I
I
United States Office of Air Quality EPA-450/4-83-004
Environmental Protection Planning and Standards August 1982
Agency Research Triangle Park NC 27711
Air
• &EPA Characterization Of
I PIN/MOAndTSP
| Air Quality Around
I Western Surface
Coal Mines
I
I
I
I
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
CHARACTERIZATION OF PM 10 AND TSP
AIR QUALITY AROUND WESTERN
SURFACE COAL MINES
by
PEDCo Environmental, Inc.
2420 Pershing Road
Kansas City, Missouri 64108
and
TRC Environmental Consultants
8775 East Orchard Road
Englewood, Colorado 80111
Contract No. 68-02-3512
Work Assignment No. 35
PN 3525-35
Project Officer
Thompson G. Pace
AIR MANAGEMENT TECHNOLOGY BRANCH
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711
August 1982
. S. Environment" 1 .:
x' 5. r/Otti.'bo.r n Gi.
, IL 60GO'i
-------
This report has been reviewed by the Office Of Air Quality
Planning And Standards of the U. S. Environmental Protection
Agency and approved for publication as received from Pedco
Environmental, Inc. Approval does not signify that the contents
necessarily reflect the views and policies of the U. S. Environ-
mental Protection Agency, nor does mention of trade names or
commercial products constitute endorsement or recommendation
for use.
EPA-450/4-83-004
ii
-------
I
I
I
I
I
I
1
1
I
I
I
I
I
I
i
I
I
I
I
CONTENTS
Figures v
Tables x
1.0 Executive Summary 1
1.1 Introduction 1
1.2 Conclusions 2
2.0 Characterization of PM 10 and TSP Air Quality around
Western Surface Coal Mines using Monitoring Data 11
2.1 Background 11
2.2 Data collected 11
2.3 Procedure to infer PM 10 concentrations from
monitored TSP concentrations 12
2.4 Description of results 23
2.5 Error analysis and assumptions 39
3.0 Characterization of PM 10 and TSP Air Quality around
Western Surface Coal Mines using Previous Modeling
Studies 43
3.1 Background 43
3.2 Previous modeling studies 44
3.3 Other descriptive models 70
3.4 Relationship of modeling results to possible
ambient standards and the PSD permitting
process 76
4.0 Characterization of PM 10 and TSP Air Quality around
Western Surface Coal Mines using New Predictive
Tools 77
4.1 Background 77
4.2 Calculation of emissions 77
4.3 Calculation of concentrations 92
4.4 Potential sources of error in the prediction
process 112
4.5 Relationship of scenario results to possible
regulatory options 116
111
-------
CONTENTS (continued)
5.0 Synthesis of Three Approaches to Characterize PM 10
and TSP Air Quality around Western Surface Coal
Mines 121
5.1 Introduction 121
5.2 Comparison of alternate approaches 121
6.0 Need for Further Study 125
6.1 Additional monitoring 125
6.2 Deficiencies in the predictive process 126
6.3 Impact of additional control measures and
alternate mine configurations on
concentrations 126
6.4 Standardized methods 127
6.5 Regulatory implications 127
References 128
Appendices
A Mass Fraction Calculations Derived for Section 2.0
Procedure to Infer PM 10 Concentrations from
Measured TSP Data 130
B Computation of PM 10, PM 5, and PM 2.5 Mass
Fractions at Various Downwind Distances
for Section 2.0 Procedure 137
C ICS Model Application to Three Mine Scenarios 147
IV
-------
I
I
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
FIGURES
Number Page
1 Maximum Annual Geometric Mean Concentrations at
Mine Boundary—TSP and PM 10 9
2 Maximum 24-Hour Concentrations at Mine
Boundary--TSP and PM 10 10
3 Mass Fraction of TSP, Stability Class D, 4.3 m/s 21
4 Mass Fraction of TSP, Stability Class D, 2.5 m/s 22
5 Annual Average Monitored TSP and Calculated PM 10
Concentrations, Northeast Wyoming, Mine 1 26
6 Annual Average Monitored TSP and Calculated PM 10
Concentrations, Northeast Wyoming, Mine 2 27
7 Annual Average Monitored TSP and Calculated PM 10
Concentrations, Northeast Wyoming, Mine 3 28
8 Annual Average Monitored TSP and Calculated PM 10
Concentrations, Northeast Wyoming, Mine 4 29
9 Annual Average Monitored TSP and Calculated PM 10
Concentrations, Northeast Wyoming, Mine 5 30
10 Annual Average Monitored TSP and Calculated PM 10
Concentrations, Northeast Wyoming, Mine 6 31
11 Annual Average Monitored TSP and Calculated PM 10
Concentrations, Northeast Wyoming, Mine 7 32
12 Annual Average Monitored TSP and Calculated
PM 10 Concentrations, Northeast Wyoming,
Mine 8 33
13 Annual Average Monitored TSP and Calculated
PM 10 Concentrations, Southwest Wyoming,
Mine 9 34
-------
FIGURES (continued)
Number Page
14 Annual Average Monitored TSP and Calculated
PM 10 Concentrations, Northeast Colorado
Mine 10 35
15 Annual Average Monitored TSP and Calculated
PM 10 Concentrations, Southwest Montana
Mine 11 36
16 Annual Average Monitored TSP and Calculated
PM 10 Concentrations, Western North Dakota,
Mine 12 37
17 Sensitivity of the Procedure to Infer PM 10
Concentrations to Windspeed Assumptions 40
18 Sensitivity of the Procedure to Infer PM 10
Concentrations to Stability Class Assumptions 42
19 Geographical Areas of Interest 45
20 Annual Average Modeled TSP Concentrations
Powder River Basin, Mine 1 48
21 Annual Average Modeled TSP Concentrations
Powder River Basin, Mine 2 49
22 Annual Average Modeled TSP Concentrations
Powder River Basin, Mine 3 50
23 Annual Average Modeled TSP Concentrations
Powder River Basin, Mine 4 51
24 Annual Average Modeled TSP Concentrations
Powder River Basin, Mine 5 52
25 Annual Average Modeled TSP Concentrations
Green River/Hams Fork Basin, Mine 6 53
26 Annual Average Modeled TSP Concentrations
Green River/Hams Fork Basin, Mine 7 54
27 Annual Average Modeled TSP Concentrations
Green River/Hams Fork Basin, Mine 8 55
VI
-------
I
I
I
I
29 Annual Average Modeled TSP Concentrations
• Green River/Hams Fork Basin, Mine 10 57
30 Annual Average Modeled TSP Concentrations
Green River/Hams Fork Basin, Mine 11 58
I
I
I
I
I
I
I
I
I
I
I
I
1
I
FIGURES (continued)
Number Page
28 Annual Average Modeled TSP Concentrations
Green River/Hams Fork Basin, Mine 9 56
31 Annual Average Modeled TSP Concentrations
Green River/Hams Fork Basin, Mine 12 59
32 Annual Average Modeled TSP Concentrations
Green River/Hams Fork Basin, Mine 13 60
33 Annual Average Modeled TSP Concentrations
Green River/Hams Fork Basin, Mine 14 61
34 Annual Average Modeled TSP Concentrations
Green River/Hams Fork Basin, Mine 15 62
35 Annual Average Modeled TSP Concentrations
Green River/Hams Fork Basin, Mine 16 63
36 Existing and Anticipated Surface Coal Mining
Operations in Campbell County 65
37 Annual Average TSP Concentrations (pg/m3),
Total Dust Sources with Deposition, no
Background Added 67
38 Annual Average TSP Concentrations (|jg/m3),
Coal Dust Sources with Deposition 68
39 Annual Average TSP Concentrations (pg/m3),
Respirable Particulate £3 [jm without
Deposition 69
40 Assumed Worst-Case Mine Configuration 71
41 Nomograph Procedure for Predicting Worst-Case
Fenceline Concentration, TSP 72
42 Distance from Mine Center over which TSP
Class II Increment is Exceeded in the Powder
River Basin 74
VI1
-------
FIGURES (continued)
Number Page
43 Profile Presentation of a Powder River Basin
Mine, Annual TSP Concentrations 75
44 Examples of Composite Particle Size Distributions
from Coal Mining Particulate Sources 89
45 Theoretical Impact of Incorrect Particle Size
Distribution on Model-Predicted PM 10 or TSP
Concentrations 91
46 ISC Modeled Annual Geometric Mean TSP
Concentrations, Powder River Basin 97
47 ISC Modeled Annual Geometric Mean PM 10
Concentrations, Powder River Basin 98
48 ISC Modeled 24-Hour TSP Concentrations, Powder-
River Basin 99
49 ISC Modeled 24-Hour PM 10 Concentrations,
Powder River Basin 100
50 ISC Modeled Annual Geometric Mean TSP Concentra-
tions, San Juan Basin 101
51 ISC Modeled Annual Geometric Mean PM 10 Concen-
trations, San Juan Basin 102
52 ISC Modeled 24-Hour TSP Concentrations, San
Juan Basin 103
53 ISC Modeled Annual Geometric Mean TSP Concentra-
tions, Green River/Hams Fork Basin 104
54 ISC Modeled 24-Hour PM 10 Concentrations,
San Juan Basin 106
55 ISC Modeled Annual Geometric Mean PM 10
Concentrations, Powder River Basin 107
56 ISC Modeled 24-Hour TSP Concentrations,
Green River/Hams Fork Basin 108
57 ISC Modeled 24-Hour PM 10 Concentrations,
Green River/Hams Fork Basin 109
Vlll
-------
I
I
I
FIGURES (continued)
Number Page
m 58 Maximum Annual Geometric Mean Concentrations at
• Mine Boundary—TSP and PM 10 122
59 Maximum 24-Hour Concentrations at Mine
• Boundary—TSP and PM 10 123
I
I
I
I
I
I
I
I
I
I
I
I
I
I
IX
-------
TABLES
Number page
1 Range of Measured TSP and Inferred PM 10
Concentrations near Western Surface Coal Mines 3
2 Maximum Concentration versus Distance by Scenario,
|jg/m3 (No Background Concentration Added) 6
3 Violations of the NAAQS and Class II PSD
Increments 7
4 Particulate Monitoring Information 13
5 Particle Size Distributions by Mass Fraction 16
6 Computation of Settling Velocity 16
7 Reflection Coefficients 17
8 Groundlevel Concentrations at 1,000 Meters, "D"
Stability, and 4.30 Meters/Second Windspeed 18
9 Relative Groundlevel Concentrations at 1,000
Meters, "D" Stability, and 4.30 Meters/Second
Windspeed 19
10 Inferred PM 10 Annual Average Concentrations 24
11 Inferred PM 10 Second-Highest 24-Hour
Concentrations 38
12 Particulate Emission Factors and Control
Efficiencies used in Previous Modeling Studies 47
13 Surface Coal Mines Projected for the Campbell
County Portion of the Powder River Basin 66
14 Mining Operations that Generate Particulate
Emissions 78
x
-------
I
I
I
I
I
I
I
I
i
i
i
i
i
i
i
i
i
i
i
TABLES (continued)
Number Page
15 Hypothetical Mine Annual Activity Parameters 80
16 Base Emission Factors 81
17 Independent Variable Values used in Emission
Factor Equations 83
18 Particulate Emission Factors for Hypothetical
Mines 85
19 Calculated Emissions 86
20 Distribution of Emissions by Particle Size 87
21 Assumed Background Concentrations used in the
Scenario Analysis 95
22 Annual Average Geometric/Arithmetic TSP
Concentrations Measured by the Hi-Vol Method 96
23 Maximum Concentration versus Distance by Scenario,
|jg/m3 (No Background Concentration Added) 110
24 Violations of the NAAQS and Class II PSD
Increments 111
25 Potential Sources of Error in the Predictive
Process for Estimating Particulate Concentra-
tions around Surface Coal Mines 113
26 Impact of Additional Particulate Controls on
Emissions, Powder River Basin Scenario Example 117
27 Impact of Additional Particulate Controls on
Maximum Off-Site Concentrations 119
XI
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
SECTION 1.0
EXECUTIVE SUMMARY
1.1 INTRODUCTION
Various regulatory changes that would impact the surface
coal mining industry are being considered. These changes relate
to the ambient particulate standard and the Prevention of Sig-
nificant Deterioration (PSD) regulations.
The Environmental Protection Agency (EPA) is currently
considering a new ambient particulate standard that would address
particulates less than 10 microns in diameter (termed PM 10).
Additionally, there is considerable debate over whether
fugitive dust from surface mines should consume PSD increment.
Currently, federal PSD regulations dictate that fugitive and
nonfugitive dust consume increment. These regulations may change
as a result of EPA's settlement agreement (Chemical Manufacturer's
Association, et al v. EPA) or as a result of changes in the Clean
Air Act. The fugitive dust issue is critically important because
the vast majority of particulate emissions at surface mines are
fugitive.
The objectives of this study are: (1) to provide data on PM
10 concentrations and total suspended particulate (TSP) concentra-
tions around western surface coal mines sufficient to assess
their relationship to possible changes to ambient standards; (2)
to apply new emission factors and the new ISC model to predict PM
10 and TSP concentrations around hypothetical surface mines; and
(3) to assess the impact of the PM 10 and TSP concentrations on
the permitting process.
The objectives of the study are addressed through analysis
of existing monitoring data (Section 2.0), previous modeling
studies (Section 3.0), and new modeling studies using recently
available improved techniques (Section 4.0). The results of the
three approaches are synthesized in Section 5.0. Recommendations
for further study are outlined in Section 6.0.
-------
1.2 CONCLUSIONS
1.2.1 Characterization of PM 10 and TSP Air Quality around
Western Surface Coal Mines using Monitoring Data
Particulate monitoring around western surface coal mines has
been almost exclusively for TSP. To support the regulatory
decision making process, it is desirable to have a knowledge of
PM 10 concentrations. Therefore, a theoretical procedure was
derived to infer PM concentrations from monitored TSP data. The
procedure is subject to several uncertainties and a qualitative
error analysis has been performed.
TSP Concentrations—
Monitoring data were collected from 12 mines with annual
coal production ranging from 0.7 to 16.0 million tons per year.
Maximum monitored concentrations were 86 and 324 |jg/m3 on an
annual and second-highest 24-hour basis (Table 1). Six of the
twelve mines had monitors located outside the mine boundaries
facilitating comparison of the monitored values with National
Ambient Air Quality Standards (NAAQS) and PSD increments;. TSP
concentrations outside the mine boundary did not exceed the
primary NAAQS on an annual or 24-hour basis. The secondary
24-hour NAAQS were exceeded at two of the six mines.
PM 10 Concentrations—
Maximum PM 10 concentrations calculated from TSP monitoring
data were 29 and 98 (jg/m3 on an annual and second-highest 24-hour
basis. The calculated maximum PM 10 concentrations outside mine
boundaries and above background concentrations were 8 and 39
|jg/m3 on an annual and 24-hour basis. If the PSD Class II incre-
ments remain the same, but the applicable particle size category
changes from TSP to PM 10, the data indicate that the 24-hour
increment may still be a restraint to obtaining a permit. This
conclusion should be considered tentative due to uncertainties in
the calculation procedure.
1.2.2 Characterization of PM 10 and TSP Air Quality around
Western Surface Coal Mines using Previous Modeling Studies
Most previous modeling studies allow only a characterization
of annual average TSP concentrations. Short-term TSP modeling
studies have been performed on a more limited basis and are
considered to be less accurate because of the difficulty in
predicting short-term activity patterns, emissions, and meteoro-
logical parameters. Modeling for sub-TSP particle size ranges,
such as for PM 10, has been limited to a few theoretical studies.
TSP Concentrations—
Previous modeling suggests that ambient concentrations
resulting from a given level of emissions vary considerably by
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
TABLE 1. RANGE OF MEASURED TSP AND INFERRED PM 10 CONCENTRATIONS
NEAR WESTERN SURFACE COAL MINES3
(jjg/m3)
Location
Within mine boundaries
concentration
above background
Outside mine boundaries
concentration
above background
Measured
TSP concentrations
Annual
17-86
2-68
18-42
3-27
Second-highest
24-hour
79-324
64-306
56-180
41-165
Inferred
PM 10 concentrations
Annual
9-29
0-20
10-17
1-8
Second-highest
24-hour
19-98
10-91
24-48
13-39
Based on monitoring at 12 mines. The data may not represent the full range
of monitored values at all coal mines.
-------
air basin because of different dispersion conditions, and because
of TSP background concentrations ranging from about 15 pg/m3 to
30 (jg/m3 . However, for mines isolated from other particulate
sources (such as other mines), previous modeling indicates that
annual average ambient fenceline concentrations seldom exceed 50
(jg/m3. At mines with an annual coal production of greater than
about 15 million tons, however, annual concentrations may approach
the primary and secondary annual standards under worst-case mine
configurations. In all cases examined, annual TSP concentrations
decreased to <_I ug/m3 within four miles of the mine boundary.
Based on previous modeling studies, PSD Class II increments
appear to be a much greater restraint than the NAAQS. Even with
application of EPA Region VIII defined Best Available Control
Technology (BACT), the annual PSD Class II increment was predicted
to be violated up to 2 miles beyond the fenceline at mines with
an annual production of about 10 million tons a year or more.
The 24-hour PSD Class II increment would be an even greater
restraint with violations as distant as 5 miles beyond the fence-
line at isolated mines with greater than 10 million tons produc-
tion.
Often mines are not isolated from other mines. This is
particularly true in the Powder River Basin. Fugitive dust
consumption of available PSD increment in that and other regions
may severely restrict the planned level of mining.
PM 10 Concentrations--
Previous modeling studies of sub-TSP size fractions do not
allow adequate characterizations of PM 10 concentrations around
western surface coal mines.
1.2.3 Characterization of PM 10 and TSP Air Quality around
Western Surface Coal Mines using New Predictive Tools
Two new tools—improved emission factors and the ISC disper-
sion model--are now available for assessing a western surface
coal mine's impact on air quality.
In this report the new predictive tools were applied to
three hypothetical surface coal mines. The mines were assumed to
have the following annual coal production rates:
0 Powder River Basin (near Gillette, Wyoming): 25 mil-
lion tons per year.
0 Green River/Hams Fork Basin (near Craig, Colorado):
3.6 million tons per year.
0 San Juan Basin (near Four Corners area): 6.5 million
tons per year.
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1
I
I
The hypothetical mines were chosen to represent a cross section
of mine size and mine location found in the west.
Using the new emission factors coupled with the ISC model,
TSP and PM 10 concentrations were computed for annual average and
24-hour time periods. The predicted concentrations are directly
comparable with existing TSP standards, with possible PM 10
standards, and with PSD increments. The results of the modeling
are summarized in Tables 2 and 3.
Based on the modeling work, several observations can be
made:
0 TSP standards. Only the largest mine, the Powder River
Basin Mine, shows violations of the annual or 24-hour
TSP NAAQS.
0 PSD increments. Presently, federal PSD regulations
dictate that fugitive and nonfugitive dust consume PSD
increment. All three of the hypothetical mines exhibit
TSP exceedances of the annual and 24-hour PSD Class II
increment outside of the mine boundary.
0 PM 10 concentrations. The peak PM 10 concentrations
estimated at the mines' fencelines are between 1/3 and
2/3 the magnitude of the TSP concentrations. PM 10
concentrations consume all of the PSD Class II incre-
ment at two of the three scenario locations.
Several regulatory options were considered and applied to
the scenario analysis results. Two of these options were: (1)
requiring additional particulate control measures; and (2) ap-
plying the PSD increment consumption determination at some dis-
tance beyond the mine boundary. All physically possible particu-
late control measures were applied to the scenarios, regardless
of their cost or other environmental consequences. For the
Powder River Basin scenario, TSP concentrations were still twice
the annual Class II increment. The PM 10 concentration also
exceeded the increment.
If the PSD program is viewed as a resource allocation pro-
gram, it may be reasonable to apply the increment consumption
determination at some distance beyond the boundary. Regarding
TSP concentrations, the 24-hour increment would still be a re-
straint for large mines at distances 5 miles beyond the mine
boundary. However, if PM 10 concentrations were used to compute
increment consumption, a 2-mile buffer around a mine boundary
would allow even the Powder River Basin scenario mine to receive
a PSD permit.
-------
TABLE 2. MAXIMUM CONCENTRATION VERSUS DISTANCE BY SCENARIO, |jg/m3
(NO BACKGROUND CONCENTRATION ADDED)
Scenario
Annual concentrations
Powder River Basin
San Juan Basin
Green River/Hams
Fork Basin
Second-highest 24-hour
concentrations
Powder River Basin
San Juan Basin
Green River/Hams
Fork Basin
TSP
At
boundary
115
23
30
867
106
104
Distance from
boundary, miles
1
115
20
20
260
80
55
2
10
8
13
240
50
25
3
7
7
7
165
45
18
4
6
6
4
80
37
14
5
4
5
3
45
32
10
PM 10
At
boundary
51
16
23
289
35
30
Distance from
boundary, miles
1
40
11
15
200
30
16
2
7
7
5
45
20
11
3
6
4
4
25
18
7
4
4
3
3
16
16
4
5
3
2
2
12
12
1
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
a:
o
a
oo
a.
to
C/)
g
•^
w
••*
in
»
*> 3 C
< O «D
A "D
Distance
from boundary,
miles
in
»
rr>
CM
fH
c >»
*> 3 t~
< O ID
A TJ
Scenario
X
X X
XX X
x
X X
X X
X X
XXX XXX
XXX XXX
X
X
X X
X X
X X
X X
u
c c o c
"-in in i in in
--> nj E ^ ID E
re co c re CM mere
i- ••- X «n — X
C O) ID S-C inO OJ ID UC
O) > CD Ol ••- Ol -^ > CD * •—
c oe c —re O) ID cc c —re
o re ce CD ••- t- re cc co
o; -5 c ^ ie * -3 c ^
•— "O OJ i- "D O) "D OJ U
ID 2 C O>O CO 2 C OJ O
3 O ID i- U. OC O ID S- u.
C Q. (/) C5 UOQ.tO(3
C 0) 0
< I/)
•o
•o
1C
c
o
ID
*J
C
QJ
U
O
u
o
O)
.*
u
ID
-------
1.2.4 Synthesis
Three approaches to characterizing PM 10 and TSP air quality
around western surface coal mines are utilized in this report.
These approaches are examination of monitoring data, review of
previous modeling studies, and application of improved tools to
the predictive process.
The data derived from the three methods are in relative
agreement. The annual and 24-hour TSP NAAQS are restraining only
for large mines (probably greater than 15 million tons of annual
production) or in areas with several nearby mines.
The PSD increments (particularly the 24-hour increment) are
much more restrictive than the NAAQS. TSP concentrations from
mines producing as little as 5 million tons/year may consume all
of the Class II increment under worst-case site configurations
(Figures 1 and 2). Inclusion of fugitive emissions in the PSD
process would severely restrict the ability of new mines to
obtain a PSD permit.
Changing the pollutant of measurement from TSP to PM 10
would be less restrictive to the coal mining industry, but may
still prohibit large mines from obtaining a PSD permit.
1.2.5 Need for Additional Study
The need for further research became apparent at several
points in the study. Required further study can be divided into
five broad categories. These are:
1. Additional monitoring to gain particle size information and
to attempt to validate the predictive process.
2. Analysis of deficiencies in the predictive process, including,
but not limited to particle deposition and pit retension.
3. Impact of additional particulate control measures and alter-
nate mine configurations on concentrations.
4. Standardized methods.
5. Regulatory implications.
8
-------
1
1
1
1
•
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
60
2J"
LU
<-> 40
t— 1 ^"
t—
LU
s: a.
o to
LU (—
C3
_J
ii a" 20
Z LU
Z Q
- Z
o: o
Q £
LU
I 1 I i i •
• MEASURED MINUS ASSUMED BACKGROUND Q
* PREVIOUS MODELING 115
D THREE SCENARIOS
.
D •
_ A
A
• D • PSD CLASS II INCREMENT _
*
* A
A. A •
» A A A •
A* A
1 m 1 f i 1 1
LU O
"-" O
s: o
I— Q
•a: z
ID
~£ 60
-------
200
150
CC 0.
gS £ 100
Z LLJ
=> 0
0 0
CO «t
LiJ Z
Z 0
»— 1 »-H
*£• 1_
^^ ^^
£§ 50
Z
Z LU
O 0
•— ' Z
t— 0
1— 0
LU =3
°9 0
1 1 1 1 1 1
• MEASURED MINUS ASSUMED BACKGROUND *
D THREE SCENARIOS 8g?
-
D D •
•
PSD CLASS II INCREMENT ~
•
ii(ii t
O CD
O
a:
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
SECTION 2.0
CHARACTERIZATION OF PM 10 AND TSP AIR QUALITY AROUND WESTERN
SURFACE COAL MINES USING MONITORING DATA
2.1 BACKGROUND
Over the last few years a base of air quality monitoring
data has been generated at proposed and operating western surface
mines. However, the data from the individual mines has not been
compiled and analyzed in a comprehensive manner. Ideally, the
compilation of these data would allow an immediate characteriza-
tion of particulate concentrations around western surface coal
mines. However, a great amount of data is required to interpret
the monitoring results.
Monitors are sited under highly varying conditions. The
resulting concentration measurements are a function of a number
of variables. These variables include: production level, size
of mining area, production methods, types of sources, source-
receptor separation, terrain features, and meteorological pat-
terns. Any rigorous interpretation of mine monitoring data
requires that monitor siting be considered in conjunction with
the concentration data.
A review of available particulate monitoring data has indi-
cated that almost all of the reported data are for TSP, not
PM 10. Since a knowledge of PM 10 concentrations is of primary
interest to support the regulatory process, a method to infer
PM 10 information from TSP monitoring data was developed, and is
described in Subsection 2.3. This section contains a description
of data collected, a description of the process to infer PM 10,
PM 5, and PM 2.5 concentrations from monitored TSP concentrations,
and a tabular and graphical summary of the results.
2.2 DATA COLLECTED
Several sources of monitoring siting data were considered.
These sources included file information in the PEDCo and TRC
offices, the coal industry as represented by members of the Tech-
nical Advisory Group, and public documents on file in state air
quality offices.
11
-------
The form shown in Table 4 was used to compile the available
information. The National Coal Association (NCA) sent letters
and the form to several of its members requesting monitoring/siting
data. In addition, project resources (time and budget) allowed
the study team to visit the Colorado and Wyoming state agencies.
During the visits to the state agencies, and during the
review of contractor file data, a large amount of air quality
data were initially considered. Certain data were rejected for
the following reasons:
1. The data were obtained as part of a background or construc-
tion phase air quality monitoring program.
2. The data required to complete Table 4 could not be obtained
from file information.
3. The data were still classified as proprietary.
4. Less than 45 data points were available (assumes monitoring
every sixth day and EPA requirements for 75 percent data
recovery).
5. Monitoring data were collected within 100 meters of an
identifiable source and were not representative of areawide
air quality. Only limited data were available to etpply this
criteria.
After application of the rejection criteria outlined above,
data from 12 surface coal mines remained. These data were used
to develop the TSP to PM 10 concentration algorithm as described
in the next subsection.
2.3 PROCEDURE TO INFER PM 10 CONCENTRATIONS FROM MONITORED TSP
CONCENTRATIONS
2.3.1 Description of Procedure
This section describes the mechanics of the procedure used
to infer PM 10 (and PM 5 and PM 2.5) concentrations from measured
TSP concentrations. The foundation of the procedure is to utilize
a dispersion model to calculate downwind concentrations of total
suspended particulate and concentrations of PM 10 from which the
PM 10 mass fraction can be determined. Measured hi-vol TSP
concentrations are then multiplied by the appropriate mass frac-
tion to yield PM 10 concentrations. For the purposes of explana-
tion, the procedure can be viewed as composed of five sequential
steps:
12
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
TABLE 4. PARTICULATE MONITORING INFORMATION
PART A—GENERAL MINE INFORMATION (ONE SHEET PER MINE)
1. Name of mine
2. Mine location, county and state
3. Average coal production, tons per year
4. Area disturbed per year for pit, acres
5. Approximate quantity of overburden moved per year,
cubic yards
6. Method of overburden removal - dragline
- shovel/truck
- scraper
7. Dust control measures for haul roads - method
- frequency
8. Do you have a monitor quality assurance program - yes
- no
9. Map or sketch (please attach) indicating north arrow and scale; mine
boundaries; location of pit during monitoring period, permanent haul
roads, coal processing plant, and coal storage, monitor locations and
monitor number.
10. Name and phone number of contact - name
- phone number
11. Above information applicable to what year
(continued)
13
-------
TABLE 4 (continued)
PART B—SPECIFIC MONITOR INFORMATION (ONE SHEET FOR EACH MONITOR LOCATION)
1. Monitor number
2. Instrument to measure TSP - hi-vol
- other
3. Instrument to measure small particles,
Instrument
0
0
0
Cascade impactor
Size selective inlet
Dichotomous
4. Measured values
No. of
Year samples
TSP
Annual Second 20th
geom. highest highest No. of
mean 24-hour 24-hour* samples
Particle size
cut-off, |jm
Size specific
Annual Second 20th
geom. highest highest
mean 24-hour 24-hour*
19
* Assumes monitoring every sixth day.
5. If this is a source-impacted site as opposed to a background site,
Source type
Approximate
source strength,
tons per year of TSP
Distance from
monitor
14
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
0 Collect measured TSP concentrations taken in the vi-
cinity of western surface coal mines; estimate distance
from each hi-vol to the major particulate sources at
the mine.
0 Adopt a universal particle size distribution reasonably
expected to simulate the distribution found at the
surface mines for which measured hi-vol concentrations
are available.
0 Using the universal size distribution, model the down-
wind concentration of various size categories of par-
ticulate. Modeled concentrations must be computed at
various downwind distances and under reasonably expected
meteorological conditions.
0 Calculate the ratio of PM 10/TSP concentration (and
PM 5/TSP and PM 2.5/TSP) for several downwind distances.
These ratios are the multipliers used to infer PM 10
(and PM 5 and PM 2.5) concentrations from the measured
TSP concentrations.
0 Multiply the ratios by the measured concentrations,
taking into account the effect of background concentra-
tions .
Each of these five steps is discussed in more detail below.
Collect Data—
The data collection criteria and procedures were described
in Subsection 2.2. Data required for this analysis are the
measured TSP concentration, assumed background TSP concentration,
mean windspeed, and nominal source to hi-vol distance. The
source to hi-vol distance was determined in some instances by
averaging the pit to hi-vol and the haul road to hi-vol distances
measured from maps; in other instances where experience suggested
that one of the sources may be dominant, the source to hi-vol
distance was chosen. In all cases the nominal source to hi-vol
distance is a representative distance over which particulate
matter contributed by the surface mine could be transported to
the hi-vol sampler. Annual windspeeds correspond to the closest
midpoints of the standard default windspeed categories used in
the ISCST dispersion model.
Particle Size Distribution—
The ISC model in its short-term mode (ISCST) simulates
gravitational settling and deposition by selectively removing
particle mass from the air as a function of downwind distance and
meteorological conditions. Just as in the atmosphere, larger
particles settle out faster in the ISCST model than do smaller
particles. As the dust plume moves downwind, the mass of airborne
15
-------
large particles is depleted sooner than the mass of smaller
particles, with the result that the overall particle size distri-
bution changes with distance.
Derivation of particle size distributions by mass fraction
is discussed in Subsection 4.2.1. The distributions are shown in
Table 5.
TABLE 5. PARTICLE SIZE DISTRIBUTIONS BY MASS FRACTION
Fraction less than
Basin
Powder River
Green River/
Hams Fork
San Juan
Universal
2.5
0.020
0.020
0.024
0.021
5.0
0.096
0.082
0.103
0.094
10
0.287
0.234
0.289
0.270
15
0.447
0.365
0.442
0.418
20
0.567
0.474
0.557
0.533
30
1.000
1.000
1.000
1.000
Once the size distribution had been established, the settling
velocity and reflection coefficient were determined using stan-
dard methods described in the Industrial SourceComplex (ISC)
Dispersion Model User's Guide. These computations are summarized
in Tables 6 and 7.
TABLE 6. COMPUTATION OF SETTLING VELOCITY3
Particle diameter
size range,
microns
0-2.5
2.5-5.0
5.0-10.0
10.0-15.0
15.0-20.0
20.0-30.0
Mean radius,
cm
6.25 x 10~;j
1.88 x 10"T
3.75 x 10~;
6.25 x 10 7
8.75 x 10"^
1.25 x IQ~*
Settling velocity,
m/s
0.000093
0.000837
0.. 003347
0,. 009297
0.018223
0.. 037189
Settling velocity determined from the equation.
16
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
V = £
2 p g r'
9 u
where g = acceleration due to gravity, 980 cm/s2
p = particle density, 2.0 gm/cm3
H = air viscosity, 1.83 x 10 4 gm/cm-s
TABLE 7. REFLECTION COEFFICIENTS
Particle diameter size range,
microns
0-2.5
2.5-5.0
5.0-10.0
10.0-15.0
15.0-20.0
20.0-30.0
Reflection coefficient
1.0
0.99
0.86
0.77
0.73
0.65
Model Concentrations—
The most important step in the inferential method is to use
the ISCST model to calculate the mass fraction of PM 10 to TSP
concentration. The mass fraction of PM 10 particulate to TSP
concentration is given by:
XPM 10
PM10
'TSP
where x = concentration
F = mass fraction
PM 10 = particulate smaller than 10 microns
TSP = total suspended particulate
The mass fraction, F „ ,Q, is a function of initial particle size
distribution, windspeed, stability class, and downwind distance.
The value of FpM -Q is not a function of wind direction, cross-
wind distance, or initial particle emission rate. To compute the
values of FpM ,» at various windspeeds, stability classes, and
downwind distances, the ISCST model was used to simulate coin-
cident groundlevel area sources of 0.75 kilometers on a side.
Each source, representing a discrete particle size range,
assumed to emit 10 x 10
v - • was
grams/meter -second of particulate
matter within one of the six particle size categories shown in
17
-------
Table 6. The ISCST model was exercised to simulate all possible
combinations of National Weather Service windspeed categories
(0-3, 4-6, 7-10, 11-16, 17-21, and greater than 21 knots) and
atmospheric stability classes (A, B, C, and D); concentrations
within each of the six particle size categories were computed by
the ISCST model at downwind distances of 1,000, 1,500, 2,000,
3,000, 4,500, 7,000, 10,000, and 20,000 meters. The tabular
output of the model is displayed in Appendix A. Each entry in
Appendix A is the groundlevel concentration that would be detected
downwind of an area source emitting 10 x 10 grams/ meter -second,
at the indicated windspeed, stability class, and downwind distance.
Compute Mass Ratios--
The next step was to weight the modeled concentrations by
the assumed universal particle size distribution by multiplying
each concentration within a given particle size range by its
appropriate emission mass fraction. The computations involved in
this step for "D" stability and two windspeed categories are
included in Appendix B. The rationale for this weighting is
straightforward: each area source used in the model was assumed
to emit at a constant 10 x 10~ grams/meter -second, but in fact
the emission rate of particulate matter within a given size range
is given by the universal distribution illustrated in Table 5.
Finally, the F 1Q, F 5, and F_M „ 5 ratios are calculated by
summing the total particulate mass witnin a given size range, and
dividing that by the TSP concentration. These computations also
appear in Appendix B.
As an example of how the inferential mass fraction method is
employed, consider the computation of FpM .._ at 1,000 meters
downwind distance, under "D" stability ar a windspeed of 4.30
meters/second. The modeling results in Appendix A indicate that
the groundlevel concentrations induced by an emission rate of 10
x 10 grams/meter -second are as shown in Table 8.
TABLE 8. GROUNDLEVEL CONCENTRATIONS AT 1,000 METERS, "D" STABILITY,
AND 4.30 METERS/SECOND WINDSPEED
Particle size range,
microns
0-2.5
2.5-5
5-10
10-15
15-20
20-30
Concentration,
(jg/m3
42.77
42.57
39.84
38.02
37.31
35.87
Note that the concentration for progressively larger particle
sizes decreases. This result is as expected since more of the
18
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
larger particles settle out before reaching the receptor 1,000
meters away.
Next, the relative frequency of occurrence of each of the
particle size categories at the source is taken into account by2
dividing the concentrations in Table 8 by 10 x 10~ grams/meter •
second, and multiplying the concentration by the previously
adopted universal particle size distribution. This step is
illustrated in Table 9.
TABLE 9. RELATIVE GROUNDLEVEL CONCENTRATIONS AT 1,000 METERS,
"D" STABILITY, AND 4.30 METERS/SECOND WINDSPEED
Particle size,
microns
0-2.5
2.5-5
5-10
10-15
15-20
20-30
Modeled concentration,
X/Q, s/m
4.28
4.26
3.98
3.80
3.73
3.59
Universal mass
fraction
0.021
0.073
0.176
0.148
0.115
0.467
Relative concentration,
(jg/m3
0.090
0.311
0.700
0.562
0.429
1.676
Finally, the fraction of particulate mass smaller than 10 microns
can be determined by summing the relative concentrations with
diameters less than 10 microns, and dividing this quantity by the
total concentration less than 30 microns. Specifically, the
relative concentrations smaller than 10 microns from Table 9 is
given by 0.090 + 0.311 + 0.700 = 1.101; the total relative con-
centration is 3.768. The ratio of these numbers, 1.101/3.768 =
0.292, represents the fraction of PM 10 particulate matter con-
tained in the total suspended particulate. Ths finding suggests
that at a distance of 1,000 meters from a dust source, under "D"
stability and with a windspeed of 4.30 meters/second, 29 percent
of the TSP collected by a hi-vol is smaller than 10 microns in
diameter. In a manner similar to the above computation, the
fraction of 2.5 micron and 5 micron diameter particulate was
computed for a number of downwind distances. Although it was
initially intended that a separate fraction would be computed for
each possible combination of stability class and windspeed, this
proved to be unnecessary since the annual average windspeeds at
the candidate mines covered only two windspeeds classes, namely
4-6 knots and 7-10 knots. Furthermore, it was decided that the
most prevalent stability class, "D", would adequately represent
stability conditions for the annual time periods (Turner 1969).
This is believed to be a reasonable approximation for the purposes
of this study. The influence on the study results of assuming D
stability is examined in Subsection 2.5.
19
-------
Multiply Ratios by Measured Concentrations—
The 2.5, 5.0, and 10 micron particle concentration fractions
of TSP are computed in Appendix B and are plotted in Figures 3
and 4 as a function of downwind distance. As expected, the ratio
of small particle mass to total suspended particulate increases
with distance since the larger particles are removed from the
plume with distance. Eventually, the mass fraction of small
particulate to TSP would equal 1.0, but Figures 3 and 4 present
the mass fraction only as far as 20 kilometers. Figures 3 and 4
provide the means to determine small particle concentrations
directly given downwind distance, meteorological conditions, and
measured TSP concentration.
2.3.2 Application of Inferential Method
The ISCST model was configured so that the southwest corner
of the area sources coincided with origin of the grid system,
whereas the distances reported for each of the hi-vol concentra-
tions were measured from the nearest edge of the mine, pit, or
haul road.
It was also necessary to account for background concentra-
tion. Since the inferential method discussed so far only applies
to particulate matter contributed by the surface mine, measured
hi-vol concentrations must be corrected by first subtracting
background concentration. Similarly, the final determination of
PM 10 (or PM 5 or PM 2.5) must also be corrected by adding a
representative background value of PM 10 (or PM 5 or PM 2.5).
Combining these corrections yields the following equations:
PM 10 = [(Measured TSP- background TSP)(PM 10 fraction)] +
PM 10 background
PM 5 = [(Measured TSP- background TSP)(PM 5 fraction)] +
PM 5 background
PM 2.5 = [(Measured TSP- background TSP)(PM 2.5 fraction)] +
PM 2.5 background
The PM 10 background concentration assumed for all of the
mines was generated from preliminary data from the 1980-1981
Western Energy Resource Development Area (WERDA) study. During
this 2-year study, 72-hour concentration measurements of PM 2.5
and PM 15 were taken twice a week. Data were collected at 40
remote sites in 8 different states. A limited data summary was
available for 22 of these sites. However, there was insufficient
information in this summary to select a background concentration
specific to each mine. Consequently, with the concurrence of the
EPA project officer, the mean PM 2.5 and PM 15 concentrations
were calculated for the limited data from the 22 sites. Because
specific data were not available, it was necessary to use these
20
-------
1
1
1
1
1
1 0.5
0.4
1 0.3
I £ 0-2
u_
o
1 1 o,
o
1
1 °°
i 0.05
| 0.04
0.03
™ 0.02
1
0.01
1
1
1
1 1 1 1 1 1 1 I 1 1 1 1 1 1 1 M_
-
-
pM5
-
-
_____ pM25
-
1 1 1 1 1 1 1 1 1 | III
2345 10 20 30 40 50 100
DOWNWIND DISTANCE, km
jure 3. Mass fraction of TSP, stability class D, 4.3 m/s.
1
1
1
-------
1.0
T i i n r
1 i i i n M_J
o
z
o
0.5
0.4
0.3
0.2
0.1
PM 10
PM 5
t/1
0.05
0.04
0.03
0.02
PM 2.5
0.01
2345 10 20 30 40 50
DOWNWIND DISTANCE, km
100
Figure 4. Mass fraction of TSP, stability class D, 2.5 m/s.
22
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
means as an approximation for use with the measured TSP back-
ground concentrations. The average concentrations for PM 2.5 and
PM 15 were 4 pg/m3 and 11 pg/m3. Background concentrations for
PM 5 and PM 10 of 6 pg/m3 and 9 |jg/m3 were obtained by inter-
polating between PM 2.5 and PM 15 and assuming a lognormal dis-
tribution of concentrations between 2.5 pm and 15 pm.
2.4 DESCRIPTION OF RESULTS
2.4.1 Annual Average Concentrations
The application of the inferential method, using the two
corrections above, is detailed in Table 10. The right-hand
columns in that table present the inferred PM 2.5, PM 5, and PM
10 concentrations. The annual average TSP and inferred PM 10
concentrations are shown graphically on Figures 5 to 32. Each
figure is drawn to the same scale and orientation and shows the
approximate location of the monitor, mine boundary, pit, main
haul road and coal preparation plant. Limited production data
are also presented.
The inferred concentrations of PM 10, PM 5, and PM 2.5
presented in Table 10 are generally very small, and only slightly
greater than the respective background concentrations. At the
distances examined (938 m to 5605 m), the mine contribution
(inferred concentration - background concentration) of PM 10, PM
5 and PM 2.5 ranges from 0.2 to 19.8, 0.1 to 7.2, and <0.1 to 1.6
pg/m3 respectively. Obviously, the background concentration is
the dominant factor in almost all cases.
2.4.2 Second-Highest 24-Hour Concentrations
The procedure for inferring PM 10 concentrations from TSP
monitored values was also applied to second-highest 24-hour TSP
monitored values. The background concentrations assumed in the
analysis of annual average concentrations were carried forward to
this analysis. The results appear in Table 11. The table indi-
cates the monitored TSP second-highest 24-hour concentration, the
70 and 90 percentile values, and the inferred PM 10 second-highest
24-hour concentration. At the distances examined (938 m to 5605
m), the mine contribution (inferred contribution-background
concentration) of PM 10, PM 5, and PM 2.5 ranges from 9.9 to
89.4, 3.6 to 32.4, and 0.8 to 7.3 pg/m3, respectively. In con-
trast to the annual average PM 10 concentrations where the back-
ground PM 10 concentration was usually the dominant factor, the
mine contribution is the dominant factor in the 24-hour averaging
period. Also of note is that the short-term PSD Class II incre-
ment of 37 pg/m3 was violated by the inferred PM 10 concentration
outside mine boundaries.
23
-------
TABLE 10. INFERRED PM 10 ANNUAL AVERAGE CONCENTRATIONS
Inferred cone. ,
ug/m3
Assumed background
cone. , pg/m3
0
rH
O.
in
s:
m
CM
Q.
0
rH
Q.
O.
in
CM
Q.
-0
C
T3 3
cu o • n
E t- 0- U E
3 cnoo c \
in -*: h- O O)
in u o a.
< TO
.Q
CU
0 t- Q.
Z 0 E
TO
w
•o
CO
£-
3 Q. E C . «
in co o TO u E
TO 1— CO CO C \
cu 0} E o cjs
s: 03.
'ro • -
3 cn m in
c ro 5 E
in
cn oo o in
in in ID m
TO TO
p». in ID o
ID rH «t in
CM CM CM CM
co co co co
oo cn oo o
«* co CM in
in pv ID cn
CM CM CM CM
CM CO rH CM
in
CO P-. 00
^1- cn CM
rH rH rH
(7^ CTl ^"
p-cnr^
*tf* o> ro
«•«* *
O*i CTl
(£>&(£>
in m in
P^ ID CO
^- m in
CM P-. cn
CO rH P^
CO LT> CM
CO CO CO
«d- o o
oo «t oo
Cn CO rH
rH rH rH
rH CM CO
ID
oo o in
in oo I'M
rH rH rH
<$• ro eo
oo cn p—
ID Pv CO
^-
rH rH rH
O rH rH
ID ID ID
TO
CM in rH
co m p-
co ^- c-J
CO CO CO
«•<*••*
P** CO 01
ro oo «3-
«*• CO C3
rH CO C-J
rH CM OO
P-
ID rH
in rH
rH rH
*t 00
00 ID
m CM
•* **•
cn cn
ID ID
m m
rH rH
m o
in ID
TO ro
in CM
p- CM
CO CM
CO CO
"* ^
CO ro
cn CM
rH «3-
rH CM
CO
TJ
0)
C
•r-
•P
C
O
U
24
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-a
ai
CQ
0
f,
U Q-
c
o
U LD
•o E s:
CU \ 0-
S- O)
5- 3.
0) LD
14-
C CM
s:
CL
Assumed background
cone. , pg/m3
PM 2.5 PM 5 PM 10
•a
c
T3 3
CD 0 . «
E S- 0. (J E
in -^ r— O O)
in u U 3.
•>» ^
C to 5 E
•
•r- O
C Z
o
CU
c •
•p- O
z z
tjDCOCM CM rHCOrv'd-rHrH
ooop*- cn CO^OLO^'**'
CMCMrH rHrHrHrHrHrH
CMCMO rH LOCniDCOOOCn
ococn ID f^p^iDOOP~-r*.
rH rH
<7) *J3 f*1*" CD CO ^^ pH I/O ^f ^J°
^-LD^- <• «-<*«•«-«-«*
cn cn cn cn cn cn cn cn cn cn
IDIDID ID iDlDlDlDIDID.
000000 CM LDLDLDLDlDlD
cninr^ CM oocnvDcn^"cn
r)- in ID ID IDIDIDLDLDLD
1 — r~«-O ID rHOO
-------
TSP = 16.7 pg/m3
PM 10 = 9.5 |jg/m3
1 all*
MINE BOUNDARIES
[Jj PIT LOCATION
— HAUL ROAD
A PUWT LOCATION
• MONITOR LOCATION
Location: Campbell County, Wyoming
Coal production: 10.0 x 106 tons/year
Area disturbed: 50 acres/year
Method of overburden removal: shovel/truck
Quantity of overburden moved: 16.0 x 106 yardVyear
Method of dust control for haul roads: water
Figure 5. Annual average monitored TSP and calculated PM 10
concentrations, northeast Wyoming, Mine 1.
26
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
TSP
PM 10
TSP
PM 10
32.7 ug/m3
14.2 ug/m3»
24.4 ug/m3.
11.8 ug/m3
TSP = 27.8 H9/nv
PM 10 = 12.8 ug/m3
41.8 ug/m3
16.8 ug/m3
MINE BOUNDARIES
] PIT LOCATION
HAUL ROAD
PLANT LOCATION
MONITOR LOCATION
Location: Campbell County, Wyoming
Coal production: 8.0 x 106 tons/year
Area disturbed: 65 acres/year
Method of overburden removal: shovel/truck
Quantity of overburden moved: 3.3 x 106 yardVyear
Method of dust control for haul roads: water
Figure 6. Annual average monitored TSP and calculated PM 10
concentrations, northeast Wyoming, Mine 2.
27
-------
TSP = 18.0 pg/m3.
PM 10 = 9.9 ug/m3
TSP =27.2 ug/m3
PM 10 = 12.6 ug/m3
MINE BOUNDARIES
£~j PIT LOCATION
— HAUL ROAD
A PLANT LOCATION
• MONITOR LOCATION
Location: Campbell County, Wyoming
Coal production: 8.2 x 10s tons/year
Area disturbed: 109 acres/year
Method of overburden removal: shovel/truck
Quantity of overburden moved: 5.8 x 106 yardVyear
Method of dust control for haul roads: water
Figure 7. Annual average monitored TSP and calculated PM 10
concentrations, northeast Wyoming, Mine 3.
28
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
TSP
PM 10
25.6 Ljg/n>3
12.1
TSP = 26.7
PM 10 = 12.4 ug/m3
TSP = 21.5
PM 10 = 10.9 ug/m3
1 Hilt
MINE BOUNDARIES
["] PIT LOCATION
— HAUL ROAO
A PLANT LOCATION
• MONITOR LOCATION
Location: Campbell County, Wyoming
Coal production: 2.0 x 10s tons/year
Area disturbed: 133 acres/year
Method of overburden removal: shovel/truck
Quantity of overburden moved: 3.5 x 106 yardVyear
Method of dust control for haul roads: water/coherex
Figure 8. Annual average monitored TSP and calculated PM 10
concentrations, northeast Wyoming, Mine 4.
29
-------
TSP = 24.6 ug/m3
PM 10 = 11.8 ug/m3
TS*P
PM 10
25.0 ug/m3
11.9 ug/m3
1 mile
I j MINE BOUNDARIES
uJi PIT LOCATION
— HAUL ROAD
A PLANT LOCATION
• MONITOR LOCATION
Location: Campbell County, Wyoming
Coal production: 4.5 x 106 tons/year
Area disturbed: 98 acres/year
Method of overburden removal: shovel/truck
Quantity of overburden moved: 5.7 x 106 yardVyear
Method of dust control for haul roads: water
Figure 9. Annual average monitored TSP and calculated PM 10
concentrations, northeast Wyoming, Mine 5.
30
-------
I
I
I
I
1
I
I
I
I
I
I
I
I
1
I
I
TSP
PM 10
TSP
PM 10
33.2 ug/m3.
14.3 ug/m3
27.9 ug/m!
12.8 ug/m3
3-4-
TSP = 51.7 ug/m3
PM 10 = 19.7 ug/m3
MINE BOUNDARIES
("\ PIT LOCATION
— HAUL ROAD
A PLANT LOCATION
• MONITOR LOCATION
Location: Campbell County, Wyoming
Coal production: 6.5 x 106 tons/year
Area disturbed: 63 acres/year
Method of overburden removal: shovel/truck
Quantity of overburden moved: 9.6 x 106 yardVyear
Method of dust control for haul roads: water
Figure 10. Annual average monitored TSP and calculated PM 10
concentrations, northeast Wyoming, Mine 6.
I
I
31
-------
TSP = 38.2 yg/m3
PM 10 = 15.8 yg/m3
- TSP = 45.5 yg/m3
PI^IO = 18.0 yg/m3
TSP = 27.1 yg/m3
PM 10 = 12.5 yg/m3
MINE BOUNDARIES
L'j PIT LOCATION
— HAUL ROAD
A PLANT LOCATION
• MONITOR LOCATION
Location: Campbell County, Wyoming
Coal production: 16.0 x 106 tons/year
Area disturbed: 197 acres/year
Method of overburden removal: shovel/truck
Quantity of overburden moved: 23.6 x 106 yardVyear
Method of dust control for haul roads: water
Figure 11. Annual average monitored TSP and calculated PM 10
concentrations, northeast Wyoming, Mine 7.
32
-------
I
I
I
I
i
I
I
I
I
I
l
l
l
l
l
l
I
l
I
TSP =37.5 ug/m3
*PM 10 = 15.6 ug/m3
TSP =22.2 ug/m3
PM 10 = 11.1
MINE BOUNDARIES
["I PIT LOC*TIOH
— HAUL ROAD
A PLANT LOCATION
• MONITOR LOCATION
Location: Sheridan County, Wyoming
Coal production: 4.3 x 106 tons/year
Area disturbed: 58.6 acres/year
Method of overburden removal: scraper/shovel-truck
Quantity of overburden moved: 13.7 x 106 yardVyear
Method of dust control for haul roads: water
Figure 12. Annual average monitored TSP and calculated PM 10
concentrations, northeast Wyoming, Mine 8.
33
-------
r
TSP =57.7 ug/m3
,,,v • PM 10 = 20.6 ug/m3
\ \
>'' TSP = 85.7 ug/m3
/ • PM 10 = 28.8 ug/m3
TSP =46.0 ug/m3
*PM 10 = 17.2 ug/m3
( J PIT LOCATION
— HAUL ROAD
A PLANT LOCATION
, ,1le ' * MONITOR LOCATION
NOTE: MINE BOUNDARIES COULD NOT BE OBTAINED
Location: Sweetwater County, Wyoming
Coal production: 2.4 x 106 tons/year
Area disturbed: 164 acres/year
Method of overburden removal: dragline
Quantity of overburden moved: 12.0 x 106 yardVyear
Method of dust control for haul roads: water
Figure 13. Annual average monitored TSP and calculated PM 10
concentrations, southwest Wyoming, Mine 9.
34
-------
I
I
I
I
I
I
I
I
t
i
i
i
i
i
i
i
i
i
i
TSP
PM 10
22.6 ug/m3
9.2 ug/m3
[~] MINE BOUNDARIES
[ J PIT LOCATION
— HAUL ROAD
A PLANT LOCATION
• MONITOR LOCATION
Location: Jackson County, Colorado
Coal production: 0.7 x 106 tons/year
Area disturbed: 15 acres/year
Method of overburden removal: scraper
Quantity of overburden moved: 4.5 x 106 yardVyear
Method of dust control for haul roads: water
Figure 14. Annual average monitored TSP and calculated PM 10
concentrations, northwest Colorado, Mine 10.
35
-------
TSP =
PM 10 =
TSP =20.6 ug/m3
PM 10 = 10.7 pg/m3
TSP = 36'7
PM10 = 15.4
•— TSP = 32.1 pg/m3
PM 10 = 14.1 pg/m3
TSP =32.4 pg/m3
PM 10 = 14.1 pg/m3
TSP = 29.1 pg/m3
PM 10 = 13.1 pg/m3
H11*
NOTE: NINE BOUNDARIES COULD NOT BE OBTAINED
[ J PIT LOCATION
— HAUL ROAD
A PLANT LOCATION
• MONITOR LOCATION
Location: Big Horn County, Montana
Coal production: 13.9 x 106 tons/year
Area disturbed: 4333 acres/year
Method of overburden removal: dragline/shovel-truck/scraper
Quantity of overburden moved: 44.8 x 106 yardVyear
Method of dust control for haul roads: water/1ignon sulfonate
Figure 15. Annual average monitored TSP and calculated PM 10
concentrations, southeast Montana, Mine 11.
36
-------
I
I
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
TSP =36.0 ug/m3
PM 10 = 14.6 ug/m3
TSP =26.0 ug/m3
PM 10 = 10.8
TSP =39.0
PM 10 = 13.7 |jg/m3
MINE BOUNDARIES
[~J PIT LOCATION
— HAUL ROAD
A PLANT LOCATION
• MONITOR LOCATION
Location: McLean County, North Dakota
Coal production: 3.1 x 106 tons/year
Area disturbed: 207 acres/year
Method of overburden removal: dragline
Quantity of overburden moved: 19.5 x 106 yardVyear
Method of dust control for haul roads: water
Figure 16. Annual average monitored TSP and calculated PM 10
concentrations, western North Dakota, Mine 12.
37
-------
TABLE 11. INFERRED PM 10 SECOND-HIGHEST 24-HOUR CONCENTRATIONS
County
Campbel 1
Campbell
Campbel 1
Campbell
Campbell
Campbel 1
Campbel 1
Sheridan
Sweetwater
Jackson
McLean
State
Wyo.
Wyo.
Wyo.
Wyo.
Wyo.
Wyo.
Wyo.
Wyo.
Wyo.
Colo.
N. Dak.
Mine
No.
1
2
3
4
5
6
7
8
9
10
12
Monitor
No.
I
1
2
3
4
I
2
1
2
3
1
2
1
2
3
1
2
3
1
2
1
2
3
1
1
2
3
Measured
TSP
second
highest
24-hour
cone. ,
HQ/m3
49
72a
79a
64a
119a
56a
149
75
92
59
93a
59a
103
150
97
149a
138
79
180a
76a
224
324
128
99
126a
133a
151a
Per-
centi le
70
32
48
43
32
53
30
42
37
38
32
39
37
48
76
42
56
73
37
56
27
93
127
59
b
b
b
b
90
44
66
63
51
82
44
74
65
71
45
67
56
76
114
70
123
106
55
103
55
165
259
102
b
b
b
b
Inferred cone. ,
|jg/m3
2.5
4.8
5.4
5.5
5.2
6.5
5.0
7.2
5.4
5.8
5.0
5.9
5.0
6.1
7.2
6.0
7.2
7.0
5.5
8.0
5.5
8.9
11.3
6.6
6.0
6.5
6.7
7.1
5.0
9.6
12.1
12.8
11.2
17.0
10.4
20.3
12.4
14.2
10.7
14.3
10.7
15.3
20.3
14.7
20.2
19.3
12.8
23.5
12.6
28.0
38.4
17.8
15.1
17.3
18.2
19.9
10
18.9
25.7
27.9
23.4
39.4
21.2
48.8
26.6
31.6
21.9
31.8
21.9
34.7
48.4
32.9
48.1
45.3
27.7
57.2
27.2
69.4
98.4
41.2
33.6
40.0
42.6
47.3
a
b
Outside mine boundary.
Data not provided.
38
-------
I
I
I
I
I
I
I
a
i
i
i
i
i
i
i
i
i
i
i
2.5 ERROR ANALYSIS AND ASSUMPTIONS
A number of assumptions were made in the derivation and
application of the inferential analysis method, and these should
be recognized and discussed along with the findings. The fore-
most assumption, and one which has the greatest potential for
influencing inferred PM 10, PM 5, and PM 2.5 concentrations, is
the choice of background concentrations. It is conceivable that
the background concentrations of PM 10 could be equal to the TSP
background, that is to say, it is possible that all of the mea-
sured background concentration could be comprised of particulate
matter smaller than 10 microns in diameter. Where this is the
case, then inferred PM 10 concentrations in Table 10 could increase
by as much as 13 |jg/m3 over and above those reported. Additionally,
the actual background TSP concentrations could be appreciably
higher than those assumed, particularly during worst-case short-
term periods, with the same effect of increasing the inferred
concentrations. The magnitude of these background concentrations
is simply not known to the degree of accuracy desirable. Addi-
tional monitoring is required to quantify background concentra-
tions by particle size under different conditions.
A second major assumption in the inferential method is that
the ISCST model correctly describes the transport and deposition
of particulate matter. If the model simulates different size
categories with varying degrees of accuracy, then the resulting
mass fractions would be altered.
The adoption of the universal particle size distribution
could also have a pronounced impact on inferred concentrations.
While Table 5 suggests that the distribution of particle mass
within specific size categories is reasonably uniform, the mass
fraction of PM 10 assumed in the universal distribution could be
in error by 100 percent. This error in turn would induce errors
in the inferred concentrations in Table 10 of roughly a factor of
two.
In light of the uncertainties associated with the calcula-
tion of the inferred concentrations presented in Table 10, these
findings should be viewed as preliminary.
Three of the assumptions employed in the procedure were
examined more closely. These assumptions are the imposition of
fixed windspeeds, fixed D stability class, and the choice of
nominal receptor to source distances.
The influence of windspeeds can be examined by comparing the
curves in previously cited Figures 3 and 4. These curves have
been replotted on Figure 17. The PM 10 fraction is higher at the
lower windspeed since the lower windspeed would cause comparitively
greater deposition of larger particles per unit of distance. The
39
-------
1.0 t
0.5
0.4
0.3
0.2
2 0.1
0.05
0.04
0.03
0.02
0.01
MAXIMUM DISTANCE
USED IN ANALYSIS
I I I I I I I I I
PM 10
PM 5
PM 2.5
I I I I I I I I
I1III I M_
2.5 m/s
4.3 m/s
2.5 m/s
4.3 m/s
2.5 m/s
4.3 m/s
2345 10 20 30 40 50
DOWNWIND DISTANCE, km
100
Figure 17. Sensitivity of the procedure to infer PM 10
concentrations to windspeed assumptions.
40
-------
I
I
I
i
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
distance between the curves represents only a portion of the
possible range of values because extreme low or high windspeeds
would widen the range. However, using an average annual wind-
speed would still cause the correct values to lie toward the
center of the range. Assumption of a lower windspeed than 4.3
m/s would increase the PM 10 fraction of the TSP concentration.
The influence of the D stability assumption was examined by
overlaying the curves from Figure 3 with curves derived assuming
A stability and 4.3 m/s windspeed. For PM 10, one additional
curve was derived for B stability and 4.3 m/s windspeed. These
curves are shown in Figure 18.
The three curves for PM 10 exemplify one of the limitations
of the ISC model. As discussed in the ISC user's manual, modeling
results for A stability are unreliable at distances greater than
3.0 kilometers. As shown in Figure 18 the model overestimates
the amount of deposition at large downwind distances. The PM 10
comparison of B and D stability classes demonstrates that less
deposition would occur in the unstable case (B stability).
However, at the maximum downwind distance considered in this
study (6 km), the results show that there is very little dif-
ference between the curves for B and D stability.
The influence of the nominal receptor distance appears to be
slight. The mass fractions change only slightly with downwind
distance, so that an error of many hundred meters in the distance
probably only has a negligible effect on concentrations.
41
-------
0.01
MAXIMUM DISTANCE
USED IN ANALYSIS
I I I I M
A STABILITY
A STABILITY
D STABILITY
D STABILITY
A STABILITY
D STABILITY
I
I
I I I I I I
45 10 20
DOWNWIND DISTANCE, km
30 40 50
100
Figure 18. Sensitivity of the procedure to infer PM 10
concentrations to stability class assumptions.
42
-------
I
I
I
I
I
I
I
I
1
I
I
I
I
I
I
I
I
I
I
SECTION 3.0
CHARACTERIZATION OF PM 10 AND TSP AIR QUALITY AROUND WESTERN
SURFACE COAL MINES USING PREVIOUS MODELING STUDIES
3.1 BACKGROUND
Particulate dispersion modeling for TSP has been performed
in support of the mine permitting process in many cases. Unfor-
tunately, there are no standardized procedures for performing the
analyses. Previous modeling efforts have utilized various lists
of sources, emission factors, and models.
The two most common sets of emission factors used are from
the EPA Region VIII Interim Policy Paper (EPA 1978a) and from the
State of Wyoming (1979). Each set of factors is a combination of
individual emission factors gathered from several references.
The Interim Policy Paper (IPP) considers more emission sources
than the Wyoming factors. The IPP factors are intended for use
with a model particulate fallout function, whereas the Wyoming
factors are not.
Two variations of the Climatological Dispersion Model (CDM)
have been predominately used for long-term modeling. The State
of Wyoming had adapted CDM into a CDM-W model and specified its
use for permit work. Therefore, all recent permit work in Wyoming
has been performed with the CDM-W model, and it has also found
some application in Colorado and Montana. Over 40 modeling
studies have been performed for the Bureau of Land Management
(BLM) in connection with their tract leasing program. For this
work, various modifications, including insertion of a fallout
function, have been made to the CDMQC model.
Short-term modeling in support of the mine permitting process
is performed much less frequently. The State of Wyoming does not
require short-term modeling. Elsewhere, the PAL (EPA 1978b), RAM
(EPA 1978c), or Valley (EPA 1977) models have been used. In
general, short-term modeling is considered less accurate than
long-term modeling because of the difficulty in defining short-
term activity patterns, emissions, and meteorological parameters.
In all mine permitting cases, the analyses have been per-
formed for the Total Suspended Particulate (TSP) size range.
Modeling for <30 (jm particle sizes has been limited to a few
theoretical studies which are reviewed in Subsection 3.3.
43
-------
This section describes several previous modeling studies and
illustrates TSP concentrations around surface coal mines with
concentration isopleths. In a change from the original scope of
work, no effort was made to adjust or evaluate the previous
modeling studies. This change was approved by the contract
officer.
3.2 PREVIOUS MODELING STUDIES
Three geographical areas were selected for the study of
model applications in this section, and the scenario analysis in
Section 4.0. The three geographical areas are the Powder River
Basin (near Gillette, Wyoming); the San Juan Basin (near Farmington,
New Mexico); and the Green River/Hams Fork Basin (southern Wyoming,
and northwest Colorado). These basins are shown in Figure 19.
Reasons for selecting these basins are:
1. The three areas are major coal producing areas with signifi-
cantly different characteristics relating to total produc-
tion level, mining methods, and meteorological conditions.
2. The characteristics of the three areas are suitable for use
in the scenario analysis described later in Section 4.0.
The new EPA emission factors were derived from field testing
in two of the three basins. The factors are more accurate
when applied to areas with emission factor correction para-
meters in the same range over which testing occurred.
3. The study team has a significant amount of file data for
these three basins allowing for a more efficient use of
project resources.
4. Focusing on three areas allows the study team to reduce the
number of variables involved in the task to a more manageable
level.
The primary source for the modeling studies analyzed in this
section was the study team files. Project resources allowed a
limited search for additional modeling studies from the Technical
Advisory Group and the State of Wyoming.
3.2.1 Single Mine Modeling
The results of 16 previous modeling studies from the Powder
River Basin and the Green River/Hams Fork Basin were collected.
No previous San Juan modeling studies were available from the
resources listed above.
44
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
<\
(U
to
(O
O
Ol
CO
0£ Z
O •-•
Z U. «/)
•— " «t
t/1 t/1) CO
co S ex:
o: i-" z
a: <:
C£
-------
The mining sources and associated emission factors used in
the previous analyses are presented in Table 12 along with the
particulate control efficiencies applied. The emission factors
and control efficiencies were taken from Region VIII Interim
Policy Paper (EPA 1978). Utilizing the source list and emission
factors presented in Table 12, along with estimated activity
parameters, an emission inventory was prepared for each mine.
These data, along with spatial parameters, were entered into a
modified version of CDMQC. The climatological data used for
input to the model were obtained from the nearest National Weather
Service station in the form of a stability rose (STAR) deck.
Details of the CDMQC model and its application can be found
in the literature and are not repeated here. However, the fol-
lowing modifications to the model were made for a more accurate
reflection of conditions in a surface coal mine.
1. Insertion of a fallout function for point and area sources.
2. Specification of a 10-meter release height.
3. Removal of a model provision to alter stability classes in a
manner that reflects urban conditions.
The results of the modeling exercises were plotted on a
series of isopleth maps as shown in Figures 20 through 35. Each
figure contains a variety of information. All of the figures are
drawn to the same scale and orientation. At the bottom of the
figures, the mine location, and limited activity data are pre-
sented along with the calculated annual TSP emissions. Also
shown are the mine boundary, approximate locations of the coal
preparation facility, main haul road, and active pit. The iso-
pleth concentrations represent TSP emissions including fugitive
dust. Two concentrations are shown for each isopleth, the pre-
dicted concentration due to the mine and the predicted concentra-
tion plus a background concentration.
The 16 mines presented in the figures represent annual coal
productions ranging from 225,000 tons to 23,200,000 tons. Calcu-
lated emission estimates range from 154 tons to 4443 tons. The
maximum TSP concentration predicted at the mine boundary varies
from 1 to 50 (jg/rn3. As shown in the figures, this value is
highly dependent on the location of the mining activity with
respect to the mine boundary.
In all cases, the predicted concentration decreases rapidly
as the distance from the mining activity increases. The 1 (jg/rn3
isopleth is predicted to occur within 4 miles of all mine bounda-
ries.
46
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
TABLE 12. PARTICULATE EMISSION FACTORS AND CONTROL EFFICIENCIES
USED IN PREVIOUS MODELING STUDIES
Mining activity
Topsoil removal - scraper
Scraper travel - topsoil
Truck dump - topsoil
Stockpile - topsoil
Overburden drilling
Overburden blasting
Overburden removal
Shovel/truck
Dragline
Haul trucks - overburden
Truck dump - overburden
Overburden stockpile
Dozer - overburden
Coal drilling
Coal blasting
Coal loading
Haul trucks - coal
Truck dump - coal
Crushing - coal
Screening - coal
Conveyor - coal
Coal storage
Open piles
Silos
Coal loadout
Road maintenance - grader
Access road travel
Exposed areas
Miscellaneous haul road travel
Powder River
Basin
Emission
factor
0.38
5.62
0.002
0.48
1.5
85.3
0.37
a
5.62
0.002
0.48
16
0.22
72.4
0.0035
9.37
0.007
0.08
0.10
0.20
32.9
0.0002
32
6.56
0.48
6.56
Percent
control
50
85
50
85
85
90
99
99
99
99
99
50
99
75
85
Green River/Hams Fork
Emission
factor
0.38
a
a
a
1.5
85.3
a
0.053
a
a
0.38
16
0.22
72.4
0.12
13.6
0.007
0.08
0.10
0.20
16.9-33.1
0.0002
32
5.3
0.38
a
Percent
control
85
85
85
99
99
90
50
99
95
99
40
Units
Ib/yd3
1 b/VMT
Ib/ton
ton/ac-yr
Ib/hole
Ib/blast
1 b/yd
Ib/yd3
1 b/VMT
Ib/ton
ton/ac-yr
Ib/h
Ib/hole
Ib/blast
Ib/ton
1 b/VMT
Ib/ton
Ib/ton
Ib/ton
Ib/ton
ton/ac-yr
Ib/ton
Ib/h
1 b/VMT
ton/ac-yr
1 b/VMT
Not applicable.
47
-------
•1 ug/m3 (16 ug/m3)
•5 ug/m3 (20 ug/m3)
,10 ug/m3 (25 ug/m3)
?0 ug/m3 (35
MINE BOUNDARIES
PIT LOCATION
— HAUL ROAD
A PLANT LOCATION
mtUTU rnoiCTD
CMcurwtm nm vt*uou» cacuiurio*
Location: Campbell County, Wyoming
Coal production: 23.2 x 106 tons/year
Area disturbed: 183 acres/year
Method of overburden removal: shovel/truck
Quantity of overburden moved: 29.5 x 106 yard3/year
Method of dust control for haul roads: water/chemical
Annual TSP emissions: 2832 tons
Figure 20. Annual average modeled TSP concentrations
Powder River Basin, Mine 1.
48
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1 ug/m3 (16 ug/m3)
5 pg/m3 (20 pg/m3
10 pg/m3 (25 |jg/m3
20 pg/m3 (35 pg/m3)
1 Bill
MINE BOUNDARIES
PIT LOCATION
HAUL ROAD
A PLANT LOCATION
i in muKnwsti IWICAIU mmcra
CMCUTUTIM run ucioou* COKUIUTIOI
Location: Campbell County, Wyoming
Coal production: 14.0 x 106 tons/year
Area disturbed: 91 acres/year
Method of overburden removal: dragline/shovel-truck
Quantity of overburden moved: 40.9 x 106 yard3/year
Method of dust control for haul roads: water/chemical
Annual TSP emissions: 2126 tons
Figure 21. Annual average modeled TSP concentrations
Powder River Basin, Mine 2.
49
-------
1 (jg/m3 (16 ug/m3)
5 ug/m3 (20 ug/m3)
10 ug/m3 (25 ug/m3)
20 ug/m3 (35 ug/m3)
50 |jg/m3 (65 ug/m3)
1 rile
NIT! HMM I> PUEHTHfUS IN01UTES HTCOlCTtO
a»ct«ii«iio« run wcuwua conce»tuiio«
MINE BOUNDARIES
PIT LOCATION
HAUL ROAD
PLAHT LOCATION
Location: Campbell County, Wyoming
Coal production: 18.2 x 106 tons/year
Area disturbed: 190 acres/year
Method of overburden removal: shovel/truck
Quantity of overburden moved: 61.3 x 106 yard3/year
Method of dust control for haul roads: water/chemical
Annual TSP emissions: 4443 tons
Figure 22. Annual average modeled TSP concentrations
Powder River Basin, Mine 3.
50
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1 ug/m3 (16 ug/m3)
-5 ug/m3 (20 ug/m3)
10 ug/m3 (25 ug/m3)
1 ulle
MINE BOUNDARIES
("I PIT LOCATION
— • HAUL TOAD
A PLANT LOCATION
•MCI » MHOncSIS IWlCATti n(01CT(B
coKwruri* MM wcuwuc co>e£«Tuiion
Location: Campbell County, Wyoming
Coal production: 12.8 x 106 tons/year
Area disturbed: 125 acres/year
Method of overburden removal: shovel/truck
Quantity of overburden moved: 23.9 x 106 yard3/year
Method of dust control for haul roads: water/chemical
Annual TSP emissions: 1947 tons
Figure 23. Annual average modeled TSP concentrations
Powder River Basin, Mine 4.
51
-------
1 |jg/m3 (16 pg/m3)
5 |jg/m3 (20 pg/m3)
10 pg/m3 (25 (jg/m3
20 pg/m3 (35 pg/m3
5 pg/m3 (20 pg/m3)
10 pg/m3 (25 pg/m3)
MINE BOUNDARIES
[~j PIT LOCATION
— HAUL ROAD
A PLANT LOCATION
NIMER IN 'UtNTHCSES IIGICATES MfOlCTED
rancepmuTieii mis ucunua CO«CE«HATIO«
Location: Campbell County, Wyoming
Coal production: 12.5 x 106 tons/year
Area disturbed: 242 acres/year
Method of overburden removal: shovel/truck
Quantity of overburden moved: 39.9 x 106 yardVyear
Method of dust control for haul roads: water/chemical
Annual TSP emissions: 3073 tons
Figure 24. Annual average modeled TSP concentrations
Powder River Basin, Mine 5.
52
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
3 (19 ug/m3)
5 ug/m3 (23 ug/m3)'
10 ug/m3 (28 ug/m3)
[j| MINE BOUNDARIES
[Jj PIT LOCATION
—• HAUL ROAD
A HA*T LOCATION
WTt: WWI III MKHTMiai !ICIC*TtS ratDtCTtl
riM MM uaaatm oMU«rurio>
Location: Carbon County, Wyoming
Coal production: 4.0 x 106 tons/year
Area disturbed: 135 acres/year
Method of overburden removal: dragline
Quantity of overburden moved: 15.6 x 106 yardVyear
Method of dust control for haul roads: water/chemical
Annual TSP emissions: 1278 tons
Figure 25. Annual average modeled TSP concentrations
Green River/Hams Fork Basin, Mine 6.
53
-------
1 ug/m3 (23 ug/m3)
5 ug/m3 (27 ug/m3)
10 ug/m3 (32 ug/m3)
MINE BOUNDARIES
• PIT LOCATION
— HAUL ROAD
A PLANT LOCATION
•art *uw« in 'uciiTHCSEs INOICATK
CDKEnrurioii run wcumw COCIHTUTIMI
Location: Carbon County, Wyoming
Coal production: 2.7 x 106 tons/year
Area disturbed: 117 acres/year
Method of overburden removal: dragline
Quantity of overburden moved: 14.8 x 106 yard3/year
Method of dust control for haul roads: water/chemical
Annual TSP emissions: 1075 tons
Figure 26. Annual average modeled TSP concentrations
Green River/Hams Fork Basin, Mine 7.
54
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1 pg/m3 (23 ug/m3)_
5 ug/m3 (27 M9/m3>
MINE BOUNDARIES
X5 yg/m3 (28 yg/m3) —MAUL ROAD
PLANT LOCATION
WMCI » MMHTKUS IWIUfCS HtDICTtD
Tiai nus iKiunM cnccmurion
Location: Carbon County, Wyoming
Coal production: 1.7 x 106 tons/year
Area disturbed: 134 acres/year
Method of overburden removal: dragline
Quantity of overburden moved: 12.7 x 106 yardVyear
Method of dust control for haul roads: water/chemical
Annual TSP emissions: 1179 tons
Figure 27. Annual average modeled TSP concentrations
Green River/Hams Fork Basin, Mine 8.
55
-------
1 ug/m3 (23 ug/m3)
MINE BOUNDARIES
J PIT LOCATION
HAUL ROAD
A PUNT LOCATION
•DTI: KMCI 11 MI»T>*SES IW1CATES 1CDICTCO
CWKinTMTIO" nil! WCUMIM COCEKIUTI0>
Location: Carbon County, Wyoming
Coal production: 1.0 x 106 tons/year
Area disturbed: 44 acres/year
Method of overburden removal: dragline
Quantity of overburden moved: 5.6 x 106 yardVyear
Method of dust control for haul roads: water/chemical
Annual TSP emissions: 632 tons
Figure 28. Annual average modeled TSP concentrations
Green River/Hams Fork Basin, Mine 9.
56
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
ug/m3 (23 ug/m3)
ug/m3 (27 ug/m3)
MINE BOUNDARIES
_"] PIT LOCATION
HAUL ROAD
A PLANT LOCATION
» nmrMscs natures
CONCENTUTiaH »1US tVg"?1"* COMCENTIUTiaM
Location: Carbon County, Wyoming
Coal production: 2.1 x 106 tons/year
Area disturbed: 91 acres/year
Method of overburden removal: dragline
Quantity of overburden moved: 11.6 x 106 yardVyear
Method of dust control for haul roads: water/chemical
Annual TSP emissions: 901 tons
Figure 29. Annual average modeled TSP concentrations
Green River/Hams Fork Basin, Mine 10.
57
-------
1 ug/m3 (23 ug/m3)
5 ug/m3 (27 Mg/m3)
20 ug/m3 (42 ug/m3)
n1le
MINE BOUNDARIES
L"i PIT LOCAT1°N
— HAUL ROAD
A PLANT LOCATION
•ore lumen in PUCKTHCICS IWIOHS "loicuo
COHCtNtUIlOH rtUS UCKCKXM CONCENIUT10M
Location: Moffat County, Colorado
Coal production: 1.4 x 106 tons/year
Area disturbed: 27 acres/year
Method of overburden removal: dragline
Quantity of overburden moved: 5.4 x 106 yard3/year
Method of dust control for haul roads: water/chemical
Annual TSP emissions: 474 tons
Figure 30. Annual average modeled TSP concentrations
Green River/Hams Fork Basin, Mine 11.
58
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1 yg/m3 (23 yg/m3)
5 yg/m3 (27 yg/m3)
10 yg/m3 (32 yg/m3)
30 yg/m3 (52 yg/m3)
NINE BOUNDARIES
PIT LOCATION
HAUL ROAD
A PLANT LOCATION
_
•en «Mfi in Mmmus IWJOTII moicrto
CMCMTMTIOII HIK IMUKIM CMCMTUT10II
Location: Moffat County, Colorado
Coal production: 1.9 x 106 tons/year
Area disturbed: 60 acres/year
Method of overburden removal: draglines/truck-shovel
Quantity of overburden moved: 9.6 x 106 yardVyear
Method of dust control for haul roads: water/chemical
Annual TSP emissions: 1039 tons
Figure 31. Annual average modeled TSP concentrations
Green River/Hams Fork Basin, Mine 12.
59
-------
1 ug/m3 (23 ug/m3)
5 ug/m3 (27 ug/m3)-
10 ug/m3 (32 ug/m3)
1 Bile
[j MINE BOUNDARIES
[ ] PIT LOCATION
—— HAUL ROAD
A PLANT LOCATION
WTt IWKI in MMHTHtSCS IKIUTES HHUCTtD
CWICtBTMTlOIC PLUS MCUttUW CWCEKTUTIW
Location: Moffat County, Colorado
Coal production: 2.7 x 106 tons/year
Area disturbed: 114 acres/year
Method of overburden removal: dragline
Quantity of overburden moved: 18.1 x 106 yardVyear
Method of dust control for haul roads: water/chemical
Annual TSP emissions: 1224 tons
Figure 32. Annual average modeled TSP concentrations
Green River/Hams Fork Basin, Mine 13.
60
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1 ug/m3 (23 ug/m3)
10 ug/m3 (32 ug/m3
1 Hi It
««t. m**a M MMHTWSU IM1UTIS MfOlCTO
caKUTuri* KM uaaam cocnruTioi
f_~j MINE BOUNDARIES
[_"j PIT LOCATION
— HAUL ROAD
A PLANT LOCATION
Location: Routt County, Colorado
Coal production: 0.2 x 106 tons/year
Area disturbed: 17 acres/year
Method of overburden removal: scrapers/truck-shovel
Quantity of overburden moved: 2.0 x 106 yardVyear
Method of dust control for haul roads: water/chemical
Annual TSP emissions: 154 tons
Figure 33. Annual average modeled TSP concentrations
Green River/Hams Fork Basin, Mine 14.
61
-------
|jg/m3 (23 ug/m3)
ug/n>3 (27 ug/in3)
10 ug/m3 (32 ug/m3)
N
I
I, j MINE BOUNDARIES
(Tj PIT LOCATION
— MAUL ROAD
A PLANT LOCATION
D in nufiincsts imuns MKOICTED
TiM rua Mcnaouw COCUTUTIOI
Location: Routt County, Colorado
Coal production: 2.8 x 106 tons/year
Area disturbed: 55 acres/year
Method of overburden removal: dragline
Quantity of overburden moved: 8.5 x 106 yardVyear
Method of dust control for haul roads: water/chemical
Annual TSP emissions: 775 tons
Figure 34. Annual average modeled TSP concentrations
Green River/Hams Fork Basin, Mine 15.
62
-------
1 ug/m3 (23 ug/m3)
5 ug/m3 (27 ug/m3)
1 milt
MINE BOUNDARIES
PIT LOCATION
HAUL ROAO
PLANT LOCATION
«M» in numnfsu IWIUTU nKoicrn
OKinwriiM run uetaam coctururioii
Location: Routt County, Colorado
Coal production: 1.3 x 106 tons/year
Area disturbed: 80 acres/year
Method of overburden removal: dragline
Quantity of overburden moved: 24.4 x 106 yardVyear
Method of dust control for haul roads: water/chemical
Annual TSP emissions: 1134 tons
Figure 35. Annual average modeled TSP concentrations
Green River/Hams Fork Basin, Mine 16.
63
-------
There are some limitations to the results presented in the
figures. Various problems have been associated with the IPP and
Wyoming emission factors, and with CDM. There are widely varying
terrain features that are associated with the mines in each area.
This is a particular problem in the Green River/Hams Fork Basin.
Neither the sketches nor the CDMQC model accurately represent the
effects of complex terrain. Also, no consideration has been
given to the cumulative effect of mines located in close proximity.
This aspect is discussed in the next subsection. Sources of
error in the concentration predictive process are discussed in
detail in Section 4.0.
3.2.2 Cumulative Impact from Multiple Mines in a Mining Region
The modeling analyses presented in Subsection 3.2.1 were
examples of modeling studies performed for mines in isolation
from other particulate sources. In actuality, mines are often
located in close proximity to other mines and ambient concentra-
tions are cumulative concentrations from more than one mine.
This is particularly true in the Powder River Basin. Envi-
ronmental Research & Technology (ERT) prepared a report for the
U.S. Department of Energy (ERT 1979) in which Powder River Basin
coal mines were modeled for three particle sizes and different
dust categories. Figure 36 shows existing and anticipated sur-
face coal mining operations in the Campbell County portion of the
Powder River Basin. Table 13 shows the anticipated production of
these mines used in the ERT study. Region VIII IPP emission
factors were used in the analyses (EPA 1978). These factors,
which are for TSP, were converted to 2.5 pm and 15 pm from com-
posite size distribution curves (PEDCo 1978). The ERT air quality
(ERTAQ) model was used, which is a Gaussian model similar to CDM.
The results are presented in a series of annual average
concentration isopleth maps. Figure 37 shows TSP concentrations
from all dust sources. Figure 38 shows TSP concentrations from
coal dust emissions only. Deposition was assumed for both model
runs. Inclusion of all dust sources results in TSP concentration
from the mines exceeding 50 pg/m3 without background added. The
coal dust isopleths (current PSD interpretation) show maximum
concentrations of 2.0 pg/m3.
Figure 39 shows concentrations of particles <3 pm. All dust
sources were included but no deposition function was used in the
model. At 3 pm particle size, the lack of a deposition function
had little impact on the results. The 3 pm isopleths show a
maximum impact of 5 pg/m3.
64
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
aero o* '.J' ''• «."*, p
ii* ^••A-:(->4'
T^ y .^ r"-*. I
Thunder'-4
r. /r-i
Figure 36. Existing and anticipated surface coal mining operations in
Campbell County.
65
-------
TABLE 13. SURFACE COAL MINES PROJECTED FOR THE CAMPBELL COUNTY PORTION
OF THE POWDER RIVER BASIN
Mine
AMAX (Belle Ayr)
AMAX (Eagle Butte)
Carter (North Rawhide)
Carter (Cabal! o)
Cordero
Kerr-McGee (Jacobs Ranch)
Kerr-McGee (East Gillette No. 16)
ARCO (Black Thunder)
ARCO (Coal Creek)
Wyodak
Consolidation Coal Co. (Pronghorn)
ARCO (Black Thunder Expansion)
Shell Oil (Buckskin)
Kerr-McGee (East Gillette)
Carter (North Rawhide Expansion)
Carter (South Rawhide)
Mobil Oil (Rojo Cabal! o)
Other mines being planned:
Gulf Oil (Wildcat)
Peabody (North Antelope, Rochelle)
Status
Active
Under const.
Active
Under const.
Active
Active
Under const.
Active
Under const.
Active
Application
Application
Application
Application
Application
Application
Application
Projected maximum
production, million
tons per year
25
20
12
12
24
16
4
10
18
5
5
20
'4
11
12
7
15
10
5
66
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
»9 \ . 1 -" > ' • ~" - - ' \. ' .
ffia; --i^^
^H?S
J- "*• • T v i
--- ^ — -_— j. - X^* t*^-" I
^nv^yiJ3t5&.T"
Figure 37. Annual average TSP concentrations (yg/m3), total dust sources
with deposition, no background added.
67
-------
Figure 38. Annual average TSP concentrations (yg/m3), coal
dust sources with deposition.
68
-------
I
I
I
I
I
I
I
Si*-^ r \ t T !•'"« =V •"- - "—vi
rf^^M ^ '•>",* - ^^
m.€fe&^-.-vs
^Jf-^y^^-^f' i
m$£*%*€*
r^^f'.--A^'.^
Figure 39. Annual average TSP concentrations (ug/m3), respirable
particulate <3 ym without deposition.
69
-------
3.3 OTHER DESCRIPTIVE MODELS
The model results presented in Subsection 3.2 were displayed
as a series of isopleth lines. Other forms of presentation are
possible which can be more graphic in displaying certain charac-
terizations.
3.3.1 Nomograph Presentations
Emissions versus Concentrations by Air Basin—
Two studies were performed by PEDCo Environmental (1981b;
1981c) which allow prediction of worst-case fenceline annual TSP
concentrations based on two data items likely to be available
during the preliminary planning stages of a mine. The procedure
was developed for use as a screening tool to determine if more
detailed air quality studies were justified. It can be applied
to surface coal mines where anticipated TSP emissions are 2500
tons/year or less.
The dispersion component was summarized in a nomograph which
relates annual TSP emissions to maximum fenceline concentrations.
Worst-case conditions were simulated with the CDM model through
the use of air basin specific meteorological data in STAR deck
form, and the worst-case annual mine configuration over the mine
life shown in Figure 40. Because the worst-case mine configura-
tion shown may never be realized over the mine life, overpredic-
tion may occur. However, this is commensurate with the proce-
dure 's intended use as a screening tool. The methodology is
described in detail in the referenced reports.
The results of the two PEDCo studies are summarized in
Figure 41. The figure indicates the worst case fenceline TSP
concentrations in the three basins of interest in this report.
The solid line indicates the concentration increase from the mine
(no provision for background concentration). The nomograph
indicates that dispersion conditions in the basin have ci strong
influence on the maximum fenceline concentration. The dashed
lines on Figure 41 indicate the ambient concentration with the
mine in operation (increased concentration plus background con-
centration). The great difference in background concentrations
(ranging from 15 [ig/m3 in the Powder River Basin to 32 (jg/m3 in
the San Juan Basin) further influence ambient concentrations
resulting from the same level of emissions.
Concentration Versus Distance by Mine Size—
Nomographs have been prepared (Radian Corporation 1979)
which relate particulate concentrations to distance from the
center of the mine by mine size. Powder River Basin mines of 2,
4, 12, and 20 million tons of coal produced per year were analyzed.
The mine configurations are not shown in the reference. Emission
factors were derived from the EPA Region VIII Interim Policy
70
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
1
I
I
I
I
I
PIT AREA
(1-YR PERIOD)
MINE
FACILITIES
WORST-CASE
WIND/STABILITY CLASS
DIRECTION
MINE
BOUNDARY
Figure 40. Assumed worst-case mine configuration.
71
-------
o>
d.
CO
80
70
60
50
40
UJ
O
UJ
(-5
30
20
o
I 10
x
SAN JUAN
X POWDER RIVER-
BASIN
GREEN RIVER/
HAMS FORK
POWDER RIVER
BASIN
SAN JUAN
GREEN RIVER/
HAMS FORK
CONCENTRATION INCREASE FROM MINE
AMBIENT CONCENTRATION (INCREASE
PLUS BACKGROUND CONCENTRATION)
I
I
500 1000 1500 2000
TSP EMISSIONS, tons/yr
2500
3000
Note: Procedure not applicable when emissions exceed 2500 tons/year.
Figure 41. Nomograph procedure for predicting worst-case
fenceline concentration, TSP.
72
-------
I
I
I
I
I
I
I
I
I
i
I
i
i
i
i
i
i
i
i
Paper (EPA 1978). An 85 percent control efficiency was assumed
for haul roads. Annual emissions for the four mine sizes were
914, 1823, 4888, and 6991 tons, respectively. Annual modeling
was performed with the CDM model modified with a source depletion
fallout function, and the RAM model with a fallout function was
used for 24-hour modeling.
The results of the study are summarized in Figure 42. The
figure indicates the distance from the mine center over which the
TSP Class II increment would be exceeded by mine production size.
Mine production size projected for the Powder River Basin ranges
from 4 to 25 million tons per year, averaging 12.3 million tons
per year. Data are included for the annual Class II increment
(19 ng/m3) and the 24-hour Class II increment (37 (jg/m3). Data
are also shown for two cases, i.e., with the inclusion of all
fugitive dust sources (same as remainder of this report), and
with the inclusion of only coal dust emissions (current inter-
pretation of existing PSD regulations). The annual increment
would be violated at a distance from the mine center of 0.5 km to
1.8 km, depending on mine size and inclusion of only coal dust
emissions or all fugitive dust sources. Violation of the 24-hour
increment would occur at much greater distances. In the case of
inclusion of coal dust emissions over both averaging periods, the
distance of violation would range from less than 1.0 km to about
3.0 km. With all fugitive emissions, violation distances from
the mine center would range from about 4.5 km to 16.0 km. The
data indicate that inclusion of all fugitive dust emissions at
larger mines would routinely violate the 24-hour Class II incre-
ment at considerable distances beyond the mine boundary. The
extent of violations of the annual increment and the short-term
increment with only coal dust emissions included would be depen-
dent on the location of the mine boundary. However, the data
were derived for an isolated mine. In the Wyoming portion of the
Powder River Basin, mines are almost never isolated and the Class
II increment would be a much greater restraint.
3.3.2 Profile Presentations
Concentration data can also be displayed in profile form.
An example of a mine shown in Subsection 3.2 is shown in profile
format in Figure 43. This presentation format is good for visu-
alizing the effect of distance from the source on concentrations,
as well as fenceline concentrations. Unfortunately, it is subject
to several distortions if used improperly. The results are
highly dependent on the cross-section used and the relationship
between the vertical and horizontal scales.
In examination of Figure 43, it is apparent that the highest
concentrations center around the pit area. Fenceline concentra-
tions range from 0 to 20 ug/m3. The profile and fenceline con-
centrations are specific for the year analyzed. They would both
change significantly during other periods of analysis.
73
-------
18
15
12
'ANNUAL CLASS II INCREMENT EXCEEDED
24-HOUR CLASS II INCREMENT EXCEEDED
ALL FUGITIVE
.• DUST EMISSIONS
__—-• COAL DUST EMISSIONS _
. «"•""" fll I FIIRTTTur
ALL FUGITIVE
DUST EMISSIONS
COAL DUST EMISSIONS
10
15
20
25
30
MINE SIZE, 10 tons coal per year
Figure 42. Distance from mine center over which TSP Class II
increment is exceeded in the Powder River Basin.
74
-------
I
I
I
I
I
I
I
I
I
i
i
i
i
i
i
i
i
i
i
NORTH/SOUTH CROSS SECTION
3 0
BOUNDARY pll
3 6
BOUNDARY
km
COAL PRODUCTION - 14 mm tons/yr
ANNUAL TSP EMISSIONS - 2126 tons/yr
EAST/WEST CROSS SECTION
6 3
BOUNDARY
6 km
PIT
BOUNDARY
Figure 43. Profile presentation of a Powder River Basin mine,
annual TSP concentrations.
75
-------
3.4 RELATIONSHIP OF MODELING RESULTS TO POSSIBLE AMBIENT
STANDARDS AND THE PSD PERMITTING PROCESS
The objectives of this study are to provide data on PM 10
and TSP concentrations around western surface coal mines suffi-
cient to assess their relationship to possible ambient standards,
and to assess the impact of the inclusion of fugitive dust and
PM 10 in the PSD permitting process.
Most existing modeling studies allow a characterizcition of
only annual average TSP concentrations. When mines are in isola-
tion from other particulate sources, the results suggest that
fenceline concentrations approach but would probably not violate
national or state ambient annual particulate standards. On an
annual basis, concentrations decrease to 1 vg/m3 or less gener-
ally within 4 miles or less of the mine boundary. However, even
with application of BACT as defined by Region VIII, the PSD Class
II annual increment of 19 ug/m3 would be violated in many cases
at the fenceline if all fugitive dust was included in the analysis.
A study performed for hypothetical mines (Radian 1979)
suggests that the 24-hour Class II increment (with inclusion of
all dust sources) is much more restrictive than the annual incre-
ment. Areas of exceedance of the short-term increment would be
as far as 10 miles from the mine center for a mine with an annual
production of 20,000,000 tons.
When mines are not isolated from other particulate sources,
existing model results suggest that both the existing TSP ambient
standards and the Class II increments (with inclusion of all dust
sources) are restraining factors at the fenceline even with
application of BACT.
A nomograph procedure (PEDCo 1981a) illustrates that ambient
concentrations resulting from a given level of emissions vary
considerably by air basin because of different dispersion condi-
tions and TSP background concentrations in western mining areas
ranging from 15 pg/m3 to 30 |jg/m3.
76
-------
I
I
I
I
I
I
I
I
I
I
f
I
I
I
I
I
I
I
I
SECTION 4.0
CHARACTERIZATION OF PM 10 AND TSP AIR QUALITY AROUND
WESTERN SURFACE COAL MINES USING NEW PREDICTIVE TOOLS
4.1 BACKGROUND
Two new tools are available to improve the state-of-the-art
for predicting particulate concentrations around surface coal
mines. The first tool is the new set of emission factors devel-
oped for EPA (PEDCo 1981a) and the second is the new Industrial
Source Complex (ISC) model (EPA 1979). No modeling has been
performed to date using the new EPA emission factors. However,
recent application of the ISC model to surface coal mines has
been made (TRC 1981).
This section describes the application of these new tools to
three hypothetical mines representative of operations in three
distinct coal basins. The three basins are the same as described
in Section 3.0 and are the Powder River, Green River/Hams Fork,
and San Juan Basins. Annual coal production for these mines was
assumed to be 25.0, 3.6, and 6.5 million tons. For each mine, a
detailed emission inventory was developed. These data were then
used as input to the ISC model to predict ambient concentration
levels around the mines.
4.2 CALCULATION OF EMISSIONS
4.2.1 Annual Emissions
The end result desired from the emission calculations was an
inventory of particulate emissions by mining operation and particle
size.
The calculation of these data involved four sequential
steps. First, a general checklist of mining operations expected
to generate particulate was developed as shown in Table 14.
Next, for each operation in the list, an activity parameter was
determined for each of the three hypothetical mines. Not all of
the operations were expected to be found at each mine. Although
the three mine scenarios were hypothetical, existing mine plans
for currently operating mines were used as examples so that the
resultant activity parameters would be physically realistic.
77
-------
TABLE 14. MINING OPERATIONS THAT GENERATE PARTICULATE EMISSIONS
Mining and Reclamation
Topsoil removal
Scraper travel - topsoil
Topsoil dump
Overburden drilling
Overburden blasting
Overburden removal
Haul truck travel - overburden
Overburden replacement
Overburden shaping - dozers
Coal drilling
Coal blasting
Coal loading
Haul truck travel - coal
Coal dump
Dozers - coal
Wind erosion from exposed areas
Light- and medium-duty vehicle travel
Road construction and maintenance - graders
Access road travel
Process and Transfer
Crushing, screening, conveying
Coal storage
Coal loadout
78
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Consequently, the resulting data are representative of real world
data but should not be construed to be accurate for any particular
coal mine. The activity parameters were developed in units
appropriate for application of the selected emission factors.
The data for each of the three hypothetical mines are shown in
Table 15. As seen in the table, the activity parameters for the
three mines are vastly different, reflecting the different mining
and operating conditions in the three basins.
The third step in the sequence was to develop emission
factors and control efficiencies to apply to the activity para-
meters. The EPA 1981 emission factors were the primary reference
for factors. These emission factors were supplemented with data
from the recent TRC study, (TRC 1981) IPP factors (EPA 1978) and
AP-42 (EPA 1980). Base emission factors are shown in Table 16.
The independent variable values associated with these factors are
shown in Table 17.
EPA Region VIII defined BACT practices and control efficien-
cies were also used. Controls assumed were for access roads
(paved, 99 percent control), haul roads (watered, 50 percent
control), coal storage (Wyoming only, enclosure, direct emission
factor), and various coal processing activities (enclosed, bag-
house, 90-99 percent control).
The activity parameters were multiplied by the calculated
emission factors (Table 18) and control efficiencies to yield
controlled emissions (Table 19). For sources for which EPA 1981
factors were available, emissions for TSP and 2.5 and 15 pm are
shown. These sources constituted about half of the total number
of sources and about three quarters of the total TSP emissions.
For sources for which TRC, IPP, or AP-42 emission factors were
used, only TSP emissions are presented in Table 19.
The fourth step was to compute the particle size distribu-
tion for each mine. It was decided that one particle size dis-
tribution should be used for the entire mine in each basin rather
than to attempt to define a different distribution for each
operation. This decision was made for three reasons:
1. Particle size distribution data were not available for all
sources.
2. Particle size analysis of measured data in several studies
suggest that the size distribution of the fugitive dust does
not vary drastically from operation to operation. Conse-
quently, one average distribution for the entire mine was
used.
3. Assuming a single distribution for each mine greatly sim-
plified the subsequent modeling analyses.
79
-------
TABLE 15. HYPOTHETICAL MINE ANNUAL ACTIVITY PARAMETERS
Mining operation
Topsoil removal
Scraper travel - topsoil
Topsoil dump
Overburden drilling
Overburden blasting
Overburden removal
Haul truck travel
overburden
Overburden replacement
Overburden scraping - dozers
Coal drilling
Coal blasting
Coal loading
Haul truck travel - coal
Coal dump
Dozers - coal
Wind erosion
Light- and medium-duty vehicles
Graders
Access road
Crushing, screening
conveying
Coal storage
Coal loadout
Activity parameters
Powder
River
Basin
2.17
1.16
2.17
4.42
312
3.75
4.35
3.75
2.36
9.10
208
25.0
1.27
25.0
2320
310
4.58
1201
0.25
25.0
b
25.0
Green River/
Hams Fork
Basin
1.49
0.70
1.49
2.35
118
1.63
a
a
0.70
4.88
128
3.6
0.11
3.6
4640
165
0.21
327
0.15
3.6
1.0
a
San Juan
River
Basin
8.84
3.90
8.84
11.79
325
4.60
a
a
3.03
59.73
500
6.5
0.35
6.5
15167
1096
0.88
62000
2.00
6.5
5.0
a
Units per
year
wl yd3
10* VMI
ID, yd
10 holes
blasts-
lol yd6
105 VMT
ioj yd3
10!I hours
10 holes
blasts
10* tons
10* VMT
10 tons
hours
acr.es
10b VMT
VMT
10* VMT
10 tons
acres
10 tons
a
b
Not applicable.
Silo storage.
80
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
CO
ry
o
r-
O
^£
U_
Z
0
1 — t
(/}
HH
LU
LU
^
CQ
.
to
rH
LU
_J
CQ
i
LU
C
E 0 0.
IL-i- l/>
•P r-
m u
. ro M-
CM s- o
v <*-
E
m
rH
V
a.
i/)
r-
C
O
•r—
4^
ro
S-
cu
o.
o
O)
c
•r—
C
• r™
^r
+•>
cu in
c i— ro r— ro roro i—
0 2 -O 0 r-
•0 T3 2
c
0
•P > >> -C -Q v^>^» ^ ^
& & & .a ^
f^~ r^ ^™ r™™ f~
X ^ ~Nt ^*
} -Q -Q -Q
-x
(.p f"i f^^ f^
CSJ O
T rH rH
S-
3
O
x:
^
m
o
0 0 0 0 rH
.
• • «
o o o o
LO
.
CVJ
^£
^.
.
iH
W U3
^O •
1 0
o <:
rH
X 0
CM in
in
10 CM
CM
3:
ro
•
rH
in
^\
m
1 CO
o
rH O
00 X » CM to
CO ro o O
O O O
rH ...
Q O O O
03
C
O)
•a
7) I-
C C i— 3
• r— tf
1— •»•
r— i — u
~ ^ Q
•> > S-
D o ai
ro •— -i- ro E >
> 0) S- i-
O > Q. T3 ^
E ro E
cu ^ o
3 S- U
cu^3c c cs- -^
s- p TJ 01 o> CU-PCUU
T3 T
r- S- r— S- I
•r- at T- 3 :
o a. o .a ^
w ro cn
i- o ro i—
Q. S- Q. HI CU O) £ S- 3
o o o > ;
> > i/i "o ro
1— to i— o o o 2:
CM
rH
O
d
^j
c
cu
cu
(J
ro
"a.
2
c
01
TJ
S-
^^
t.
cu
o
.
o
^^
.
rH
21
m
.
rH
in
0
.
rH
00
.
rH
SI
\
CM
.
rH
U)
[^
.
m
c
ai
•a
S-
^
J2
S_
01
o
1
in
S-
cu
N
O
Q
^-»
O) U)
r— ro
O r—
-^ ^
p-Q ^3
o
ro
O
•
O
to ro
d CM
< 2:
o in
m .
in rH
CM Q
CTl
CO
. rH
o s:
CM CO
CM rH •
tO rH
O CT) Q
O5 O)
C C
•r— 'r—
i— -P
i — in
•i- ro
•a .a
^_ ^_
ro ro
o o
0 O
c
0
+J
£
CTl
r-H
O
•
0
O"l
»
o
yr
"**s^
CT»
rH
rH
*
0
CM
.
rH
^*
\
ID
rH
.
rH
O)
C
•a
ro
o
i—
ro
o
CJ
•o
O)
r;
c
•r™
-p
c
o
u
81
-------
^^s
^
OJ
D
C
•r-
•p
c
o
u
1C
rH
LU
_J
00
r-
1/1
-P
C
^3
C
E _0 0.
•P 1—
m u
. 03 4-
CM 5- O
V *
S- 1
1— C 3 U
S O O 03
^> > > -P -P OJ -P
JD JD JD JD JD JD JD
O rH
*t 00
O O
o o'
0
oo
CM
«* CO
y
^v rH
CM IT)
CM O
OO O
m
o
CM
2:
^ O CM 3 00
CT)
^.
+J
3
•0 CD
1 C
E
3 C
•r- CU
-a cu
CO i-
£ U CU -P
"O in CD 3
TJ 03 03 O
C O - CD i- ~O
03 £- CD C O 03
in c -i- -P o
i s_ in •!- >, in i—
-p cu in .c cu
JZ "O CU in > r- r—
CD 03 U 3 C 03 03
•.- S- U S- O 0 0
—I C3 < O CJ O 0
•o
0)
o
S-
p
c
o
o
OJ
3
•P
U
OJ
CU
c
o
c
o
•t—
+J .
3 CU
.o c:
•r- T-
S- E=
in s-
•i- CU
•o :>
CU Q-
N
•r- S-
in cu
cu :t
r- C3
0 Q-
-P CU
S- -CI
03 4-J
O.
CU C3
-P
cu s:
c >
•i— ^x
J- r~
OJ
HI •
T3 m
O U)
-P ra
CU T3
CU
U3
3
cu
cu
m
c
o
•r—
-P
03
CT
CU
S-
C3
•P
U
c:
C3
•r—
n
(.O
•r—
CU
CU
in
CU Q-
JZ U->
r- \~
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
to
0
H-t
cr
LU
rv*
O
|»
CJ
z
o
1— I
OO
OO
1— )
^r
LU
Z
1— 1
LU
to
^3
to
LU
5!
^>
LU
1
CD
1— t
rx
^£
^>
t-
z
LU
Z
LU
a.
LU
o
t—l
rH
LU
_J
CO
s-
5 u_
c in
01 E
0) ro
S- X
C3
Oi
fy
S_
a*
•a
^
o
a.
in
-P
•r-
C
=
s_
o
•p
a.
si
u
in
ai
o
Ol
r—
-O
(0
•r—
ro
C
O
•^
•p
ro
S-
0)
a.
0
o>
c
•^
c
•r*
s:
CM 0
r-. oo
in
<*• oo
to oo
rH in
«t 00
to oo
rH ID
in
c
o
*? +J
^_>
c
Ol
+J
c
o
U -P
•P D)
1— — tfm.
•r- Q)
00 jB
in 3
r—
Ol
>
ro
S-
•p
S-
01
a.
&_
u
to
o o o
en oo o
•3- ao
^"
00
f>x
o to o
rH i^ in
**• oo
00 rH
in
CM
o to o
o f*^ in
m oo
i-H i-H
00
i-H
CM
-P -P
t- ^ >*-
in
0)
•a i—
0) O
+j j=
in
ro 0) >*-
•— S- O
J3 3
•p f
ro in +J
oi -i- a.
i- O Ol
< s: Q
< 3: Q
D)
c
•r™
+J
in
(0
r—
J3
c
Ol
T>
S-
3
o
t.
O>
>
0
o o
m oo
oo
rH tO
oo r^
CSJ
^J
f- »?
ai
u
c
ro
•p
in at
•r- S-
T3 3
a. in
o -i-
i. O
a 2:
•a s:
^—
ro
>
O
S
0)
s.
C Ol
Ol C
•a -r-
S- r—
3 ai
£. J-
01 -a
>
0
rti m
'V 'W
0 00
to o
CM
E
I O)
in
^~ O)
Ol C
Q) ••-
-c -a
3 ro
o
1*- I—
o
•p
O •!-
z to
2 -1
c
ai
•a
S-
3
S-
Ol
>
o
1
in
u
3
S.
-P
r—
3
ro
O 00
r^ CM
en en
to r>»
en en
to r^«
^ ^
-P
c
Ol
-p
c o>
0 S-
U 3
•P
i— *r—
•r- 0
oo s:
m s:
c
Ol
TJ
S-
3
XI
J-
Ol
>
o
1
in
S-
o>
N
O
a
o
CM
^^
f^
r^
<^
o
CM
CvJ
00
00
CM
O
in
CM
CM
CM
CM
14-
T3
Ol
•P
in
rO
f^
rO
Ol
i.
^^
<
O)
c
^J
in
ro
o
r—
rO
O
U
O 0
00 CD
rH CM
r^ ^^
•^ CM
rH CM
^- o
O CM
OO CM
-P
s« *-
in
Oi
'o
JZ
01 M-
S- O
3
•P ^
in -P
•r- O.
O 01
s: o
^. o
o
OO
rH
^>
^
rH
•*
O
OO
^
Ol
S-
3
•P
in
• r«
0
s.
s:
O)
c
•a
ro
o
r—
ro
o
0 rH
O OO
rH i-H
O 00
CO CD
^*
O 00
to o
CM
E
1 D)
in
r~ O)
o> c
0) T-
j= -a
3
N
o
o
TJ
Ol
C
C
o
u
83
-------
T3
01
+J
C
O
u
CO
S-
0)
•r—
P*
C
03
13
C
03
to
i-
0) -^
> t-
• r- O
01 LJ_
C in
0) E
0) 03
S- I
0
S-
Ol
>
•r-
a:
S-
O)
T3
5
O
Q.
in
•P
•P—
c
=5
J-
o
4J
a.
•r—
S-
u
in
o*
0
Ol
_a
03
S-
03
C
0
-p
03
t-
0)
D.
O
O5
C
•r-
C
•^.
^T
IT)
CM «*•
OOO 0000 O > D. \
i -P i ii i ^J= E E
S-
o
•p i. 1
C >, U £L O TJ
O +i 03 T3 -P -P C
•r- •!- in M- O) T3 Ol U -i-
+Jr—in I. T- > 03 S
U T- Ol UOI5 •!-!*- O)
03X2OICJ--I-+J t+J i_ Ol
j_ ip^ (j r" Q ^_> t— "O O 03 t- 3 O3 "O
<^- *X3 (^ rj) ^ fQ 0 ^~ ^j +j ^j 40 ~rj {Q Q)
^— o *^ 3 U E ^" 0)OOI^ I/) Ol ^OJ
^L *r- C. &_ O 03 *r~ in •!-- 03 O5 O *r- Ol Ol Q.
I/) o Ol 3 S- 'f— ' — C 4— *+— 0> U O Q. >in
r-^tO OZ3 > 2! tO<
_ _
03 1— 1 ^ «_3 — 1 > 2T 1/03
•a
i
E
3
• r"
^3J
O)
C E 01
0 D)
•i- T3 03
W C i-
O 03 Ol O
S. r- W 4J
01 i U i- in
4J .r- 0)
T3 ^1 -C TJ i—
C O) 01 03 03
•r- -r- > f- O
-3. _i o o
OJ
03
u
a.
a.
03
•p
o
03
84
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
to
LU
Z
s
_i
<
o
h— 1
1
LU
X
0
>-
oe.
0
u_
l/>
Q£
O
<
u_
z
t— (
to
CO
t—4
5-
LJJ
1
o
O£
<
Q.
s
LU
_l
GO
<
1—
s~
o>
>
ac:
c
s-
•r- O
Q£ U_
c in
OJ fS
01 (0
S- X
CD
S-
Ol
•^
Q£
S-
01
"O
O
0.
l/>
p
•^-
c
0
to
(—
LO
rH
V
m
CM
V
a.
h-
LO
t-H
V
m
eg
V
a.
h-
LO
rH
V
m
CM
V
Mining operation
i.
-P -P >>
0) in 0) i
PO |_,>>,j:ja>>>>)X:i:^2-P>-P-c^x >>> 4J\-P
oor>>o OCT>
OCDOrH^fO mOLOO^l-OCOO OOlDO OtH
If) rH 00 rH M CSJ
rH
<* Csl
r-l rH
rH OOO «S- r-l O r-» O1 r-ir^
csj «*o <• o o od CM mcsj
csi m rH
0 O
iH rH
co no cMino'd-H moo
iH COO COinOCSJOO COM
o r-io r-iinooo oo
00
r^OLn coco CMCOOO
OOPO ^J- CM^-COOO rHOCM
ocsiocorHO p»csjroopoocoro cni^-o or^o
OCOOr-lOO COOr-IOOOOrHO CSJLOO OCSJO
r-l r-l 1^ CO rH
CSJ r-l
rH 00 rH
o ooo OOOOOI^OT oor^
1X3 COO OrHOCOO r-HCM
CM rH
00 CT>
0 0
•3- OO O^tO^-O CMOO
CO COO rH rH O I— 1 r^ rH rH,
d do dcsiddd do
J3
O r>-rH CT1CO CSJinOD
OOCO COCMCMrHCOOO rHOCSJ
ocsjococoovoorxCMOoocoocsjco cfir-»o o o
ocoorHr^oinoroocnoinocMO csjtno o o
rH rH r-H
CO
00 O
o L0«j-oo*ro)
T3 C +J
i. 0) 3
O) Oi 3 E C -D CT
C. C. i— .O 0) 0) 1 C
•r- -r- (O 5- (J T3 r— E •!-
r— +J>OlflJt- r— 3 O •!- O)
for— T-(OEoo.ja u "a OI 1- r— Ol O) S- D) CT r— 0) S-
O > Q.T3 J3 1- 1 S- 0) C C CO 1 (OCE U OI-P
SrtJE >T--r-C OO T3WCTCn3
O)5-3CCCJ^COi— 4-)'i-J^ U'i-"DW (0 CITJO
t.+j-rja)aiaiuai r— UITJOQ. wcoi o «v-s.-a
T3TJ-03-0|T-(0(03Sl OiTJr- i-O)>,O(TJ
r-1-r—t.i.t.s.t. i-r— o i- 3 t- um COI-PO
•r-O»T-D33+J3U)T3^3r— P "O in O» 1 T-J-U)T->V)i—
00.0x2^3^3 -Q t- i- 4->j=a>u>^:c
a.s-o.cua)ai3aiN(0(an}3(ONCo)>in(j3u(cn}
ooo>>>m>oooo(ooo-r--r- j-us- oo
i— cot- oooxoooc_)oxc_>a3:-J C3
O. Q.
Q.
tO r-
(0
•P P
O O
to .a
85
-------
z
o
1— 1
LO
in
i — i
2:
LU
Q
LU
i
p—
_l
—^
O
1
o
,
0s*
rH
LU
1
CO
I—
TO
0)
1
-P
*«
i/)
C
O
•r"
1/5
E
0)
^—
TO
3
C
C.
^£
£-
0)
•r—
C
TO
3
^
C
TO
LO
^
J-
S-
•r- O
O£. LL.
C U)
0) E
CD
S-
o>
.^
^
s-
0)
TJ
5
o
Q.
o.
1—
in
rH
V
m
•
CM
V
r\
LO
m
rH
V
in
.
CM
V
O_
LO
1—
in
rH
V
in
CM
V
c
o
+3
TO
S-
0)
Q.
O
0s
C
C
y
in o oo r- co o cMLOooLOOocMOt-Hr-o«d-^-
oo r-» oo r"~ co oo ^•LOLOoooors--oooLQLDCMCMO'>
OO rH i — 1 O) OO *Et* F** LD rH 00 LD 00 F^ rH rH O
rH CM CM rH «*• rH CM O OO rH iH
rH CM rH
O
00 OOO LO LO«sl-. rH LOf>.
. . . . . rH . . .
O «*rH LD CMCOr-l 00 rHOO
^" rH LD rH OO *!f CH CM 00
00 rH CM
O OOO LD ^tOOCTI CM **m
• •• • ••• • ..
00 OCM «3- rHOOrH LO LOin
CM CM «^- rH
t f"i L^ fyj rH f™ ^ in CM f^ ^^ CM CM OO OO CM *~^ f^ f*™^ tf^ c\j 0s*
^^* tf^ rH QO ^^ to f*^ t o fv} rH ^^
OO <1-
O"> ^" in rH ^" O*i in ^^ CM ^^
• •• » •>• . ••
O Of--- OO r-HOOLD LO O1O
CM CTl rH O CM rH
CM
CM rHCM »* rHLDO^ LD CMrH
t •• • **. , ..
r-l OLD r-l OrHt^- rH r-HO
V V
JD
r^ ^o ^^ CT^ ^^J oo ^j* to r^^ f~^ I~H ^^ ^\i t*~^ ^^ QQ p^ <^ L/^ r^* f~ *^
CT^ ^^ OO ^^J ^^ CO ^^ ^H CO r^ ^^ f^ C^ ^^ <^ f^ ^f* PO ^^ 1^^ C3
r-- OlrHCn«* ^j-LOLOrHrHLO LO LO
LDCMoo CMOI*J-T-HLD oo
rH OO
Is-. en •* «s- moovD LD CSILO
^t O 00 O OOOin <* CMrH
OO CM rH LD rH rH
m r-- ^-
rH
O rH P-- LD rHLOOO OO LDrH
• • • • .*. . ..
CM O O «* O<3-f-- O LOO
V CM LO CM
in
0) D5
r— C
(J T-
r* QJ
QJ >
> C
c o
>, U
T3 C -P
i- Q) 3
CT O5 3 E C TJ O)
C C r- J3 0) 0) 1 C
•r- •<- <0 S- L> -0 <— E -r-
i— -P > 0) TO S- TO 3 C
r— r— in O > i — 3 O -i- ttl
TO<— •I-TOEOQ.J3 U T3 Q)
>OI J-r— 0) OIS-O1D) i — OJ S-
O > Q.T3 ^3S-l5-QJCCCni TOCE UCO-P
ETOE >•<-•<- C. OO •OinO)3
OJS-3CCC^COr-+J-r-^ U-i-TJ TO TOO
s--pt3Q)a)0)(jaj i — m'ouo- me o -t-'a
TJ'O'as-oiT-rorosEi OTO s-cnoro
r— S-i— S-S-S-t-S- J. r- O t- 3 1- in C+JO
ODLOX3^3^ X15- "~ S- -POIifl-C
Q.S-Q.O)>>TO>OOOOTOOO'r--r-S-US-OO
I— LOh- OOOXOOOUOXOOjE— ICJeCOOO
U
TO
C
O
r- TO
JD S-
(0 TO
U Q.
•i- O)
i— S-
o- a.
o.
TO r—
TO
•P -P
O O
Z. r-
TO
86
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
For the purposes of this analysis it was further assumed
that the TSP emissions could be reasonably approximated as par-
ticles <30 pm in aerodynamic size. This was a basic assumption
used to develop some of the EPA 1981 emission factors. Hence, it
was also used here.
The first step in determining the distribution was to sum
the <2.5, <15, and TSP emissions for those sources with a particle
size breakdown. Next, the <2.5, and <15 emissions were expressed
as a fraction of the TSP emissions. Based on these fractions it
was possible to interpolate the fractions for <5 pm and <10 pm,
and to extrapolate the distribution to <20 pm by assuming a
lognormal distribution.
The mass distribution determined in this manner was then
applied to the total TSP emissions from the mine based on the
lognormal relationship. Table 20 presents the final distribution
of emissions by particle size determined for each mine.
TABLE 20. DISTRIBUTION OF EMISSIONS BY PARTICLE SIZE
Cumulative emissions less than stated size, tons/year
Hypothetical mine
Powder River
Green River/Hams Fork
San Juan River
2.5 pm
166
26
157
5 pm
797
105
676
10 pm
2,383
301
1,896
15 pm
3,711
469
2,899
20 pm
4,708
610
3,653
30 pm
8,303
1,286
6,559
These distributions were used as input to the ISCST and ISCLT
models for predicting the annual and 24-hour mean concentrations.
In further examination of Table 20, the emissions for the
San Juan Basin mine 6.5 million appear high when compared to the
Powder River Basin mine (25 million tons/year of coal produced).
This is a direct result, however, of the activity parameters
assumed in the scenarios (Table 4-2). Because of the lower coal
to overburden ratio in the San Juan Basin, it is necessary to
disturb significantly more acres of land and cubic yards of
overburden per unit coal mine.
The approach described above represents one possible way to
define the particle size distribution of the emissions. Certainly
other methods could be used to produce a distribution composed of
more coarse particles or more fine particles. The method selected
was based on the methodology used to develop the EPA 1981 emis-
sion factors. The emissions defined by these factors represent
about 75 percent of the total mine emissions. As these factors
were developed from field sampling for the particular particle
87
-------
sizes described (2.5, 15, TSP), using some other distribution
based on another type of analysis did not seem to be justified.
In critical evaluation of the particle size distributions
shown in Table 20, these distributions are compared to particle
size distributions found by other researchers in Figure 44. Only
a few such distributions are presented, others are available.
All of the distributions shown in the figure represent the por-
tion of the material less than 30 (jm. This change in the original
data was made so that the distributions could be compared directly.
A further simplification was to assume a lognormal distribution
of the data.
The two upper curves represent the results of microscopic
analysis of open faced filters exposed near haul roads (EPA 1978;
TRC 1981). The particle sizes represented by these two curves
are observed physical diameters. There are four basic limita-
tions to the microscopy data that prevent their application to
measured TSP concentrations.
I. The filters are exposed open faced. Consequently larger
particles are collected than would be collected with a
hi-vol, thereby, biasing the distribution towards coarser
particles.
2. The distributions are based on physical rather than equiva-
lent aerodynamic diameters. This limitation conceivably
creates a bias towards the finer particles.
3. The distributions include the small sizes of particulate
present in the background concentrations, adding an addi-
tional bias towards the fine particles.
4. Microscopy cannot accurately identify fine particulate (less
than about 2 urn in physical diameter) because of the lack of
optical resolution and because some of the fine particulate
may be located out of view behind the larger particulate.
Therefore, the distribution will be biased towards the
larger, easier to perceive, particle sizes.
It is not known to what extent these biases affect the overall
distribution.
The lower curve is a composite of ambient data obtciined with
a cascade impactor within 5 m of a haul road. The data represent
the best attempt to correct measured data to account for one of
the serious limitations of this instrument, particle bounce
through. Additionally, no consideration is usually given to the
density of the collected particulate or to the portion of the
distribution attributable to background. The net effect of these
three problems is to bias the resulting distribution towards the
88
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
3.
ft
UJ
Q.
30
20
10
9
8
7
6
5
4
1
I I I I I I I
DATA SOURCE
'••• MICROSCOPY (EPA 1978)
MICROSCOPY (TRC 1981)
STABLE 20 COMPOSITE
—•CASCADE IMPACTOR (EPA 1981 a)
— DICHOTOMOUS (EPA 1981a)
I \ \ I
I i II
I I I I I
0.01 0.1 0.5 1 5 10 50
CUMULATIVE FREQUENCY OF OCCURRENCE, PROBABILITY, %
80
Figure 44. Examples of composite particle size distributions
from coal mining particulate sources.
89
-------
fine particle sizes, drastically limiting the usefulness of data
collected with this instrument.
The middle curve represents ambient data collected with a
dichotomous sampler near haul roads. The virtual impaction
principle used in this instrument theoretically gives concentra-
tions for particulate with aerodynamic diameters <2.5 and <15 urn.
However, there are also several problems associated with this
instrument. The inlet design of this device is sensitive to the
ambient windspeed requiring questionable corrections to obtain
the <15 |jm concentration. Also, the mass of <2.5 pm particulate
routinely collected during field sampling is very near the detec-
tion limit of the gravimetric analysis method. Additional prob-
lems are related to the filter media and handling procedures used
in fugitive dust testing. One identifiable effect of these prob-
lems noted in a recent study (EPA 1981) was that the <2.5 pm
concentrations are artifically high by about 10 percent. This
has the influence of biasing the 10 (jm data towards a finer
particle distribution by about 1 percent.
In addition to the sampling biases outlined above there is
one underlying assumption that is critical to the analysis, the
assumption of the lognormal distribution. This assumption is
especially appealing, as it greatly simplifies many aspects of
particle size analyses. The fugitive dust generated by a single
mechanism can be approximated by the lognormal distribution,
especially for coarse (>2.5 jjm) particulate (EPA 1981). However,
additional research needs to be performed in order to justify or
invalidate the use of this assumption.
The sensitivity of coal mine modeling results to size dis-
tribution has not been critically analyzed. It is an area where
additional research is needed. Theoretically, assumption of a
distribution composed of more coarse particles than the true
distribution would result in lower initial emissions for the
smaller particle sizes and a greater amount of deposition over
distance. The effect of this distribution would be to underesti-
mate downwind concentrations. Conversely, assumption of a finer
distribution would be associated with greater emissions for the
smaller sizes, less particle deposition, and larger downwind
concentrations. These concepts are displayed in Figure 45 for PM
10 or TSP. Additional research is required in order to define
these curves.
4.2.2 Twenty-Four-Hour Emissions
In addition to the annual emissions estimates, 24-hour emis-
sions were required for the modeling. Unfortunately, there is no
standardized approach for determining short-term emission levels.
This has led to a wide variety of practices that further compromise
the accuracy of 24-hour modeling. A simple procedure was developed
90
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
o
z
o
D-
1/1
o
o
FINER DISTRIBUTION
CORRECT DISTRIBUTION
COARSER DISTRIBUTION
Figure 45.
DISTANCE
Theoretical impact of incorrect particle size distribution
on model-predicted PM 10 or TSP concentrations.
91
-------
for this study to convert annual emissions to 24-hour emissions.
For all sources except blasting and wind erosion, the annual
emissions were divided by 365 to obtain the daily average. Then
the daily averages were increased by 25 percent to represent peak
24-hour emissions to represent worst-case activity parameters.
Other methods could have been used. In general, the procedure
should be standardized to facilitate a uniformity of short-term
modeling assumptions.
For blasting it was assumed that the most blasts that would
occur in a given day were: 2 coal and 2 overburden (Powder
River); 1 coal and 1 overburden (Green River/Hams Fork); 3 coal
and 3 overburden (San Juan River). The fraction of the annual
emissions for the 24-hour period was then calculated based on the
total number of blasts of each type for the year.
Wind erosion emissions for 24-hours were based on the mean
number of days with <0.01 inch of precipitation. These were 273,
275, and 285 for the three mines. Annual wind erosion emissions
were simply divided by the number specific to scenario location.
The same particle size distribution developed for the annual
modeling was used with the 24-hour emissions.
4.3 CALCULATION OF CONCENTRATIONS
The ISC model was used to predict both PM 10 and TSP par-
ticulate concentrations, for annual average and 24-hour time
periods at each of the previously discussed hypothetical mines.
The goal of the simulation was to model each mine as if an air
quality permit application were being prepared. All assumptions
and idealizations used in the modeling effort are those routinely
used by permit applicants, air quality consultants, and reviewing
agency personnel.
4.3.1 Fundamentals of the ISC Model
The ISC dispersion model can be used to perform air quality
impact analyses for a wide variety of facilities, including
surface coal mines. It should be noted, however, that certain
features of coal surface mines, such as pit retention and wet
deposition of particles, are not treated and that dry deposition
and particle reentrainment are handled somewhat simplistically.
The ISC model combines various analytical dispersion modeling
algorithms in two computer programs, a short-term version and
long-term version. The ISC model short-term program (ISCST) is
an updated version of the EPA's CRSTER model. The ISC model
long-term program (ISCLT), a sector-averaged model, utilizes some
of the features of two EPA UNAMAP models: the Climatological
Dispersion Model (CDM) and the Air Quality Display Model (AQDM).
Both programs (ISCST and ISCLT) of this comprehensive model offer
92
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
some very useful features that are especially attractive for use
in simulating surface mines. The feature options exercised for
this modeling task include:
0 Emissions from a combination of area and volume sources
with physical separation of the multiple sources.
0 Application-defined receptor grid.
0 Effects of gravitational settling and dry deposition.
0 Dispersion coefficients and mixing depths for a rural
environment.
0 Simulation of site-specific atmospheric conditions.
The same basic dispersion model assumptions apply to both pro-
grams (ISCST and ISCLT). The steady-state Gaussian plume equa-
tion for a continuous source is used to calculate groundlevel
particulate concentrations for point, area, and volume, sources.
The area source model is based on the dispersion equation for a
continuous and finite cross-wind line source. This extremely
flexible model calculates particulate concentrations using se-
quential hourly meteorological data for the short-term (24-hour)
and a joint frequency distribution of windspeed, wind direction,
and Pasquill stability class for the long-term (annual) values.
The ISC model offers some very real advantages over other
widely used dispersion models for simulating the air quality
impact of surface coal mines. Most important, the ISC model
includes a treatment of particle settling and deposition which is
essential if concentrations of large particulate (including TSP)
are to be accurately predicted. Secondly, the ISCLT and ISCST
programs permit a great deal of latitude in how the modeler
idealizes sources such as haul roads, coal handling facilities,
mine pits, etc., so that a surface coal mine can be approximated
more closely than with most other models. The ability to employ
volume sources to simulate haul roads, for example, permits a
better representation of the road than do the line source or
sequential point source methods of other models. Finally, the
ISC's detailed meteorological input ensures that the convective
and diffusive parameters governing particulate transport will be
properly described. A technical description of each of the
models features is contained in Appendix C.
4.3.2 Application of the ISC Model
Source Data—
Calculation of emissions was described in Subsection 4.2.
Appendix C contains source apportionment data for the three
location scenarios (annual and 24-hour), and source characteris-
tics (number of sources, source type, source size, a , and a ).
93
-------
Receptor Data—
Model grid receptors used in the annual average ISCLT simu-
lations were placed in a uniform Cartesian coordinate system,
equally spaced at intervals of 0.5 kilometers over a range of 40
kilometers in both the N-S and E-W directions. The hypothetical
mine sources were situated near the center of the grid array.
Model receptors used in the 24-hour ISCST model runs were
located on Cartesian coordinates nodes confined to a 90 degree
wide sector in the predominant downwind direction of the mine.
Receptor spacing ranged from 0.5 to 2.0 kilometers in distance,
with additional random receptors situated near major mine sources
to better define peak concentrations. The range of the receptor
array extended to roughly 30 kilometers in the downwind direction.
Meteorological Data--
The meteorological data input to the ISCLT models included
joint frequency of occurrence of windspeed, wind direction, and
stability class (STAR deck) collected at surface stations in the
vicinity of the hypothetical mines:
0 Craig, Colorado STAR data used to model Green River/Hams
Fork Basin mine.
° Farmington, New Mexico STAR data used to model San Juan
Basin mine.
0 Moorcroft, Wyoming STAR used to model Powder River
Basin mine.
Annual average ambient temperatures were set to 281 degrees
Kelvin for all three mines, and annual mixing heights were input
as 5,000 meters for stability classes A-D, and 10,000 meters for
stability classes E and F. Values of vertical potential tempera-
ture gradient and wind profile power law exponents were set to
the ISCLT default values.
For the 24-hour ISCST model runs, the combination of wind-
speed, wind direction, and stability class was chosen so as to
yield maximum groundlevel concentrations that could be reasonably
expected to occur twice in a one-year period. Concentrations
predicted in this manner represent second-highest values, directly
comparable to existing ambient air standards. The windspeeds and
stability classes were set equal to those identified by D. B.
Cabe as prompting peak 24-hour particulate concentrations down-
wind of surface coal mines, based on previous studies using the
RAM model (Radian 1979). Wind directions were centered about a
mean coinciding with the major axis of each of the hypothetical
mines' sources so that as many sources as possible were in line
with the wind direction. Individual hourly wind directions were
then randomized within a 22.5 degree sector about this mean wind
94
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
direction. The windspeeds, wind directions, and stability classes
modeled at each of the three hypothetical mines are shown in
Appendix C. Like the long-term model runs, air temperatures were
set to 281 degrees Kelvin and mixing heights were assumed equal
to 5,000 meters.
Background Concentrations—
The background concentrations used in the analyses are the
same as were used elsewhere in this report and are shown in Table
21.
TABLE 21. ASSUMED BACKGROUND CONCENTRATIONS
USED IN THE SCENARIO ANALYSIS
Bas i n
Powder River Basin
San Juan Basin
Green River/Hams
Fork Basin
Background
concentration, (jg/m3
TSP
15
32
22
PM 10
9
9
9
Two items are of note. First, the same background concen-
trations were used for the annual and 24-hour averaging periods.
This undoubtedly is an underestimate of the 24-hour level but no
data are available to determine a short-term background concen-
tration. Secondly, for the reasons noted in Subsection 2.3.2,
the same background concentration for PM 10 had to be used in all
basins.
Pit Retention—
For all of the modeling runs, no terrain adjustments were
made to receptors or sources. Emissions from the mine pits were
assumed to occur at groundlevel, and therefore, no correction was
made for retention of particulate matter within the pit. These
procedures are common modeling practice.
Lack of adjustment for pit retention is contrary to accepted
intuitive knowledge about surface coal mine particulate emission
and dispersion characteristics. However, the technical ability
to accurately simulate the effect of pit retention is not avail-
able. For example, while pit retention is known to occur, pre-
liminary estimates of pit retention range from about 25 to 90
percent under almost identical conditions. Additional research
is required in order to gain the technical ability to account for
pit retention.
95
-------
Conversion of Arithmetic Concentrations to Geometric Concentra-
tions—
The long-term computer models calculate arithmetic concen-
trations which, for comparison with annual geometric mean stan-
dards, must be corrected. The commonly accepted means of con-
verting long-term modeled concentrations to geometric means is to
multiply the modeled arithmetic concentrations by a factor equal
to the ratio of measured annual average geometric concentration
over measured annual average arithmetic concentration. Of course,
the value of this factor differs according to geographic location.
A collection of characteristic factor values gathered from mea-
sured hi-vol networks in the West are shown in Table 22 below:
TABLE 22. ANNUAL AVERAGE GEOMETRIC/ARITHMETIC TSP
CONCENTRATIONS MEASURED BY THE HI-VOL METHOD
Location
Geom/arith
factor
Powder River Basin (Gillette, Wyo.)
Powder River Basin (Colstrip, Mont.)
Little America, Wyoming
Sheridan, Wyoming
0.75
0.80
0.86
0.79
The factor values in Table 22 do not differ greatly. In
order to predict geometric mean concentrations that are conserva-
tively high, it was decided to use the factor corresponding to
the Little America, Wyoming area (0.86). All annual average
arithmetic concentrations, both TSP and PM 10, were multiplied by
0.86 to obtain annual geometric means displayed in Figures 46
through 57.
4.3.3 Results
The results of the long- and short-term modeling are dis-
played in Figures 46 through 57, numbered as follows:
Figure 46:
Figure 47:
Figure 48:
Figure 49:
Figure 50:
Figure 51:
Figure 52:
Figure 53:
Powder River Basin, Annual TSP
Powder River Basin, Annual PM 10
Powder River Basin, 24-Hour TSP
Powder River Basin, 24-Hour PM 10
San Juan Basin, Annual TSP
San Juan Basin, Annual PM 10
San Juan Basin, 24-Hour TSP
San Juan Basin, 24-Hour PM 10
96
-------
1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1 •
-------
1 yg/m3 (10 yg/m3)
5 yg/m3 (14 yg/m3)
10 yg/m3 (19 yg/m3)
20 yg/m3 (29 yg/m3)
yg/m3 (49 yg/m3)
f J HIKE BOUNDARIES
£j PIT LOCATION
HAUL ROAD
A PLANT LOCATION
NOTE: NUMBER IN PARENTHESES INDICATES PREDICTED
CONCENTRATION PLUS BACKGROUND CONCENTRATION
Location: Powder River Basin
Coal production: 25 x 106 tons/year
Area disturbed: 130 acres/year
Method of overburden removal: shovel/truck
Quantity of overburden removed: 37.5 x 106 yardVyear
Method of dust control: water
TSP emissions: 8303 tons/year
Figure 47. ISC modeled annual geometric mean PM 10
concentrations, Powder River Basin.
98
-------
,5 ug/m* (16 pg/n»3)
1 •
-------
1 •
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1 yg/m3 (33 yg/m3)
5 yg/m3 (37 yg/m3)
10 yg/m3 (42 yg/m3)
20 yg/m3 (52 yg/m3)
NINE BOUNDARIES
f IT LOCATICH
HAUL ROAD
A PLANT LOCATICK
I •«!•
NOTE: NUMBER IN PARENTHESES INDICATES PREDICTED
CONCENTRATION PLUS BACKGROUND CONCENTRATION
Location: San Juan Basin
Coal production: 6.5 x 106 tons/year
Area disturbed: 548 acres/year
Method of overburden removal: dragline
Quantity of overburden removed: 46 x 106 yardVyear
Method of dust control: water
TSP emissions: 6559 tons/year
Figure 50. ISC modeled annual geometric mean TSP
concentrations, San Juan Basin.
101
-------
1 yg/m3
^5 yg/m3
/lO yg/m3
(10 yg/m3)
(14 yg/m3)
(19 yg/m3)
[Tj MINE BOUNDARIES
£j «T LOCATIOR
—• HAUL ROW
A PLANT LOCATION
NOTE: NUMBER IN PARENTHESES INDICATES PREDICTED
CONCENTRATION PLUS BACKGROUND CONCENTRATION
Location: San Juan Basin
Coal production: 6.5 x. 106 tons/year
Area disturbed: 548 acres/year
Method of overburden removal: dragline
Quantity of overburden removed: 46 \ 106 yardVyear
Method of dust control: water
TSP emissions: 6559 tons/year
Figure 51. ISC modeled annual geometric mean PM 10
concentrations, San Juan Basin.
102
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
- ug/m3
10 ug/m3
20 ug/m3
40 pg/m3
60 Lig/m3
80 Lig/m3
(33 Lig/m3)
(37 Lig/m3)
(52 Lig/m3)
(72 Lig/m3)
(92 ug/m3)
(102
1 •
-------
ug/m» (23 ug/m3)
5 ug/m3 (27 ug/m3)
10 ug/m3 (32 ug/m3)
20 ug/m3 (42 ug/m3)
NINE BOUNDARIES
_J PH LOCATION
— HAUL ROAD
A PLANT LOCATION
NOTE: NUMBER IN PARENTHESES INDICATES PREDICTED
CONCENTRATION PLUS BACKGROUND CONCENTRATION
Location: Green River/Hams Fork Basin
Coal production: 3.55 x 106 tons/year
Area disturbed: 70 acres/year
Method of overburden removal: dragline
Quantity of overburden removed: 16.3 x 106 yard3/year
Method of dust control: water
TSP emissions: 1286 tons/year
Figure 53. ISC modeled annual geometric mean TSP
concentrations, Green River/Hams Fork Basin.
104
-------
I
I
I
I
I
I
I
I
I
I
I
I
1
I
I
I
I
I
I
Figure 54: Green River/Hams Fork Basin, Annual TSP
Figure 55: Green River/Hams Fork Basin, Annual PM 10
Figure 56: Green River/Hams Fork Basin, 24-Hour TSP
Figure 57: Green River/Hams Fork Basin, 24-Hour PM 10
Table 23 indicates maximum concentration versus distance
from the mine boundary for the annual and 24-hour time periods.
Table 24 relates violations of the NAAQS and Class II PSD incre-
ments versus distance from the mine boundary.
Total suspended particulate concentrations at the mine
boundary for the three scenarios all exceed the NAAQS and or PSD
Class II increments. The Powder River Basin scenario violates
both the annual and 24-hour NAAQS, as well as consumes all PSD
Class II annual and 24-hour increment. The San Juan and Green
River/Hams Fork Basin scenarios do not violate the primary or
secondary NAAQS, but do consume all of the annual and 24-hour
increment.
PM 10 concentrations were compared only to PSD increments.
At the mine boundary, the Powder River Basin scenario consumes
all of the annual and 24-hour increment, while the Green River/
Hams Fork Basin scenario consumes all the annual increment.
If surface mines were required to secure PSD permits, then
the TSP concentrations compated in this modeling study at all of
the scenario mines would exceed PSD increments. Furthermore, the
modeling results suggest that PM 10 concentrations alone would
exceed PSD increments at the Powder River Basin mine and at the
Green River/Hams Fork Basin mine.
Throughout this discussion, a key factor in judging whether
standards would be violated is the distance between the major
mine particulate matter sources and the mine boundary. Because
concentrations of TSP, and to a lesser extent PM 10, decrease
dramatically with downwind distance, the proximity of pits and
haul roads to the mine boundary may dictate whether a violation
of standards is predicted off site. It would be misleading to
assume that the problem of small separation distances between
pits and mine boundaries may never occur, or would occur only in
a small fraction of proposed surface coal mines, because just the
opposite is true. Mineral recovery laws require that all of the
recoverable coal within a permit boundary be mined, and to do
this pits and haul roads must be situated adjacent to mine bound-
aries at some point during the life of the mine. The results of
the modeling work presented in this report suggest that mine
source and boundary configuration may be just as important a
determinant in predicting whether a mine will meet standards as
other factors such as annual production rate.
105
-------
5 pg/m3 (14 M9/«n3)
10 ng/m3 (19 pg/m3)
20 pg/m3 (29 pg/m3)
30 pg/m3 (39
I •
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
10 pg/m3 (19 pg/m3)
20 pg/m3 (29 pg/m3)
NINE lOUNCMRfES
PIT LOCATION
w^
— HAUL KOAO
A nxcr LOCATION
NOTE: NUMBER IN PARENTHESES INDICATES PREDICTED
CONCENTRATION PLUS BACKGROUND CONCENTRATION
Location: Green River/Hams Fork Basin
Coal production: 3.55 x 106 tons/year
Area disturbed: 70 acres/year
Method of overburden removal: dragline
Quantity of overburden removed: 16.3 x 106 yardVyear
Method of dust control: water
TSP emissions: 1286 tons/year
Figure 55. ISC modeled annual geometric mean PM 10
concentrations, Powder River Basin.
107
-------
5 pg/m3 (27 pg/m3)
10 pg/m3 (32 pg/m3)
20 pg/m3 (42 pg/m3)
60 pg/ra3 (62 pg/m3)
1 Bill
NOTE: NUMBER IN PARENTHESES INDICATES PREDICTED
CONCENTRATION PLUS BACKGROUND CONCENTRATION
Location: Green River/Hams Fork Basin
Coal production: 1.2 x 104 tons/day
Area disturbed: 165 acres
Method of overburden removal: dragline
Quantity of overburden removed: 0.56 x 10s yardVday
Method of dust control: water
TSP emissions: 373 Ibs/h
Figure 56. ISC modeled 24-hour TSP concentrations,
Green River/Hams Fork Basin.
HIKE IOUNOAHIES
PIT LOCATION
HAUL ROAD
PLANT LOCATION
108
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
5 ug/m3 (14 ug/m3)
20 ug/m3 (29 M9/™3)
I •
-------
TABLE 23. MAXIMUM CONCENTRATION VERSUS DISTANCE BY SCENARIO, pg/m3
(NO BACKGROUND CONCENTRATION ADDED)
Scenario
Annual concentrations
Powder River Basin
San Juan Basin
Green River/Hams
Fork Basin
Second-highest 24-hour
concentrations
Powder River Basin
San Juan Basin
Green River/Hams
Fork Basin
TSP
At
boundary
115
23
30
867
106
104
Distance from
boundary, miles
1
115
20
20
260
80
55
2
10
8
13
240
50
25
3
7
7
7
165
45
18
4
6
6
4
80
37
14
5
4
5
3
45
32
10
PM 10
At
boundary
51
16
23
289
35
30
Distance from
boundary, miles
1
40
11
15
200
30
16
2
7
7
5
45
20
11
3
6
4
4
25
18
7
4
4
3
3
16
16
4
5
3
2
2
12
12
1
110
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
1
I
I
I
I
I
TABLE 24. VIOLATIONS OF THE NAAQS AND CLASS II PSD INCREMENTS
o
1-H
s
a.
«/>
H-
Class II Increment
Class II Increment
Secondary NAAQS*
10
co
£
i
i_
Q.
Distance
from boundary,
miles
m
»
m
(M
rH
*,§£
« O W
.0 -o
Distance
from boundary,
miles
If)
»
CO
CM
iH
c >»
*> 3 C
« O <0
A -O
Distance
from boundary,
miles
m
»
fO
(M
fH
*,§£
*£•§
Distance
from boundary,
miles
m
«•
PO
CM
i-t
-i^
< o n
J3 -O
Scenario
X
X X
XX X
X
X X
X X
X X
XXX XXX
XXX X X X
X
X
X X
X X
ac 9<
X X
«/> 3
c c o c
0 ••- f. .-
"-in tn i i/> tn
•u ic e «• v
C O) ro L.C 01 O Ol (0 UC
01 > 00 O> •— Ol — > CO Ol f
U--- > in £ «J — >v>
c er c — 10 D>iecr c — co
O (O Q: CD — t. -5 C^C
•— T> O) t- -D OJ -O Olt-
10 X C O>O CU X C O) O
3 O 10 I- u. O C O 10 Uu.
CQ.OOU U O Q. 00 O
C 01 0
< to
T3
>O
O
m
t-
c
Ol
o
u
O
O)
.*
U
10
CO
(O
111
-------
while no direct comparisons of the new techniques (emission
factors and model) with the IPP/CDM (with fallout function) tools
could be made, it appears that the new techniques result in
higher concentrations over a larger area. This comparison should
be studied further. Certainly one reason for the difference is
that the EPA 1981 emission factors are higher than the IPP fac-
tors. There is also probably a difference in fallout functions
that is also influencing comparative results.
4.4 POTENTIAL SOURCES OF ERROR IN THE PREDICTION PROCESS
The purpose of this subsection is to identify and evaluate
sources of error in the predictive process of developing a par-
ticulate concentration. Six error categories are presented in
Table 25. Ideally, each potential source of error in each cate-
gory would be quantified with a ± error term. The error cate-
gories would in turn be combined into a ± error term. Unfor-
tunately, a complete statistical analysis has never been attempted,
and it is not within the scope of the present study. It is an
area of recommended further research. At this time, the sensi-
tivity of predicted concentrations to each error category is not
even known. Therefore, this discussion will be limited to a
qualitative analysis of each error category.
The first error category is the emission factors. More
error analyses have been performed on this error category than
perhaps any other. The average 80 percent confidence interval
for the 1981 EPA TSP emission factors was calculated to be -20 to
+24 percent when applied at western mines in the range of correc-
tion parameter conditions over which testing was conducted.
Error analyses for the IPP, TRC, and AP-42 emission factors have
not been performed. As important as the uncontrolled emission
factor is the assumed control efficiency. Since the uncontrolled
emission factor term is simply multiplied by a percentage control
term for use in the emissions calculations, each term is of equal
importance. Control efficiency testing has been very limited and
the error band has not been comprehensively evaluated. Other
sources of emission factor error are listed in Table 25.
The second error category is activity parameters. The
controlled emission factor is multiplied by an activity parameter
to estimate emissions. The first source of error is identifying
all sources of particulate emissions. The very different source
lists used in previous modeling exercises illustrate this problem.
The level of source activity can be extremely difficult to predict,
with predictions required for such parameters as vehicle miles
traveled, grader speed, dragline drop distance, etc. Predictions
of activity are probably more accurate for longer averaging
periods.
112
-------
I
I
I
I
1
I
I
I
I
I
i
i
I
i
I
i
i
i
i
TABLE 25. POTENTIAL SOURCES OF ERROR IN THE PREDICTIVE PROCESS FOR
ESTIMATING PARTICULATE CONCENTRATIONS AROUND SURFACE COAL MINES
Error category
Emission factors
Activity parameters
Source location
Meteorological inputs
Model-related
(continued)
Potential sources of error
Measurement error in derivation of factors
Errors in assumed values for independent
variables
Not all factors are based on actual field
testing
Applicability of factors to mines other
than those tested
Assumptions in particle size distributions
Identification of independent variables
affecting emission rates
Control effectiveness assumptions
Averaging time
Ability to predict source types
Ability to predict level of source activity
Averaging time
Ability to predict location
Ability to predict spatial extent of sources
Release height
Averaging time
Windspeed
Wind direction
Stability class
Dispersion parameters
Mixing height
Representativeness of
area
data over entire mine
Averaging time
Gaussian dispersion algorithm
Plume rise algorithm
Deposition algorithm
Effects of complex terrain
Source representation (point, area, line)
Effects of pit retention
Receptor locations relative to assumed
source locations
Averaging time
113
-------
TABLE 25 (continued)
Error category
Potential sources of error
Verification
Measurement error of ambient monitors
vol, particle sizing devices)
Source/receptor relationship
Representativeness of data
Quality assurance program
Conversion algorithm for model results
(arithmetic to geometric mean)
Background concentration
Averaging time
(hi-
114
-------
I
I
I
I
i
i
I
i
I
i
i
i
i
i
I
i
i
i
i
Another error category is source location. Sensitivity of
model results to errors in source location has not been quanti-
fied. Overprediction or underprediction of concentrations is
possible, as well as errors in the location of predicted concen-
trations.
Meteorological inputs are required by the dispersion model
and are the variables that determine dispersion. There are at
least seven potential sources of error in this error category
that impact predicted concentrations to an unknown degree.
Several semi-rigorous evaluations of model performance have
been performed. Sources of error are listed in the table. While
is was not possible to evaluate the overall accuracy of the model
results, it is possible to discuss the relative accuracy of
certain types of model applications. For example:
0 Relative versus Absolute Application: Using
a model as a predictive tool to project
differences between two or more alternatives
is inherently more accurate than attempting
to calculate an absolute number.
0 Magnitude versus Location: Models generally
predict the magnitude of a concentration more
accurately than the precise location of that
magnitude because of inaccuracies in the
meteorological fields used by the models.
0 Long-Term versus Short-Term Estimates: The
accuracy of calculated annual average con-
centrations is better than calculated worst-
case 1-, 3-, or 24-hour concentrations because
mean value is more accurately determined than
extreme values (National Commission on Air
Quality 1980).
Other model-related problems that are particularly troublesome
are the inability to simulate the complex terrain often found in
surface coal mining areas, and the influence of pit retention.
The final error category is verification. There are two
problems that make it difficult to determine the accuracy of the
predictive process. First, most air quality analyses are per-
formed for mine permitting. The mine is not in operation and
therefore there are no data to use to compare predicted impact
versus actual monitored data. The second problem relates to the
error in particulate measurement devices. If the predictive
process were applied to an existing mine, it would still be
difficult to obtain an accurate actual term. For example, col-
located hi-vols often yield differences of >10 percent. Particle
115
-------
sizing instruments have equal or greater error. The accuracy of
measured concentrations has its own associated error bounds.
The preceding discussion illustrates that there are many
potential sources of error, any of which can be quite large. In
order to keep the model results in proper perspective, further
research is needed to define the uncertainty in the predicted
concentrations and the relationship of these uncertainties to the
various sources of error shown in Table 25.
4.. 5 RELATIONSHIP OF SCENARIO RESULTS TO POSSIBLE REGULATORY
OPTIONS
It is beyond the scope of this project to comprehensively
evaluate the relationship of the scenario results to possible
regulatory changes. However, four regulatory options are con-
sidered briefly. They are:
1. Apply additional dust controls to particulate sources.
2. Change PSD pollutant of measurement from TSP to PM 10.
3. Apply NAAQS and/or PSD increments at some distance beyond
the mine boundary.
4. Change the period for comparison to standards from worst-
case to a percentile or average period.
4.5.1 Additional Controls
Application of additional dust controls would reduce emis-
sion levels, and consequently, concentrations. Table 26 examines
the level of control assumed for certain sources at the Powder
River Mine and additional controls that could possibly be imple-
mented if cost and other environmental effects were totally
neglected. For example, in the scenario analyses, controls were
assumed for only haul and access roads, and the coal preparation
facility. Other sources are conceivably controllable, but are
usually not controlled due to the nature of the sources and the
costs associated with the controls. The control methods and
efficiencies beyond those assumed in the scenario analysis shown
in Table 26 are purely theoretical and should not be construed as
being practical or accurate. Additional research is needed to
determine whether or not the controls are feasible, and if fea-
sible what level of control could be achieved. Table 26 is
included as an extreme example of the maximum level of emissions
control conceivable.
Using a simple rollback technique, the effect of these
additional controls on predicted concentations was examined for
116
-------
I
I
I
I
I
1
I
i
I
i
i
i
i
i
i
i
I
i
i
LU |
— 1 1
O-
Si
!
2
LU
O
00
( ( I
oo
03
LU
t— t
Q£
LU
Q
3
O
Q. '
ft
O
00
•— •* '
LU !
O
oo i
-J 1
o
z
o
CJ
LU
H-
^•4
Q-:
a.
_i
o
i— i
t—
t— 4
0
Q
U.
O
r—
rf
r\
s:
vo
CM
LU
-J
CO
"^—
Ol
*-> —
O O
ii
re c
(1 Q
LJ
^j
«j a*
£ -Q
w —
c* re
(- CJ
3 —
O *•*
iyi u
Q.
re
t-
OJ */*
c c
i/i Ol O
i/) re
>, *J Ol
c —
O tJ C
"L" — E
re — c
! C O O
Ol b b
i U «J •—
1 ^ e >
o c
C (JO)
'•^ .So
•— o -
— W *J
i w or o
1 io
u «•»
0 — 3
*/i •••' T3
Ol */) 01
b ,C —
,3 0.0
0 £
° « C
u o
(- U
o
1 VJ
1
-—
W)
1 *•«
j c
a
1
u
IA
C
•a
01
*o
b.
c
o
u
tA
3
t3
•— ifl fc-
— c re
o o at
o •— c
u E o
C 0) W
Ol
u
t_
3
o
trt
O> i/> L.
"~ O 4*
C ••- C
0 E 0
<_> W *•>
•a
Ol 4-> •—
•«-* c o
« Si
— i- C
*j 0) O
a. u
LU
O
u
c
o
•o
•—1/11-
— c re
o o oi
"c Ul ^»
o -~ c
U E 0
3 * **
O)
0
/» w»
c -~ c
0 B 0
0 41 ->
•a
*> C 0
i Si
IA CX U
LU
o
t-
1
? .
.— Wl W
^ c re
o o oi
C tfl (A
o — c
u » o
01
u
f-
3
5
fk..
cn
re
o
1
•~
o
1
CO
o
o
a> at
> i-
— 3
cn oy
a* u
z a.
CM
c
0)
u O)
b. C
3 -^
"Sz
> U
O "O
g
01
X
CO
s
CM
1
* C
II
ri
i!
CM
|
—
o
Ul
a.
o
CM
in
lO
m
CO
"re
u
I
00
m
CNJ
fl
*^ 3
— •£
X 0
(M
CM
in
S
5
(O
^>
o
1
1
S3
X U
Csj
C
Ol
3 -S
oi re
O Z
r*.
CO
1— t
GO
"re
u
I
o
0
r«-
i
4->
U
re o
x u
o
0
s
b. >>
rtJ Q.
2 IA
0
I
1
s
00
s
c
3 re
,n >
li
o i-
0
un
m
^*
in
CO
> b
— 3
re *"
z a
o
s
cn
u
I
•o
_
o
u
I/I
cn
0)
°-
o
g
^
IV
2
(A
u
un r^ *-«
t-t m csj
PO
c
*J •—
C C C *J
oi a* oi w
"OS I "O "0
b. O> I- —
3 U (A3 .0
^ re i- «o
u — oi u f—
oi a. IM oi re
> Of O > O
Ob. a o u
•^ LO «r
en ^-4
o un un
Ol 0* — —
> u re re
•~ 3 U (J
4-> l/» •*- ••-
re in S fi
CD 01 o> oi
Ol b. £ £
Z Q. O CJ
o ^ m
- s s
CD •*/>'«
C "O 41 >
•- « "~ 2
b. C Ol b.
•— i x re
0 •«-> 3 U
cj — j *a
in
cn
an
•0 oi
Ol lA
U> 3
O O
"if
LU .fl
O
r*.
S.
• * re
wJ'J'e
"- ^^1 IA
x: ai *
M tl > ^
3 t e re
fc. y O O
U (A U O
* CM on
cn v ao
re c
0 0
u --^
*" >*
I O b
b. ui o> b
oi u o
*o oi "O a.
re N c E
b 0—01
CO un O •-«
»-^ un r"s
n
un un
co cn
„
— -o a» a» o>
re oi > b w>
u */» •- s 3
•— O W wi O
S^- re m £
u ai oi 01
£ c cu b re
CJ UJ C r^ Q
cn o «*•
SC3 CM
un w
t-4
tJ^
C
C Ol *J
•Si 1
ob re
b O> O
oi a. —
•a — *—
•^ 0) O
2 »/* o
o «r
o *
00
^
CM
*
e ui u
• O 3 W —
- 0 CO
3IA U *J
•^^ b C
o E— a* o
t— oi re o- u
u
I
w
>
•a
I
o
a
a> ai
W I/I
s-|«
"Is
^e-J
Ol O IA
> — o
-^a
5 I!
€=•£
ki!S
a,iu
J3 «>
b. W
S10?
"SoS
e —
O 3 -
CJ O lA
* *tn
S2^.
— w c n
E e ™ ai
>,l_> O ID
'u-o
X . fl 01
u — c in
ID * ai o
> w u a.
c i/> x
*-:
-— 3 "O **
£-«3
+J V* r—
C — O **-
O b O
u x+*
*/ C CM
O *^" O **•>
** f— O i-l
2-.51S
=tjll
ilH
117
-------
each of the three scenarios. These results appear in Table 27.
Three very critical assumptions make the results only approxima-
tions: (1) the controls, which are far greater than any in place
at an existing mine, are feasible, (2) the control efficiencies
are accurate, and (3) concentration reductions are directly
related to emission reductions. Consequently, the results should
be considered with a high degree of skepticism.
Even with the extreme level of control, TSP concentrations
with the Powder River Basin scenario would be more than twice the
PSD Class II increment. PM 10 concentrations would also consume
all of the PSD Class II increment.
4.5.2 Change PSD Pollutant of Measurement from TSP to PM 10
Repeated reference throughout the report has been made to
the possibility of changing the PSD pollutant of measurement from
TSP to PM 10. Because PM 10 concentrations are significantly
lower than TSP concentrations, this has the affect of making the
PSD increments less restrictive.
Tables 23 and 24, which summarize the scenario analysis,
indicate that PM 10 at two of the three mines examined would
still consume all of the increment at the mine boundary, although
by a much smaller margin than if TSP concentrations were used for
the comparison. Certainly, however, more mines could obtain PSD
permits if this change was made.
4.5.3 Apply NAAQS and/or PSD Increments at some Distance Beyond
the Mine Boundary
Another possible regulatory option is to apply the NAAQS
and/or PSD increments at some distance beyond the mine boundary.
Possible justifications are that high concentrations decrease
rapidly with distance, and that the worst-case conditions (where
major dust producing activities are adjacent to the boundary
resulting in maximum off-site concentrations) may not be indica-
tive of concentrations over the mine life.
It would probably be difficult to use this concept, for
application of the NAAQS since they are health-related standards.
Public access would be possible to areas where concentrations
exceeded healthful concentrations. Conversely, if the PSD pro-
gram is viewed as a resource allocation program (as opposed to
protecting human health), is may be reasonable to apply the
increment consumption determination at some distance beyond the
boundary.
The consequences of applying the increment determination at
distances beyond the mine boundary can be examined by refering to
the results of the scenario analysis summarized in previously
118
-------
I
I
I
1
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
TABLE 27. IMPACT OF ADDITIONAL PARTICIPATE CONTROLS
ON MAXIMUM OFF-SITE CONCENTRATIONS
Scenario
Powder River Basin
San Juan Basin
Green River/Hams
Fork Basin
Parameters from scenario
analysis
TSP
emissions,
tons/year
8303
6559
1286
Max. annual
off-site
concentration,
(jg/m3
TSP
115
23
30
PM 10
51
16
23
Parameters with additional
controls
TSP
emissions,
tons/year
3353
3752
764
Max. annual
off- site ,
concentration,
Mg/m3
TSP
46
13
18
PM 10
21
9
14
Additional controls listed in Table 4-13. Control may be prohibitively
. expensive or have adverse environmental impacts.
Based on rollback method, not dispersion modeling. Results should be
regarded as preliminary and approximate.
119
-------
cited Table 24. Regarding TSP concentrations, the 24-hour incre-
ment would still be a restraint as far as four to five miles
beyond the mine boundary. However, if PM 10 concentrations were
used to compare to increment consumption, a two mile buffer
around a mine boundary would allow even the Powder River Basin
scenario mine to receive a permit.
4.5.4 Change the Period for Comparison to Standards from Worst-
Case to a Percentile or Average Period
Several of the analyses in this report have shown that
worst-case off-site concentrations occur when the major dust-
producing activities are located adjacent to the mine boundary.
During this period, off-site concentrations can be two to three
times greater than when the major dust-producing activities are
not near the mine boundary. Thus some period shorter than the
mine life dictates maximum concentrations that will be permitted
throughout the mine life.
The percentage of the time that the major dust producing
activities, such as the pit, are adjacent to the mine boundary,
is a function of several parameters. These include the shape of
the mine (square versus long and narrow), regularity of boundaries,
ratio of the pit area to the size of the mine, and mining method.
Virtually no two mines would be the same, and many would not even
fall within the same range.
After a very preliminary analysis, the system appears to be
impossible to implement during the preliminary mine planning
stages. After the pit locations have been delineated, the com-
putational procedure would be extremely cumbersome.
120
-------
I
I
I
I
I
I
f
I
I
I
I
I
I
I
I
I
I
I
I
SECTION 5.0
SYNTHESIS OF THREE APPROACHES TO CHARACTERIZE PM 10 AND TSP
AIR QUALITY AROUND WESTERN SURFACE COAL MINES
5.1 INTRODUCTION
Three approaches to characterizing PM 10 and TSP air quality
around western surface coal mines are utilized in this report.
ideally, a direct comparison of results from the three methods
would be applied to several existing identifiable mines, i.e.,
monitoring, IPP/CDM modeling results, and EPA 1981/ISC modeling
results could be directly compared. Unfortunately, this could
not be accomplished with existing data. No mine was found that
had coincident measured and modeled concentration data. The
comparative study would be useful but additional monitoring data
are required before the comparisons could be made (see Section
6.0, Need for Further Study). Some limited comparisons are made
below.
5.2 COMPARISON OF ALTERNATE APPROACHES
A comparison of the three alternate approaches is presented
in Figures 58 and 59. The data plotted in these figures repre-
sent maximum mine boundary concentrations as a function of annual
production. Background concentrations have been removed from all
data so that the results could be compared.
The results are in relative agreement when considering the
possible causes of data scatter. These are:
1. Maximum boundary concentrations vary over time and are only
an approximate indicator of air quality impact since both
total emissions and resulting concentrations are a function
of a number of variables: annual production, mine configura-
tion, acres disturbed per year, and other factors.
2. Ambient monitors are seldom located to measure maximum
fenceline concentrations. This would cause the monitored
data to appear lower than the model results.
3. The previous modeling and scenario modeling used different
emission factors and models.
121
-------
MAXIMUM ANNUAL CONCENTRATION AT MINE BOUNDARY, ANNUAL GEOMETRIC MEAN,
ug/m3 (NO BACKGROUND CONCENTRATION ADDED)
PM 10 TSP
rv> j> CTI INS J>
-------
1
1
200
1
| 150
1
Q£ 0.
ISs £ 100
ig
0 Q
CD
-------
Figures 58 and 59 are a graphic summary of the conclusions
set forth in Sections 2.0 through 4.0. The NAAQS are constraining
only for very large mines. The PSD Class II increments are
considerably more constraining, particularly the 24-hour incre-
ments .
124
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
SECTION 6.0
NEED FOR FURTHER STUDY
The need for further research became apparent at several
points in the study. The importance of the problem is related to
(1) the proximity of western surface coal mines to Class I lands
and other areas of relatively pristine air quality; and (2) the
importance of the surface coal mining industry to the nations
energy and economic welfare.
Further study can be divided into five broad categories.
These are:
1. Additional monitoring.
2. Analysis of deficiencies in the predictive process.
3. Impact of additional particulate control measures and alter-
nate mine configurations on concentrations.
4. Standardized methods.
5. Regulatory implications.
6. Development of better PM 10 measurement methods.
A brief outline of study for each category is presented in this
section.
6.1 ADDITIONAL MONITORING
Additional monitoring and monitoring methods are required to
meet three objectives. These are:
0 Quantify pit retention. The impact of pit retention,
while acknowledged in principle by most investigators,
has not been conclusively quantified. No allowance for
pit retention is allowed in the permitting process. It
is desirable to quantify pit retention so that impacts
are not overpredicted.
125
-------
0 Monitor PM 10 concentrations. PM 10 sampling data are
very limited. Monitored PM 10 data are required at
impacted and background sites to adequately characterize
PM 10 concentrations around surface coal mines.
0 Measure the change in concentration with distance and
model validation. There are very little sampling data
to measure the change in concentration over distance.
Theoretical dispersion models have been used to simu-
late deposition. Measured particle size specific
values could be used to validate dispersion models.
Such a monitoring program could best be carried out in the
Powder River Basin where the largest monitored TSP data base
exists. PM 10 monitors could be collocated at existing TSP
monitoring locations. Other collocated samplers could be located
around pits, at mine boundaries, and beyond. Such a sampling
array at one to three mines would allow achievement of all three
objectives.
t
6.2 DEFICIENCIES IN THE PREDICTIVE PROCESS
The predictive process for projecting particulate concen-
trations is a multistep process. Six error categories are emis-
sion factors, activity parameters, source locations, meteorological
inputs, dispersion modeling (including deposition and pit reten-
tion), and model verification. Within each error category are
several sources of error (see previously cited Table 2l). Ideally,
potential source of error in each category would be quantified
with a ± error term. All sources of error would then be combined
into a ± error term which described the reliability of the predic-
tive process. This analysis has never been conceived of in this
comprehensive form.
The error analysis should be undertaken for annual predic-
tions and for 24-hour predictions. It is likely that short-term
predictions are considerably less accurate than annual predic-
tions. The error associated with 24-hour predictions may invali-
date their legitimate use for determining compliance with stan-
dards .
6.3 IMPACT OF ADDITIONAL CONTROL MEASURES AND ALTERNATE MINE
CONFIGURATIONS ON CONCENTRATIONS
The scenario analysis in this report was performed based on
a specified level of control measures and assuming certain mine
configurations. The sensitivity of concentrations to incremental
changes in particulate controls, and to changes in mine configura-
tion, has never been comprehensively analyzed. The objective of
126
-------
I
I
I
I
I
I
I
I
I
I
I
I
1
1
I
I
I
I
I
this research would be to identify the principles that would
allow optimization of mine controls and layout from an air quality
perspective.
6.4 STANDARDIZED METHODS
All new mines must apply for various air quality related
permits at the state and/or federal levels. Issuance of the
permits is usually contingent upon prediction of air quality
levels and attainment of standards. While the standards are
relatively uniform across the county, the method to predict
concentrations is not standardized.
This is particularly critical for short-term predictions and
would be critical for PM 10 predictions if they were required.
To illustrate with 24-hour predictions, guidance is required on
how to define 24-hour meteorological conditions, 24-hour activity
parameters, and source locations. Variation in practice for
these three parameters alone can result in differences in projected
concentrations by several orders of magnitude for the same mine.
This would nullify the creditibility of regulatory review.
Likewise, an initial particle size distribution is very important
to PM 10 predictions, as well as several other possible variations
in the analysis procedure.
6.5 REGULATORY IMPLICATIONS
A number of regulatory changes are currently being explored
by EPA and others. The changes revolve primarily around a par-
ticle size specific standard for the NAAQS and/or PSD process,
and fugitive emissions and the PSD process. These regulatory
changes have high potential to drastically impact the ability of
a new mine to obtain the necessary permits, and may also influence
existing mines.
The regulatory impact on the following parameters should be
investigated as a minimum.
0 Coal production by region.
0 Coal production costs
0 Impact on ambient concentrations near mines.
0 Impact on Class I lands.
0 Economic impact.
127
-------
REFERENCES
Burt. 1977. Valley Model User's Guide.
Environmental Protection Agency. 1978a. Interim Policy Paper on
the Air Quality Review of Surface Mining Operations. Environmental
Protection Agency, Region VIII, Denver, Colorado.
Environmental Protection Agency. 1978b. User's Guide for PAL.
EPA-600/4-78-013.
Environmental Protection Agency. 1978c, User's Guide for RAM.
EPA-600/8-78-016a.
Environmental Protection Agency. 1979. Industrial Source Complex
(ISC) Dispersion Model User's Guide. EPA-450/4-79-031.
Environmental Protection Agency. 1980. Compilation of Air
Pollutant Emission Factors. Third Edition, AP-42.
Environmental Research & Technology, Inc. 1979. A comparison of
Alternate Approaches for Estimation of Particulate Concentrations
Resulting from Coal Strip Mining Activities in Northeastern
Wyoming. Prepared for U.S. Department of Energy.
National Commission on Air Quality. 1980. Summary Report of the
NCAQ Atmospheric Dispersion Modeling Panel Volume 1, Recommenda-
tions. NTIS Publication No. PB80-174964.
PEDCo Environmental, Inc. 1978. Survey of Fugitive Dust from
Coal Mines. Prepared for Environmental Protection Agency, Region
VIII, Denver, Colorado.
PEDCo Environmental, Inc. 1981a. Improved Emission Factors for
Fugitive Dust from Western Surface Coal Mining Sources.
PEDCo Environmental, Inc. 1981b. Nomograph Method for Determining
Impact of TSP Concentrations Resulting from Surface Coal Mining
in Colorado.
PEDCo Environmental, Inc. 1981c. Nomograph Method for Determining
Impact of TSP Concentrations Resulting from Surface Coal Mining
in Wyoming.
128
-------
I
I Radian Corporation. 1979. Influence of Alternate Definitions of
Exempt Fugitive Dust Sources on the Impact of PSD Regulations on
. Surface Coal Mines.
* State of Wyoming, Division of Air Quality. 1979. Guideline for
Fugitive Dust Emission Factors for Mining Activities.
I
I
I
I
I
I
I
1
I
I
I
I
I
I
I
I
TRC Environmental Consultants, Inc. 1981. Coal Mining Emission
Factor Development and Modeling Study.
Turner. 1969. Workbook of Atmospheric Dispersion Etimates. PHS
Publication No. 999-AP-26.
129
-------
APPENDIX A
MASS FRACTION CALCULATIONS DERIVED FOR SECTION 2.0
PROCEDURE TO INFER PM 10 CONCENTRATIONS FROM
MEASURED TSP DATA
130
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
ISCST RESULTS, 0-2.5
(H9/m3)
STABILITY
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
WIND
SPEED (nips)
0.75
2.50
4.30
6.80 *
9.50
12.50
1000m
16.2
69.61
131.68
245.3
4.86
20.88
39.50
73.56
2.83
12.14
22.97
42.77
1.79
7.68
15.19
27.04
1.28
5.50
10.39
19.36
0.97
4.18
7.90
14.71
1500
4.98
32.28
^68.38
58.95
1.50
9.68
20.51
47.67
0.87
5.63
11.92
27.71
0.55
3.56
7.89
17.52
0.39
2.55
5.40
12.54
0.30
1.94
4.10
9.53
2000
2.47
21.19
46.16
115.08
0.74
6.36
13.85
34.51
0.43
3.70
8.05
20.06
0.27
2.34
5.33
12.69
0.19
1.67
3.64
9.08
0.15
1.27
2.77
6.90
3000
1.08
12.39
28.59
77.84
0.32
3.72
8.58
23.34
0.19
2.16
4.99
13.57
0.12
1.37
3.30
8.58
0.085
0.98
2.26
6.14
0.065
0.74
1.72
4.67
4500
0.84
6.90
18.24
56.16
0.25
2.07
5.47
16.84
0.15
1.20
3.18
9.79
0.092
0.76
2.10
6.19
0.066
0.54
1.44
4.43
0.050
0.41
1.09
3.37
7000
0.60
3.25
10.36
39.81
0.18
0.97
3.11
11.94
0.10
0.57
1.81
6.94
0.066
0.36
1.20
4.39
0.048
0.26
0.82
3.14
0.036
0.19
0.62
2.39
10,000
0.45
1.67
6.03
28.49
0.13
0.50
1.81
8.54
0.08
0.29
1.05
4.97
0.049
0.18
0.70
3.14
0.035
0.13
0.48
2.25
0.027
0.10
0.36
1.71
15,000
0.32
0.77
3.07
18.11
0.10
0.23
0.92
5.43
0.06
0.13
0.54
3.16
0.035
0.085
0.35
2.00
0.025
0.061
0.24
1.43
0.019
0.046
0.18
1.09
20,0
0.25
0.44
1.86
12.58
0.07
0.13
0.56
3.77
0.04
0.77
0.33
2.19
0.02;
0.04S
0.22
1.39
0.02Q
0.035
0.15
0.99
0.015
0.026
0.11
0.75
131
-------
ISCST RESULTS, 2.5-5
(pg/m3)
STABILITY
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
WIND
SPEED (nips)
0.75
2.50
A. 30
6.80 *
9.50
12.50
1000m
16.13
69.28
131.09
244.53
4.84
20.78
39.31
73.24
2.81
12.08
22.86
42.57
1.78
7.64
15.12
26.91
1.27
5.47
10.34
19.26
0.97
4.16
7.86
14.64
1500
4.96
32.13
68.09
158.79
1.49
9.64
20.41
47.49
0.87
5.60
11.87
27.59
0.55
3.54
7.85
17.44
0.39
2.54
5.37
12.48
0.30
1.93
4.08
9.49
2000
2.45
21.09
45.97
115.02
0.74
6.33
13.78
34.39
0.43
3.68
8.01
19.98
0.27
2.33
5.30
12.63
0.19
1.66
3.63
9.04
0.15
1.27
2.76
6.87
3000
1.07
12.33
28.47
77.80
0.32
3.70
8.54
23.26
0.19
2.15
4.96
13.51
0.12
1.36
3.28
8.54
0.084
0.97
2.25
6.11
0.064
0.74
1.71
4.65
4500
0.83
6.87
18.16
56.11
0.25
2.06
5.44
16.78
0.14
1.20
3.17
9.75
0.091
0.76
2.09
6.16
0.065
0.54
1.43
4.41
0.050
0.41
1.09
3.35
7000
0.59
3.23
10.31
39.74
0.18
0.97
3.09
11.89
0.10
0.56
1.80
6.91
0.065
0.36
1.19
4.37
0.046
0.26
0.81
3.13
OJ035
0.19
0.62
2.38
10,000
0.43
1.66
6.00
28.41
0.13
0.50
1.80
8.51
0.07
0.29
1.05
4.95
0.047
0.18
0.69
3.13
0.034
0.13
0.47
2.24
OJ026
0.10
0.36
1.70
15,000
0.29
0.76
3.05
18.03
0.09
0.23
0.92
5.41
0.05
0.13
0.53
3.14
20,000
i
0.20
0.44
1.85
12.51
0.06
0.13
0.56
3.76:
0.0401
0.07
0.32
2.18
0.03U 0.02
0.081 0.04
0.35
1.99
0.02'
OJD6(
0.24
1.42
OJDI:
OJ04(
- 0.21
1.38
0.016
0.03.'
0.15
0.99
cm:
0.021
0.181 0.11
1.081 0.75
132
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
ISCST RESULTS, 5-10 \*m
STABILITY
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
WIND
n .L^IX/
SPEED (ops)
0.75
2.50
4.30
6.80 '
9.50
12.50
1000m
15.08
64.79
122.76
230.10
4.52
19.43
36.7
68.60
2.63
11.29
21.37
39.84
1.66
7.14
14.13
25.18
1.19
5.11
9.67
18.02
0.90
3.88
7.34
13.69
1500
4.63
30.05
63.81
150.11
1.39
9.01
19.10
44.56
0.81
5.24
11.10
25.85
0.51
3.31
7.34
16.33
0.37
2.37
5.02
11.68
0.28
1.80
3.82
8.87
2000
2.29
19.72
43.05
108.69
0.69
5.91
12.89
32.28
0.40
3.44
7.49
18.72
0.25
2.17
4.95
11.82
0.18
1.56
3.39
8.46
0.14
1.18
2.58
6.43
3000
0.99
11.53
26.64
73.22
0.30
3.46
7.98
21.83
0.17
2.01
4.64
12.67
0.11
1.27
3.07
8.00
0.07J
0.91
2.10
5.72
0.05<
0.69
1.60
4.35
4500
0.73
6.42
16.97
52.42
0.22
1.93
5.09
15.74
0.13
1.12
2.96
9.14
0.08
0.71
1.96
5.77
0.057
0.51
1.34
4.13
0.044
0.39
1.02
3.14
7000
0.44
3.02
9.63
36.66
0.13
0.91
2.89
11.14
0.08
0.53
1.68
6.47
0.049
0.33
1.11
4.09
0.035
0.24
0.76
2.93
0.027
0.18
0.58
2.22
10,000
0.25
1.56
5.50
'5.85
0.07
0.47
1.68
7.96
0.04
0.27
0.98
4.63
0.028
0.17
0.65
2.93
0.020
0.12
0.44
2.09 -
0.015
0.093
0.34
1.59
15,000
0.10
0.71
2.85
16.0
0.03
0.21
0.86
5.05
0.018
0.12
0.50
2.94
0.011
0.079
0.33
1.86
0.0082
0.056
0.23
1.33
OJ3062
0.043
0.17
1.01
20,00
0.05
0.41
1.73
10.90
0.01
0.12
0.52
3.50
0.008
0.071
0.30
2.04
0.0054
0.045
0.20
1.29
0.0038
0.032
0.14
0.92
OJ0029
0.025
0.10
0.70
133
-------
ISCST RESULTS, 10-15
(M9/ni3)
STABILITY
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
WIND
SPEED (mps)
0.75
2.50
4.30
6.80 -
9.50
12.50
1000m
14.35
61.73
117.27
222.05
4.31
18.49
35.03
65.59
2.50
10.75
20.35
38.02
1.58
6.70
13.46
24.00
1.13
4.86
9.20
17.17
0.86
3.70
6.99
13.04
1500
4.41'
28.62
60.91
144.81
1.32
8.58
18.21
42.76
0.77
4.99
10.57
24.73
0.49
3.15
7.00
15.59
0.35
2.26
4.78
11.14
0.26
1.71
3.63
8.46
2000
2.18
18.77
40.97
103.38
0.65
5.63
12.29
30.97
0.38
3.27
7.14
17.92
0.24
2.07
4.72
11.20
0.17
1.48
3.23
8.07
0.13
1.13
2.45
6.13
3000
0.94
10.95
25.21
67.21
0.28
3.29
7.60
20.89
0.16
1.91
4.42
12.11
0.10
1.21
2.92
7.64
0.07'
0.87
2.00
5.46
0.05<
0.66
1.52
4.15
4500
0.66
6.09
L5.97
45.54
0.20
1.83
4.85
15.00
0.12
1.07
2.82
8.72
0.073
0.67
1.86
5.51
0.052
0.48
1.27
3.94
0.040
0.37
0.97
2.99
7000
0.37
2.87
9.01
29.22
O.lll
0.86
2.75
10.53
0.064
0.50
1.60
6.16
0.040
0.32
1.06
3.90
0.029
0.23
0.72
2.79
0.022
0.17
0.55
2.12
10,000
0.18
1.48
5.22
i8.73
0.05
0.44
1.60
7.46
0.032
0.26
0.93
4.40
0.020
0.16
0.62
2.79
0.014
0.12
0.42
2.00
0.011
0.089
0.32
1.52
15,000
0.07
0.68
2.65
9.92
0.02
0.20
0.81
4.66
0.011
0.12
0.47
2.78
OJ0073
0.075
0.31
1.77
0.0052
0.054
0.21
1.27
OD039
0.041
0.16
0.96
20.00C
0.03
0.39
1.60
5.80
0.01
0.12
0.49
3.19
0.00!
0.06*
0.29
1.92
OJ003
0.04:
0.19
1.23
o.oo:
0.03
0.13
0.88
OJOO
0.02
O.OS
0.6:
134
-------
I
I
I
I
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
ISCST RESULTS, 15-20 pm
.
STABILITY
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
1 A
B
C
D
A
B
C
D
WIND
" JLtvU
SPEED (taps)
0.75
2.50
4.30
*
6.80
9.50
12.50
1000m
14.03
60.42
115.13
220.51
4.21
18.09
34.30
64.54
2.45
10.51
19.91
37.31
1.55
6.64
13.16
23.52
1.11
4.76
9.00
16.81
0.84
3.61
6.84
12.76
1500
4.31
27.95
59.33
139.50
1.29
8.39
17.83
42.18
0.75
4.88
10.35
24.34
0.48
3.08
6.84
15.31
0.34
2.21
4.68
10.93
0.26
1.68
3.55
8.29
2000
2.13
18.28
39.50
94.14
0.64
5.50
12.02
30.45
0.37
3.20
6.98
17.62
0.24
2.02
4.62
11.09
0.17
1.45
3.16
7.92
0.13
1.10
2.40
6.01
3000
0.91
10.64
23.93
54.22
0.27
3.22
7.43
20.34
0.16
1.87
4.32
11.88
0.10
1.18
2.86
7.50
0.072
0.85
1.95
5.35
0.055
0.64
1.49
4.06
4500
0.63
5.91
L4.93
30.64
0.19
1.79
4.72
L4.37
0.11
1.04
2.75
8.51
0.070
0.66
1.82
5.39
0.050
0.47
1.25
3.86
0.038
0.36
0.95
2.93
7000
0.34
2.78
8.31
14.85
0.10
0.84
2.68
9.83
0.059
0.49
1.56
5.96
0.037
0.31
1.03
3.80
0.027
0.22
0.71
2.73
0.020
0.17
0.54
2.07
10,000
0.16
1.43
4.76
7.00
0.05
0.43
1.55
6.77
0.028
0.25
0.91
4.21
0.018
0.16
0.60
2.71
0.013
0.11
0.41
1.95
0.0096
0.087
0.31
1.48
15,000
0.06
0.66
2.39
2,22
0.02
0.20
0.79
4.04
0.0097
0.12
0.46
2.62
0.0062
0.073
0.31
1.71
OJD044
0.052
0.21
1.23
OJ0033
0.040
0.16
0.94
20,00
0.02
0.38
1.43
0.79
0.01
0.11
0.48
2.64
OJ0043
0.066
0.28
1.78
0.0027
0.042
0.19
1.18
OJ0019
0.030
0.13
0.85
OJ0015
0.023
0.097
0.65
135
-------
ISCST RESULTS, 20-30
STABILITY
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
WIND
SPEED (mps)
0.75
2.50
4.30
6.80 *
9.50
12.50
1000m
13.38
57.70
110.22
213.19
4.01
17.27
32.83
62.34
2.33
10.03
19.03
35.87
1.48
6.34
12.57
22.56
1.06
4.54
8.59
16.10
0.80
3.45
6.53
12.21
1500
4.11
26.38
54.47
112.25
1.23
8.00
17.04
40.51
0.72
4.65
9.89
23.45
0.45
2.94
6.53
14.73
0.32
2.10
4.47
10.50
0.25
1.60
3.39
7.95
2000
2.03
17.09
34.70
58.46
0.61
5.25
11.44
28.69
0.35
3.05
6.67
16.89
0.22
1.93
4.41
10.66
0.16
1.38
3.01
7.60
0.12
1.05
2.29
5.76
3000
0.86
9.84
19.71
19.87
0.26
3.06
7.02
18.31
0.15
1.78
4.12
11.22
0.095
1.13
2.73
7.16
0.068
0.81
1.86
5.13
0.062
0.61
1.42
3.89
4500
0.58
5.43
LI. 58
5.15
0.17
1.70
4.44
.2.06
0.10
0.99
2.62
7.85
0.064
0.63
1.74
5.11
0.046
0.45
1.19
3.68
0.035
0.34
0.90
2.80
7000
0.29
2.54
6.08
0.75
0.09
0.80
2.50
7.41
0.050
0.47
1.48
5.30
0.032
0.30
0.98
3.55
0.022
0.21
0.67
2.58
0.017
0.16
0.51
1.97
10,000
0.13
1.31
3.34
0.096
0.04
0.41
1.45
4.53
0.022
0.24
0.86
3.59
0..014
0..15
0.57
2.48
0.010
0.11
0.39
1.83
OJ0076
0.083
0.30
1.40
15,000
0.04
0.60
1.60
0.0034
0.01
0.19
0.73
2.21
OX)073
0.11
0.44
2.09
0.0046
0.070
0.29
1.52
OJ0033
0.050
0.20
1.14
0.0025
0.038
0.15
0.88
20,000
0.02 ,
0.35
0.94
OJ0001J
0.01
0.11
0.44
1.20
OJ0032
0.063
0.26
1.33
0.0020
0.040
0.18
1.02
0.0014
0.029
0.12
0.78
OJ0011
0.022
0.092
0.61
136
-------
APPENDIX B
COMPUTATION OF PM 10, PM 5, AND PM 2.5
MASS FRACTIONS AT VARIOUS DOWNWIND DISTANCES
FOR SECTION 2.0 PROCEDURE
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
STABILITY; 4.30 m/s WINDS; 1000 m
Size range
0 -
2.5 -
5.0 -
10.0 -
15.0 -
20.0 -
2.5p
5. Op
10. Op
15. Op
20. Op
30. Op
I < 2.5p
total
I < 5. Op
total
I < 10.Op
total
0.090
3.768
0.401
3.768
1.101
3.768
X/QCsnf1)
4.28
4.26
3.98
3.80
73
3.59
0.024
0.106
0.292
Mass fraction
0.021
0.073
0.176
0.148
0.115
0.467
X (pg/m3)
0.090
0.311
0.700
0.562
0.429
1.676
3.768
STABILITY; 4.30 m/s WINDS; 1500 m
Size range
0 -
2.5 -
5.0 -
10.0 -
15.0 -
20.0 -
2.5p
5. Op
10. Op
15. Op
20. Op
30. Op
I < 2.5p
total
I < 5.Op
total
I < 10.Op
total
= 0.024
= 0.105
0.291
X/QCsnf1)
2.77
2.76
2.59
2.47
2.43
2.35
Mass fraction
0.021
0.073
0.176
0.148
0.115
0.467
X (pg/m3)
0.058
0.201
0.456
0.366
0.279
1.097
2.457
138
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Size range
0 - 2.5n
2.5 - 5.OH
5.0 - 10.OH
10.0 - 15.OH
15.0 - 20.OH
20.0 - 30.OH
< 2.5M _
total
I < 5.OH
total
i < 10.OH
total
= 0.024
= 0.106
= 0.292
Size range
0 -
2.5 -
5.0 -
10.0 -
15.0 -
20.0 -
2.5n
5. OH
10. OH
15. OH
20. OH
30. OH
D
STABILITY; 4.30 m/s WINDS; 2000 m
X/QCsm"1)
2.
2.
I.
I.
I.
01
00
87
79
76
1.69
Mass fraction
0.021
0.073
0.176
0.148
0.115
0.467
D
STABILITY; 4.30 m/s WINDS; 3000 m
X/QCsm"1)
1.
1.
1.
1.
1.
36
35
27
21
19
1.12
Mass fraction
0.021
0.073
0.176
0.148
0.115
0.467
I < 2.5|j
total
i < 5. OH
total
i < IO.QIJ
total
0.024
0.107
0.295
139
0.042
0.146
0.329
0.265
0.202
0.789
1.773
0.028
0.098
0.224
0.179
0.137
0.523
1.190
-------
'STABILITY; 4.30 m/s WINDS; 4500 m
Size range
0 - 2.5|j
2.5 - 5.Op
5.0 - 10.0(j
10.0 - 15. Op
15.0 - 20.OH
20.0 - 30.OH
I < 2.5n
total
I < 5.OH
total
I < 10.Op
total
0.024
0.108
0.298
X/QCsnf1)
0.979
0.975
0.914
0.872
0.851
0.785
Mass fraction
0.021
0.073
0.176
0.148
0.115
0.467
0.020
0.071
0.161
0.129
0.098
0.367
0.846
D
STABILITY; 4.30 m/s WINDS; 7000 m
Size range
0 - 2.5n
2.5 - 5. OH
5.0 - 10.OH
10.0 - 15.OH
15.0 - 20.OH
20.0 - 30.OH
X/QCsm"1)
0.694
0.691
0.647
0.616
0.596
0.530
Mass fraction
0.021
0.073
0.176
0.148
0.115
0.467
0.014
0.050
0.114
0.091
0.068
0.248
0.586
total
= 0.025
I < 5. Op
total
I < 10. OH
01 ~i T
. ill
= n ^rm
140
-------
I
Size range
0 - 2.5M
2.5 - 5.0M
5.0 - 10.OM
10.0 - 15.OM
15.0 - 20.OM
20.0 - 30.OM
Size range
0 -
2.5 -
5.0 -
10.0 -
15.0 -
20.0 -
2.5M
S.OM
10. OM
15. OM
20.0|j
30.0(j
'STABILITY; 4.30 ra/s WINDS; 10,000 m
X/QCsnf1)
0.497
0.495
0.463
0.440
0.421
0.359
Mass fraction
0.021
0.073
0.176
0.148
0.115
0.467
0.010
0.036
0.081
0.065
0.048
0.168
0.409
I < 2.5u
total
I < 5. On
total
I < 10. On
total
— n noc
— U. U
-------
D
STABILITY; 4.30 m/s WINDS; 20,000 m
Size range
0 - 2.5|j
2.5 - 5. Op
5.0 - 10.Op
10.0 - 15.OM
15.0 - 20.0|j
20.0 - 30.0|j
I < 2.5p
total
i < S.QM
total
I < 10.Op
total
0.027
0.122
0.337
X/QCsm"1)
0.219
0.218
0.204
0.192
0.178
0.133
Mass fraction
0.021
0.073
0.176
0.148
0.115
0.467
X (pg/m3)
0.004
0.016
0.036
0.028
0.020
0.062
0.167
0
STABILITY; 2.50 m/s WINDS; 1000 m
Size range
0 - 2.5p
2.5 - 5.Op
5.0 - 10.Op
10.0 - 15.Op
15.0 - 20.Op
20.0 - 30.Op
X/QCsm"1)
I < 2.5p
total
I < 5. Op
total
I < 10. Op _
total
0.024
0.106
0.291
7.
7.
,36
32
6.86
6.56
6.45
6.23
Mass fraction
0.021
0.073
0.176
0.148
0.115
0.467
X (pg/m3)
0.154
0.534
1.207
0.971
0.742
2.909
6.518
142
-------
'STABILITY; 2.50 m/s WINDS; 1500 m
Size range
X/QCsnf1)
0 -
2.5 -
5.0 -
10.0 -
15.0 -
20.0 -
2.5M
5. OH
lO.Ofj
15. OH
20. OH
so. OH
4.77
4.75
46
,28
4.22
4.05
4.
4.
Mass fraction
0.021
0.073
0.176
0.148
0.115
0.467
I < 2.5p .
total
I < 5. OH
total
I < 10. OH
total
Ono/i
. U£4
01 nc
. lUb
OOQrt
. zyu
0
STABILITY; 2.50 m/s WINDS; 2000 m
Size range
0 - 2.5n
2.5 - 5.OH
5.0 - 10. OH
10.0 - is. OH
15.0 - 20.OH
20.0 - 30.OH
X/QCsirf1)
45
44
23
10
05
2.87
Mass fraction
0.021
0.073
0.176
0.148
0.115
0.467
I < 2.5n
total
I < 5. OH
total
I < 10.OH _
total
0.024
0.106
143
0.100
0.347
0.785
0.633
0.485
1.891
4.242
X (H9/m3)
0.072
0.251
0.568
0.459
0.351
1.340
3.042
-------
'STABILITY; 2.50 m/s WINDS; 3000 m
Size range
o -
2.5 -
5.0 -
10.0 -
15.0 -
20.0 -
2.5p
5. Op
10. Op
15. Op
20. Op
30. Op
I < 2.5p
total
I < 5. Op
total
I < 10. Op
— n n?d
— U. UcH
= 0.110
= n im
X/QCsnf1)
2.33
2.33
18
09
03
1.83
Mass fraction
0.021
0.073
0.176
0.148
0.115
0.467
X (pg/m3)
0.049
0.170
0.384
0.309
0.233
0.855
2.000
D
STABILITY; 2.50 m/s WINDS; 4500 m
Size range
0 -
2.5 -
5.0 -
10.0 -
15.0 -
20.0 -
2.5p
5. Op
10. Op
15. Op
20. Op
30. Op
I < 2.5p
total
I < 5.Op
total
I < 10.Op _
total
0.025
0.114
0.313
X/QCsm"1)
68
68
57
50
44
1.21
Mass fraction
0.021
0.073
0.176
0.148
0.115
0.467
X (pg/m3)
0.035
0.123
0.276
0.222
0.166
0.565
1.387
144
-------
D
STABILITY; 2.50 m/s WINDS; 7000 m
Size range
0 - 2.5M
2.5 - S.OM
5.0 - 10.OM
10.0 - is.OM
15.0 - 20.OM
20.0 - 30.OM
X/QCsm"1)
1.19
1.19
1.11
1.05
0.98
0.74
Mass fraction
0.021
0.073
0.176
0.148
0.115
0.467
< 10-
total
-0.121
= o 334
U'^4
D
STABILITY; 2.50 m/s WINDS; 10,000 m
Size range
0 - 2.5|j
2.5 - 5.0(j
5.0 - 10.OM
10.0 - 15.0(j
15.0 - 20.0|j
20.0 - 30.0(j
X/QCsnf1)
0.85
0.85
0.80
0.75
0.68
0.45
Mass fraction
0.021
0.073
0.176
0.148
0.115
0.467
I < 2.5M
total
Z < S.OM
total
I < 10.On
total
= 0.029
= 0.129
0.356
145
X (M9/m3)
0.025
0.087
0.195
0.155
0.113
0.346
0.921
0.018
0.062
0.141
0.111
0.078
0.210
0.620
-------
D
STABILITY; 2.50 m/s WINDS; 15,000 m
Size range
2.5M
0
2.5
5.0
10.0
15.0
20.0
10. OM
15. OM
20. OM
30. OM
1 < 2.5M
total
I < 5.0M
total
I < 10. OM
OD79
. \J3£
= 0.141
. - n 1Q9
X/QCsm"1)
0.54
0.54
0.51
0.47
0.40
0.22
Mass fraction
0.021
0.073
0.176
0.148
0.115
0.467
X (pg/m3)
0.011
0.039
0.090
0.070
0.046
0.103
0.359
D
STABILITY; 2.50 m/s WINDS; 20,000 m
Size range
0 -
2.5 -
5.0 -
10.0 -
15.0 -
20.0 -
2.5|j
5. Op
10. On
15.0(j
20.0|j
30. OM
I < 2.5M
total
i < S.QM
total
i < 10.OM
total
= 0.
0.155
0.'
X/QCsnf1)
0.38
0.38
0.35
0.32
0.26
0.12
Mass fraction
0.021
0.073
0.176
0.148
0.115
0.467
X (pg/m3)
0.008
0.028
0.062
0.047
0.030
0.056
0.231
146
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
APPENDIX C
ICS MODEL APPLICATION TO
THREE MINE SCENARIOS
147
-------
MODEL FEATURES
The ISC model programs (ISCST and ISCLT) accept the fol-
lowing source types: point, area, and volume. The line sources
are simulated by multiple volume sources. The transport of
particulate emissions from stacks (point sources) are determined
using the steady-state Gaussian plume equation for a continuous
elevated source. The area and volume source options are used to
simulate the impact of particulate emissions from a variety of
surface coal mining operations, such as reclaimed land and striped
overburden (area sources) and coal and overburden haul roads
(volume sources). The area source model is based on the equation
for a continuous and finite cross-wind line source. Since each
area source must be square, the effects of an irregularly-shaped
area source can be simulated by a series of multiple squares
approximating the source's actual geometry. In general, no plume
rise exists with area sources; consequently, the effective emis-
sion height is equivalent to the physical height of the source of
particulate emissions, which for many surface mine sources will
be groundlevel.
The steady-state Gaussian plume equation for a continuous
source is also used to calculate groundlevel particulate concen-
trations contributed by volume source emissions. A coal or
overburden haul road (line source) is represented by a series of
multiple volume sources. To represent a line source exactly, the
line source is divided into N volume sources, where N is deter-
mined by the length of the line source divided by its width.
Because this exact procedure is not always practical, a haul road
is often simulated by an approximate representation in which a
lesser number of equally spaced volume sources is used. In
spacing a smaller number of volume sources at equal intervals, an
approximate representation of the line source is achieved. The
initial lateral dispersion coefficient, a , is set equal to the
distance between adjacent volume sources divided by 2.15 so that
the road is simulated by a series of overlapping Gaussian distri-
butions resulting from each discrete volume source. As with area
sources, generally no plume rise occurs from volume sources.
Receptor Grid
Selection of a Cartesian or a polar receptor grid system
allows the user to design the ISC model output for the specific
application. Since a surface coal mine approximates a combina-
tion of multiple sources, not located at the same point, the
148
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Cartesian coordinate system is usually used to model mines.
Model receptors are generally placed outside what would normally
comprise the mine permit boundary for two reasons: first, most
state and federal regulations do not require that ambient air
standards be met inside plant property, and second, the ISC model
itself ignores the source contribution induced by sources within
100 meters of the nearest receptor.
Settling and Deposition
The effects of gravitational settling and dry deposition on
small particulates which tend to remain suspended in the atmo-
sphere for long distances is ignored by the ISC model. The
larger particulates, however, are brought to the ground surface
by a combination of gravitational settling and deposition.
Additionally, small particulates are generally reflected from the
ground surface, whereas the large particulates that come in
contact with the surface are usually completely or partially
retained at the surface. The ISC model includes the effects of
both dry deposition and gravitational settling. The Dumbauld dry
deposition model, used in the ISC dispersion model, assumes that
a user specified fraction of the material that comes into contact
with the ground surface is reflected from the surface back into
the atmosphere. The reflection coefficient is a function of the
gravitational settling velocity. Gravitational settling of large
particulates result in a tilted plume with the plume axis inclined
to the horizontal.
The particulate emissions from each modeled source are
subdivided into contiguous particle size categories so that the
full spectrum of particle sizes is represented by discrete cate-
gories. For each category, the mass fraction of particulates,
the particle reflection coefficient, and the gravitational settling
velocity for the size range within the category are specified.
The emission rates of particulate matter from each of the
three mine scenarios—Powder River Basin, San Juan Basin, and
Green River/Hams Fork Basin—have been computed in Subsection 4.2
of this report. The first step in modeling these mines was to
collect or combine emission rates of coincident particulate
producing activities. Haul roads, pit activities, and coal
handling facility sources were grouped together for subsequent
simulation as volume or area sources. This source collection and
apportionment is summarized in Tables C-l through C-6, which
illustrate annual average emission rates and worst-case 24-hour
emission rates for each of the three mines.
In all cases, haul roads and access roads were simulated as
spaced volume sources, emitting at groundlevel. Coal handling
facilities and mine pits were idealized as groundlevel area
sources. Details of source characteristics are summarized in
149
-------
Table C-7. As shown in Table C-7, the representation of a mine
during an annual time period and during a worst-case 24-hour
period sometimes differ. In a year's time, activities associated
with the mine pit cover a larger spatial area than they do during
a 24-hour period. At the San Juan Basin hypothetical mine, for
example, the pit can be idealized as five area sources, while in
a 24-hour time period, activity may only take place in a spatial
area represented by two area sources. All computations of a
and a in Table C-7 are in keeping with standard modeling practice
zo
A very important input to the ISC model are the particle
deposition parameters, summarized in Table C-8. Each of the
settling velocities and reflection coefficients was computed
using methods recommended in the ISC Dispersion Model User's
Guide. The distribution of mass within each particle size range
for each of the three mines was discussed previously in Subsec-
tion 4.2 of this report.
TABLE C-l. PARTICLE DEPOSITION PARAMETERS
Particle diameter, Settling velocity Reflection
microns m/s coefficient
0-2.5 0.000093 1.0
2.5-5.0 0.000837 0.99
5.0-10.0 0.003347 0.86
10.0-15.0 0.018223 0.73
20.0-30.0 0.037189 0.65
150
-------
I
I
I
I
I
TABLE C-2. APPORTIONMENT OF EMISSIONS, POWDER RIVER BASIN MINE: ANNUAL
TSP emissions
Source group tons/year
151
I Coal loading 240.9
Wind erosion 117.8
Overburden replacement 391.5
Dozers - overburden 43. 7
I Mine pit
Topsoil removal 9.7
Scraper travel 76.3
_ Topsoil dump 3.2
• Overburden drilling 2.9
• Overburden blasting 1.2
Overburden removal 693.8
I Dozers - coal 14.2
Coal drilling 1.0
Coal blasting 2.1
Coal loading ""* "
Wind erosion
Overburden replacement
Dn7pr<; - nv/prhurripn
I
I
I
I
I
I
I
I
I
1598.3
Haul roads
Haul trucks - coal 3952.2
Lt.- and med.-duty vehicles 664.1
Graders 3.4
Access road 1.5
Haul trucks - overburden 1217.4
5838.6
Facilities
Coal dump 450.0
Crushing, screening, conveying 65.7
Coal loadout 350.0
865.7
-------
TABLE C-3. APPORTIONMENT OF EMISSIONS, POWDER RIVER BASIN MINE: 24-HOUR
TSP emissions
Source group pound/hour
Mine pit
Topsoil removal 2.79
Scraper travel 21.93
Topsoil dump 0.92
Overburden drilling 0.83
Overburden blasting 0.64
Overburden removal 199.37
Dozers - coal 4.18
Coal drilling 0.29
Coal blasting 1.68
Coal loading 69.22
Wind erosion 35.95
Overburden replacement 112.50
Dozers - overburden 12.60
462.90
Haul roads
Haul trucks - coal 1135.69
Lt.- and med.-duty vehicles 190.80
Graders 0.98
Access road 0.81
Haul trucks - overburden 349.80
1678.08
Facilities
Coal dump 129.31
Crushing, screening, conveying 18.88
Coal loadout 100.57
248.76
152
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
TABLE C-4. APPORTIONMENT OF EMISSIONS, SAN JUAN BASIN MINE: ANNUAL
TSP emissions
Source group tons/year
Mine pit
Topsoil removal 38.5
Scraper travel 117.0
Topsoil dump 13.3
Overburden drilling 7.7
Overburden blasting 8.8
Overburden removal 1293.0
Dozers - coal 280.2
Coal drilling 6.6
Coal blasting 46.3
Coal loading 173.6
Wind erosion 1063.0
Dozers - overburden 234.2
3282.2
Haul roads
Haul trucks - coal 2463.0
Lt.- and med.-duty vehicles 386.1
Graders 176.7
Access road 12.0
3037.8
Facilities
Coal dump 117.0
Crushing, screening, conveying 12.4
Coal storage 109.4
Coal loadout -
238.8
153
-------
TABLE C-5. APPORTIONMENT OF EMISSIONS, SAN JUAN BASIN MINE: 24-HOUR
TSP emissions
Source group pound/hour
Mine pit
Topsoil removal 11.06
Scraper travel 33.62
Topsoil dump 3.82
Overburden drilling 3.10
Overburden blasting 4.51
Overburden removal 371.55
Dozers - coal 80.52
Coal drilling 1.90
Coal blasting 23.15
Coal loading 49.88
Wind erosion 310.82
Dozers - overburden 67.30
961.23
Haul roads
Haul trucks - coal 707.76
Lt.- and med.-duty vehicles 110.95
Graders 50.78
Access road 3.45
872.94
Facilities
Coal dump 33.62
Crushing, screening, conveying 3.56
Coal storage 31.44
68.62
154
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
TABLE C-6. APPORTIONMENT OF EMISSIONS, GREEN RIVER/HAMS
FORK MINE: ANNUAL
TSP emissions
Source group tons/year
Mine pit
Topsoil removal 6.5
Scraper travel 45.9
Topsoil dump 2.2
Overburden drilling 1.5
Overburden blasting 0.6
Overburden removal 365.0
Dozers - coal 73.4
Coal drilling 0.5
Coal blasting 4.6
Coal loading 82.2
Wind erosion 62.7
Dozers - overburden 12.9
658.0
Haul roads
Haul trucks - coal
Lt.- and med.-duty vehicles
Graders
Mine facilities
Coal dump
Crushing, screening, conveying
Coal storage
Coal loadout
155
494.4
133.0
Access road
Access road 0.9
0.9
-------
TABLE C-7. APPORTIONMENT OF EMISSIONS, GREEN RIVER/HAMS
FORK BASIN MINE: 24-HOUR
TSP emissions
Source group pound/hour
Mine pit
Topsoil removal 1.87
Scraper travel 13.19
Topsoil dump 0.63
Overburden drilling 0.43
Overburden blasting 0.42
Overburden removal 104.88
Dozers - coal 21.09
Coal drilling 0.14
Coal blasting 2.99
Coal loading 23.62
Wind erosion 19.00
Dozers - overburden 3.71
191.97
Haul roads
Haul trucks - coal 132.93
Lt.- and med.-duty vehicles 8.88
Graders 0.27
142.08
Mine facilities
Coal dump 18.36
Crushing, screening, conveying 1.93
Coal storage 3.85
Coal loadout 14.28
38.42
Access road
Access road 0.24
0.24
156
-------
I
I
I
I
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
O
>E
T3 E
*r™
3:
in
•ft-
O O O
o o o
in CM O
o
in
CM
CO
CO
CM
CM
m
ID
o o
O in
O CM
ro ro
ro ro
CM CM
CM CM
o o
ro ro
01 CT>
ro ro
ro ro
CM CM
CM CM
o o
ro ro
o
O
o o
o o
ro in
o
o
ro
CO O r-t
CM
r-t O
CM
m
CM
CM
o
t—i
I—
i/)
HH
UJ
cu re
(J 4J
S- C
3 CU
O I/)
oo cu
Q.
CU
s-
cu
E
re 3 re
cu •— cu
S- 0 i-
^£
cu
E
CU i— CU
s- o t.
^C ^^ ^1
cu
E
re 3 re
cu •— cu
s- o s_
4^ t- *r~
•^ ^~
Q. r— •r-
3 U
re re
.c: <*-
in
•D CU
re ••-
o •*->
4^ ^ °t~
•r~ r^
Q- i— ••-
3 0
re re
T3
re
•a o cu
re i- ••-
o •«->
+j t- -^
•r* V) r-*
Q. i— CU -i-
3 (J U
re u re
x: re M-
•o
re i/>
-a o cu
re t- "-
o +*
+J t- W -^
•i- in i—
a_ i— cu •!-
3 «J (J
re u re
x: re «»-
> re
•r- CO
Of
.*
c t-
cu o
cu u_
t-
CJ
S-
3
O
CM
in
re
t. -r-
cu in
> re
T- CO
ce.
^
c t-
cu o
cu u_
S-
C3
157
-------
•.- W
JD
to
W
X3 TJ
C 0) (A O
0. E in
in
O
-3-
in
CM
O
•a-
o
CM
O
vO
m
m
sr
m
in
m
m
o
m
o
eo
o o
o
ON
o
in
o
vo
o
m
o
vo
o
CO
o o
00 r-
o
oo
in
re
o
. C T-
tr- 4J V>
in u OJ
• ID OJ QJ
m t- t-
o
m
m
ce •!-
m
sr
oo
oo
o o
m m
Psl
m
in
m
sr
CM CO -3- m
in in -» sr
CJ
O
O
Q£
O
s-
o
01
C
'5
ZD
O
I
O
•r- O
in -P in
CO O 0>
OJ Q>
C (- (-
ro -i- o)
3 TJ O)
O T3
TJ
C C
ro -i-
CM
CO
CM
vO
CM
00
CN
•3-
CO
CO CM
ro
CO
S- (J OJ
QJ OJ OJ O
> t- t- o
QC TJ OJ
•o
t~ TJ
OJ C
TJ •!-
O
OL.
vD
00
oo
CO
CO
CO
co
vD
co
CN
CO
o
o
r-. o\
ON 00
r-
oo
O O P^- CO P*»
O> 00 CO CO CO
co oo co
CO
co
t-
=j
O
CM CO
in vo
CO
CM CO
in vo
CO
O
CM
rH CM
CM CM
CO
CM
158
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO,
EPA 450/4-83-004
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE ANDSUBTITLE
Characterization Of PM,jo And TSP Air Quality Around
Western Surface Coal Mines
5. REPORT DATE
August 1982
6. PERFORMING ORGANIZATION CODE
7. AUTHORIS)
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Pedco Environmental, Incorporated
Kansas City, Missouri 64108 and
TRC Environmental Consultants
Englewood, Colorado 80111
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-3512
12. SPONSORING AGENCY NAME AND ADDRESS
Monitoring And Data Analysis Division
Office Of Air Quality Planning And Standards
Office Of Air, Noise And Radiation
U.S. Environmental Protection Agency
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
EPA Project Officer:
Thompson G. Pace
16. ABSTRACT
This document is directed to those managers and technical staff of coal
industry and air pollution regulatory agencies needing information on the
general impact of surface mines on ambient particulate matter concentrations.
The document addresses both PM10, which is a measure of particles generally
smaller than 10 yra by a sampler with a 50% efficiency at 10 ym, and Total
Suspended Particulate (TSP), as measured by a high volume sampler. Estimates
of PMiQ and TSP concentrations are developed from actual measurements, from
previous dispersion modeling and from modeling three scenarios representing a
range of mine sizes and configurations. The results are compared to the
Prevention of Significant Deterioration (PSD) regulations and the National
Ambient Air Quality Standards (NAAQS). . Both maximum controls and typical
controls are considered, and distances from mine boundaries where PSD and/or
NAAQS are exceeded are discussed.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS c. COSATI Held/Group
Air Quality Maintenance
Coal Mining
Particulate - Size
Control Strategies
18. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
21. NO. OF PAGES
170
20. SECURITY CLASS (This page)
22. PRICE
EPA Form 2220-1 (R«v. 4-77) PREVIOUS EDITION is OBSOLETE
I
------- |