PAVED ROADS
AP-42
Section 11.2.5
Reference Number
1
RESUSPENSION OF PARTICULATE MATTER
Standards" Implementation Branch
Control Programs Development Division
Office of Air Quality Planning and Standards
U. S. Environmental Protection Acency
Research Triangle ?ark,»V. C.
March 15, 1975
-------
TABLE OF CONTENTS
Executive Summary ... i
Chapter I. Documentation of Resuspension Issue .1
Chapter II. Available Control Techniques ......... 8
Appendix A. '-Jhat is on the Street and How Does it Get
There? i A-1
Appendix 3. Theoretical Emission Factor Estimates
for Resuspension . 8-1
-------
Executive Summary
Over the past several years, various studies have been conducted
in numerous cities to investigate the urban total suspended oarticulate
matter problem. These studies have generally cone to one basic con-
clusion; that is, in addition to other various sources of oarticulate
matter, vehicular traffic activity, including resusoension of street
!dust by wind or traffic, is an underlying source which significantly
contributes to the TSP problem within the urban area.
These studies have concluded that:
(a) Vehicular traffic activities (i.e., direct tailpipe emissions,
rubber tire wear, and resuspension of street dust) contribute
approximately 50% (Philadelphia) to 90S (Chicago) of the TSP
collected on certain high-volume air samplers in close
proximity to the roadway.*
(b) Resusoension of material caused by vehicular movement alone
accounts for approximately 15% of the annual TSP average in
Chicago (averaged for 20 monitoring locations throughout
the city). Resuspension due to i-dnd action, generally greater
than 13 miles per hour, may contribute an additional 5% to
the annual TSP average in Chicago.
(c) Resuspension is generally a localized problem similar to CO
in that TSP levels vary depending uoon distance from the roadway.
(d) Resuspension of anti-skid materials (e.g., sand or salt for snow
control) may be significant on certain days of the year but
generally have a minimal effect on annual air quality levels.
In Detroit, ambient sodium chloride levels increased the
total TSP loading by 15« to 20% on days after salting for snow
control. •
*Sorne have questioned the results of these studies, however, there has
been no data to substantiate their claims to date.
-------
(e) Sources of material found on street surfaces vary greatly
depending on such factors as land use, aeopraohic location,
season, weather, traffic activity, and the length of time
which has elapsed since the site was last cleared.
(f) The major sources of street dust include:
(1) Erosion of the street surface itself,
(2) Motor vehicle emissions from both tailpioe and tire wear,
etc.,
(3) Atmospheric fall-out of both natural and man-made materials,
(4) Run-off and carryout of materials from adjacent lands,
(5) Spills of material from vehicular transoort, and
(6) Use of anti-skid materials, such as salt and sand during
oeriods of snow cover.
(g) One study indicated that half of the mineral material on the
roadway is derived from agareqate limestone (i.e., erosion
of the pavement), and the remainder of the material is from
adjacent soil.
(h) On the average for all types of streets (commercial, industrial,
and residential), 350 Ibs. of material, less than 100 microns,
is found alonq a curb mile of roadway at a aiven point in time.
(f) Within 3 or 4 days, dust loadings on a street approach a
maximum and begin to level off.
(j) Emission factors .available in the literature to estimate the
amount of material that becomes airborne due to traffic along
a "dirty" street range from 0.8 to 77 grams VMT. Based imon
available information, it is believed that 1 to 5 prams VMT is
*
a reasonable estimate of the amount of material less than
40 microns that becomes suspended.
ii
-------
(k) Street cleaning technology consists of sweeping and flushing.
Flushing is conducted mainly for aesthetics to displace dirt
and debris from the center of the street to the curb area.
Traditional brush-type sweepers have an overall removal
t . efficiency of 50% and only a 151 to 20S efficiency for particles
less than 100 microns.
However, the results from some tests have indicated an actual
increase of particles less than 100 microns after the test site had
been swept with a brush-type sweeper. Vacuum sweepers are more efficient
and are used quite extensively in Europe but have not been used in this
country.
Though one could question the accuracy of the various semi-
quantitative techniques used in some of the studies, the.findings seem
«
clear - resuspension has been and still is being identified as a common
source of TSP in many areas. While the results of each individual study
taken alone may not be conclusive, when looked at collectively the pre-
ponderence of evidence leaves little doubt that resuspension is a
significant problem for attainment of the TSP standards within urban
areas. In summary it is difficult with the information available to
say quantitatively (without question) how much of an ambient impact
resuspension has, however, because of the confirming results from city
to city, it is difficult to deny that resusoension is a contributing
factor to high ambient levels in urban areas.
Presently there are several ongoing projects and some additional
work planned which is associated with the resuspension problem.
-------
{a} 14 Cities Study conducted by GCA on the National Assessment
of the Particulate problem, scheduled for completion in the
Sprinq of 1976.
(b) Emission Factor Study, conducted by MR! to determine an emission
factor for resuspended dust from paved roads to be completed
in Spring of 1976.
(c) American Public Storks Association study with the National Science
Foundation to design an environmentally desirable street sweeper.
(d) Contract Study by Region II! 'to assess the particulate problem
for Philadelphia.
(e) Contract Study to review and update the siting criteria for
particulate monitors.
(f) Control technology assessment study to evaluate the available
control techniques associated with resuspension. This work is
presently in the planning stages.
As the above studies are completed, the results will be forwarded
to the Regional Offices and States 'for their use in the development of
revised implementation plans for total suspended oarticulates in non-
attainment areas.
-------
CHAPTER I
DOCUMENTATION OF RESUSPENSION ISSUE
!• Background & Defjnition
5 - Wipespread failure to attain the national ambient air quality
•standards for particulate matter in many urban areas has spurred a
reexamination of the nature of the urban particulate problem.
Routinely the control strategy development for these areas included
an analysis of the contribution of conventional point and area
sources without consideration of other less conventional sources of
particulate. One of these non-conventional sources is vehicular
traffic-related particulate, including resuspended particulate matter
commonly referred to as just "resuspension" or reentrainment. For
the purposes of this discussion, resuspended particulate matter (RPM)
is defined as that particulate matter which becomes airborne through
the action of the wind across a paved surface or by traffic upon the
roadway,
II. Urban Sjtudies
Several studies have been conducted over the past few years to
investiqate the urban total suspended particulate matter problem. Many
of these studies were concerned with the identification of suspended
particulate particles and the source of such particles. Some of the
studies have indicated that vehicular traffic related activities in
general are the key sources while other studies have more specifically
identified resusoension of roadway material as the problem.
(a) Vehicular traffic-related studiets
One of the first studies was by Scott Research Laboratories in
-------
Philadelphia in 1972. Suspended particulate samples were collected
from four sampling sites which were part of the existing air monitoring
network for the City of Philadelphia. The site locations ranged from
12 to 75 feet above the ground and 25 to 90 feet from the nearest roadway.
The data were analyzed by two semi-quantitative techniques, microscopic
and elemental analysis. Based on a combination of microscopic and
chemical analyses, the Philadelphia samples show that 30 to 40% of the
airborne particulates come from stationary sources and approximately 50%
of the particulate is traffic related with a bulk of the particulate large
enough to settle quickly. However, some are fine exhaust particles
which can be potentially hazardous. Resuspended material was estimated
at approximately 30%.
2
Another study in Chicago, Illinois, by I IT Research Institute
identified the types and sources of suspended particles in Chicago via
microscopic analysis and examined the relationship between sampling height
and TSP. Historical samples from existing sites in various portions
of the city with varying heights and distances from the roadway, as well
as new samoles from temporary sampling locations, were analyzed. The
results of the study indicated that calcite and auto tailpipe emissions
were the major particles found in every samole analyzed. Their combined
mass accounted for more than 90% of the filter weight in many cases.
With regard to samel ing height, the study concluded that;
(1) Hi Vol samplers close to street level show higher mass loading
due to collection of entrained mineral particulate matter.
(2) Two sites almost identical in all resoects except for differences
in height above ground were found to have identical type particles
but differed in TSP values solely due to differences in height.
-------
3
In a study in Denver , microscopic analysis of particles revealed
that an estimated 801 of the mass collected on the hi-volume filter con-
sisted of road dusts (e.g., quartz, limestone and mica). These"minerals
are common to (1) the soils of Denver, (2) the dust on its street pave-
ment, and (3) the pavement itself (monitoring site located 10 - 20 miles
*
NE of Denver). Because of the relatively low wind speeds observed during
the sampling period of the study, it seemed quite probable that auto traf-
fic entrained most of,the minerals collected, although no specific estimate
of the amount entrained was given. In addition, rubber tire dust and
auto exhaust particles were also found on the filters which provides some
additional support that the particulates in Denver are hiqhly related
to vehicular traffic.
(b) Res_u_sp_e_nsion Studies
In an investigation of Chicago ambient air quality data, Abel
and Neuman ^analyzed the TSP levels from January 1966 to Harch 1974 to
determine the impact of resuspension (or refloatation) due both to wind-
induced and traffic-induced causes on Chicago's total suspended oarticu-
late levels. The results indicated that resuspension below wind speeds
of 13 mph is generally regarded as vehicular suspended particles and
accounts for a time weighed average (on annual basis) of 10 ug/m or 15/» of
of Chicago's annual average. Wind speeds in excess of 13 mph cause refloat
tion of surface dust such that the total annual TSP average is increased
nearly 5%. In conclusion, "the total amount of TSP due to resuspension
3
may be 18 to 20 ug/m in Chicago no matter what the yearly average TSP
value is because resuspension is a function of traffic volume, precipi-
tation and high wind velocities, rather than a function of industrial
emission sources."
-------
In a study in 1973 by PEDCO Environmental, a resuspension problem
was analyzed for Denver. In that city, dust emissions are created
by traffic movement over sanded highways and streets resulting from
the mechanical grinding of sand which remains after the snow melts.
On an annual basis, this has only a minimal effect on air quality, but
it has considerable air quality impact on those few days a year when
3
it does occur (average increase of 50 ug/m ).
In 1972 the Montana Department of Health and Environmental Sciences
conducted a sampling program which indicated that a very large portion
of the material collected by the high volume air sampling is dust from
streets (no estimate of concentration was given). Of this street dust,
a large part comes from winter sanding operations. Through the use of
questionnaires, it was determined that few communities used any form
of dust removal when they treated the sand for use. Further, the sand
on the streets is ground into finer dust by the tires of vehicles.' Re-
sults of a special study indicated that though only about 1 percent of
the sand applied was dust (• 79u)» by the spring because of the traffic
activity, 16% of the sand found on the roadway surface had been ground
into dust. The dust which has been carried away by rain or melting snow
and by the wind was not included in the sample. Therefore, it is quite
likely that the actual percentage of dust may well have exceeded 20*.
The impact of this dust on ambient levels was not quantified, however.
Another study conducted in Detroit during late 1973 and early 1974'
indicated a close correlation between local street salting and general
ambient levels of aerosol chloride in suspended particulate as measured
by the hi-vol. The samplinq network was established using high volume
air samplers to samole aerosol at urban locations which would be expected
-------
to experience various degrees of influence from traffic-related street
salting. Actual percentage of sodium chloride measured as chloride
at 8 metro stations in Detroit was found to be from 4 to 8% of the
total suspended oarticulate matter for normal winter months to highs
«f 20 to 30 percent following incidents of street salting. Assuming
chloride could be used as a tracer for resuspended materials, the study
would indicate that 15 to 20 ug/m results from resuspended salt.
8
A study was conducted by Pennsylvania State University to deter-
mine the original source of reentrained material. In the study, selec-
ted samples (grab) from the road and adjacent soil were subjected to
emission spectrographic analysis. The dominant particle size for this
source is in the range of-10 to 60 microns. While there were slight
differences between the road samples, these were within the expected
exDerimental error'of the whole procedure. The difference that appears
significant is a definite lower silica and higher limestone composition
for the road dusts than for the soil sample, showing that roughly half
of the dust on the roadway appears to be derived from the aggregate
limestone and the remainder from adjacent soil dust.
Q
Similarly, Harrison and Rahn undertook a separate study to deter-
mine the chemical composition of Chicago street dust. The study consisted
of a survey of the elemental composition of Chicago street dust using
nondestructive neutron activation as the analytical tool. The study
suggested that the variability of the concentration of enriched elements
in the street dust provided evidence that the material was deposited
locally and that the street dust was not well mixed. In other words,
deoosition and resusoension of materials is local in nature.
-------
Mr. G. A. Sehmel of Battelle, Pacific Northwest attempted to
quantify the particle suspension caused by auto and truck generated
surface stresses by using solid ZnS as a tracer {-'25u particles).
The results of the study indicate the following: (1) the resus-
pension rate increases with vehicle speed, (2) the resuspension
rates are proportional to car-generated turbulence, (3) resuspension
decreased nearly an order of magnitude when the vehicle is driven on
a lane adjacent to the tracer as compared to driving through the tracer
and (4) oarticles are less readily resuspended the longer the particles
remain in contact with the road surface (due to weathering).
From the data collected during the study on the day the tracer
is placed on the road, it is estimated that II of the material becomes
resuspended at an automotive speed of 50 mph and 0,3% for speeds of
30 mph. However, in just 5 days after placing the tracer on the surface,
the resuspension rates have dropped to .}% for 50 mph and .006% for •
30 mph. It should be noted that the resuspension rates for trucks are
greater than the resuspension rates for cars at the same weathering time
by a factor of 10,
Soon after the particles become resuspended4 they begin to deposit
on the ground immediately adjacent to and downwind of the road. Samplers
were located a varying distance up to 30 feet from the.roadway. Deposition
is a function of weathering time and for a car driving through the tracer
at a speed of 30 mph after 30 days, the cumulative deposition fraction
(the fraction of particles resuspended and leaving the road during a
vehicle pass which is deposited up to the distance of interest) at 30 feet
from the road, is .32. That is, 32% of the faterial that was resuspended
would have been deposited between the road and 30 feet from it.
-------
Late in 1972 a study was benun in Seattle's Dumwamish Air Basin
to measure the emissions from dusty roads. The study indicated that
road dust, from both paved and unpaved (gravel) roads is the largest
source of suspended oarticles in Seattle, 2100 tons/year. Table VIII
(Appendix B) provides some estimates of the impact of mud carryout from
unpaved roads and parking lots, and dust from paved streets.
Additional Work and' Conclusions
In addition to the above 'auto-related activity and resuspension
studies, some additional work has been done by 6CA and others, both within
the United States and Australia, to determine the variation of TSP levels
with height and slant distance (i.e., vertical and horizontal distance
of a sampler) from a roadway. In all cases the sites further from the
roadway consistently had lower TSP concentrations than those nearer the
road when average daily traffic (ADT) was considered. Specific relation-
ships will become available as the GCA study nears completion.
Thus, from the above results one can reasonably conclude that while
resuspension is a significant source of TSP, its impact upon receptors
will vary, depending upon the receptor's distance from the roadway and
height above the ground. Thus from th'ii sense the resuspension problem
may be like CO in that it may have more of a localized, rather than a
regionwide, effect.
-------
CHAPTER II
AVAILABLE CONTROL TECHNIQUES
The present methods of cleaning streets fall into two categories:
sweeoing and flushing. Hachine sweeping accounts for the great
majority of street cleaning in most communities. This may be supple-
mented by some manual sweeping in a few areas. Rainfall also acts
to clean streets in the short term, however, it is responsible for
mud carryout in the long term such that overall rainfall may not be an
effective "control,''1
Mechanical street sweepers are designed to loosen dirt, trans-
port it onto a conveyor and deposit it in a hopper. Typically sweepers
do have some type of dust control system.
There are three basic types of sweepers in use: (1) pickup broom
which uses a rotating gutter broom to move matter into the main pickup
broom -water spray is used to control dust, (2) regenerative air which
blasts dirt from surface into a hopper with some of the air being
recycled - also uses water soray for dust control, (3) vacuum (limited
use in U.S. - wide use in Europe} where all material picked up by the
vacuum nozzle is saturated with water on entry.
Street flushing is presently conducted to mainly displace dirt and
debris from street into the gutter, at which point the material is
concentrated for sweeper oickuo. Most agencies use flushers for
aesthetic purooses and moving material out of travel lanes quickly.
Table I is a summary of cleaning oractices in selected cities.
All cities surveyed were found to have a comnrehensive sweeping program.
*.
About half had a flushing program. Several cities relied on manual
8
-------
TABLE I
CLEANING PRACTICES IN SELECTED CITIES
MAJOR CLEANING
San Jose
Phoenix
Mi Iwaukee
Baltimore
Atlanta
Seattle
Minneapolis
St. Paul
San Francisco
Lawrence
NOTE:
SWEEPER
X
X
X
X
X
X
X
X
X
X
•*
PROGRAMS
FLUSHER MANUAL
0
M
M
X
*\
X
M
X
X
X
0
M
X
X
X
X
M
M
X
M
, » * *
EQUIPMENT
SWEEPER
15
21
22
26
24
18
. 18
14
14
3
FLUSHER
0
1
2
11
3
8
3
7
10
2
EQUIPMENT/l.OOO-mi. STREET
SWEEPER
12.9
14.5
12.9
13.0
13.7
14.1
18.0
15.6
16.5
20.0
FLUSHER
_
0.7
1.2
5.5
1,7
6.3
3.0
7.8
11.8
13.3
x = major use.
0 = none
M = minor use.
Source: Reference 12
-------
cleaning programs, mainly in downtown areas. The study " also surveyed
the sweeping frequency, sweeper utilization factor, oickun per sweeper,
and pickup rate. This data is summarized in Figure I,
Three general types of tests have been conducted to determine
sweeoing effectiveness: (1) jn situ street tests, (2) controlled
tests in which paved areas are artificially given a variable or uniform
loading, and (3) strip test in which a narrow path of material is laid
down to be removed. Since the last test is easily run and readily
reproducible, most of the data available is for strip tests. Since
strip tests provide nearly ideal operating conditions, it is not
surprising that results from such tests result in very high removal
efficiency (90%). Controlled tests have reported less efficient, results
and JJ1 sjtu street tests even lower. It should be noted, however,
that the in situ street tests represent actual'real world conditions.
A summary of these tests results is shown 4n Figure II.
In an in situ test conducted for- the 1972 study, the contractor
found that there was an overall removal effectiveness of 50% for
sweeping ooerations. However, when considering the removal effectiveness
in terms of oarticle size, the results are quite different. Larger
oarticles are nicked up very effectively (70 - 80%), however, the
smaller fraction range (43 - 104u) in some cases showed an increase
(See Table II).
Since aoproximately 80 - 90% of the material found on the road
accumulates within 12 inches of the curb, most of the street cleaning
operations concentrate on this area. Thus, since the smaller oarticles
are not effectively picked uo, they may be jictually redistributed across
the entire street, due to the action of the brushes near the curb.
10
-------
t/l
o
c
-t
n
ID
•*)
(D
-5
n>
n
ro
o
o
o
Oi
-J
o
3
O
l/J
ro
fO
T3
fO
-$
to
"I
-h
O
3
O
fO
CO
in
Baltimore
San Francisco
San Jose
Phoenix
o.
Baltimore
"* San Francisco
o
San Jose
Phoenix
SWEEPING FREQUENCY
(times swept/year)
ro
O
o
0%
o
Baltimore
San Francisco
San Jose
Phoenix
PICKUP RATE (cu yd/curb mile) OP
Baltimore
San Francisco
San Jose
Phoenix
SWEEPER UTILIZATION FACTOR
(curb miles swept^yehicle)
o
o
o
ro
o
o
o
OJ
o
o
o
O
O
o
01
o
o
o
cr>
o
o
o
o
o
o
ex
c
c
PICKUP PER SWEEPER (cu yd/vehicle)
o
o
o
o
r\>
o
o
o
o
o
o
-------
O >. i—
100
I/)
o- z:
"O
3
in
"O
C
(O
oo
CO
Q£
=3
o
cc
50
O)
o
tt>
C,
o
UJ
(X
C
a)
Street
Tests
Controlled
Tests
Strip
Tests
FIG. II. Comoarison of Results from Sweeping Effectiveness Tests Conducted
Under Various Conditions: For Dirt/Dust Fraction
Source - Reference 12.
12
-------
TABLE I!
REMOVAL EFFECTIVENESS VERSUS PARTICLE SIZE DISTRIBUTION
ATLANTA
PARTICLE
SIZE RANGE
(micron)
> 2,000
840-2,000
246-890
104-246
43-104
< 43
Total (g)
(Derail Eff.
INITIAL
LOADING
(g)
175
103,
375
231
' 66
43
993
(*)
RESIDUAL
LOADING
(g)
76
14
56
29
136
187
498
50
TULSA
INITIAL
LOADING
(9)
1,438
418
690
544
415
324
3,829
RESIDUAL
LOADING
(9)
142
181
588
595
549
431
2,486
35
PHOENIX
INITIAL
LOADING
(9)
535
308
2,190
1,273
425
175
4,906
RESIDUAL
LOADING
(9)
240
107
224
381
614
498
2,064
62
SCOTTSDALE
INITIAL
LOADING
(9)
217
439
915
421
213
87
2,292
RESIDUAL
LOADING
(9)
43
124
115
287
134
44
1,017
56
Source: Reference 12
-------
These results were corroborated by another study which indicates
the particles that are most important as far as air and water pollution
is concerned are the most poorly reduced by conventional street sweep-
ing procedures. Table III provides a summary by particle size of the
results of the two studies mentioned above. The second study utilized
an equation to determine the removal efficiencies.
The efficiency of street sweepers can be increased by (1) oper-
ating the sweeper at a slower rate, (2) increasing the number of
oasses on a given street, and (3) increasing the frequency of
cleaning. In fact, it was shown that by decreasing the speed from
5 to 2% mph almost as much dirt could be picked up in one pass as with
two at the higher speed. Also, it was found that vacuum sweeping was
more effective than motorized sweeping.
Two studies are currently underway to obtain more information about
street cleaning operations. The first is a study by the National Science
Foundation (NSF) under contract with the American Public Works Associ-
ation to develop performance specifications for street cleaning as it
oertains to air and water pollution requirements. The study has four
major activities: '!) literature search relating to street sweeping,
(2) survey of street cleaning equipment maintenance and costss (3) survey
of street cleaning practice and costs, and (4) develop performance
specifications for existing technology and orescriptive specifications
for environment-oriented technology.
The second study is being done as part of the EPA's New York City
demonstration grant to the Interstate Sanitation Commission. As part
of the study, Brooklyn Polytechnical Institute is evaluating the resuspen-
sion of dust from paved streets in ^lew York City. Further information
14
-------
Table III.
KSTIMATHD STaKKT .V'-'. K;-;:"H:; Ki'F] >" ! KNCV
PARTICLE SIZE
; X SIT:' TEST
(i: S v 1 W :11 C 3
> 2,000
S'iO - 2,000
246 - »-iO
•13 - 10-;
< 43
< C)
< 0
Source - Reference 12.
15
-------
on this project will be forthcoming as no specific date has been
set for its completion.
In the past, the major concern of the Agency in regard to street
cleaning, etc., has been with respect to the water oolution and solid
12 13
waste aspects associated with these activities. ' Thus, up to now
little emphasis has been placed uoon street sweeping and cleaning for air
oollution considerations. The .major studies therefore that have been
completed in the past and a majority of those going on presently,
except for the above work, have been mainly associated with the solid
waste and water pollution aspects of the problem.
Little information is available in the literature on the cost
of street sweeping. What information is available has shown wide
variations between cities of comparable size. One study reported in
the literature was done by The American, City, magazine in 1970 - 1971.
In this survey only about one third of the survey-reporting cities sup-
plied data on their unit street-sweeping costs, though sweeping is a
major budget item. What information was available indicated that the
average curb mile sweeping costs vary from $8.42 in the surveyed cities
of 500,000 population and above to $2.18 per curb mile among those
cities of 25,000 to 50,000 population. Some relatively old total expendi-
ture data (1955) indicates the costs for street cleaning varies fron
500,000 to 3 or 4 million dollars per year per city.
Summary
Since most of the work oh street cleaning practices has been done
in regard to the water pollution aspects, there is no real information
available which indicates how effective these practices might be in
controlling resuspension. To our know!edge,*no work has been done to
date in which ambient air quality samples have been collected before
16
-------
and after various street cleaning procedures have been used. In
fact, we are unsure if the actual sweeping operation using brush-
type sweepers does not in fact create more of a resuspension problem
11*
than it solves. "* A good ambient study to further quantify the impact
of these measures is needed before we recommend that they be used as
part of the control strategy alternatives for resuspension.
In addition to the above control techniques, steps can be taken
to minimize the amount'of dust which can get onto the street in the
first place. This concept in the past has generally been associated with
abating of local nuisances. However, recently some have looked to
these regulations as a source of control of resuspended particulate
matter in that one could minimize the amount of material upon the
roadway that has the potential of being resuspended. These regulations
take many different forms or approaches. The first is the so called
"mud-carryout" regulations. This regulation basically requires
those involved in construction type operations » etc., to wash down
the surrounding streets at the end of every day or so to minimize the
amount of mud or-soil carryout upon the surrounding streets.
The second type of control involves the watering or chemically
stabilizing construction haul roads and other areas on the construction
site to control wind induced or vehicle induced particulate emissions
from construction activities.
The third type of regulation is the limitation of the number of
acres that may be disturbed at any one time prior to stabilization.
This greatly reduces the potential for vast areas of the construction
site to be subjected to potential wind erosion,
»
The fourth type of control regulation deals with the issuance of
17
-------
a permit to construct or excavate. In this type of regulation the
agency places stipulations on the permit that requires that the owner
or operator take all reasonable precautions to minimize the dust
emissions from this activity.
The fifth type of regulation deals with the requirement that all
trucks carrying aggregate type material must be covered to minimize
the emissions from transport and reduce losses due to spills.
Hhile these type of regulations have been part of the air program
in many cities for several years, there is no information available
at present to indicate how actively these regulations are enforced if
they are enforced at all. In addition, there have not been any studies
done to indicate how effective these measures may be if they are
properly used and actively enforced in reducing the TSP levels due
to resuspended particulate matter. It is believed however that
these types of minimizing regulations do have considerable promise
in reducing the amount of material that can get onto the roadway and
have the potential for resuspension.
-------
APPENDIX
The following appendices contain information on the more techni-
cal aspects of the resuspension problem. They are included to oro-
vide a more thorouah understand!"nq of what is on the street and how
does it get there? (Appendix A) and the range of enission factors
<*
associated with the resuspended problem {Appendix B).
Appendix A provides some insight into the nature and amount of
material that is found on urban streets as a result of studies done
to identify and quantify street surface contaminants.
Aopendix 8 attempts to calculate some theoretical emission factors
using all the available information found in the literature to date.
It should not be interpreted to be the last word on emission factors
for resuspension but should act as a guide to determine if the nore
empirical numbers are reasonable and in line with the rannes proposed
in this paoer.
-------
APPENDIX A
NHAT IS ON THE STREET AND HOW DOES IT GET THERE?
A- Suspension Process
While particulate emissions from paved surfaces are primarily
generated by vehicle motion, participate emissions may also be
generated when the wind velocity across a surface exceeds the
threshhold velocity value at which dust becomes airborne from streets
(i.e., erosion threshhold velocity). In Chicago, this threshhold
velocity has been estimated to be 13 mph and is resoonsible for approxi-
mately 5% of the annual TSP concentration,
A threshhold value of air caused surface stress on a particle
must be exceeded before a particle is suspended. "The threshhold
stress is a function of particle properties as well as surface prooerties.
Particles for a given set of physical prooerties are resuspended more
easily from a smooth surface than from one of irregular shape, such a's
an asphalt surfaced road. However, in-some cases, the particles may
become attached to the surface. This attachment of the particles to
the surface or particles of the surface is referred to as weathering.
Particles weather as a function of time and become more firmly attached
•to the surface, thus making them less susceptible to resuspensionJO
Basically, there are three types of movements associated with the
action of the wind across a surface (usually soil, although it has an
aoplication to paved surface): surface creep, saltation and suspension.
Surface creep is associated with particles in the size range of 500 to
1000 u. As the wind exerts its force on these large particles (500 to
1000 u) they are rolled along the ground, beijig pushed instead of lifted.
A-l .
-------
Saltation consists of individual particles jumping and bouncing within
a few centimeters of the surface. The particles that .ire acted upon
by saltation are those within the 100 to 500 u range. The third type
of movement caused by the action of the wind consists of particles below
*
100 u being lifted off the surface and becoming comoletely airborne.
These particles will stay suspended as long .-s the uoward vertical
velocities of the wind are greater than the terminal settling velocities
of the particles. However, the mechanism whereby fine particles are
suspended is slightly different than that described above. Some work by
Dr. Chepil has shown that soil or dust composed only of small particles
lie in a somewhat laminar layer next to the surface and therefore do
not protrude as much as larcjer particles do into the turbulent air layers.
Therefore, if the dust contains larger particles, they are the first to be
*
affected by surface creep which in turn causes saltation and finally
suspension of the finer particles. It should be noted, however, that
the fine dust is susceptible to suspension by means other than wind,
such as the action of moving vehicles, people, etc.
3. Accumulation Process
Before discussing the sources and amount of contaminants on
a street, it is important to understand the basic principles involved
with the accumulation of contaminants on a street surface. "Consider
a hypothetical area of street surface which is (for the purposes of
discussion) subjected to a continual and uniform loading of contaminants
(uniform with respect to both time and spatial distribution). If
there were no other activities to disturb the contaminants, the loading
intensity would increase linearly with respect* to time." -^
A-2
-------
Where periodic cleaning is practiced, the plot looks like below.
Note -that this represents a case of uniform, continuous loading and
a regular cleaning (with the same degree of efficiency each time and a
uniform frequency).
Accumulation o£ Contaminants - Hypothetical Case
(natural buildup with periodic s-.vcopiug but no rainfall)
Intermittent rains would also have an effect on accumulation. Large
storms would remove more than sweepers; small storms, less.
o
Accumulation ui Coiil'nmijumLs -
Typical Cntso (natural buildup
with periodic sweeping and
in tcnui ctuiH' rninCnll)
The main purpose for the above discussion is to place into context
the meaning of the "observed loading intensities" to be discussed and
A-3
-------
provide a general understanding of the accumulation process. Streets
were sampled to determine their contaminent loading intensities. At
each samolirtq location, historical information was obtained as to when
the street had last been swept and when the last rain occurred. While
thele-data are of value, they can by no means be used to describe the
shape of the overall curve. "Observed loading intensities" provide an
answer to the question, how much material resides on a typical street
12
at any point in time?
C. Sources of Material Found on the Urban Roadway
Figure 3 is a block diagram depicting a qualitative mass balance
for urban resuspension problems. Two studies have been done by Municipal
Pollution Control Division (MPCD) of the Office of Research and Development,
12 13
EPA, * on the composition of materials on roadways in relation to
the water pollution aspects of .street surfaces and urban roadways
which are applicable to air pollution problems.
The materials found on street surfaces vary greatly. Obviously,
the material observed at any given location will be a composite of
several sources depending on such factors as land use, geographical
location, season, weather, .traffic activity, etc. There are 6 major
sources of materials found on the street^
(a) Eros ton of street surface material itself. ("'On a weight
basis, aggregate materials account for the largest contribution
from this source.")
(b) Motor vehicle related particles (broad range of materials,
such as tire wear, particulate exhaust emissions and dirt
from undercarriages). *
A-
-------
NATURAL BACKGROUND
Localized Wind-
Induced Suspension
Construction
Unoaved Roads
Storage Piles
Fallow Fields
Total Suspended Particle
as Measured by Hi Vol
Urban Sources
Point
Area
Resuspension
(Vehicle "lotion and Wind)
Vehicular Deposits
Carryout from Unpaved
Roads
Tire Hear
Tailpipe Emissions
Accumulated
Paved Street
Deoosits
Runoff
Sanding
Salting
Spills
I
Mechanical Removal
(Street Cleaners)
Mashing
i
I •
Dry Sweeoina
Figure 3. Urban Resuspensiorv
A-5
-------
(c) Atmospheric fallout. (This material may be
of a size fine enough that it could have been transported
by air currents prior to being deposited on the street
surface. Major sources of such materials would be industrial
stacks, construction projects, agricultural operations and
exposed vacant land.)
(d) Runoff from adjacent land areas, (Areas where soil is exposed
rather than protected by vegetative cover, oaving or other
means,)
(e) Spills from vehicular transport. (While this source is well-
known, it is virtually impossible to quantify. The types of
material that may be included are dirt, sand, qravel, cement,
etc.)
(f) Anti-skid conoounds. (Sand, salt, ashes which are applied
with the intent of melting ice or providing better traction
during the winter months.)
It should be noted that the amount of material residing on a
street surface will vary considerably from olace to olace and from time
to time depending on several factors:
—time since last cleaning or rainfall
--season of year
--land use activity for a particular location
"Microscopic examination of the bulk of the materials found on street
surfaces consists of -inert minerals of various types which effect the
components of street paving compounds and local geology." This inert or
inorganic material is probably blown, washed, or tracked in from surrounding
1 A 1*
land areas.
A-6
-------
The quantity and character of contaminants found on streets is
summarized in Table I. These data are weighted averages for the 12
cities samples (San Jose, Phoenix, Milwaukee, Bucyrus, Baltimore,
Atlanta, Tulsa, Seattle, Decatur, Scottsdale, fiercer Island and
Owasso). Table II contains the values for each land use category
where samples were collected. These values are the average loadings
one would find if the material were spread uniformly across the full
width of the street. However, some studies have been done which
indicate that some 80 - 90* of the material found on a street is
within 12 inches of the curb. Therefore, the numbers in Tables jv and V
should be used with caution.
The quantity of contaminant material existing at a given test site
was found to depend upon the length of time which had elapsed since the
site was last cleaned; intentionally {by sweeping or flushing) or by
rainfall. The field sampling program focused on collecting materials
from street surfaces at single points in time. However, information was
collected for each site to define the elapsed time since the last sub-
stantial rain storm and/or cleaning. Accumulation patterns as calculated
here are shown in Figure IV.
As can be seen, contaminant loading intensities were found to vary
with respect to land-use activities. In general, industrial areas
have substantially heavier than average loadings. This is probably
due to the fact that these areas are swept less often and because
"fallout", spills, unpaved roads, etc., tend to be high in these areas.
However, commercial areas have substantially lighter loading. This
may be due in part to the fact that they are sweot more often than other
A-7
-------
TABLE TV _^
V.'KICHTKD MEANS
:n FOR ALL SAMPLF.S
CONST!TUKNTS (1 h/eu ri> mile)
ToLnl Solids MOO
Oxypcii HewaiKl
1101)5 in. 5
COIJ !T>
Vnlnlilo So]ids 100
Algal Ntilr ic n Is
* Phospha tcs _ I.I
Nilrnlcs • ,09'i
- Kjeldahl Nitrogen • 2.2
Bacter iol oc;i en 1
Total Conforms (nrc.Aairl) ml I c) 99 x 10'
Fecal Col if onus (org/curb inilu) 5,G x 1U*
Heavy Me in Is
7.1 tic . nr>
Lend .57
N'ic'UcI • .Uf>
Me reury .073
Chromium . 11
Pesticides
0 I x I0~G
l)i pi dri» 2-1 x !0~f'
I'ol yi-li) i>ri 11:1 i I'd I! i phcny 1 s ll()0 x 1 0~b
Source: Reference i2
Note-; Tlie term "org" refers to "number of-coll form orcnnisms observed,"
TABLE V
.!r,. Ki^'MttCAl- WKII.II IKU
£ Ih/curh fi J
1, 2DU
(<-.»,•. i If! .'« I iij l<; ' ti'ift
|.>iv nl»I 'will ; i H"it
"M-d< ni"% 'j i MI: l f •! :u'
menl '(Mfl ' <•• I rn: l
-------
Fiaure IV
1400
1200
1000
800
600
400
200
0
Industrial
categories..
Residential
Commercial
5
8
9 10 11 12
Elaosed time since last cleaning by sweepina or rain (days)
Source: Aooendix I Reference 12
A-9
-------
areas. Finally, residential areas were found to have an average loading
intensity comparable to the average for all land use of all cities.
However, the loading varied widely from site to site. One possible
explanation that the more affluent neighborhoods tended to be cleaner
possibly because they are better maintained by residents.
In order for the above information on loadings to be useful to
air oollution work, one must know what percent of this material may
become suspended and collected by the high volume air sampler. In
theory particles less than lOOu can readily become suspended and be
measured as total suspended particles by the high volume air sampler
technique.
Particle size distributions were determined from composite samples
collected from 5 representative cities. The data were determined by
summing values obtained by dry sieving, wet sieving and sedimentation
pipette analysis. The results are given in Table I-/. 'The average
percent composition of particles less than lOOu was approximately 25%.
Thus, in order to obtain the amount of material which has the
potential for resuspension, one must multiply the value for a given land
use category by 25% (That percent of material which is less than lOOu).
If the overall weighted mean average of 14001b/curb mile is used, then
{.25 x 1400 Ib/curb mi, = 350 Ib/curb mile) is in the potential size range
for resuspension, and collection by the high volume air sampler.
In a second study by HPCO work was done to analyze roadway dust and
dirt. (See Table VII.) Street surface contaminants are deposited on the
roadway from many sources as stated above. Metropolitan Washington, O.C.,
»
with its low background of industrial emissions was chosen for a study to
A-10
-------
determine the contribution of motor vehicle usage to urban roadway loading.
Specific sites were selected to provide minimal interference from non-
traffic related land use activites and thus- isolate traffic-related depositions,
Less than 5% by weight of the material on the streets originates
directly from motor vehicles, however, these pollutants are
important because of their potential toxicity (lead, zinc, other metals,
asbestos from clutch and brake linings), the bulk of the traffic-related
material is representative of local geology and to lesser extent, products
of abraded roadway surfaces.
An examination of the street surface contaminants found that they
consisted largely of roadway surfacing materials and various mineral forms
representative of the local geology. The results of the study showed
that dust and dirt is composed of over 95% inorganic, material, most of
which is insoluable. Close examination of the particles under the micro--
scope .revealed many individual particles appeared to be fractured mineral
crystals. A semi quantitative emission spectrographic analysis of eight
samples were performed to determine the major metallic constituents. The
results are summarized in Table VII.
.Loadings intensities of street surface contaminants measured during
a 12-month field study were examined and observed loadings were plotted
as the dependent variable against total traffic and least square equation
of the linear relationships were calculated. For examole, tha equation
of the least squares line obtained upon plotting total dust and dirt
bv weight in pounds per mile against traffic in axles is:
Ib/mile = 96 + .00238 tires axles »
A-n
-------
Tnblc VI
pAiri'in.r. sr/.K I.HSTUIHUTION OK SOLIDS
SKLKCITU) CITY COMPOS 1TKS
MIl.l'-'MiKKK ilUCYKUS IWI.TIMOUE ATLANTA
TULSA
> -1, W>0 i.
2,000-1. MOO .1
8-10-2, OiiQ U
21G-U-IO 'i
10-1 -2 16 ti
•13- 10-1 U
30-13 tl
1-1-30 u
•l-l'l U
< -1 .1
12.
12,
•It),
20
0
1
•1
2
1
0
,O7,
.1 •
. H
,'1
,5
.3
.2
.0
*i
.r>
"" ,p
30. 1
7.3
ao.o
15.5
20.3
13,3
7 , 'J
'** . 7
-
17 .
•1 .
ti.
22,
20.
11 .
10.
-1 .
2.
"•
•17=
G
0
3
3
5
1
'1
ii
n
-
M.
G.
30,
29,
10.
5.
1 .
0.
0.
r*
/e
»
6
9
5
1
1
8
D
3
-
37.
9.
1C,
17.
12.
3.
3.
0.
0.
%
1
4
7
1
0
7
0
9
1
Sand •'.
•13-4, y 7,,
'' -i il
U2.1
7..!
7-1.1
25.9
ti'J.l
17.1
Ul.9
7.8
0.3
92.3
7.G
0.1
LI) Saiiil/curt) mi U.-lHO 1,020
Lli Sil I /curb rat 200 356
!.!. rinv/<-tirli mi I 3 . f.
« 15
170
U.3
39-S
33,5
1 .3
300
30
0.3
Noli; : u = jni crons .
Source: Reference Vl
A-12
-------
TABLE VII
SEMIQUANTITATIVE EMISSION SPECTRQGRAPHIC ANALYSES
OF ROADWAY DUST AND DIRT SAMPLES
(WASHINGTON, D.C. METROPOLITAN AREA)
Concentration
High
Med-high
Low
Low-med
Trace
Trace-med
Trace-low
n.d.-trace
Silicon
Iron
Aluminum
Calcium
Magnesium
Titanium
Zinc
Lead
Boron
Barium
Cobalt
Strontium
Chromium
Copper
Vanadium
Nickel
Sodium
Zirconium
Manganese
Molybdenum
Tin
Reference 13
A-13
-------
Although the deposition of traffic-related materials occurs at a constant
rate, the accumulation of material along the roadway tends to level
off after some period of time due to traffic-related removal mechanisms.
The y intercept, 96 Ib/roadway mile, is the amount of total dust which
*
appears as a result of phenomena not related to actual traffic. This
is dependent upon geographic location and the intensity of the local particu-
late air oollution problem. Only a very small portion of the 96 Ib/roadway
mile is due to materials abraded from the roadway during sample collection.
Therefore, in any event a majority of the y intercept results from the
transport of particulate pollutants from air currents from some distance.
Since the rate at which airborne material is deposited is more nearly time
dependent than traffic related, the y-intercept is no doubt a function of
ti me.
As mentioned previously, the deposition of the pollutants on a
street through traffic-related mechanisms appears to occur at a relatively
constant rate and is independent of loadings already present. However,
the accumulation of surface contaminants is not linear and levels off
due to a combination of factors, other than cleaning or rainfall.
One hundred twenty-seven (127) roadv/ay samples were collected during
the study, 94 of which samples were taken to be used in calculation of •
traffic-related deposition rates. Of the 94, 75 were collected after a
one-day accumulation. The remaining 19 samples were gathered after either
3 or 4 days accumulation. Comparisons of the loading samples for one
day accumulation with those of multiday accumulation revealed that
within 3 or 4 days, the loadina began to level off and approach a
»
maximum value. This occurs as particles of the material are nicked un
A-14
-------
by passing traffic and by other mechanisms and displaced onto areas adjacent
to the roadways. Mechanical fracture to smaller particle size, as well as
physical transport, is postulated as the mechanism responsible for this •
leveling off. The relationship between total dust and dirt dry-weight
loading and accumulation period are shown in Figure II for one
street in Washington, D.C. using above estimates of K? (removal rate) and
K-j (deposition rate) = 2.38 x 10" Ib/axle-mile and average daily
traffic (ADT) of 40,000 axles. The speed limit on this street is
30 men. Thus, given the proper information to calculate K~ and ADT,
one can calculate the maximum oollutant loading on any given street
for traffic-related sources of street dust. However, as mentioned
above, motor vehicle or traffic-related sources of contaminants are
only one source of dust and dirt found on the street surface, therefore,
in order to calculate the amount of total material that is susceptible
to resuspension (<100u)» one should use the total loading intensities
1 2
approach from the first HPCD study for particles less than lOOu.
A-15
-------
300
= 1 x 10 , Lm =233 Ibs./raile
m = 79.3 Ibs./mile
34567
Accumulation Period (Days)
12
,-3
16
20
28
+4
Traffic (Axles x 10 )
(k1 = 2.38 x 10 Ibs./axle-mile, AI)T = 40,000 axles)
Figure V Total dust and clii'C dry weight accumulation
10 11
36 40
Source: Reference 13
A-1S
-------
APPENDIX B
THEORETICAL EMISSION FACTOR ESTIMATES FOR RESUSPENSION
The previous information in Appendix A on loading intensities
couoled with the information on the fraction resusoended from Chaoter I
allows one to calculate a resuspension factor in Ib/veh-mile.
I. Loading Intensity Approach (HDCD/Sehjne_l Dajta^
Since the accumulated loading begins to level off after the third
or fourth day, the resuspension rate which utilized the 5-day weathering
factor (from Sehmel's work, see Chapter I) is used to calculate a resus-
pension factor. This factor would estimate the amount of material
Generated (<100u) which would be suspended beyond 30 feet from the roadway
and theoretically would be measured by a high volume air sampler.
{4.44X10"5)(350 lb) .015 lb = 7.05g
Residual
Street
Industrial
Street
vehic
1 e mi 1 e
10"5)(300)
10~5)(70Q)
_
veh-mile
« .nisib _
veh-mile-
= '.03l1b
veh-mi 1
6.0Sq
veh-mi le
14.11g
veh-mile veh-mtle
(4.44 X l0-5)(72.5) =
v /\ /
ven-mile veh-mile
It should be noted that these factors represent averages based unon
loadings which were highly variable from site to site and city to city.
Also no ambient measurements were made in the vicinity of the roads in the
water pollution studies to provide any cross-check on the validity of using
the values obtained by Sehmel-in his tracer work. Sehmel also was only
concerned with asohalt roads and the streets used in the water oollution
studies were constructed of various materials not just asohalt. -In fact,
the first studv by HPCH states that pavement comoosition and coordination
-------
.05 TMVEF - .34
TMVEF - 6.8
Road dust = TMVEF - exhaust - tire wear - brake wear
= 6.8 - .34 - .2 - .02
=6.24 grams/vmt
m- Lead Tracer Approach.
Another approach for estimating the impact of resuspension via an
1R
emission factor is to utilize the lead tracer concept. From previous work
on auto exhaust particles, lead is believed to comprise 26% of the suspended
1 9
particle emissions from automobiles. Thus, if one has ambient lead data,
he can obtain the associated auto tailpipe emissions by multiplying by 3.8
(3.8 x 26% - 100% TSP auto). In the case of Denver, Colorado, the annual
3
arithmetic mean for TSP at the one site with lead data is 139 ug/m . The
•3
average lead concentration at that site is 1.51 ug/o .
139 ug/m3
- 35 ug/m3 background
109 ug/m3 from man-made sources
1.61 x 3.8 = 6.12 ug/m3 from auto TSP tailpipe
Thus if the site were solely traffic-originated, then the worst case
situation would assume that all the remaining TSP were from resuspension alone,
Thus if all the remaining TSP were from resuspension then
AQ auto _ EF auto
AQ res up" ~ EF resup
6.12 ug/m3 = .34 .g/vjot
109 - 6.i2ug/mJ X
EF = 5.72 g/vmt (Probable worst casa number)
However, if the site is not strictly influenced by traffic-related
*
activities, then one might assume that only 15* of the annual TSP is
-------
auto-related resuspension as estimated in Abel's work in Chicago as
referenced above:
6.12 ug/m3 _ .34
"
20.85 ug/nP X
; X = 1.16 g/vmt
W - - Afr Quality Impact AgpToacFi
Another approach is to attempt to define an air quality impact type
factor rather than just an emission source type. This was attempted by
using the line source model and air quality data from a site in Philadelphia
which is located very near a heavily traveled paved road and which is believed
to record TSP values which are highly related to traffic activity.
The line source equation when the wind direction is normal to the line
is as follows:
X » - 2 °z 20
&n c?2 u
X = cone, mg/m
q = source strength g/sec-meter
cj = vertical dispersion coefficient
H = source or receptor height (m)
U = average wind spted (m/sec)
Now solving for source strength line source equation becomes
X /2n e, U
2 exp -% (£ )2
az
For the particular situation in Philadelphia the following values
were used: *•
B-4
-------
X = AQ - Bkg = 115-35 = 8T mg/m3
18
o = 4.6 m (Figure 3-3 Turner } for x = ,006 km
U = 4.47 m/sec
H = 3.96 m
_ .08 /2T37T4) (4.6) (4.471 4. 1.2
q " , _ , /3,96S? 1.38
2 exp -% (jT]*-)*
q = 2.9855 mg/m/sec
The street 'near where the monitor is located has an average weekday
traffic count of 30,000 vehicle/day. Thus, in order to develop a factor
in grams/vmt, one must use the ADT data coupled with the calculated source
strength asfollows:
2. 9855 xVQ~ v 3600 sec v 8760 hr v H v yr v wkday
....... ........ ' ' - «" •»" A ................... ™ ........ """•— A r ...... ..... "" '"" "" ........... ..... A .......................... "" ..... " A ^ - A -~ * " —^
~3 v
" A
m - sec hr yr 6.21x10 26/wkday 30,000
miles
q =19.44 grams/vmt
The above estimate is simply used as a rough estimate of how the approach
would be used; however, it does attempt to provide a "ball park estimate" of
resuspension influence as related to existing air quality levels. Certainly
such things as particle size and density of the material a hi-vol might
collect at 6 m from the road at a height of 13 feet off the ground is a
factor along with the particle size and density of the material on the
roadway. More accurate estimates of wind speed and direction along with
a better estimate of AOT and daily air quality levels would provide a
V
better estimate of the source strength from the road as calculated via its
impact on existing hi-volume air samolers. The number above again would
probably be a worst case estimate. Also, if there were other hi-vols at
varying distances from the roadway, one could use this technique to calcu-
late the source strength from the road based on e«ech hi-vol measurement
and average these to provide an estimate of what had to be the calculated
-------
source strength to have the TSP levels at varying distances from the
roadway as measured by the hi-vol,
V, Summary and Status of Current Work on Emission. Factors.
Thus with the exception of the Seattle data, all the values for vehicle
resuspension are in the range of 1 to 20 g/vmt. (see Table VIII.) The most
reasonable values appear to be somewhere between 1-5 g/vrat as the number
generated using the MPCD values for loading could be reduced substantially
if one assumes that only particles of 30-40u or less instead 'of lOOu or
less would actually be picked up and deposited beyond the immediate vicinity
of the street. If one does make this assumption, {i.e., only particles less
than 40u), then the overall factor from the Sehmel/IPCD data would become
(4.44 X 10"5)(1400}(J35) = •=
' '^ '
veh mi veh mi
Recently another approach has: been used by RTI under contract wfth EPA
in their efforts to identify problems associated with non -attainment in
North Carolina. In this approach RTI modeled two cities in North Carolina,
and in both cases the model calibrated quite poorly. As a result of this
poor calibration, RTI found that they had overlooked resuspended participate
emissions from paved streets. By a trial -and-error method, they used various
emission factors for resuspensed particulate matter from paved streets until
they obtained a good correlation between predicted and measured air quality,
They found their best correlation when 6.1 g/vmt was used as an emission
factor, They tried their same approach in another city and found the value
to be 4.2 g/vmt. While the factors are relatively close considering two
different cities were involved, further investigation found that the city with
the lower emission factor had some vacuum sweeoimg as part of its street
cleaning program, which is more effective than brush-type sweepers used in
B-6
-------
Study
TABLE VIII
Road Surface
ParticulaLe Emission Factor
Dumavnsh Valley
Ousty paved road--
no curb
Paved streets--
washed regularly
Hud carryout from gravel
roads
Carryout from unpaved
parking lots
376 g/vmt {77 g/vmt
63 g/vmt
3174 g/vmt
336 g/vmt
Nashville Work
Dust from paved streets
.8 g/vmt
Nashvile Work
Modified
Dust from paved streets
6.24 g/vmt
Lead Tracer Approach Oust from paved streets
5,72 g/vmt
Lead Tracer
Approach Modified
Oust from paved streets 1.16 g/vmt
Air Quality
Impact Modified
Dust from paved streets 19.44 g/vmt
Sehmel/MPCD
Dust from paved streets
7.05 g/vmt
Sehmel/HPCD
Modified
Oust from paved streets
4,54 g/vmt
RTI Approach
Oust from paved streets 4-6 g/vmt
MRI Preliminary
Dust from paved streets 10 g/vmt (6.1 g/vmt' <3Qu)
B-7
-------
the other city. More work is proceeding on this study and the final
report should be available shortly.
Preliminary indications from some emission factor work currently being
done by MRI under contract with EPA indicate that the emission factor may
be Jpp-roximately 10 p/vmt(5.1 p/vmt<30y). The final report on this study is
due -in March of 1976.
B-8
-------
REFERENCES
1. A s_tudy_ofthe natureand Origin of Airborne Particujate Matter Jn
Philadelphia^ Scott Environmental Technology, Inc., January 1975..
2. Draftz, R.G., Types and Sources of Suspended Particles in Chicago,
Kay 1975. "~~ |J
3. Wilson, VI., et al, "Identification and Sources of Denver Aerosol,"
Presented it the 1974 APCA convention in Denver, Colorado.
4 Abel M 0 "The Impact of Refutation of Chicago's Total Suspended
, " Particuiate Levels," Thesis in partial fulfillment of Master of
Science Degree, Purdue University, August 1974.
5- Investigation of Fugitive Dust - Sources, Emissions and_ Control -
For At t a jinm'e n t o f _ Seep nd a ryArab "i e n tTA 1 r Qualify Stand a rd _s _, C o 1or ado
PEDCo EnvironmentaT Control Ho. 68-02-0044Task Order No. 16^
6, Brant, L.A., "Winter Sanding Operation and Air Pollution" Pub1ic
Works, pp. 94-97, September 1972.
7, Warner, P.O., et al., "Effects of Street Salting on Ambient air
Monitoring of Particuiate Pollutants in Detroit" presented at 6th
Center Regional Meeting American Chemical Society, April 21-24,
1974.
8. Shuler, L.M. and Heqmon, R.R., Road Oust as Related to Pavement
Po 1 j!shing, Automotive Research Program, Penn. State Univ. May 1972
1.. Rahn, K.A., and Harrison, P.R., "The Chemical Composition of Chicago
Street Dust"
10, Sehmel, G.A., "Particle Resuspension from an Asphalt Road Caused
by Car and Truck Traffic," Atmospheric Environment
11. Roberts, J.W., Rossano, A.T., Jr., Maters, H.A., "Dirty Roads =
Dirty Air," APWA Report, November 1973, p. 10-12.
12. '^ater Pollution Aspects of Street Surface Contaminants, Sartor, J.D.,
and Boyd, G.B., Contract No. 14-12-921, November 1972.
13, Contribytions of Urban Roadv/ay Usage to Water Pollution, Shaheen, O.G.
Contract No. 68-01-0197 Task Order 005, April 1975^
14. Laird, C. H. and John Scott, "How Street Sweepers Perform Today,"
American City_» pg, 58-62, March 1971,
-------
15. Buckwald, H., and Schrag, K.R., "Dust Exposure in Mechanical Street
Sweepers," American Industrial Hygiene Association Journal, pp. 485-487,
Sept-Oct. 1967.
16, Chepil, W.S., "Relation of Wind Erosion to the Dry Aggregate
Structure of a Soil," Sci Agr 21:488-SQ7, 1941,
^* ParticulateArea Source Emission Inventory for Nashville, PEDCo
Environmental, EPA Contract No. 68-02-1375 Task Order 9, Feb. 1975.
18. Personal Communication with, Mr. Tom Pace.
19. LilTis, E. J., and Dunbar, D. ft., "Imoact of Automotive Particle Exhaust
Emissions on Air Quality," Draft Report.
20. Turner, D, B., Workbook of Atmospheric Dispersion Estimates, U. S. EPA,
1970. ''"
------- |