United States EPA-600 /7-84~077
Environmental Protection
Agency July 1984
<>EPA Research and
Development
PAVED ROAD
PARTICULATE EMISSIONS
Source Category Report
Prepared for
Office of Air Quality Planning and Standards
Prepared by
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-84-077
July 1984
PAVED ROAD PARTICULATE EMISSIONS
Source Category Report
by
Chatten Cowherd, Jr. and Phillip J. Englehart
Midwest Research Institute
425 Volker Boulevard
Kansas City, Missouri 64110
EPA Contract 68-02-3158, Task 19
EPA Project Officer: Dale L. Harmon
Industrial Environmental Research Laboratory
Research Triangle Park, North Carolina 27711
Prepared for:
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
PREFACE
This report was prepared for the Environmental Protection Agency's In-
dustrial Environmental Research Laboratory under EPA Contract No. 68-02-3158,
Technical Directive No. 19. Dale L. Harmon was the project officer and
William B. Kuykendal was the task manager for the preparation of this
report. Dennis C. Drehmel and William B. Kuykendal served as tech-
nical project officers for the field testing portion of the study.
i1
-------
CONTENTS
Preface ii
Figures jv
Tables v
1.0 Introduction 1
2.0 Background Information 3
2.1 Street dust composition 3
2.2 Street dust loadings 4
2.3 Deposition and removal processes 4
2.4 Traffic-generated emissions 7
3.0 Sampling Site Selection 9
3.1 Site presurveys 9
3.2 Site selection 12
4.0 Sampling Equipment 19
4.1 Air sampling equipment 19
4.2 Roadway dust sampling equipment 26
4.3 Vehicle characterization equipment 26
5.0 Sampling and Analysis Procedures 29
5.1 Preparation of sample collection media 29
5.2 Pre-test procedures/evaluation of sampling
conditions 33
5.3 Air sampling 34
5.4 Sample handling and analysis 34
5.5 Emission factor calculation 36
6.0 Test Results 37
6.1 Test site conditions 37
6.2 Street surface particulate loadings 41
6.3 Airborne particulate concentrations 41
6.4 Emission factors 45
7.0 Multiple Regression Analysis 51
7.1 Introduction 51
7.2 Analysis and results 53
7.3 Comparative evaluation 58
7.4 Extension of the predictive equation to
different particulate size fractions 61
7.5 Emissions inventory applications 62
8.0 Summary and Conclusions 67
9.0 References 69
Appendices
A. Emission Factor Calculation Procedure A-l
B. Correction Parameter Calculation Procedures B-l
C. Proposed AP-42 Section C-l
111
-------
FIGURES
Number
1 Empirical relationship between sampler catch and traffic
volumes
2 Parameters for calculations of angle of unobstructed wind
flow 16
3 MRI exposure profiler 21
4 Sampling equipment deployment for winter testing in Greater
Kansas City Area 23
5 Sampling equipment deployment "A" for spring testing in
Greater St. Louis Area 24
6 Sampling equipment deployment "B" for spring testing in
Greater St. Louis Area 25
7 Predicted versus observed IP emission factors by land use
category 57
8 Comparison of emission factor models 60
A-l Downwind particle size distribution measured at a height of
1 m for Run M-3 A-5
A-2 Downwind particle size distribution measured at a height of
3 m for Run M-3 A-6
A-3 Exposure profile for Run M-3 ' A-9
IV
-------
TABLES
Number
1 Contaminant Loadings on Urban Street Surfaces 5
2 Estimated Deposition and Removal Rates 6
3 General Sampling Areas Designated by EPA 10
4 Field Data Requirements for Each Sampling Site Presurvey. ... 11
5 Application of Selection Criteria to Candidate Sampling Sites . 18
6 Field Measurements 20
7 Quality Control Procedures for Sampling Flow Rates.. 30
8 Quality Control Procedures for Sampling Media 31
9 Quality Control Procedures for Sampling Equipment 32
10 Quality Control Procedures for Data Processing and
Calculations 33
11 Criteria for Suspending or Terminating an Exposure Profiling
Test 35
12 Winter Test Site Conditions 38
13 Spring Test Site Conditions 39
14 Acceptable Tests for MLR 40
15 Paved Road Surface Dust Loadings 42
16 Particulate Concentrations and Plume Height 43
17 Summary of Particulate Size Ratios 44
18 Paved Road Emission Factors 46
19 Source Characterization Parameters 47
20 Summary of Paved Road Emission Factors 48
21 Summary of Source Characterization Parameters 49
22 Correlation Matrix for Entire Data Set 54
23 Correlation Matrix for "MRL" Data Set 54
24 Predicted Versus Observed IP Emission Factors 56
25 Paved Road Emission Factor Equation Parameters 62
26 Paved Roadway Classification 63
27 Summary of Silt Loadings (g/m2) for Urban Paved Roadways by
City ' 64
28 Recommended Emission Factors for Specific Roadway Categories
and Particle Size Fractions 65
A-l Inhalable Concentrations and Exposures A-3
B-l Silt Analysis Procedures B-2
-------
1.0 INTRODUCTION
Traffic-entrained participate from paved roads has been identified as
a major cause of nonattainment of air quality standards for total suspended
particulates (TSP) in urban areas.1 Therefore, the quantification of this
source is necessary to the development of effective strategies for the at-
tainment and maintenance of the TSP standards, as well as the anticipated
standard for inhalable particulate.
Based on previous limited field testing of this source,2 suspended par-
ticulate emissions have been found to vary in direct proportion to traffic
volume and surface loading of fines on the traveled portion of the street.
Measured emission factors for street particulate reentrainment added to ve-
hicle exhaust have been found to be an order of magnitude larger than the
factors for vehicle exhaust alone.3
This document presents the results of an expanded measurement program
to develop particulate emission factors for paved roads. The emission sam-
pling procedure used in this program provided emission factors for the fol-
lowing particle size ranges:
IP = Inhalable particulate matter consisting of particles equal to or
smaller than 15 urn in aerodynamic diameter
PM-10 = Particulate matter consisting of particles equal to or smaller
than 10 urn in aerodynamic diameter
FP = Fine particulate matter consisting of particles equal to or
smaller than 2.5 urn in aerodynamic diameter
-------
Results are presented for winter testing in the Kansas City, Missouri area
and spring testing in areas of St. Louis, Missouri and Granite City, Illinois
These results are used as a basis for the derivation of a matrix of emission
factors for specific road categories and particle size ranges.
The presentation of this report is organized in the following sequence:
Section 2 - Background Information
Section 3 - Sampling Site Selection
Section 4 - Sampling Equipment
Section 5 - Sampling and Analysis Procedures
Section 6 - Test Results
Section 7 - Test Data Reduction and Analysis
Section 8 - Conclusions and Recommendations
Appendix A - Emission Factor Calculation Procedure
Appendix B - Correction Parameters Calculation Procedures
Appendix C - Proposed AP-42 Section
-------
2.0 BACKGROUND INFORMATION
This section reviews published background information on the dynamics
of the paved roadway dust emissions problem.
2.1 STREET DUST COMPOSITION
In a comprehensive study of runoff from street surfaces as a source of
water pollution,4 81 samples were taken from streets in 12 cities by vacuum
sweeping and/or flushing. The samples were dry sieved and chemically ana-
lyzed to determine composition. The major constituent of street surface
contaminants was consistently found to be mineral-like matter similar to
common sand and silt. Typically, 78% of the material was located within
6 in. from the curb and 88% within 12 in. from the curb. The silt content
of the material (particles smaller than 75 micrometers (urn) in diameter)
fell in the 5 to 15% range reported elsewhere5 7 for surface dust from
paved streets and parking lots and from gravel roads and parking lots. In
addition, it was found that 5.9% of the material was less than 43 urn in
size. The silt size fraction, which is readily suspendable in the atmo-
sphere, was found to contain more than proportional amounts of the total
heavy metals and pesticides.
In a study which entailed a comprehensive review on the topic of re-
entrained dust from paved streets,8 129 samples of street surface materials
were taken in Kansas City and Cincinnati by means of broom sweeping and
subsequent vacuuming. The samples were weighed and analyzed by microscopy
to determine the particle size distribution. The results of the sample
analyses showed that approximately 9.5% of the paved road surface material
was less than 44 urn in size.
-------
2.2 STREET DUST LOADINGS
Table 1 summarizes the results of field measurements of surface load-
ings at sites in 12 cities.4 In addition to land use characteristics, dust
loadings were found to depend on:
Time elapsed since the last cleaning by mechanical means or by
substantial rainfall (exceeding 0.5 in. accumulation).
Street surface characteristics: asphalt streets had loadings
that were 80% higher than concrete-surfaced streets; and streets
in fair-to-poor condition had loadings about twice as high as
streets in good-to-excellent condition.
Public works practices: average loadings were reduced by regular
street cleaning (as reflected by lower values for commercial
areas), and loadings were increased during winter in areas where
sand and salt were applied.
Although traffic speed and density were believed to be important factors,
effects of these parameters could not be separated from more dominant fac-
tors such as land use.
2.3 DEPOSITION AND REMOVAL PROCESSES
On the average, vehicular carry-out from unpaved areas (unpaved roads
and parking lots, construction sites, demolition sites) may be the largest
source of dust on paved streets.8 Maximum carry-out occurs in wet weather
when dust emissions from open sources are at a minimum. In a study con-
ducted in the Seattle area,7'9 a car driven at 10 miles/hr on a wet gravel
road collected approximately 80 Ib of mud on tires and underbody, and carry-
out on tires from a wet unpaved parking lot averaged about 3/4 Ib/vehicle.
-------
TABLE 1. CONTAMINANT LOADINGS ON URBAN STREET SURFACES4
Land use
Mean initial
accumulation rate
(Ib/mile/day)
Loading intensity (Ib/curb mile)
Numerical Weighted
Minimum Maximum mean mean
Residential 373
Low/old/single
Low/old/multi
Med/new/s ingle
Med/old/single
Med/old/multi
Industrial 447
Light
Medi urn
Heavy
Commercial 226
Central business
district
Shopping center
Overall 348
120
31
180
260
140
260
280
240
60
63
1,900
1,300
1,200
1,900
6,900
12,000
1,300
12,000
1,200
640
850
890
430
-
1,400
2,600
890
3,500
290
290
1,200
2,800
290
1,500
There are 2 curb miles per street mile.
An American Public Works Association study10 found that 10.2 Ib of
dust under 1/8 in. in size comes onto each 100 ft of curb!ess paved road
in Chicago each day; this amount is cut by a factor of four if curbs are
added.
As evidence of the importance of the carry-out process, a positive
correlation has been observed between TSP concentration and the occurrence
of precipitation several days before sampling, i.e., after sufficient time
for the carry-out residue to dry out.11
In addition to vehicular carry-out, other potentially significant
sources of street dust are:
-------
Water and wind erosion from adjacent exposed areas (sparsely vege-
tated land, unpaved parking lots, etc.).
Motor vehicle exhaust, lubricant leaks, and tire and brake wear.
Truck spills.
Street repair.
Winter sanding and salting.
Atmospheric dustfall.
Vegetation and litter.
Table 2 presents typical annualized deposition and removal rates for
street surface material estimated by one study.8 The values were derived
by applying assumptions to data found in other literature sources. One as-
sumption was that the typical street has four lanes, is 50 ft wide, and has
an average daily traffic volume of 10,000 vehicles.
TABLE 2. ESTIMATED DEPOSITION AND REMOVAL RATES
Deposition Typical rate Removal Typical rate
process (Ib/curb-miles/day) process (Ib/curb-miles/day)
Mud and dirt 100 Reentrainment 100
carry-out . Displacement 40
Litter 40 Wind erosion 20
Biological debris 20 Rainfall 50
Ice control compounds 20 runoff
Dustfall 10 Sweeping 35
Pavement wear and 10
decomposition
Vehicle-related 17
(including tire
wear)
Spills < 2
Erosion from adjacent 20
areas
-------
In a recent field study of street surface contaminants in the
Washinton, D.C. area,12 roadway deposition of traffic-related materials was
found to be directly proportional to the traffic volume, at a rate of about
10 3 Ib/vehide-mile. The rate appeared to be independent of the loading
already present.
However, the accumulation of materials on the roadway has been found
to level off within a period of 3 to 10 days after a rain storm or street
cleaning.4'12 This leveling off occurs when traffic-related removal rates,
which increase with loading intensity, balance traffic-related deposition
rates. The equilibrium is established more rapidly with increasing traffic
speed.
2.4 TRAFFIC-GENERATED EMISSIONS
Few data on directly measured dust emissions from paved streets are
available in the literature. An isolated study of dust emissions from a
paved road in the Seattle area yielded an emission factor of 0.83 Ib/ve-
hicle-mile at 20 mph.7'9 The test road was noticeably dusty and had no
curbs or street cleaning program; it was located adjacent to gravel roads
and unpaved parking lots from which dirt was tracked. Dust emissions gen-
erated by vehicular traffic with average daily traffic exceeding 200 ve-
hicles was estimated to equal the amount removed by sweeping every 2 weeks.9
A single-valued emission factor of 3.7 g/vehicle-kilometer for dust en-
trainment from paved roads was developed from another field study.8 Emis-
sion measurements were obtained using the upwind-downwind technique with
four high-volume samplers,, one located 10 m upwind, with the remaining three
located at 10, 20, and 30 m downwind. Thirty-five successful tests were
completed. It was determined through microscopy that 78% (by weight) of the
emissions consisted of particulate less than 30 urn in size. Also through
optical microscopy it was found that 59% of the particulate collected was
mineral matter while 40% consisted of combustion products. It was also
concluded in this study that particulate emissions from a street are propor-
tional to traffic volume but independent of street surface dust loading.
-------
In a third field study quantitative emission factors for dust entrain-
ment from paved urban roads were developed using exposure profiling.2 Field
testing was conducted at three representative sites in the Kansas City area.
At one location, controlled amounts of pulverized top soil and gravel fines
were applied to the road surface. Eight tests were performed at the artifi-
cially loaded site, and five tests were made at a different site under ac-
tual traffic conditions. Emissions were found to vary directly with traffic
volume and surface loading of silt (fines). The dust emission factor for
normally loaded urban streets ranged from 1 to 15 g/vehicle-kilometer, de-
pending on land use. Approximately 90% of the emissions (by weight) were
found to be less than 30 urn in Stokes diameter and 50% less than 5 urn in
Stokes diameter, based on a particle density of 2.5 g/cm3.
-------
3.0 SAMPLING SITE SELECTION
Eight candidate sampling areas in Kansas. Missouri and Illinois were
designated by the Environmental Protection Agency (EPA) as representative
sites for the field study. As indicated in Table 3, these areas represent
a range of typical road, traffic, geographical, and environmental condi-
tions within residential, commercial, and industrial land uses. Each sam-
pling area contained a TSP monitoring site providing historical air quality
data. In 1975, ambient TSP concentrations in the candidate sampling areas
ranged from annual geometric means of 52 ug/m3 at Brauer School, Wyandotte
County, Kansas, to 157 ug/m3at 2001 East 20th, Granite City, Illinois.
3.1 SITE PRESURVEYS
Before going to the field, liaison was established with the appropriate
state and local environmental and transportation authorities. Support data
were compiled for each proposed sampling area to aid in careful site selec-
tion. This information included local street maps, topographic maps, street
maintenance and traffic data, and 1976 microinventories supplied by EPA.
Based on this research, previous Midwest Research Institute (MRI) road dust
sampling experience, and EPA recommendations, presurvey data requirements
were developed. Table 4 identifies specific field data that were obtained
during the presurveys for use in final sampling site selection.
It was decided to presurvey two or three sites within each sampling
area to provide roadway orientations suitable for sampling under various
wind direction ranges. Similarly, street segments were surveyed where mini-
mum obstruction to wind flow existed to provide a wide spread of wind fetch
corresponding to acceptable sampling conditions.
-------
TABLE 3. GENERAL SAMPLING AREAS DESIGNATED BY EPA
o
Sampling area
Preaurvey
Identification Name
I 321 Delaware
2 Brauer School
3 Baltimore and Miami
4 Slireve and 1-70
5 River dea Peres
Address of associated Ill-Vol
321 Delaware, Tonganoxle,
Kanaaa
K-7 and Leavenuortll Roadv
Wyandotte County, Kansas
Baltimore and Miami, Kansas
City, Kanaaa
Slireve and 1-70, St. Lou la,
Missouri
E. of Sulnlier, between
SAROAD
Identification
17-2000-001-F01
17-3840-008-1101
17-1800-002-1101
26-4280-061
26-4280-062
Ill-Vol
currently
operat iona I
No
No
No
Yes
Yea
Annual geometric
TSP concentrations
1974
82
45
107
111
-
1975
90
52
110
105
90
mean
.ilJB/JiLl
1976
107
-
-
96
-
Manchester and 1-44,
St. louls, Missouri
(15th and Madison 15cli and Madison, Granite City, 14-2960-10
III* j _
2001 f. 20th
Illinois
2001 E. 20tli. Granite City, 14-2960-09
Illinois
23rd and Madlaon 23rd and MaJlson, Granite City, 14-2960-07
Illinois
Yea
Yea
Yes
158
93
137
157
105
154
205
122
-------
TABLE 4. FIELD DATA REQUIREMENTS FOR EACH SAMPLING SITE PRESURVEY
1. Accurate location of each presurvey site on street and topographic map
2. Location of site with respect to reference Hi-Vol monitor.
3. Primary land use in the surrounding area.
4. Street information including:
Direction of travel
Number of travel and parking lanes
Presence of curbs and sidewalks
Street surface composition
Street surface roughness {qualitative - smooth, medium, rough)
5. Road maintenance information including:
Cleaning activities and frequency
Winter snow mitigation procedures
6. Street surface particulate loading in curb area, parking lanes, and
travel lanes (qualitative - light, medium, heavy)
7. Detailed sketch of the road dimensions.
8. Detailed sketch of surrounding area including:
Topography
Buildings (type, dimensions, address)
Open areas (use, dimensions)
Street names and locations
Fences, trees, billboards, and other miscellaneous information
9. A 15 to 30 min traffic count by vehicle- type.
10. A photographic survey including views of:
The sampling street (both directions)
The sampling set-up area
The fetch area
The road surface (travel lane and curb area)
11
-------
Seven sites in Areas 1, 2, and 3 (identified in Table 1) were surveyed
on August 2 and 3, 1979. An additional 11 sites were surveyed in Areas 4,
5, 6, and 7 on August 7 and 8, 1979. The 2001 East 20th location and the
15th and Madison location in Granite City, Illinois were combined into one
sampling area (Area 6) because of their proximity.
A wide variety of road and traffic characteristics were found in the
areas presurveyed. Equivalent hourly traffic volume ranged from 36 vehicles
at Site 2A to 2,944 vehicles at Site 5A. Road width varied from 22 ft at
Site 1C to 216 ft at Site 28. Both asphalt and concrete street surfaces,
curbed and uncurbed, were included. Qualitative evaluation of street surface
conditions indicated that the surfaces ranged from smooth to rough, and that
surface particulate loadings varied from light to heavy in comparison with
typically observed loadings.
3.2 SITE SELECTION
Three major criteria were used to determine the suitability of each
candidate site for sampling of road dust emissions by the exposure profiling
technique.6
1. Adequate space for sampling equipment,
2. Sufficient traffic and/or surface dust loading so that adequate
mass would be captured during a reasonable sampling time period, and
3. A wide range of acceptable wind directions.
3.2.1 Adequate Space
Adequate space for equipment deployment and easy accessibility to the
area is required for road dust sampling. All of the 18 candidate paved road
sites were chosen so as to provide necessary space, as well as accessibility
for the setup of all sampling equipment and to ensure the safety of the sam-
pling crew.
12
-------
3.2.2 Adequate Mass Catch
To provide for accurate determination of the dust emission rate from
exposure profiling data, at least 5 mg of sample should be collected by each
profiling head. Particulate concentration and sampling time must be suffi-
cient to provide the 5 mg weight gain under isokinetic sampling conditions.
This requirement is the most difficult to achieve for the highest sampling
head (located at 4 m above ground) because of the significant decrease in
particulate concentration with height.
An empirical relationship between sampler catch and traffic volume ob-
tained in MRI's previous testing of traffic entrained dust emissions is
illustrated in Figure 1. Assuming a typical silt loading (excluding curbs)
of 5 kg/km, approximately 3,600 vehicle passes are required to collect 5 mg
of sample (above background) on the top sampler; for roads with heavier
loadings, fewer passes are required.
Roads with light traffic are excluded from consideration because (a) it
is not possible to collect sufficient sample mass within an acceptable sam-
pling period (4 to 6 hrs), and (b) such roads probably do not contribute sub-
stantially to total emissions of traffic entrained dust in urban areas. In
any case, the emission factor equations developed in this study are expressed
in terms of emissions per unit of traffic volume (Kg/VKT); therefore these
equations should be applicable regardless of traffic density.
3.2.3 Adequate Traffic Volume
During the presurvey of each candidate testing site, traffic was counted
visually during a 15 to 40 min period. These traffic counts were then con-
verted to an average hourly count (AHT) by simple linear extrapolation in time
In order to evaluate each site with respect to the requirement of 3,600
vehicle passes in a 4-hr test period, it was necessary to convert the ob-
served AHT into an equivalent 4-hr count. This was accomplished by using
reported data on the diurnal variation of hourly traffic in Detroit, Chicago,
13
-------
100,000
3
o
•«*•
c
« 10,000
l>» *
o
a.
ju
u
1,000
O Data Point from
Reference 2
10.0
Mass (mg) Collected by Top Sampler (h = 4m. )
100.0
Figure 1. Empirical relationship between sampler catch and
traffic volume.
14
-------
Pittsburgh, and Toronto.13 In order to maximize the projected vehicle count,
it was assumed that testing would be conducted between 3:30 PM and 7:30 PM
which encompasses the evening traffic peak. After analysis of the collected
data, eleven of the eighteen site candidates met the traffic requirement and
were eligible for selection.
3.2.4 Acceptable Wind Directions
Wind directions that would successfully transport the traffic entrained
dust from paved streets to the exposure profiler depend on the following
factors:
Street Orientation - the mean (15-min average) direction of the wind
must lie within 45 degrees of the perpendicular to the road.
• Wind Fetch - the wind flowing toward the test roadway should not be
blocked by obstacles on the upwind side.
In order to evaluate the candidate sites for the wind fetch requirement,
the arc of wind direction for which the wind would flow freely between the
two nearest upwind obstacles (houses, buildings, or trees) was calculated
as follows:
6 = arctan ^7
where 9 represents the half angle of the arc, b is half the distance be-
tween the two blocking obstacles (fetch), and a is the perpendicular
distance from the line joining the rear corners of the obstacles to the
proposed location of the profiler (5 m from the downwind edge of the road-
way.) Figure 2 illustrates these parameters.
3.2.5 Summary of Selection Criteri?
Selection criteria for sampling sites included, in descending order of
importance:
15
-------
Wind
b/2
Obstacle
to Flow
b/2
Obstacle
to Flow
Test Roadway
v
o
'Exposure Profiler
Figure 2. Parameters for calculations of angle
of unobstructed wind flow.
16
-------
Adequate space - for operation of equipment and for safety of crew.
Adequate mass - as determined by number of vehicle passes in a 4-hr
test period.
• Wind direction - range of unobstructed wind directions.
A summary of the selection criteria as applied to each site, is shown in
Table 5. It should be noted that accessibility was determined during the
presurveys, and all candidates were assured of this.
Suitability was determined by an examination of all criteria, and rat-
ings were assigned as follows: (A = primary choice, B = alternate choice,
C = emergency choice, R = rejected). Those sites designated A or B were
selected to be considered for source testing. It should be noted that sam-
pling sites 1A and 1C were considered primary because it was desirable to
sample at rural locations. These sites were easily accessible to the sam-
pling crew so that a longer sampling period was possible.
17
-------
TABLE 5. APPLICATION OF SELECTION CRITERIA TO CANDIDATE
SAMPLING SITES
Site
1-A
1-B
1-C
2-A
2-B
3-A
3-B
4- A
4-B
5-A
5-B
5-C
6-A
6-B
6-C
7-A
7-B
7-C
Traffic
count
(peak
4 hr)a
Low
Low
Low
Low
High
High
Low
Medi urn
Low
High
Medium
High
Medi urn
Medium
Medi urn
Low
Medium
Low
Curbed
Yes
Yes
No
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
One side
One side
Street
parti culats
loading
Moderate
Moderate
Moderate
Moderate
Light
Light
Moderate
Moderate
Moderate
Light
Moderate
Light
Moderate
Moderate
Heavy
Moderate
Moderate
Heavy
Adequate Wind Suitability
sample direction, for
mass versatility testing
No
No
No
No
Yes
Yes
No
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
No
N/2Q°
E/40°
W/90°
N/90°, 5/90°
W/90°, E/90°
W/70°, E/50°
N/40°, S/20°
W/5Q°, E/70°
WNW/900
N/90°, S/90°
N/90°
ESE/200
NE/400
SE/200
ESE/400
SE/400
NNW/400, SE/900
ENE/700
Pd
Rd
Pa
R
P
P
R
P
R
P
S
E
P
E
P
R
P
S
Four-hour traffic count: low = 1,000 to 4,000; medium = 4,000 to
8,000; high = > 8,000.
Centerline directions and ranges of unobstructed wind flow.
P = prime site; S = alternate site; E = emergency site; R = rejected
site.
Sampling will be attempted for periods longer than 4 hr (see text).
18
-------
4.0 SAMPLING EQUIPMENT
A variety of sampling equipment was utilized in this study to measure
particulate emissions, roadway surface particulate loadings, and traffic
characteristics.
Table 6 specifies the kinds and frequencies of field measurements
that were conducted during each run. "Composite" samples denote a set of
single samples taken from several locations in the area; "integrated" sam-
ples are those taken at one location for the duration of the run.
4.1 AIR SAMPLING EQUIPMENT
The primary tool for quantification of emissions was the MRI exposure
profiler, which was developed under EPA Contract No. 68-02-0619.6 The pro-
filer (Figure 3) consists of a portable tower (4 to 6 m height) supporting
an array of four sampling heads. Each sampling head is operated as an isoki-
netic total particulate matter exposure sampler directing passage of the
flow stream through a settling chamber (trapping particles larger than about
50 urn in diameter) and then upward through a standard 8 in. by 10 in. glass
fiber filter positioned horizontally. Sampling intakes are pointed into the
wind, and sampling velocity of each intake is adjusted to match the local
mean wind speed, as determined prior to each test. Throughout each test,
wind speed is monitored by recording anemometers at two heights, and the
vertical wind profile of wind speed is determined by assuming a logarithmic
distribution. Normally, the exposure profiler is positioned at a distance
of 5 m from the downwind edge of the road.
19
-------
TABLE 6. FIELD MEASUREMENTS
ro
CD
1.
2.
3.
4.
lest Parameter
Meteorology
a. Wind speed
b. Wind direction
c. Cloud Cover
d. Temperature
e. Relative humidity
lload Surface
a. Pavement type
b. Surface condition
c. Particulate loading
d. Particulate texture
Vehicular Traffic
a. Mix
b. Count
Atmospheric Particulate
a. Total participate
b. Total suspended
participate
c. Inhalable particulate
d. Inhalable particulate
Units
m/s
deg
%
"C
%
q/m2
% silt
-
mass cone. (|ig/m3)
mass cone. (|ig/m3)
mass cone. (|ig/m3)
mass size dist. (ug)
Samp I i no Mode
continuous
continuous
s i ng 1 e
single
single
compos i te
composite
multiple
multiple
multiple
cumulative
integrated
integrated
integrated
integrated
Measurement/ Instrument
method
warm wire anemometer
wind vane
visual observation
sling psychrometer
sling psychrometer
observation
observation
dry vacuuming
dry sieving
observation
pneumatic tube axle
counters
Iso-kinetic prof i ler
Hi-Volume sampler
size selective inlet
slotted high-volume
Manufacturer/Model
Kurz Model 410
Wong Eco-System 111
Taylor cat. no. 146-761
Taylor cat. no. 146-761
Hoover, Model S2015 Quick Broom
Forney, Inc. , IA-410 Sieve Shaker
Streeter Amet, J R traf f {counter.
Model No. 160
MRI developed under EPA Contract
No. 68-02-0619
Sierra Instruments, Inc., Model 305
Andersen Samplers, Inc., Model 7001
Sierra Instruments, Inc., Model 23(
cascade impactor
-------
Flow Control Circuit Box
Warm-Wire
Anemometer
Electronic Readout/Control Unit
Figure 3. i'1RI exposure profiler.
21
-------
The recently developed EPA version of the size selective inlet (SSI)
for the high volume air sampler was used to determine the IP concentrations.
To obtain the particle size distribution of IP, a high-volume parallel-slot
cascade impactor (CI) with greased substrates was positioned beneath the
SSI. This five stage cascade impactor has, at a flow rate of 40 SCFM, 50%
efficiency cutpoints at 7.2, 3.0, 1.5, 0.95, and 0.49 urn aerodynamic diam-
eter.
The cascade impactors were used in conjunction with the SSI's for two
reasons. First, the 15 urn cutpoint for inhalable particulate (IP) was not
well established as a standard at the time of this study. With the use of
the cascade impactor data, alternate cutpoints for IP could be determined.
The second reason for using the cascade impactors was to obtain a fine par-
ticle (FP) cutpoint of 2.5 urn.
Other air sampling instrumentation used included standard high-volume
air samplers to measure total suspended particulate matter (TSP) consisting
of particles smaller than about 30 urn in aerodynamic diameter.
Three variations of air sampling equipment deployment were used in
this study. The deployment used in the winter testing (Kansas City area)
is shown in Figure 4. The two deployments of sampling equipment for the
spring testing (St. Louis/Granite City areas) are shown in Figures 5 and 6.
The basic downwind equipment included an exposure profiling sampling
system with four sampling heads positioned at I- to 4-m heights. In addi-
tion, size selective inlets fitted with high-volume cascade impactors were
placed at 1- and 3-m heights to determine the respective IP and FP mass
fractions of the total particulate emissions. A standard high-volume air
sampler was also operated at a height of 2 m.
Optional equipment operated downwind in the winter testing included a
1-m high size-selective inlet, fitted with a cascade impactor with ungreased
substrates. No optional equipment was operated downwind in the St. Loui.s
testing.
22
-------
2m A
3m
IV)
U)
-#gg#£ Roadway
4m
2m
^L.A
mm
r4m
3m
2m
U0m
LEGEND:
A Standard Hi Vol
D Profiler Head
Q Hi Vol with SSI
Q Hi Vol with SSI/lmpactor
O Hi Vol with Cyclone/lmpactor
G Greased Filters
Figure 4. Sampling equipment deployment for winter testing in Greater Kansas City Area.
-------
3m
4m f"
3m -
2m
1m
A 2m
v
^:^:::^:^::: Roadway
~4m
-3m LEGEND:
A Standard Hi Vol
h2m D Profiler Head
^Hi Vol with SSI
-1m ^ Hi Vol with SSI/Impactor
O Hi Vol with Cyclone/Impactor
L0m G Greased Filters
Figure 5. Sampling equipment deployment "A" for spring testing in Greater St. Louis Area.
-------
3m
A 2m
ro
en
Roadway ;::::::::::::>::^:o:::::::::::::::::::::::::::::::::::::::::^:::::::::::::::::::::::::::::::::::
4m
3m
2m
1m
Om
LEGEND:
A Standard Hi Vol
D Profiler Head
Hi Vol with SSI
Hi Vol with SSI/lmpactor
Hi Vol with Cyclone/lmpactor
G Greased Filters
Figure 6. Sampling equipment deployment "B" for spring testing in Greater St. Louis Area.
-------
The basic upwind equipment included SSIs and a standard high-volume
air sampler. In the Kansas City testing, two SSIs at heights of 2 and 4 m
were used to obtain the IP concentration of upwind particulate matter. In
the St. Louis testing, the primary upwind equipment included a high-volume
air sampler and an SSI/CI with greased substrates. For the secondary de-
ployment array, two SSIs were used to obtain the vertical distribution of
IP.
4.2 ROADWAY DUST SAMPLING EQUIPMENT
Samples of the dust found on the roadway surface were collected during
the source tests. In order to collect this surface dust, it was necessary
to close each traffic lane for a period of approximately'15 min. Normally,
an area that was 3 m by the width of a lane was sampled. For each test,
collection of material from all travel lanes and curb areas (extending to
about 25-30 cm from the curbing) was attempted. A hand held portable
vacuum cleaner was used to collect the roadway dust. The attached brush
on the collection inlet was used to abrade surface compacted dust and to
remove dust from the crevices of the road surface. Vacuuming was preceded
by broom sweeping if large aggregate was present.
4.3 VEHICLE CHARACTERIZATION EQUIPMENT
The characteristics of the vehicular traffic during the source testing,
were determined by both automatic and manual means. The vehicular charac-
teristics included: (a) total traffic count, (b) mean traffic speed, and
(c) vehicle mix.
Total vehicle count was determined by using pneumatic-tube counters.
In order to convert the axle counts to total vehicles, visual 1-min vehicle
mix summaries were tabulated every 15 min during the source testing. The
vehicle mix summaries recorded vehicle type, number of vehicle axles and
number of vehicle wheels. From this information, the total axle counts
were corrected to the total number of vehicles by type.
26
-------
The speed of the traveling vehicles was determined by noting the posted
speed limits of the roadway test section. As a check against this determin-
ation method, speeds of the vehicles were determined through the occasional
use of a hand-held radar gun.
The weights of the vehicle types were estimated by consulting (a) auto-
mobile literature concerning curb weights of vehicles and (b) distributors
of medium duty and semi-trailer type trucks as to their curb weights.
27
-------
(This page intentionally blank)
28
-------
5.0 SAMPLING AND ANALYSIS PROCEDURES
The sampling and analysis procedures employed in this study were sub-
ject to the Quality Control guidelines summarized in Tables 7 to 10. These
procedures met or exceeded the requirements specified by EPA.14'15
As part of the QC program for this study, routine audits of sampling
and analysis procedures were performed. The purpose of the audits was to
demonstrate that measurements were made within acceptable control condi-
tions for particulate source sampling and to assess the source testing data
for precision and accuracy. Examples of items audited include gravimetric
analysis, flowrate calibration, data processing, and emission factor cal-
culation. The mandatory use of specially designed reporting forms for sam-
pling and analysis data obtained in the field and laboratory aided in the
auditing procedure. Further detail on specific sampling and analysis pro-
cedures are provided in the following sections.
5.1 PREPARATION OF SAMPLE COLLECTION MEDIA
Particulate samples were collected on type A slotted glass fiber im-
pactor substrates and on Type AE (8 x 10 in.) glass fiber filters. To
minimize the problem of particle bounce, the glass fiber cascade impactor
substrates were greased. The grease solution was prepared by dissolving
140 g of stopcock grease in one liter of reagent grade toluene. No grease
was applied to the borders and backs of the substrates. The substrates
were handled, transported and stored in specially designed frames which
protected the greased surfaces.
29
-------
TABLE 7. QUALITY CONTROL PROCEDURES FOR SAMPLING FLOW RATES
Activity
QC check/requirement
Calibration
Profilers, hi-vols, and
impactors
Single-point checks
Profilers, hi-vols, and
impactors
• Alternative
Orifice calibration
Calibrate flows in operating ranges
using calibration orifice every two
weeks at each regional site prior
to testing.
Check 25% of units with rotameter
calibration orifice or electronic
calibrator once at each site prior
to testing (different units each
time). If any flows deviate by
more than 7%, check all other units
of same type and recalibrate non-
complying units. (See alternative
below.)
If flows cannot be checked at test
site, check all units every two
weeks and recalibrate units which
deviate by more than 7%.
Calibrate against displaced volume
test meter annually.
30
-------
TABLE 8. QUALITY CONTROL PROCEDURES FOR SAMPLING MEDIA
Activity
QC check/requirement
Preparation
Conditioning
Weighing
Auditing of weights
(Tare and Final)
Correction for handling
effects
Calibration of balance
Inspect and imprint glass fiber
media with ID numbers.
Equilibrate media for 24 hr in
clean controlled room with relative
humidity of less than 50% (vari-
ation of less than ± 5%) and with
temperature between 20° ± and 25°C
(variation of less than ± 3%).
Weigh hi-vol filters and impactor
substrates to nearest 0.1 mg.
Independently verify weights of 10%
of filters and substrates (at least
4 from each batch). Reweigh batch
if weights of any hi-vol filters
(8 x 10 in.) or substrates deviate
by more than ±1.0 and ± 0.5 mg
respectively.
Weigh and handle at least one blank
for each 1 to 10 filters or sub-
strates of each type for each test.
Balance to be calibrated once per
year by certified manufacturer's
representative check prior to each
use with laboratory Class S weights.
31
-------
TABLE 9. QUALITY CONTROL PROCEDURES FOR SAMPLING EQUIPMENT
Activity
QC check/requirements
Maintenance
• All samplers
Equipment siting
Operation
• Timing
Isokinetic sampling
(profilers only)
Prevention of static
mode deposition
Check motors, gaskets, timers, and
flow measuring devices at each re-
gional site prior to testing.
Separate colocated samplers by 3 to
10 equipment widths.
Start and stop all samplers during
time spans not exceeding 1 min.
Adjust all sampling intake orienta-
tations whenever mean (15 min. average)
wind direction changes by more than
30 degrees.
Adjust all sampling rates whenever mean
(15 min average) wind speed approach-
ing samplers changes by more than 20%.
Cap sampler inlets prior to and im-
mediately after sampling.
32
-------
TABLE 10. QUALITY CONTROL PROCEDURES FOR DATA PROCESSING AND CALCULATIONS
Activity QA check/requirements
Data recording Use specially designed data forms to
assure all necessary data are re-
corded. All data sheets must be
initialed and dated.
Calculations Independently verify 10% of calcu-
lations of each type. Recheck all
calculations if any value audited
deviates by more than ± 3%.
Prior to the initial weighing, the greased substrates and filters were
equilibrated for 24 hr at constant temperature and humidity in a special
weighing room. During weighing, the balance was checked at frequent inter-
vals with standard weights to assure accuracy. The substrates and filters
remained in the same controlled environment for another 24 hr, after which
a second analyst reweighed them as a precision check. Substrates or filters
that could not pass audit limits were discarded. Ten percent of the sub-
strates and filters taken to the field were used as blanks. Paper bags
for the vacuum cleaner were conditioned and tared in a similar manner.
5.2 PRE-TEST PROCEDURES/EVALUATION OF SAMPLING CONDITIONS
Prior to equipment deployment, a number of decisions were made as to
the potential for acceptable source testing conditions. These decisions
were based on forecast information obtained fom the local U.S. Weather Ser-
vice office. A specific sampling location was identified based on the
prognosticated wind direction. Sampling would ensue only if the wind speed
forecast was between 4 and 20 mph. Sampling was not planned if there was a
high probability of measurable precipitation (normally > 20%) or if the
road surface was damp.
33
-------
If conditions were considered acceptable, the sampling equipment was
transported to the site, and deployment was initiated. This procedure nor-
mally took 1 to 2 hr to complete. During this period, the samples of the
road surface particulate were collected at a location within 100 m of the
air sampling site. For a 4-lane roadway, the collection of road surface
particulate samples required approximately 1 hr to complete.
5.3 AIR SAMPLING
Once the source testing equipment was set up and filters put in place,
air sampling commenced. Information recorded for each test included: (a)
exposure profiler - start/stop times, wind speed profiles and sampler flow
rates (determined every 15 min) and wind direction (relative to roadway per-
pendicular); SSI/CIs, Hi-Vols - start/stop times, and sampler flow rates,
(c) vehicle traffic - total count, vehicle mix count, and speed; and (d)
general meteorology - wind speed and direction, temperature, relative hu-
midity and solar radiation.
Sampling usually lasted 4 to 6 hr. Occasionally, sampling was inter-
rupted due to occurrence of unacceptable meteorological conditions and then
restarted when suitable conditions returned. Table 11 presents the criteria
used for suspending or terminating a source test.
The upwind-background samplers were normally operated concurrent with
the downwind samplers. Care was taken to position the upwind samplers away
from any influencing particulate emission source.
5.4 SAMPLE HANDLING AND ANALYSIS
To prevent particulate losses, the exposed media were carefully trans-
ferred at the end of each run to protective containers within the MRI instru-
ment van. Exposed filters and substrates were placed in individual glassine
envelopes and numbered file folders and then returned to the MRI laboratory.
Particulate that collected on the interior surfaces of each exposure probe
was rinsed with distilled water into separate glass jars.
34
-------
TABLE 11. CRITERIA FOR SUSPENDING OR TERMINATING AN EXPOSURE PROFILING TEST
A test will be suspended or terminated if:a
1. Rainfall ensues during equipment setup or when sampling is in progress.
2. Mean wind speed during sampling moves outside the 4 to 20 mph accept-
able range for more than 20% of the sampling time.
3. The angle between mean wind direction and the perpendicular to the path
of the moving point source during sampling exceeds 45 degrees for more
than 20% of the sampling time.
4. Mean wind direction during sampling shifts by more than 30 degrees from
profiler intake direction.
5. Mean wind speed approaching profiler sampling intake is less than 80% or
greater than 120% of intake speed.
6. Daylight is insufficient for safe equipment operation.
7. Source condition deviates from predetermined criteria (e.g., occurrence
of truck spill).
a "Mean" denotes a 15-min average.
35
-------
When exposed substrates and filters (and the associated blanks) were
returned from the field, they were equilibrated under the same conditions
as the initial weighing. After reweighing, 10% were audited to check pre-
cision.
The vacuum bags were weighed to determine total net mass collected.
Then the dust was removed from the bags and was dry sieved. The screen
sizes used for the dry sieving process were the following: 3/8 in., 4, 10,
20, 40, 100, 140, and 200 mesh. The material passing a 200 mesh screen is
referred to as silt content.
5.5 EMISSION FACTOR CALCULATION
The primary quantities used in obtaining emission factors in this study
were the concentrations measured by the size selective inlet/cascade im-
pactor sampler combinations. This combination not only provides a reliable
cut point for 15 urn but also permits the determination of concentrations in
other particle size ranges. The MRI exposure profiler collects total par-
ti cul ate matter and enables one to determine the plume height. A knowledge
of the vertical distributions of plume concentration is necessary in the
numerical integration required to calculate emission factors. The emission
factor calculation procedure is presented in Appendix A.
36
-------
6.0 TEST RESULTS
6.1 TEST SITE CONDITIONS
As indicated in Table 12, the winter testing was conducted during the
months of February and March 1980 at three sites in the Kansas City area:
7th Street in Kansas City, Kansas; Volker Boulevard/Rockhill Road in Kansas
City, Missouri, and 4th Street in Tonganoxie, Kansas. The spring testing
(Table 13) was conducted during the month of May 1980, at two sites in
St. Louis (1-44 and Kingshighway) and at three closely spaced sites in
Granite City, Illinois.
The sites where source testing occurred can be classified into four
land use categories, based upon source parameters such as road type, vehi-
cle mix, and vehicle speed. The categories are: commercial/industrial;
commercial/residential; expressway, and rural town. Much of the data pre-
sented in the following sections is broken out according to these categor-
ies.
Table 14 presents an evaluation of the source tests according to estab-
lished QA criteria. Seven of the nine Kansas City tests (Runs M-l, -2, -3,
-6, -7, -8, and -9) met all of the. QA criteria, while only three of the ten
tests conducted in the St. Louis, Granite City area (Runs M-ll, -12, and
-15) met the QA criteria. The spring testing, in particular, was hampered
by unseasonably light wind conditions as wind speed for four of the ten
tests did not meet the minimum wind speed criterion of 4 mph.
The results of the ten runs which met the QA criteria were used as
input to Multiple Linear Regression (MLR) analysis (see Section 7.0). These
runs are subsequently referred to as the "MLR" data set.
37
-------
TABLE 12. WINTER TEST SITE CONDITIONS (Kansas City Area Paved Roads)
Profiler Operation
Run
M-l
M-2
M-3
M-4
M-5
M-6
H-7
M-8
M-9
Site
7 Lh Street
7th Street
7th Street
Volker Boulevard
Volker Boulevard
Roc kill 11 Road
Volker Boulevard
Tonganoxie - 4th St.
7th Street
Date
2/7/80
2/11/80
2/12/BO
2/26/80
2/27/80
2/28/80
3/5/80
3/6/80
3/10/80
Start
Time
13:08
14:21
10:54
11:22
10:35
10:19
11:30
11:17
10:30
Stop
Time
15:08
15:47
12:54
15:22
14:21
15:00
15:41
17:02
12:51
Sampling
Duration
120
86
120
240
226
281
251
345
136
Ambient
Temperature
°C
-2.0
-2.8
-2.5
3.5
11.5
1.8
-5.0
10.0
10.0
"F
28
27
28
38
53
35
23
50
50
Wind Speed
m/sec
3.3
2.9
3.5
3.5
1.0
2.5
2.4
2.1
3.3
mph
7.4
6.5
7.8
7.8
2.2
5.6
5.4
4.7
7.4
Number of
Vehicle Passes
Cars/
Vans
1,932
1.952
1,848
2,732
2,419
3,672
3,017
1.936
2.705
Light
Trucks
359
107
120
31
54
32
57
39
270
Heavy
Trucks
336
107
176
0
0
0
17
0
273
Mean Vehicle
Speed
mph
30
30
30
35
35
30
35
20
30
kph
48
48
48
56
56
48
56
32
48
OJ
oo
-------
TABLE 13. SPRING TEST SITE CONDITIONS (St. Louis Area Paved Roads)
Profiler Operation
Run
M-10
M-ll
M-12
M-13
M-14
M-15
M-16
M-17
M-18
M-19
M-20
Site
1-44, Hampton by Sublette
1-44, Hampton by Sublette
1-44, Hampton by Sublette
Kingshighway and Penrose Park
Kingshighway and Penrose Park
Kingshighway and Penrose Park
1-44, Hampton by Sublette
24th and Madison, Granite City
24th and Madison, Granite City
Benton Road and Oregon,
Granite City
Ben ton Road and Hameok,
Granite City
Date
5/7/00
5/8/80
5/B/BO
5/9/00
5/11/80
5/13/80
5/14/80
5/15/80
5/15/80
5/19/80
5/21/80
Start
Time
11:30
9:42
15:27
11:16
9:20
13:46
10:18
12:23
16:19
9:45
10:00
Stop
Time
14:32
12:43
17:57
14:30
12:18
16:01
14:32
14:53
19:11
17:53
10:20
Sampling
Duration
182
181
150
194
178
135
254
150
172
488
Ambient
Temperature
~V
15.5
13.3
18.3
15.5
12.7
24.9
21.1
23.8
23.8
21.1
Test Aborted -
"F
60
56
65
60
55
77
70
75
75
70
light
Wind Speed
in/sec mph
1.3 2.9
3.9 8.7
2.1 4.7
1.2 2.7
4.1 9.2
5.1 11.4
1.8 4.0
1.8 4.0
2.3 5.1
1.2 2.7
and variable
Number of
Vehicle Passes
Cars/
Vans
9.595
9.563
7.751
5,000
3,800
3,900
13.200
3.390
3,670
5,500
winds
Light
Trucks
450
520
850
40
30
30
1.010
-
-
300
Heavy
Trucks
1.095
1.016
1.211
150
110
110
1,220
-
-
-
Mean Vehicle
Speed
mph
55
55
55
35
35
35
55
30
30
30
kph
89
89
89
56
56
56
89
48
48
48
Sampling
Equipment
Config-
uration
A
A
B
A
A
B
A
A
A
A
CO
-------
TABLE 14. ACCEPTABLE TESTS FOR MLR
"'"'Sim Runibor
Criteria 12345670 9 10 11 12 13 14 15 16 17 18 19
. Mean angle of Yes Yes Yes No No Yes Yes Yes Yes No Yes Yes No Yes Yes No Yes Yes Yes
wind direction
and profiler
orientation
less llian 20°.
. Mean angle of Yes Yes Yes No No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
profiler orien-
tation and road-
way less tlian 45°.
. Adequate sam- Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes No Yes Yes No Yes Yes No
pi ing wind
speed conditions,
greater than 4 inpli.
. Acceptable back- Yes Yes Yes No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes
ground particle
concentration re-
lative to downwind
samplers
-£;
0 Results are not Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes No Yes Yes
based on average
emissions data of
other test runs.
Acceptable test for Yes Yes Yes No No Yes Yes Yes Yes No Yes Yes No No Yes No No No No
MLR
-------
6.2 STREET SURFACE PARTICIPATE LOADINGS
During each emissions sampling run and at other times when emissions
sampling was not being conducted, samples of street surface particulate
were collected to determine total particulate loadings and silt percentages.
The silt percentage corresponds to that fraction of the surface sample
< 75 urn in equivalent physical diameter. As shown in Table 15, silt load-
ings on active travel lanes ranged from about 0.022 g/m2 on a freeway (1-44)
to more than 2.5 g/m2 on a lightly traveled rural road in Tonganoxie. As
expected, loadings in curb areas substantially exceeded loadings in travel
lanes. The range of day-to-day variations in loadings at a given site was
generally within a factor of 2. Higher loadings tended to occur after a
precipitation event.
6.3 AIRBORNE PARTICULATE CONCENTRATIONS
Table 16 lists the upwind and downwind particulate mass concentrations
for the various particle size fractions measured during the field program.
These concentration data were collected under a broad range of environmental
conditions, some of which did not meet the QA criteria established for a
valid profiling test (see page 40). The latter data are included in Table 16
because they reflect the air quality impact of the roadway under meteoro-
logical conditions which occur a significant portion of the time. Also
shown in this table is the effective plume height found by extrapolating the
upper net (i.e., due to the source) TP concentrations to a value of zero.
Table 17 provides a summary of the mass fraction ratios. As indicated,
the IP concentration measured downwind of the test road segment was found to
decrease with height. The mean ratio of downwind IP to TSP concentration
(2 m) was 0.45 (a = 0.14), and the corresponding mean upwind ratio was 0.54
(a = 0.18). This indicates that background TSP, although lower in concen-
tration , contains a higher percentage of IP. Similar differences are also
evident in the mean upwind versus downwind < 10 urn to TSP ratios and FP to
TSP ratios.
41
-------
TABLE 15. PAVED ROAD SURFACE DUST LOADINGS
1X3
Test site by land use category
COMMERCIAL/INDUSTRIAL
7th Street
7th Street
7th Street
7th Street
7th Street
7th Street
COMMERCIAL/RESIDENTIAL
Volker Boulevard
Volker Boulevard
Volker Boulevard
Volker Boulevard
Volker Boulevard
Volker Blvd. (2 blks. E of prev. site)
Rockhill Road
Kingshighway and Penrose Park
Kingshighway and Penrose Park
Kingshighway and Penrose Park
24th and Madison, Granite City
24th and Madison, Granite City
Uenton Road and Oregon
20th Street (E. of Steel Plant)
EXPRESSWAY
1-44, Hampton by Sublette
1-44, Hampton by Sublette
1-44, Hampton by Sublette
1-44, Hampton by Sublette
1-44, Hampton by Sublette
RURAL 10WN
Tonganoxie - 4th St. (W of Church St.)
Tonganoxie - 4th St. (W of Church St.)
Tonganoxie - 4th St. (3 blocks W of
previous site)
Tonganoxie - 4th St. (3 blocks W of
previous site)
Tonganoxie - Main Street
Tonganoxie - Main Street
Date
2/11/80
2/12/80
2/12/80
2/22/80
3/03/80
3/10/80
2/13/80
2/26/80
2/27/80
3/03/80
3/05/80
3/05/80
2/28/80
5/09/80
5/11/00
5/13/80
5/15/80
5/15/80
5/19/80
5/21/80
5/07/80
5/08/80
5/08/80
5/14/80
5/21/80
3/16/80
3/17/80
3/06/80
3/17/80
3/06/80
3/17/80
Run
M-l
M-2
M-3
-
-
M-9
_
M-4
M-5
-
-
-
M-6
M-13
M-14
M-15
M-17
M-18
M-19
-
M-10
M-ll
M-12
M-16
-
-
-
M-8
-
-
No. of
traffic
lanes
4
4
4
4
4
4
2
2
2
4
4
2
4
8
8
8
8
B
2
2
2
2
2
2
Average
width
(m)
3.4
3.4
3.4
3.4
3.4
3.4
2.7
2.7
2.7
2.7
2.7
2.7
2.7
3.4
3.4
3.4
3.0
3.0
3.0
2.7
3.6
3.6
3.6
3.6
3.6
3.0
3.0
3.0
3.0
3.0
3.0
lane
(ft)
11
11
11
11
11
11
9
9
9
9
9
9
9
11
11
11
10
10
10
9
12
12
12
12
12
10
10
10
10
10
10
Total
Curb
(gr/ftz) (g/mz)
203 142
a a
a a
5,670 3,160
-
-
a h
2.590 I.BIO"
3,640 2,540°
-
h
1,330 929"
910 635°
41.3 28.
-
— -
~K
560 391,
168 117D
245 171
-
-
-
-
-
-
-
-
-
-
loading
Travel
(gr/ftzi
6.2
6.1
6.0
7.7
4.5
3.4
3.1
3.3
6.7
3.2
3.8
9.8
4.7
8 1.12
-
0.81
20.3
14.8
15.4
20.3
-
-
-
-
0.07
322.0
238.0
24.5
35.7
13.3
14.7
lanes
(9/B)Z)
4.3
4.2
4.2
5.4
3.1
2.4
2.2
2.3
4.7
2.2
2.6
6.8
3.3
0.8
-
0.6
14.2
10.3
10.8
14.2
-
-
-
-
0.05
224.8
166.2
17.1
24.9
9.3
10.3
Silt Content
{%)
Curb Travel lanes
15.7 10.7
a 6.2
a 3.5
8.7 11.9
5.5
12.2
a 5.9
12.3 18.8
13.7 21.4
20.2
22.7
11.3° 20.7
21.2 21.7
8.2 13.7
-
2.6 8.1
5.7
12.5° 7.1
15.5° 8.6
13.2 6.7
-
-
-
-
46.0
9.6
6.9
14.5
7.4
11.0
41.1
Curb area wet.
Average of two samples.
-------
TABLE 16. PARTICIPATE CONCENTRATIONS AND PLUME HEIGHT
Run number
Upwind concentration (|iq/m*)
IP
(2 m)
£10 |im
(2 in)
FP
(2 m)
TSP
(2 m)
IP
(1 m/3 m)
Downwind concentration dig/in*}
£10 pin
(1 m/3 m)
FP
(1 m/3 m)
TSP
(2 m)
IP
(2 m)
Downwind
plume height
(m) '
COMMfRClAt/INDUSTRIAL
M-l
M-2
M-3
M-9
COMMERCIAL/RESIDENTIAL
to
EXPRESSWAY
RURAL TOWN
M-fl
No daTa collected
a
19
12
35
65
M-4
M-5
M-6
M-7
M-13
M-14
M-15
M-17
M-18
M-19
M-10
M-ll
M-12
M-16
37
83
57
31
60
37
59a
65
75
47a
83
71
65a
42
62
54
31
60
60
70
63
38
33
16
39
47
34
40
24
120
46
77
128
65
99
78
42
79
80
161
98
113
72
131
154
198
67
100
146/100
35/33
93/62
207/106
37/32
110/95
69/71
154/70
00/71
34/38
100/13
64/61
92/85
57/50
171/125
124/94
33/31
79/60
169/06
31/27
90/85
63/63
130/61
70/63
20/34
09/10
56/53
82/74
51/42
140/109
68/60
28/23
44/40
77/31
17/11
55/46
41/38
57/26
42/39
15/22
56/6
39/35
51/48
33/31
121/110
99/03
96/79
71/63
104/100
86/75
84/70
65/56
53/50
51/46
55/45
44/37
275
233
228
128
323
322
161
191
59/49
l)
TSP
(2 m)
490
68
BO
400
76
190
127
190
164
116
200
160
187
109
275
233
226
128
IP
(2 m)
368
124
183
437
85
248
138
210
199
184
242
228
230
112
323
322
161
191
Downwind
plume height
(m) '
8K5
106
0.1
0.5
4.6
6.1
7.2
V
I0b
7.1
6.0
6.5
M
10"
4.0
4.5
4.8
5.8
348
307
0.5
Represents the average of 1 and 3 meter measurements.
Assumed value.
-------
TABLE 17. SUMMARY OF PARTICULATE SIZE RATIOS
Statistical
parameter
X
o
RSI)
t)
JP/TSP
0.54
O.lb
0.2fl
19
<.10 pm/TSP
0.54
0.11
0.20
7
Upwind
FP/1SP
0.33
0.090
0.27
7
<10 ,.n,/IP
0.88
0.034
0.039
7
FP/TP
0.53
0.08'j
0.16
7
IP/TSP
0.45
0.14
0.3J
19
<10 iMU/TSP
0.40
0.13
0.32
19
Downwind
FP/TSP
0.24
0. 092
0.38
19
<10 pin/IP
0.87
0.028
0.032
19
FP/IP
0.52
0.098
0.19
19
-------
The FP to IP mean ratio measured downwind was 0.52 (cr = 0.098) while
the mean upwind ratio was 0.53 (a = 0.085). This finding implies that
there is no significant enrichment of fine particles attributable to the
paved road source.
6.4 EMISSION FACTORS
Tables 18 and 19 present for each test the calculated emission factors
(IP, < 10 |jm, and FP) and corresponding source characterization parameters
which are thought to affect the intensity of emissions from paved roads.
Appendix A describes the procedures used to calculate the emission factors
from field testing data.
Tables 20 and 21 summarize, by land use category and test series qual-
ity, the emission factor and associated parameter data. As can be seen,
the smallest emission factors were measured in the freeway category which
also had the lowest surface silt loadings. The highest emission factor was
measured in the rural town category which showed a correspondingly high
surface silt loading.
Intercomparison of emission factors by land-use category indicates
that relative to the mean expressway IP emissions: (a) mean commercial/
residential IP emissions were approximately 10 times larger; (b) commercial/
industrial emissions were approximately 20 times larger; and (c) the rural
town roadway produced IP emissions that were roughly 60 times larger. Rela-
tive to mean expressway silt loading: (a) the silt loading for commercial/
residential roadways was approximately 25 times higher; (b) the silt loading
for commercial/industrial roadways was roughly 15 times higher; and (c) silt
loading on the rural town roadway was approximately 115 times higher.
45
-------
TABLE 18. PAVED ROAD EMISSION FACTORS9
cn
Inhalable particulate < 10 pin particulate Fine particulate
emission factor , emission factor , emission factor ,
Run No.
Site
(g/VKT)
(Ib/VMT x 101)" (g/VKT)
(Ib/VMT x 10T (g/VKT)
(Ib/VMf x 10')
COMMERCIAL/INDUSTRIAL
M-l
M-2
M-3
M-9
7th Street
7th Street
7th Street
7th Street
3.52
1.01
2.39
2.80
125.0
35.7
84.8
99.3
3.10
0.96
2.20
2.01
110.0
34.0
78.1
71.2
1.78
0.86
1.54
1.14
63.2
30.4
52.0
40.5
COMMERCIAL/RESIDENTIAL
EXPRESSWAY
RURAL TOWN
M-4
M-5
M-6
M-7
M-13
M-14
M-15
M-17
M-18
M-19
M-10
M-ll
M-12
M-16
M-8
Volker Boulevard
Volker Boulevard
Rockhill Road
Volker Boulevard
Kingshighway and
Penrose Park
Kingshighway and
Penrose Park
Kingshighway and
Penrose Park
Granite City. 24th
and Madison
Granite City. 24th
and Madison
Granite City, Benton
Road and Oregon
1-44, Hampton .by
Sublette
1-44, Hampton by
Sublette
1-44, Hampton by
Sublette
1-44, Hampton by
Sublette
longanoxie, 4th Street
0.19
0.47
0.93
3.30
0.22
0.93
1.01
1.92
0.32
0.16
0.12
0.22
0.059
0.16
8.77
6.7
16.8
32.9
117.0
7.8
33.0
35.8
68.0
11.4
5.8
4.1
7.8
2.1
5.8
311.0
0.11
0.43
0.86
2.62
0.19
0.85
0.91
1.64
0.23
0.11
0.11
0.20
0.052
0.15
6.96
4.0
15.3
30.4
92.8
6.8
30.1
32.3
58.2
8.0
3.9
3.9
7.0
1.9
5.3
247.0
c
0.16
0.59
1.04
0.11
0.62
0.62
c
0.052
d
0.063
0.097
0.039
d
1.42
c
5.6
20.9
36.8
4.0
22.0
22.0
c
1.9
d
2.3
3.4
1.4
d
50.4
Emission factors for < 10 and FP require use of interpolated concentrations.
To convert these factors to Ib/VMT multiply by 10 4.
Downwind and upwind concentrations were equal.
Data
preclude the
ination of an emission factor.
-------
TABLE 19. SOURCE CHARACTERIZATION PARAMETERS
Sill loading
Hun number
COMMERCIAL/INIHISIKIAL
M-l
M-2
M-3
M-9
COMMERCIAL/RESIDENTIAL
M-4
M-5
M-6
M-7
M-13
M-14
M-15
M-17
M-18
M-19
EXPRESSWAY
M-10
M-ll
M-12
M-16
Site
7th Street
7th Street
7th Street
7lh Street
Volker Boulevard
Volker Boulevard
Rockhill Road
Volker Boulevard
Kingshighway and
Penrose Park
Kingshighway and
Penrose Park
Kingshighway and
Penrose Park
Granite City. 24th
and Madison
Granite City. 24lh
and Madison
Granite City, Benton
Road and Oregon
1-44, Hampton by
Sublette
1-44, Hampton by
Sublette
1-44, Hampton by
Sublette
1-44, Hampton by
Sublelte
fiTi T
0.46
0.26
0.15
0.29
0.43
1.00
0.68
0.59
0.11
0.079
0.047
0.83
0.73
0.93
0.022
0.022
0.022
0.022
(gr/ft*)
0.66
0.37
0.21
1.41
0.61
1.43
0.97
0.84
0.16
0.11
0.067
1.19
1.04
1.33
0.03]
0.031
0.031
0.031
Moan vehicle speed
(W)
48
40
48
48
56
56
48
56
56
56
56
48
48
48
89
89
89
89
(mph)
30
30
30
30
35
35
30
35
35
35
35
30
30
30
55
55
55
55
Mean vehicle weight
(Mg)
5.1
3.4
4.1
3.7
1.9
2.0
1.9
2.1
2.4
2.4
2.4
1.8
1.8
2.2
4.1
4.4
3.4
3.9
(tons)
5.6
3.8
4.5
4.1
2.1
2.2
2.1
2.3
2.7
2.7
2.7
2.0
2.0
2.4
4.5
4.8
3.8
4.3
RURAL TOWN
M-0
Tonganoxie, 4th St.
2.50
3.58
32
20
2.0
2.2
-------
TABLE 20. SUMMARY OF PAVED ROAD EMISSION FACTORS
Land use
category
Commercial/
Industrial
Commercial/
Residential
Lxpresswoy
Rural Town
Test
series
quality
All tests
MLR tests
All tests
MLR tests
All tests
MIR tests
All tests
MLR tests
Test numbers
M-l, 2. 3, 9
M-l, 2, 3, 9
M-4, 5, 6. 7, 13, 14,
15, 17. 18, 19
M-6, 7. 15
M-10. 11, 12, 16
M-ll, 12
M-8
M-8
IP
x
(fl/VKl)
2.43
2.43
0.94
1 75
0.14
0.14
8.77
8.77
emission factor
(9/VKT)
1.06
1.06
0.99
1.35
0.069
0.12
-
RSI)/ROa
0.44
0.44
1.05
0.77
0.50
1.16
-
SJO |iin emission factor
x
(gMi)
2.07
2.07
0.80
1.46
0.13
0.12
6.96
6.96
i)
(g/VKI)
0.88
0.99
0.80
1.00
0.062
0.10
-
RSO/RO3
0.42
0.42
1.01
0.68
0.48
1.16
-
f-P emission factor
x
(g/VKl)
1.31
1.31
0.46
0.75
0.66
0.068
1.42
1.42
a
(g/VKf)
0.40
0.40
0.36
0.25
0.029
0.041
-
HSI)/KI)a
0.30
0.30
0.79
0.34
0.44
0.85
-
USD (relative standard deviation) calculated when more than two data points are present.
Ix - xj
RO
deviation) =
-*- calculated when two data points are present.
-------
TABLE 21. SUMMARY OF SOURCE CHARACTERIZATION PARAMETERS
,„.., Silt loading
Land use
category
Commercial/
Industrial
Commercial/
Residential
Expressway
Rura 1 Town
series
qual ily
All runs
MLR runs
Al 1 runs
MLR runs
All runs
MLR runs
A 1 1 runs
MLR runs
Test numbers
M-l. 2, 3, 9
M-l. 2. 3. 9
M-4. 5. 6. 7. 13.
15. 17. 18. 19
M-6. 7, 15
M-10. 11, 12, 16
M-ll, 12
M'8
M-8
x o
(a/w2) (9/m2)
0.29 0.13
0.29 0.13
14. 0.51 0.36
0.14 0.34
0.022
0.022
2.5
2.5
x
RSD/RO3 (kph)
0.45 48
0.45 48
0.67 53
0.77 53
89
89
32
32
Vehicle speed Vehicle weight
0 a x
(kph) RSD/KIT (Mfl)
4.1
4.1
4.2 0.079 2.1
4.6 0.087 2.1
4.0
3.9
2.0
2.0
u
(Mfl)
0.75
0.75
0.25
0.25
0.42
0.71
-
RSI)/RDa
-
0.12
0. 12
0.26
:
KSD (relative standard deviation) calculated when more than two data points are present.
£Jx - x I
leviation) = -' A
RO (relative deviation)
calculated when two data points are present.
-------
(This page intentionally blank)
50
-------
7.0 MULTIPLE REGRESSION ANALYSIS
7.1 INTRODUCTION
Stepwise Multiple Linear Regression (MLR) was the method used to evalu-
ate independent variables for possible use as correction factors in a pre-
dictive emission factor equation. It is available as a computer program in
the Statistical Package for the Social Sciences (SPSS).17 The MLR program
outputs of interest in evaluating the data sets for the paved road source
tests are the multiple regression coefficient, significance of the variable,
and reduction in relative standard deviation due to each variable. Further
information on MLR can be found elsewhere.16 18
It is desirable to have correction factors in the emission factor equa-
tions multiplicative rather than additive; consequently all independent and
dependent variable data are transformed to natural logarithms before being
entered in the MLR program.
The stepwise regression program: (a) selects the potential correction
factor that is the best predictor of IP emission factors; (b) changes the
dependent variable values to reflect the impact of this independent vari-
able; and, (c) repeats this process with remaining potential correction fac-
tors until all have been used in the MLR equation or until no improvement
in the predictive equation is obtained by adding another variable. Not all
variables included in the MLR equation are necessarily selected as correc-
tion factors.
51
-------
The steps followed in developing correction factors are listed below:
1. Create a data array of all monitored independent variables with
corresponding emissions measurements.
2. Input these data into the MLR program using a COMPUTE statement
to transform both independent and dependent variables to their
natural logarithms.
3. From the summary statistics, find variables that have a signifi-
cance less than 0.05. These are definite correction factors.
4. Next, evaluate those variables with a significance of 0.05 to
0.20. If any of these variables are judged to be pertinent in-
dependent variables they may also be included as correction
factors.
5. Determine the form of the emission factor equation, exclusive of
the coefficient (base emission factor).
6. Assume typical values for the correction parameters.
7. Calculate adjusted emission factors at the average conditions for
all the correction parameters, using the relationships established
in the emission factor equation.
8. Determine the geometric mean for the adjusted data set. This mean
is the base emission factor or coefficient in the emission factor
equation.
9. Finalize the emission factor equation as the base emission factor
times each correction parameter normalized to average conditions.
10. Determine the precision factor for the emission factor equation.
52
-------
7.2 ANALYSIS AND RESULTS
The independent variables evaluated initially as possible correction
factors were silt loading (g/m2), total loading (g/m2), average vehicle
speed, (Kph), and average vehicle weight (Mg). The rationale for includ-
ing measures of roadway particulate loading stems from findings of an
earlier MRI program3 which indicated that the magnitude of roadway emis-
sions was directly related to variations in surface loadings. The vehi-
cle parameters—mean weight and speed—were included largely by analogy to
MRI's unpaved road equation,19 although it was recognized that the dust gen-
eration mechanism for paved roads may differ from that for unpaved roads.
The moisture content of the road surface particulate was not included as a
correction parameter because of the difficulty of collecting a sample with-
out altering its moisture content.
The correlation matrix associated with a preliminary MLR analysis of
the entire data set is shown in Table 22. Examination of the matrix indi-
cated that all the independent variables except vehicle weight were highly
intercorrelated. Although the stepwise algorithm would include vehicle
speed first in a predictive equation, silt loading and total loading show
essentially the same correlation with IP emissions (r = 0.60). In other
words, the variables represent a common set of source conditions—either
low vehicle speed, high surface loadings and emissions or high vehicle
speed, low loadings and emissions.
The decision was made to use silt loading rather than total loading or
vehicle speed in the development of the emission factor equation from the
"MLR" data set. This decision was based on the perception that (a) silt
loading is the most physically plausible indicator of the magnitude of IP
emissions, and (b) it will yield more reproducible results in independent
applications than total loading, a parameter which can be biased by the
presence of macro size particles (i.e., gravel).
53
-------
TABLE 22. CORRELATION MATRIX FOR ENTIRE DATA SET
(n = 19)
eIP
IP Emission factor 1.0
C.IP)
Silt loading
Total loading
Vehicle speed
Vehicle weight
Silt
loading
0.56
1.0
Total
loading
0.63
0.94
1.0
Vehicle
speed
-0.74
-0.86
-0.94
1.0
Vehicle
weight
0.02
-0.62
-0.56
0.48
1.0
The correlation matrix associated with the "MLR" data set is presented
in Table 23. Including silt loading as the primary predictor effectively
precludes total loading or vehicle speed from entering the equation. This
follows from the high intercorrelations (multicollinearity) mentioned above.
Examination of the regression statistics indicated that inclusion of vehicle
weight as a second correction parameter could not be justified.
TABLE 23. CORRELATION MATRIX FOR "MLR" DATA SET
(n = 10)
SIP
IP Emission factor
(elp) 1.0
Silt loading
Total loading
Vehicle speed
Vehicle weight
Silt
loading
0.85
1.0
Total
loading
0.91
0.92
1.0
Vehicle
speed
-0.89
-0.89
-0.97
1.0
Vehicle
weight
-0.08
-0.46
-0.31
0.37
1.0
54
-------
The raw MLR equation for the "MLR" data set , as output from the SPSS
package is as follows:
elp = 4.37 (sL)°-8 (1)
where:
6jp = IP emission factor expressed in grams per vehicle
kilometer traveled (g/VKT)
sL = Silt loading of road surface part icu late matter ex-
pressed in grams per square meter (g/m2).
This equation explains 73% of the variation in the emission factors. As
noted earlier, the "MLR" data set does contain data from all the land use
categories sampled during the field program.
Equation 2 presents the comparable predictive IP emission factor equa
tion normalized to a typical value for silt loading:
0
The normalization procedure consists of steps 6 through 10 as outlined in
Section 7.1 (p. 52).
Table 24 presents the predicted versus measured IP emission factors,
and provides a comparative statistic—the ratio of predicted to measured
emission factors for each test. The same information is presented graphi-
cally by land use category in Figure 7. As can be seen, there is consider-
able variation between predicted and measured emission factors, both overall
and within individual categories. The only discernible, predictive bias ap-
pears in the commercial/industrial subset where the tendency appears to be
for the emission factor equation to underpredict observed emissions.
55
-------
TABLE 24. PREDICTED VERSUS OBSERVED IP EMISSION FACTORS
Land use
category
IP Emission factor
(g/VKT)
Observed
Predicted
Ratio'
Commercial/
industrial
Commercial/
residential
Expressway
Rural town
M-l
M-2
M-3
M-9
M-6
M-7
M-15
M-ll
M-12
M-8
3.52
1.01
2.39
2.80
0.928
3.30
1.01
0.222
0.0589
8.77
2.37
1.51
0.970
1.64
3.25
2.90
0.384
0.209
0.209
9.20
0.67
1.50
0.41
0.58
3.50
0.88
0.38
0.94
3.55
1.05
Predicted divided by observed.
56
-------
10
u
a
E
UJ
a.
•o
4)
.3
o
0.1
0.01
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
/
.
/
/
1 1 1 1 1 1 1 1 1
/:
° /
/
/
/ o
Land Use Category
O Commercial/Industrial
C3 Commercial/Residential
A Expressway
• Rural town
0.01
0.1 1
Predicted IP Emission Factor (g/VKT)
10
Figure 7. Predicted versus observed IP emission factors by land use category.
57
-------
This tendency may be the result of a combination of the high percentage of
heavy-duty vehicles (~ 20%) coupled with vehicle idle, acceleration, and
deceleration typically associated with proximity to traffic lights. The
latter condition normally produces a significant increase in the exhaust
emissions component, which would not be incorporated in the silt loading
model.
7.3 COMPARATIVE EVALUATION
The emission factor equation predicts the "MLR" series test data with
a precision factor of 2.0. The precision factor (f) for an emission factor
is defined such that the 68% confidence interval for a predicted value (P)
extends from P/f to Pf. The precision factor is determined by exponentiat-
ing the standard deviation of the differences (standard error of the esti-
mate) between the natural logarithms of the predicted and observed emission
factors.
The precision factor may be interpreted as a measure of "average" error
in predicting IP emissions from the regression equation. Assuming that the
actual IP emission factors are log normally distributed about the regression
line, it can be stated that approximately 68% of the predictions are within
a factor of 2. The effective outer bounds of predictability are determined
by exponentiating twice the standard error of the estimate. The resultant
estimate of predictive accuracy, in this case 4.0, then encompasses approxi-
mately 95% of the predictions.
To put the precision factor of the IP predictive emission factor equa-
tion emission factor into perspective, two comparisons were undertaken
utilizing the single-value emission factor found in the current AP-42 man-
ual.3 However, before valid comparisons could be made, it was necessary to
convert the AP-42 single value factor which represents TSP emissions, to an
approximate IP emission factor.
58
-------
This was accomplished by multiplying the AP-42 value by 0.4 which is the
mean ratio of net IP (downwind minus upwind) to net TSP concentrations as
determined from the data collected in this study. This ratio may be ex-
pressed as follows:
" IPUp)
Because this ratio reflects net emissions, that is, the emissions directly
attributable to the source, it is preferable to one based on sizing infor-
mation given in AP-42 which describes emissions due to both source and back-
ground. As noted in AP-42, the latter information will be biased toward
small particle sizes.
The first comparison involved the calculation of a precision factor for
the AP-42 data set. The resulting value of 2.1 is a measure of the ability
of the single-value factor to represent the 40 pieces of data which were
averaged originally to produce the AP-42 factor. The second comparison in-
volved the calculation of a precision factor using the single value AP-42
factor to represent the "MLR" data set, as collected in this study. This
comparison yielded a precision factor of 4.4.
The precision factors and the range of the data values (emission fac-
tors) upon which they are based, are presented graphically in Figure 8.
The ideal model has a precision factor of 1.0, implying that each predicted
value is identical to the corresponding observed valuer over an infinite
range of emission factors. The most important conclusion that can be drawn
from Figure 8 is that the emission factor equation, though far from ideal,
does predict IP emissions more accurately over a much greater range of val-
ues than does the AP-42 single-value factor over a considerably smaller
range of data values corresponding to the AP-42 data set. Furthermore, ap-
plication of the single-value AP-42 factor to represent the wide range of
IP emissions from paved roads as measured during this program, yields a
59
-------
5p-
u
o
AP-42 Applied to Present Data
c
o
Present Regression Equation
r AP-42 Emission Factor
Ideal Model
I I I I I II I I I I I I I I I I
0.01 0.1 1 10
Emission Factor ~(g/VKT)
Figure 8. Comparison of emission factor models.
60
-------
precision factor which is more than double (4.4 versus 2.0) that associated
with the predictive equation. This ability of the predictive equation to
more accurately represent variations in IP emissions is directly attribut-
able to the relatively strong relationship between roadway surface silt
loading and IP emissions.
7.4 EXTENSION OF THE PREDICTIVE EQUATION TO DIFFERENT PARTICLE SIZE
FRACTIONS
The particle sizing data obtained from the SSI/CI combinations was also
used to develop emission factors and predictive emission factor equations for
the ^ 10 urn and FP particle size fractions. These analyses used the same pro-
cedure as that applied in developing the equation for IP (see Section 7.1).
Derivation of TSP emission factors for use in developing a predictive equation
required different initial calculations, since only two TSP samplers (one up-
wind, one downwind) were operated during the measurement phase of the program.
In essence, the initial calculation involved multiplication of the IP emission
factor for each run in the "MLR" series data set by the corresponding net ratio
of TSP to IP concentration as measured by appropriate samplers (see Figures 4
to 6). This procedure assumes that the TSP/IP ratio is constant over the ver-
tical extent of the plume.
The general form of the emission factor equation applicable to all
particle size fractions, is as follows:
p
e = k (^) (metric) (3)
p
e = K (f) (English) (4)
The base emission factor coefficients (k, K) , exponent (P), and precision
factor for each size fraction are listed in Table 25. For the metric equa
tion, silt loading is expressed as grams per square meter; silt loading
for the English equation is expressed as grains per square foot.
61
-------
TABLE 25. PAVED ROAD EMISSION FACTOR EQUATION PARAMETERS
(by particle size fraction)
Particle size fraction
TSP
IP
^ 10 |jm
FP
k (g/VKT)
5.87
2.54
2.28
1.02
K (Ib/VMT)
0.0208
0.0090
0.0081
0.0036
P
0.9
0.8
0.8
0.6
Precision factor3
2.4
2.0
2.2
2.2
a Represents the interval encompassing 68% of the predicted values.
It should be noted that the tendency for the power term in the equation
to increase with larger particle size fraction is generally consistent with
MRI's previous paved road equation in which silt loading to the 1.0 power
was employed to account for variations in TSP emissions.
7.5 EMISSIONS INVENTORY APPLICATIONS
For the majority of emissions inventory applications involving urban
paved roads, actual measurements of silt loading will probably not be made.
Therefore, in order to facilitate the use of the previously described equa-
tions, it is necessary to characterize silt loadings according to a param-
eter(s) more readily available to persons developing emissions inventories.
After examination and analysis of silt loading and traffic data collected
during relevant MRI sampling programs, as well as surface loading data
gathered in connection with an extensive study of urban water pollution,
the decision was made to characterize variations in silt loading based
upon a roadway classification system. The roadway classification system
developed by MRI for this purpose is presented in Table 26.
62
-------
TABLE 26. PAVED ROADWAY CLASSIFICATION
Average daily traffic
Roadway type (ADT) No. of lanes
Freeway/expressway
Major street/highway
Collector street
Local street
> 50,000
> 10,000
500-10,000
< 500
> 4
> 4
2a
2b
a Total roadway width g 32 ft.
Total roadway width < 32 ft.
This system generally corresponds to the functional classification systems
employed by transportation agency personnel; and thus the data necessary
for emissions inventory—number of road miles per road category and traffic
counts—should be easily estimated. It should be noted that in some situa-
tions it may be necessary to combine this silt loading information with
sound engineering judgment in order to approximate the loadings for roadway
types not specifically included in Table 26.
It should be recalled from Section 2.0 that traffic volume is not the
only factor affecting roadway silt loadings. For all roadways that provide
access to immediately adjacent areas, land use, particularly as it relates
to the potential for mud and dirt "tracking," is important. Silt loadings
may also be affected by street surface type and condition, the presence or
absence of curb, as well as public works practices and season of the year.
However, given the present data base, it is not possible to incorporate
relationships between these factors and silt loadings in a manner applicable
to the majority of emissions inventories.
The data base made up of 44 samples collected and analyzed according
to the procedures outlined in Sections 4.2 and 5.4 may be used to character-
ize the silt loadings for each roadway category. These samples, obtained
63
-------
during MRI field sampling programs over the past 3 years, represent a broad
range of urban land use and roadway conditions. Geometric means for this
data set are broken out by sampling location (i.e., city) and roadway cate-
gory in Table 27
TABLE 27. SUMMARY OF TYPICAL SILT LOADINGS (g/m2) FOR URBAN PAVED
ROADWAYS3 BY CITY
Roadway
Local Collector
City Xgb
Baltimore1" 1.42
Buffalod 1.41
Granite City (111.)8
Kansas City6
St. Louis6
n Xg
2 0.72
5 0.29
.
2.11
- -
n
4
2
-
4
-
category
Major
Xg
0.39
0.24
0.82
0.41
0.16
n
3
4
3
13
3
Overal 1
xg
0.68
0.56
0.82
0.60
0.16
n
9
11
3
17
3
a Freeway/expressway loading measurement (0.022 g/m2) from Table 19 not
included.
Xg's are geometric means based on the corresponding n sample size.
c Reference 20.
Reference 21.
6 From this report.
The sampling locations can be considered representative of most large
urban areas in the United States with the possible exception of those lo-
cated in the Southwest. Except for the collector roadway category, the over-
all mean silt loadings do not vary greatly from city to city, though the
St. Louis mean for major roads is somewhat lower than the other four cities.
The substantial variation within the collector roadway category is probably
attributable to the deposition effects of land use associated with the spe-
cific sampling locations. It should also be noted that an examination of
data collected at three cites in Montana during early spring, indicates that
winter road sanding may produce loadings five to six times higher than the
means of the loadings given in Table 27 for the respective road categories.22
64
-------
Typical silt loadings by roadway category (from Table 27) are as follows:
Local streets 1.41 g/m2
Collector streets 0.92 g/m2
Major streets and highways 0.36 g/m2
Expressways/freeways 0.022 g/m2
It should be noted that regression analysis indicates a significant (a = 0.01)
relationship between silt loading and traffic volume of the following form.
sL = 21.3 (ADT)
-0.41
This equation explains 35% of the sample variation.
Table 28 presents the emission factors broken out by roadway category
and particle size. These were obtained by inserting the typical silt loadings
of each roadway category into the emission factor equations found in Section
7.4, Table 25. These emission factors can be utilized directly for emission
inventory purposes. It is important to note that the current AP-42 paved road
emission factors3 for TSP agree quite well with those developed in this study.
For example, those cited in connection with MRI's previous testing2 were con-
ducted at two roadway sites in the major street and highway category. Those
tests yielded a mean TSP emission factor of 4.3 g/VKT versus 4.4 g/VKT as
determined from the data presented here.
TABLE 28. RECOMMENDED EMISSION FACTORS FOR SPECIFIC ROADWAY CATEGORIES
AND PARTICLE SIZE FRACTIONS
Emission factor by parti cl
Roadway
category
Local
Collector
Major street
g/VKT
15
10
4.4
TSP
Ib/VMT
0.053
0.035
0.016
^ 15
g/VKT
5.8
4.1
2.0
urn
1 b/VMT
0.021
0.015
0.0071
^
g/VKT
5.2
3.7
1.8
e size fraction
10 urn
1 b/VMT
0.018
0.013
0.0064
^ 2.
g/VKT
1.9
1.5
0.84
5 |jm
1 b/VMT
0.0067
0.0053
0.0030
and highway
Freeway/ 0.35
Expressway
0.0012 0.21 0.00074 0.19 0.00067 0.16 O.OOOS7
65
-------
(This page intentionally blank)
66
-------
8.0 SUMMARY AND CONCLUSIONS
The purpose of this study was to quantify inhalable particulate emis-
sions generated by traffic entrainment of paved road surface particulate
matter. Paved road source testing was performed at sites representative
of significant emission sources within a broad range of urban land-use
categories.
The measured inhalable particulate emission factors ranged from 0.06
to 8.77 g/VKT. Lowest mean emissions were measured for the "Expressway"
use category; highest mean emissions were measured for the "Rural Town1' use
category. Approximately 90% of the IP emissions consisted of particles
smaller than 10 urn in aerodynamic diameter, and approximately 50% of the IP
emission consisted of particles smaller than 2.5 |jm in aerodynamic diameter
Correlation analysis of IP emissions with parameters characterizing
the source conditions showed the existence of a relatively strong positive
relationship between intensity of emissions and roadway surface silt load-
ing. This confirms the findings of earlier testing..2 Based on regression
analysis of a subset of acceptable ("MLR") test runs, the following predic-
tive IP emission factor equation was developed:
o- s
elp=2.54 (gtj) (5)
where e,p = Inhalable particulate emission factor (g/VKT).
sL = Road surface silt loading (g/m2).
67
-------
This predictive equation has an associated precision factor of 2.0 in
relation to the "MLR" data set. By way of comparison, the AP-42 single-
value factor (corrected to represent IP emissions) has a precision factor
of 2.1 for its data set and a precision factor of 4.4 for the "MLR" data
set, which spans a much larger range of values than the AP-42 data set.
Therefore the predictive equation, though far from ideal, does represent
IP emissions more accurately over a much larger range of values than does
the AP-42 single-value factor. This fact is directly attributable to the
relationship of IP emissions to silt loading.
Extension of the regression analysis to include emission factor equa-
tions for other particle size fractions—FP, < 10 urn, and TSP—yielded a
set of equations in which the power term for silt loading increased with
larger particle size fraction. This result is generally consistent with
MRI's previous paved road equation in which silt loading to the 1.0 power
was employed to account for variations in TSP emissions.
To facilitate the use of these particle size specific equations in the
development of emission inventories, a classification system of mean or
typical silt loadings as a function of roadway category was derived. These
mean silt loadings were then inserted into the respective emission factor
equations. The resultant emission factors for specific roadway category
and particle size fractions can be utilized directly for emissions inven-
tory purposes. By accounting for variations in silt loading, these emis-
sion factors are significantly more reliable than an overall average emis-
sion factor in developing components of an urban paved road emission
inventory.
68
-------
9.0 REFERENCES
1. Lynn, D. L. , G. Deane, R. Galkiewicz, R. M. Bradway, and F. Record.
National Assessment of Urban Particulate Problem. Volume I - Summary
of National Assessment. U.S. Environmental Protection Agency, Publi-
cation No. EPA 450/3-76-024, July 1976.
2. Cowherd, C. , Jr., C. M. Maxwell, and D. W. Nelson. Quantification of
Oust Entrainment From Paved Roadways. U.S. Environmental Protection
Agency, Publication No. EPA-450/3-77-027, July 1977.
3. Compilation of Air Pollutant Emission Factors, Third Edition, U.S.
Environmental Protection Agency, Publication No. AP-42, August 1977.
4. Sartor, J. D., and G. B. Boyd. Water Pollution Aspects of Street Sur-
face Contaminants. U.S. Environmental Protection Agency, Publication
No. EPA-R2-72-081, November 1972.
5. Abel, M. P. The Impact of Refloatation on Chicago's Total Suspended
Particulate Levels. Master's Thesis, Purdue University, August 1974.
6. Cowherd, C., Jr., K. Axetell, Jr., C. M. Guenther, and G. Jutze.
Development of Emission Factors for Fugitive Dust Sources. Final
Report, Midwest Research Institute for U.S. Environmental Protection
Agency, Publication No. EPA-450/3-74-037, NTIS No. PB 238262/AS,
June 1974.
69
-------
7- Roberts, J. W., A. T. Rossano, P. T. Bosserman, G. C. Hofer, and H. A.
Watters. The Measurement, Cost and Control of Traffic Dust and Gravel
Roads in Seattle's Duwamish Valley. Paper No. AP-72-5, Presented at
the Annual Meeting of the Pacific Northwest International Section of
the Air Pollution Control Association, Eugene, Oregon, November 1972.
8. Axetell, K., and J. Zell. Control of Reentrained Dust from Paved
Streets. EPA Publication No. EPA-907/9-77-007, August 1977.
9. Roberts, J. W., H. A. Watters, C. A. Margold, and A. T. Rossano. Cost
and Benefits of Road Dust Control in Seattle's Industrial Valley.
Paper No. 74-83, Presented at the 67th Annual Meeting of the Air Pol-
lution Control Association, Denver, Colorado, June 9 to 13, 1974.
10. American Public Works Association. Water Pollution Aspects of Urban
Runoff, APWA, Chicago, 1969. pp. 171-175.
11. Hanna, T. R., and T. M. Gilmore. Applicability of the Mass Concentra-
tion Standards for Particulate Matter in Alaskan Areas. Alaska Depart-
ment of Environmental Conservation, Juneau, Alaska, 1973.
12. Shaheen, D. G. Contribution of Urban Roadway Usage to Water Pollution.
U.S. Environmental Protection Agency, Publication No. EPA-600/2-75-004,
March 1975.
13. Transportation and Traffic Engineering Handbook. Institute of Traffic
Engineers. Prentice-Hall, Inc., London, 1976. pp. 162-163.
14. Quality Assurance Handbook for Air Pollution Measurement Systems.
Volume II - Ambient Air Specific Methods. U.S. Environmental Protec-
tion Agency, Publication No. EPA 600/4-77-027a, May 1977.
15. Ambient Monitoring Guidelines for Prevention of Significant Deteri-
oration. U.S. Environmental Protection Agency, Publication No. EPA
450/2-78-019, May 1978.
70
-------
16. Draper, N. R. and H. Smith. Applied Regression Analysis. John Wiley
and Sones, New York, 1965.
17. Nie, N. H., et al. Statistical Package for the Social Sciences, Sec-
ond Edition. McGraw-Hill, Inc., New York, 1975.
18. Snedecor, G. W. Statistical Methods. Fourth Edition. The Iowa State
College Press, Ames, Iowa, 1946.
19. Cowherd, C. , Jr., R. Bohn, and T. Cuscino, Jr. Iron and Steel Plant
Open Source Fugitive Emission Evaluation. Final Report, Midwest
Research Institute for U.S. Environmental Protection Agency, Publica-
tion No. EPA-600/2-79-103, May 1979.
20. Cuscino, T. , Jr. Total Suspended Particulate Matter Analysis in
Baltimore, Maryland. State of Maryland, Baltimore, Maryland, October
1981.
21. Bohn, R. Evaluation of Open Dust Sources in the Vicinity of Buffalo,
New York. EPA Contract No. 68-02-2545, Assignment 1, Environmental
Protection Agency, New York, New York, March 1979.
22. Bohn, R. Update and Improvement of the Emission Inventory for MAPS
Study Areas. State of Montana, Helena, Montana, August 1979.
71
-------
APPENDIX A
EMISSION FACTOR CALCULATION PROCEDURES
A-l
-------
INTRODUCTION
This appendix describes the calculation of participate emission fac-
tors from exposure profiling data. The example calculation presented here
is based on actual data obtained from an exposure profiling test (M-3) per-
formed at the 7th Street site in Kansas City, Kansas on February 12, 1980.
The following definitions for particulate matter will be used in this
appendix:
TP Total airborne particulate matter.
IP Inhalable particulate matter consisting of particles smaller than
15 urn in aerodynamic diameter.
PARTICULATE CONCENTRATION
The concentration of airborne particulate matter measured by an air
sampler is given by
C = 103 _|
Where C = particulate concentration (ug/m3)
m = particulate sample weight (mg)
Q = sampler flow rate (m3/hr)
t = duration of sampling (hr)
The specific particulate matter concentrations from the various par-
ticulate catches are as follows:
Size range Particulate catches
TP Profiler filter and intake catches
IP Size Selective Inlet (SSI) filter and
impactor substrate catches
A-2
-------
The measured IP concentrations for the sample test are found in Table A-l.
TABLE A-l. INHALABLE PARTICULATE CONCENTRATIONS FOR RUN M-3
Height
(m)
1.0
2.0
3.0
4.0
Location
Qownwi nd
Upwind
Downwi nd
Upwi nd
Total
sample mass
(nig)
12.75
5.25
8.45
4.45
Sample
flow rate
(nrVhr)
68.0
68.0
68.0
68.0
Sampling
duration
(min)
120
130
120
130
Measured IP
concentration
(ug/m3)
93.8
35.6
62.1
30.2
To be consistent with the National Ambient Air Quality Standard for TSP,
all concentrations are adjusted to standard conditions (25°C and 760 mm of
Hg).
ISOKINETIC FLOW RATIO
The isokinetic flow ratio (IFR) is defined only for a directional sam
pler. It is the ratio of intake air speed to the mean wind speed approach
ing the sampler. It is given by
(2)
where Q = sampler flow rate (m3/hr)
a = intake area of sampler (m2)
U = mean wind speed at height of sampler (m/hr)
This ratio is of interest in the sampling of TP, since isokinetic sampling
assures that particles of all sizes are sampled without bias. For Run M-3,
the profiler IFRs at 1.0, 2.0, 3.0, and 4.0 m heights were 0.98, 0.96, 0.96,
A-3
-------
and 0.96, respectively. The profiler was the only directional sampler used
in this study.
PARTICLE SIZE DISTRIBUTIONS
The particle size distribution at a given height is determined using
concentration measurements from the profiler head (or Hi-Vol for upwind dis-
tributions) and the SSI/cascade impactor at the same height and at the same
distance from the source. The determination of concentrations corresponding
to particulate fractions < 10 urn and < 2.5 urn requires an interpolation of
the particle size-mass distribution. In this study, a spline fit of the
natural logarithms of the SSI/cascade impactor data was used to determine
these concentrations. The downwind particle size data for Run M-3 are
plotted on log-probability paper in Figures A-l and A-2.
NET IP EXPOSURES
The upwind IP concentrations from Table A-l are averaged to produce a
representative upwind (uniform) concentration. This value is subtracted
from the downwind concentrations at each height to obtain net IP concentra-
tions (i.e., due to vehicular traffic on the road). The net concentrations
are used to produce net exposure values at each downwind sampling height by
the expression.
E = 10~7C U t (3)
where E = net IP exposure (mg/cm2)
C = net IP concentration, (ug/m3)
U = mean wind speed (m/s)
t = duration of sampling (s)
Exposure represents the net mass flux of airborne particulate matter at the
downwind sampling point, integrated over the time of sampling, or equiva-
lently, the total net particulate mass passing through a unit area normal
to the mean wind direction during the test. Net IP concentrations and ex-
posures for the sample test are presented in Table A-2. The sample test
lasted 120 min.
A-4
-------
WEIGHT % GREATER THAN STATED SIZE
993 99J 99 98 98 90 80 70 60 30 40 30 20 10 3 2 I 0.5 0.2 O.I
100 r
so
20
IO
o
I s
IT
LJ
O.S
0.2
OJ
100
30
20
10
0.3
0.2
0,
ai 0.2 0.3 I 2 5 10 20 30 4O 50 60 TO 80 90 93 98 99 994 99.9
WEIGHT % LESS THAN STATED SIZE
Figure A-l. Downwind particle size distribution measured at a height of
1 m for Run M-3.
A-5
-------
100
90
WEIGHT % GREATER THAN STATED SIZE
993 99.8 99 98 99 90 80 70 60 50 4O 30 20 10 8 2 OS 02 01
20
10
to
I
o
I 9
j5
LU
UJ
IT
2
0.9
0.2
OJ
100
so
10
09
0.1 02 0.3 I 2 S 10 20 30 40 50 60 70 80 90 99 9« 99 994 9*9
WEIGHT % LESS THAN STATED SIZE
0.2
0.1
Figure A-2. Downwind particle size distribution measured at a height of
3 m for Run M-3.
A-6
-------
TABLE A-2. NET INHALABLE CONCENTRATIONS AND EXPOSURES
Height Concentration (|jq/m3) Wind speed
Cm; Downwind Upwind Net (m/s)
1.0 93.8 32.9 60.9 2.78
3.0 62.1 32.9 29.2 3.48
Net IP
exposure
(mg/cm2)
0.122
0.0732
EXPOSURE PROFILE
Typically the (net) exposure values decrease with increasing height in
the plume. If exposure is mathematically integrated over the vertical ex-
tent of the plume, then the quantity obtained represents the total passage
of airborne particulate matter due to the source, per unit length of road-
way. This quantity is called the integrated exposure A and is found by:
TH
A = / E dh (4)
QJ
where: A = integrated IP exposure (m-mg/cm2)
E = net IP exposure (mg/cm2)
h = height (m)
H = vertical extent of plume above ground (m)
The exposure must equal zero at the vertical extremes of the profile,
i.e., at the ground where the wind velocity equals zero and at the vertical
extent of the plume where the net concentration equals zero. Because ex-
posure increases sharply over the first few centimeters of plume height,
the value of exposure at the ground level is set equal to the value at a
height of 1 m.
A-7
-------
The vertical extent of the plume is found by linear extrapolation of
the uppermost net TP concentrations to a value of zero. Net TP concentra-
tions are found by subtracting the upwind TSP concentration from the down-
wind profiler concentration. In the case of Run M-3, Table 16 of the text
shows that a plume height value of 8.1 m was found by extrapolation. For
cases in which extrapolation was not possible, a plume height of 10 m was
used.
Linear interpolation is used to generate the intermediate exposure
values (at 1 m intervals) needed for the Simpson's rule integration of A.
Because Simpson's Rule requires an odd number of equally spaced points,
additional points are added (if needed) by setting exposures of heights
greater than H equal to zero. From the data presented in Table A-2, the
exposure profile of Figure A-3 is thus obtained.
Application of Simpson's rule to perform the integration in Eq. (4)
for Run M-3 yields:
A = § (E0 * 4EX + 2E2 + 4E3 + 2E4 + 4ES + 2E6 + 4E7 + 2E8 + 4E9 + E10) (5)
where: A = Integrated IP exposure (m-mg/cm2)
E. = Net IP exposure at i m above ground (mg/cm2)
h = Distance between exposure values (i.e., 1 m)
EQ = Net IP exposure at ground level = Ex
When the values from Figure A-3 are substituted into Eq. (5), it is found
that the integrated exposure for Run M-3 equals 0.512 m-mg/cm2).
A-8
-------
10.
• Vertical Extent of Plume
Determined by Extrapolation
of Net TP Concentration Profile
O)
1
0
0.05 0.10
f\
Net .IP Exposure ( mg/cm )
0.15
Figure A-3. Exposure profile for Run M-3.
A-9
-------
INHALABLE PARTICIPATE EMISSION FACTORS
The emission factor for IP generated by vehicular traffic on the paved
road is given by
e = 104 j} (6)
where e = IP emission factor (g/VKT)
A = integrated IP exposure (m-mg/cm2)
N = number of vehicle passes
Note that the leading term of Eq. (-6) is a conversion factor. The IP emis-
sion factor for Run M-3 is 2.39 g/VKT based on 2,144 veh.icle passes during
the 120 min sampling period. To convert g/VKT to Ib/VMT, multiply by
0.00355.
OTHER EMISSION FACTORS
Particulate emission factors for other size ranges are found in a man-
ner analogous to that described above for IP. The concentrations for the
other size ranges are determined using the sizing information presented
earlier. Once the net concentrations are obtained, the exposure values and
emission factors are found in the same manner as those for IP.
A-10
-------
APPENDIX B
CORRECTION PARAMETER CALCULATION PROCEDURES
B-l
-------
Silt loading is calculated as the product of total loading and frac-
tional silt content. The total loading is simply the mass of street sur-
face particulate sample divided by the surface area from which the sample
was obtained. The tare weights of sample containers are subtracted from
the total weights to obtain the sample weights. Table B-l gives the
procedure for determination of silt content.
Mean vehicle weight is the arithmatic average of the weights of
vehicles passing over the test road segment during the emissions sam-
pling period. Vehicle weights are assigned to vehicle types as des-
cribed in the body of this report.
TABLE B-l. SILT ANALYSIS PROCEDURES
1. Select the appropriate 8-in diameter, 2-in deep sieve sizes. Recom-
mended U.S. Standard Series sizes are: 3/8 in., No. 4, No. 20, No. 40,
No. 100, No. 140, No. 200, and a pan. Comparable Tyler Series sizes
can also be utilized. The No. 20 and the No. 200 are mandatory. The
others can be varied if the recommended sieves are not available or if
buildup on one particular sieve during sieving indicates that an inter-
mediate sieve should be inserted.
2. Obtain a mechanical sieving device such as a vibratory shaker or a
Roto-Tap (without the tapping function).
3. Clean the sieves with dry compressed air and/or a soft brush. Material
lodged in the sieve openings or adhering to the sides of the sieve
should be removed (if possible) without handling the screen roughly.
4. Obtain a scale (capacity of at least 1,600 g (3.5 lb)) and record ;nake,
capacity, smallest division, date of last calibration, and accuracy.
B-2
-------
TABLE B-l (concluded)
5. Tare sieves and pan. Check the zero before every weighing. Record
weights.
6. After nesting the sieves in order from the largest to the smallest
openings with pan at the bottom, dump dried laboratory sample (immedi-
ately after drying) into the top sieve. The sample should weigh be-
tween 800 and 1,600 g (1.8 and 3.5 lb).a Brush fine material adhering
to the sides of the container into the top sieve and cover the top
sieve with a special lid normally purchased with the pan.
7. Place nested sieves into the mechanical device and sieve for 10 min.
Remove pan containing minus 200 mesh and weigh. Replace pan beneath
the sieves and sieve for another 10 min. Remove pan and weigh. When
the difference between two successive pan sample weighings spaced
10 min apart (where the tare of the pan has been subtracted) is less
than 3.0%, the sieving is complete. Do not sieve longer than 40 min.
8. Weigh each sieve and its contents and record the weight. Check the
zero before every weighing.
9. Calculate the percent of mass passing the 200 mesh screen (75 urn phys-
ical diameter). This is the silt content.
a This amount will vary for the finer textured materials; 100 to 300 g may
be sufficient when 90 percent of the sample passes a No. 8 (2.36 mm)
sieve.
B-3
-------
APPENDIX C
PROPOSED AP-42 SECTION
The reader is cautioned that this proposed AP-42 section is subject
to probable change resulting from internal EPA reviews before it is published
in AP-42.
C-l
-------
11.2.5 PAVED URBAN ROADS
General - Various field studies have indicated that dust emissions from
paved streets are a major component of the material collected by high vol-
ume samplers.1 Reentrained traffic dust has been found to consist primarily
of mineral matter similar to common sand and soil, mostly tracked or depos-
ited onto the roadway by vehicle traffic itself. Other particulate matter
is emitted directly by the vehicles from, for example, engine exhaust, wear
of bearings and brake linings, and abrasion of tires against the road sur-
face. Some of these direct emissions may settle to the street surface, sub-
sequently to be reentrained. Appreciable emissions from paved streets are
added by wind erosion when the wind velocity exceeds a threshold value of
about 20 km/hr (13 mi/hr).2 Figure 11.2.5-1 illustrates particulate trans-
fer processes occurring on urban streets.
Emission Factors and Correction Parameters - Dust emission rates may vary
according to a number of factors. " The most important are thought to be
traffic volume and the quantity and particle size of loose surface material
on the street. As shown in Figure 11.2.5-1, various activities add or re-
move street surface material. On a normal paved street, an equilibrium is
reached whereby the accumulated street deposits are maintained at a rela-
tively constant level. On average, vehicular carryout from unpaved areas
may be the largest single source of street deposit. Accidental spills,
street cleaning and rainfall are activities that disrupt the street loading
equilibrium, usually for a relatively short duration.
The lead content of fuels also becomes a part of reentrained dust from
vehicle traffic. Studies have found that, for the 1975-76 sampling period,
the lead emission factor for this source was approximately 0.03 gram per
vehicle mile. With the reduction of lead in gasoline and the use of cata-
lyst equipped vehicles, the lead factor for reentrained dust was expected
to drop below 0.01 grams per mile by 1980.3
The quantity of dust emissions of vehicle traffic on a paved roadway
per vehicle kilometer of travel may be estimated using the following empir-
ical expression4:
••*(& "
where: e = particulate emission factor (g/VKT)
L = total road surface dust loading (g/m2)
s = surface silt content, fraction of particles
< 75 |Jm diameter (American Association of
State Highway Officials)
k = base emission factor (g/VKT)
p = exponent (dimensionless)
The total loading (excluding litter) is measured by sweeping and vacuuming
lateral strips of known area from each active travel lane. The silt frac-
tion is determined by measuring the proportion of loose dry road dust that
passes a 200 mesh screen, using the ASTM-C-136 method. Silt loading is the
product of total loading and silt content.
Miscellaneous Sources n 25-1
02
-------
Cn
I
t-l
co
t-3
o
50
DEPOSITION
PAVEMENT WEAR AND DECOMPOSITION
VEHICLE RELATED DEPOSITION
OUSTFALL
LITTER
MUD AND OIRT CARRYOUT
EROSION FROM ADJACENT AREAS
SPILLS
8) BIOLOGICAL DEBRIS
?) ICE CONTROL COMPOUNDS
REMOVAL
REENTRAINMENT
WIND EROSION
DISPLACEMENT
RAINFALL RUNOFF TO CATCH BASIN
STREET SWEEPING
>-~^ s*-*-
3*W8ft».ftfe^
11.2.5-1, Deposition and removal processes.
-------
The base emission factor coefficients (k) and exponents (p) in the
equation for each size fraction are listed in Table 11.2.5-1. Total sus-
pended particulate (TSP) denotes that particle size fraction of airborne
particulate matter that would be collected by a standard high volume
sampler.
TABLE 11.2.5-1. Paved Urban Road Emission Factor Equation Parameters3
Particle Size Fraction k (g/VKT)
TSP
< 15 |jm
£ 10 Mm
< 2.5 Mm
5.87
2.54
2.28
1.02
0.9
0.8
0.8
0.6
, Reference 4. See p. 11.2.5-1 for equation.
Aerodynamic diameter. TSP is total suspended particulate.
Microscopic analysis indicates the origin of material collected on high
volume filters to be about 40 weight percent combustion products and 59 per-
cent mineral matter, with traces of biological matter and rubber tire par-
ticles. The small particulate is mainly combustion products, while most of
the large material is of mineral origin.3
Emissions Inventory Applications4 - For most emissions inventory applica-
tions involving urban paved roads, actual measurements of silt loading will
probably not be made. Therefore, to facilitate the use of the previously
described equation, it is necessary to characterize silt loadings according
to parameters readily available to persons developing the inventories. It
is convenient to characterize variations in silt loading with a roadway
classification system, and this is. presented in Table 11.2.5-2. This sys-
tem generally corresponds to the classification systems used by transporta-
tion agencies, and thus the data necessary for an emissions inventory -
number of road miles per road category and traffic counts - should be easy
to obtain. In some situations it may be necessary to combine this silt
loading information with sound engineering judgment in order to approximate
the loadings for roadway types not specifically included in Table 11.2.5-2.
A data base of 44 samples analyzed according to consistent procedures
may be used to characterize the silt loadings for each roadway category.4
These samples, obtained during recent field sampling programs, represent a
broad range of urban land use and roadway conditions. Geometric means for
this data set are given by sampling location and roadway category in Table
11.2.5-3.
Miscellaneous Sources 11.2.5-3
C-4
-------
TABLE 11.2.5-2. Paved Urban Roadway Classification"
Roadway
Category
Freeways / exp r es sways
Major streets /highways
Collector streets
Local streets
Average Daily Traffic
(ADT)
> 50,000
> 10,000
500 - 10,000
< 500
Lanes
> 4
> 4
2b
2C
Reference 4.
Road width £ 32 ft.
Road width < 32 ft.
TABLE 11.2.5-3. Summary of Silt Loadings (sL) for Paved
Urban Roadways
Roadway
Local
Streets
X (g/m2) n
g
Baltimore 1.42 2
Buffalo 1.41 5
Granite City (IL)
Kansas City
St. Louis
All 1.41 7
Collector
Streets
X (g/m2)
g
0.72
0.29
-
2,11
-
0.92
n
4
2
-
4
-
10
Category
Major Streets/
Highways
X (g/m2)
g
0.39
0.24
0.82
0.41
0.16
0.36
n
3
4
3
13
3
26
Freeways/
Expressways
X (g/m2) n
g
-
-
-
-
0.022 1
0.022 1
Reference 4. X = geometric mean based on corresponding n sample size.
11.2.5-4
EMISSION FACTORS
05
-------
These sampling locations can be considered representative of most large
urban areas in the United States, with the possible exception of those in
the Southwest. Except for the collector roadway category, the mean silt
loadings do not vary greatly from city to city, though the St. Louis mean
for major roads is somewhat lower than those of the other four cities. The
substantial variation within the collector roadway category is probably at-
tributable to the effects of land use around the specific sampling locations.
It should also be noted that an examination of data collected at three cities
in Montana during early spring indicates that winter road sanding may produce
loadings five to six times higher than the means of the loadings given in
Table 11.2.5-3 for the respective road categories.5
Table 11.2.5-4 presents the emission factors by roadway category
and particle size. These were obtained by inserting the above mean silt
loadings into the equation on page 11.2.5-1. These emission factors can be
used directly for many emission inventory purposes. It is important to note
that the paved road emission factors for TSP agree quite well with those
developed from previous testing of roadway sites in the major street and high-
way category, yielding mean TSP emission factors of 4.3 g/VKT (Reference 6)
and 2.6 g/VKT (Reference 7).
TABLE 11.2.5-4.
Recommended Particulate Emission Factors for Specific
Roadway Categories and Particle Size Fractions
Emission Factor (by oarticle
Roadway
Category
Local streets
Collector streets
Major streets/
highways
Freeways/
expressways
TSP <
g/VKT (Ib/VHT) g/VKT
15
(0.053)
10
(0.035)
4.4
(0.016)
0.35
(0.0012)
15 Mm <
Ub/VMT) g/VKT
5-3
(0.021)
4.1
(0.015)
2.0
(0.0071)
0.21
(0.00074)
size fraction)
10 Mm
Ub/VMT)
5.2
(0.018)
3.7
(0.013)
1.8
(0.0064)
0.19
(0.00067)
< 2.5 Mm
g/VKT (lb/VMT)
1.9
(0.0067)
1.5
(0.0053)
0.34
(0.0030)
0. 16
(0.00057)
References for Section 11.2.5
1. D. R. Dunbar, Resuspension of Particulate Matter, U. S. Environmental
Protection Agency, Research Triangle Park, NC, March 1976.
2. M. P. Abel, "The Impact of Refloatation on Chicago's Total Suspended
Particulate Levels", Purdue University, Purdue, IN, August 1974.
Miscellaneous Sources
C-6
11.2.5-5
-------
2. M. P- Abel, "The Impact of Refloatation on Chicago's Total Suspended
Participate Levels", Purdue University, Purdue, IN, August 1974.
3. C. M. Maxwell and D. W. Nelson, A Lead Emission Factor for Reentrained
Dust from a Paved Roadway, EPA-450/3-78-021, U. S. Environmental
Protection Agency, Research Triangle Park, NC, April 1978.
4. C. Cowherd, Jr., et al., Paved Road Particulate Emissions, EPA Contract
No. 68-02-3158, Midwest Research Institute, Kansas City, MO, April 1982.
5. R. Bonn, Update and Improvement of the Emission Inventory for MAPS Study
Areas. State of Montana, Helena, MT, August 1979.
6. C. Cowherd, Jr. et al., Quantification of Dust Entrainment from Paved
Roadways, EPA 450/3-77-027, U. S. Environmental Protection Agency,
Research Triangle Park, NC, July 1977.
7. K. Axetell, and J. Zell, Control of Reentrained Dust from Paved Streets,
EPA Contract No. 68-02-1375, PEDCo Environmental, Inc. Cincinnati, OH,
July 1977.
11.2.5-6 EMISSION FACTORS
07
-------
TECHNICAL REPORT DATA .
(Please read Instructions on the reverse before completing)
1 . REPORT NO.
EPA-600/7-84-077
4. TITLE AND SUBTITLE
Paved Road Particulate Emissions—Source Category
Report
6. PERFORMING ORGANIZATION CODE
3. RECIPIENT'S ACCESSION-NO.
5 REPORT DATE
July 1984
7. AUTHORIS)
Chatten Cowherd, Jr. and Phillip J. Englehart
8. PERFORMING ORGANIZATION REPORT NO,
4892-L
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Midwest Research Institute
425 Volker Boulevard
Kansas City, Missouri 64110
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-3158, Task 19
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13 TYPE OF REPORT AND PERIOD COVERED
Final; 1/80-4/84
14. SPONSORING AGENCY CODE
EPA/600/13
is. SUPPLEMENTARY NOTESlERL-RTP project officer W.B. Kuykendal is no longer with the
Laboratory; for report details, contact Dale L. Harmon, Mail Drop 61, 919/541-2429.
is. ABSTRACT The report gives results of extensive field tests to develop emission fac-
tors for particulate emissions generated by traffic entrainment of paved road surface
particulate matter. Emission factors were developed for the following particle size
ranges: < or = 30, 15, 10, and 2. 5 micrometer aerodynamic diameter. Sites tested
represented commercial/industrial, commercial/residential, expressway, and
rural town land-use categories. The measured inhalable particulate (IP--< or = 15
micrometer aerodynamic diameter) emission factors ranged from 0.06 to 8.8 g/VKT
(vehicle km traveled). Lowest emissions were measured for the expressway cate-
gory; highest emissions were measured for the rural town category. About 90% of the
IP emissions consisted of particles < or = 10 micrometers in aerodynamic diameter,
and about 50% of the IP emissions consisted of particles < or = 2. 5 micrometers in
aerodynamic diameter. Using roadway surface silt loading as the basis, predictive
emission factor equations for each particle size range were derived. To facilitate
the use of these particle-size-specific equations in developing emission inventories,
a classification system was derived of mean or typical silt loadings as a function of
roadway category. These mean silt loadings were then inserted into the respective
emission factor equations for specific, roadway categories and particle size fractions.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Pollution
Pavements
Roads
Particles
Dust
Silts
Pollution Control
Stationary Sources
Particulate
Emission Factors
Silt Loading
13 B
14G
11G
3. DISTRIBUTION STATEMEN1
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
97
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
C-8
------- |