, the AQDM is known to have a tendency
to overpredict the impact of area sources (of which motor
vehicles are a subset).
After consideration was given to the drawbacks of all
the available options, the AQDM was judged to be the best
for this task. The capabilities and usage of this model are
discussed fully in the referenced document, Air Quality Dis-
play Model.
4.1 NONMOTOR VEHICLE EMISSIONS
Base Year Emissions
Because no BaP emissions data are available from point
and nonvehicular area sources in the Kansas City area,
no effort is made to predict overall concentrations of that
pollutant. Rather, efforts are directed toward defining the
impact of BaP emissions from diesel vehicles.
Particulate emissions data covering point and area
sources (summarized in Table 4-1) are from a recent report
4-6
-------
Table 4-1. PARTICULATE EMISSIONS IN TEST CITY (1974)
- " 1
Source category
Point sources
Power plants
Mineral products
Grain mills and elevators
Refineries
Chemical process
Metals
Automotive
Fiberglass
Miscellaneous
Subtotal
Stationary area sources
Natural gas combustion
LPG combustion
Distillate oil combustion
Residual oil combustion
Coal combustion
Wood combustion
Incinerators
Subtotal
Mobile sources
Highway vehicles (diesel exhaust)
Highway vehicles (gasoline exhaust)
Highway vehicles (tire wear)
Railroad
Aircraft
River vessels
Subtotal
Fugitive dust sources
Paved streets
Unpaved streets, parking lots
Cleared areas, storage areas
Construction, aggregate storage
Railroad yards
Agriculture
Subtotal
Total
1
Annual
emissions,
ton/yr
37,253
2,563
4,113
1,186
135
1,639
92
2,871
296
50,148
418
48
63
106
0
1
10
646
171
835
676
167
4
n. d.
1,853
10,377
110
r- f\
50
420
204
225
11,789
64,436
Percent
of total
57.8
4.0
6.4
10
. 8
00
. I
2.5
0*1
. 1
4r-
. D
0.5
77.8
0.6
OT
. 1
01
.1
0*-)
. 2
Of\
. 0
Or\
. o
1f\
. 0
1f\
.0
0.3
1.3
1.0
0.3
0.0
0.0
2.9
16.1
0.2
01
. i
0.7
0«-N
. 3
OO
. J
18.3
4-7
-------
entitled, Analysis of Probable Participate Nonattainment in
the Kansas City AQCR.4 These data, which are used to pre-
dict base year concentrations, have been modified in the
following ways. First, emissions from one of the power
plants have been increased substantially to reflect findings
concerning the effect of equipment malfunctions, shutdowns,
*
and inefficiencies; second, paved road emissions (reen-
trained dust) have been reduced by approximately one-third
to reflect the findings of a recent study;5 and third,
emissions from motor vehicles are separated into tire-
wear emissions and exhaust emissions. The latter category
has been further split into emissions from diesel-powered
and gasoline-powered vehicles. Total emissions from motor
vehicles also differ slightly from those in the reference
material as a result of the use of recently revised emission
factors.
These modified data are used to compare annual con-
centrations predicted by AQDM with measured values, and the
resulting relationship is used to adjust predicted values.
As shown in Table 4-2, these predicted values agree fairly
well with the measured TSP data from 18 monitoring sites:
an average predicted arithmetic mean of 73.1 yg/m3, compared
Communication with Mr. P. Stablein, Kansas City Health
Department, Air Quality Division. November 1977.
4-8
-------
Receptor
No.
I
vo
Average
Table 4-2. COMPARISON OF MEASURED VERSUS PREDICTED TSP
CONCENTRATIONS, KANSAS CITY (1974)
No. Brighton
Waterworks
No. Kansas City
KC Health Department
Morse School
NASN site
Leeds
6600 Independence
Fairfax
Deramus
Municipal Airport
Independence Courthouse
Claycomo
ASB Bridge
No. Liberty
UMKC Campus
Klamm Park
Turner H.S.
Deviation of
predicted from
measured values,
yg/m3
Measured
Predicted
arithmetic
mean,
yg/m3
arithmetic
mean
Receptor name
73.1
86.8
Ratio of measured to predicted values = 86.8/73.1 - 1.187
17.3
-------
with an average measured concentration of 86.8 yg/m3. The
mean deviation of the predicted from the measured concentra-
tions is 17.3 yg/m . it is considered neither useful nor
appropriate to use a regression equation to correct pre-
dicted values, however, because none of the predicted or
measured values is below 55 yg/m3. Because all data pairs
tend to cluster at the upper end of the concentration
range, predicted values are corrected by a ratio of average
measured to predicted concentrations, i.e., 1.187. A factor
of 0.933, obtained from Reference 4, is used to convert
arithmetic mean predictions to geometric means. This value
is based on a statistical analysis of arithmetic and geo-
metric means observed at the 18 Kansas City TSP monitoring
sites.
Projection Year Emissions
Numerous uncertainties are associated with projecting
future emissions. Perhaps the most basic relate to the
location and magnitude of new sources, how rapidly point-
source compliance schedules are met, and how extensively
nontraditional fugitive dust sources are controlled.
Because of these uncertainties, it is assumed that all
emission sources other than diesel vehicles will remain
constant in the test city through 1990, both as to location
and emission rate. This assumption probably causes future
4-10
-------
emissions to be overestimated, but, concurrently, it focuses
attention on the impact of diesel vehicles alone (should all
other factors remain constant).
4.2 MOTOR VEHICLE EXHAUST EMISSIONS
This section explains the derivation of 1974 motor
vehicle exhaust information presented in Table 4-1 and
provides additional information necessary to project emis-
sions in 1981, 1983, 1985, and 1990. The focus is on esti-
mating overall vehicle miles traveled (VMT) by grid within
the study area, allocating those VMT to six vehicle cate-
gories, developing weighted particulate and BaP emission
factors, and using these data to calculate emissions for the
base year and each of the projection years.
Estimating Vehicle Miles Traveled
Vehicle miles traveled are estimated for six different
vehicle-engine classes: gasoline-powered and diesel-powered
light-duty vehicles (LDVG and LDVD); gasoline-powered and
diesel-powered light-duty trucks (LDTG and LDTD); gasoline-
powered heavy-duty trucks (HDG); and diesel-powered heavy-
duty trucks (HDD). National percentage of VMT by vehicle
type, vehicle type distribution by age, and assumed rate of
diesel-powered vehicle introduction are used in conjunction
with Kansas City traffic distribution and growth rate data.
4-11
-------
Base Year VMT - According to data provided by the Mid-
American Regional Council (MARC), the agency responsible for
transportation planning in the Kansas City area, 1974 VMT
totalled 2851.5 x 106 in the study area during the course of
the work reported in reference 4. Traffic volume was
plotted by link segment on a map of the area and assigned to
the 2 km by 2 km grid network shown in Figure 4-1.
Projection Year VMT - Local data were used to project VMT to
1990. A growth of 36 percent is predicted for the Kansas
City metropolitan area from 1970 to 1990. 6 Assuming this
growth occurs linearly, the following rates are calculated
from 1974 figures.
Projection year Fraction of 1974
1981 i
1983 1.151
1985 i.ise
1990 1.270
No data are readily available on which to base growth
projections by geographical area; therefore, it is assumed
that growth will occur uniformly throughout the urban area.
This assumption probably results in an overestimation of
both VMT and emissions in the central city core and an
underestimation of VMT in the suburban ring.
Distributing VMT Among Vehicle-engine Classes
Once VMT totals have been generated for the base and
projection years, the next step is to distribute these
4-12
-------
totals among the six vehicle categories described above.
The distribution varies with the year for which calculations
are made because of the impact of increasing use of diesel
vehicles. The following paragraphs describe the techniques
used to distribute VMT for the base year and each of the
four projection years.
Base Year Distribution - Figure 4-2 summarizes the procedure
used to distribute VMT among the six vehicle-engine classes.
In essence, the base-year national urban VMT distribution is
calculated, and it is assumed that the resulting distribu-
tion applies uniformly throughout the Kansas City study
area. The fractions arrived at are then applied to the
grid VMT generated previously.
The following base-year national VMT data were ob-
tained from the Federal Highway Administration (FHWA).
Vehicle type National VMT x 10
All personal passenger 1,013,068
vehicles
Commerical buses 2,610
School and other non- 2,450
revenue buses
Single-unit trucks 211,460
Combination trucks 56,059
All motor vehicles 1,285,647
4-13
-------
Data necessary to calculate urban VMT for light-duty
vehicles are assumed to be equivalent to those reported for
personal passenger vehicles. In the case of light- and
heavy-duty trucks, however, some data manipulating is re-
quired.
The U.S. Environmental Protection Agency (EPA) recently
calculated that light-duty trucks contribute 60 percent of
total truck VMT. This percentage is applied to the truck
VMT data reported above.
To calculate HDG and HDD shares of single-unit and
combination truck VMT, commercial and school bus VMT must be
assigned to each of these two classes. Again, the method-
ology used by the EPA is applied. A synthesis of the
calculation procedures used yields the following equations:
HDG VMT = (single-unit trucks - LDT) 0.91 + Eq. (1)
school bus VMT + (combination
trucks) 0.16
HDD VMT = (single-unit trucks - LDT) 0.09 + Eq. (2)
commercial bus VMT + (combination
trucks) 0.84
These national VMT totals must then be converted to
national urban VMT values, which is accomplished by applying
urban factors developed by EPA:
*~
Memorandum from M. E. Williams, U.S. EPA, re Urban/Rural
Vehicle Miles Travelled Split by Mobile Source Cateqorv
dated December 4, 1975.
4-14
-------
Category Urban fraction
LDV 0.57
LDT 0.47
HDG 0.43
HDD 0.33
Urban VMT are then converted to fractions of total VMT
by dividing them by the total urban VMT. The following
fractions result:
, Fraction of total
Category VMT x 10 urban VMT (1974)
./ J.
LDV
LDT
HDG
HDD
577,449
75,440
24,847
17,914
0.830
0.108
0.036
0.026
In the absence of better data, it is assumed that diesel-
powered vehicles comprise 0.5 percent of the LDV and LDT VMT
in 1974.
At this point, these fractions are applied to the grid
VMT developed previously to produce a VMT total for each
vehicle-engine class in each study area grid.
Projection Year Distribution - Figure 4-2 summarizes the
procedure used to distribute VMT among the six vehicle-
engine classes for each projection year. This procedure is
complicated by the need to calculate fractions of assumed
VMT by model year for each of the six vehicle-engine classes
(reasons discussed under Assigning Emission Factors later in
this chapter). In the discussion of the procedure that fol-
lows, emphasis is given to the method of generating fractions
4-15
-------
OBTAIN REGIONAL
TRAFFIC GROWTH PRO-
JECTIONS. AND INTER-
POLATE FOR PROJEC-
TION YEARS
APPLY GROWTH RATES
TO BASE YEAR VMT
AND ASSUME GROWTH
IS UNIFORMLY DISTRIBUTED
OBTAIN NEW SALES
DISPOSAL INTRODUCTION
RATE DATA (BEST
ESTIMATE AND MAXIMUM)
FROM REF 11
(LOT RATES ASSUMED
TO BE IDENTICAL WITH
LDV RATES)
OBTAIN VEHICLE CATEGORY
FRACTIONS OF URBAN VMT
FROM BASE YEAR CALCULATIONS
ASSUME LDV/LOT/HEAVY-
DUTY TRUCK SPLIT REMAINS
CONSTANT OVER PROJECTION
PERIOD
OBTAIN FRACTION OF TOTAL
VEHICLES IN USE (LDV, LOT,
HOO. AND HOG) BY MODEL
YEAR FROM REF. 10 LOV AND
LOT THROUGH 1990, HDD AND
HD6 CHANGE)
MULTIPLY INTRODUCTION RATES
BY MODEL YEAR VMT FRACTIONS
AND PERFORM REMAINING
CALCULATIONS SPECIFIED IN
REF. 10 (LDV AND LOT CALCU-
LATIONS PERFORMED SEPARATELY)
CALCULATE VEHICLE CATEGORY
FRACTIONS BY MODEL YEAR FOR
EACH PROJECTION YEAR AND
INTRODUCTION ASSUMPTION
SUM FOR VEHICLE CATEGORY
FRACTION OF URBAN VMT
USE RESULTING FRACTIONS TO
SPLIT GRID VMT UNIFORMLY
FOR EACH PROJECTION YEAR i
CALCULATE FRACTIONS OF
TOTAL HOT'S IN USE NATION-
WIDE BY MODEL YEAR (TO
INCLUDE DIESEL FRACTIONS
THEREOF)
BACK-CALCULATE NO. OF
VEHICLES BY MODEL YEAR
FOR HDD'S AND HDG'S
OBTAIN NATIONAL URBAN HDD
AND HOG VMT FROM BASE
YEAR CALCULATIONS
CALCULATE TOTAL HOT MODEL
YEAR DISTRIBUTION AND
DIESEL FRACTIONS
DETERMINE NEW VEHICLE
DIESEL FRACTIONS FOR
EACH OF EIGHT TRUCK
CLASSES IDENTIFIED IN
REF. 13
DETERMINE NEW VEHICLE
SALES FOR EACH OF
EIGHT TRUCK CLASSES
SYNTHESIZE EIGHT CLASS
DATA AND CALCULATE HDD
INTRODUCTION RATES (LOW
SALES/LOW FRACTION, AND
HIGH SALES/HIGH FRACTION
INTERPOLATE INTRODUCTION
RATES FOR PROJECTION YEARS
Figure 4-2. Procedures for calculating projection years'
grid VMT by vehicle category for two diesel
introduction rate assumptions.
4-16
-------
of annual VMT by model year for HDD's and diesel introduc-
tion rates for each vehicle class.
The first step is to obtain vehicle-engine class frac-
tions of urban VMT from the base-year calculations discussed
above. Diesel- and gasoline-powered fractions for each
vehicle class are combined to obtain the following urban
fractions:
Vehicle class Fraction of Urban VMT (1974)
LDV 0.830
LOT 0.108
HDT 0.062
It is assumed that these fractions will remain constant
through 1990.
Next, data concerning the fractions of total LDV, LOT,
HDG, and HDD vehicles used nationwide (by model year) are
obtained from the EPA.8 Again it is assumed that these
fractions, with the exception of those for HDD and HDG, will
remain constant through 1990. Table 4-3 presents the base-
year fractions used in the study.
The third step is to convert these vehicle-in-use
factors to VMT factors by model year for each of the six
vehicle-engine classes. To do that, however, it is nec-
essary to account for the projected influx of diesel vehicles
This is a rather straightforward process for light-duty
vehicles and trucks, but is somewhat more complex for heavy-
duty trucks. Two stages are required: first, the calcula-
4-17
-------
Table 4-3. BASE-YEAR FRACTIONS OF TOTAL
VEHICLES IN USE NATIONWIDE
Age,
years
1
2
3
4
5
6
7
8
9
10
11
12
I13
Fraction of total
vehicles in use nationwide
LDV
0.081
0.110
0.107
0.106
0.102
0.096
0.088
0.077
0.064
0.049
0.033
0.023
0.064
LOT
0.061
0.095
0.094
0.103
0.083
0.076
0.076
0.063
0.054
0.043
0.036
0.024
0.185
HDD
0.077
0.135
0.134
0.131
0.099
0.090
0.082
0.062
0.045
0.033
0.025
0.015
0.064
HDG
0.037
0.070
0.078
0.086
0.075
0.075
0.075
0.068
0.059
0.053
0.044
0.032
0.247
Source: Reference 8.
4-18
-------
tion of base-year fractions of total HDT1s in use nationwide
by model year (to include diesel fractions thereof), and,
second, the development of a set of diesel introduction
rates.
Calculating Fractions of Annual HDT VMT by Model Year - To
perform this step, nationwide urban HDD and HDG VMT for
19747 and AP-42 Supplement 8 VMT fractions for diesel- and
gasoline-powered HDT's are used to back-calculate the number
of diesel- and gasoline-powered HDT's in use in urban areas
nationwide. These vehicle-in-use data are then combined,
and fractions of total HDT's by model year and diesel frac-
tions of each model year are generated. Table 4-4 presents
the vehicle-in-use fractions arrived at by this method.
Determining Diesel Vehicle Introduction Rates - Two dif-
ferent diesel introduction rates for each of the projection
years are specified: a "best estimate" and a "maximum"
case. Introduction rates for LDV's were obtained from the
*
EPA Emission Control Technology Division (ECTD) and are
presented in Table 4-5. After discussion with ECTD per-
sonnel, ^ it was decided that these introduction rates should
also be assumed to be representative of LDT's.
Basic data used to develop HDD introduction rates are
from a recent Michigan Technological University (MTU) report
Memorandum from J.P. DeKany, U.S. EPA, re Request for an
Air Quality Assessment of Particulate Emissions from Diesel-
powered VEhicles, dated September 19, 1977.
f Communication with J. Somers, U.S. EPA, November 1977.
4-19
-------
Table 4-4. BASE-YEAR FRACTIONS OF TOTAL HEAVY-DUTY TRUCKS
IN USE NATIONWIDE (AND DIESEL FRACTIONS THEREOF)
Age,
years
1
2
3
4
5
6
7
8
9
10
11
12
I13
Fraction of total
HOT' s in use
nationwide (1974)
(urban only)
0.042
0.078
0.085
0.092
0.078
0.077
0.076
0.067
0.057
0.051
0.041
0.030
0.224
Fraction of
model year (1974)
HDT's that are
diesel- powered
0.233
0.221
0.201
0.182
0.162
0.148
0.137
0.119
0.100
0.085
0.076
0.068
0.037
4-20
-------
Table 4-5. CLASSIFICATION OF TRUCKS BY GVW
Truck
class
2A
2B
3
4
5
6
7
8
GVW range, lb
i '
< 6,000
6,000-8,500
8,500-10,000
10,001-14,000
14,001-16,00
16,001-19,500
19,501-26,000
26,001-33,000
> 33,000
MTU Projection Comments
^^^^^^^^^^^,^^1 ' ^i^-"^"^
"Low fractions are expected to
occur with a high degree of
probability..
High fractions...are not very
probable"
"Low estimates of sales represent
a slowly expanding economy...
Moderate sales volume...expresses
a steady or healthy growth
High truck sales projections are
...probable, if the economy grows
exceptionally well and at the
same time technical breakthroughs
occur"
Source: Reference 9.
4-21
-------
entitled The Development of an Emission and Fuel Economy
Computer Model for Heavy-duty Trucks and Buses.9 This model
classifies truck population by gross vehicle weight (GVW)
ranges and other relevant criteria.
The referenced report notes that past trends are no
longer satisfactory for projecting diesel truck sales.
General economic conditions, energy supply/demand constraints,
and government fuel-economy, pollution, and safety regula-
tions have placed the diesel engine "...in a position to
dominate the future truck market because of its cost ef-
fectiveness in terms of power applications and utilization."9
Recognizing a great margin of uncertainty associated
with making such projections, the authors of the MTU report
generated three sets of new truck sales and three sets of
diesel penetration fractions for each of the eight GVW
classes listed in Table 4-5. Because these sales and
penetration fractions were presented graphically, they are
difficult to interpolate. They were, however, used to
generate the sales and penetration fractions shown in Table
4-6. The penetration fractions are weighted by the sales
data so as to produce cumulative penetration fractions for
each projection year. For the purposes of this report, a
"best estimate" introduction rate is defined to be low
penetration fractions weighted by low sales estimates. A
"high" rate is defined to be high penetration fractors
weighted by high sales estimates.
4-22
-------
Table 4-6. TRUCK SALES AND DIESEL PENETRATION FRACTIONS
=====
Truck
class
2B
3
4
5
6
7
8
Total
Vehicle Sales (x 106) and Diesel Fractions*
Projection
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
1981
0.14 (0.03)
0.16 (0.14)
0.00 (0.03)
0.04 (0.14)
0.00 (0.03)
0.04 (0.14)
0.00 (0.11)
0.03 (0.20)
0.17 (0.11)
0.28 (0.20)
0.02 (0.46)
0.04 (0.70)
0.10 (0.95)
0.19 (1.00)
0.43 (0.33)
0.78 (0.40)
1983
0.14 (0.05)
0.16 (0.32)
0.01 (0.05)
0.05 (0.32)
0.00 (0.05)
0.05 (0.32)
0.00 (0.16)
0.03 (0.47)
0.18 (0.16)
0.29 (0.47)
0.02 (0.50)
0.04 (0.80)
0.10 (0.96)
0.20 (1.00)
0.45 (1.31)
0.82 (0.57)
1985
0.14 (0.06)
0.17 (0.57)
0.01 (0.06)
0.05 (0.57)
0.00 (0.06)
0.05 (0.57)
0.00 (0.20)
0.04 (0.79)
0.19 (0.20)
0.31 (0.79)
0.02 (0.53)
0.04 (0.89)
0.10 (0.97)
0.20 (1.00)
0.46 (0.33)
0.86 (0.78)
1990
0.15 (0.11)
0.18 (0.97)
0.01 (0.11)
0.06 (0.97)
0.00 (0.11)
0.07 (0.97)
0.00 (0.29)
0.05 (1.00)
0.21 (0.29)
0.35 (1.00)
0.02 (0.58)
0.04 (1.00)
0.11 (0.98)
0.23 (1.00)
0.50 (0.64)
0.98 (0.99)
a Figures in parentheses represent fractions of truck class sales
which are projected to be diesel-powered.
Source: Reference 9.
4-23
-------
Diesel introduction rates for the years in between the
four projection years are determined through linear inter-
polation.
Table 4-7 summarizes the diesel vehicle introduction
rates provided by ECTD and synthesized from the MTU report.
The diesel introduction rates by model year shown in this
table are used to modify the fractions of total vehicles in
use shown in Tables 4-3 and 4-4. This is accomplished by
multiplying the diesel fraction for a given model year times
the fraction of vehicles in use that model year. These
fractions are then used to perform the VMT fraction calcula-
tions described in AP-42, Supplement 8.8 Tables A-l through
A-3 (in Appendix A) present details of the resulting VMT
fractions for each of the projection years; Table 4-8 Cin
this section) summarizes this information.
At this point it is necessary to apply these fractions
to the grid VMT developed previously. The result is a VMT
total for each vehicle-engine class in each study area grid
for two diesel-introduction-rate assumptions affecting each
of the four projection years.
Assigning Emission Factors
Emission factors are available for two pollutants:
particulate matter and BaP. For other pollutants (e.g., CO
or HC), emission correction factors are available for such
4-24
-------
Table 4-7. DIESEL VEHICLE INTRODUCTION RATES
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
Share of new sales by model year
Light-duty vehicles
and trucks
(Perceni
Best
estimate
0.5
0.5
0.5
0.5
2.0
4.0
6.0
8.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
Maximum
0.5
0.5
0.5
0.5
5.0
10.0
15.0
20.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
Heavy-duty trucks
;ages)
Best
estimate
28.0
28.0
28.0
30.0
31.0
31.0
31.0
31.0
31.0
31.0
33.0
39.0
45.0
52.0
58.0
64.0
Maximum
28.0
28.0
28.0
35.0
36.0
38.0
40.0
48.0
57.0
67.0
78.0
82.0
86.0
90.0
94.0
99.0
4-25
-------
Table4-8. FRACTION OF URBAN VMT BY MOBILE SOURCE CATEGORY IN PROJECTION YEARS
. -- - - -r
urce category
asoline vehicles
asoline trucks
asoline trucks
iesel vehicles
iesel trucks
iesel trucks
=====
1974
0.826
0.107
0.036
0.004
0.001
0.026
Fraction of urban VMT
1981
Best
est.
0.815
0.106
0.035
0.015
0.002
0.027
Max.
0.796
0.104
0.033
0.034
0.004
0.029
1983
Best
est.
0.798
0.104
0.034
0.032
0.004
0.028
Max.
0.750
0.098
0.029
0.080
0.010
0.033
1985
Best
est.
0.779
0.102
0.034
0.051
0.006
0.028
Max.
0.704
0.093
0.023
0.126
0.015
0.039
1990
Best
est.
0.754
0.098
0.025
0.076
0.010
0.037
Max.
0.639
0.084
0.010
0.191
0.024
0.052
-------
variables as average speed, ambient temperature, percent
cold starts, truck weight, weight/power rates, and age
deterioration. Such is not the case for particulate or BaP
emissions. Further, the base emission factors for these two
pollutants are based on fewer samples, hence can be expected
to have much larger confidence intervals.
Recognizing the need for better emission factors,
EPA generated the exhaust emission factors given in Table
4-9.*t§ Except in the case of gasoline-powered light-
duty vehicles and trucks, it is assumed that the emission
factors will not vary from those given in Table 4-9. In
the case of gasoline-powered light-duty vehicles and trucks,
however, the effect of phasing out leaded fuel and phasing
in innovative technology needs to be taken into account.
Two assumptions are made in calculating weighted emis-
sion factors for gasoline-powered light-duty vehicles and
trucks in the four projection years:
0 that all vehicles prior to the 1975 model year
were noncatalyst and used leaded fuel, and
0 that from 1975 on 70 percent of each new model
year fleet will use catalysts and 30 percent will
use catalysts with excess air.9
Memorandum from J.P. DeKany, U.S. EPA, re Request for an
Air Quality Assessment of Particulate Emissions from Diesel-
powered Vehicles, dated September 19, 1977.
* Communication with J. Somers, U.S. EPA, November 1977.
§ Memorandum from J.P. DeKany, U.S. EPA, re Transmittal of
Particulate Emission Factors for Heavy-duty Diesels, dated
November 8, 1977.
4-27
-------
Table 4-9. EXHAUST EMISSION FACTORS
Vehicle category
Light-duty gasoline-powered
vehicles and trucks
Catalyst
Catalyst (excess air)
Noncatalyst (leaded fuel)
Noncatalyst (unleaded fuel)
Heavy-duty gasoline-powered
trucks
Catalyst
Catalyst (excess air)
Noncatalyst (leaded fuel)
Noncatalyst (unleaded fuel)
Light-duty diesel-powered
vehicles and trucks
Heavy-duty diesel-powered
trucks
Emission factors
Particulates,
g/VMT
0.006
0.015
0.25
0.002
0.02
0.05
0.90
0.007
0.5
2.0
BaP,
M g/VMT
0.1
0.1
1.0
1.0
0.3
0.3
3.0
3.0
1.0*
6.0b
4.6a
24. 6b
Low estimate.
High estimate.
4-28
-------
The latter data are based on the further assumption
that the technology used by Chrysler, General Motors, and
Ford will be representative of the entire new fleet; that
the current technological split between catalysts and
catalysts with excess air will remain constant through 1990;
9
and that the current technological split is as follows:
Share of new car fleet, percent
Manufacturer Catalyst Catalyst (excess air)
Chrysler 90 10
General Motors 100
Ford 100
When combined with 1974 new car sales data, these tech-
nological splits yield the 70/30 ratio described above.
Weighted emission factors for a given year and diesel
introduction rate assumption are calculated by using the age
distribution data generated during the process of calculat-
ing VMT factors by vehicle-engine class. (Appendix A
presents VMT factors.) Table 4-10 presents the resulting
weighted emission factors for gasoline-powered light-duty
vehicles and trucks.
Calculating Projection Year Exhaust Emissions
The emission factors just discussed are combined with
the vehicle-engine class VMT by grid data to yield exhaust
emissions by vehicle-engine class for each grid and each
projection year. Tables 4-11 and 4-12 summarize the par-
ticulate and BaP emission totals for each projection year.
4-29
-------
Table 4-10. WEIGHTED EMISSION FACTORS FOR
GASOLINE-POWERED VEHICLES
1974
1981
Best est.
Max.
1983
Best est.
Max.
1985
Best est.
Max.
1990
Best est.
Max.
Particulates , g/VMT
Lt-duty
vehicle
0.25
0.061
0.061
0.035
0.035
0.022
0.022
0.017
0.017
Lt-duty
trucks
0.25
0.070
0.072
0.047
0.049
0.033
0.036
0.026
0.029
BaP, y g/VMT
Lt-duty
vehicle
1.0
0.267
0.271
0.178
0.183
0.136
0.141
0.119
0.123
Lt-duty
trucks
1.0
0.299
0.304
0.219
0.227
0.174
0.182
0.149
0.158
4-30
-------
Table 4-11. PROJECTED MOTOR VEHICLE EXHAUST EMISSIONS (PARTICULATES)
(ton/yr)
Vehicle category
Gasoline-powered
Lt-duty vehicles
Lt-duty trucks
Hv-duty trucks
Subtotal
Diesel-powered
Lt-duty vehicles
Lt-duty trucks
Hv-duty trucks
Subtotal
Total
1974
649
84
102
835
6
2
163
171
1006
1981
Best
est.
174
26
111
311
26
3
190
219
530
,
Max.
171
26
104
301
60
7
205
272
573
1983
Best
est.
101
18
111
230
58
7
206
271
501
Max.
76
18
94
187
145
18
239
402
589
1985
Best
est.
64
12
111
187
95
11
209
315
502
Max.
58
12
76
146
230
28
291
549
695
1990
Best
est.
51
10
90
151
171
19
295
485
636
Max.
43
10
36
89
381
48
415
844
933
"Best est." and "Max." refer to diesel introduction rate assumptions.
-------
Table 4-12. PROJECTED MOTOR VEHICLE EXHAUST EMISSIONS (BaP)
(ton/yr x 10 3)b
Vehicle category
Gasoline-powered
Lt-duty vehicles
Lt-duty trucks
Hv-duty trucks
Subtotal
Diesel-powered
Lt-duty vehicles
Lt-duty trucks
Hv-duty trucks
Subtotal
Total
a
1974
2.596
0.336
0.340
3.272
0.013
(0.075)
0.003
(0.019)
0.376
(2.009)
0.392
(2.103)
3.664
(5,375)
1981
Best
est.
0.763
0.111
0.369
1.243
0.053
(0.316)
0.008
(0.042)
0.426
(2.332)
0.497
(2.690)
1.740
(3.933)
Max.
0.758
0.111
0.348
1.217
0.119
(0.716)
0.014
(0.084)
0.469
(2.506)
0.602
(3.306)
1.819
(4.523)
1983
Best
est.
0.513
0.082
0.369
0.964
0.116
(0.694)
0.014
(0.087)
0.466
(2.492)
0.596
(3.273)
1.560
(4.237)
Max.
0.497
0.080
0.315
0.892
0.289
(1.141)
0.036
(0.217)
0.549
(2.937)
0.874
(3.272)
1.766
5.783)
1985
Best
est.
0.395
0.066
0.380
0.841
0.190
(1.141)
0.022
(0.134)
0.480
(2.568)
0.692
(3.843)
1.533
(4.684)
Max.
0.369
0.063
0.257
0.689
0.470
(2.818)
0.056
(0.335)
0.669
(3.577)
1.195
(6.730)
1.884
(7.419)
1990
Best
est.
«,
0.358
0.058
0.299
0.715
0.303
(1.820)
0.040
(0.239)
0.680
(3.634)
1.023
(5.693)
1.738
(6.408)
Max.
-
0.315
0.053
0.120
0.488
0.762
(4.575)
0.096
(0.575)
0.955
(5.106)
1.813
10.256)
2.301
10.744)
^ "Best est." and "Max." refer to diesel introduction rate assumptions.
Figures in parentheses represent values obtained with high BaP emission factor.
-------
4.3 PROJECTED IMPACT OF DIESEL EMISSIONS ON AIR QUALITY
The impact of diesel emissions on ambient TSP concen-
trations is predicted by using the calibrated AQDM discussed
earlier. This is done by running the AQDM for each projec-
tion year, using only diesel emissions as input. The impact
of diesels on ambient BaP concentrations is determined
similarly, but a ratio of BaP to particulate emissions is
also applied to the calculations for each projection year.
Table 4-13 presents the ratios used for each projection case.
Total Suspended Particulate Concentrations
The maximum annual TSP concentration from diesel-
powered vehicles in 1974 (0.35 yg/m3) occurred in the
immediate downtown area. This rather low maximum concen-
tration suggests that diesels contributed a relatively small
amount of particulate matter that year.
The emission data presented earlier, however, indicate
that TSP concentrations attributable to diesels will in-
crease through 1990. Table 4-14 summarizes the predicted
changes in regional concentrations over the projection
period. The regional TSP concentrations resulting from
diesel usage are projected to increase steadily to a high in
*%
1990 of either 0.96 or 1.73 yg/m (annual geometric mean),
depending upon the assumed rate of diesel introduction. The
higher value constitutes 2.3 percent of the primary national
ambient air quality standard (NAAQS) for TSP.
4-33
-------
Table 4-13. RATIO OF BaP TO PARTICULATE EMISSIONS
(Diesels only)
Year
Low BaP
High BaP
1974
1981
1983
1985
1990
Best estimate
Max. diesel
Best estimate
Max. diesel
Best estimate
Max. diesel
Best estimate
Max. diesel
2.2864 x 10~6
2.2592 x 10~6
2.2261 x 10~6
2.2271 x 10~*
2.1783 x 10"6
2.1989 x 10
2.1575 x 10
2.1895 x 10
2.1474 x 10
12.2866 x 10~6
12.2618 x 10~6
12.2259 x 10~6
12.2269 x 10~6
12.1778 x 10~6
12.1984 x 10~6
12.1569 x 10~6
12.1879 x 10~f
12.1465 x 10~6
4-34
-------
Table 4-14. PROJECTED REGIONAL ANNUAL AVERAGE
CONCENTRATIONS OF TSP FROM DIESEL EXHAUST FOR
TEST CITY.
Year
1974
1981
1983
1985
1990
Diesel Growth Case
Best
Est.
yg
0.35
0.45
0.57
0.65
0.96
Max.
.j Growth
/m3
0.56
0.83
1.13
1.73
4-35
-------
The method for predicting BaP concentrations attribut-
able to diesels differs from that described for TSP only in
that BaP concentrations are indirectly modeled by AQDM. It
is assumed that a ratio of BaP to particulate emissions can
be applied to predicted TSP concentrations to yield pre-
dicted BaP concentrations for each grid in a given year.
The ratio changes from year to year because of variations in
emission factors and vehicle category mix.
The ratios presented in Table 4-15 are calculated by
defining the VMT/yr (see Table 4-8), the particulate emis-
sion factor, and the BaP emission factor for each diesel
vehicle category. Two BaP emission factors are used for
each diesel type to correspond with the factors given in
Table 4-10. The data are combined and totaled to yield
total particulate and BaP emissions from diesels, and the
ratios between the two totals are developed. The process is
repeated for each projection year.
4.4 ASSESSING POPULATION EXPOSURE
The difficulties involved in assessing exposure of a
given population to varying concentrations of a pollutant
are well documented in the air pollution control literature.11
Two examples are cited below:
4-36
-------
Table 4-15. PROJECTED REGIONAL ANNUAL AVERAGE
CONCENTRATIONS OF BaP FROM DIESEL EXHAUST FOR TEST CITY
Projection
case
Year
1974
1981
1981
1983
1983
1985
1985
1990
1990
Diesel
growth
case
Best est.
Max . growth
Best est.
Max . growth
Best est.
Max . growth
Best est.
Max . growth
Regional annual geometric mean
BaP cone, ng/m^ x 10 ~ 3
Emission factor case
Low
0.8
1.0
1.2
1.3
1.8
1.4
2.9
2.1
3.7
High
4.3
5.5
6.8
6.9
10.1
7.9
13.8
11.7
21.1
4-37
-------
0 The representativeness of measured air quality
data is uncertain. Data obtained in monitoring a
pollutant like carbon monoxide may be representa-
tive of little more than exposure at the exact
location of the monitor. Conversely, data on TSP
obtained with a hi-vol sampler in a rural area may
well be representative of hundreds of square
kilometers.
0 People are mobile rather than stationary, tending
to live in one place, work in another, arid travel
often among various points. Thus, the exposure of
people to a given pollutant may differ signifi-
cantly from the concentrations measured at any one
site.
Both of these limitations are especially serious when
averaging periods are short-term (1, 3, or 8 hours). When
averaging periods are longer, both concentrations and
exposures tend to homogenize. In assessments of particu-
lates and BaP, the focus is primarily on annual concentra-
tions or those representing longer-term averaging periods.
Thus, in this report it is assumed that the place of resi-
dence (as defined by the U.S. Bureau of Census and similar
data) is generally representative of exposure to annual
concentrations. Residences located near roadways are an
exception. It cannot be assumed that measured air quality
data are representative of air quality at residences near
heavily travelled roadways. On the contrary, recent empir-
ical data clearly indicate that TSP concentrations increase
as the slant distance from roadways decreases, and that TSP
concentrations increase with increasing traffic volume.^
4-38
-------
The exposure of persons living in such locations cannot
be assessed in the test city, however, because no data are
available on the number of people living at x distance from
roadways of y traffic.
Exposure of the test city population to varying levels
of TSP and BaP is assessed by assigning locally generated
data on 1970 origin-destination (OD) zone population to a
grid system established about the AQDM receptor network.
Thus 165 grids, 2 by 2 km, are centered upon the 165 AQDM
receptors. The OD zone populations are assigned to the
grids on the basis of land area by assuming that populations
are uniformly distributed within each OD zone.
It is assumed that the total population and its dis-
tribution in the test city will not change from 1970 to
1990. Such an assumption is necessary to minimize the
computation required to generate projections. This assump-
tion should not introduce significant error, because the
rate and distribution of population growth are expected to
vary greatly from city to city.
The number of persons exposed to varying concentrations
is predicted by combining these population data with the
concentrations predicted by AQDM for the projection years.
The AQDM program produces the TSP concentration at a
series of grid points throughout the urbanized population
4-39
-------
area, but it does not indicate the extremes in concentration
levels to which the population may be exposed. Further, no
population exposure data are available that account for
distances from roadways. This analysis attempts to predict
a range of population exposure; therefore the procedure
incorporates a distribution that estimates the upper expo-
sure. The dosage spectrum distribution developed by Horie
and Stern for data from the New York-New Jersey-Connect-
icut Tri-State Region is used for this purpose. Dosage
extremes are estimated from a mean value for a given area,
assuming a population evenly distributed over the area.
This dosage relationship is represented mathematically:
S(D) = /r N(r,D)dr/Ao
where S(D) is the dosage spectrum
r is the receptor site
Ao is the area under consideration
N(r,D) is a threshold function such that
N(r,D) = 1 if D(r) >_ 5
N(r,D) = 0 otherwise ,.
D is the dosage threshold
Simply stated, this equation presents the fraction of a
total area S(D) that is polluted more than D. Figure 4-3
shows this relationship for data from the Tri-State Region.
These data allow an extrapolation of the TSP concentration
at highest exposure level.
A dosage spectrum is generated from each grid point con-
centration. The dosage spectra-population data over the
4-40
-------
0.001
0 30 40 50 60 70 80 90 100 110 120 130
Figure 4-3. Dosage spectrum distribution in the
Tri-State Region (19).
4-41
-------
entire gridded area are then sxonmed. The Horie and Stern11
relationship is based on data from the Tri-State Region.
These data do not represent either national or Kansas City
data; however this is the only one data base readily avail-
able.
4-42
-------
REFERENCES FOR SECTION 4
1 Berry B.J.L., et al. Land Use, Urban Form, and
' Environmental Quality. University of Chicago, Chicago,
Illinois. Prepared for U.S. Environmental Protection
Agency, Office of Research and Development. Research
Paper Number 155. 1974.
2 Busse, A.D., and J.R. Zimmerman. User's Guide for the
Climatological Dispersion Model. U.S. Environmental
Protection Agency, Research Triangle Park, North Caro-
lina. Publication No. EPA-R4-73-024. December 1973.
3 Air Quality Display Model. TRW Systems Group. Pre-
pared for National Air Pollution Control Administra-
tion, Washington, D.C. November 1969.
4 PEDCo Environmental, Inc., Cincinnati, Ohio. Analysis
of Probable Particulate Non-Attainment in the Kansas
City AQCR. Prepared for U.S. Environmental Protection
Agency, Kansas City, Missouri. February 1976.
5 Control of Reentrained Dust from Paved Streets. U.S.
Environmental Protection Agency, Kansas City, Missouri.
Publication No. 907/9-77-007. August 1977.
6 PEDCo Environmental, Inc., Cincinnati, Ohio. Trans-
portation Controls for the Kansas City Air Quality
Control Region. Prepared for U.S. Environmental Pro-
tection Agency, Research Triangle Park, North Carolina.
May 1973.
7 Estimated Motor Vehicle Travel in the United States and
Related Data, 1975 and Revised 1974. Table VM-1.
U S. Department of Transportation, Federal Highway
Administration, Highway Statistics Division, Office of
Highway Planning. January 1977.
8. Mobile Source Emission Factors. Interim Document.
U.S. Environmental Protection Agency, Washington, D.C.
June 1977.
4-43
-------
9. Jambekan, A.B. and J.H. Johnson. Development of an
Emission and Fuel Economy Computer Model for Heavy-duty
Trucks and Buses. Michigan Technological University,
Houghton, Michigan. Prepared for U.S. Environmental
Protection Agency, Ann Arbor, Michigan. August 1977.
10. Motor Vehicle Facts and Figures, 1976. Motor Vehicle
Manufacturers Association of the United States, Inc.,
New York, New York. 1977.
11. Horie, Y., and A.C. Stern. Analysis of Population Ex-
posure to Air Pollution in New York-New Jersey-Con-
necticut Tri-State Region. U.S. Environmental Pro-
tection Agency, Research Triangle Park, North Carolina.
March 1976.
12. Record, F.A. Evaluation of the Suspended Particulate
Problem. GCA Corporation, Bedford, Massachusetts,
Prepared for U.S. Environmental Protection Agency.
1976.
13. Estimates and Projections: Kansas City Metropolitan
Region. Mid-American Regional Council, Kansas City,
Missouri. September 1974.
14. R.I. Larsen. A New Mathematical Model of Air Pollution
Concentration Averaging Time and Frequency. Journal of
the Air Pollution Control Association. 19:24-30.
January 1969.
4-44
-------
5.0 ESTIMATES OF POPULATION EXPOSURES TO TSP AND BaP
Estimates of exposures of the national population to
TSP and BaP from diesel-powered vehicles are based on the
Kansas City data and on national trends. Estimates of TSP
and BaP concentrations expected to occur near roadways are
also presented on the basis of regional maximum short-term
(1-hour and 24-hour) and annual measurements. The cor-
relation methods and results of the analyses are discussed
in this section.
5.1 NATIONAL POPULATION IMPACT
5.1.1 TSP Dosage Analysis
Because the major constraint built into this analysis
is that a population-dose relationship is available only for
the test city, a method was sought to exprapolate the test
city dose data to all SMSA1s.
Several approaches were evaluated for characterizing
the national population with regard to total TSP exposure.
This list is by no means exhaustive, and because of time
limitations imposed on development of this report, considera-
tion was given only to those variables that appeared most
likely to correlate with the TSP level and for which a
comprehensive empirical data base was readily available.
5-1
-------
i) Annual geometric mean concentrations of TSP were
analyzed at ambient monitoring sites in 15 cities of various
sizes. Information from the National Air Data Bank (NADB)
was correlated with data on SMSA population, urbanized SMSA
population density, and percentage of urban blue-collar
workers. The percentage of blue-collar workers was included
in an effort to characterize the degree of industrialization
in an urban area that could have an impact on the TSP level.
Values for these three parameters are readily available from
census compilations, and information at the census tract
level should best characterize the individual monitoring
station location. Correlations of TSP with these three
parameters were very poor, however, probably because the
monitoring sites often do not reflect the average TSP level
in a census tract and the three parameters do not indicate
the types of particulate sources in an SMSA. Thus, this
approach was rejected.
ii) An effort was made to analyze dosage in terms of
the annual TSP loading in selected urbanized counties. Data
from the National Emissions Data System (NEDS) are readily
available for point sources in all U.S. counties, but data
on area sources are not systematically compiled or updated.
Because NEDS considers roadways as area sources, this
approach was rejected.
5-2
-------
iii) Finally, the annualized geometric mean TSP levels
for all monitoring stations reporting to NADB were averaged,
and the average level was correlated with SMSA population
and urbanized SMSA population density. The percent of urban
blue-collar workers did not correlate with TSP and was
excluded from consideration. The TSP data for selected
cities were the most recent year's values reported to the
NADB. This approach was selected as the means of charac-
terizing the exposure of the total U.S. population to TSP.
Averaging all monitoring station TSP levels in an SMSA
smoothed out the data and reduced the impact of local con-
ditions on single station analysis attempted in i) above.
The annual TSP levels from monitoring stations in 66
SMSA's were used to develop the correlations. These data
are presented in Appendix B. The relationship based on
correlation of TSP with SMSA population and urban population
density is shown in Figure 5-1.
In development of the exposure relationships, the U.S.
population is summed by SMSA for the 4 target years, within
ranges of population and population density. The summed
data in 15 cells are shown in Table 5-1. Since the regres-
sion line correlations in Figure 5-1 are not good, these
data are presented in a two-dimensional array with each cell
given a TSP mean and standard deviation to characterize the
5-3
-------
fr-S
ANNUAL GEOMETRIC MEAN TSP LEVEL, ug/m
in
fD
CD
O
rt
fD
cn
3
cn
w
fD
cn
d
in
cn
3
cn
O
d
0)
rt
H-
O
D)
QJ
H-
O
fD
cn
H-
rt
d
i-<
fD
§
§
o
O
O
M
fD
rt
M-
O
O
H)
Qi
fD
fD
cn
2
3
d
0)
§
3
fD
rt
H-
O
3
(D
0)
3
cn
13
fD
fD
M
cn
8
r
u>
co
rv>
A
^
cn
i
o
CO
J>
<
i
o
X
o
-------
Table 5-1.
DISTRIBUTION OF U.S. POPULATION BY SMSA POPULATION RANGE AND POPULATION
DENSITY FOR PROJECTION YEARS - NATIONAL SMSA POPULATIONS
Millions of people in each SMSA population group
I
Ul
Urbanized
population
density
10-* people/mi^
<2
2-2.99
3-3.99
4+
Cell
no.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Metropolitan
population
10° people
<0.5
0. 5-1.0
1.0-1.5
<0.5
0.5-1.0
1.0-1.5
>1.5
<0.5
0.5-1.0
1.0-1.5
>1.5
<0.5
0.5-1.0
1.0-1.5
>1.5
, 1981 .
ua
4.984
2.651
0.687
11.267
9.153
4.720
5.049
5.916
8.615
4.498
11.338
1.756
0.412
2.819
55.826
Tu
7.074
2.927
1.342
15.847
11.286
5.530
5.706
9.211
10.742
5.126
13.309
2.429
0.697
3.509
60.667
1983
U
5.079
2.741
0.702
11.505
9.390
3.520
6.564
6.017
7.902
5.566
11.575
1.852
0.418
2.867
56.895
T
7.214
3.033
1.372
16.183
11.574
4.167
7.430
9.369
9.968
6.323
13.590
2.540
0.707
3.572
61.846
1985
U
5.179
2.828
0.718
11.321
9.007
4.643
6.789
6.136
7.256
3.779
14.608
1.835
0.424
2.913
58.199
T
7.356
3.129
1.402
16.004
11.322
5.317
7.686
9.537
9.174
4.471
16.942
2.532
0.717
3.63}3
63-223
1990
0
5.452
3.061
0.758
11.269
9.393
3.029
10.219
5.804
6.676
5.637
15.548
1.942
0.440
3.036
61.716
T
7.743
3.388
1.481
15.844
11.990
3.618
11.540
8.944
8.669
6.818
18.019
2.670
0.744
3.795
66.565
U - urbanized part of SMSA population.
T - total urbanized population.
-------
data scatter. Table 5-2 presents the annual geometric mean
TSP and intercity standard deviations corresponding to each
combination of SMSA population and population density from
the data in Figure 5-1. where cells have few representative
test cities, values are extrapolated from Figure 5-1 and
from values for adjoining cells in Table 5-2.
The TSP-population exposure distribution described in
Section 4.4 is applied to each of the 15 population cells.
Further, each cell population is divided into five equal
fractions using the cell mean and standard deviation:
Population cell
fraction
1
2
3
4
5
Fraction of cell
population
0.20
0.20
0.20
0.20
0.20
Mean of
subfraction
X - 1.28 S.D.
X - 0.538 S.D.
X
X + 0.53 S.D.
X + 1.28 S.D.
The population distribution of each population subcell is
estimated by multiplying the Kansas City population-exposure
distribution times the subcell mean divided by the Kansas
City mean. The Kansas City mean is the mean TSP level for
all the monitoring stations, which in 1976 was 63.6 vg/m3.
A total population exposure range is determined by applying
the appropriate population data from Table 5-1 to the
individual subcell exposure distributions and suTiming all
the subcell TSP versus population relationships.
5-6
-------
Table 5-2. SUMMARY OF TSP DATA FROM
TEST CITY MONITORING STATIONS
Population
x 106
<0.5
0. 5-1.0
1.0-1.5
>1.5
Mean
S.D.b
NC
Mean
S.D.
N
Mean
S.D.
N
Mean
S.D.
N
Population density - urbanized,
10 3 people/mi'*
<2
57.7
3.0
4
62.0
2.5
1
66.5
3.6
0
d
2-<3
58.7
3.1
17
62.5
2.6
6
66.5
3.6
4
69.0
3.6
2
3-<4
59.5
2.7
7
63.0
4.2
8
66.5
4.0
3
73.0
3.6
2
4
60.5
2.7
2
63.2
4.2
1
66.5
4.3
2
75.0
4.4
7
a Geometric mean TSP level in yg/m .
Standard deviation of mean.
c Actual number of cities in test data analyzed in each
population region.
No population in this range.
5-7
-------
The diesel participate dose distribution developed for
Kansas City was projected for the national population,
assuming correlations similar to those developed for SMSA
population and population density versus TSP.
The overall national population dose distribution
procedure is summarized in Figure 5-2.
5.1.2 Population Exposures to TSP
Table 5-3 presents estimates of population exposure
relationships to TSP for each projection case. in every
case, the distribution consists of the minimum annual geo-
metric mean diesel emission particulate concentration to
which increments of the population are exposed. In both
best estimate and maximum diesel vehicle growth cases, the
concentrations are shown to increase progressively between
1981 and 1990. Emphasis is placed on the most exposed
fraction of the population because the attributable health
effects are expected to be more pronounced in this group.
In none of the cases is the diesel fraction of total TSP
emissions large enough to allow a meaningful graphic repre-
sentation of its relative contribution to the total TSP
level. Table 5-4 does show the contribution of diesels to
total population exposure as a function of TSP range. The
values on the table represent the percent of the total
population exposed to different TSP concentrations attrib-
5-8
-------
(1)
(2)
(3)
(4)
DIVIDE ALL SMSA'S INTO
15 SEGMENTS BASED ON
THEIR POPULATION AND
URBAN POPULATION DENSITY
X » SEGMENT MEAN TSP
EXPOSURE
P SEGMENT POPULATION
SD ' SEGMENT STANDARD
DEVIATION EXPOSURES
DIVIDE EACH SEGMENT
INTO FIVE SUBSEGMENTS
OF EQUAL POPULATION
BASED ON SD:
X.. = SUBSEGMENT MEAN
1J TSP EXPOSURE
0.2 P1 = SUBSEGMENT
POPULATION
GENERATE A POPULATION
DOSE DISTRIBUTION FOR _
EACH SUBSEGMENT BASED
TEST CITY DISTRIBUTION
FOR EACH DOSE DISTRIBUTION
INCREMENT (Xj, the k-th
EXPOSURE IS:
SUM POPULATION DISTRIBUTION
SUBSEGMENTS TO DEVELOP A
NATIONAL POPULATION DOSE
DISTRIBUTION.
n
-*J-
₯
Atc
AND
INCREMENT POPULATION IS
0.2P. x fk
0 1.0
CUMULATIVE FRACTION OF
POPULATION EXPOSED
01
I
TEST TEST DISTRIBUTION
(SECTION 4)
o
o
2
X
tc
= TEST CITY MEAN
EXPOSURE CONC.
k-th TEST CITY
POPULATION FRACTION
FRACTION OF POPULATION - f
Figure 5-2. Information flow diagram for the development of the
population TSP dose relationship for each projection case.
-------
Ul
I
Table 5-3. ESTIMATED ANNUAL EXPOSURE CONCENTRATIONS
OF TSP FROM DIESEL VEHICLE EXHAUST.
Millions
of people
Exposed
100
50
25
10
8
6
5
4
3
2
1
Exposure Concentration,
Best Est* i mfi-t-f^ Pa co I "
1981
0.24
0.34
0.42
0.52
0.54
0.57
0.58
0.59
0.61
0.63
0.66
1983
0.26
0.38
0.48
0.59
0.61
0.64
0.65
0.67
0.69
0.70
0.73
1985
0.31
0.44
0.56
0.71
0.74
0.78
0.80
0.84
0.87
0.93
1.02
1990
0.50
0.69
0.87
1.08
1.12
1.17
1.20
1.22
1.25
1.29
1.36
1981
0.25
0.38
0.47
0.58
0.60
0.63
0.65
0.66
0.68
0.70
0.72
jg/m
Max. Gro
1983
0.39
0.57
0.72
0.90
0.93
0.98
1.01
1.03
1.07
1.10
1.16
wtn Case
1985
0.58
0.81
1.02
1.26
1.30
1.36
1.40
1.42
1.46
1.51
1.59
1990
0.88
1.23
1.55
1.91
1.98
2.07
2.12
2.16
2.21
2.29
2.41
-------
Table 5-4. PERCENT OF POPULATION EXPOSURE TO TSP
ATTRIBUTABLE TO DIESEL EXHAUST EMISSIONS.
TSP
Cone ,
yg/m3
40
50
60
70
80
90
100
110
120
130
TSP - All sources
Millions of people
exposeda
152
134
103
74
52
36
24
14
8
4
Best Estimate Case
1981
0.1
0.3
0.3
0.2
0.8
1.4
2.5
3.6
3.4
2.8
1983
0.1
0.4
0.3
0.3
0.8
1.6
2.7
3.8
3.7
3.2
1985
0.2
0.7
0.8
0.8
0.9
1.9
3.2
3.8
3.7
3.2
1990
0.2
1.2
1.8
1.6
1.7
2.3
3.2
3.8
3.8
3.4
Max. Growth Case
1981
0.1
0.4
0.3
0.3
0.8
1.6
2.7
3.8
3.7
3.2
1983
0.2
0.9
1.2
1.1
1.7
2.4
3.2
3.8
3.9
3.4
1985
0.3
1.2
2.5
2.3
2.4
2.6
3.6
4.4
4.3
3.5
1990
0.2
1.5
3.0
3.3
3.3
4.2
5.8
8.8
10.1
8.3
Data based on 1981, best estimate diesel growth case.
-------
utable to diesel exhaust. Although no correlation was found
between total TSP levels and diesel particulate contribu-
tion, this table clearly shows that the higher the TSP
concentrations, the greater the diesel-contributed percent
impact is expected to be on the total population exposed.
Diesel vehicles contribute the greatest percentage at total
TSP levels between 110 and 120 pg/m3 (3.6 to 10.2 percent).
National TSP exposure relationships are also normalized
to show the population exposed to TSP concentrations ex-
ceeding the national standard (75 yg/m3 annual). Table 5-5
presents this information. This table also presents the
diesel vehicle contribution for each projection case. These
estimates are based on the following projected population
growth data:
1981
1983
1985
1990
Population x 10
SMSA
155.4
158.9
162.4 .
171.8
Total
261.0
266.3
277.2
297.2
The 1990 maximum diesel growth case shows the highest diesel
TSP contribution (2.2 million of the total 72.3 million
people exposed to greater than the TSP standard). Thus it
appears that the most sensitive indicator of the impact of
diesel TSP is obtained by matching the diesel TSP contribu-
5-12
-------
I
!-
U>
Table 5-5. ESTIMATED POPULATION EXPOSED TO MORE THAN THE
FEDERAL STANDARD FOR TSP.
Projection
year
1981
1983
1985
1990
Millions of people exposed to more than 75 yg/m
Diesel vehicle growth case
Best estimate
Total
exposed
62.7
64.3
66.4
71.1
Diesel
contrib.
0.4
0.4
0.6
1.0
Maximum growth
Total
exposed
62.8
64.7
67.3
72.3
Diesel
contrib.
0.4
0.8
1.5
2.2
-------
tion to the total TSP level at high exposure locations, as
presented in Table 5-4.
5.1.3 BaP Dosage Analysis
Data are not adequate to attempt a comprehensive in-
ventory of BaP emission sources in Kansas City. Thus
national BaP exposure data are used, and the contribution of
diesel vehicles is added.
Total BaP exposure is estimated by summing the exposure
of population to BaP emitted from coke ovens and BaP ex-
posure in locations without impacting coke ovens. The
reason for this approach is that NASN data on BaP concentra-
tions show that in cities with coke ovens the average BaP
concentration is 1.21 ng/m , with a range of 0.3 to 4.7
ng/m in 21 samples, whereas in cities without coke ovens
the average is 0.35 ng/m , with a range of 0.03 to 0.90
3 2
ng/m in 15 samples.
Locations for which coke ovens have a significant
impact are taken from a recent report from Stanford Research
2
Institute (SRI), in which population exposures to BaP
emitted from coke ovens are estimated by determining the
populations within a series of concentric rings around each
coke oven and estimating the annual average BaP concentra-
tion by use of monitoring data and an extrapolative modeling
technique. Impact of BaP from other sources is also in-
5-14
-------
eluded. The overall BaP exposure relationship developed by
SRI is shown in Table 5-6.
The BaP impact not attributable to coke ovens is based
on data from NASN's urban stations at locations without coke
ovens. The annual average BaP concentrations reported by
NASN are presented in Table 5-7. The mean concentration is
0.35 ng/m , with a standard deviation of 0.21, based on data
from 15 cities. In generation of a distribution from these
data it is assumed that the concentration is normally dis-
tributed. This analysis, shown in Table 5-8, is based on
the total 1976 U.S. urbanized population of 144.95 million
people minus 17.1 million exposed to coke oven emissions.
The diesel component of the total population exposure
to BaP is based on diesel BaP relationships developed from
the Kansas City emission inventory analysis. Each projec-
tion case for BaP includes 4 projection years, each with a
best estimate and maximum diesel growth case and low and
high diesel BaP emission factors. For each case the BaP
concentration is correlated with population density for 165
grid points. Grid areas, all of known population, are
divided into four groups as a function of population den-
sity:
5-15
-------
Table 5-6. ANNUAL AVERAGE EXPOSURE CONCENTRATIONS OF BaP
EMITTED BY COKE OVENS 2
Subgroup
concentration
range , ng/m^
95-100
50-55
45-50
40-45
35-40
30-35
25-30
20-25
15-20
10-15
8-10
6-8
5-6
4-5
3-4
2-3
1-2
0.5-1
0.2-0. 5
Cumulative number of people exposed
Background plus
coke oven emissions
1, 800
2,670
2,720
4,220
5,920
9,320
14,120
19,120
82,820
630,220
705,320
981,020
1,097,720
1,345,920
3,069,020
7,335,620
15,148,620
16,754,020
17, 106,620
Coke oven
emissions only3
1,800
2,670
2,720
4,220
5,920
8,320
9,920
18,920
82,620
219,920
662,620
798,920
995,220
1,182,320
1,971,620
3,216,820
8,243,520
12,923,120
17,106,620
4
Number exposed to indicated concentration or more.
5-16
-------
Table 5-7. ANNUAL AVERAGE AMBIENT BaP CONCENTRATIONS AT
NASN URBAN STATIONS WITHOUT COKE OVEN IMPACT
City
Montgomery
New York
Toledo
Charleston, WV
St. Louis
Spokane
Jacksonville
Honolulu
Baton Rouge
New Orleans
Duluth
Houston
Norfolk
Seattle
St. Paul
BaP
concentration ,
ng/m3
0.3
0.9
0.4
0.5
0.3
0.6
0.4
<0.1
0.1
0.2
0.3
0.2
0.2
0.4
0.4
Population
x 106
0.14
11.6
0.49
0.16
1.88
0.23
0.53
0.44
0.25
1.96
0.14
1.68
0.67
1.24
1.70
Urban
population
density range
2
4
2
2
4
3
1
3
2
4
1
3
2
3
2
Population density ranges
1 is 0-1.99 x 103 people/mi2
2 is 2-2.99 x 103 people/mi2
3 is 3-3.99 x 103 people/mi2
4 is 4 or more x 103 people/mi2
5-17
-------
Table 5-8. POPULATION EXPOSURE TO BaP IN URBAN AREAS WITHOUT COKE OVEN IMPACT
in
I
t->
oo
z = ^ for a given concentration level X where
BaP
concentration,
ng/m^
0. 90 +
0. 8-0.9
0.7-0.8
0.6-0.7
0.5-0.6
0.4-0.5
0.3-0. 4
0.2-0.3
0.1-0.2
<0.1
Z3
<2.75
2.25-2.75
1.75-2.25
1.25-1.75
0.75-1.25
0.25-0.75
-0.25-0.25
-0.75-0.25
-1.25-0.75
<-1.25
Fraction of popula-
tion less than X
concentration
0.997
0.988
0.960
0.894
0.773
0.599
0.401
0.227
0.106
0.040
% of
population in
interval
0.003
0.009
0. 028
0.066
0.121
0.174
0.198
0.174
0.121
0.106
Population
in cone.
interval x 10
0.38
1.15
3.58
8.44
15.47
22.25
25.31
22.25
15.47
X is the mean (0.35 ng/m3) and a is the standard deviation on the population sample,
-------
Population density,
Group 1000 people/mi2
1 <2
2 2 - 2.99
3 3 - 3.99
4 4 or more
The mean and standard deviations of the BaP values are
determined for each group. To develop a more sensitive
prediction of high BaP exposures, each group is then divided
into five equal population groups, and a subgroup mean is
estimated from the group mean and standard deviations. This
approach is identical to that used to generate subcell TSP
groups in Section 5.1.2.
A population exposure relationship for diesel BaP is
also developed from the Kansas City data for the 165 grid
points. An exposure distribution is developed by summing
individual grid area populations for a series of diesel BaP
concentration ranges.
National SMSA populations for each projection year are
summed within four urban population density ranges. The
four groups are then divided into the five equal subgroups
described above. A population-versus-dose distribution is
developed for each subgroup within each case by multiplying
the Kansas City exposure distribution by the ratio of the
subgroup over the Kansas City mean. Subgroups are then
summed to produce the overall national population distri-
butions.
5-19
-------
5.1.4 Population Exposure to BaP
The estimate of total U.S. exposure to BaP is presented
in Table 5-9. The values are based primarily on 1975 BaP
monitoring data and 1976 population estimates. There are no
projections to 1981-1990, because although ambient BaP
concentrations are tending downward,3 no reliable projec-
tions are available. The decline in BaP concentration
between 1966 and 1975 has more than compensated for the
increase in urbanized U.S. population.
Table 5-10 shows the population exposure for the 16 BaP
projection cases. The concentration range is far below the
lowest range of total BaP exposures presented in Table 5-9.
The highest exposure case (1990, with maximum diesel fleet
growth and high emissions) shows the 1 million most-exposed
people in the nation being exposed to 0.034 ng/m3, whereas
Table 5-9 shows 116 million exposed to 0.2 to 0.4 ng/m3 or
greater. A more sensitive means of estimating ambient BaP
exposure would help to produce a more realistic assessment
of the diesel contribution to ambient BaP; however, the
present paucity of empirical data appears to preclude
further refinement of the analysis.
5.2 PROJECTED MAXIMUM IMPACT OF DIESEL EMISSIONS ON AIR
QUALITY
Larsen's statistical transforms and Record's empirical
relationships are used to predict the regional maximum 24-
5-20
-------
Table 5-9. ESTIMATED TOTAL POPULATION DOSAGE OF BaP IN 1976
BaP
cone. ,
ng/m3
95-100
50-55
45-50
40-45
35-40
30-35
25-30
20-25
15-20
10-15
8-10
6-8
5-6
4-5
3-4
2-3
1-2
0.8-1
0.6-0.8
0.4-0.6
0.2-0.4
Population exposure to concentration greater
or equal to BaP level shown
Coke oven
exposure-*-
1,800
2,670
2,720
4,220
5,920
9,320
14,120
19,120
82,820
630,220
705,320
981,020
1.098 x 106
1.346 x 106
3.069 x 106
7.336 x 106
15.149 x 106
15.791 x 106
16.433 x 106
16.873 x 106
17.107 x 106
Urban exposure
x 106
1.534
13.552
51.268
98.828
Total
x 106
0.002
0.003
0.003
0.004
0.006
0.009
0.014
0.019
0.083
0.630
0.705
0.981
1.098
1.346
3.069
7.336
15.149
17.325
29.985
68.141
115.935
% of
population
<0.1
<0.1
<0.1
<0.i
<0.1
<0.1
<0.1
<0.1
<0.1
0.3
0.3
0.5
0.5
0.6
1.5
3.5
7.2
8.3
14.3
32.4
55.2
5-21
-------
Table 5-10. ANNUAL AVERAGE EXPOSURE CONCENTRATIONS OF BaP
FROM DIESEL EXHAUST EMISSIONS.
a) Low diesel exhaust emission rate case
Population
exposed
millions
100
50
25
10
8
6
5
4
3
2
1
Exposure level - ng/m x 10
Best estimate growth case
1981
0.4
0.6
0.9
1.1
1.2
1.2
1.3
1.3
1.4
1.5
1.6
1983
0.4
0.8
1.0
1.4
1.4
1.5
1.6
1.6
1.7
1.8
2.0
1985
0.5
0.9
1.2
1.6
1.7
1.8
1.8
1.9
2.0
2.2
2.4
1990
0.8
1.4
1.8
2.3
2.5
2.6
2.7
2.8
3.0
3.2
3.4
Max. qrowth case
1981
0.4
0.8
1.0
1.3
1.4
1. 5
1.5
1.6
1.7
1.8
2.0
1983
0.7
1.1
1.5
1.9
2.0
2.2
2.2
2.4
2.5
2.6
2.8
1985
0.9
1.6
2.1
2.7
2.8
1990
1.4
2.4
3.2
4.0
4.3
i
2.9
3.0
3.1
3.2
3.4
3.6
4.5
4.7
4.9
5.1
5.3
5.9
b) High diesel exhaust emission rate case
Population
exposed
millions
100
50
25
10
8
6
5
4
3
2
1
Exposure level - ng/m x 10
Best estimate growth case
1981
1. 0
1.7
2.3
3. 1
3.3
3.5
3.6
3.8
4.0
4.3
4.8
1983
2. 5
4.3
5.6
7.4
7.7
8.2
8.5
8.9
9.2
9.7
10.4
1985
2.9
4.9
6.6
8. 6
9.1
9.7
10.2
10.6
11.3
12.2
13.0
1990
4.6
7.6
10.0
13.2
13.7
14.5
15.1
16.0
16.7
17.8
19.0
Max. growth case
1981
2.4
4.1
5.5
7.1
7.5
7.9
8.2
8.4
8.9
9.3
10.0
1983
3.7
6.2
8.4
10.9
11.6
12.3
12.8
13.4
14.0
15.0
16.3
1985
1990
5.3 8.4
8.9 13.8
11.8 18.4
15.4 23.7
16. 1 '25.0
16.9
17.6
18.0
19.2
19.9
21.9
26.5
27.3
28.2
30.0
32.2
34.5
5-22
-------
hour and 1-hour TSP concentrations and maximum annual, 24-
hour, and 1-hour concentrations expected to occur near
roadways. The corresponding BaP concentrations are deter-
mined by applying a ratio of BaP to particulate emissions in
calculations for each projection year (presented in Table 4-
13).
5.2.1 Estimate of Maximum TSP Impact
The maximum annual TSP concentration from diesel-
powered vehicles in 1974 was 0.35 yg/m , occurring in the
immediate downtown area. This rather low maximum concen-
tration suggests that diesels contributed a relatively small
amount of particulate matter in that year.
The emission data presented earlier, however, indicate
that TSP concentrations attributable to diesels will in-
crease through 1990. Table 5-11 summarizes the predicted
changes in maximum concentrations over the projection
period. Regional annual mean TSP levels, taken from Table
4-13, are also presented.
Maximum 24-hour concentrations can be calculated by
4
application of Larsen's statistical transform. This
technique expresses air pollutant concentrations as a func-
tion of averaging time and frequency, and it assumes that
the following characteristics hold true for any given data
set under consideration:
5-23
-------
Table 5-11. PROJECTED CONCENTRATIONS OF TSP FROM DIESEL EXHAUST
(yg/m3)
Regional annual geometric mean
Regional 2 4 -hour maximum
Roadside annual geometric mean
Roadside 2 4 -hour maximum
1974
0.35
1.05
3.85
11.48
1981
Best
est.
0.45
1.34
4.95
14.76
Max.
0.56
1.66
6.16
18.36
1983
Best
est.
0.57
1.68
6.27
18.69
Max.
0.83
2.46
9.13
27.22
1985
Best
est.
0.65
1.93
7.15
21.31
Max.
1.13
3.38
12.43
37.05
1990
Best
est.
0.96
2.86
10.56
31.48
Max.
1.73
5.16
19.03
56.73
Jl
I
to
-------
0 pollutant concentrations are lognormally distri-
buted for all averaging times; and
0 median concentrations are proportional to averag-
ing time raised to an exponent.
Given these assumptions, Larsen's model may be ex-
pressed as:
C = M (S )Z
max g g
where C = the maximum concentration expected for
max the time period of concern (24 hours for
TSP)
M = annual geometric mean
S = standard geometric deviation
z = an empirical value representing the number
of standard deviations from the geometric
mean that corresponds to the desired averaging
period (or, in other words, to the desired
percentile on a normal probability curve)
The numerical value for z, obtainable from any standard
statistical text, is 2.94 for a 24-hour period and 3.81 for
a 1-hour period. The value for S is derived from the
r g
standard geometric deviations at the 18 Kansas City sampling
sites, which range from 1.29 to 1.69 and average of 1.45.
The average value is used in the Larsen computations.
The maximum regional 24-hour concentrations predicted
by this technique are presented in Table 5-11. The 1974
maximum is 1.05 yg/m , and the predicted maximum in 1990 is
either 2.86 or 5.16 yg/m . Again these values are rela-
tively low; 5.16 yg/m3 constitutes 3.4 percent of the 24-
hour secondary NAAQS.
5-25
-------
A critical limitation of any regional dispersion model
is that it predicts concentrations representative of average
conditions over relatively large geographical areas. In
application of the test city AQDM, predicted values can be
regarded as representative of areas 2 km by 2 km square.
Other investigations, however, demonstrate clearly that
particulate concentrations are not uniform over such areas;
rather, they vary considerably according to distance from
roadway and elevation above ground. '
In examining annual mean TSP concentrations at several
Q
stations. Record has found that the portion of the average
TSP concentration attributable to a nearby roadway can be
described by the following empirical expression:
C = (T/r) (0.265sin2 9 + 0.07cos2 0)
where C = average contribution of paved road to measured
TSP, yg/m~3
T = average daily traffic, vehicles/day
r = slant distance between monitor and roadway, ft
9 = arctan(z/x)
z = sampler height, ft
x = horizontal distance between roadway and
monitor, ft
Ludwig et al. used this expression to generate the set
of curves shown in Figure 4-4, which relate the measured TSP
concentrations to the ratio of average daily traffic (ADT)
Q
to height and setback of high volxmie samplers (T/r) .
On the basis of 1) the assximption that emissions due to
tire wear, vehicle exhaust, and reentrained dust share the
5-26
-------
same dispersion characteristics, 2) the calculation that
roadway-impacted high-volume samplers in the test city are
an average of 7 meters above ground and 31 meters from
roadways with 17,000 ADT, and 3) the data presented in >
Figure 4-4, it is estimated that a concentration measured 3
meters above ground and 4 meters from a roadway with 25,000
ADT would exceed a regionally representative concentration
by a factor of approximately 11. This value is an approxi-
mation that depends heavily upon the assumptions concerning
existing samplers. The true factor applicable to the test
city probably lies somewhere between 5 and 15.
Application of this estimated roadside adjustment
factor yields a maximum annual TSP concentration of 3.85
yg/m3 attributable to diesels in 1974. As shown in Table
5-11, the maximum annual concentration is projected to reach
10.56 or 19.03 yg/m3 by 1990 depending upon the assumption
concerning rate of introduction of diesel vehicles. Maximum
24-hour concentrations near roadways are calculated in a
similar fashion. The estimated concentration rises from
11.48 yg/m3 in 1974 to 31.48 or 56.73 yg/m in 1990.
The TSP concentrations projected to be contributed by
diesels near roadways in 1990 are not trivial. Calculations
with the maximum introduction rate yield concentrations that
are 25.3 and 37.8 percent of the primary and secondary
NAAQS, respectively.
5-27
-------
5.2.2 Estimate of Maximum BaP impact
BaP concentrations attributable to diesels are pro-
jected in a manner identical to that described for TSP,
except that BaP concentrations are indirectly modeled by
AQDM. It is assumed that a ratio of BaP to particulate
emissions can be applied to predicted TSP concentrations to
yield predicted BaP concentrations for each grid in a given
year. The ratio changes from year to year because of varia-
tions in emission factors and vehicle category mix.
The rates in Table 5-12, are calculated by defining the
VMT/yr (see Table 4-8), the particulate emission factor, and
the BaP emission factor for each diesel vehicle category.
Two BaP emission factors are used for each diesel type to
correspond with the factors given in Table 4-10. The data
are combined and totaled to yield total particulate and BaP
emissions from diesels, and ratios between the two totals
are developed. The process is repeated for each projection
year.
5.3 DISCUSSING EXPOSURE DATA
The assessment of diesel exhaust particulate contribu-
tion to ambient exposures of TSP and BaP includes national
annual exposure estimates and regional short-term and annual
exposure estimates of the most exposed group. Projections
of the national exposure distributions indicate that diesel
5-28
-------
Table 5-12. PROJECTED MAXIMUM CONCENTRATIONS OF BaP FROM DIESELS
Benzo-a-pyrene
(low emission factor) , ng/m x 10
Regional annual geometric mean
Regional 2 4 -hour maximum
Regional 1-hour maximum
Roadside annual geometric mean
Roadside 2 4 -hour maximum
Roadside 1-hour maximum
Benzo-a-pyrene- (high emission
factor) , ng/m x 10~3
Regional annual geometric mean
Regional 24-hour maximum
Regional 1-hour maximum
Roadside annual geometric mean
Roadside 24-hour maximum
Roadside 1-hour maximum
1974
0.8
2.4
3.3
8.9
26.6
36.7
4.3
12.9
17.7
147.6
142.0
196.1
1981
Best
est.
1.0
3.0
4.1
11.2
33.5
46.1
5.5
16.5
22.7
60.7
181.0
250.0
Max.
1.2
3.7
4.9
13.6
40.7
56.0
6.8
20.3
28.0
74.8
273.0
308.1
1983
Best
est.
1.3
3.8
5.4
13.9
41.3
57.3
6.9
20.6
28.4
76.0
226.6
313.1
Max.
1.8
5.4
7.4
19.8
59.0
81.6
10.1
29.9
41.6
110.6
329.5
455.6
1985
Best
est.
1.4
4.2
5.8
15.6
46.6
64.3
7.9
23.5
32.5
86.8
258.7
357.5
Max.
2.4
7.3
9.9
26.9
80.0
110.8
13.8
41.0
56.8
151.4
451.2
623.6
1990
Best
est.
2.1
6.3
8.7
23.1
68.9
95.2
11.7
34.8
48.2
128.4
382.7
528.9
Max.
3.7
11.1
15.2
41.0
122.3
168.9
21.1
62.8
86.9
231.6
690.3
954.0
-------
exhaust will constitute a relatively minor source of TSP and
BaP. Diesel vehicles are projected, however, to increase by
1.0 to 2.2 million the total population (over 70 million)
exposed to more than the primary annual TSP limit set by
standards. The diesel contribution to ambient BaP exposure
was found to be quite small. Neither national nor regional
estimating procedures are adequately refined to provide a
sensitive estimate of the levels of TSP and BaP to which the
most exposed segments of the population will be subjected.
Such a refinement would require a demographic data base of
populated residential areas relative to major roadways and
rough-estimate time-and-motion studies. The maximum impact
of diesel vehicle emissions near roadways does show the
anticipated peak exposures, although no population estimates
could be made. A regional 24-hour maximum TSP exposure for
the maximum 1990 diesel growth case was only 5 yg/m ; how-
ever, the maximum exposure case defined as a roadside loca-
3 3
tion resulted in 56.7 yg/m 24-hour exposure and a 19 yg/m
annual geometric mean exposure (25% of the primary NAAQS).
Similar BaP exposure estimates for 1990 resulted in a road-
side 24-hour maximum of 0.07 to 0.69 ng/m and an annual
geometric mean of 0.02 to 0.23 ng/m . Thus, it appears that
diesel vehicles are a potentially significant source of TSP
and BaP.
5-30
-------
REFERENCES FOR SECTION 5
1. U.S. Department of Commerce, Obers Projections, Volume
2. Washington, B.C. April 1974.
2 Suta, B.E., Human Population Exposures to Coke Oven
Atmospheric Emissions. U.S. Environmental Protection
Agency, Washington, D.C. August 1977.
3. Faoro, R.B., and J.A. Manning. Trends in Benzolajpy-
rene (1966-1975). U.S. Environmental Protection
Agency. Research Triangle Park, North Carolina.
Unpublished report.
4. R.I. Larsen. A New Mathematical Model of Air Pollution
Concentration Averaging Time and Frequency. Journal of
the Air Pollution Control Association. 19^24-30.
January 1969.
5. PEDCo Environmental, Inc., Cincinnati, Ohio. Analysis
of Probable Particulate Nonattairanent in the Kansas
City AQCR. Prepared for U.S. Environmental Protection
Agency, Kansas City, Missouri. February 1976.
6. Control of Reentrained Dust from Paved Streets. U.S.
Environmental Protection Agency, Kansas City, Missouri.
Publication No. 907/9-77-007. August 1977.
7. National Assessment of the Urban Particulate Problem.
Volume I: Summary of National Assessment. U.S.
Environmental Protection Agency, Research Triangle
Park, North CArolina. Publication No. EPA-450/
3-76-024. July 1976.
8. Record, F.A. Evaluation of the Suspended Particulate
Problem. GCA Corporation, Bedford, Massachusetts.
Prepared for U.S. Environmental Protection Agency.
1976.
9. Selecting Sites for Monitoring Total Suspended Partic-
ulates. U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina. Publication No. EPA-
450/3-77-018. June 1977.
5-31
-------
APPENDIX A
A-l
-------
Table A-l. FRACTIONS OF LIGHT-DUTY VEHICLE VMT IN PROJECTION YEARS
i
NJ
Fraction of light duty vehicle VMT
1974
Diesel vehicles
Age, years 1
2
3
4
5
6
7
8
9
10
11
12
>13
Total
0.0005
0.0008
0.0006
0.0006
0.0005
0.0005
0.0003
0.0003
0.0003
0.0002
0.0001
0.0001
0.0002
0.005
Gasoline vehicles
Age, years 1 0-1115
2 0.1422
3 0.1294
4 0.1204
5 0.1075
6 0.0935
7 0.0787
8 0.0627
9 0.0467
10 0.0318
11 0.0189
12 0.0129
>13 0.0388
Total 0.995
1981
Best
est.
0.0068
0.0057
0.0025
0.0006
0.0005
0.0005
0.0003
0.0003
0.0003
0.0002
0.0001
0.0001
0.0002
0.0181
0.1052
0.1373
0.1275
0.1204
0.1075
0.0935
0.0787
0.0627
0.0467
0.0318
0.0189
0.0129
0.0388
0.9819
Max.
0.0169
0.0143
0.0066
0.0006
0.0005
0.0005
0.0003
0.0003
0.0003
0.0002
0.0001
0.0001
0.0002
0.0409
0.0951
0.1287
0.1234
0-1204
0.1075
0.0935
0.0787
0.0627
0.0467
0.0318
0.0189
0.0129
0.0388
0.9591
1983
Best
est.
0.0112
0.0115
0.0078
0.0048
0.0021
0.0005
0.0003
0.0003
0.0003
0.0002
0.0001
0.0001
0.0002
0.0394
0.1008
0.1315
0.1222
0.1126
0.1059
0.0935
0.0787
0.0627
0.0467
0.0318
0.0189
0.0129
0.0.388
0.9606
Max.
0.0281
0.0287
0.0196
0.0121
0.0054
0.0005
0.0003
0.0003
0.0003
0.0002
0.0001
0.0001
0.0002
0.0959
0.0839
0.1143
0.1104
0-1089
0.1026
0.0935
0.0787
0.0627
0.0467
0.0318
0.0189
0.0129
0.0388
0.9041
1985
Best
est.
0.0112
0.0143
0.0130
0.0096
0.0064
0.0037
0.0017
0.0003
0.0003
0.0002
0.0001
0.0001
0.0002
0.0611
0.1008
0.1287
0.1170
0.1114
0.1016
0.0903
0.0773
0.0627
0.0467
0.0318
0.0189
0.0129
0.0388
0.9389
Max.
0.0281
0.0359
0.0326
0.0242
0.0163
0.0094
0.0039
0.0003
0.0003
0.0002
0.0001
0.0001
0.0002
0.1516
0.0839
0.1071
0.0974
0.0989
0.0917
0.0846
0.0751
0.0627
0.0467
0.0318
0.0189
0.0129
0.0388
0.8484
1990
Best
est.
0.0112
0.0143
0.0130
0.0121
0.0108
0.0094
0.0079
0.0063
0.0037
0.0019
0.0008
0.0003
0.0002
0.0919
0.1008
0.1287
0.1170
0.1089
0.0972
0.0846
0.0711
0.0567
0.0433
0.0301
0.0182
0.0127
0.0388
0.9081
Max.
0.0281
0.0359
0.0326
0.0302
0.0270
0.0236
0.0198
0.0157
0.0095
0.0049
0.0019
0.0007
0.0002
0.2301
0.0839
0.1071
0.0974
0.0908
0.0810
0.0704
0.0592
0.0473
0.0375
0.0271
0.0171
0.0123
0.0388
0.76»9
-------
Table A-2. FRACTIONS OF LIGHT-DUTY TRUCKS VMT IN PROJECTION YEARS
>
u>
Fraction of light duty truck VMT
1974
Diesel trucks
Age, years i
2
3
4
5
6
7
8
9
10
11
12
>13
Total
0.0005
0.0008
0.0007
0.0007
0.0005
0.0005
0.0004
0.0003
0.0003
0.0002
0.0001
0.0001
0.0004
0.0055
Gasoline trucks
Age, years i
2
3
4
5
6
7
8
9
10
11
12
>13
Total
0.0935
0.1372
0.1263
0.1303
0.0975
0.0825
0.0756
0.0567
0.0437
0.0318
0.0229
0.0159
0.0806
0.9945
1981
Best
est.
0.0057
0.0055
0.0026
0.0007
0.0005
0.0005
0.0004
0.0003
0.0003
0.0002
0.0001
0.0001
0.0004
0.0173
0.0883
0.1325
0.1244
0.1303
0.0975
0.0825
0.0756
0.0567
0.0437
0.0318
0.0229
0.0159
0.0806
0.9827
Max.
0.0141
0.0138
0.0064
0.0007
0.0005
0.0005
0.0004
0.0003
0.0003
0.0002
0.0001
0.0001
0.0004
0.0378
0.0799
0.1242
0.1206
0.1303
0.0975
0.0825
0.0756
0.0567
0.0437
0.0318
0.0229
0.0159
0.0806
0.9622
1983
Best
est.
0.0094
0.0110
0.0076
0.0052
0.0020
0.0005
0.0004
0.0003
0.0003
0.0002
0.0001
0.0001
0.0004
0.0375
0.0846
0.1270
0.1194
0.1258
0.0960
0.0825
0.0756
0.0567
0.0437
0.0318
0.0229
0.0159
0.0806
0.962S
Max.
0.0235
0.0276
0.0191
0.0131
0.0049
0.0005
0.0004
0.0003
0.0003
0.0002
0.0001
0.0001
0.0004
0.0905
0.0705
0.1104
0.1097
0.1179
0.0931
0.0825
0.0756
0.0567
0.0437
0.0318
0.0229
0.0159
0.0806
0.9095
1985
Best
est.
0.0194
0.0138
0.0129
0.0104
0.0059
0.0033
0.0015
0.0003
0.0003
0.0002
0.0001
0.0001
0.0004
0.0586
0.0846
0.1242
0.1141
0.1206
0.0921
0.0797
0.0745
0.0567
0.0437
0.0318
0.0229
0.0159
0.0806
0.9414
Max.
0.0235
0.0346
0.0319
0.0261
0.0137
0.0083
0.0038
0.0003
0.0003
0.0002
0.0001
0.0001
0.0004
0.1433
0.0705
0.1034
0.0951
0.1049
0.0843
0.0747
0.0722
0.0567
0.0437
0.0318
0.0229
0.0159
0.0806
0.8567
1990
Best
est.
0.0094
0.0138
0.0129
0.0131
0.0098
0.0083
0.0076
0.0057
0.0036
0.0019
0.0009
0.0003
0.0004
0.0877
0.0846
0.1242
0.1141
0.1179
0.0882
0.0747
0.0684
0.0513
0.0404
0.0301
0.0221
0.0157
0.0806
0.9123
Max.
0.0235
0.0346
0.0319
0.0327
0.0246
0.0208
0.0190
0.0144
0.0089
0.0047
0.0023
0.0008
0.0004
0.2186
0.0705
0.1034
0.0951
0.0983
0.0734
0.0622
0.0570
0.0426
0.0351
0.0273
0.0207
0.0152
0.0806
0.7814
-------
Table A-3. FRACTIONS OF HEAVY-DUTY TRUCK VMT FOR PROJECTION YEARS, URBAN ONLY
1974
Gasoline vehicles
Age, years 1
2
3
4
5
6
7
8
9
10
11
12
>13
Total
0.036
0.065
0.068
0.071
0.054
0.047
0.039
0.033
0.027
0.024
0.018
0.012
0.089
0.583
Diesel vehicles
Age, years 1
2
3
4
5
6
7
8
9
10
11
12
>13
Total
0.041
0.070
0.070
0.070
0.047
0.032
0.027
0.020
0.015
0.007
0.004
0.003
0.012
0.417
Fraction of heavy duty vehicles VMT, urban only
1981
Best "
est.
0.034
0.060
0.064
0.068
0.053
0.046
0.038
0.033
0.026
0.023
0.018
0.012
0.088
0.563
0.045
0.077
0.077
0.073
0.047
0.032
0.027
0.020
G.G15
0.007
0.004
0.003
6.012
0.437
Max.
0.029
0.053
0.057
0.061
0.051
0.044
0.037
0.031
0.026
0.022
0.017
0.012
0.085
0.525
0.055
0.090
0.086
0.083
0.045
0.031
0.026
0.019
0.014
0.007
0.004
0.003
0.011
0.475
1983
Best
est.
0.034
0.060
0.064
0.066
0.051
0.044
0.038
0.032
0.026
0.023
0.017
0.012
0.087
0.554
0.044
0.077
0.077
0.077
0.050
0.035
0.026
0.020
0.015
0.007
0.004
0.003
0.012
0.446
Max.
0.019
0.042
0.052
0.056
0.043
0.038
0.035
0.030
0.025
0.021
0.016
0.011
0.081
0.469
0.075
0.109
0.090
0.086
0.053
0.037
0.025
0.018
0.013
0.006
0.004
0.003
0.011
0.531
1985
Best
est.
0.032
0.059
0.063
0.066
0.050
0.043
0.036
0.031
0.026
0.023
0.017
0.012
0.086
0.544
0.048
0.076
0.076
0.076
0.049
0.037
0.031
0.022
0.014
0.007
0.004
0.003
0-011
0.456
Max.
0.009
0.025
0.034
0.042
0.037
0.033
0.028
0.025
0.022
0.020
0.015
0.010
0.074
0.374
0.092
0.140
0.120
0.099
0.054
0.037
0.029
0.021
0.012
0.006
0.004
0.002
a. 010
0.626
1990
Best
est.
0.015
0.032
0.039
0.046
0.039
0.037
0.032
0.027
0.022
0.019
0.015
0.010
0.077
0.410
0.078
0.128
0.114
0.096
0.057
0.033
0.028
0.020
0.013
0.008
0.004
0.003
0-010
0.590
Max.
0.0004
0.004
0.006
0.010
0.009
0.010
0.012
0.013
0.013
0.013
0.011
0.008
0.061
0.157
0.099
0.163
0.155
0.144
0.092
0.064
0.046
0.028
0.017
0.007
0.004
0.003
0-008
0-843
-------
APPENDIX B
The TSP levels shown in Table B-l represent the mean of
annual geometric average TSP levels for all ambient moni-
toring stations reporting in each SMSA shown for the case
year.
B-l
-------
Table B-l. AVERAGE SMSA TSP LEVELS CORRELATED WITH POPULATION PARAMETERS
CO
I
to
SMSA
population
' range, x 10*
<0.5
City - State
Topeka, KS
Roanoke, VA
South Bend, IN
Evansvllle, IN
Terre Haute. IN
Anderson, IN
Rockford, IL
Jollet, IL
Saginaw, MI
Green Bay, WI
Charlotte, NC
W1nston-Salem, NC
Tulsa, OK
Austin, TX
Billings, MT
Sioux Falls, SO
Spokane, HA
Tacoma, WA
Lancaster, PA
Columbia, SC
Greenville, SC
Chattanooga, TN
Knoxvllle, TN
Lincoln. NE
Oe$ Koines, IA
SMSA
population,
x 10°
0.16
0.18
0.28
0.23
0.18
0.14
0.27
0.16
0.22
0.16
0.41
0.3
0.48
0.30
0.09
0.10
0.29
0.41
0.32
0.32
0.30
0.31
0.40
0.17
0.29
Urban density,
x Io3/ni1l2
2.5
2.4
2.8
3.5
2.5
1.9
3.4
2.8
3.4
1.7
2.6
2.2
2.1
3.1
2.6
2.8
2.9
2.6
3.0
2.3
2.2
1.9
2.2
3.0
2.3
TSP X .
*g/m3
60.0
47.0
55.4
66.2
74.6
55.0
48.0
77.0
55.0
54.8
51.0
62.6
69.6
67.4
46.8
55.5
74.8
46.8
60.6
46.2
44.4
54.2
65.0
63.4
86.4
No. of
monitoring
stations
7
11
7
9
8
6
4
8
8
5
12
8
11
7
5
4
5
8
7
6
8
11
6
8
5
Year
1976
1975
1976
1976
1976
1976
1975
1975
1976
1976
1975
1975
1976
1975
1976
1976
1976
1976
1976
1975
1975
1975
1975
1976
1976
(continued)
-------
Table B-l (continued)
SMSA
population,
range, x 106
<0.5
0.5-1.0
City - State
Lexington, KY
Owensboro, KY
Morchester. MA
Flint. MI
Duluth, MN
Schenectacty, NY
Population
density range
Mean X
0
No. of SMSA's
Total
Albany. NY
Akron, OH
Toledo. OH
Columbus. OH
San Antonio, TX
Omaha. NE
Providence, RI
Norfolk-Portsmouth, VA
Louisville. KY
Grand Rapids, MI
OklataM City. OK
SMSA
population >
x 106
0.17
O.OB
0.35
-0.5
0.27
0.20
Summary
1 2
54.3 60.0
0.8 12.8
4 18
31
Urban-density.
x 103/m1l2 >
4.0
4.4
2.9
3.4
1.2
2.0
3 4
59.3 55.8
7.1 24.8
7 2
0.72
0.68
0.69
0.92
0.86
0.54
0.91
0.68
0.83
0.54
0.64
3.2
2.7
2.9
3.4
3.5
3.3
3.3
2.2
3.5
2.4
1.7
TSP if j
mg/rir
38.3
73.3
57.6
54.6
53.3
46.5
66.3
64.8
69.3
80.1
50.7
75.3
59.0
60.0
67.4
50.5
76.7
No. of
monitoring
stations
4
6
5
11
8
7
6
5
9
10
10
13
6
4
8
8
26
Year
1975
1975
1975
1976
1976
1974
1974
1976
1976
1976
1974
1976
1975
1975
1975
1976
1976
(continued)
-------
Table B-l (continued)
w
i
SWA
population,
range, x 10
0.5-1.0
1.0-1.5
City - State
Salt Lake City, UT
Nashville. TN
Richmond, VA
Rochester, NY
Syracuse, NY
Population
density range
Mean I
a
No. of SMSA's
Total
Buffalo, NY
Denver, CO
Portland, OR
Seattle-Everett, WA
Kansas City, KS
Atlanta, GA
Indianapolis, IN
New Orleans, LA
SWA
population,
x 10°
0.56
0.54
0.52
0.88
0.64
Sunmary
1
68.0
12.4
1
15
2
60.9
6.3
6
1.35
1.23
1.01
1.42
1.25
1.39
1.11
1.05
3
64.8
11.9
8
Urban density,
x 103/m1l2
2.6
1.3
2.9
4.1
3.9
4
54.4
0
1
5.1
3.6
3.1
3.0
2.2
2.7
2.2
11.5
TSP Jt,
ng/m3
62.3
59.2
58.7
54.4
72.7
78.5
87.6
50.2
63.6
63.6
53.4
70.6
56.4
No. of
monitoring
stations
7
17
11
7
9
8
7
11
9
]1
12
16
9
Year
^MOBBM
1976
1975
1975
1974
1973
1974
1976
1976
1976
1974
1975
1975
1976
(continued)
-------
Table B-l (continued)
i
ui
5MSA
population
range, x 106
1.0-1.5
>1.5
City - State
Milwaukee, WI
Population
density range
Mean I
o
No. of SMSA's
Total
Dallas. TX
Houston, TX
Chicago, IL
Philadelphia, PA
Baltimore, MD
Detroit. NI
Minneapolis, MN
Cleveland, OH
St. Louis, MO
New York, NY
Boston. MA
Population
density range
Mean X
o
No. of SMSA's
Total
SMSA
population,
x 106
1.40
SuMnary
1
-
-
0
9
2
63.0
7.1
4
3
67.1
18.9
3
1.56
1.99
7.0
4.82
2.07
5.2
1.81
2.06
2.36
9.00
2.76
S unwary
1
-
-
0
11
2
56.8
7.6
2
3
89.8
8.1
2
Urban density,
x 103/mlK
2.7
4
67.5
15.6
2
2.0
3.1
5.3
5.3
5.1
4.6
2.4
3.0
4.1
5.3
4.0
4
75.6
11.5
7
TSP X\-
mg/m
64.3
53.4
84.1
81.0
86.0
83.2
76.8
64.1
96.6
77.8
73.3
51.3
No. Of
monitoring
stations
24
16
8
23
12
10
8
18
25
20
25
20
Year
1975
1974
1974
1975
1976
1976
1976
1976
1976
1976
1976
1976
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
EPA-450/3-78-038
4. TITLE ANDSUBTITLE
Air Quality Assessment of Particulate Emissions from
Diesel-Powered Vehicles
3. RECIPIENT'S ACCESSION-NO.
5. REPORT DATE
March 1978
6. PERFORMING ORGANIZATION CODE
Terrence Briggs, Jim Throgmorton, Mark Karaffa
8. PERFORMING ORGANIZATION REPORT NO.
iGANIZATION NAME AND ADDRESS
PEDCo Environmental, Inc.
Chester Towers
11499 Chester Road
Cincinnati, Ohio 45246
10. PROGRAM ELEMENT NO.
68-02-2515
Work Assignment #17
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. EPA, OAQPS.SASD
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
FINAL
14. SPONSORING AGENCY CODE
200/04
5. SUPPLEMENTARY NOTES
Performed at the request of.OMSAPC/OAWM for an air quality assessment of the
potential impact of diesel vehicles.
The report presents estimates of the impact projected diesel-powered vehicle
sales will have on the levels of total suspended particulates (TSP) and benzo(a)p'yrene
(BaP).to which the population is exposed. A detailed particulate emission inventory
is developed for a representative test city (Kansas City, MO) for a base year (1974)
Emissions & population exposure to TSP & BaP are projected for 1981, 1983, 1985, &
1990. Emissions from all sources except diesel are assumed to remain constant in
order that the full impact of diesels can be seen & because insufficient time was
available to vary the model. An abbreviated discussion of possible health effects
attributable to organic emissions from diesel powered vehicles is included.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Air pollution
Air Quality assessment
TOtal suspended particulate matter (TSP)
3enzo(a)pyrene (BaP)
Dolynuclear aromatic hydrocarbons (PAH)
Ames test
Diesel emissions
Mutagenicity
Population exposed
Projected emissions
Diesel-powered vehicles
8. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (ThisReport)
21. NO. OF PAGES
154
20. SECURITY CLASS (Thispage)
22. PRICE
EPA Form 2220-1 (9-73)
------- |