EPA-450/3-74-037-
Jane 1974
DEVELOPMENT OF EMISSION
FACTORS FOR FUGITIVE
DUST SOURCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers. Copies are
available free of charge to Federal employees, current contractors and
grantees, and nonprofit organizations - as supplies permit - from the Air
Pollution Technical Information Center, Environmental Protection Agency,
Research Triangle Park, North Carolina 27711; for a fee, from the
National Technical Information Service, 5285 Port Royal Road, Springfield,
Virginia 22151.
This report was furnished to the Environmental Protection Agency by
Midwest Research Institute, in fulfillment of Contract No. 68-02-0619.
The contents of this report are reproduced herein as received from
Midwest Research Institute. The opinions, findings, and conclusions
expressed are those of the authors and not necessarily those of the
Environmental Protection Agency. Mention of company or product names
is not to be considered as an endorsement by the Environmental
Protection Agency.
Publication No. EPA-450/3-74-037
ii
-------
EPA-450/3-74-037
DEVELOPMENT OF EMISSION
FACTORS FOR FUGITIVE
DUST SOURCES
by
Chat ten Cowherd, Jr.
Kenneth Axetell, Jr.
Christine M, Guenther
George A. Jutze
Midwest Research Institute
425 Volker Boulevard
Kansas City, Missouri 64110
Contract No. 68-02-0619
EPA Project Officer:
Charles 0. Mann
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
June 19 74
-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers. Copies are
available free of charge to Federal employees, current contractors and
grantees, and nonprofit, organizations - as supplies permit - from the Air
Pollution Technical Information Center,
Research Triangle Park, North Carolina
Environmental Protection Agency,
27711; for a fee, from the
National Technical Information Service, 5285 Port Royal Road, Springfield,
Virginia 22151.
This report was furnished to the Environmental Protection Agency by
Midwest Research Institute, in fulfillment of Contract No. 68-02-0619.
The contents of this report are reproduced herein as received from
Midwest Research Institute. The opinions, findings, and conclusions
expressed are those of the authors and not necessarily those of the
Environmental Protection Agency. Mention of company or product names
is not to be considered as an endorsement by the Environmental
Protection Agency.
Publication No. EPA-450/3-74-037
11
-------
PREFACE
This report was prepared for EPA's Office of Air Quality Planning
and Standards, under Contract No. 68-02-0619. Mr. David Anderson and
Mr. Charles Mann served as EPA project officers. The work was performed
in the Physical Sciences Division of Midwest Research Institute under
the supervision of Dr. Larry Shannon, Head, Environmental Systems Section.
Dr. Chatten Cowherd, Jr., Principal Investigator, was the principal
author of this report. He was assisted by Mr. Kenneth Axetell, Jr., and
Mr. George Jutze of PEDCo-finvironmental (Subcontractor), who drafted the
first part of Chapter 6, Chapter 7, Appendix A and Appendix B. Other
MRI staff members who contributed to the program included Miss Christine
Guenther and Mr. Francis Bennett.
Approved for:
[WEST RESEARCH INSTITUTE
Hubbarc
Physical Sciences Division
111
-------
ACKNOWLEDGMENTS
Special appreciation goes to those who voluntarily assisted in the
testing of agricultural tilling and unpaved roads in western Kansas. In
Morton County, Mr. Herbert R. Williams, Agricultural Extension Agent,
provided valuable assistance in test site selection, and Mr. Dale Coen
and Mr. Bob G. Smith permitted use of their land and equipment for
agricultural testing. In Wallace County, Mr. Donald McWilliams,
Agricultural Extension Agent, aided in agricultural site selection and
Mr. Art Mai permitted use of his land and equipment for agricultural
testing.
IV
-------
ABSTRACT
This report presents the results of an extensive field testing pro-
gram to develop emission factors for certain common sources of fugitive
dust. The source categories that have been investigated are: agricultural
tilling, unpaved roads and air strips, heavy construction activities, and
aggregate storage piles. Characterization and quantification of emissions
from these sources are necessary to the development of effective control
strategies so that the national air quality standards for total suspended
particulates may be achieved.
Because little reliable emissions data existed for these sources prior
to this study, an extensive program of field sampling was required to gen-
erate the data which would provide the basis for emission factor determina-
tion. To this end, fugitive dust sampling techniques and associated data
reduction schemes were developed to quantify emissions from moving and
stationary dust sources. The basic measurements consisted of isokinetic
dust exposure profiles with specially designed sampling equipment, dust
concentrations with conventional high-volume samplers, particle size
classification with high-volume cascade impactors, deposition profiles
and dust transport by saltation. A description of the measurement tech-
niques and summaries of calculated test results are presented.
For each source type, emissions are related to meteorological and
source parameters, including properties of the emitting surface and
characteristics of the vehicle or implement which causes the emission.
This information is used to derive correction factors which appropriately
adjust basic emission factors to reflect regional differences in climate
and surface properties.
v
-------
TABLE OF CONTENTS
Page
Executive Summary xiii
Chapter 1 - Introduction. . ....... 1
Background 1
Unpaved "Roads and Air Strips 2
Agricultural Tilling 3
Aggregate Storage Piles 3
Construction Sites 4
Objectives. 4
Chapter 2 - Summary of Pertinent Literature ., 6
Emissions from Dirt Roads 6
Emissions from Gravel Roads ....... 9
Emissions from Agricultural and Construction Activities .... 10
Chapter 3 - Dust Emission Sampling Strategy 11
Emissions from Agricultural Tilling and Unpaved Roads ..... 11
Emissions from Aggregate Storage Piles. ,. 19
Chapter 4 - Unpaved Road Emissions 23
Sampling Site Description 23
Gravel Road Sites 23
Dirt Road Sites 23
Field Measurements. ............ 24
Test Results 28
Computed Emission Factors . . . 36
Correction Parameters .......... 40
Average Vehicle Speed ...... 40
Vehicle Mix 43
Surface Texture . 43
Surface Moisture. ... ......... 43
Corrected Emission Factor .......... , 43
vii
-------
TABLE OF CONTENTS (Continued)
Page
Chapter 5 - Agricultural Tilling Emissions 46
Sampling Site Description 46
Morton County, Kansas 46
Wallace County, Kansas. . 48
Field Measurements 50
Test Results 55
Computed Emission Factors 60
Correction Parameters 64
Surface Soil Texture 66
Surface Soil Moisture 66
Implement Speed . . ' ., 66
Corrected Emission Factor , 67
Chapter 6 - Aggregate Storage Pile Emissions 69
Total Emissions from Aggregate Storage Operations 69
Sampling Site Description 69
Field Measurements 70
Test Results 76
Correction Factors. „ 80
Computed Emission Factors 85
Emissions from Aggregate Loadout 89
Sampling Site Description 89
Field Measurements 89
Test Results 91
Computed Emission Factors 95
Comparison of Aggregate Emission Factors 99
Corrected Emission Factor 101
Chapter 7 - Building Construction Emissions .... 103
Paradise Valley Construction Study 103
VII1
-------
TABLE OF CONTENTS (Continued)
Page
Test Results 105
Calculated Emission Factors ......... 108
Correction for Activity Level . 110
Las Vegas Construction Study. ,....*.... 110
Test Results. 115
Computed Emission Factors ...... . 118
Correction for Activity Level . . 120
Summary and Conclusions ....... 124
Chapter 8 - Emissions Inventory Procedures. ... 125
Source Data Requirements. ........ . 125
Particle Drift Potential. ... 128
Windblown Dust 128
Chapter 9 - Conclusions 134
References „.. 136
Appendix A - Procedure for Estimating Windblown Dust. ....... 139
Appendix B - Dispersion Calculations for Construction Emissions . . 166
Appendix C - Photographs of Field Testing ... ..... 170
LIST OF TABLES
Mo. Title Page
1 Tests of Emissions from Unpaved Roads 8
2 Field Measurements—Unpaved Roads 25
3 Vehicle Mix (Unpaved Roads) 26
4 Emissions Test Parameters (Unpaved Roads). . 27
5 Dust Emission Sampler Locations (Unpaved Roads) 29
6 Measured Dust Emissions (Unpaved Roads). .... 30
ix
-------
TABLE OF CONTENTS (Continued)
LIST OF TABLES (Continued)
No. Title Page
7 Plume Sampling Data (Unpaved Roads). ...... 31
8 Road Surface Properties (Unpaved Roads) , 35
9 Calculated Emission Factors (Unpaved Roads). 41
10 Estimated vs Actual Emissions (Unpaved Roads). ...... 45
11 Agricultural Site Characteristics » 49
12 Field Measurements—Agricultural Tilling « 52
13 Emissions Test Parameters (Agricultural Tilling) ...... 53
14 Dust Emission Sampler Locations (Agricultural Tilling) . , 54
15 Measured Dust Emissions (Agricultural Tilling) 56
16 Plume Sampling Data (Agricultural Tilling) ... 57
17 Soil Properties (Agricultural Tilling) .......... 61
18 Calculated Emission Factors (Agricultural Tilling) .... 65
19 Estimated vs Actual Emissions (Agricultural Tilling) ... 68
20 Field Measurements—Aggregate Storage Piles. .. 72
21 High-Volume Sampling Data (Sand and Gravel Storage Piles). 74
22 Sampling Site Data (Sand and Gravel Storage Piles) .... 75
23 Suspended Dust Concentrations (Sand and Gravel Storage
Piles) 77
24 Aggregate Size Ranges. 79
25 Average High-Volume Concentrations During Wet and Dry
Sampling Periods 81
-------
TABLE OF CONTENTS (Continued)
LIST OF TABLES (Continued)
No. Title
26 Sampling Data for Working and Nonworking Periods ..... 86
27 Calculated Emission Factors (Aggregate Storage Piles)... 88
28 Field Measurements—Aggregate Loadout 90
29 Emissions Test Parameters (Crushed Stone Storage Piles). . 92
30 Measured Dust Emissions (Crushed Stone Storage Piles)... 93
31 Plume Sampling Data (Aggregate Storage). ......... 94
32 Aggregate Properties (Crushed Stone Storage Piles) .... 97
33 Calculated Emission Factors (Crushed Stone Storage Piles). 100
34 Aggregate Storage Emissions Breakdown. 102
35 Suspended Particulate Concentrations (]ig/m ) (Paradise
Valley: 31 August - 22 October 1972) 106
36 Calculated Emission Factors (Paradise Valley Construction
Site) . 109
*3
37 Activity Level vs Particulate Concentration (ug/mj)
(Paradise Valley). . Ill
38 Dust Concentration vs Activity Level 112
39 Activity Level vs Concentration (ug/nP) for W, SW and S
Winds. . . 113
40 Las Vegas Site - Sample Values (ug/m3) Sampling Period -
21 August - 22 October 1972 116
41 Measured Concentrations During N, NE, S and SW Winds
(Ug/m3) 119
42 Results of Dispersion Calculations a 121
-------
TABLE OF CONTENTS (Continued)
LIST OF TABLES (Concluded)
No. Title Page
43 Las Vegas Construction Study Activity Level vs
Concentration (ug/nP) ............ . ..... 122
44 Las Vegas Construction Study Activity Level vs
Concentration (ug/tiP) ........ . ........ . 123
45 Area Source Data ..................... 126
46 Acres of Construction - 1973 ............... 127
A-l Soil Erodibility for Various Soil Textural Classes .... 147
A-2 Values of K, L and V for Common Field Crops ........ 156
A-3 Calculation Sheet for Estimation of Dust from Wind
Erosion ......................... 163
B-l Measured Concentrations During S, SW and W Winds (ug/m3) . 167
LIST OF FIGURES
No. Title Page
1 Overhead View of Dust Plume from Moving Point Source ... 12
2 Comparison of Wind Speed Profiles ............. 15
3 MRI Dust Exposure Profiler ................ 17
4 Dust Exposure Profiler for Elevated Emissions Source ... 20
5 Positioning of Test Equipment—Aggregate Loadout ..... 21
6 Exposure Profiles — Unpaved Roads ............. 32
7 Particle Size Distributions — Dirt Road Emissions ..... 34
8 In-Place Road Dust Texture ....... . ........ 37
xn
-------
TABLE OF CONTENTS (Continued)
LIST OF FIGURES (Continued)
No. Title Page
9 Drift Potential of Road Emissions 38
10 Effect of Vehicle Speed on Gravel Road Emissions 42
11 Climatic Factor Used in Wind Erosion Equation 47
12 Exposure Profiles—Agricultural Tilling. .... 58
13 Particle Size Distributions—Agricultural Emissions. ... 59
14 Surface Soil Texture 62
15 Drift Potential of Tillage Emissions 63
16 Aggregate Storage Sampling Site, Dravo Corporation,
Camp Dennison, Ohio 71
17 Stockpile Emissions vs Wind Speed 83
18 Stockpile Emissions vs Aggregate Size . 84
19 Particle Size Distribution--Aggregate Loadout Emissions. . 96
20 Aggregate Size Distribution—Crushed Stone 98
21 Paradise Valley Construction Site 104
22 Pollution Roses—Paradise Valley Construction Site .... 107
23 Las Vegas Construction Site 114
24 Pollution Roses—Las Vegas Construction Site 117
25 Map of PE Values for State Climatic Divisions. 130
26 Moisture Correction Factors 131
27 Map of Precipitation Frequency ........ 132
28 Particle Settling/Suspension Regimes 133
xiii
-------
TABLE OF CONTENTS (Concluded)
LIST OF FIGURES (Concluded)
No. Title
A-l Soil Erodibility as a Function of Particle Size 146
A-2A Major Soil Types in Northeastern States. . 148
A-2B Major Soil Types in Southeastern States 149
A-2C Major Soil Types in the Northern Great Plains States . . . 150
A-2D Major Soil Types in the Southwestern States 151
A-2E Major Soil Types in the Western States 152
A-2F Legend for Soil Maps in Figures A-2A through A-2E 153
A-3 Determination of Surface Roughness Factor , . 155
A-4 Typical Monthly Climatic Factors for the U.S.. ...... 158
A-5 Effect of Field Length on Relative Emission Rate ..... 159
A-6 Effect of Vegetative Cover on Relative Emission Rate . . . 161
C-l Testing of Gravel Site Emissions (Site R2) 171
C-2 Testing of Agricultural Tilling Emissions ,, 172
xiv
-------
EXECUTIVE SUMMARY
This report presents the results of an extensive field testing pro-
gram which was conducted to determine emission factors for four categories
of fugitive dust sources:
1. Unpaved roads and airstrips
2. Agricultural tilling
3. Construction sites
4. Aggregate storage piles
The testing was necessitated by the lack of reliable data on the char-
acteristics of these sources.
Special dust sampling techniques and associated data reduction
schemes were developed to quantify emissions from moving and stationary
fugitive dust sources. The two basic plume sampling techniques were
isokinetic exposure profiling and conventional high-volume sampling with
wind direction activators. The effective dust cut-off diameter for the
standard high-volume sampler was found to be about 30 urn.
During each field test, source activity and meteorological conditions
were continuously monitored. In addition, samples of the emitting surface
material were collected for laboratory analysis.
Test sites were concentrated in the dust bowl area of the Great Plains.
However, emissions from aggregate storage piles were tested in the
Cincinnati and Kansas City areas.
For each source type, the observed relationship between emission rate
and source activity was used to derive a basic emission factor. In addi-
tion, test data were analyzed to determine the dependence of the emission
rate on properties of the emitting surfaces and characteristics of the
vehicle or implement which caused the emissions.
xv
-------
The corrected emission factors which were developed for each source
category and the associated particle size breakdowns are presented In the
following paragraphs.
UNPAVED ROADS
The equation for estimating the total amount of road dust emissions
with drift potential greater than 25 ft, i.e., particles smaller than
100 um in diameter, is as follows:
e(roads) = 0.81 s (S/30)
where e = emission factor (pounds per vehicle-mile)
s = silt content of road surface material (percent)
S = average vehicle speed (miles per hour)
The precision of this equation in predicting the results of the emission
tests of unpaved roads is ± 10%,
The aggregate silt* content (i.e., particles smaller than 75 pai in
diameter) of the road surface is determined by measuring the amount of
loose (dry) surface dust which passes a 200 mesh screen. The silt content
of gravel roads is approximately 12%.
The above equation applies to "dry" days. Emissions are assumed to
be negligible on days with rainfall exceeding 0.01 in.
The test results indicate that, on the average, dust emissions from
unpaved roads have the following particle size characteristics;
Particle Diameter Weight Percent
< 2 urn 25
2 urn - 30 urn 35
30 pm - 100 urn 40
AGRICULTURAL TILLING
The equation for estimating the total amount of tillage dust emissions
with drift potential greater than 25 ft, i.e., particles smaller than 75 pm
in diameter, is as follows:
* As defined by American Association of State Highway Officials,
xvx
-------
_ 1.4 s (S/5.5)
e(tilling) " (PE/50)2
where e = emission factor (pounds per acre)
s - silt content of surface soil (percent)
S = implement speed (miles per hour)
PE = Thornthwaite's precipitation-evaporation index
(corrected for irrigation, if any)
The precision of this equation in predicting the results of the emission
tests of agricultural tilling is ± 157o.
The soil silt* content (i.e., particles between 50 urn and 2 urn in
diameter) may be determined by the Buoyocous hydrometer method. Surface
soil samples should be extracted with a plugging device to a depth of 4 in.
The test results indicate that, on the average, dust emissions from
agricultural tilling have the following particle size characteristics:
Particle Diameter Weight Percent
< 2 um 35
2 um - 30 um 45
> 30 um 20
AGGREGATE STORAGE PILES
The corrected emission factor for estimating the total amount of dust
emissions with drift potential greater than 1,000 ft, i.e., particles
smaller than 30 u in diameter, is given by the following expression:
e(aggregate) =
(PE/100)2
where e = emission factor (pounds per ton placed in storage)
PE = Thornthwaite's precipitation-evaporation index
Total dust emissions from aggregate storage piles can be divided into
the contributions of several distinct source activities which occur within
the storage cycle:
* As defined by U.S. Department of Agriculture.
xvi i
-------
Loading of aggregate onto storage piles (12%)
. Equipment traffic in storage area (40%)
. Wind erosion (33%)
Loadout of aggregate for shipment (15%)
The numbers in parentheses are the relative contributions of each
activity to the total emissions.
CONSTRUCTION SITES
The emission factor for medium-type construction activities (e.g.,
townhouses, shopping center) averaged about 1.2 tons/acre/month. However,
because of the use of water for dust control and interferences from other
dust sources in the vicinity of the test sites, correlations between
emission rate and potential correction parameters could not be estab-
lished. There was strong evidence that the level of activity could change
emissions by a factor of two or more.
The probable correction parameters for construction emissions are
(1) soil silt content and (2) surface moisture and level of activity.
The value reported above is thought to be fairly representative of un-
controlled emissions in areas less arid (PE ~ 50) than the Arizona-Nevada
test sites, but having a similar soil silt content (~ 30%). Approximately
407o of the dust emissions are smaller than 3 pm.
xviii
-------
CHAPTER 1
INTRODUCTION
This report presents the results of a program conducted by Midwest
Research Institute to develop emission factors for estimating atmospheric
emissions from certain common sources of fugitive dust.* The source
categories studied were:
Unpaved roads and airstrips
Agricultural tilling
Construction sites
Aggregate storage sites
In this chapter, the background of the fugitive dust problem is re-
viewed, and the objectives of the investigative program are stated.
BACKGROUND
Natural dust, commonly termed "dust rise by wind," is a major source
of global aerosol, accounting for as much as 20% of the total yearly pro-
duction. J:/ Recent studies have demonstrated that fine particles of soil
and minerals drift for thousands of miles on high altitude wind currents t
On a regional scale, sources of natural dust have been associated
primarily with the background particulate matter in the ambient air. The
occurrence of high background dust loadings during periods of dry, windy
weather has supported the widely held contention that the generation of
natural dust is an uncontrollable climatic phenomenon. Except for major
wind erosion damage to croplands, the effects of natural dust emissions
have often been viewed as relatively inconsequential.
Fugitive emissions are defined as pollutant emissions which are not
confined in process streams.
-------
In recent years, however, with the development of the national
effort to abate air pollution, the public has become more discerning
about the differences between a purely natural dust generation process
and the generation of "natural" dust resulting from the anthropogenic
disturbance of a surface exposed to the air environment. For example,
when the land is stripped of vegetation in preparation for a construc-
tion project, the enhanced vulnerability to wind erosion is no longer
viewed as a natural phenomenon. Likewise the generation of soil and
rock dust by vehicular traffic on an unpaved road is recognized as a
man-made source of air pollution.
Nevertheless, the problems of localized fallout of atmospheric dust
in the vicinity of common fugitive dust sources still draw markedly
different reaction from different segments of the population. In rural
areas the dust fallout from unpaved roads and agricultural tilling is
normally accepted by local residents as a nuisance which can be tolerated.
However, in the larger population centers, dust fallout from mineral
mining, processing and storage operations is often decried as an intoler-
able nuisance and a potential health hazard.
Recently, the development of State Implementation Plans to achieve the
national ambient air quality standards for suspended particulates, has
revealed that fugitive dust sources (including strictly natural sources) in
many areas of the country, both urban and rural, may have a much more sub-
stantial impact than once thought. In addition to large dust particles which
settle out near the source and cause the nuisance problem, fine particles
are also emitted and dispersed over much greater distances from the source.
Although common sources of fugitive dust generally have not been regarded
as serious air pollution problems, the cumulative effect of widely scattered
emissions in many areas has been suggested as a major cause of noncompliance
with air quality standards.
For the source categories treated in this report, there are two basic
mechanisms of dust generation by disturbance of exposed surface material:
1. Pulverization and abrasion of surface material by application of force
through implements (wheels, blades, etc.)
2. Entrainment of dust particles by the action of turbulent air currents.
The characteristics of dust generation for each source type will be dis-
cussed briefly in the following paragraphs.
Unpaved Roads and Air Strips
Unpaved roads are the most common transportation surface in the rural
areas of the country. Dust plumes trailing behind vehicles are a common
sight in these areas.
2
-------
When a vehicle travels over an unpaved road, the forces of the wheels
on the road surface cause pulverization of surface material. Particles are
lifted and dropped from the rolling wheels and the road surface is exposed
to strong air currents in turbulent shear with the surface. The turbulent
wake behind the vehicle continues to act on the road surface after the
vehicle has passed.
Unpaved airstrips are also common to the rural areas of the country.
Emissions from unpaved airstrips are caused almost entirely by the
turbulent wake generated by the propulsion systems.
Agricultural Tilling
The two universal objectives of agricultural tilling are the creation
of the desired soil structure to be used as the crop seedbed and the
eradication of weeds. A desirable soil structure is one in which large
pores extend from the surface to the water table or drains; this structure
helps to provide the right proportion of air and water for plant roots to
absorb nutrients from the soil. Plowing, the most common method of
tillage, consists of some form of cutting loose, granulating, and inverting
the soil and turning under the organic litter. Sweeps or undercutters
which loosen the soil and cut off the weeds but leave the surface trash in
place, have recently become more popular for tilling in dryland farming
areas.
During a tilling operation, dust particles from the loosening and
pulverization of the soil are injected into the atmosphere as the soil
is dropped to the surface. Dust emissions are greatest when the soil is
dry and during final seedbed preparation.
Aggregate Storage Piles
An inherent part of the operation of plants that utilize minerals in
aggregate form is the maintenance of outdoor storage piles. Storage piles
are usually left uncovered, partially because of the necessity for fre-
quent transfer of material into or out of storage.
Dust emissions occur at several points in the storage cycle—during
loading of material onto the pile, whenever the pile is acted on by strong
wind currents, and during loadout of material from the pile. The truck
and loading equipment traffic in the storage pile area is also a sub-
stantial source of dust emissions.
When freshly processed aggregate is loaded onto a storage pile, its
potential for dust emissions is at a maximum. Fines are easily disaggre-
gated and released to the atmosphere upon exposure to air currents re-
sulting from aggregate transfer or high winds.
-------
As the aggregate weathers, however, the potential for dust emissions
is greatly reduced. Moisture causes aggregation and cementation of fines
to the surfaces of large particles. A significant rainfall soaks the
interior of the pile and the drying process is very slow.
Construction Sites
Heavy construction is a source of dust emissions which may have sub-
stantial temporary impact on air quality. Building and road construction
are the prevalent construction categories with the highest emissions
potential.
Emissions are generated by a wide variety of operations over the
duration of the construction of a building or road. These include land
clearing, blasting, ground excavation, cut and fill operations, and the
construction of the facility itself. Dust emissions vary substantially
from day to day depending on the level of activity, the specific opera-
tions and the prevailing weather. A large portion of the emissions result
from the equipment traffic over temporary roads at the construction site.
In all of the above cases, dust generation from a mechanical contact
process with the exposed surface is insensitive to the ambient wind speed.
However, the wind speed does determine the drift distance of large dust
particles and, therefore, the localized impact of the fugitive dust
source.
On the other hand, the generation of suspended particulates by wind
erosion of exposed surface is very sensitive to the wind speed. The total
surface removal by wind erosion, which consists mostly of transport of
large particles close to the ground, depends on the cube of the wind
speed above a threshold value of about 12 mph.ft/
OBJECTIVES
The principal objective of the investigation reported herein was the
development of emission factors for estimating atmospheric dust emissions
from the source categories listed above. In each case, the emission fac-
tors were to incorporate correction factors to account for major varia-
tions in emissions with source conditions. Correction factors would in-
clude the effect of geographical differences in surface properties and
climate. An attendant objective was the development of field testing
procedures for measurement of dust emission rate and the particle size
distribution of suspended dust.
-------
This report is organized by subject area as follows:
. Chapter 2 presents a summary of the published literature dealing with
quantitative studies of fugitive dust emissions.
Chapter 3 outlines the plume sampling techniques and the data reduction
schemes used to derive emission factors.
Chapters 4-7 present for the four source categories, a complete,
self-contained discussion of the field testing and the calculated test
results, and conclude with the presentation of the corrected emission
factor.
Chapter 8 discusses the development of an emissions inventory for the
specified source categories.
. Chapter 9 states the conclusions of this investigation.
-------
CHAPTER 2
SUMMARY OF PERTINENT LITERATURE
The literature search conducted as part of this program yielded only
scattered quantitative information on the characteristics of fugitive dust
sources. Most of the reported studies were directed to the characteriza-
tion of dust generation from unpaved roads. Measurement of suspended
particulate levels (by standard high-volume filtration) in the vicinity
of a fugitive dust source has been the most commonly used technique for
quantification of the source impact.
EMISSIONS FROM DIRT ROADS
In an early study by the Albuquerque Air Management Division,—'
dust emissions from a dirt road in Bernallilo County were measured. A
small filtration sampler was positioned first at the edge of the road
and then directly behind the test car which traveled at 30 mph. The
measured concentrations, coupled with assumptions about the configuration
of the plume, yielded an emission factor in the range of 0.5-0.7 Ib/vehicle-
mile.
The first effort to measure the particle size of dust emissions from
an unpaved road was conducted by engineering students at the University of
New Mexico,^' at a site just north of the Albuquerque campus. A standard
high-volume filtration unit and a rotorod impactor were positioned 60-90
ft downwind of the test road. During each 30-min test, a total of 50 passes
of the (two) test cars were sampled. Meteorological data for the test
period were obtained from the local weather bureau. Background dust levels
were determined by sampling with no traffic on the road. Particle size
distribution was determined by microscopic examination of rotorod impaction
samples. Emission factors were calculated from test results by applying
a dispersion equation to account for expansion of the dust cloud from the
point of generation. The factor for particles smaller than 6 pm in
diameter (i.e., particles which would remain suspended under dry, windy
conditions) was 0.93 Ib/vehicle-mile.
-------
A detailed study of emissions from dirt roads was conducted by
PEDCo-EnvironmentalZ/on a test roadway near Santa Fe, New Mexico, and at
two sites in Tucson, Arizona. At the primary test site in Santa Fe,
a GCA beta-gauge detector was used to measure vertical concentration
profiles at distances of 50-300 ft downwind of the road during each of
six 1-hr tests. Standard high-volume filtration samplers were also
operated at downwind locations during each test, to provide a basis for
correcting the measurement of the beta-gauge detector to an equivalent
high-volume measurement. The high-volume readings averaged 1.68 times
the beta-gauge measurements with a correlation coefficient of 0.87.
Several test vehicles were used to provide between 100 and 200 passes
per test. A recording wind instrument was operated near the site.
Emission factors were calculated from corrected beta-gauge measurements
and meteorological conditions, through the application of a dispersion
equation for an infinite line source. The results are given in Table 1.
In addition to the intensive beta-gauge study, longer term (24- and
48-hr) high-volume dust samples were collected over a period of 2 months.
Andersen high-volume cascade impactors were operated during 48-hr periods
to measure particle size distribution of suspended dust. The purpose of
this longer term study was to measure the impact of normal road traffic
and, in particular, to determine the contribution of traffic dust emissions
to the total suspended dust level in the vicinity of the test road.
Identical high-volume measurements were conducted during the same
2-month period at the two test sites in Tucson, Arizona. Application of
the dispersion formulae to data from the Tucson sites for days when the
wind conditions were fairly constant, yielded apparent emission factors
(scaled against the traffic load) ranging from 4-6 lb/vehicle-mile.
Taking into account the contributions of background dust and the low-
level of wind erosion from the test roadways, the investigators concluded
that there was substantial uniformity in emission rates from the three
roads, in spite of differences in geographical location and traffic
patterns.
The results of the particle size measurements for the three PEDCo sites
were as follows:
Suspended Particulate
Mass < 3.3 um (%)
Santa Fe 48
Tucson A 37
Tucson B 36
-------
Table 1. TESTS OF EMISSIONS FROM UNPAVED ROADS
00
Vehicle
Type of Speed
Site Road (mph)
Bernalillo County, Dirt 30
New Mexico!/
University of Dirt 25
New Mexico6-/
Santa Fe, Dirt 15
New Mexico!/ 25
35
40
Poweshiek County, Dirt
Iowa!/
Duwamish Valley, Gravel 10
Washington!/
20
30
Gravel 20
Sampler No. of Passes
Type Location Tests per Test
Small filter In plume 2
Hi-vol filter 60-90 ft 2 50
Rotorod from road
Beta gauge 50-300 ft 1 150
Hi-vol filters from road 1 240
Hi-vol cascade 3 200
impactor 1 130
Dustfall con- Shoulder to 1 3,000
tainers 500 ft from
road
Isokinetic 7 ft behind 2
cascade automobile
impactor
25
2 -™
1
Emission
Factor
Hb/vehic le-mi le)
0.5 - 0.7
0.93
0.04
0.67
1.0
2.0
3.5
5.5
2.2
0.41
0.11
8.5
2.3
0.29
13.9
5.2
0.43
8.8
2.4
Dust Size
Cut-Off
__
< 6 pm
< 3 urn
_.
—
--
—
< 10 pm
< 2 pm
...
< 10 urn
< 2 urn
,-
< 10 pm
< 2 um
__
< 10 pm
-------
Recently, Hoover.2/ reported the results of the measurement of dust
deposition near the edge of a test gravel road in Poweshiek County, Iowa.
Dustfall collectors were positioned 3 ft above the ground and at dis-
tances (along a line perpendicular to the test road) ranging from 12 ft
(shoulder) to 500 ft from the center line of the road. The containers
were left in place for 21 days. Based on the amount of dust which
settled within 500 ft of the road, the calculated emission rate was
5.5 Ib/vehicle-mile. The results at the primary test site were confirmed
by the results at a site near Iowa State University in Ames.
EMISSIONS FROM GRAVEL ROADS
A definitive study of emissions from gravel roads was conducted by
the Puget Sound Air Pollution Control Agency^/ on test roads in Seattle's
Duwamish Valley. The primary sampling device was a University of
Washington Mark II Cascade Impactor, which separated the particulate
catch into size fractions. The impactor was operated isokinetically at
successive grid points (5-10 min per point) on a rack which was towed
behind the test car. Twenty-five tests were conducted for a vehicle
speed of 20 mph, with an average dust concentration of 370 mg/m^ in the
plume; tests were also run at 10 and 30 mph. The test results are shown
in Table 1. As indicated, the total emissions factor and the size dis-
tribution for the two gravel roads tested at 20 mph are nearly identical.
Also worthy of mention is Sehinel's studyl£/ of particle resuspension
from an asphalt road caused by car and truck traffic. Solid zinc sulfide,
which was used as the tracer material, was applied to the 10-ft wide by
100-ft long area on one lane of a two-lane seasoned asphalt road. Filtra-
tion samplers (nonisokinetic), mounted on 8-ft towers, and ground-level
deposition samplers were positioned in an array at distances of 3.5-100 ft
downwind from the edge of the test area. A meteorological tower with a
vector vane at 3-ft elevation and 3-cup anemometers at 1- and 7-ft eleva-
tions, was also operated downwind of the road.
The fraction of the tracer dust resuspended from the road per vehicle
pass was calculated from a graphical integration of the downwind airborne
tracer exposure and the tracer ground deposition. The mass balances were
accurate within a factor of three. The following significant results
were obtained:
1. The resuspension rate increased as the square of the vehicle speed and
was independent of wind velocity.
2. Twenty to thirty percent of the particulate mass resuspended was
deposited on the ground within 20 to 30 ft of the road.
-------
3. The relative deposition rate passed through a minimum for a vehicle
speed of 30 mph.
EMISSIONS FROM AGRICULTURAL AND CONSTRUCTION ACTIVITIES
The only available data on dust emissions from agricultural and con-
struction activities were generated in the study, mentioned above, by
PEDCo-Environmental.Z/ At agricultural sites in Five Points, California,
and Mesa, Arizona, standard high-volume filtration samplers were operated
for a period of 2 months downwind (based on prevailing wind direction) of
the test sites. Atmospheric dispersion formulae were used to calculate
emission factors from the measured increase in particulate concentration
(downwind minus upwind value) for selected days when the wind direction
matched the alignment of the samplers. It was assumed that each sampler
measured emissions from a test area of about 500 acres. The resulting
factors, which were judged to be strongly affected by wind erosion emis-
sions, ranged from about 1-2 tons/acre/year.
PEDCo's tests of emissions from residential construction activities!/
will be discussed thoroughly in Chapter 6.
No quantitative data were found for dust emissions from aggregate
storage piles. An estimated value of 10 Ib/ton for storage pile losses
has been reported.!!/
10
-------
CHAPTER 3
DUST EMISSION SAMPLING STRATEGY
This chapter summarizes the sampling strategy which was utilized for
each source type. In particular, the dust emission sampling techniques
are described and the schemes for calculating source emission rates from
the field measurements are presented.
EMISSIONS FROM AGRICULTURAL TILLING AND UNPAVED ROADS
An agricultural implement tilling a field or a vehicle traveling
an unpaved road may be treated as a moving point source which emits dust
at a relatively constant rate. If the mean wind direction is roughly
perpendicular to the path of motion of the point source, the dust plume
drifts laterally as shown in Figure 1. As the plume is convected by the
mean wind, atmospheric turbulence effectively disperses fine particles
(and, to a lesser extent, moderate-sized particles) over an increasing cross-
sectional area. The large particles settle to the ground as a result of
the dominance of gravitational and inertial forces over turbulent mixing
forces.
Since there is no net transport of dust in the direction of equip-
ment motion, the settled and airborne dust within an incremental length
in the direction of source motion directly represents what was emitted
by an equivalent length of disturbed surface. This may be expressed as
a mass balance which traces the fate of the dust emissions.
In the case of emissions from an agricultural tilling operation, the
mass balance per unit length of tillage path is as follows:
Dust generated by _ Dust Integrated atmospheric
N implement passes* deposition exposure ,
* Over adjacent strips of land.
11
-------
Wind
Dust
Sampler
Movi'ng
Vehicle
Figure 1. Overhead view of dust plume from moving point source.
12
-------
or
XP
eabN = D(x)dx + r
- o JQ
where ea = agricultural dust emission factor (mass/area),
b = working width of implement (length),
D = dust deposition (mass/length squared),
x = distance downwind from the source (length),
x = location of exposure sampler (length),
m = dust catch by exposure sampler after subtraction of
background contribution, measured at x^ (mass),
a = intake area of exposure sampler (length squared), and
h = height above ground (length) .
The exposure — is the integrated passage of airborne dust per area normal
to the direction of passage. The background contribution to the exposure
is given by
mb = QCb ,
where Q = volume of air sampled (length cubed) , and
Cb = background dust concentration measured upwind of
the source (mass/length cubed).
In the case of emissions from an unpaved road, the mass balance per
unit length of road is as follows:
Dust generated by _ Dust Integrated atmospheric
N vehicle passes deposition exposure
erN = ,(
where e = road dust emission factor (mass/ length-vehicle) , and the
other symbols are as defined above.
In order to collect a representative sample of airborne particulate,
the sampling rate must be isokinetic; that is, the streamlines, along
which the air flows as it passes into the sampler, must be rectilinear.
Two requirements must be met to achieve isokinesis:
1. The magnitude of the sampling velocity must equal the local mean wind
speed; and
2. The sampling intake must be perpendicular to the wind vector.
13
-------
Near the surface, the mean wind speed has been found to increase in
proportion to the logarithm of the height.
I) = ku^ln(h/h0),
where k = von Karman's constant (0.4 for clear fluids),
u. = friction velocity, and
75°
h = apparent roughness height.
The roughness height of a plowed field is approximately equal to 1 cm..—'
The wind speed profile over a larger vertical range may also be
expressed as a power law,
u -
where Uj_ = wind speed at reference height h.. , and
n = 0.2 for daytime conditions..!!/
Using hj_ = 12 ft as the reference height, the above expressions
may be rewritten as follows:
log law power law
U In (336 h) U /h \°'2
U12 5*90 U12 12
As shown in Figure 2, over the range of height utilized for exposure
sampling (3 ft £ h < 12 ft), the two expressions agree to within about 1%,
If the sampling is nonisokinetic by virtue of the failure to meet
condition 1 above, corrections must be made to the nonisokinetic par-
ticulate catch n.
m = %'o fine Particles (d < 5 Jim)
m = nijj coarse particles (d > 50 um) ,
where U = the local wind approach speed and
u = the magnitude of the sampling velocity at the sampler intake.
14
-------
— — Power-Law Profile
Log- Law Profile
Reference Wind Speed
Dust Exposure
Dust Exposure
fParticIe Size
LHi-Vcl Concentration
Dust Exposure
Dust Exposure
Saltation
Deposition
0.2 0.4 0.6 0.8
WIND SPEED/REFERENCE WIND SPEED
Figure 2. Comparison of wind speed profiles.
15
-------
For intermediate-sized particles,
m =
2*1 /
where Fj = |j = the isokinetic ratio.
The above connections for nonisokinesis were derived from the correc-
tion for nonisokinetic particulate concentration presented in the Federal
Register-M' and the basic relationship between exposure and concentration (C)
™ = CU
a
Most conventional samplers for airborne participates (e.g., the
high-volume filtration sampler) are nondirectional with sampling intakes
usually aimed downward. While particles smaller than about 10 um in
diameter are readily drawn into the sampler, particles larger than about
50 um (for moderate wind speed) are sampled with very low efficiency.
Consequently, the large particle mode (> 3 pn diameter) of the typical
bimodal size distribution of atmospheric particulateii/ is largely missed,
even though it may comprise more than half of the total mass in an area
influenced by sources of dispersion* particulate aerosol (e.g., soil and
mineral particles).
Since most of the mass of the particles emitted by agricultural
tilling and unpaved roads would fall into the large particle mode, con-
ventional samplers were judged to be less suitable than isokinetic sam-
plers for the subject program.
The exposure profiling unit which was designed for this study is
pictured in Figure 3. It consists of a vertical array of isokinetic
high-volume filtration devices attached to a mobile support tower. Each
sampler accommodates an 8-in. x 10-in. glass fiber filter (Type E). The
reduced sampling intake area (2 in. x 2 in.) increases the allowable wind
speed maximum for isokinetic sampling to 20 mph. Flexible hose (4-in.
diameter) connects each sampler to a suction manifold. Each leg of the
manifold is fitted with a calibrated orifice (connected to 0-1 in w.c.
inclined manometer) and a butterfly valve for flow control. The vacuum
source is a 2-hp centrifugal blower. Electrical power is supplied by a
gasoline-engine generator.
The exposure profiling tower was positioned close enough to the
source to measure the vertical extent of the plume (by reasonable ex-
trapolation) , but far enough downwind from the source to allow for ade-
quate plume development prior to sampling. The minimum acceptable plume
* Generated by mechanical forces,
16
-------
WIND
Figure 3. MRI dust exposure profiler.
-------
travel distance from the downwind edge of the source was judged to be about
20 ft. (In the case of agricultural tilling, the source-to-sampler dis-
tance was maintained by advancing the profiling tower downwind between
tillage implement passes.) Since dust-producing conditions were fairly
uniform along the emitting surface, the specific sampling location in the
direction of source motion was not critical.
Dust deposition was measured by standard 1-ft high dustfall buckets,
which were positioned downwind of the source along a line perpendicular
to the direction of source motion. The deposition samplers may also have
collected some particles transported by saltation.
Sand-sized dust particles injected into the atmosphere by a tilling
operation or by a vehicle traveling an unpaved road may be transported
by "saltation"* over substantial distances if the wind velocity exceeds the
wind erosion threshold. Since these particles are never truly suspended
in the atmosphere, they are not considered part of the atmospheric dust
emissions from a fugitive dust source. Nevertheless, limited measurements
of saltation dust would yield useful information on the magnitude of
saltation transport relative to suspended dust transport and saltation
transport by wind erosion.
The saltation catcher which was designed and fabricated for this
study, consisted of a dustfall bucket fitted with an 18-in. high sheet
metal tube with a 1-in. wide vertical sampling slot. The slot is pointed
upwind and captures saltating particles within the height interval o£
12-30 in. The capture efficiency is estimated to be about 50%.!§/
The Andersen high-volume cascade impactor was selected as the primary
device for suspended dust particle sizing. The impactor is designed to be
attached to a standard high-volume sampler. It has five glass fiber im-
paction surfaces, followed by a glass fiber back-up filter. A sampling
height of 6 ft was chosen to represent average plume conditions and to
correspond to the ground level breathing zone.
The standard high-volume filtration unitiZ' was selected for measure-
ment of background (upwind) dust concentration. The 3-ft sampling height
is above the saltation zone and should, in the absence of wind erosion,
trap most of the background particulate. Limited downwind measurements
of suspended dust by standard high-volume filtration were also included
in the experimental design as a. check on the large-particle trapping
efficiency of the standard high-volume sampler.
Saltation is particle motion by a series of jumps
18
-------
EMISSIONS FROM AGGREGATE STORAGE PILES
A distribution of aggregate storage piles with associated truck
traffic and transfer operations is a diffuse area source. Emissions vary
substantially from day to day because of variations in consumer demand.
Therefore, emissions must be sampled over a widespread area for a period
of several days. Because of changes in wind speed and direction over
extended time periods, isokinetic sampling of aggregate storage emissions
is a virtual impossibility.
Standard high-volume filtration unitsiZ' with wind-direction activators
were selected as most suitable for sampling of diffuse aggregate storage
operations. Sampling units were strategically positioned so that when
the wind had a nonzero component in the prevailing wind direction for the
locality (a condition for activation of the samplers), one unit was upwind
of the storage area and the others were distributed downwind of the
storage area. Emissions were calculated from the measured average down-
wind flux (average concentration multiplied by atmospheric ventilation rate)
of aggregate dust, over an assumed cross-sectional transport area.
The greatest intensity of dust emissions in the aggregate storage
cycle occurs during the transfer of material onto the stockpiles and the
loadout of material from stockpiles into trucks.
In order to measure the dust emission rate from the loadout operation,
a special sampling apparatus was designed and constructed. This apparatus,
shown in Figure 4, consisted of a grid of six samplers mounted on top of
a mobile van and controlled by auxiliary equipment inside the van.
Dust-laden air passes into the intake nozzle (1/2-in. diameter by
4 in. long) of each sampler and through the dust collection mediutn--a
circular glass fiber filter (2-in. diameter). The filtered air then
passes through a matched critical orifice, common manifold, vacuum pump,
and dry test meter. Sampling rates were preset to be isokinetic for a
10-mph wind speed. The dry test meter provided a check on the total sam-
ple volume. Electrical power was supplied by a generator located on top
of the van.
During testing, the sampling van was positioned downwind of the
truck being loaded, as shown in Figure 5. The dust, which was generated
when the high loader dumped into the truck, passed across the sampling
grid.
In the case of emissions from aggregate loadout, the mass balance
(neglecting deposition) is given by:
19
-------
SUPPORT FRAME
NJ
O
SUPPORT FRAME
BRACE
SAMPLER-
r
79"
SUPPORT
FRAME
BRACE
-JOS' —
72"
-•-SUPPORT FRAME
Figure 4. Dust exposure profiler for elevated emissions source.
-------
t
AGGREGATE STORAGE PILE
/^"
HIGH LOADER
WIND-
DUMP TRUCK
•SAMPLING GRID
SAMPLING VAN
Figure 5. Positioning of test equipment—aggregate loadout.
-------
Dust generated by _ Integrated atmospheric
aggregate loaded exposure
or
f. CO ,. CO
TT I I m(h.w) ,, ,
e_W = / / ' i" /dh dw
P J - CD JO a
where e = loadout dust emission factor (mass/weight loaded)
W = weight of aggregate loaded (mass)
w = lateral distance from center-line of truck (length)
and the other symbols are defined as above.
22
-------
CHAPTER 4
UNPAVED ROAD EMISSIONS
SAMPLING SITE DESCRIPTION
Franklin County, Kansas, was selected for the study of atmospheric
dust emissions from gravel roads; Morton and Wallace counties in Kansas,
were selected for the study of emissions from dirt roads. The test roads
were chosen on the basis of their representativeness of unpaved roads in
the dry, windy area of the Great Plains.
Detailed descriptions of the individual test sites are given in the
following paragraphs.
Gravel Road Sites
Two sites in Franklin County, Kansas, were selected for the study of
atmospheric dust emissions from gravel roads. Franklin County is located
in the east-central part of the state.
Site Rl was a lightly traveled section of east-west road located about
1 mile east of Williamsburg, Kansas; this road was covered with a con-
siderable amount of loose gravel. Site R2 was a section of north-south
county road located just north of a nearly completed section of Interstate
35; this road was well worn, with little loose gravel.
Dirt Road Sites
Two sites were selected for the study of atmospheric dust emissions
from dirt roads--one in Morton County, Kansas, and the other in Wallace
County, Kansas.
Site R3 was a section of east-west county road located in Morton
County between T35S, R42W, Section 2, and T34S, R42W, Section 35. The
soil type in the area was Richfield fine sandy loam. This road, although
lightly traveled, had a large proportion of heavy truck traffic.
23
-------
Site R4 was a section of north-south road located in Wallace County
between T13S, R40W, Section 31, and T13S, R41W, Section 36. The soil
in the area of this lightly traveled road was of the Keith/Colby silt
loam association.
FIELD MEASUREMENTS
Field testing of dust emissions from unpaved roads was conducted at
the Franklin County sites (Rl and R2) in April 1973, and at the Morton
and Wallace counties sites (R3 and R4) in May and June 1973.
Table 2 specifies the kinds and frequencies of field measurements
that were conducted during each run. "Composite" samples denote a mix-
ture of single samples taken from several locations in the area; "inte-
grated" samples are those taken at one location for the duration of the
run.
Composite samples of in-place road dust were obtained by manually
sweeping the loose material from lateral strips of road surface into
plastic bags. Samples were returned to MRI for laboratory determination
of texture and moisture content.
At the end of each run, the collected samples of dust emissions were
carefully transferred to shipping containers within the MRI instrument
van, to prevent dust losses. High-volume filters (from the MRI exposure
profiler and from standard high-volume units) were folded and placed in
individual folders. Dust that collected on the interior surfaces of each
exposure probe was rinsed with distilled water into a glass jar. The con-
tents of the deposition samplers were also rinsed into glass jars. Cas-
cade impactor collection papers were left in place within each impactor
unit.
Most of the traffic volume for each run was provided by local resi-
dents who were hired to drive their own vehicles at the prescribed speed
over a 1/2-mile section of test road. Vehicle spacing was maintained
to eliminate possible vehicle interaction effects on dust generation.
As indicated in Table 3, all of the test vehicles were four-wheel
vehicles--either passenger cars or pick-up trucks.
Table 4 presents information on the time of each run, the prevailing
meteorological conditions and the vehicular traffic. Over the typical
1-hour test duration, meteorological conditions and traffic characteristics
did not vary significantly.
24
-------
Table 2. FIELD MEASUREMENTS —UNPAVED ROADS
Test Parameter
Units
Sampling Mode
K5
Meteorology
a. Wind speed mph
b. Wind direction deg
c. Cloud cover 7,
d. Temperature °F
e. Relative humidity 7,
Road Surface
a. Type
b. Texture
c. Moisture content 7>
d. Embankments
Vehicular Traffic
a. Type
b. Count
Continuous
Continuous
Single
Single
Single
Composite
Composite
Composite
Composite
Multiple
Cumulative
pm
pg/
4, Suspended Dust
a. Exposure mg/in'1
(vs height)
b. Size distribution
(by weight)
c. Concentration Pg/m"
d. Background ug/m-'
concentration
e. Duration of sampling min
5, Deposition (vs distance Ib/ft^/hr
from source)
Integrated
Integrated
Integrated
Integrated
Cumulative
Integrated
Measurement Method
Recording instrument at "background"
station; sensors at reference height
Visual observation
Sling psychrometer
Observation (photographs)
Dry sieving
Weight loss on oven drying
Observation (photographs)
Observation (car, truck, number of
axles, etc.)
Observation
Isokinetic high-volume filtration
(MRI method)
Cascade Impact ion
High-volume filtration (EPA methodiZ/)
High-volume filtration (EPA method!!/)
Timing
Dust fall buckets (ASTM method—/)
-------
Table 3. VEHICLE MIX (Unpaved Roads)
to
No. of
Vehicle Passes
Normal Traffic
No. of Test
Run
1
2
3
4
8
10
13
Site
Rl
R2
R2
Rl
R3
R3
R4
Vehicles
3
5
5
3
4
5
1
Description of Test Vehicles
2 cars, 1 truck
3 cars, 2 trucks (one with camper)
3 cars, 2 trucks (one with camper)
3 cars
3 cars, 1 pick-up truck
3 cars, 2 pick-up trucks
1 car
Test Vehicles
Cars
108
112
144
148
109
71
51
Trucks
54
91
109
0
42
60
0
Total
162
203
253
148
151
131
51
Cars
3
7
11
—
0
0
1
Trucks
4-Wheel
1
7
7
--
3
0
3
Other
2
6
2
12
13
1
0
Total
6
20
20
12
16
1
4
% 4 -Wheel
Vehicles
98.8
97.3
99.1
>92.0
92,2
91.7
100.0
-------
Table 4. EMISSIONS TEST PARAMETERS (Unpaved Roads)
Time
Run
1
2
3
4
8
10
13
Site
Rl
R2
R2
Rl
R3
R3
R4
Date
4 April
5 April
5 April
7 April
25 May
6 June
12 June
1973
1973
1973
1973
1973
1973
1973
Start
1522
1411
1555
0945
1502
1141
1701
Finish
1622
1511
1655
1215
1602
1241
1801
Duration of
Exposure
Sampling
(rain)
60
60
60
150
60
60
60
Ambient
Road Temp,
Direction (°F)
E-W 50
N-S 60
N-S 63
E-W 46
E-W 85
E-W 75
N-S 75-^
Cloud
Wind Direction/ Cover Pasquill
Speed (12 ft) (1) Stability—
NNW/17 mph 75 D
SW/13 mph 0 C-D
SW/13 mph 0 C-D
N/15 mphk/ 100 D
S/19 mph 50 D
SW/10 mph-'' 0 B
NE/10 mph— 80-i D
Vehicle
Speed
' (mph)
30
30
40
30-35
30
40
30
No. of
Passes
168
223
273
160
167
132
55
a./ Pasquill Stability Classes:
.19/
A - Extremely unstable
B - Unstable
C - Slightly unstable
D - Neutral
E - Slightly stable
F - Stable to extremely stable
b/ Estimated value
-------
Table 5 gives the locations (intake height and distance from road)
of the various plume sampling devices that were used for each run. The
dust particle size classifiers included two types of high-volume cascade
impactors (Andersen and Sierra) operated within standard high-volume en-
closures. The drift distance multiplier, given in the last column of
the table, takes into account the effect of the horizontal wind-road
angle on the plume travel distance.
TEST RESULTS
Dust samples from the field tests were analyzed gravimetrically in
the laboratory. Filters were conditioned in a controlled temperature-
humidity environment prior to weighing. Water rinses from exposure
probes, deposition samplers and saltation catchers were evaporated on a
steam bath in tared beakers, after which the beakers were conditioned
and weighed.
The measured dust emission from the tests of unpaved roads are pre-
sented in Table 6. The dust quantities are the amounts generated per
vehicle-mile of travel.
The total dust emissions for a given run are the sum of the inte-
grated exposure (above the background exposure) and the amount of
deposition between the edge of the road and the downwind location of the
exposure profiler.
The suspended dust measurements used to compute the integrated ex-
posure are presented in Table 7. Point values of exposure are converted
to concentration. The concentration measured by the standard high-volume
unit, which was positioned to the side of the profiler, is also presented.
The exposure profiles are shown in Figure 6,
Through regression analysis of all of the deposition measurements,
the local deposition (scaled against the integrated exposure measurements)
was found to correlate best with plume travel time. The generalized de-
position distribution (vs travel time) exhibited a sharp decrease within
the first second of travel time followed by a gradual decay with in-
creasing travel time. Because no simple (two-parameter) mathematical ex-
pression described the abrupt change in the deposition distribution, it
was decided to treat only the gradual decay portion of the distribution.
28
-------
Table 5. DUST EMISSION SAMPLER LOCATIONS
(Unpaved Roads)
Perpendicular Distance from Downwind Edge of Unpaved Road (ft)
Run
1
2
3
4
8
10
13
Site
Rl
R2
R2
Rl
R3
R3
R4
MRI Particle Size Deposition Saltation
Exposure Classifier Sampler Catcher
Profiler3/ (h = 6 ft)b/ (h - 1 ft) (l<;h«;2.5 ft)
20 — 9
24 — 6.5, 13
24 « 6.5, 13
18 (SI)
20 ~- 8, 90-100 20
20 20 (SI, AI, AC) 10, 20, 50 20
20 20 (AI) 20 20
Standard Drift
Hi-Vol Sampler Distance
(h = 6 ft) Multiplier^/
1.1
24 1.4
24 1.4
1.0
20 1.0
1.4
1.4
a./ Sampling heights: h = 3, 5.5, 8, and 10.5 ft above grade.
b_/ Sizing device: SI = Sierra impaetor
AC — Aerotec cyclone
AI = Andersen impactor
c/ Ratio of drift distance to perpendicular distance.
-------
Table 6. MEASURED DUST EMISSIONS (Unpaved Roads)
Background Wind Plume Travel Time Unit Dust Catch (Ib /vehicle-mile) Hi-Vol
Run
1
2
3
4
8
10
13
Concentration Speed to Profiler
Site (pg/nP) (mphj (sec)
Rl 46.9 17 0.88
R2 38.5 13 1.8
R2 38.5 13 1.8
Rl « 15^
R3 102.0 19 0.72
R3 — 10^ 1.9
R4 — !(£/ 1.9
Integrated MMDJ/
Saltation Deposition Exposure (pm)
2.1 9.98
1.1 10.3
2.1 13.9
--
1.7 350 16.3
0.26 5.6 6.03 2.3
0.41 45.1 55.9 2.5
a/ Mass mean diameter of suspended dust, measured with Andersen high-volume cascade impactor.
b/ Estimated value.
-------
Table 7. PLUME SAMPLING DATA (Unpaved Roads)
Run
I-/
2£/
3fi/
8
10
13
a/
b/
c/
d/
Height
Site (ft)
Rl 10.5
8
5.5
3
R2 10.5
8
6J/
5.5
3
R2 10.5
8
&
5.5
3
R3 10.5
8
6k/
5.5
3
R3 10.5
8
6£/
6^/
5.5
3
R4 10.5
8
&£/
5.5
3
Sampling rate was
Sampling
Rate
(cfm)
29.0
27.5
26.0
24.1
24.1
22.7
49.3
21.4
19,3
24.1
22.7
46.5
21.4
19.3
35.7
34.5
43.0
32.2
28.2
24.1
22.8
20.0
38.9
21.3
19.2
24.3
23.2
20.5
21.5
19.2
corrected
Concentrat ion
(mg/m3)
0.90
3.33
7.20
8.13
2.82
6.60
6.53
10.8
18.4
3.66
10.4
9.50
18.1
30.9
2.65
4.81
1.37
9.08
21.9
1.94
3.29
2.74
2.31
4.10
8.27
4.61
9.20
8.61
16.4
28.0
for 80% isokinetic
Unit Exposure
(mg/in. /vehicle)
0.082
0.289
0.591
0.619
0.162
0.357
—
0.552
0.843
0.172
0.459
--
0.753
1.158
0.238
0.418
--
0.737
1.56
0.150
0.242
._
--
0.281
0.511
0.866
1.65
--
2.73
4.15
*
Standard high-volume sampler.
Andersen impactor.
Sierra impactor.
"
31
-------
w
O
J_
JL
Unpoved Roods
Run Site
O 2
D 3
A 8
• 10
• 13
Rl
R2
R2
R3
R3
R4
2 3
UNiT EXPOSURE (mg/in -vehicle)
Figure 6. Exposure profiles--unpaved roads.
-------
Deposition measurements for distances greater than 8-10 ft from the
road edge (i.e., beyond the high fallout strip adjacent to the road) were
fit to the function a exp(-j3t) where a and |3 are parameters and t is
the travel time. If only one deposition measurement were available, an
average value of p from the other runs was used and a new value of a
was determined.
The measurements of dust transport by saltation are shown only for
purposes of comparison. Saltation, which is confined to about 30 in. of
height, is not considered to be a form of atmospheric emissions. Also
it should be noted that the saltation catchers used in this study did
not sample below 12 in. above the ground.
Also given in Table 6 is the mass mean diameter of suspended dust
particles measured with the Andersen high-volume cascade impactor. The
diameter values are aerodynamic measures which treat particles as
equivalent spheres with a density of 2.5 gm/cm . The complete size
distributions are shown in Figure 7.
Two potentially significant sources of error in the particle size
measurements deserve special mention:
1. The impactor samples nonisokinetically through the high-volume en-
closure openings and captures large particles with low efficiency.
2. Unlike urban aerosol, road dust particles are dry and brittle and
are subject to bouncing and reentrainment from impaction surfaces. Recent
empirical evidence obtained by SehmeLrP' indicates that this effect is
most pronounced for particles larger than 20 pm in diameter.
Both of these factors cause apparent size determinations to be biased
in the direction of small diameter. The second factor seemed to be sub-
stantial with the Sierra slotted impactor (MMD » 1 lira); for this reason
the Sierra measurements were not used.
Table 8 gives the results of the laboratory analyses of the samples
of loose material from the road surface. Moisture content was determined
by weight loss on oven drying and particle size analysis by dry sieving.
The low moisture content of the surface material is indicative of
its tendency to dry quickly after the nighttime addition of moisture from
the road substrate.
The particle size analyses of the road surface samples indicate that
the well-worn gravel road (R2) had more sand-sized fines than the less-
traveled gravel road (Rl), but both had about the same percentage of silt.
The dirt road in Wallace County (R4) had a much larger percentage of silt
than the gravel roads.
33
-------
WEIGHT % GREATER THAN STATED SIZE
100.0
10.0
§ ,.0
0.1
r -
1
wtts
Andersen Results
£ (Dirt Roads) ill
~ A Run 10, Site R3 -
-- O Run 13, Site R4 ^Lp£
0.01 ' b
WEIGHT % LESS THAN STATED SIZE
Figure 7. Particle size distributions--dirt road emissions.
34
-------
Table 8. ROAD SURFACE PROPERTIES (Unpaved Roads)
Surface Texture (% by weight)-^
Run
1
2
3
4
8
10
13
Site
Rl
R2
R2
Rl
R3
R3
R4
Surface
Surface Type Moisture (%)
Gravel 3 . 8
Gravel 1.4
Gravel 1.4
Gravel
Dirt
Dirt - —
Dirt 3.2
Loose
Surface
Material
(lb/ft2)
1,4
1.0
1.0
1.4
2.2
Gravel
C>2000 urn)
38
29
29
38
__
—
12
Coarse Fine
Sand Sand
(2000- (420-
420 pm) 74 urn)
29 21
31 26
31 26
29 21
__
—
12 8
Silt
(<74 urn)
12
13
13
12
—
—
68
a/ Determined by dry sieving method; results accurate to within ± 5% of true value
(e.g., 12 ± 0.6).
-------
The size distributions for the road surface samples are plotted in
Figure 8. The samples from dirt road R4 was also analyzed for size dis-
tribution by the Buoyocous hydrometer method^!/ with sodium hexameta-
phosphate as a dispersing agent. As shown in the figure, the hydrometer
method disaggregates clay particles and produces a better representation
of the "ultimate" size distribution of the material.
COMPUTED EMISSION FACTORS
The environmental impact of dust emissions from unpaged roads varies
greatly with particle size. Large particles (d > 100 um) drift short
distances from the road during the settling process and create mainly a
nuisance problem. On the other hand fine particles (d < 2 um), which
represent a potential health hazard and which effectively reduce atmospheric
visibility, are dispersed to high altitudes, and may remain suspended for
long periods of time. Thus, it is imperative that emission factors be
developed for specific particle size ranges.
Gillette and Blifford—have recently developed criteria for the
maximum sized particle which can be supported in suspension by a given
turbulent wind and the minimum sized particle which settles unimpeded
by the vertical velocity fluctuations of the air. These size cut-offs
are related to specific ratios of particle settling velocity to friction
velocity. This work is reviewed further in Chapter 8.
The drift distance as a function of particle size may be estimated
from the initial height of injection into the atmosphere, the settling
velocity and the mean wind speed. For emission from unpaved roads, the
average height of injection is assumed to be 5 ft. The mean wind speed
at 5 ft is related to the speed at the 12-ft reference height through
the profile presented in Figure 2. The settling velocity is based on
o 9 / *^ 23/
the drag coefficient for spheres^-' and a particle density of 2.5 g/cmj.—
Figure 9 shows the calculated drift distance as a function of
particle size and mean wind speed. The boundaries of the settling-sus-
pension regimes were derived from the Gillette-Blifford criteria—using
a friction velocity based on a roughness height of 1 cm.—' As indicated
in the figure, particles which are not significantly affected by atmospheric
turbulence will settle to the ground within a drift distance of 15 ft.
Because particles which drift beyond 15 ft are affected by vertical
velocity fluctuations, the average drift distance will be greater than
the values shown.
36
-------
WEIGHT % GREATER THAN STATED SIZE
100.0» — ..-='.*». 2—2 S-
** •*"L
10.0
E
E
5
Q
UJ
_j
u
1,0
0.1
TIC in in
Road Surface Texture f
Run Site
-i] A 1
R1
Method ___
Gravel Dry Sieving 4-f-
Q 2&3 R2 Gravel Dry Sieving T^
^ O 13 R4 Gravel Dry Sieving ^7_
13 R4 Dirt Hydrometer ^
WEIGHT % LESS THAN STATED SIZE
Figure 8. In-place road dust texture.
37
-------
210
OJ
00
180
150
E
o£ 120
<
Q
LLJ
u 90
60
30
\Jn\mpeded Settling
I I I I I I
I I I I I I
UNPAVED ROADS
Indefinite Suspension
I I
Figure 9.
10
DRIFT DISTANCE DOWNWIND (ft)
Drift potential of road emissions.
100
-------
8/
It can be shown that Hoover s data" on the deposition near a gravel
road is consistent with Figure 9. Assuming that all wind directions were
equally likely over the 21-day test period (which means that the average
drift distance is 1.57 times the perpendicular distance from the road),
particles larger than 75 um settled within a drift distance of 75 ft.
The normal average wind speed for the test period was 9 mph.—'
Lundgren's study—' of the capture efficiency of a standard high-
volume sampler is also useful to the interpretation of particle size
spectra associated with the exposure measurements. He found that for
wind speeds in the 3-10-mph range, the suspended dust mass fraction not
collected by the high-volume samplers (operating at 55 cfm) was approxi-
mately equal to the total mass fraction greater than 60 um diameter, for
a particle density of 1-1.5 g/cm .
The effective cut-off diameter for capture of dust by a standard
high-volume sampler (or a high-volume cascade impactor operated within
a standard enclosure) is taken to be 30 um for a particle density of
2.5 g/cm^. This value is based on (1) Lundgren's result, (2) the settling
characteristics of road dust particles and (3) the observed ratios of
total high-volume concentration to isokinetic profiler concentration.
In the determination of emission factors for unpaved roads, dust
which settled out before reaching the exposure profiler (within 20-30
ft of drift distance from the downwind edge of the road) was not
included in the emission factor; these particles are larger than 100 um
for winds exceeding 10 mph.
The equations for calculation of the emission factors for three
particle size ranges (< 2 um, 2-30 um, > 30 um) are as follows:
1. For particles less than 2 um in diameter:
e< 2 = ER6F< 2
where e< 2 = mass of dust emissions less than 2 um in diameter per
vehicle-mile of travel (pounds per vehicle-mile)
E = integrated exposure measurement (pounds per vehicle mile)
R£ = ratio of the dust concentration measured by the standard
high-volume sampler to the concentration measured by
the isokinetic profiler at 6-ft height
F< 2 = fraction of the particles less than 2 um in diameter,
measured by high-volume cascade impaction.
39
-------
2. For particles with diameters between 2 and 30 um:
e2_30 = ER6(1 - F< 2)
where e?-30 ~ mass °^ dust emissions with diameters between 2 and 30 um
per vehicle-mile of travel (pounds per vehicle-mile)
and the other symbols are defined above.
3. For particles greater than 30 um in diameter, but excluding particles
which settled out over the first 20-30 ft of drift distance:
e> 30 = E(l - R6)
where e-> 30 = mass of dust emissions greater than 30 um in diameter
per vehicle-mile of travel.
Table 9 presents the calculated emission factors.
CORRECTION PARAMETERS
Atmospheric dust emissions from unpaved roads depend on the follow-
ing local parameters:
1. Average vehicle speed,
2. Vehicle mix,
3. Surface texture, and
4. Surface moisture.
Each of these factors is discussed below.
Average Vehicle Speed
The test results reported above indicate the total dust emissions
from unpaved roads increase in proportion to the average vehicle speed,
in the speed range of 30 to 40 mph. As shown in Figure 10, this depen-
dence is corroborated by the results of Duwamish Valley study.—' Sehmel's
data on the resuspension of tracer dust from asphalt roadsilQ.' indicates
that the linear dependence extends up to 50 mph. Below 30 mph, however,
both Duwamish Valley study and Sehmel's measurements indicate that emis-
sions increase in proportion to the square of the vehicle speed.
Since the typical speed range on unpaved roads is 30-50 mph, the
linear dependence of dust emissions on vehicle speed was used in
developing the correction factor.
40
-------
Table 9. CALCULATED EMISSION FACTORS (Unpaved Roads)
Run
1
2
3
8
10
13
Site
Rl
R2
R2
R3
R3
R4
Integrated
Exposure
(lb/ vehicle-mile)
10.0
10.3
13.9
16.3
6.0
55.9
Ratio Fraction of
Hi-Vol Catch: Hi-Vol Catch Emission Factors (lb/vehicle-mile)^/
Profiler Catch
0.60
0.66
0.57
0.50
0.65
0.57
< 2 u
0.45k/
0.45^
0.45^/
0.4^
0.46
0.41
d > 30 um
4.0 (40%)
3.5 (34%)
6.0 (43%)
8.2 (50%)
2.1 (35%)
24.0 (43%)
2 < d < 30 urn
3.3 (33%)
3.7 (36%)
4.3 (31%)
4.4 (27%)
2.1 (35%)
18.8 (34%)
d < 2 um
2.7 (27%)
3.1 (30%)
3.6 (26%)
3.7 (23%)
1.8 (30%)
13.1 (23%)
Total
10.0
10.3
13.9
16.3
6.0
55.9
aj d = particle diameter
b/ Estimated value
-------
_2
o
(U
i
ar
o
Z
O
1/1
I
20
15
10
9
8
7
6
A
10
I
GRAVEL ROADS
• Rl (Franklin Co.)
O R2 (Franklin Co.)
A Duwamish Valley Site 1
A Duwamish Valley Site 2
I
J_
l
ilt
20 30 40 50 60 70 80 90 100
VEHICLE SPEED (mph)
Figure 10. Effect of vehicle speed on gravel road emissions.
-------
Vehicle Mix
Based on the limited data presented in this report, a vehicle
traveling an unpaved road generates dust in proportion to the number of
its wheels. The emission factors presented above are based on equiva-
lent four-wheeled vehicles. For roads with a significant volume of heavy-
duty trucks or other vehicles, the traffic volume should be adjusted to
the equivalent volume of four-wheeled vehicles.
Surface Texture
Since the dust emissions which drift more than a few feet from an
unpaved road are smaller than 75 p in diameter, (i.e., defined as silt
particles), a linear dependence of emission on silt content of the road
surface material may be assumed. The average silt content of the loose
material on gravel roads was found to be 12.5%.
The amount of surface fines on an unpaved road is normally close to
an equilibrium value. The fines which are injected into the atmosphere
by vehicular traffic, are replaced in the same process by new fines which
are generated by abrasion of surface material. As was the case for
Site R3 in Morton County this equilibrium can be upset by a windstorm or
other severe phenomenon, and for a time emissions are reduced.
Surface Moisture
Unpaved roads have a hard, nonporous surface which dries quickly
after a rainfall. The temporary reduction in emissions because of rain-
fall is accounted for by neglecting emissions on "wet" days, i.e., days
with more than 0.01 in. of rainfall.
CORRECTED EMISSION FACTOR
The correction parameters discussed above have been incorporated into
a single mathematical expression for the amount of dust generated per
vehicle-mile of travel. The equation for estimating the total amount of road
dust emissions with drift potential greater than 25 ft, i.e., particles
smaller than 100 um in diameter, is as follows:
e(roads) = °-81 s
where e = emission factor (pounds per vehicle-mile)
s = silt content of road surface material (percent)
S = average vehicle speed (miles per hour).
43
-------
As shown in Table 10, the precision of this equation in predicting
the results of the emission tests of unpaved roads is ± 107=,.
The silt content (i.e., particles smaller than 75 urn in diameter) of
the road surface is determined by measuring the amount of loose (dry) sur-
face dust which passes a 200 mesh screen. The silt content of gravel
roads is approximately 12%.
The above equation applies to "dry" days. Emissions are assumed to
be negligible on days with rainfall exceeding 0.01 in.
The test results presented above indicate that, on the average, dust
emissions from unpaved roads have the following particle size characteristics:
Particle Diameter Weight Percent
< 2 urn 25
2 urn - 30 um 35
30 urn - 100 urn 40
44
-------
4>
Ul
Table 10. ESTIMATED VS ACTUAL EMISSIONS (Unpaved Roads)
Run
1
2
3
8
10
13
Site
Rl
R2
R2
R3
R3
R4
Vehicle
Speed
(raph)
30
30
40
30
40
30
Emission Factor
Percent (Ib/vehicle-mile)
Silt Estimated
12 9.7
13 10.5
13 14.0
205/ 16.2
&l 5.4
68 55.1
Actual
10,0
10.3
13.9
16.3
6.0
55.9
Percent
Difference
-3
2
1
-1
-10
-1
a/ Estimated value
-------
CHAPTER 5
AGRICULTURAL TILLING EMISSIONS
SAMPLING SITE DESCRIPTION
Morton and Wallace counties in Kansas were selected for the study of
atmospheric dust emissions from agricultural tilling. Located in extreme
southwest and west-central Kansas, respectively, both counties are in the
dry, windy area of the Great Plains referred to as the "dust bowl," where
problems of windblown dust are severe. The climatic potential for wind
erosion in the dust bowl area is illustrated in Figure 11, which presents
the distribution of annual average values of the climatic factor used in
iyc I
the wind erosion equation. —'
Detailed descriptions of the characteristics of the individual test
sites are given in the following paragraphs.
Morton County, Kansas
Morton County is located in the southwest corner of Kansas, near the
center of the dust bowl area of the Great Plains. The annual rainfall
in the county averages 16 in. and the average wind speed is 14 mph with
prevailing winds from the southwest.
Morton County is a part of the southern High Plains section of the
Great Plains physiographic province. About 85% of the county consists of
upland plains and rolling to hilly sandy land and the rest is stream
flood plains and intermediate slopes. Large areas on the upland are com-
paratively flat and featureless. In detail, however, most parts of the
flat upland are more or less uneven and consist of broad, gentle swells
or hills and shallow depressions.
The Cimarron River passes through the central part of the couaty.
In this county it is an intermittent stream that flows only when there
is a large amount of rainfall upstream.
46
-------
ANNUAL
CLIMATIC FACTOR C'
OEIGIKAL DRAWIIG ^-17-68, D, V. ARMBRUST.
ARK.. IA., KY., LA,, TEKN., W. VA. ADDED
11-24-71, N. P. WOODRUBT.
Figure 11. Climatic factor used in wind erosion equation.
-------
About 50% of the county is drained by the Cimarron liver and its
tributaries; the rest has no exterior drainage. Rain that falls on flat
upland and sandhills drains into temporary ponds or small, shallow lakes,
where it evaporates or percolates downward.
The elevation of the upland ranges from about 3,700 ft above sea
level in the southwestern part of the county to 3,150 ft on the eastern
county line. In general, the county slopes to the northeast and east about
15 ft/mile. The Cimarron River is more than 100 ft below the upland areas.
A soil survey of Morton County is complete and fully documented,~' and
it is tied in with aerial photographs. The two major soil associations are
Richfield/Ulysses and Dalhart/Richfield which cover 58 and 17% of the
county, respectively, and comprise the agricultural soils which are cul-
tivated to produce crops.
The Richfield/Ulysses association occurs in two nearly level to
gently sloping areas of the uplands, mostly in the northern half of the
county. It is composed mainly of soils with a loamy surface layer. Most
of this association is used for crops, principally grain sorghum and
wheat, which are often grown on a crop-fallow system. Most of the irriga-
tion in the county is done on soils of this association.
The Dalhart/Richfield association occurs south of the Cimarron
River and is composed of soils with a sandy surface layer. Most of this
association is used to produce crops. Sorghum is the main crop, but wheat
is grown on a small portion of the acreage.
Two individual sites in Morton County were selected for the study of
atmospheric emissions from agricultural tilling. Site Al, located in the
south-central part of the county, was a section of fallow acreage with a
surface of fine sandy loam; the terrain was level and there was little
vegetative cover. Site A2, located in the west-central part of the
county, was a section of fallow acreage with a surface of silt loam.
Additional details of the site characteristics are given in Table 11.
Wallace County, Kansas
Wallace County is situated on the western-most tier of Kansas counties
about one-third of the way downstate, in the dust bowl area of the Great
Plains. The annual rainfall in the county averages 22 in. and the average
wind speed is 14 mph with prevailing winds from the southwest.
48
-------
Table 11. AGRICULTURAL SITE CHARACTERISTICS
Site Location Soil Type Slo{>e
Al Morton County: T35S, R41W, Dalhart/Richfield 0-1%
Section 8 fine sandy loam
A2 Morton County: T33S, R43W, Ulysses/Richfield 0-1%
Section 22 silt loam
A3 Wallace County: T13S, R40W, Ulysses/Colby 1-2%
Section 19 silt loam
.P, A4 Wallace County: T13S, R41W, Keith/Colby Terraced
Section 26 silt loam
-------
The soil of Wallace County is derived from three major soil associa-
tions: (1) Canyon/Colby (immature and shallow soils on steep slopes) in
the north; (2) Keith/Colby in a band from west-central to southeast, and
(3) Richfield/Colby in the southwest part of the county. The Keith/Colby
and Richfield/Colby associations are chestnut-colored soils developed
under prairie vegetation and are representative of a large area of the
Great Plains. An extensive soil survey is underway and is being tied to
aerial photographs.
The Keith/Colby and Richfield/Colby soils are well suited to culti-
vation for crop production. The area has traditionally grown a crop of
winter wheat every second year in rotation with summer fallow.
Two individual sites in Wallace County were selected for the study of
atmospheric emissions from agricultural tilling. Both sites were located
in the central portion of the county, just west of Sharon Springs. Site
A3 was a section of gently sloping fallow land with light vegetative
cover. Site A4 was a terraced section of fallow land with light vegeta-
tive cover. The surface soil at both sites was a silt loam. Additional
details of the site characteristics are given in Table 11.
FIELD MEASUREMENTS
Field testing of dust emissions from agricultural tilling was con-
ducted at the Morton County sites (Al and A2) in May and June 1973} and
at the Wallace County sites (A3 and A4) in June 1973. The testing of
agricultural tilling emissions had to be postponed from dates scheduled
in March and April because of adverse weather conditions, as explained
below.
The spring of 1973 was one of the wettest in history in the Great
Plains. During March and April flooding was widespread and received
extensive news coverage. Because fugitive dust emissions are highly
dependent on surface moisture, the decision was made not to test under
these highly nonrepresentative conditions. As a result, testing was
curtailed until mid-May.
Because of persistent wet weather in March and April, the tilling
operations in preparation for spring planting were very atypical and were
not tested. Instead of the originally scheduled testing of tilling emis-
sions from spring seedbed preparation, testing was conducted on the tilling
of fallow ground which was later planted in winter wheat (at the end of
the summer).
50
-------
The tillage implements which were selected for testing were the
one-way disk plow and the sweep-type plow. These implements were chosen,
with the advice of area agricultural specialists, as representative of
implements used in dryland farming in the Great Plains.
Table 12 specifies the kinds and frequencies of field measurements
that were conducted during each run. "Composite" samples are made up
of single samples taken from several locations in the area; "integrated"
samples are those taken at one location for the duration of the run.
Composite samples of soil (8-12 cores) were obtained with a plugging
device from randomly selected locations within 100 yards of the exposure
profiler. The soil was sampled separately to depths of 4 and 6 in. The
soil samples were stored in polyethylene bags and returned to MRI for
laboratory determination of texture and moisture content.
At the end of each run, the collected samples of dust emissions were
carefully transferred to shipping containers within the MRI instrument
van to prevent dust losses. High-volume filters (from the MRI exposure
profiler and from standard high-volume units) were placed in individual
folders. Dust that collected on the interior surfaces of each exposure
probe was rinsed with distilled water into a glass jar. The contents
of the deposition samplers and saltation catchers were also rinsed into
glass containers. Cascade impactor collection papers were left in place
within each impactor unit.
Table 13 presents information on the time of each run, the prevail-
ing meteorological conditions and the tillage implement. The duration of
sampling for the exposure profiler was a fraction of the total elapsed
test time because the profiler was operated only when the tillage imple-
ment was nearby. The other sampling devices were operated continuously
during the run.
Table 14 gives the locations (intake height and distance from tillage
path) of the various plume sampling devices that were used for each run.
The dust particle sizing samplers included Andersen and Sierra high-volume
cascade impactors (operated within a standard high-volume enclosure). The
drift distance multiplier, given in the last column of the table, takes
into account the effect of the angle between the horizontal wind direction
and implement path direction, on the plume travel distance.
51
-------
Table 12, FIELD MEASUREMENTS--AGRICULTURAL TILLING
Test Parameter Units
Meteorology
a. Wind speed mph
b. Wind direction deg
c. Cloud cover %
d. Temperature °F
e. Relative 7,
humidity
Field Surface
a. Soil texture pm
b. Soil moisture %
content
c. Vegetative
cover
Sampling Mode
Continuous
Continuous
Single
Single
Single
Composite
Composite
Multiple
Measurement Method
Recording instrument at "back-
ground" station; sensors at
reference height
Visual observation
Sling psyehrometer
Hydrometer method
Weight loss on oven drying
Observation (photographs)
3. Tillage Equipment
a. Type
b. Dimensions
c. Translational
speed
d. Number of
passes
ft
mph
Single Observation (photographs)
Single Observation (photographs)
Multiple Elapsed time between
reference points
Cumulative Counting
4, Suspended Dust (downwind unless indicated)
mg/in^
pm
Pg/m
a. Exposure
(vs height)
b. Size distribu-
tion (by wt.)
c. Concentration
d. Background
concentrat ion
e. Duration of min
sampling
5. Large Particle Transport
a. Deposition Ib/ft2/hr
(vs distance
from source)
b. Saltation ing/in
Integrated Isokinetic high-volume
filtration (MRI method)
Integrated Cascade impaction
Integrated High-volume filtration
(EPA method!!/)
Integrated High-volume filtration
(EPA methodil/)
Cumulative Timing
Integrated Dustfall buckets (ASTM
method!!/)
Integrated Saltation catcher
52
-------
Run
5
6
7
9
11
12
14
Site
Al
Al
Al
A2
A3
A3
A4
Date
24 May
24 May
25 May
5 June
12 June
12 June
1 3 June
1973
1973
1973
1973
1973
1973
1973
Tin
Start
1415
1818
0940
1125
0829
1117
1111
e
Finish
1727
1958
1108
1350
1009
1409
1222
Duration of
Exposure Direction
Sampling of
(niin) Travel
32 . 5 E-W
13.7 E-W
24.3 E-W
31.0 E-W
25.0 E-W
22.0 E-W
25.0 N-S
Ambient
Temp.
(°F)
74
74
70
75
68
80
70
Wind Di
Speed
N/13
N/13
S/12
NW/10
NE/12
NE/10
SE/8
rection/'
(12 ft)
mph
niph
mph
mph
mph
mph
mph
Cloud
Cover Pasquill
(") Stability-
0 B
0 D
30 A-B
5 A
80 D
80 C
100 C
lilla^e Implement
Speed No. of
Type (niph) Passes
12-ft disk 5-6 15
12-ft disk 5-6 10
12-ft disk 5-6 16
30-ft disk 4 10
30-£t sweep 6-7 12
30-ft sweep 7 12
20-ft disk 5-^ 12
a/ Pasquill Stability Classes:!!' A - Extremely unstable D - Neutral
B - Unstable E - Slightly stable
C - Slightly unstable F - Stable to extremely stable
b/ Estimated value
-------
Table 14. DUST EMISSION SAMPLER LOCATIONS (Agricultural Tilling)
Perpendicular Distance from Downwind Edge of Tillage
MRI Particle Size Deposition Saltation
Exposure Classifier—' Sampler Catcher
Run Site Profiler!/ (h = 6 ft) (h = 1 ft) (1 < h < 2.5 ft)
5 Al 20
6 Al 22
7 Al 20
9 A2 20
11 A3 20
12 A3 20
14 A4 22
20 (Al) - 20
22 (SI) - 22
20 (AC) - 20
20 20
20 (Al) 20 20
20 (Al) - 20
22 (Al) - 22
Path (ft)
Standard Drift
Hi-Vol Sampler Distance
(h = 6 ft) Multiplier-/
1.0
1.0
1.0
20 1.4
1.4
1.4
1.4
a/ Sampling heights: h = 3, 5.5, 8, and 10.5 ft above grade.
b_/ Sizing device: Al = Andersen impactor
SI = Sierra impactor
AC = Aerotec cyclone
c/ Ratio of drift distance to perpendicular distance.
-------
TEST RESULTS
Dust samples from the field tests were analyzed gravimetrically In
the laboratory. Filters were conditioned in a controlled temperature-
humidity environment prior to weighing. Water rinses from exposure
probes, deposition samples and saltation catchers were evaporated on a
steam bath in tared beakers, after which the beakers were conditioned and
weighed,
The measured dust emissions from the tests of agricultural tilling
are shown in Table 15. The dust quantities are the amounts generated per mile
of 12-ft tilling cut. This normalization basis has been chosen for com-
parison with unpaved road emissions.*
The total dust emissions for a given run are the sum of the integrated
exposure (above the background exposure) and the amount of deposition
between the edge of the road and the downwind location of the exposure
profiler.
The suspended dust measurements used to compute the integrated ex-
posure are presented In Table 16. Point values of exposure are converted
to concentration. The concentration measured by the standard high-volume
unit, which was positioned to the side of the profiler, is also presented.
The exposure profiles are shown in Figure 12.
In general, deposition measurements were not obtained for agricultural
tilling because most of the deposition occurs on the tilled land, A deposi-
tion measurement was made for Run 11 and the cumulative deposition between
the downwind edge of the tillage path and the exposure profiler, was
determined by the method described in Chapter 4,
The measurements of dust transport by saltation are shown only for
purposes of comparison. Saltation, which is confined to about 30 in. of
height, is not considered to be a form of atmospheric emissions. Also It
should be noted that the saltation catchers used in this study did not
sample below 12 in. above the ground.
Also given in Table 15 is the mass mean diameter of suspended dust
particles measured with the Andersen high-volume cascade irapactor. The
diameter values are aerodynamic measures which treat particles as equiva-
lent spheres with a density of 2.5 g/cm3. The complete size distribu-
tions are shown in Figure 13.
A typical roadway lane is 12 ft in width.
55
-------
Table 15. MEASURED DUSTEMISSIONS (Agricultural Tilling)
Ul
Run
5
6
7
9
11
12
14
Site
Al
Al
Al
A2
A3
A3
A4
Background
Concentration
(ug/m3)
44.5
44.5
-
40.6
30.7
30.7
87.6
Wind
Speed
(mph)
13
13
12
10
12
10
8
Plume Travel Time
to Profiler
(sec)
1.0
1.0
1.1
1.9
1.6
1.9
2.6
Unit Dust Catch
(Ib/mlle of 12 ft
Saltation Deposition
1.8
1.9
5.4
0.66
0.84 11.4
0.80
1.3
cut)
Integrated
Exposure
81.4
75.4
86.6
50.5
92.4
124
114
Hi-Vol
MMM/
(pm)
2.3
-
-
-
2.5
2.0
2.9
a/ Mass mean diameter of suspended dust, measured with Andersen high-volume cascade impaetor.
-------
•Table 16. PLUME SAMPLING DATA (Agricultural Tilling)
Run
5
6
7
9
11
12
14
Height
Site (ft)
Al 10.5
8
6S/
5.5
3
Al 10.5
8
&
5,5
3
Al 10.5
8
5.5
3
A2 10.5
8
6£/
5.5
3
A3 10.5
8
6S/
5,5
3
A3 10.5
8
<£/
5.5
3
A4 10.5
8
&£/
5.5
3
Sampling
Rate
(cfm)
27.5
26.8
18.5
25.0
22.8
27.5
26.8
40.3
25,0
22.8
27.5
26.8
25.0
22.8
24.1
22.8
42.6
21.3
19.2
24.3
23.2
22.0
21.5
19.2
28.5
27.0
24.0
25.3
23.0
19.8
18.5
27.0
17.0
15.2
Concentration
2.00
3.13
8.23
10.3
21.8
2.01
7.35
5.32
14.9
34.3
0.864
4.29
13.4
44.0
6.17
9.52
13.5
15.8
25.4
12.3
17.2
10.5
34.3
57.7
15.6
23.9
27.9
40.7
75.9
14.3
22.9
15.3
37.1
62.3
Unit Exposure
o
(mg/ in. /equivalent pass)
0.804
1.23
__
3.77
7.27
0.537
1.92
__
3.60
10.8
0.256
1.24
3.60
10.8
1.30
1.91
--
2.96
4.29
1.76
2.35
__
4.35
6.53
2.31
3.34
__
5.35
9.06
2.53
3.74
__
5.59
8.38
a_/ Andersen impact or.
b/ Sierra impactor.
c/ Standard high-volume sampler.
57
-------
U1
oo
18
16
14
12
10
x
O
Agricultural Tilling
Run Site
A 5
O 6
a 7
• 9
AH
T 12
• 14
Al
Al
Al
A2
A3
A3
A4
468
UNIT EXPOSURE (mg/in2 - equiv. pass)
Figure 12. Exposure prof lies—agricultural tilling.
10
12
-------
WEIGHT % GREATER THAN STATED SIZE
5
Q
LU
_l
u
Andersen Results
(Agricultural Tilling)
O Run 5, Site Al
Run n, Site A3
a Run 12, Site A3
o Run 14, Site A4
Figure 13.
WEIGHT % LESS THAN STATED SIZE
Particle size distributions—agricultural emissions.
59
-------
Two potentially significant sources of error in the particle size
measurements as mentioned in Chapter 4 were:
1. The impactor samples nonisokinetically through the high-volume enclo-
sure openings and collects large particles with low efficiency.
2. Unlike urban aerosol, tillage dust particles are dry and brittle and are
subject to bouncing and reentrainment from impaction surfaces. Recent
empirical evidence obtained by SehmelJ=M' indicates that this effect is
most pronounced for particles larger than 20 urn in diameter.
Both of these factors cause apparent size determinations to be biased
in the direction of small diameter. The second factor seemed to be sub-
stantial with the slotted impactor (HMD «* 1 u); for this reason, the
Sierra measurements were not used.
Table 17 gives the results of the laboratory analyses of the soil
samples. Moisture content was determined by weight loss on oven drying
and particle size analysis by the Buoyocous hydrometer method=i' (with
sodium hexametaphosphate as a dispersing agent) and by wet sieving.
The significantly higher moisture content of the soil at the 4-6 in.
depth in comparison with the 0-4 in. depth, indicates the transfer of
moisture from beneath the exposed soil surface to replace moisture lost by
atmospheric drying.
As indicated in Table 17, the soil from Site Al is rich in fine sand
and Site A3 has the highest total silt content. The size distributions for
the soil samples are plotted in Figure 14.
COMPUTED EMISSION FACTORS
The approach that was used in the development of emission factors
for agricultural tilling, is the same as that presented in Chapter 4.
Emission factors for three particle size ranges (d < 2 urn, 2 urn <, d ^30 um,
d > 30 Jim) were determined from the integrated exposure measurements, the
cascade impactor measurements of particle size and the ratio of high-
volume concentration to the isokinetic profiler concentration for a height
of 6 ft.
Figure 15 shows the estimated drift distance as a function of the
size of the particle injected into the atmosphere and the mean wind speed.
For emissions from agricultural tilling, the average height of injection
is assumed to be 2 ft. The mean wind speed at 2 ft is related to the
speed at the 12-ft reference height by the profile presented in Figure 2,
The settling velocity is based on the drag coefficient for spheres-==' and
a particle density of 2.5 g/cm^.—'
60
-------
Table 17. SOIL PROPERTIES (Agricu1tura1 Tilling)
Texture, 0-4 in, depth (/
Run
5
6
7
9
11
12
14
Site
Al
Al
Al
A2
A3
A3
A4
Moisture ( / )
Soil Type 0-4 in. depth 4-6 in. depth
Sandy Loam 10.5 10.7
Sandy Loam
Sandy Loam -
Loatn 11.0
Silt Loatn 15.9 19.7
Silt Loam 13.4 17.5
Silt Loair. 12.3 18.5
Medium-
Coarse Sand
(2000-250 pm)
2
2
2
8
1
1
5
Fine
Sand
(250-105 pm)
35
35
35
10
1
1
3
Very Fine
Sand
(105-50 pm)
17
17
17
22
20
21
18
, b , vei ht)a/
Coarse
Silt
(50-20 urn)
14
14
14
16
30
30
27
Fine
Silt
iiSrJLMll
12
12
12
16
18
19
19
Clay
(< 2 urn)
20
20
20
28
30
28
28
al Determined by Buoyocous hydrometer method.
-------
WEIGHT % GREATER THAN STATED SIZE
100.0
10.0
E
E
<
Q
< 1.0
U
o.
0.01 ' '...
ttl
H H H
M H «fl
i j
fl _ 9t t\
± i
[Agricultural Soil Texture]
[
I a Run 5, Site Al ;
; O Run 9, Site A2
• v Run 11, Site A3
| O Run 12, Site A3 '
' A Run 14, Site A4
WEIGHT % LESS THAN STATED SIZE
Figure 14. Surface soil texture.
62
-------
w
AGRICULTURAL TILLING
Indefinite Suspension
100
Figure 15.
DRIFT DISTANCE DOWNWIND (ft)
Drift potential of tillage emissions.
-------
The boundaries of the settling and suspension regimes were derived
from the Gillette-Blifford criteria!^.' using a friction velocity based
on a roughness height of 1 cm.—' As indicated in Figure 15, particles
which are not significantly affected by atmospheric turbulence will settle
to the ground within a drift distance of 5 ft. Because particles which
drift beyond 5 ft are affected by vertical velocity fluctuations, the
average drift distance will be greater than the values shown.
The effective cut-off diameter for capture of dust by a standard
high-volume sampler (or a high-volume cascade impactor operated within
a standard enclosure) is taken to be 30 p.m for a particle density of
2.5 g/cnP. This figure is based on (1) Lundgren's result,!!' (2) the
settling characteristics of agricultural dust particles and (3) observed
ratios of dust concentration by high-volume measurement to dust concentra-
tion by isokinetic profiler measurement.
In the determination of emission factors for agricultural tilling,
dust which settled out before reaching the exposure profiler (within
20-30 ft of drift distance from the downwind edge of the tilling path)
was not included in the emission factor; these particles are larger than
75 um in diameter for winds exceeding 10 mph.
The equations for calculation of the emission factors for three
particle size ranges (< 2 urn, 2-30 urn, > 30 urn) are as follows:
e
< 2
e2-30 - EV1 ' F<
e> 30
where e, = mass of dust emissions with diameter i per acre tilled
E = integrated exposure measurement
Rg = ratio of dust concentration measured by the standard high-
volume sampler to the concentration measured by the
isokinetic profiler, at 6 ft height
;p< 2 = fraction of the particles less than 2 um in diameter,
measured by high-volume cascade impaction
The calculated emission factors are presented in Table 18.
CORRECTION PARAMETERS
Atmospheric dust emissions from agricultural tilling exhibit signifi-
cant dependence on the following variable factors:
64
-------
Table18. CALCULATED EMISSION FACTORS (ARricultural Tilling)
Ui
Run
5
6
7
9
11
12
14
Site
Al
Al
Al
A2
A3
A3
A4
Integrated
Exposure
(Ib/mile
of 12-ft cut)
81.4
75.4
86.6
60.6
92.4
124
114
Ratio
Hi-Vol Catch:
Profiler Catch
0.
0.
0.
0.
0.
0.
0.
90
90
90
90
75
75
75
Fraction of
Hi-Vol Catch
< 2 microns
0.44
0.44^
0.44^
0.44^
0.42
0.50
0.38
Emission Factors
d > 30 ]im
5.6
5.2
6.0
4.2
15.9
21.3
19.5
(10%)
(10%)
(10%)
(10%)
(25%)
(25%)
(25%)
2 < d
28.2
26.2
30.0
21.0
27.7
31.9
36.3
(Ib/acre)-'
< 30 (itn d <
(50%)
(50%)
(50%)
(50%)
(44%)
(37%)
(46%)
22.1
20,5
23.6
16.4
20.0
32.0
22.3
2 Jim
(40%)
(40%)
(40%)
(40%)
(31%)
(38%)
(29%)
Total
55.9
51.9
59.6
41.6
63.6
85.2
78.1
a/ d = particle diameter
b/ Estimated value
-------
1. Surface soil texture,
2. Surface soil moisture content, and
3. Implement speed.
Each of these factors is discussed below:
Surface Soil Texture
There is good reason to infer a linear dependence of dust emissions
from agricultural tilling on the silt content (i.e., particles between
2 and 50 u in diameter) of the surface soil. Firstly, dust emissions
which drift more than a few feet from a tillage operation are smaller
o o /
than 50-75 um in diameter. Secondly, Gillette±£' has found that clay
particles (smaller than 2 um in diameter) remain bound to larger par-
ticles during wind erosion because of the relatively large amount of
energy required to disaggregate particles in that size range; the same
reasoning should apply to dust generated by tilling.
Surface Soil Moisture
Those familiar with agricultural tilling are well aware that dust
emissions increase substantially in dry weather. Moisture tends to bind
fine dust particles together.
The developers of the Wind Erosion Equation-^' which is used to pre-
dict the susceptibility of a given area of land to wind erosion, have
found that erosion is inversely proportional to the square of the mois-
ture content of the surface soil. They have adopted Thornthwaite1 s; pre-
cipitation-evaporation index-=-^-' as a useful approximate measure of
average soil moisture.
The inverse square dependence of dust emissions from agricultural
tilling on the moisture content of the surface soil (0-4 in. depth) was
demonstrated on a very limited basis at Site R3 in Wallace County, Kansas.
Test 11 was conducted in the morning and Test 12 in the early afternoon
of the same day; the measured increase in emissions from the same tillage
tool was approximately inversely proportional to the square of the de-
crease in soil moisture.
Implement Speed
Dust emissions from agricultural tilling are dependent on the rate
at which mechanical energy is consumed by working the soil. Since tillage
implements are designed to operate over a narrow speed range, a linear
dependence of emissions on implement speed may be assumed. As a practical
matter, data on implement speed is not recorded and emission estimates
must be based on the average implement speed.
66
-------
CORRECTED EMISSION FACTOR
The correction parameters discussed above have been incorporated
into a single mathematical expression for the amount of dust generated
per acre of land tilled.
The equation for estimating the total amount of tillage dust emissions
with drift potential greater than 25 ft, i.e., particles smaller than
75 um in diameter, is as follows:
e/ .IT ^ = t .5)
e(txllxng)
where e = emission factor (pounds per acre)
s = silt content of surface soil (percent)
S = implement speed (miles per hour)
PE - Thornthwaite1s precipitation-evaporation index
As shown in Table 19, the precision of this equation in predicting
the results of the emission tests of agricultural tilling is ± 15%.
The soil silt content (i.e., particles between 50 um and 2 um in
diameter) may be determined by the Buoyocous hydrometer method.—'
Surface soil samples should be extracted with a plugging device to a
depth of 4 in.
The PE index is determined from total annual rainfall and mean
annual temperature; rainfall amounts must be corrected for irrigation.
The test results presented above indicate that, on the average,
dust emissions from agricultural tilling have the following particle
size characteristics:
Particle Diameter Weight Percent
< 2 um 35
2 um - 30 um 45
> 30 um 20
67
-------
oo
Table 19. ESTIMATED ¥S ACTUAL EMISSIONS
(Agr icu1tural Til1ing)
Run
5
6
7
9
11
12
14
Site
Al
Al
Al
A2
A3
A3
A4
Implement
Speed
(mph)
5.5
5.5
5.5
4
6.5
7
5
a/
Soil Properties-
Percent
Silt
26
26
26
32
48
49
46
Emission Factor
Moisture Equivalent (Ib/acre)
(% by
10
10
10
11
15
13
12
weight)
.5
.5
.5
.0
.9
.4
.3
PEk/
40
40
40
41
59
50
46
Estimated
57
57
57
49
56
87
70
Actual
56
52
60
42
64
85
78
Percent
Difference
2
9
-5
16
-12
2
-10
a/ 0-4 in, depth.
b/ Precipitation-evaporation index, adjusted to soil moisture.
-------
CHAPTER 6
AGGREGATE STORAGE PILE EMISSIONS
This chapter presents the results of two separate emission testing
studies which were conducted to characterize dust emissions from aggre-
gate storage piles. The first sampling program was designed to quantify
total dust emissions from the various constituent sources associated with
a representative aggregate storage operation. The second study had as
its purpose the quantification of emissions from a specific storage
transfer operation--aggregate loadout.
TOTAL EMISSIONS FROM AGGREGATE STORAGE OPERATIONS
Sampling Site Description
The Dravo Corporation sand and gravel pit located at Camp Dennison,
Ohio (just east of Cincinnati), was selected for testing of dust emissions
from aggregate storage piles. A survey of this pit and processing area
indicated that its stockpile operations were representative of those at
many aggregate quarrying operations of medium and large size.
The Dravo sand and gravel pit at Camp Dennison is situated in the
Little Miami River Valley about 7 miles northeast of the point where it
meets with the Ohio River Valley. Prevailing winds In this area during
the spring and early summer, reinforced by channeling in the river
valley, are from the southwest and south.
The Camp Dennison pit produces about 800,000 tons of aggregate
annually. The operation is year-round, with production rates changing
seasonally with demand for aggregate from local construction projects.
For most of the year, excavation, processing, and loading are on a
5-day week, 8-hr day schedule. During the June and July sampling period,
the operation was at its peak annual level and was active 5-1/2 days a
week, 10 to 12 hr a day.
69
-------
The gravel pits and stockpiles, as shown in Figure 16, are adjacent
to each other. However, they are separated by 40 to 70 ft vertically—
the distance from the floor of the pit to the grade level in the process-
ing and storage areas. This separation effectively eliminates the impact
of dust emissions from quarrying on the storage area.
The active crushing and screening equipment and loading hoppers are
north of the stockpile cluster. The crushing and screening plant shown
in Figure 16 as being located in the storage area is not currently in
use and was not operated during the sampling period.
The storage area covers approximately 17 acres. There were 15 stock-
piles in this area at the time of the field study, ranging in height from
5 to 30 ft. The average height, weighted on the basis of exposed surface
area, was 23 ft (7.0 m). The total estimated weight of the aggregates in
storage was 50,000 tons, and the approximate total surface area of the
15 piles was 96,000 ft2 (9,000 m2).
All stockpiled stone and gravel has been washed and screened, but
none has been crushed. Stockpiled sand has been dredged and put into
storage without washing or screening. Material processed through the
crusher is loaded directly for shipment.
By comparing the amount of material in storage to the annual produc-
tion rates or daily rates of movement into and out of storage, it is obvious
that the stockpiles have a high turnover and that there is significant
activity in the storage area on a daily basis. This activity in the storage
area affects the rate of dust generation. In other words, dust in aggregate
storage areas is produced not just by wind erosion on exposed surfaces, but
also by vehicle movement between piles and by disturbances of the aggregate
in moving it into and out of piles.
Field Measurements
Field testing of dust emissions from aggregate storage piles at the
Camp Dennison site was conducted during a 1-month period beginning 9 June
1973. The test program consisted of 11 24-hr runs and eight 12-hr runs.
Table 20 specifies the kinds and frequencies of field measurements that
were performed during each run.
Because of the diffuse and variable nature of the source, conventional
high-volume samplers with wind direction activators were used to measure
dust emissions. A 180-degree sector of sampling was employed, so that any
wind with a southerly component activated all the samplers. This effected
the isolation of the storage area from the various processing and truck
traffic emissions to the north of the storage area and from the pit
operations.
70
-------
T)/(6)- Sampling Sites
PIT
NO. 5? % (MO. 9 GRAVEL
(D
BY-PASS US50-I26
Figure 16. Aggregate storage sampling site, Dravo Corporation,
Camp Dennison, Ohio.
-------
Table 20. FIELD MEASUREMENTS—AGGREGATE STORAGE PILES
Test Parameter
1. Meteorology
a. Wind speed mph
b. Wind direction deg
c. Cloud cover °L
d. Temperature °F
e. Rainfall in.
Units Sampling Mode
Continuous
Continuous
Multiple
Multiple
Cumulative
mm
2. Aggregate
a. Size
b. Pile
configurat ion
3. Suspended Dust
a. Concentration
(vs location)
b. Background
c. Size distribu- um
tion (by wt.)
d. Duration of min
sampling
4. Operations Log (only for
a. Material tons
loaded
b. Material tons
excavated
c. Material tons
sized
Single
Single
Integrated
Integrated
Integrated
Cumulative
weekday samples)
Cumulative
Cumulative
Cumulative
Measurement Method
Recording instrument on site
Recording instrument on site
Hourly readings at Lunken
Field
Hourly readings at Lunken
Field
Daily readings at Lunken
Field
NCSA standard ranges
Observation
High-volume filtration
w/directional control
High-volume filtration
w/directional control
Cascade impaction
High-volume time meters
Operator's records and
estimates
Operator's records and
estimates
Operator's records and
estimates
72
-------
A series of five directional high-volume samplers were installed at
representative locations immediately downwind from storage piles holding
different sizes of aggregate. Locations of the high-volumes are shown
in Figure 16.
The samplers were placed at various heights above grade from 3 ft to
20 ft. The assumptions were made that, during periods with winds blowing
out of a southerly direction:
1. Particulates were emitted parallel to the wind direction over the
entire 980-ft width of the storage area;
2. The emissions occurred from ground level to a height approximated
by the average height of the storage piles; and
3. The average particulate concentration at the five downwind sampling
stations was representative of the particulate concentration in the
assumed rectangular cross section which contained all of the emissions
from the stockpile area.
An additional high-volume sampler with the same 180-degree sampling
sector was located south of the storage area (at station 1 in Figure 16)
to measure the incoming, or background, particulate levels in the air-
stream. In the data analysis phase, this upwind particulate concentra-
tion was deducted from the measured downwind concentration to determine
the net contribution from the stockpile area.
The sampling schedule was designed to obtain the maximum possible
number of independent samples within a 1-month period. In addition, an
effort was made to obtain some of the samples during periods when only
wind erosion was causing emissions—12-hr samples from 6:00 PM to 6:00 AM
and 24-hr samples from noon Saturday to noon Sunday—for comparison with
samples taken during periods when there was movement of the piles and
traffic in the stockpile area. The sampling periods are shown in Table 21.
All six samplers were operated on the same schedule.
The number of minutes that the directional controls activated the
high-volumes were usually almost the same for all six samplers during
each sampling period, indicating that wind directions were uniform over
the sampling area. The values for running time shown in Table 21 were
obtained from time meters attached to the high-volume samplers.
Wind speed and direction data were also measured and recorded at the
study site. The weather vane and anemometer at the study site were located
on a mast at Station 4, and were about 25 ft above grade with no nearby ob-
structions. The continuous data have been summarized for 6-hr periods in
Table 22. All other meteorological data were obtained from the FAA Weather
Station at Lunken Airport, located about 5 miles southwest of the Dravo
73
-------
Table 21. HIGH-VOLUME SAMPLING DATA
(Sand and Gravel Storage Piles)
Test Period
Date Start Time
6/9/73
6/11/73
6/12/73
6/13/73
6/14/73
6/16/73
6/18/73
6/19/73
6/20/73
6/21/73
6/23/73
6/25/73
6/26/73
6/29/73
6/30/73
7/2/73
7/3/73
7/5/73
7/6/73
1200
1800
1200
1800
1200
1200
1800
1200
1800
1200
1200
1800
1200
1800
1200
1800
1800
1200
1200
(hr)
24
12
24
12
24
24
12
24
12
24
24
12
24
12
24
12
12
24
24
Sta. 1
1130
484
1009
276
695
1126
532
1149
410
1205
1087
586
1181
_
1233
611
1139
770
1093
Sampling Duration (min)
Sta. 2
1140
403
1074
70
424
1192
340
1127
Void
1032
1011
578
1440
_
1119
613
1058
508
734
Sta. 3
1082
415
1103
73
347
1128
381
1160
Void
1301
1440
721
1365
_
1066
620
1031
420
842
Sta. 4
1074
413
1039
80
360
1082
406
1134
201
940
Void
301
1290
-
1190
596
869
1311
751
Sta. 5
1165
423
1073
280
661
1168
619
1440
719
1423
1352
719
240
-
982
378
1249
1280
1432
Sta. 6
1064
355
1090
62
285
378
Void
940
Void
1009
1024
510
Void
-
1032
Void
1054
375
706
-------
Table 22. SAMPLING SITE DATA
(Sand and Gravel Storage Piles)
•M
tn
Date/Hour
6/9/73
6/10/73
6/11/73
6/12/73
6/13/73
6/14/73
6/15/73
6/16/73
6/17/73
6/18/73
6/19/73
6/20/73
6/21/73
6/22/73
6/23/73
6/24/73
6/25/73
6/26/73
6/27/73
6/28/73
6/29/73
6/30/73
7/1/73
7/2/73
7/3/73
7/4/73
7/5/73
7/6/73
7/7/73
Material Processed
Wind Speed (mph)
00-0600
2.5
3.5
4.5
5.0
3.0
7.0
3.5
6.5
4.0
3.5
4.0
4.5
8.0
9,0
4.0
2.5
4.0
4.0
2.5
2.5
06-1200
4.5
5.0
5.5
10.0
12.0
6.5
9.0
8.5
3.5
6.5
8.5
4.5
5.5
12-1800
14,5
15.0
11.0
16.0
17.5
9.0
7.0
25,0
19.5
18.5
9.5
7.5
11.0
18-2400
8.0
6.0
6.0
8.0
7.0
7.5
5,5
7.5
7.5
4.0
4.5
5.0
10.5
9.0
11.0
7.0
9.5
8.0
4.5
6.0
00-0600
SSE
NNW
W
N
N
NW
WSW
SSW
NW
WSW
W
W
SSW
sw
N
N
SW
WSW
SSE
N
Wind Direction
06-1200 12-1800
WNW
WSW
W
NNW
ESE
WNW
W
SSW
SW
WSW
W
WSW
N
WSW
SW
WSW
W
SW
WSW
SSW
SSW NE
N
WSW SW
N
18-2400
WNW
W
WSW
N
NE
WSW
WSW
SSE
NW
SE
WSW
SW
W
WSW
N
WSW
WSW
WNW
NE
SSW
recipitation Excavated and
(in.)
0
0
0
0
trace
0
0
0.39
0.59
0
1.96
0.03
trace
0
0
0
0
0,12
1.37
trace
trace
0
0.94
trace
0.55
0.26
0
0
0
Sized (tons)
_
3830
3850
3970
4155
4020
805
_
3600
3645
3855
3070
3775
_
_
3385
4020
3375
1145
875
3670
_
-
-
-
-
.
Loaded
(tons)
-
6461
6461
4616
3946
5140
-
_
4555
5334
3059
3813
4791
_
.
4854
5644
2655
2782
-
-
-
-
-
-
-
-------
pit in the Little Miami River Valley. Daily rainfall for the sampling
period is also presented in Table 22.
As a check of the on-site wind measurements, the 6-hr average wind
speeds shown in Table 22 were compared by linear regression analysis with
corresponding measurements from Lunken Airport. For the 66 data points
considered, the slope of the regression line was 1.11 and the correlation
coefficient was 0.86. Thus, the measurements on-site were generally about
11% higher than at the airport, and the two data sets showed a good
correlation.
Test Results
The measured background dust concentrations and the net concentrations
(background subtracted) at the five downwind stations are shown in Table 23,
In the analysis of the concentration data, several observations were
made. First, it was noted that the concentrations at all five stations
tended to change together from one sampling period to another, indicating
that some external factors such as weather conditions were influencing
the emission rate. Also, there was no set pattern in relative concen-
trations measured at the five stations, i.e., one station did not always
have the highest reading and another the lowest. This appeared to show
that the points of emission within the storage area were not constant.
The background values recorded at sampling Station 1 were consistent
from the standpoint of three different evaluation criteria. First, the
concentrations at Station 1 were, with few exceptions, lower than those
at the downwind stations. Second, the arithmetic average concentration
*3
for the 4-week sampling period was 73.4 ug/m , certainly a reasonable
value for this area of the Cincinnati AQCR. Finally, the average con-
centrations for samples taken during working and nonworking periods were
not significantly different--76.1 and 71.7 ug/m^, respectively. This
indicated that measurements at the upwind station were not influenced by
emissions from the sand and gravel operation.
In addition to calculating emission rates for each of the 19 sampling
periods, an evaluation of the effects of four different factors on the
emission rates was desired. These factors were rainfall, wind speed,
type of aggregate, and amount of activity in the piles. Appropriate data
on these four variables for periods concurrent with the sampling were re-
quired for this evaluation. The sources of these data are described
below.
Daily rainfall data at Lunken Airport, shown in Table 22, were used
to determine the effect of a wet aggregate surface on emission rates.
76
-------
Table 23. SUSPENDED DUST CONCENTRATIONS (Sand and Gravel Storage Piles)
Dust Concentration ((j,g/m )
Date
6/9/73
6/11/73
6/12/73
6/13/73
6/14/73
6/16/73
6/18/73
6/19/73
6/20/73
6/21/73
6/23/73
6/25/73
6/26/73
6/29/73
6/30/73
7/2/73
7/3/73
7/5/73
7/6/73
Start Time
1200
1800
1200
1800
1200
1200
1800
1200
1800
1200
1200
1800
1200
1800
1200
1800
1800
1200
1200
Test Period
(hr)
24
12
24
12
24
24
12
24
12
24
24
12
24
12
24
12
12
24
24
Background
Sta. 1
94
95
60
65
139
75
71
49
61
7
67
86
58
-
61
64
50
95
124
Net Downwind
Sta. 2
8
107
85
215
575
3
21
93
Void
152
8
55
121
-
16
20
28
231
362
Sta. 3
23
152
113
125
134
ok/
16
57
Void
140
6
19
134
_
31
17
24
138
170
Sta. 4
49
184
252
15
239
ok/
37
105
48
249
Void
89
50
-
ok/
11
28
146
332
Sta. 5
13
172
208
ok/
175
7
42
74
2
154
9
33
202
-
31
71
22
150
183
Sta. 6
4
76
147
125
259
26
Void
170
Void
108
27
210
Void
-
42
Void
19
40
241
Average
19a/
138
161
96
276
?*/
29
100
25
161
12-/
81
127
-
?4§/
30
24
141
258
Average
73
124
76
108
86
107
a/ Weekend sample.
k/ Slightly negative net value; assumed = 0.
-------
Since the high-volume samples ran from noon of one day until noon of the
next or from 6:00 PM until 6:00 AM of the next day, a wet sampling period
was taken to be one in which there was measurable rainfall on either of
the 2 days or the day preceding the first day of the sampling period. If
only a trace of precipitation were recorded on one of the sampling days,
it was still counted as a wet period. However, trace precipitation on
the day preceding sampling did not classify the period as wet.
Since the on-site wind speed data agreed well with corresponding data
from Lunken Airporta the on-site readings were used in the analysis.
Average wind speeds for periods coincident with the high-volume sampling
periods were obtained directly from the already-prepared wind speed
summaries.
Dravo personnel at the sand and gravel pit provided information on
the grade of aggregate in each storage pile. The grade of gravel or stone
is shown in Figure 16 for each pile. Equivalent aggregate size ranges for
these grades are presented in Table 24.
The amount of activity in the stockpile area on sampling days could
only be obtained indirectly from Dravo's available records. Weights of
total material excavated/sized and material loaded onto trucks for ship-
ment were kept for each day, and are presented in Table 22. The difference
between these two values provided one estimate of the net weight of material
put into or taken out of storage for the day. However, these data proved
to be inadequate for comparison with the calculated emission rates for
individual sampling periods for the following reasons:
1. The difference between the two values included the weight of material
processed and then shipped directly, so was not a good indicator of
storage area activity;
2. The time periods for recording material movement were not coincident
with sampling periods; and
3. Complete records were not maintained for the entire sampling period.
As an alternate evaluation procedure, the emission rates during
working periods were simply compared with those during nonworking periods,
when only wind erosion of the piles caused emissions. Since all the sam-
ples taken of working periods were 24-hr samples and therefore contained
12 to 14 hr when no activity occurred in the storage area, an emission
rate was also calculated for just the portion of these periods when
activity actually took place. This was accomplished by determining the
equivalent concentration for the 12 working hours that would result in
a normal 24-hr concentration when combined with the average 12-hr measure-
ment for nonworking periods.
78
-------
Table 24. AGGREGATE SIZE RANGES
Grade Range of Aggregate Sizes (mm)
No. 6
No. 8
No. 9
No. 57
No. 67
No. 304
Construction Sand
9.5-19
2.9- 9
1.3- 4
4.8-25
4.8-19
0.2-25
0.2- 2
.0
.5
.8
.4
.0
.4
.0
Source: National Crushed Stone Association
79
-------
The calculations and graphical analyses employed to determine the
effect of the four factors on emission rate are presented in the follow-
ing section.
Correction Factors
The effect of potentially important correction parameters on dust
emissions from aggregate storage piles was assessed by examining the
correlation between net downwind dust concentrations and parameter
values. The results are described in the following paragraphs.
Rainfall - Using the criteria established above to separate the sampling
periods into wet and dry periods, average particulate concentrations were
calculated for the two different conditions. On days when the piles were
dry, the average concentration caused by the piles (background subtracted)
was 141 ug/m , while on rainy days when the piles were wet this average
concentration was only 70 ug/rn^. Wind speeds were approximately the same
for the rain and no-rain sampling periods, so the emission rates esti-
mated by the procedure explained in the previous section would be in
the same ratio as the high-volume measurements—approximately twice as
great during dry periods.
A similar relationship was observed for the background readings
measured at Station 1. The average values during wet and dry periods
were 59 and 102 ug/m^, respectively. This may indicate that relative
emissions from wet and dry storage piles are part of a much broader re-
lationship of fugitive dust sources during wet and dry periods. Under
this premise, much more data should be available and should be utilized
in developing a correction factor for the effect of surface moisture on
stockpile emission rates.
There were no extended periods without rain during the month of sam-
pling to investigate whether the emission rates increased proportionately
with the time span since the last rainfall.
An additional subdivision of the data into periods when the piles were
(a) active and (b) inactive, as shown in Table 25, showed that wet piles
did not reduce the emission rate by half for either data subset. However,
the wet piles emitted significantly less dust in both cases. Thus, it
appears that emission rates may vary by at least a factor of two-fold
between wet and dry periods or between wet and dry climates.
Wind Speed - Based on theory, the wind speed should affect high-volume
measurements downwind from the storage piles in at least two different
ways. First, atmospheric dispersion equations such as those presented
in the Workbook of Atmospheric Dispersion Estimates^' almost universally
80
-------
Table 25. AVERAGE HIGH-VOLUME CONCENTRATIONS DURING
WET AND DRY SAMPLING PERIODS
At FiveDownwind Sites At Background Site
Average No, of Average No. of
Concentration^:' Sampling Concentration Sampling
Stockpile Condition (ug/m ) Periods (iig/mr) Periods
Wet piles, all
sampling periods
Wet piles, active
Wet piles, inactive
70
141
44
11
59
44
67
12
Dry piles, all
sampling periods
Dry piles, active
Dry piles, inactive
141
225
57
102
119
85
a/ Background concentration subtracted.
81
-------
show that the pollutant concentration downwind from a source is inversely
proportional to the average wind speed. These equations assume that the
source strength is independent of wind speed. However, for particulate
emissions from aggregate storage piles and other fugitive dust sources,
it is the force of the wind that at least partially creates the emissions.
Thus, some positive relationship also exists between wind speed and
particulate concentration.
The high-volume measurements shown in Table 23 were plotted against
average wind speeds for corresponding periods in an effort to determine
the resultant, or net, function of concentration versus wind speed. The
plotted data, shown in Figure 17, indicated no well-defined relationship.
In addition to this plot, similar diagrams (not shown) were prepared for
each sampling site, with similar results. Also, data subsets such as wet
days and dry days were evaluated to find an effect of wind speed on down-
wind concentrations. Only 12- and 24-hr periods could be studied, since
particulate concentrations were not available for any shorter averaging
times. The only significant conclusion that could be drawn from these
analyses was that high particulate concentrations were not associated with
periods of high average wind speed.
Therefore, based on these test results, wind speed did not appear to
be a candidate as a correction factor for estimating emission rates from
aggregate storage piles.
Aggregate Size - With the available sampling data, the only method of
evaluating the effect of aggregate size on emission rate was to compare
the average particulate concentration for each site with the size of
aggregate in the nearest pile. This procedure was executed, as shown in
Figure 18. However, this simple analysis did not indicate any apparent
correlation for several reasons:
1. There were only five high-volume sites and therefore only five data
points;
2. Each site was actually impacted by several piles, depending on wind
direction; and
3. The range of aggregate sizes in the separate piles was quite large
(see Table 24), and the size difference between different piles was not
distinct.
As previously noted, the data did not demonstrate a continuing
pattern in the relative concentrations measured at the five sites, so no
"hot spots" of emission within the storage area were suspected.
Also from a theoretical viewpoint, it is doubtful that emission rates
are closely related to aggregate size. Fines that are loosely attached to
the surface of the aggregate, not the aggregate particles themselves,
82
-------
oo
w
12
11
10
9
s 8
UJ J
lit I
Cu.
in
§ 6
l^_f
3=
-------
oo
1 J
12
11
10
E 3
E
•
UJ Q
M 8
H-i
£ 7
-------
become airborne by mechanical entrainment or by wind erosion. Smaller
aggregate may contain more fines because of its greater surface area per
unit volume or because of additional crushing during its production. On
the other hand, rock which is crushed may have more attached fines than
sand or gravel which is mined from dry river beds and processed by just
screening. No data were found to substantiate or quantify either of
these hypotheses.
In summary, aggregate size was not found to be a significant factor
In determining the emission rate from an aggregate storage pile.
Activity in the Storage Area - For reasons already explained the data ob-
tained for activity levels in the processing and loading operations were
not representative of relative activity in the storage area. If good
data for activity in the storage area were available, It is suspected
that a relationship could be established. However, such data probably
would not be available for other sand and gravel operations either, so
would be of very limited use as a correction factor.
Next, a simple analysis was performed comparing measurements taken
on working days with those taken overnight or on weekends, when there
was no activity in the storage area. The average of all samples from
periods with activity was 182.7 ug/m^, while the average for all periods
with no activity was 47.4 ug/nP. Both of these values were after back-
ground had been subtracted.
With this significant finding, the readings for working and nonworking
periods from each individual site were compared to determine how consistent
this observed relationship was. The ratios of working to nonworking periods
varied from 2.4/1 at Station 6 up to 5.2/1 at Station 2, as shown in Table 26.
At all five stations, significantly higher particulate concentrations were
measured when there was activity in the storage area. These results cannot
be attributed to differences in meteorology between the 24-hr sampling
periods and the 12-hr night samples, because the four 24-hr weekend samples
included in the nonworking category had lower readings than the 12-hr night
samples.
Therefore, with no exceptions the data pointed to a definite relation-
ship between emission rates from storage piles and activity in the piles,
and this relationship should be reflected in the development of an emis-
sion factor.
Computed Emission Factors
The general methodology for estimating emission rates from the aggre-
gate storage area has already been described in the preceding section.
85
-------
Table 26. SAMPLING DATA. FOR WORKING AND NONWORK.ING PERIODS
oo
Average Hi-Vol Concentration (ug/nr)
Site
1
2
3
4
5
6
Gross for
Working Period
76.1
325.8
201.4
297.3
233.4
236.9
Gross for Non-
working Period
71.7
119.8
113.0
117.8
108.2
137.8
Net for
Working Period
0
249.7
125.3
221.2
157.3
160.8
Net for Non-
working Period Ratio
0
48.1
41.3
46.1
36.5
66.1
5.2
3.0
4.8
4.3
2.4
-------
Briefly, it was assumed that all emissions from the stockpiles passed
through an imaginary vertical plane with the dimensions of the width of
the storage area by the average height of the piles (300 m by 7 m); that
the five samplers located downwind of the piles sampled particulate con-
centrations representative of average particulate concentrations passing
through this vertical cross section; and that the total air volume con-
taining this average concentration could be approximated as the average
wind speed times the area of the cross section (2,100 m^).
Emission rates were calculated for two conditions—active piles and
inactive piles. The air volume per day was estimated as 2,100 m times
the average wind speed of 3.12 m/sec, or 5.66 x 10^ m . For average con-
centrations of 182.7 Pg/m^ and 47.4 pg/nr* for working and nonworking days,
the emissions from the study area were calculated to be 103 and 26.8 kg/day,
respectively. —
Since the 24-hr samples included a time period during which there was
no activity, it also appeared reasonable to estimate a shorter-term emis-
sion rate for just that portion of the 24 hr during which the activity
actually took place. This was accomplished by determining the equivalent
concentration for the 12 working hours that would result in a 24-hr
average of 182.7 ug/trP when combined with a value of 47.4 ug/m for the
12 nonworking hours. This value was calculated to be 318.0 pg/itr and
resulted in an estimated hourly emission rate of 7.5 kg/hr by using the
same methodology as above. This value would be applicable only for
short-term emission rates, not for general emission inventory work.
Emission rates from the study area can be used to estimate emission
rates from other similar operations only after they have been normalized
with an appropriate parameter of the operation's size or production rate.
The two parameters which appear to be appropriate for aggregate storage
areas, and for which survey data could be obtained, are the acreage of
the storage area and the tons of material placed in storage (eliminating
the time variable). The calculated emission factors are shown in Table 27.
As specified previously, the above emission factors include the
emission contributions from the movement of traffic among the storage
piles and from loading and unloading operations, plus wind erosion. They
do not include emissions from the mining or processing of the aggregate
or from traffic movement in other parts of the plant. It should also be
restated that these factors are not universally applicable, but are
intended to be representative for storage piles in areas of the country
with climatic conditions similar to Cincinnati, Ohio.
As noted in Chapter 2, the only published emission factor for aggre-
gate storage pile losses (in rock handling operations) was reported in
87
-------
Table 27. CALCULATED EMISSION FACTORS
(Aggregate Storage Piles)
Storage Pile Activity
Active^/
Inactive (wind erosion)
Normal mix-
Emission Factor
(Ib/acre of storage/day)
13.2
3.5
10.4
(Ib/ton placed in storage)
0.42
0.11
0.33
£/ Eight to twelve hours of activity per 24-hr period.
b/ Five active days per week.
-------
the April 1973, edition of Compilation of Air Pollutant Emission Factors-—'
as 10 Ib/ton (5 kg/metric ton). This value is approximately 24 times as
high as the factor developed in the present study--0.42 Ib/ton, for sand
and gravel storage piles with daily activity.
EMISSIONS FROM AGGREGATE LOADOUT
Sampling Site Description
Originally a crushed limestone operation in South Kansas City was
designated for the study of atmospheric dust emissions from aggregate
loadout from storage piles. Although testing was scheduled in August
1973, a period of record-breaking wet weather ensued, lasting through
September. Even after 2 weeks of dry weather, the storage piles remained
wet just below the surface and emissions were barely visible. (No
freshly crushed, dry rock had been stockpiled during this period.)
Because at that time the crushed stone sales season was coming to a
close, no further stockpiling was anticipated either at the designated
test site or at other area quarries. This made it necessary to shift the
test site to a crushed stone user operation which stockpiled freshly
crushed rock.
The Royal Asphalt plant in Kansas City, Missouri, was selected for
the testing of emissions from aggregate storage loadout operations.
Royal Asphalt maintained stockpiles of four sizes or blends of crushed
rock.
To avoid possible interference with normal plant operations and to
better control test conditions, testing was scheduled for a weekend. A
truck and high-loader were reserved for the testing.
Field Measurements
Field testing of dust emissions from storage pile loadout of crushed
rock was conducted at the Kansas City site on 17 November 1973. The
asphalt plant was not in operation during testing. A high-loader and a
dump truck with a load capacity of about 15 tons were rented for this
study.
Table 28 specifies the kinds and frequencies of field measurements
that were conducted during each run. "Composite" samples denote a mix-
ture of single samples taken from several locations in the area;
"integrated" samples are those taken at one location for the duration of
the run.
89
-------
Table 28. FIELD MEASUREMENTS--AGGREGATE LOADOUT
Test Parameter
Units
Sampling Mode
Measurement Method
Meteorology
a. Wind speed mph
b. Wind direction deg
c. Cloud cover 7<,
d. Temperature °F
e. Relative humidity 7<,
Continuous
Continuous
Single
Single
Single
Recording instrument at "background"
station; sensors at reference height
Visual observation
Sling psychrometer
Sling psychrometer
Aggregate
a. Size mm
b. Moisture content %
c. Age days
d. Pile configuration
Composite
Composite
Single
Single
Dry sieve analysis
Weight loss on oven drying
Plant records
Observation (photographs)
Loading Operations
a. Type of equipment
b. Load capacities
c. Number of loads
tons
Multiple
Multiple
Cumulative
Observation (photographs)
Plant records
Observation
Suspended Dust
a. Exposure profiles
b. Size distribution
(by weight)
c. Concentration
d. Background
concentration
e. Duration of
sampling
o
mg/in'1
urn
Jig/m3
Jig/in-
min
Integrated
Integrated
Integrated
Integrated
Cumulative
Isokinetic high-volume filtration
(MRI method)
Cascade impaction
High-volume filtration
(ErA method)
Timing
-------
Composite samples of aggregate (12 scoops) were obtained from
various points on the worked area of the pile being loaded. The aggre-
gate samples were sealed in polyethylene bags and returned to MRI for
laboratory determination of texture and moisture content.
At the end of each run, the collected samples of dust emissions were
carefully transferred to shipping containers within the MRI instrument
van, to prevent dust losses. After tapping each grid sampler tip so that
dust was dislodged onto the filter, the filters were carefully inserted
into glycene envelopes which were, in turn, put into paper envelopes.
High-volume filters were folded and placed in individual folders. Cas-
cade impactor collection papers were left in place within the impactor
unit.
Table 29 presents information on the time of each run, the prevailing
meteorological conditions and the weight of aggregate loaded. The exposure
profiler was not operated while the truck was dumping its load, but the
other sampling instruments were operated continuously during the run.
Test Results
Dust samples from the field tests were analyzed gravimetrically in
the laboratory. Filters were conditioned in a controlled temperature-
humidity environment prior to weighing. Water rinses from exposure probes,
deposition samplers and saltation catchers were evaporated on a steam
bath in tared beakers, after which the beakers were conditioned and
weighed.
The measured dust emissions from aggregate storage loadout are pre-
sented in Table 30. The dust quantities are the amounts generated per
ton of aggregate loaded.
The total dust emissions for a given run are the sum of the integrated
exposure (above the background) and the amount of deposition between the
back of the truck and the exposure profiler, a distance of 5-6 ft. Since
only very large particles, which settle quickly, would not reach the ex-
posure profiler, this fraction of the deposition was not considered as a
significant air pollution problem.
The suspended dust measurements used to compute the integrated ex-
posure are presented in Table 31. Point values of exposure are converted
to concentration. The concentration measured by the standard high-volume
unit, which was positioned to the side of the profiler, is also presented.
91
-------
Table 29. EMISSIONS TEST PABAMETERS (Crushed Stone Storage Piles)
Time S
Run
15
16
Start
0831
1107
Finish
1018
1226
Duration of
ampling Exposure
(min)
61.2
59.1
Ambient
Temperature Wind Direction/
(°F) Speed (12 ft)
51 S/12.6
57 S/14.0
Cloud
Cover
5
20
Pasquill
Stability a/
D
D
Aggregate Loaded
Dumps
86
80
Tons
150
150
N!
ji/Pasquill Stability Classes; A - Extremely unstable
B - Unstable
C - Slightly unstable
D - Neutral
E - Slightly stable
F - Stable to extremely stable
-------
Table 30. MEASURED DUST EMISSIONS (Crushed Stone Storage Piles)
Run
15
16
Background Concentration
Cug/m3)
334
334
Z
Integrated Exposure
(Ib/ton)
0.11
0,11
Hi-Vol
MMDS/
((J,ra)
1.4
_
_a/ Mass mean diameter of suspended dust, measured with Andersen high-volume
cascade impactor.
-------
Table 31. PLUME SAMPLING DATA (Aggregate Storage)
Height above
Run Truck (ft)
15 0
3
3
3
3
6
1.75S/
16 0
3
3
3
3
6
1.75&
Sampler Location
Lateral Displacement from
Center of Truck (ft)
0
4.5 Right
1.5 Right
1.5 Left
4.5 Left
0
1 Right
0
4.5 Right
1.5 Right
1.5 Left
4.5 Left
0
1 Right
Sampling
Rate
(cfm)
0.570
0.570
0.570
0.570
0.570
0.570
18
0.577
0.577
0.577
0.577
0.577
0.577
33.5
Concentration
(mg/m3)
43.4
46.3
49.0
62.9
43.7
27.5
26.7
18.4
25.4
45.6
56.8
39.1
63.2
25.7
Unit Exposure
(mg/in.2/ton)
2.59
2.76
2.92
3.74
2.60
1.64
1.07
1.48
2.66
3.32
2.28
3.68
a/ Andersen impactor.
b_/ Standard high-volume sampler.
-------
Also given in Table 30 is the mass mean diameter of suspended dust
particles measured with the Andersen high-volume cascade impactor. The
diameter values are aerodynamic measures which treat particles as equiva-
lent spheres with a density of 2.5 g/cm^. The complete size distribu-
tions is shown in Figure 19.
Two potentially significant sources of error in the particle size
measurements as mentioned in Chapter 4 are:
1, The impactor samples nonisokinetically through the high-volume en-
closure openings and collects large particles with low efficiency.
2. Unlike urban aerosol, aggregate particles are dry and brittle and are
subject to bouncing and reentrainment from impaction surfaces.
Both of these factors cause apparent size determinations to be biased.
in the direction of small diameter.
Table 32 gives the results of the laboratory analyses of the samples
of aggregate from the test piles. Moisture content was determined by
weight loss on over drying and particle size analysis by dry sieving.
As expected the moisture content of the aggregate was very low. This
confirms near maximum dust generating potential of the aggregate.
The particle size analyses of the aggregate samples indicate that
the 3/8-blend had more fine sand than the 1/2-straight rock, but slightly
less silt. The size distributions are plotted in Figure 20.
The effective cut-off diameter for capture of dust by a standard
high-volume sampler (or a high-volume cascade impaction operated within
a high-volume enclosure) is taken to be 30 urn for a particle density of
2.5 g/cm^. This value is based on (1) Lundgren's result, (2) the
settling characteristics of aggregate particles and (3) the observed
ratios of total high-volume concentration to isokinetic profiler concen-
tration.
Computed Emission Factors
In the determination of emission factors for aggregate loadout, dust
which settled out before reaching the exposure profiler (within 6 ft of
drift distance from the downwind edge of the truck bed) was not included
in the emission factor; these particles are larger than 150 pm for winds
exceeding 10 mph.
95
-------
WEIGHT % GREATER THAN STATED SIZE
100.0 "
10.0 -
LU
I—
LU
LU
_l
0.1
0.01 • b
Mlf *»*
M M
it!
Andersen Results
g(Storage Pile Loadout):
I
• _
fST
it±
WEIGHT % LESS THAN STATED SIZE
Figure 19. Particle size distribution—aggregate loadout emissions.
96
-------
Table 32. AGGREGATE PROPERTIES (Crushed Stone Storage Piles)
Aggregate Size (% by weight)!*/
Aggregate Moisture Gravel Coarse Sand
Run (Type) Age (%) (>2000 um) (2000-420 urn)
15 3/8 in. Blend 1 week 0.3 97.3 0.3
16 1/2 in. Straight 1 week 1.1 94.9 2.5
Fine Sand Silt
(420-74 iam) (<74 pm)
1.1 1.3
0.7 1.9
a/ Determined by dry sieving method.
-------
WEIGHT % GREATER THAN STATED SIZE
100.0
MM Wl
10.0
1.0
Q
LU
_I
U
0.1
0.01
MM *t
JSljCrushed Stone Aggregate Size
O 3/8 - In Blend
A 1/2 - In Straight
WEIGHT % LESS THAN STATED SIZE
Figure 20. Aggregate size distribution—crushed stone.
98
-------
The equations for calculation of the emission factors for three
particle size ranges (< 2 um, 2-30 um, > 30 um) are as follows:
e< 2 = ER6F< 2
e2_30 = ER6(1 - F< 2)
e> 30 =
where e^ = mass of dust emissions with diameter i per ton placed in storage,
E = integrated exposure measurement,
Rg = ratio of dust concentration measured by the standard high-
volume sampler to the concentration measured by the
isokinetic profiler, at 6-ft height,
F< 2 = fraction of the particles less than 2 ^m in diameter,
measured by high-volume cascade impaction.
The calculated emission factors are presented in Table 18.
Emissions during testing visually appeared to be very high, and may
have approached a maximum for the following reasons:
1. The aggregate tested had been crushed within the previous week and
had remained completely dry.
2. The wind velocity was high (beyond the point of incipient wind
erosion).
3. The two sizes of aggregate were relatively small and contained a
substantial amount of fines.
As indicated in Table 33 there is little difference in emission fac-
tors for the two sizes. Because the potential dust generation during
these tests was near the maximum, an average value for the emission fac-
tor is thought to be about 0.05 Ib/ton.
COMPARISON OF AGGREGATE EMISSION FACTORS
Total dust emissions from aggregate storage piles can be divided into
the contributions of several distinct source activities which occur within
the storage cycle:
1. Loading of aggregate onto storage piles,
2. Equipment traffic in storage area,
3. Wind erosion, and
4. Loadout of aggregate for shipment.
99
-------
o
o
Table 33. CALCULATED EMISSION FACTORS
(Crushed Stone Storage Piles)
Integrated
Exposure
Run (Ib/ton)
15 0.11
16 0.11
Ratio Fraction of
Hi-Vol Catch: Hi-Vol Catch Emission Factors (lb/ton>§/
Profiler Catch < 2 u d > 50 jjpi 2 < d < 50 yjm.
0.48 0.67 0.057 (52%) 0.018 (16%)
0.57 0,6?k/ 0.047 (43%) 0.021 (19%)
d < 2 yea. Total
0.035 (32%) 0.11
0.042 (38%) 0.11
a/ d = partial diameter.
b/ Estimated value.
-------
Although the test results presented in this chapter are limited, a
comparison can be made to estimate the relative contributions of each
of the source activities. The validity of the comparison of test results
for different types of aggregate is best substantiated by the consistency
of the data.
Table 34 shows the contribution of each source activity to the total
dust emissions from aggregate storage piles. The total emission factor
and the wind erosion contribution were determined from the testing in the
Cincinnati area, and the contributions from the aggregate transfer opera-
tions were estimated from the results of the aggregate loadout tests in
the Kansas City area. The contribution of vehicle traffic was determined
by difference; its relatively high value is confirmed by visual observa-
tion of dust emissions from aggregate storage areas.
CORRECTED EMISSION FACTOR
Also shown in Table 34 are the correction parameters which differen-
tiate the emissions potential of one aggregate storage area from another.
For every contributing source activity, the correction parameter is
climatic in nature. Overall the preceipitation-evaporation index best
characterizes the regional variability of total emissions from aggregate
storage piles. The PE index is 103 for Cincinnati and 96 for Kansas City.
The corrected emission factor which can be used to estimate the total
amount of dust emissions with drift potential greater than 1,000 ft, i.e.,
particles smaller than 30 urn in diameter, is given by the following
expression:
(aggregate) (PEAOO)2
where e *= emission factor (pounds per ton placed in storage), and
PE = Thornthwaite1s precipitation-evaporation index.
101
-------
Table 34. AGGREGATE STORAGE EMISSIONS BREAKDOWN
o
ro
Source Activity
Loading onto piles
Vehicular traffic
Wind erosion
Loadout from piles
Total
Correction
Parameter
PE index
Rainfall
frequency
Climatic
factor
PE index
Emission Factor (total
storage cycle) (Ib/ton)
0.04
0.13
0.11
0.05
0.33
Approximate
Percentage of Total
12
40
33
15
100
-------
CHAPTER 7
BUILDING CONSTRUCTION EMISSIONS
Under a separate contract from EPA, PEDCo-Environmental conducted a
field investigation of atmospheric dust emissions from construction
activities in the Southwest. A preliminary report-!' on the findings was
submitted to EPA during February 1973. This section provides a further
analysis of the sampling data from two construction sites in order to
develop an emission factor for this source category and to evaluate sev-
eral factors which affect the emission rate.
The original analysis of fugitive dust emissions from construction
activities was based upon limited data available at the time of report
preparation, and as such the conclusions derived therefrom were con-
sidered only preliminary. This supplemental evaluation is based upon
all the sampling data which were collected at two locations, namely,
Paradise Valley in Phoenix, Arizona, and a construction area in Las Vegas,
Nevada. The conclusions which are derived from this larger data base,
while not significantly different from the initial findings, do point
to a slightly lower emission factor from construction activities.
The Paradise Valley construction site was an 80-acre residential
development with a shopping center. Because atmospheric dust emissions
from the construction activity were generated by diffuse and variable
operations, conventional high-volume samplers, operated for 24-hr periods,
were used to measure emissions.
PARADISE VALLEY CONSTRUCTION STUDY
Figure 21 shows the locations of six sampling stations in relation
to the construction site in Paradise Valley. Samples were collected
periodically at these stations between 31 August and 22 October 1972. A
daily record of construction activity at the site was maintained through-
out this period.
103
-------
C-12
-------
TestResults
An examination of the particulate concentrations obtained at the
sampling locations revealed that Station C-12 usually recorded abnormal
values which were not representative of either normal background concen-
trations or concentrations expected to be contributed from the construc-
tion activity. An on-site examination earlier had revealed that this
sampling location was far from an ideal exposure and therefore data ob-
tained from this location were not used for evaluation purposes.
Station C-16 was located farthest from the construction site. Since
it was seldom downwind from the site, it did not show an impact from con-
struction activity. Consequently, data obtained from this location was
also judged unsuitable for evaluation purposes.
Suspended dust concentrations measured at Stations Oil, C-13, C-14
and C-15, grouped according to wind directions, are listed in Table 35.
This breakdown facilitated proper documentation of concentrations at back-
ground and downwind stations and subsequent evaluation of the contribution
from the construction activity.
A cursory examination of pollution roses presented in Figure 22 in-
dicates that the effect of the construction activity was reflected at
sampling Stations C-13, C-14 and C-15 when they were downwind from the
construction site. This occurred during periods when the wind was from
the southwest quadrant, the predominant wind direction during the sampling
period. Under these conditions, Station Oil served as the background
station. It had an average concentration of 130 ug/m .
Station C-13, located just east of the construction site, recorded an
average concentration of about 260 ug/m^. During the periods of southerly,
southwesterly and westerly winds, this station recorded its highest con-
centrations. This definitely reflects the contribution from the construc-
tion site to the concentration at this location.
Station C-14, located northeast of the construction site, also reflects
higher concentrations. The average concentration recorded at this site was
about 225 pg/m^. This was as expected in view of its relative distance
from the construction site compared to C-13, but is definitely indicative of
contribution from the construction activity.
It is also important to note that the respective ordinate lengths of
the pollution rose for this station were smaller than those at Station C-13,
a trend which has been exhibited at Station C-15 as well. Apparently there
were no localized activities downwind from the construction site impacting
on these sampling stations; the effect of the construction activity was
105
-------
Table 35. SUSPENDED PARTICULATE CONCENTRATIONS (ug/m3)
(Paradise Valley: 31 August - 22 October 1972)
Wind Direction
Station
C-ll
Average
C-13
Average
C-14
Average
C-15
N NE E
219 137
130
160
219 142
254 236
166
285
254 229
593 296
161
131
593 190
105 117
130
SE
105
256
155
136
129
156
130
492
349
239
201
282
176
296
171
187
192
204
163
374
198
114
94
S SW W
203 347 152
212 152 95
163
185
170
208 250 153
353 461 212
389 487 375
371 474 294
370 324 280
258 368 251
336
312
70
346 346 250
328 363 169
292 365 141
240
415
NW
28
138
102
42
114
73
83
168
123
47
49
127
103
23
166
194
49
126
112
65
118
24
57
78
Average 105 124 189 310 364 241 68
106
-------
Station C-15
PARADISE VALLEY POLLUTION ROSES
Dust Concentration vs Wind Direction
Station C-14
Station C-il ;i
* '
Figure 22. Pollution Roses - Paradise Valley Construction Site.
107
-------
felt at all these stations, but to a progressively lesser degree depend-
ing on the distance away from the construction site.
Calculated Emission Factors
Since the wind was predominantly from the southwest quadrant during
the sampling study and since the stations were aligned in that direction
from the site, it was possible to determine the construction site source
strength values using dispersion equation calculations. The procedure
is outlined below.
For aparticular wind direction of interest:
I. (a) Determine the average concentrations recorded at downwind stations
(in this case, Stations C-13, C-14 and C-15).
(b) Determine the average concentration recorded at background station
(in this case, Station C-ll).
(c) Determine the source strength using dispersion equations.
II. (a) Determine the average concentration recorded at one of the down-
wind stations. For this purpose, it is desirable to use the
closest station downwind from the construction site, since the
distance of plume travel will be short and as such the cumulative
effects of local terrain features will be small.
(b) Determine the average concentration recorded at background station.
(c) Determine the source strength using dispersion equations.
If the source strength values obtained in steps I(c) and II(c) above
are approximately the same, and if similar values are obtained for S, SW,
and W winds, it can be concluded that this estimation technique provides
reproducible results and is descriptive of the actual emission rates.
The calculations for the three wind directions are presented in
Appendix B and summarized in Table 36.
It is evident from these results that the source strength values cal-
culated for the southwesterly winds are comparable and closer to each
other than the other two pairs of values. This is probably because the
sampling stations are lined up best for the southwesterly winds. Con-
sequently, it may be concluded that the values of 1.37 and 1.41 tons/acre/
month are closer to the actual emissions from the construction site. A
value of 1.4 tons/acre/month will be used for the average dust emission
factor.
108
-------
Table 36. CALCULATED EMISSION FACTORS
(Paradise Valley Construction Site)
Q, Emissions (tons/acre/month)
Based on Average of
Wind Direction Based on C-13 Only C-13, C-14 and C -15
Southwest 1.37 1.41
South 1.13 1.51
West 0.42 0.65
o
-------
Correction for Activity Level
An activity log was maintained during the sampling period on daily
activity level at the Paradise Valley construction site. Information
obtained on the activity level was grouped into one of three categories—
no activity, light to moderate activity, and heavy activity. Granted
that such categorization was based more upon subjective evaluation
rather than quantifiable parameters, it was hoped that such an analysis
might yield a significant difference in respective fugitive dust emission
rates.
Table 37 presents the measured particulate concentrations at the
four sampling stations subdivided by activity level. The average concen-
trations for the various levels of activities do indicate a correlation
between emission rate and activity level, as shown in Table 38.
Quantification of emissions associated with the level of activity
should not be determined using just the above breakdown, since this break-
down includes data collection from all wind directions. Therefore, a
further breakdown was made to separate the data collected when the wind
was from the southwest quadrant (W, SW and S winds). This data analysis
is shown in Table 39.
It is evident from Table 39 that there is not sufficient data to
quantify the source emissions associated with each activity level. For
the "no activity" category, there are insufficient data with, at best,
one value. The comparison is further complicated by the fact that
emissions were reduced during some of the sampling periods by application
of water on the construction site.
For these reasons, it was not possible to quantify emissions associated
with activity level. However, from the above two tables and from an ex-
amination of individual readings, it can generally be concluded that:
1. Light to moderate activity does not produce significantly higher
emissions than no activity; and
2. Watering does not always show reduced emissions. This may be explained
by the fact that watering is applied only on days that are extremely dusty
or when heavy activity is expected.
LAS VEGAS CONSTRUCTION STUDY
Figure 23 shows the locations of five sampling stations in relation
to the construction site in Las Vegas. The sampling program was conducted
during the period between 21 August and 22 October 1972.
110
-------
Table 37,
ACTIVITY LEVEL VS PARTICULATE CONCENTRATION
(Paradise Valley)
No Activity
Sta-
tion Date
C-ll 9-30-72
9-2-72
9-24-72
10-4-72
10-8-72
10-18-72
10-22-72
9-10-72
10-14-72
Avg .
C-13 9-2-72
9-24-72
10-8-72
10-18-72
10-22-72
9-10-72
10-14-72
Avg,
C-14 9-30-72
9-2-72
9-24-72
10-4-72
10-8-72
10-18-72
10-22-72
9-10-72
10-14-72
Avg .
C-15 9-30-72
9-2-72
10-8-72
10-18-72
10-22-72
9-10-72
10-14-72
Average
Hind
Dir.
W
SE
E
NW
NW
NW
NW
D.I.
D. I,
SE
E
NW
NW
NW
D.I.
D. I.
W
SE
E
NW
NW
NW
NW
D. I.
D.I.
W
SE
NW
NW
NW
D.I.
D. I.
Light/
Wind Conccn-
Spee-.l tration
5
7
2
2
2
3
2
8
Calm
7
2
2
3
2
8
Calm
5
7
2
2
2
3
2
8
Calm
5
7
2
3
2
8
Calm
185
105
160
28
138
42
73
97
95
103
130
285
168
47
127
147
186
156
312
176
113
23
166
49
126
117
205
143
415
163
65
24
76
103
121
138
Date
9-20-72
9-28-72
10-6-72
10-2-72
9-18-72
9-22-72
10-10-72
10-20-72
9-20-72
9-28-72
10-2-72
9-18-72
9-22-72
10-10-72
10-20-72
9-20-72
9-28-72
10-6-72
10-2-72
9-18-72
9-22-72
10-10-72
9-20-72
9-28-72
10-2-72
9-18-72
9-22-72
10-10-72
10-20-72
Moderate Activity
Wind
Dir .
W
W
W
SE
E
E
:-;w
NW
w
W
SE
E
E
NW
NW
W
W
w
SE
E
E
NW
W
W
SE
E
E
MW
NW
Wind
Speed
2
2
2
3
2
2.5
2
2
2
2
3
2
2.5
2
2
2
2
2
3
2
2.5
2
2
2
3
2
2.5
2
2
Concen-
trat ion
95
163
170
136
137
130
102*
114*
131
212
375
239
236
166
125*
99*
207
251
336
70
187
296
161
194*
214
141
240
114
117
130
118*
57*
131
Heavy Activity
Wind
Date Dir.
8-31-72
9-6-72
9-12-72
9-4-72
9-14-72
9-8-72
9-26-72
10-16-72
10-12-72
9-6-72
9-12-72
9-4-72
9-14-72
9-8-72
9-26-72
10-16-72
10-12-72
8-31-72
9-6-72
9-12-72
9-4-72
9-14-72
9-8-72
9-26-72
10-16-72
10-12-72
8-31-72
9-6-72
9-12-72
9-4-72
9-14-72
9-8-72
9-26-72
10-16-72
10-12-72
SW
sw
SW
s
s
SE
SE
SE
N
SW
sw
s
s
SE
SE
SE
N
SW
sw
sw
s
s
SE
SE
SE
N
SW
SW
SW
s
s
SE
SE
SE
N
Wind Concon-
Saeed tration
9
3
6
6
7
6
1 .5
3
2
1
6
6
7
6
1. 5
3
2
9
3
6
6
7
6
1.5
3
2
9
3
6
6
7
6
1.5
3
2
347
152
152
203*
212
256
155
129
219
203
461
487
353*
389
492
349
201
254
373
324
280
368
370*
258
296
171
192
593
317
363
169
365
328*
292
374
198
94
105
254
* Indicates no watering applled
D.I. means direction indeterminate
in
-------
Table 38, DUST CONCENTRATION YS ACTIVITY LEVEL
Average Concentration (iig/nr)
Light to Moderate
Station No Activity Activity Heavyr: Activity
C-ll 103 131 203
C-13 156 207 373
C-14 143 214 317
C-15 138 131 254
Average 135 171 287
112
-------
Table 39. ACTIVITY LEVEL VS CONCENTRATION
FOR W, SW AND S WINDS
Light to Moderate
Station No Activity Activity Heavy Activity
C-ll 185 95 347
163 152
170 152
203
212
Average 185 143 213
C-13 212 461
375 487
353
389
Average -- 294 423
C-14 312 251 324
336 280
70 368
370
Average 312 219 336
C-15 415 141 363
240 169
365
328
292
Average 415 191 303
113
-------
RESIDENTIAL
CO
at:
:D
i—
(-)
LU
Q
in
I/)
CAPRI
CASHMAN MOBILE
JR. HIGH PARK
SCH.
D
RESIDENTIAL
W. PENNWOOD AVE
CLARK
HIGH
SCH.
X
C-23
W. DESERT INN RD
|CfCO.
I FIR
W. SPRING MT. RD.l STA
C-21-—
DRAKE
ESTATES
CONSTRUCTION
SITE
PLEASANT VALLEY
CASCADE MOBILE
HOMES
*=SAMPLING STATION
Figure 23. "Las Vegas construction site.
114
-------
Test Results
Data collected during this sampling program have been grouped accord-
ing to wind direction and are shown in Table 40, and in Figure 24 in the
form of pollution roses for each sampling station.
An examination of tabulated data and the pollution roses developed
therefrom indicates that Station C-21, which was located just south of the
construction site (see Figure 23), recorded higher particulate concentra-
tions during northerly winds than during the periods when the wind was
from other directions. Therefore, it was concluded that the only local
activity which contributed particulate emissions to this station was the
construction activity under study.
Station C-22, which was located north of the construction site,
recorded higher concentrations during southerly and southwesterly winds,
which may be attributed to the construction activity. However, this
sampling station also recorded high concentrations during northerly and
westerly winds. With winds from those directions, the effect of the
construction site should not be felt at this sampling station, thus
strongly indicating that there were other localized activities in the
vicinity of this station which contributed to higher concentration.
Data collected at Station C-23, which was located northeast of the
construction site, also indicate possible contribution from localized
activities other than the construction activity. This is evident from
the higher concentrations recorded during northerly, northeast, southeast
and perhaps westerly winds also. Higher concentrations recorded during
southwesterly winds may be attributed to construction activity but can
possibly be attributed to localized activities immediately west of the
sampling station.
Station C-24 might have had interference from localized activities as
evidenced by higher readings during northerly winds. The interfering
source(s) could be the same located north of this station, which con-
tributed to higher concentration at C-23 during southeasterly winds.
Station C-25, which was located on the premises of Clark High School,
recorded concentrations comparable to expected ambient concentrations.
From the above analysis, it appears that all the sampling data
collected at these stations cannot be used to evaluate the effect of
the construction site activity because of possible interferences at
some stations from other localized activities, even though the predom-
inant wind as determined from the collected meteorological data was from
the southwest and the locations of the sampling stations appear to be
115
-------
Table 40. IAS VEGAS SITE - SAMPLE VALUES (pg/mj)
SAMPLING PERIOD - 21 AUGUST - 22 OCTOBER 1972
Station
C-21
Average
C-22
Average
C-25
Average
C-23
Average
C-24
Average
N ME E SE
48 48 60
717 143 68
204 18 73
255 49
3Q6 70 63
46 56 64
314 97 69
152 44 52
71 80
146 66 66
47 77
67
61
47 68
102 109 89
336 205 196
112 142
133 164
188
171 157 156
73 39 56
206 74 94
64 79 86
88 52
89
108 67 75
S SW W NW
49 66 83
S3
147
19
100
196
122
46
37
34
41
42
45
49 75 83
122 38 102
125
127
126
151
79
220
135
0 f\
Q'J
99
132
94
104
263
122 127 102
57 74
74
54
73
115
83
33
46
27
57
32
85
61 74
85 228 127
300
238
236
128
194
104
127
148
69
139
71
37
68
76
85 144 127
75 173 99
97
84
97
115
230
114
47
106
128
72
54
57
128
75 107 99
116
-------
• tation C-25
Station C-21
LAS VEGAS POLLUTION ROSES
Dust Concentrotion vs Wind Direction
Stotion C-22 _N
,«?
Station C-23
100
Figure 24. Pollution Roses - Las Vegas Construction Site.
117
-------
lined up best for this wind. On the other hand, it would appear that for
northerly winds, all the sampling data collected can be used to estimate
the contribution from the construction site with Station C-21 serving as
downwind station and Stations C-22, C-23 and C-25 serving as background
stations. It should be mentioned that for this wind, even though the
background stations' readings might reflect interferences from other
sources, the contribution of the construction site will be superimposed
upon these readings and will be reflected at Station C-21.
ComputedEmissionFactors
With the knowledge that the sampling stations originally were located
to reflect only the contribution from the construction activity, a check
on the validity of the collected data was made using the following
methodology. The collected data have been separated out for the desired
wind directional analysis and are given in Table 41.
I. For southwesterly wind
(a) Determine average concentration recorded at Stations C-22 and
C-23 and assume this value to reflect particulate contribu-
tion from the construction site.
(b) Determine the average concentration at background station
(Station C-21).
(c) Determine source emission strength of the construction activity
using dispersion calculations (calculations similar to the
ones performed earlier).
II. For northerly wind
(a) Determine average concentration recorded at Station C-21 and
assume this to reflect contribution from the construction
site.
(b) Determine average concentration at background Stations C-22 and
C-23.
(c) Determine source emission strength using dispersion calculations.
If the source strength values obtained in steps I(c) and II(c) are
comparable to each other, then we can assume that the effect of localized
sources were negligible during southwesterly winds and the apparent dis-
tortion of pollution rose might be due to the micrometeorology of the study
area. On the other hand, if these values are not comparable, then we can
assume that the localized sources did have an effect in the recorded con-
centrations at some of these stations. In this case, the value determined
in step II(c) for northerly wind can be considered to be representative of
118
-------
Table 41. MEASURED CONCENTRATIONS DURING N, NE, S AND SW WINDS
(Hg/m3)
Date C-21
8-21-72 48
8-23-72 717
8-25-72 204
8-27-72 255
Avg=306
9-2-72 48
9-4-72 143
9-18-72 18
Avg=70
10-8-72 49
8-29-72 66
9-10-72 147
9-22-72 19
9-24-72
9-26-72
9-30-72 100
10-2-72 196
10-4-72 122
10-6-72 46
10-10-72 37
10-12-72 34
10-16-72 41
10-18-72 42
10-20-72 45
10-22-72
Avg-75
C-22
46
314
152
71
Avg=146
56
97
46
Avg=66
122
38
-
127
126
151
79
220
135
80
99
132
94
104
263
-
Avg=127
C-23
102
336
112
133
Avg=71
Avg=142
109
-
205
Avg=157
Avg=84
85
228
238
236
128
194
104
127
148
69
139
-
71
37
68
76
Avg=130
Avg=107
C-25
_
-
-
-
_
-
47
46
_
-
57
74
54
73
115
83
33
27
-
57
32
85
-
Avg=63
Wind
C-24 Direc.
73 N
206
64
88
Avg=108
39 NE
74
79
Avg=64
75 S
173 SW
97
84
-
97
115
230
114
47
106
128
72
54
57
128
Avg=107
Speed
8
9
8
6
Avg=7.8
7
9
8
Avg=8. 0
11
8
12
8
8
11
6
9
7
6
9
5
5
5
5
8
Avg=7.5
mph
mph
mph
-------
emissions from the construction site since there are no interferences
surrounding Station C-21.
The results of the calculation exercise as outlined in steps I and II
are given in Table 42.
It is apparent from Table 42 that the source emission strength values
derived for southwesterly and northerly winds are not comparable to each
other. Since the northerly wind direction apparently had the least inter-
ference from other emission sources, a"Q" value of approximately 1.0 tons/
acre/month should be representative of the actual emission rate from this
site.
Correction for Activity Level
An attempt was made to correlate the data obtained from the sampling
program with the activity level at the construction site. The data were
broken down into three categories of activity level (namely, no activity,
light to moderate activity, and heavy activity) for each sampling station,
as shown in Table 43. Within each category, further breakdown was nade by
grouping the data into different sectors of wind directions, and analyzing
for any correlation which existed between the measured concentrations and
the activity level. As can be seen from the summaries in Table 44, it is
not possible to derive any meaningful correlation factors or to quantify
the source emission strengths associated with each activity level.
The reasons for lack of any correlation are suspected to be the same
as those for the Paradise Valley data: (a) the categorization of activity
at the construction site into three groups was based upon subjective rather
than definite emission quantifying parameters; and (b) apparent localized
emissions surrounding some of the sampling locations in this study area
possibly have rendered the data unsuitable for this type of analysis. It
is of interest to note that the data collected during periods of northerly
and northeasterly winds reflect a trend between expected concentration and
activity level. However, these data are insufficient to quantify the
emissions.
120
-------
Table 42, RESULTS OF DISPERSION CALCULATIONS
Wind Direction
Southwesterly
Southwesterly
Receptor
Station (s)
C-22, C-23
C-22 only
Background
Station(s)
C-21
C-21
Stability
Class
C
C
g/sec
20.9
20.3
Q = Emission
ton/year
730
703
Strength
ton/acre /month
0.61
0.59
Northerly C-21 C-22, C-23 C 32.8 1,150 0.96
t->
NJ
-------
Table 43. LAS VEGAS CONSTRUCTION STUDY ACTIVITY LEVEL
VS CONCENTRATION (ug/m3)
Sta-
tion
C-21
Avg.
C-22
Avg.
C-23
Avg.
C-24
Avg.
e-25
Avg.
No
Date
8-27-72
9-2-72
9-4-72
9-16-72
10-8-72
9-10-72
9-30-72
8-27-72
9-2-72
9-4-72
9-16-72
10-8-72
9-24-72
9-30-72
8-27-72
9-2-72
9-16-72
10-8-72
9-10-72
9-24-72
9-30-72
10-22-72
8-27-72
9-2-72
9-4-72
9-16-72
10-8-72
9-10-72
9-30-72
10-22-72
10-8-72
9-24-72
9-30-72
Activity
Wind
Dir.
N
NE
NE
SE
S
sw
sw
N
NE
NE
SE
S
SW
sw
N
NE
SE
S
SW
sw
sw
sw
N
NE
NE
SE
S
SW
sw
sw
S
sw
sw
Concen-
tration
255
48
143
49
49
147
100
113
71
56
97
52
49
126
79
76
133
109
164
85
238
128
104
76
130
88
39
74
52
75
97
115
128
84
46
74
73
64
Light/Moderate Activity Heavy Activity
Date
8-25-72
9-8-72
9-18-72
9-22-72
10-4-72
10-6-72
10-10-72
10-12-72
10-16-72
10-18-72
9-28-72
8-25-72
9-18-72
9-22-72
9-26-72
10-4-72
10-6-72
10-10-72
10-12-72
10-16-72
10-18-72
9-28-72
8-25-72
9-18-72
9-8-72
9-22-72
9-26-72
10-4-72
10-6-72
10-10-72
10-16-72
10-18-72
9-28-72
9-18-72
9-8-72
9-22-72
9-26-72
10-4-72
10-6-72
10-10-72
10-12-72
10-16-72
10-18-72
9-28-72
9-18-72
9-22-72
9-26-72
10-4-72
10-6-72
10-10-72
10-16-72
10-18-72
9-28-72
Wind
Dir.
N
NE
NE
SW
SW
SW
sw
sw
sw
sw
w
N
NE
SW
SW
sw
sw
sw
sw
sw
sw
w
N
NE
SE
SW
sw
sw
sw
sw
SW
sw
w
NE
SE
SW
SW
SW
sw
sw
sw
sw
sw
w
NE
SW
sw
sw
sw
sw
sw
sw
w
Concen-
tration
204
60
18
19
122
46
37
34
41
42
83*
64
152
46
127
151
135
80
99
132
94
104
102*
111
112
205
89
236
194
148
69
139
71
37
127*
130
79
56
84
97
114
47
106
128
72
54
99*
85
47
51
54
83
33
27
57
32
74*
51
Date
8-21-72
8-23-72
9-12-72
9-14-72
8-29-72
8-31-72
10-2-72
!°-29r!2
8-21-72
8-23-72
9-12-72
9-14-72
9-20-72
8-29-72
8-31-72
10-2-72
10-20-74
8-21-72
8-23-72
9-12-72
9-14-72
9-20-72
8-29-72
8-31-72
10-2-72
10-20-72
8-21-72
8-23-72
8-25-72
9-12-72
9-14-72
9-20-72
8-29-72
10-2-72
I°~2i~Z2
9-12-72
9-14-72
9-20-72
10-2-72
10-20-72
Wind
Dir.
N
N
SE
SE
SW
SW
SW
_sw _
N
N
SE
SE
SE
SW
SW
SW
SW
N
N
SE
SE
SE
SW
SW
SW
SW
N
N
N
SE
SE
SE
SW
sw
sw
SE
SE
SE
SW
sw
Concen-
tration
48
717
68"
73
66"
83
196'*
4_5_
162
46
314
64
69
80
38
125
225*
263
136
102
336
196
142
188
228
300
127
68
187
73.
206
64
94
86
8S
17;;
230
57_
119
77
67
61
115
85
8]
indicates no watering applied.
122
-------
Table 44. LAS VEGAS CONSTRUCTION STUDY ACTIVITY LEVEL VS CONCENTRATION
Sta-
tion Wind Direction
C-21 All Directions
C-22
C-23
C-24
C-25
C-21 S, SW
C-22
C-23
C-24
C-25
C-21 N, NE
C-22
C-23
C-24
C-25
Average Concentration (pg/m^)
No Activity
113
76
130
84
64
99
84
126
104
64
149
75
121
67
--
Light to Moderate
64
111
130
85
51
49
115
128
88
48
94
99
159
79
47
Activity Heavy Activity
162
135
187
119
81
97
162
181
153
100
383
180
219
114
_ _
123
-------
SUMMARY AND CONCLUSIONS
The estimated emission values from the two construction sites in
Phoenix and Las Vegas were 1.4 and 1.0 tons/acre/month, respectively.
Based on the same methodology, except for the division of data into in-
dividual wind directions, the preliminary data (first half of sampling
period) had indicated the values to be 1.8 and 1.0. The observed
difference in estimated emission rates between the two construction
sites is attributed to differences in soil texture and to meteorological
factors such as frequency of precipitation, atmospheric turbulence, etc.
For development of an emission factor for widespread use, these two
numbers should certainly not be considered as representative of the full
range of emission rates that might be encountered. To the contrary, both
sampling locations were in the desert southwest, and are therefore probably
much higher than emission rates from similar construction projects located
in more moderate climates. The average of the two values, 1.2 tons/acre/
month, is recommended for use as the high end of the range for this fac-
tor, i.e., appropriate for application in arid areas with watering for
dust control.
Construction activity levels were shown to influence emission rates
from the sites significantly. However, this variation could not be quan-
tified. The final factor represents emission rates during the period of
active construction, including some days with no activity, some with
moderate activity, and some with heavy earth-moving equipment and con-
siderable truck traffic. Substantial error may result if the factor is
applied to a site during a period of extended inactivity.
124
-------
CHAPTER 8
EMISSIONS INVENTORY PROCEDURES
The rational development of an emissions control strategy for a county
or other jurisdiction requires an adequate assessment of the nature and
extent of air pollution in the region involved. This chapter outlines
the procedures for inventorying the source categories treated in earlier
chapters, by applying the corrected emission factor formulations.
SOURCE DATA REQUIREMENTS
Two types of data are needed for the emissions inventory:
Measure of source extent and
Parameters for correction factors.
The specific data requirements for each source category are presented
in Table 45.
Based on information available to us at this time, the following
data on source extent will have to be estimated:
1. Traffic volume on unpaved roads as a function of surface type,
2. Number of agricultural tilling operations as a function of crop
grown, and
3. Acres per dollar value of construction, as a function of construction
type.
With reference to the last item, MRI has developed factors for conversion
of dollar value of construction to acres of construction for major con-
struction categories. These factors are presented in Table 46.
125
-------
Table 45. AREA SOURCE DATA
Source Category
Unpaved roads
Heavy construction sites
Agricultural land tilling
Unpaved airstrips
Measure of Extent
Vehicle-miles traveled, by
road type
Acres of active construction, by
type of construction
Acres by crop grown
Landing/take-off cycles
Correction Parameter
. Average vehicle speed
. Vehicle mix
. Surface texture (silt
content)
. Surface moisture ("dry"
days)
. Soil texture (silt
content)
. Soil moisture
. Activity index
. Surface soil texture
(silt content)
. Surface soil moisture
. Surface soil texture
(silt content)
. Surface soil moisture
Aggregate storage piles
Tons put through storage cycle
. Precipitation-evaporation
index
-------
Table 46. ACRES OF CONSTRUCTION - 1973
Type of Construction
Private residential
Private commercial
Private industrial
Highways and streets
Estimated
Acres per
$106
8.0
2.5
3.0
25.0
All other new construction
Total new construction
1973 New
29/
Construction —
C$106)
60,084
16,259
6,108
10,350
92,801
45,752
138,553
1973
Total
Acres
480,672
40,648
18,324
258,750
798,394
127
-------
As indicated in Table 45, a frequently required climatic parameter for
use in correcting emission estimates is Thornthwaite's precipitation-evapora-
tion index. Figure 25 shows a map of PE values calculated from annual pre-
cipitation and temperature data. Figure 26 gives the conversion of PE
values to the form used in the correction factor. The precipitation fre-
quency for use in the corrected emission factor for unpaved roads, is shown
in Figure 27.
PARTICLE DRIFT POTENTIAL
The impact of a fugitive dust source on the air quality depends on the
drift potential of the particles injected into the atmosphere. This sec-
tion presents a brief analysis of particle drift potential.
The distance that a dust particle will travel from its point of injec-
tion into the atmosphere depends on (1) the injection height of the par-
ticle, (2) the terminal settling velocity of the particle, and (3) the
interaction of the particle with atmospheric turbulence. If the vertical
velocity fluctuations of the turbulent air are of the same order asi the
terminal settling velocity of a particle, the drift potential of the par-
ticle is significantly increased.
Using the fact that the root-mean square vertical velocity fluctuation
is approximately proportional to the wind friction velocity,—' Gillette
and Bliffordi^' have derived ratios of sedimentation velocity to friction
velocity which represents the boundaries of extremes in particle behavior.
These limits have been incorporated into Figure 28, which characterizes
particle behavior as a function of aerodynamic particle diameter and wind
speed. In the development of the curves shown, the friction velocity was
calculated from reference wind speed (12-ft height) and an assumed rough-
ness height of 1 cm (see Figure 2).
The area of Figure 28 labeled "suspension" describes those particles
which have the potential for long-range transport in the atmosphere. For
a given wind speed, this information can be used with the total emission
factor and the particle size data to determine the long-range impact of
dust emissions from a particular source.
WINDBLOWN DUST
As discussed in Chapter 2, soil erosion by wind is recognized as an
important source of atmospheric aerosol. However, relatively little is
known about magnitude of the suspended dust fraction (a relatively minor
portion) of wind erosion transport. Much of the information on the
128
-------
physics of wind erosion has been incorporated into the Wind Erosion
Equation,2£/ which relates the soil loss from an eroding field (i.e.,
the horizontal flux of sand-sized soil aggregate) to individual field
and climatic parameters.
As part of the investigation reported herein, a procedure was developed
for estimating suspended dust emissions from wind erosion. This procedure,
which utilizes the Wind Erosion Equation as a starting point, is delineated
in Appendix A.
129
-------
THORNTHWAITE'S PRECIPITATION - EVAPORATION INDEX
OJ
o
Figure 25, Map of PE values for state climatic divisions,
-------
10
0.907
0,826
0.756
0.694
0.640
0.592
0.549
0.510
0.476
0.444
0.416
0.391
0.367
0.346
327
0.309
0.292
0.277
0.263
0.250
30 50 70 100 110
PRECIPITATION-EVAPORATION RATIO
Figure 26. Moisture correction factors.
130
150
190 200
-------
MEAN NUMBER OF DAYS WITH 0.01 INCH OR MORE OF PRECIPITATION, ANNUAL
U)
A%0
'S0
-.jo loo no 120
i J
~ "i L. J.-^-
Figure 27. Map of precipitation frequency.
-------
200 r-
180 -
: '•'.-.':' Impeded Settling:•.••." .: ••'•;'••• •;;.••• :•.:•••'.•,;•'•
6 8 10 12
REFERENCE WIND SPEED (mph)
Figure 28. Particle settling/suspension regimes.
133
-------
CHAPTER 9
CONCLUSIONS
The major conclusions of this investigation relate to the quantity
and nature of dust emissions from the four source categories studied (i.e.,
unpaved roads, agricultural tilling, aggregate storage piles and con-
struction sites). In addition to the basic emission factors, the analysis
of test results has yielded significant information on correction factors
which account for the variability of emissions from one locality to
another because of differences in climate and in the properties of the
emitting surface.
The emissions of dust from unpaved roads (per vehicle-mile of travel)
is directly proportional to the average traffic speed and to the silt
content of the road surface. The silt content of gravel roads does not
vary significantly, which accounts for the uniformity of emissions from
gravel roads with similar traffic patterns. Emissions are reduced during
periods of rainfall, but quickly return to normal levels. Of the total
dust emissions, i.e., those particles which drift beyond about 25 ft from
the edge of the road, about one-fourth have localized impact, one-third have
medium range drift potential and about half are in the fine particle range.
Although emissions from unpaved air strips were not measured in this
program, the basic emission factor (mass emitted per landing/take-off
cycle) and the correction factors can be approximated by the factors for
unpaved roads.
The dust emitted by agricultural tilling (per acre of land tilled)
is directly proportional to the silt content of the soil and the implement
speed, and inversely proportional to the square of the surface moisture
content. The equilibrium surface moisture for a locality is represented
by Thornthwaite1s precipitation-evaporation index. Of the total dust
emissions, i.e., those particles which drift beyond 25 ft from the edge
of the tillage path, about 40% have medium range drift potential and about
one-third are in the fine particle range.
134
-------
Dust emissions from aggregate storage piles (per ton of material put
through the storage cycle) may be divided into contributions from four
basic source activities:
1. Transfer to storage pile,
2. Equipment traffic in storage area,
3. Wind erosion, and
4. Loadout from storage pile.
Test results indicate that during a typical 3-month storage cycle, about
407o of the dust comes from road traffic in the storage area, 30% from
wind erosion and 3070 from aggregate transfer operations.
Emissions from medium-type construction activities could not be
correlated with potential correction parameters because of the use of
water for dust control and interferences from other dust sources. The
values reported are thought to be fairly representative of uncontrolled
emissions in less arid areas (PE ~ 50) than the Arizona-Nevada test sites,
but having a similar soil silt content (~ 307»).
135
-------
REFERENCES
1. Hidy, G. M., and J. R. Brock, "An Assessment of the Global Sources
of Tropospheric Aerosols," Proc. Second Clean Air Congress,
Washington, D. C., December 1970.
2. Clayton, R. N., R. W. Rex, J. K. Syers, and M. L. Jackson, "Oxygen
Isotope Abundance in Quartz from Pacific Pelagic Sediments,"
J. Geophys. Res., _11, 3907-3915 (1972).
3. Carlson, T. N., and J. M. Prospero, "The Large-Scale Movement of
Saharan Air Outbreaks over the Northern Equatorial Atlantic,"
J. Applied Meteorology, ll_, 283-297 (1972).
4. Chepil, W. S., F. H. Siddoway, and D. V. Armbrust, "Climatic Index
of Wind Erosion Conditions in the Great Plains," Soil Science
Society of America Proceedings, 27 (4) 449-452, July-August 1963.
5. Anderson, C. , "Air Pollution from Dusty Roads," as presented at the
17th Annual Highway Engineering Conference, 1 April 1971.
6. "Air Pollution from Unpaved Roads," a research paper by the School
of Engineering, University of New Mexico, ,12 January 1971.
7. "Investigation of Fugitive Dust Emissions Impact in Designated Air
Quality Control Regions," Final Report, EPA Contract No. 68-02-0044
(Task 9), prepared by PEDCo-Environmental Specialists, Inc.,
May 1973.
8. Hoover, J. M., Surface Improvement and Dust Palliation of Unpaved
Secondary Roads and Streets, Final Report, by Engineering Research
Institute, Iowa State University, ERI Project 856-S, submitted to
the Iowa State Highway Commission, July 1973.
136
-------
9. 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 Inter-
national Section of the Air Pollution Control Association, Eugene,
Oregon, November 1972.
10. Sehmel, G. A., "Particle Resuspension from an Asphalt Road Caused
by Car and Truck Traffic," Atmospheric Environment, 7, 291-309,
July 1973. "~
11. Compilation of Air Pollutant Emission Factors, U.S. Environmental
Protection Agency, Office of Air and Water Programs, Office of Air
Quality Planning and Standards, Publication No. AP-42, April 1973.
12. Gillette, D. A., and P. A. Goodwin, "Microscale Transport of Sand-
Sized Soil Aggregates Eroded by Wind," manuscript prepared for
publication in the J. Geophys. Res_._, submitted in 1973.
13. Fan, L. T., and Y. Horie, "Review of Atmospheric Dispersion and
Urban Air Pollution Models," CRC CriticalReviews in Environ-
mental Control, 431-457, October 1971.
14, "Standards of Performance of New Stationary Sources," Method 5,
Federal Register, 15717, 17 August 1973.
15. Lundgren, D, A., and H. J. Paulus, "The Mass Distribution of Large
Atmospheric Particles," paper presented at the 66th Annual Meeting
of the Air Pollution Control Association, Chicago, Illinois,
24-28 June 1973.
16. Gillette, D. A., and I. H. Blifford, Jr., "The Influence of Wind
Velocity in Size Distributions of Soil Wind Erosion Aerosol
Particles," manuscript prepared for publication in the J^
Geophys. Res., submitted in 1973.
17. "Reference Method for the Determination of Suspended Particulates
in the Atmosphere (High Volume Method)," Federal Register, 36, 28
Appendix B, 22388-22390, 25 November 1971. '
18. "Standard Method for Collection and Analysis of Dustfall," ASTM
Method D 1739-62.
19. Pasquill, F., "The Estimation of the Dispersion of Windborne
Material," Meteorol. Mag., 90, 1063 (1961).
137
-------
20. Sehmel, G. A., "An Evaluation of a High-Volume Cascade Itnpactor,"
Proc. Second Joint Conference on Sensing of Environmental
Pollutants. Washington, D. C.5 December 1973.
21. Buoyocous, G. J., "Recalibration of the Hydrometer Method for Making
Mechanical Analysis of Soils," Agron. J., 43, 434-438 (1951).
22. Streeter, V. L., Fluid Mechanics, McGraw-Hill, New York, 555 pp.
(1962).
23. Gillette, D. A., I. H. Blifford, Jr., and C. R. Fenster, "Measure-
ments of Aerosol Size Distributions and Vertical Fluxes of
Aerosols on Land Subject to Wind Erosion," J. Applied Meteorology,
_U, 977-987, September 1972.
24. Climatic Atlasof the United States, U.S. Department of Commerce,
Environmental Science Services Administration, Environmental Data
Service, available from the Superintendent of Documents, U.S.
Printing Office, Washington, D. C., June 1968.
25. Woodruff, N. P., and F. H. Siddoway, "A Wind Erosion Equation,"
Soil Science Society of AmericaProceedings, 29 (5) 602-608,
September-October 1965.
26. Dickey, H. P., W, R. Swafford, and Q. L. Markley, Soil Survey of
Morton County, Kansas, U.S. Department of Agriculture, Soil
Conservation Service, Series 1960, No. 8, December 1963.
27. Thornthwaite, C. W. , "Climates of North America According to a New
Classification," Geograph. Rev. , 21., 633-655 (1931).
28. Turner, D. B., Workbook of Atmospheric Dispersion Estimates, U.S.
Environmental Protection Agency, Office of Air Programs, Publica-
tion No. AP-26.
29. Construction Review, U.S. Department of Commerce, August 1973.
30. Lumley, J. L., and H. A. Panofsky, The Structure of Atmospheric
Turbulence, Wiley and Sons, New York, 239 pp. (1959).
138
-------
APPENDIX A
PROCEDURE FOR ESTIMATING WINDBLOWN DUST
139
-------
BACKGROUND
Only scattered information is presently available on total emissions
of dust from agricultural areas. PEDCo-Environmental conducted field
sampling studies with directional high-volume networks at two locations
in the Southwest during 1972.—' The results indicated uniformly high
concentrations at all sampling sites at both locations, but no emission
factor could be established because both areas had such intensive farming
that the contributions from individual fields could not be isolated.
The emission factor for tillage operations accounts for the limited
periods when the farming equipment is actually used in the fields; it does
not account for the lower level emissions that occur periodically as a
result of wind erosion across the tilled fields. However, annual emissions
from tilling may be quite small in comparison with suspended particulate
emissions generated by wind erosion.
A recent report indicated that from 37 to 551 million tons of suspended
particulate a year are created by dust storms in the 10 Great Plains states,—/
with an average of 77 million tons per year during the 1960's. Based on these
data, wind erosion contributes more particulate emissions than all other
particulate source categories combined. The same publication estimated that
55 million acres of the approximately 70 million acres of land in the U.S.
from which significant wind erosion occurs is active cropland. Even if these
reported values are high by an order of magnitude, wind erosion emissions
from agricultural lands are still far greater than those from the tillage
operations, in areas where dust storms are common.
Estimation of the wind erosion emissions is not easily accomplished for
several reasons:
1. The sources are not well defined in area and emissions are highly
erratic over time; some sources are temporary and others are seasonal
in nature;
2. Meteorological factors, themselves quite variable, cause large vari-
ations in emission rates due to factors such as periods between rain-
fall and frequency of high wind speeds and atmospheric turbulence;
140
-------
3. Emission rate is a function of soil type, clod structure, and ridging
of the fields;
4. Emission rates are not uniform for large areas;
5. Due to the high settling rate for agricultural dust, a large portion
of the emissions fall out in the immediate area of their origin.
Therefore, the point of measurement greatly affects the apparent
emission rate; and
6. Wind erosion emissions from agricultural lands are indistinguishable
in composition from naturally-occurring dust (background) from nearby
non-agricultural areas.
141
-------
APPLICATION OF WIND EROSION EQUATION
For the reasons outlined above, a major field sampling effort would
be required to develop a comprehensive emission factor for suspended par-
ticulate emissions from wind erosion. As an alternative, it is proposed
that a procedure developed by the U.S. Department of Agriculture for esti-
mating topsoil losses from wind erosion be adapted for use in estimating
emissions from tilled fields. This procedure, called the wind erosion
equation, is thought to be appropriate because the same variables which
affect the rate of topsoil losses also affect the generation of suspended
particulate.
There are several arguments that can be presented for use of the wind
erosion equation in this application and several reasons why it may not
yield good results. These are summarized below:
Pro
1. Relationships in the basic wind erosion equation are based on extensive
data and research;
2. The procedure considers several major parameters which affect the
emission rate;
3. It requires input data which are usually readily obtainable; and
4. Its use of data descriptive of annualized and average conditions is
acceptable since the procedure estimates long-term average emission
rates (tons/year).
Con
1. The adaption assumes that a relatively constant percent of the total
soil losses from tilled land becomes suspended, without any substanti-
ating data;
2. Only sketchy data are available to provide any estimate of the percent
of total soil losses that become suspended;
142
-------
3, The procedure requires a complex series of calculations and much
input data; and
4. It is not capable of estimating short-term emission rates.
It should be stated that the USDA researchers who developed the wind
erosion equation are not in agreement with this application of the equation.
Their objection is not clear, but it probably centers around the assumption
that a constant fraction of the estimated soil losses become suspended.
They cite data which indicates that from 3 to 40% of soil movement over
test fields is in suspension rather than moving by surface creep or salta-
tion. However, the material moved by "suspension" is not equivalent to the
portion that is suspended particulate, because the former contains a signif-
icant amount of material that is settleable and falls out in proximity to
its point of origin. Also, the range of suspended fraction is normally not
as broad as indicated by the USDA data. These percentages are for extreme
soil types which are probably not suitable for cropland.
The preliminary value proposed for percent suspended material is 2.5.
This value was taken from the previous PEDCo study, where it was derived
from particulate size distributions of soils and windblown material from
agricultural lands. Obviously, the proposed number is subject to substantial
modification based on better experimental data.
143
-------
SIMPLIFIED VERSION OF WIND EROSION EQUATION
Presented below is a procedure for estimating windblown or fugitive
dust emissions from agricultural fields. The overall approach and much of
the data have been adapted from the wind erosion equation, which was devel-
oped as the result of nearly 30 years of research by the U.S. Departrtient of
Agriculture to predict topsoil losses from agricultural fields.
Several simplifications have also been incorporated during the adapta-
tion process. The simplified format is not expected to affect accuracy in
its present usage, since wind erosion estimates using the simplified equation
are almost always within 5% of those obtained with the original USDA equation.
Most of the input data are not accurate to ±57».
WINDBLOWN DUST EQUATION
The modified equation is of the form:
E_ = AIKCL'V1 (1)
o> 3(
where: Es = suspended particulate fraction of wind erosion losses of
tilled fields, tons/acre/year
a = portion of total wind erosion losses that would be measured
as suspended particulate, estimated to be 0.025
I = soil erodibility, tons/acre/year
K = surface roughness factor, dimensionless
C = climatic factor, dimensionless
L1 = unsheltered field width factor, dimensionless
V1 = vegetative cover factor, dimensionless.
As an aid in understanding the mechanics of this equation, "I" may be
thought of as the basic erodibility of a flat, very large, bare field in a
climate highly conducive to wind erosion (i.e., high wind speeds and tempera-
ture with little precipitation) and K, C, L1 and V' as reduction factors
for a ridged surface, a climate less conducive to wind erosion, smaller-
sized fields, and vegetative cover, respectively.
144
-------
This same equation can be used to estimate emissions from: (1) a
single field, (2) a medium-sized area such as a valley or county, or (3)
an entire AQCR or state. Naturally, more generalized input data must be
used for the larger land areas, and the accuracy of the resulting estimates
decreases accordingly.
PROCEDURES FOR COMPILING INPUT DATA
Procedures for quantifying the five variable factors in equation (1)
are explained in detail below:
Soil Erodibility, I
Soil erodibility by wind is a function of the amount of erodible fines
in the soil. The largest soil aggregate size normally considered to be
erodible is approximately 0.84 mm equivalent diameter. Soil erodibility, I,
is related to the percentage of dry aggregates greater than 0.84 mm as shown
in Figure A-l. The percentage of non-erodible aggregates (and by difference
the amount of fines) in a soil sample can be determined experimentally by a
standard dry sieving procedure, using a No. 20 U.S. Bureau of Standards sieve
with 0.84-mm square openings.
For larger areas than can be field sampled for soil aggregate size
(e.g., a county) or in cases where soil particle size distributions are not
available, a representative value of I for use in the windblown dust equation
can be obtained from the predominant soil type(s) for farmland in the area.
Measured erodibilities of various soil textural classes are presented in
Table A-l.
If an area is too large to be accurately represented by a soil class
or by the weighted average of several soil classes, the maps in Figures
A-2A through A-2E and the legend in Figure A-2F can be used to identify
major soil deposits and average soil erodibility on a regional basis.
Values of I obtained from Figure A-l, from Table A-l, or from the
national soil maps can be substituted directly into equation (1).
145
-------
Figure A-l. Soil credibility as a function of particle size.
146
-------
Table A-l. SOIL ERODIBILITY FOR VARIOUS SOIL TEXTURAL CLASSES
Predominant Soil Erodlbility, I,
Textural Class tons/acre/year
Sand* 220
Loamy sand* 134
Sandy loam* 86
Clay 86
Silty clay 86
Loam 5 6
Sandy clay loam* 56
Sandy clay* 56
Silt loam 47
Clay loam 47
Silty clay loam 38
Silt 38
*Very fine, fine, or medium sand
147
-------
00
Figure A-2A. Major soil types in northeastern states.
-------
•fs
VO
e A-2B
states.
-------
U1
o
Figure A-2C. Major soil types in the northern Great Plains states.
-------
Ln
Figure A-2D. Major soil types in the southwestern states.
-------
Figure A-2E. Major soil types in the western states.
152
-------
SYMBOL SOIL TYPE
Al, A2 Seasonally wet soils with subsurface clay
accumulation
A3- A5 Cool or cold soils with subsurface clay accumu-
lation
Afi- A8 Clays
A9, A10 Burnt clay soils
All- A13 Dry clay soils with some cementation
Dl- D6 Arid soils with clay and alkali or carbonate
accumulation
E] Poorly-drained loamy sands
E2 Loamy or clayey alluvial deposits
E3- E8 Shallow clay loam deposits on bedrock
E9 Loamy sands in cold regions
E10, E12 Loamy sands in warm regions
Ell, E13, Loamy sands in warm, dry regions
E14
HI, H2 Wet organic soils; peat and muck
11 Ashy or amorphous soils in cold regions
12 Infertile soils with large amounts of amorphous
material
13 Fertile soils of weathered volcanic ash
14 Tundra; frozen soils
15, 16 Thin loam surface horizon soils
17 Clay loams in cool regions
18- 110 Wide varying soil material with some clay horizons
111 Rocky soils shallower than 20 inches, to bedrock
112 Clay loams in warm, moist regions
113 Clay loams in cold regions
114 Clay loams in temperate climates
Ml- M4 Surface loam horizon underlain by clay
M5 Shallow surface loams with no underlying clays
M6- M8 Surface loamy soils
M9- M14 Semiarid loams or clay loams
M15, M16 Dry loams
01, O2 Clays and sandy clays
SI- S4 Sandy, clay, and sandy clay loams
Ul Wet silts with some subsurface clay accumulation
U2- U6 Silty loams with subsurface clay accumulation
U7 Dry silts with thin subsurface clay accumulation
VI- V2 Clays and clay loams
V3- V5 Silty clays
XI- X5 Barren areas, mostly rock with some included soils
Figure A-2F. Legend for soil maps in Figures A-2A through A-2E.
153
-------
Surface Roughness Factor, K
This factor accounts for the resistance of wind erosion provided by
ridges and furrows or large clods in the field. The surface roughness
factor, K, is a function of the height and spacing of the ridges, and
varies from 1.0 (no reduction) for a field with a smooth surface to a
minimum of 0.5 for a field with the optimum ratio of ridge height (h) to
ridge spacing (w).
,2
The relationship between K and -— is shown in Figure A-3. The value
W
of K to be used in equation (1) should be rounded to the nearest 0.1 because
of the large variations inherent in ridge measurement data. In cases where
there are extreme variations of h or w within a field, determination of the
K value should be limited to either 0.5 for a ridge surface or 1.0 for an
unridged surface.
For county or regional areas, K can best be determined as a function
of crop type, since field preparation techniques are relatively uniform for
a specific crop. Average K values of common field crops are shown in Table
A-2. When the K (or L1 or V1) factors are based on crop type, separate
calculations of windblown dust emissions must be made for each major crop
in the survey area. This procedure is explained and demonstrated later in
this presentation.
Climatic Factor, C
Research has indicated that the rate of soil movement by wind varies
directly as the cube of wind velocity and inversely as the square of soil
surface moisture. Surface moisture is difficult to measure directly, but
precipitation-evaporation indices can be used to approximate the amount of
moisture in soil surface particles. Therefore, readily available climatic
data can provide a quantitative indicator of relative wind erosion potential
at any geographic location.
The C factor has been calibrated using the climatic conditions at the
site of much of the research—Garden City, Kansas--as the standard base
(C = 1.00). At any other geographic location, the C factor for use in
equation (1) can be calculated as:
C = 0.345 W 7- (2)
(PE)2 , ^ ;
154
-------
Ui
Ln
tr
o
t-
o
<
LL
CO
w
UJ
z
X
o
o
o
tr
OJ
o
<
u_
tr
3
CO
Figure A-3. Determination of surface roughness factor.
-------
Table A-2. VALUES OF K, L AND V FOR FIELD CROPS
Crop
Alfalfa
Barley
Beans
Corn
Cotton
Grain Hays
Oats
Peanuts
Potatoes
Rice
Rye
Saf flower
Sorghum
Soybeans
Sugar Beets
Vegetables
Wheat
K
1.0
0.6
0.5
0.6
0.5
0.8
0.8
0.6
0.8
0.8
0.6
1.0
0.5
0.6
0.6
0.6
0.6
Lf ft.
1000
2000
1000
2000
2000
2000
2000
1000
1000
1000
2000
2000
2000
2000
1000
500
2000
V,lb/acre
3000
1100
250
500
250
1250
1250
250
400
1000
1250
1500
900
250
100
100
1350
156
-------
where: W = mean annual wind velocity, in mph, corrected to a standard
height of 30 feet
PE = Thornthwaite's precipitation-evaporation index
= 0.83 (sum of 12 monthly ratios of precipitation to actual
evapotranspiration).
Monthly or seasonal climatic factors can be estimated from equation (2)
by substituting the mean wind velocity of the period of interest for the mean
annual wind velocity. The annual PE value is used for all calculations of C.
Climatic factors have been computed from Weather Bureau data for many
locations throughout the country. Figure A-4 presents several maps showing
some typical monthly climatic factors for the USA. C values for use in
equation (1) may be taken from appropriate maps like these when preparing
regional emission surveys. For emission estimates covering smaller areas,
either equation (2) or the map may be used to obtain C.
Unsheltered Field Width Factor, L'
Soil erosion across a field is directly related to the unsheltered
width along the prevailing wind direction. The rate of erosion is zero
at the windward edge of the field and increases approximately proportion-
ately with distance downwind until, if the field is large enough, a maximum
rate of soil movement is reached.
Correlation between the width of a field and its rate of erosion is
also affected by the soil erodibility of its surface: the more erodible
the surface, the shorter the distance in which maximum soil movement is
reached. This relationship between the unsheltered width of a field (L),
its surface erodibility (IK), and its relative rate of soil erosion (L1)
is shown graphically in Figure A-5. If the curves of Figure A-5 are used
to obtain the L' factor for the windblown dust equation, values for the
variables I and K must already be known and an appropriate value for L
must be determined.
L is calculated as the distance across the field in the prevailing
wind direction minus the distance from the windward edge of the field that
is protected from wind erosion by a barrier. The distance protected by a
barrier is equal to 10 times the height of the barrier, or 10 H. For
example, a row of 30-ft high trees along the windward side of a field
reduces the effective width of the field by 10 x 30 or 300 ft. If the
prevailing wind direction differs significantly (more than 25 degrees) from per-
pendicularity with the field, L should be increased to account for this
additional distance of exposure to the wind. The distance across the field,
157
-------
-^' ffi
> \w
Figure A-4. Typical monthly climatic factors for the U.S.
158
-------
Figure A-5. Effect of field length on relative emission rate.
-------
L is equal to the field width divided by the cosine of the angle between
the prevailing wind direction and the perpendicular to the field:
For multiple fields or regional surveys, measurement and calculation
of L values become unwieldy. In region-wide emission estimates, average
field widths should be used. Field width is generally a function of the
crop being grown, topography of the area, and the amount of trees and other
natural vegetation in or adjacent to the farming areas that would shelter
fields from erosive winds. Since the windblown dust calculations are
already split into individual crop type to accurately consider variations
in K by crop, average L values have also been developed by crop; they are
presented in Table A-2. These values are representative of field sizes in
relatively flat terrain devoid of tall natural vegetation, such as found in
large areas of the Great Plains. The L values in Table A-2 should be divided
by 2 in areas with moderately uneven terrain and by 3 in hilly areas.
Additionally, the average field width factors should be divided by 2 to
account for wooded areas and fence thickets interspersed with farmland.
Vegetative Cover Factor, V'
Vegetative cover on agricultural fields during periods other than the
primary crop season greatly reduces wind erosion of the soil. This cover
most commonly is crop residue, either standing stubble or mulched into the
soil. The effect of various amounts of residue, V, in reducing erosion is
shown quantitatively in Figure A-6, where IKCL1 is the potential annual soil
loss (in tons /acre/year) from a bare field, andV is the fractional amount
of this potential loss which results when the field has a vegetative cover
of V, in Ib. of air-dried residue/acre. Obviously, the other four variables
in equation (l)--I, K, C, and L'--must be known before V* can be determined
from Figure A-6.
The amount of vegetative cover on a single field can be ascertained by
collecting and weighing clean residue from a representative plot or by visual
comparison with calibrated photographs. The weight obtained by either
measuring method must then be converted to an equivalent weight of flat
small-grain stubble before entering Figure A-6, since different crop residues
vary in their ability to reduce wind erosion. Detailed descriptions of the
measuring methods or conversion procedures are too complex for this presen-
tation. Interested readers are referred to a USDA publication for these
descriptions.—
The residue left on a field when using good soil conservation practices
is closely related to the type of crop. Table A-2 presents representative
160
-------
m m : JOG'
KpL'
Figure A-6. Effect of vegetative cover on relative emission rate.
-------
values of V for common field crops when stubble or mulch is left after the
crop. These values should be used in calculating windblown dust emissions
unless a knowledge of local farming practices indicates that some increase
or decrease is warranted. Note that three of the five variables in the
windblown dust equation are determined as functions of the crop grown on
the field.
SUMMARY
The estimated emissions in tons/acre/year may now be calculated for
each field or group of fields as the product of the five variables times
the constant "a".
For regional emission estimates, the acreage in agriculture should be
determined for each jurisdiction (e.g., county) by crop. "I" and "C" values
can be determined for individual jurisdiction, with the remaining three
variables being quantified as functions of crop type. The emission calcu-
lations are best performed in a tabular format such as the one shown in
Table A-3. The calculated emissions from each crop are summed to get agri-
cultural wind erosion emissions by jurisdiction and these are totaled to
get emissions for this source category for the entire region.
162
-------
Table A-3. CALCULATION SHEET FOE ESTIMATION OF DUST FROM WIND EROSION
Juris- I, c, K, L, V, L', V, E, = Total
diction Based on Climatic Surface Field Veget. Length Veget. alKC- Emissions
J_County) Soil Type Factor Crop Acres Roughness _LengthCover Factor Factor L_'_V' By
Alfalfa
Barley
Beans
Corn
Cotton
Potatoes
Sorghum
Soybeans
Sugar
Beets
v_ vegets.
£ Wheat
Etc.
Total
(List of
Crops
Grown in
Juris-
diction)
Total
-------
APPROPRIATE USAGE OF RESULTS
Inherent variabilities In the many parameters used in the windblown
dust equation cause the results to be less accurate than emission estimates
for most other sources. However, the rough estimates provided by the pro-
posed procedure are better than not considering this source at all in par-
ticulate emission inventory work. Inclusion of this source category,
possibly with some qualifying statement as to its relative accuracy,, gives
an indication of its contribution to regional air quality.
The estimation procedure is not intended for use in predicting emissions
for short time periods, nor can it be used in determining emission rates for
enforcement purposes.
164
-------
REFERENCES
Al. PEDCo-Environmental Specialists, Inc., "Investigation of Fugitive
Dust--Sources, Emissions and Control," Environmental Protection
Agency Contract No. 68-02-0044, Task Order No. 9, May 1973.
A2. Hagen, L. J., and N. P. Woodruff, "Particulate Loads caused by Wind
Erosion in the Great Plains," presentation at the 66th Annual
Meeting of Air Pollution Control Association, June 1973.
A3. Craig, D. G., and J. W. Turelle, "Guide for Wind Erosion Control on
Cropland in the Great Plains States," USDA Soil Conservation
Service (1964).
165
-------
APPENDIX B
DISPERSION CALCULATIONS FOR CONSTRUCTION EMISSIONS
166
-------
Table B-l below presents the concentrations recorded during periods
of southerly, southwesterly and westerly winds at the Paradise Valley
construction site.
Table B-l. MEASURED CONCENTRATIONS DURING S, SW AND W WINDS (ug/m3)
Station
Date
8-31-72
9-6-72
9-12-72
Average
9-4-72
9-14-72
Average
9-20-72
9-28-72
9-30-72
10-6-72
Average
Station
C-13
-
461
487
474
353
389
371
212
375
_
_
294
Station
C-14
324
280
368
324
370
258
314
251
336
312
70
241
Station
C-15
363
169
365
299
328
292
310
141
240
415
_
299
Station
C-ll
(bkgnd)
347
152
152
217
203
212
208
95
163 .
185
170
153
Wind
Speed Dir.
9 mph SW
3
3
5
6 S
7
6.5
2 W
2
5
2
2.75
167
-------
FOR SOUTHWESTERLY WIMP
I. (a) Average concentration at downwind stations:
Station C-13 474 pg/m3
C-14 324
C-15 299
Average 1097/3 = 366 pg/m3
(b) Average concentration at background station:
Station C-ll 217 pg/m3
Contribution from the construction site =
366 - 217 = 149 pg/m3
(c) Q = 2.78 Xuay az
where Q = source strength (grams per second)
X = concentration (grams per cubic meter)
u = wind speed (meters per second)
°"v = horizontal dispersion coefficient (meters)
az = vertical dispersion coefficient (meters).
[NOTE: The factor of 2.78 was derived from Table 5-1, page 38
of "Workbook of Atmospheric Dispersion Estimates," PHS
Publication No. 999-AP-26. Ratio of calculated 24-hr
concentration to 3-min concentration = 1.00/0.36 = 2.78.]
u = wind speed = 5 mph = 2.23 m/sec
For wind speed of 2.23 m/sec and assuming moderate
to strong solar radiation based on Table 3-1 of above
reference, stability class = B.
Using the method of approximation outlined for area
sources in the above reference (pages 39 and 40),
a and 0 values were obtained from appropriate
figures for Xj_ = X + X
where X is the distance of sampler from the source,
Xvo is virtual distance corresponding to
ayO = S/^«3 and
S is the length of a side of the area source.
Distance from center of construction area to sampler
locations (measured from the map):
C-13 1,000 ft
C-14 2,400 ft
C-15 3,350 ft
168
-------
Average Distance = X = 6750/3 = 2250 ft » 690 in
0yo = S/4.3 = 650/4.3 = 150 m
For ° = 150 meters, Xyo = 960 m (from chart, stability class = B)
Xx = X + X = 690 + 960 = 1,650 m
For Xj = 1650 meters, ay = 250 m
a_ - 190 m
£j
Substituting these values in the expression for Q = 2.78
XuCTycrz, we get:
Q = 2.78 (0.000366 - 0.000217) (2.23) (250) (190)
- 43.8 g/sec, or
= 1,525 tons/ year, or
= 1.41 tons/acre/month of active construction
(based on 90 acres under active construction
at this location)
[NOTE: Particulate concentration used was the difference between upwind
and downwind sampling locations and is thought to represent only
the contribution from the construction site.]
II. (a) Average concentration at the closest downwind station -
station C-13 only = 474 jig/nr*
(b) Average concentration at background station -
station C-ll = 217 ug/m3
o
Contribution from the construction site = 474 - 217 = 257 ug/m
(c) Q =2.78 Xuayaz
u =5 mph = 2.23 m/sec
X = 0.000257 g/sec
X = distance of sampler from center of construction area
2: 1,000 ft or 305 m
ayo = 650/4.3 = 150 m
Xy0 (from graph in the Reference, Stability Class = B)
= 960 m
Xl = X + X o = 305 + 906 = 1,265 m
dy = 190 m
CTZ - 140 m
Q = (2.78) (0.000257) (2.23) (190) (140)
= 42.5 g/sec, or
= 1,480 tons/year, or
= 1.37 toiis/acre/month of active construction
Performing these calculations for the other two wind directions of
interest, namely, southerly and westerly winds, the source strength values
shown in Table 36 (Chapter 7) were obtained.
169
-------
APPENDIX C
PHOTOGRAPHS OF FIELD TESTING
This appendix presents representative photographs of field equipment
used in testing dust emissions from unpaved roads and agricultural tilling.
Figure C-l shows the dust sampling equipment used at gravel road Site R2,
and Figure C-2 shows the tillage equipment used at the agricultural
sites in Wallace County, Kansas.
170
-------
Test Gravel Road
Plume Sampling Equipment
Exposure Sampling
Background Station
Figure C-l. Testing of gravel site emissions (Site R2).
171
-------
20-ft Disk (Site A4)
Plume Sampling
Dust Generation
30-ft Sweep (Site A3)
Figure C-2. Testing of agricultural tilling emissions,
172
-------
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing]
1. REPORT MO.
£PA-450/3-74-037
4. TITLE AND SUBTITLE
Development of Emission Factors For Fugitive
Oust Sources
7. AUTHORiS)
Chatten Cowherd, Jr., et al.
3. RECIPIENT'S ACCESSION-NO.
5, REPORT DATE
June 1974
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Midwest Research Institute
425 Volker Boulevard
Kansas City, Missouri 64110
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-0619
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agencv
Research Triangle Park, North Carolina 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final" Report - 7-72 - 3-74
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This report presents the results of an extensive field testing program to
develop emission factors for certain common sources of fugitive dust. A description
of the measurement techniques and summaries of calculated test results are presented.
The basic measurements consisted of isokinetic dust exposure profiles with specially
designed sampling equipment, dust concentrations with conventional high-volume
samplers, particle size classification with high-volume cascade impactors, depostition
profiles and dust transport by saltation.
For each source type, emissions are related to meteorological and source
parameters, including properties of the emitting surface and characteristics of the
vehicle or implement which causes the emission. This information is used to derive
correction factors which appropriately adjust basic emission factors to reflect
regional differences in climate and surface properties.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Emission factors Sampling techniques
Fugitive dust , Climatic factors
Agricultural tilling
Unpaved roads
Construction activities
Aggregate storage piles
Particle size
b.IDENTIFIERS/OPEN ENDEDTERMS
c. COSATI Field/Group
8. DISTRIBUTION STATEMENT
Release unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-------
------- |