United States Office of ATr Quality EPA-450/4-82-006
Environmental Protection Planning and Standards February 1982
Agency Research Triangle Park NC 27711
Air
TESTS OF THE
INDUSTRIAL
SOURCE
COMPLEX (ISC)
DISPERSION
MODEL AT THE
ARMCO
MIDDLETOWN,
OHIO STEEL MILL
-------
. I?-
EPA-450/4-82-006
Tests of the Industrial Source Complex (ISC)
Dispersion Model at the Armco,
Middletown, Ohio Steel Mill
Prepared by
J.F. Bowers, A.J. Anderson and W.R. Margraves
Prepared for
Source Receptor Analysis Branch
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
EPA Contract No. 68-02-3323
Work Assignment No. 8
H.E. Cramer Company, Inc.
University of Utah Research Park
Post Office Box 8049
Salt Lake City, Utah 84108
February 1982
-------
This report has been reviewed by the Office of Air Quality Planning and
Standards, U. S. Environmental Protection Agency, and approved for pub-
lication as received from H. E. Cramer Company, Inc., University of Utah
Research Park, P. 0. Box 8049, Salt Lake City, Utah 84108. Approval does
not signify that the contents necessarily reflect the views and policies
of the U. S. Environmental Protection Agency, nor does mention of trade
names or commercial products constitute endorsement or recommendation
for use. Copies are available for a fee from the National Technical
Information Service, 5285 Port Royal Road, Springfield, Virginia 22161.
Publication No. EPA 450/4-82-006
- ; 2-ncy
-------
ACKNOWLEDGEMENTS
Many individuals and organizations provided assistance to the
H. E. Cramer Company, Inc. during the course of the study described in this
report. We are indebted to Mr. James Dicke, Mr. Joseph Tikvart and Mr.
Edward Burt of the Source Receptor Analysis Branch, U. S. Environmental
Protection Agency for their technical guidance and their assistance in
acquiring source, meteorological and air quality data. The particulate
emissions inventory for the Armco Steel Mill at Middletown, Ohio was
developed from information provided by Armco, Inc. and PEDCo Environmental,
Inc. Mr. James Grantz and Mr. Bruce Steiner of Armco spent many hours in
assisting PEDCo and the H. E. Cramer Company in developing the source
inputs to the Industrial Source Complex (ISC) Dispersion Model, Mr. Gopal
Annamraju of PEDCo was principally responsible for compiling the detailed
emissions data and Mr. George Schewe of PEDCo assisted Mr. Annamraju in
interpreting and understanding the ISC Model's source input requirements.
The meteorological data for the 100-meter tower located 10 kilometers
southwest of the Armco Mill were provided by the Middletown Area Chamber
Foundation through its consultant, Environmental Research and Technology,
Inc. (ERT).
In addition to the authors, other staff members of the H. E.
Cramer Company, Inc. made important contributions to the preparation of
this report. We are especially indebted to Mr. Jay Bjorklund, Mr. Craig
Cheney and Ms. Margret Boes for their assistance in performing the computer
calculations. The technical illustrations were prepared by Mr. Kay Memmott
and the report was typed by Ms. Sarah Barlow, Ms. Cherin Christensen and
Ms. Bonnie Swanson.
11
-------
Section
3
4
Appendix
A
B
C
D
TABLE OF CONTENTS
Title Page
ACKNOWLEDGEMENTS 111
EXECUTIVE SUMMARY Vii
INTRODUCTION 1
1.1 Background and Purpose 1
1.2 Description of the Armco Air Quality Monitoring
Program 2
1.3 Summary of the Armco Air Quality Data 10
1.4 Factors Affecting the Accuracy of the Dispersion
Model Calculations 15
1.5 Report Organization 19
SOURCE AND METEOROLOGICAL DATA 21
2.1 Source Input Parameters 21
2.2 Meteorological Data 79
CALCULATION PROCEDURES 97
COMPARISONS OF CALCULATED AND OBSERVED PARTICULATE
CONCENTRATIONS 103
4.1 Background Particulate Concentration Estimates 103
4.2 Measures of Model Performance 116
4.3 Results of the ISCST Calculations 125
4.4 Results of the ISCLT Calculations 137
EVALUATION OF THE AIR QUALITY MONITORING NETWORK 145
CONCLUSIONS AND RECOMMENDATIONS 151
6.1 Conclusions 151
6.2 Recommendations 152
REFERENCES 155
DESCRIPTION OF THE INDUSTRIAL SOURCE COMPLEX (ISC)
DISPERSION MODEL A-l
OBSERVED PARTICIPATE AIR QUALITY DATA B-l
SUPPLEMENTARY METEOROLOGICAL DATA " C-l
PARTICULATE EMISSION RATES USED IN THE ISCLT
CALCULATIONS D-l
DETAILED RESULTS OF THE ISCST CALCULATIONS AND
STATISTICAL ANALYSES FOR THE INDIVIDUAL SAMPLE
DAYS E-l
-------
EXECUTIVE SUMMARY
BACKGROUND AND PURPOSE
During the period March through October 1980, Armco, Inc. and the
U. S. Environmental Protection Agency (EPA) conducted a cooperative par-
ticulate air quality monitoring program in the vicinity of the Armco Steel
Mill at Middleton, Ohio. The detailed particulate emissions and air
quality data available from this program provide a comprehensive set of
measurements which can be used to test the performance of the gravitational
settling/dry deposition option of the Industrial Source Complex (ISC)
Dispersion Model (EPA, 1979). The only previous tests of this ISC Model
option consisted of comparisons of model predictions with the measurements
made during three sets of field experiments in which controlled releases
were made of glass microspheres or spray droplets with significant gravi-
tational settling velocities (Bowers and Anderson, 1981). The primary
purpose of the study described in this report was to use the Armco particu-
late emissions and air quality data in the first test of the performance of
the gravitational settling/dry deposition option of the ISC Model under the
conditions for which this option was designed (particulate emissions from a
large industrial complex).
DESCRIPTION OF THE ARMCO AIR QUALITY MONITORING PROGRAM
Figure I shows the locations of the particulate air quality
monitoring sites used in the joint Armco/EPA monitoring program and Table I
gives the site names, the types of air quality monitors used at the various
sites and the normal sampling schedule. The standard high-volume (hi-vol)
samplers measured total suspended particulate concentrations, the size-
selective hi-vol samplers measured particulate concentrations for particle
diameters less than about 15 micrometers, and the dichotomous samplers
measured particulate concentrations for particle diameters less than about
2.5 micrometers, between about 2.5 and 15 micrometers and less than 15
vn
-------
FIGURE I. Map of the area surrounding the Armco Steel Mill at Middletown,
Ohio. The numbered ® symbols show the locations of particu-
late air quality monitoring sites.
Vli
-------
TABLE I
IDENTIFICATION OF THE PARTICULATE MONITORING SITES, TYPES OF
PARTICULATE AIR QUALITY MONITORING EQUIPMENT AND
NORMAL SAMPLING SCHEDULE
Site
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Site
Name
Verity School
Hook Field
SREPCO
Coke Plant
Yankee Road
Oneida School
Research Center
Wilson School
SW Ohio Steel
Oxford Road
Coil Paint
Main Gate
Reeds Yard
Lefferson Gate
Monitoring Equipment
Standard
Hi-Vol
X
X
X
X
X
X
X
X
X
X
X
X
X
Size Selective
Hi-Vol
X
X
X
X
X
X
X
X
Dichotomous
Sampler
X
X
X
X
Normal Sampling
Schedule
Every
3rd Day
X
X
X
X
X
X
X
X
X
X
X
Every
6th Day
X
X
X
-------
micrometers. Size selective hi-vol and dichotomous samplers were colocated
at the Yankee Road, Research Center, SW Ohio Steel and Lefferson Gate sites
in order to obtain independent estimates of particulate concentrations for
particle diameters less than about 15 micrometers.
CALCULATION PROCEDURES
The ISC Model consists of two computer codes, one for short-term
air quality impact analyses and one for long-term impact analyses. The ISC
Model short-term program ISCST uses sequential hourly meteorological inputs
to calculate ground-level concentration or dry deposition patterns for time
periods ranging from 1 hour to 1 year. The ISC Model long-term program
ISCLT uses statistical wind summaries to calculate seasonal and/or annual
concentration or dry deposition values. In the study described in this
report, the ISCST program was used to calculate 24-hour average particulate
concentrations for two sets of sample days. The first set of sample days
covered the period 6 May through 26 July 1980, immediately prior to the
initiation of a new fugitive dust control program at the Armco Mill. The
second set of sample days covered the period 29 July through 30 October
1980 which followed the initiation of the dust control program on 27 July
1980. The ISCLT program was used to calculate "seasonal" average parti-
culate concentrations for the two "seasons" defined by the two sets of
sample days. Meteorological inputs for the 24 sample days used in the
model calculations for the sample days before the initiation of dust
controls were developed from wind and temperature measurements made on a
100-meter tower located 10 kilometers southwest of the Armco Mill and
concurrent Dayton, Ohio International Airport cloud-cover and mixing height
data. (The Dayton Airport is 49 kilometers north-northeast of the Armco
Mill.) The 100-meter tower was struck by lightning on 2 August 1980.
Because of the damage sustained by the sensors and other equipment, full
measurement capabilities were not completely restored until 3 October 1980.
The tower meteorological data for most of the sample days during the period
-------
with dust controls are therefore incomplete and not suitable for use in
model calculations. Consequently, after consultation with the EPA Project
Officer, it was decided that Dayton Airport surface and upper-air meteoro-
logical data should be used in the ISCST and ISCLT calculations for the 32
sample days in the period with dust controls.
FACTORS AFFECTING THE ACCURACY OF DISPERSION MODEL PREDICTIONS
Because the results of this study will be used to assess the
performance of the ISC Model in predicting ambient particulate concentra-
tions, it is important to address the factors affecting the accuracy of the
dispersion model predictions. These factors are (Fox, 1981 and others):
(1) the quality of the emissions data, (2) the quality and representative-
ness of the meteorological data, (3) the quality and representativeness of
the air quality measurements, and (4) the capability of the model to
represent the natural events.
The particulate emissions inventory for the Armco Mill is perhaps
the most comprehensive and accurate particulate emissions inventory ever
developed for a large industrial source complex. It is important to
recognize, however, that the majority of the emissions from the Armco Mill
are from non-traditional (i.e., non-stack) sources. As shown by Table II,
non-traditional sources account for about 86 percent of the total emissions
from the Armco Mill during the period before dust controls and for about 77
percent of the total emissions during the period with dust controls. Par-
ticulate emissions from non-traditional sources are difficult to quantify,
especially on an hour-by-hour rather than average basis. As discussed in
detail in Section 2.1 in the main body of the report, PEDCo Environmental,
Inc. classified the particulate emission rates for the Armco Mill into four
reliability categories. The column at the far right of Table II shows the
reliability of the emissions data estimated by PEDCo for the various types
of Armco sources. According to Table II, the reliability of the emissions
data for most of the non-traditional Armco sources is judged to be fair to
poor.
-------
CO pti
2 O
O U
M W
co H
w •<
H CC
< I
rJ W
u o
h-l M
H H
^C ^i
[\y ^4
O P
O
K
O
o
os
O
4J
•H
i-H
•H
ft
CO
•H
r-l
0)
Pi
CO
C
O
•H
CO
CO
•H
E
W
9"
>5
>-i
4J
01
E
CO
Q
01
r-J,
C
•H
4J
V.
cc
PH
t.
C
CO
Ol
4_l
CO
E
•H
4-J
CO
K
^
4J
0
H
O
CO
A
O
CO
in
r— 1
iA
i-H
I
•
CM
m
CNI
V
a.
3
•j
3
a
j
s
n
CO
rH
0
4J
C
O
u
4J
CO
3
Q
01
r-l
O
m
01
CQ
""O
O
•H
CD
P-i
CO
M
O
0
PH r-l M r-l r-l
O O O O
O O O O O
4J PH PH P_t PH
4J O O O O
C 4J 4J 4J 4-1
0)
rH *O r-l r-l r-l
tH O 1-1 -H -H
0) O CO CO cc
U O Pt4 &H &H
S^
w
o sr ^ OC
C 6C
CNI
^
^
^j
O
f— 1
CO
t-H
Q
4_)
fj
O
4J
CO
3
C)
r".
4J
•H
^
"O
O
•H
r-l
&1
'-»
jfj-
o
o
PH rH r-l rW r-l
O O O O
O O O O O
4J PH PH PH PH
4-1 O O O O
C! 4-1 4J 4-1 4J
eu
r-l 13 J-l r-l r-l
•-I C -H -H -i-l
Q) O cO CO cO
U U Cu b (ZH
^
w
O OO 00 vO OO
OO ro CNI ON O
CM ^}- ^-1 r^
•-i r-. O O O
rH -* O O O
r-H
CNI CNI !-- in CNI
i— i -3- ro
4J E
en c
3 0
"e «8
o
C_) CO
cc
ofl OJ
CJ
(T. O K
CO rH 1) CO
0) P. i— 1 u
cj — 1 O
O 4-1 BH 1-^
IH C CO
PM TH « >-. OC
c oo co c
±J p , c^3 5 *^
C 1 -H T3 .*
TH C C CO r-
O O 4J O CO
P- 2 CO PC ft,
o
o
0
OO
in
r— 1
00
CO
r-H
m
•
-a-
rn
oo
m
i_3
^
H
C
f-H
xn
-------
No onsite meteorological data are available for the Annco Mill
during the period May through October 1980. Comparison of concurrent wind
data from the 100-meter meteorological tower and the Dayton Airport shows
that the hour-to-hour correspondence between the wind directions at the two
sites often is poor, a result that is expected because of the distance from
the tower to the airport and other factors. Also, the 24-hour average
surface wind speed at the Dayton Airport typically is about twice the
24-hour average wind speed at the 10-meter level of the tower. We do not
know of any explanation for this difference in mean wind speeds. Many of
the sample days were days with light and variable winds and/or days with
thunderstorms and rain showers. The wind data from the 100-meter tower or
the Dayton Airport are least likely to be representative of conditions at
the Armco Mill on these days.
Inspection of the air quality data for the sample days reveals
uncertainties in the concentration measurements. For example, the 24-hour
average particulate concentrations measured by the size-selective hi-vol
samplers for particle diameters less than 15 micrometers typically are
about 10 micrograms per cubic meter higher than the corresponding concen-
trations measured by the dichotomous samplers. Also, there are large
uncertainties in our estimates of the "background" particulate concentra-
tions, the ambient particulate concentrations attributable to emissions
from all sources (natural and anthropogenic) not included in the model
calculations. The background concentrations were subtracted from the
observed concentrations prior to performing the comparisons of calculated
and observed concentrations. Thus, our assessment of the performance of
the ISC Model is critically dependent on the background concentration
estimates. The only objective basis we found for estimating the background
concentration on a sample day was to assume that the 24-hour average
concentration at the monitoring site least likely to have been affected by
emissions from the Armco Mill during the day was representative of a
uniform background concentration over the entire Middletown area. Thus,
the background concentration estimates do not account for the actual
xm
-------
spatial variations in background concentrations and, in many cases, are
likely to include the effects of some emissions from the Armco Mill.
The only previous tests of the ISC Model's gravitational
settling/dry deposition option consisted of the application of the ISCST
program to three sets of field experiments (Bowers and Anderson, 1981).
These experiments, which included detailed onsite meteorological measure-
ments, involved controlled releases of glass microspheres or spray droplets
with diameters ranging from about 10 to 200 micrometers. Bowers and
Anderson (1981) concluded from a comparison of calculated and observed
deposition values for the three experiments that ISCST has an approximate
factor-of-two accuracy for particles with appreciable gravitational
settling velocities. The same accuracy should be expected to apply in the
case of the Armco data set for sample days which satisfy the model's
assumptions of steady-state meteorological conditions and no precipitation.
As noted above, winds on many of the sample days were highly variable.
Also, significant precipitation occurred on many of the sample days.
MEASURES OF MODEL PERFORMANCE
A current major topic of discussion addresses the question of
what specific measures of performance should be applied in testing atmo-
spheric dispersion models. EPA has entered into a cooperative agreement
with the American Meteorological Society (AMS) to obtain guidance on the
development and application of dispersion models. To assist in the identi-
fication of possible measures of model performance, the AMS conducted a
Workshop on Dispersion Model Performance at Woods Hole, Massachusetts on
8-11 September 1980. The summary report on the Workshop (Fox, 1981) pro-
vides several potential techniques for measuring model performance and
recommends that they be tested using field data to see if they are suitable
measures of model performance. Consequently, EPA specified that the
measures of performance proposed by the AMS Workshop be used in this study.
XIV
-------
The measures of dispersion model performance proposed by the AMS
Workshop are principally based on the differences between observed and
predicted concentrations. The first measure of performance is the bias
(average difference between observed and calculated concentrations) which
is defined as
AX
N
N
AX,
(1)
AX,
Xoi ~ Xci
(2)
where x • is the i observed concentration, x • is tne i calculated con-
centration and N is the number of paired observed and calculated concentra-
tions. The second measure of performance is a measure of the "noise" in
2
the results of the model calculations and is provided by the variance (a )
of the differences
N
N
(AX1 ~ Ax)'
(3)
The third measure of performance is a measure of the gross variability of
the differences and is given by the root mean square (RMS) error
RMS =
N
1/2
(A)
An additional measure of performance suggested by the AMS Workshop is the
linear correlation coefficient between observed and calculated concentra-
tions. The absolute value of the correlation coefficient ranges from zero
(no correlation) to unity (perfect correlation). The correlation coeffici-
ent is an indicator of the degree to which changes in the magnitude of the
model predictions are linearly related to changes in the magnitude of the
XV
-------
observations. The correlation coefficient is insensitive to the rate of
change and is not a measure of absolute accuracy.
The AMS Workshop suggested that the proposed measures of perform-
ance be applied to: (1) maximum or total fields of observed and calculated
concentrations paired in space and time, (2) maximum observed and calcu-
lated concentrations paired in time only, and (3) maximum observed and
calculated concentrations paired in space only. The measures applied to
concentrations paired in space and time, whether they consider all concen-
trations or only maximum concentrations, are the most rigorous tests of
model performance. The measures applied to maximum concentrations paired
in time only are based on the premise that the model can predict the
magnitude of the maximum concentration during any time period with greater
accuracy than it can predict the location of the maximum concentration. A
major difficulty in applying this premise to the Armco data set is that the
limited number of monitoring sites and receptors is inadequate to determine
the actual locations of the maximum observed and calculated concentrations.
The measures applied to maximum concentrations paired in space only are
based on the hypothesis that, for a long period of record, the maximum
concentration predicted by the model at a specific location at any time
within the period of record should equal the maximum concentration observed
at that location during the period of record. This hypothesis assumes that
the maximum observed and calculated concentrations at a specific location
are not paired in time because of factors such as uncertainties in model
inputs and the stochastic nature of atmospheric turbulence. The principal
problem with this hypothesis as applied to the Armco data set is the
limited number of sample days (i.e., the short period of record). Concen-
trations paired in space and time, in time only and in space only were
compared in the study described in this report. However, because of the
limited number of sample days and the limited number of monitoring sites,
only total fields of differences for concentrations paired in space and
time are considered in this Executive Summary.
XVI
-------
RESULTS OF THE ISCST CALCULATIONS
Table III summarizes, for the sample days before and after the
initiation of fugitive dust controls at the Armco Mill, the statistical
analyses of total fields of differences between observed (minus background)
and calculated 24-hour average particulate concentrations. The particle-
size categories used in Table III are: (1) particle diameters less than
about 100 micrometers (total suspended particulates), (2) particle diam-
eters less than 15 micrometers, (3) particle diameters less than 2.5
micrometers, and (4) particle diameters between 2.5 and 15 micrometers.
Observed Category 2 is separated into Category 2S for concentrations .
measured by size-selective hi-vol samplers and Category 2D for concentra-
tions measured by dichotomous samplers. Two sets of concentrations were
calculated for the first size category. The concentrations calculated for
Category 1A include the effects of gravitational settling and dry deposi-
tion, while the concentrations calculated for Category IB do not include
these effects. The concentrations calculated for Category IB are represen-
tative of the concentrations that would be calculated using the modeling
techniques recommended for application to particulate sources in the
current Guideline on Air Quality Models (EPA, 1978).
We believe that the comparisons of observed (minus background)
and calculated 24-hour average total suspended particulate concentrations
(Category 1 in Table III) provide the best indication of the performance of
ISCST because the largest number of paired samples are found in this
category. Also, the results of the ISCST calculations for particle-size
categories other than the first category are qualitatively the same as the
results for the first category. Consequently, only the comparisons of
observed (minus background) and calculated total suspended particulate
concentrations are discussed in the following paragraphs.
The biases of the observed (minus background) and calculated
total suspended particulate concentrations in Table III show that: (1) For
the sample days before fugitive dust controls, ISCST overpredicts the
xvii
-------
TABLE III
STATISTICAL ANALYSES OF TOTAL FIELDS OF DIFFERENCES BETWEEN OBSERVED
AND CALCULATED 24-HOUR AVERAGE PARTICULATE CONCENTRATIONS
PAIRED IN SPACE AND TIME
Observed (- Background)
Size Category
Calculated Size Category
*
1
1A
IB
2S
2
2D
2
3
3
4
4
(a) Sample Days Before Dust Controls
No. of Paired Samples
Bias (yg/m3)
O £
Variance (yg /m )
RMS Error (yg/m3)
Correlation Coefficient
229
-60.7
9,799.3
115.9
0.558
229
-79.4
16,527.3
150.8
0.527
136
-46.1
6,137.3
90.6
0.420
81
-45.2
4,477.6
80.4
0.576
82
-17.9
897.4
34.8
0.379
81
-26.8
1,496.2
46.9
0.624
(b) Sample Days After Dust Controls
No. of Paired Samples
3
Bias (yg/m )
7 f\
Variance (yg /m )
RMS Error (yg/m3)
Correlation Coefficient
291
-11.6
1,202.6
36.5
0.576
291
-18.1
1,766.7
45.7
0.558
199
-10.4
637.4
27.2
0.516
83
-9.0
523.9
24.5
0.387
87
-4.7
131.5
12.3
0.328
83
-4.3
157.8
13.2
0.364
The comparisons of observed (minus background) and calculated concentrations
for Category 1A consider the effects of gravitational settling and dry depo-
sition, while the comparisons for Category IB do not consider these effects.
xvm
-------
impact of Armco emissions by an average of 61 micrograms per cubic meter if
the effects of gravitational settling and dry deposition are considered and
by 79 micrograms per cubic meter if these effects are not considered; and
(2) For the sample days with dust controls, ISCST overpredicts the impact
of Armco emissions by an average of 12 micrograms per cubic meter if the
effects of gravitational settling and dry deposition are considered and by
18 micrograms per cubic meter if these effects are not considered. Thus,
the average model performance for the period with dust controls is better
than for the period before controls. This result could possibly be ex-
plained by the use of meteorological data from different locations in the
ISCST calculations for the two periods. However, we believe that the
100-meter meteorological tower data used in the calculations for the period
before dust controls are more likely to be representative of conditions at
the Armco Mill than the Dayton Airport meteorological data used in the
calculations for the period with dust controls.
To gain insight into the reasons for the differences in average
model performance for the two sets of sample days, we examined the spatial
variations in the differences between observed (minus background) and
calculated concentrations. The average overprediction for the entire
monitoring network for the period before dust controls is principally
determined by large overpredictions at monitoring sites internal or immedi-
ately adjacent to the Armco Mill (Sites 3, 4, 5, 12, 13 and 14 in Figure
I). Also, the large overpredictions for the sample days before dust
controls are primarily determined by roadway emissions. As shown by Table
II, the roadway emissions account for 52 percent of the total emissions
estimated for the Armco Mill before dust controls. The Armco dust control
program consists of wetting, when necessary, the paved and unpaved roads
and storage piles. It is therefore important to note that significant
rainfall occurred in the Middletown area during the period before dust
controls. Additionally, the high relative humidities (50 to 80 percent)
and the cloudy skies (average cloud cover of 60 percent) during this period
precluded the rapid evaporation of moisture from the fugitive dust sources
during the intervals between the periods with precipitation. We conclude
XIX
-------
that the apparent poor model performance during the period before dust
controls is primarily attributable to the use in the ISCST calculations of
roadway and storage pile emission rates that were too large, especially in
view of the occurrence of rainfall and high humidities during the period.
This conclusion is supported by two additional facts. First, the
differences in the particulate concentrations measured before and after the
addition of dust controls are small and not statistically significant.
Second, because of the occurrence of rainfall and high humidities during
the first month of the dust control program, it frequently was not
necessary for Armco to wet the fugitive dust sources. It follows that the
dust controls were in fact present during the first sampling period because
of the rainfall and high humidities.
It is important to recognize that uncertainties in the background
particulate concentration estimates probably result in an apparent bias of
ISCST toward overestimation of the impact of emissions from the Armco Mill.
In most cases, the background concentration estimates include some effects
of Armco emissions. These background concentrations were subtracted from
the observed concentrations to obtain estimates of the concentrations
attributable to Armco emissions. Thus, the observed (minus background)
concentrations for most of the sample days probably underestimate the
actual impact of Armco emissions. It follows that comparisons of the
observed (minus background) concentrations with the corresponding concen-
trations calculated by a "perfect" model with "perfect" source and meteoro-
logical inputs should indicate an apparent bias toward overestimation. (We
do not know of any objective basis for quantifying the expected average
overestimation arising from the uncertainties in the observed (minus back-
ground) concentrations.) Additionally, because the effects of precipi-
tation scavenging were not included in the model calculations, the model
should tend to overestimate concentrations on sample days with significant
precipitation, especially at the more distant monitoring sites.
The variances in Table III show that the use of ISCST's gravita-
tional settling/dry deposition option reduces the noise in the results and
XX
-------
the RMS errors show that this option improves the model's absolute accu-
racy. The differences between the correlation coefficients for the sets of
ISCST calculations which did and did not consider the effects of gravita-
tional settling and dry deposition (see Categories 1A and IB in Table III)
are not statistically significant. Although the use of ISCST's gravita-
tional settling/dry deposition option improves the model's performance for
both the sample days before dust controls and the sample days with dust
controls, this improvement is small in comparison with the improvement in
model performance between the sample days before and after controls. As
noted above, we believe that the improvement in model performance between
the sample days before and after controls is primarily explained by the
differences in the emission rates assumed for the roadways and storage
piles.
RESULTS OF THE ISCLT CALCULATIONS
There are several features of the ISCLT "seasonal" average
concentration calculations described in this report that should be
considered in interpreting the comparisons of observed (minus background)
and calculated concentrations for the two "seasons" defined by the sample
days before and after the initiation of fugitive dust controls at the Armco
Mill. ISCLT divides the area surrounding a point source into sectors of
equal angular width corresponding to the sixteen standard 22.5-degree
wind-direction sectors and partitions the seasonal emissions according to
the wind-direction frequencies. If there are sufficient occurrences of
each combination of wind-direction, wind-speed and stability categories to
generate a representative joint frequency distribution and if all wind
directions within each wind-direction sector are assumed to be equally
probable, it can be shown that the horizontal distribution of emissions
within each sector is uniform (Calder, 1971). For sixteen wind-direction
sectors, six wind-speed categories and six stability categories, there are
576 combinations of wind-direction, wind-speed and stability categories.
Although some of these combinations do not occur, the number of potential
XXI
-------
combinations is exactly equal to the number of hours in the set of sample
days before dust controls. Thus, the number of hours in the first "season"
may be too small to derive a representative joint frequency distribution of
wind-speed and wind-direction categories, classified by stability cate-
gories. The same reasoning also applies to the 768 hours which define the
"season" with dust controls. We have no basis for determining the repre-
sentativeness of the wind frequency distributions used in the ISCLT calcu-
lations.
Table IV summarizes, for the "seasons" before and after the
initiation of fugitive dust controls, the statistical analyses of total
fields of differences between observed (minus background) and calculated
"seasonal" average particulate concentrations. The results of the ISCLT
concentration calculations for all particle-size categories are qualita-
tively the same as the results of the corresponding ISCST concentration
calculations. Because the comparisons of observed (minus background) and
calculated total suspended particulate concentrations are based on data
from more monitoring sites than are the comparisons for the other particle-
size categories, the following discussion focuses on the comparisons of
observed (minus background) and calculated total suspended particulate
concentrations.
The biases of the observed (minus background) and calculated
total suspended particulate concentrations in Table IV show that: (1) For
the "season" before fugitive dust controls, ISCLT overpredicts the impact
of Armco emissions by an average of 80 micrograms per cubic meter if the
effects of gravitational settling and dry deposition are considered and by
142 micrograms per cubic meter if these effects are not considered; and (2)
For the "season" with dust controls, ISCLT overpredicts the impact of Armco
emissions by an average of A micrograms per cubic meter if the effects of
gravitational settling and dry deposition are considered and by 12 micro-
grams per cubic meter if these effects are not considered. As explained in
the discussion of the ISCST results, we believe that the average ISCLT
performance for the period with dust controls is better than for the period
XX11
-------
TABLE IV
STATISTICAL ANALYSES OF TOTAL FIELDS OF DIFFERENCES BETWEEN OBSERVED AND
CALCULATED "SEASONAL" AVERAGE PARTICIPATE CONCENTRATIONS PAIRED IN
SPACE AND TIME
Observed (- Background)
Size Category
Calculated Size Category
*
1A
IB
2S
2
2D
2
3
3
4
4
(a) "Season" Before Dust Controls
No. of Paired Samples
0
Bias (yg/m )
7 f\
Variance (yg /m )
RMS Error (yg/m3)
Correlation Coefficient
12
-79.8
5,290.4
105.9
0.930
12
-141.5
16,758.2
188.1
0.941
7
-70.7
2,373.6
83.9
0.820
4
-65.5
2616.8
79.1
0.760
4
-27.8
404.2
32.8
0.052
4
-37.8
973.9
46.5
0.725
(b) "Season" After Dust Controls
No. of Paired Samples
Bias (yg/m )
7 (\
Variance (pg /m )
RMS Error (yg/m3)
Correlation Coefficient
13
-4.4
109.4
11.0
0.870
13
-12.1
229.4
18.9
0.869
8
-11.1
74.6
13.7
0.804
4
-11.2
196.8
16.5
0.189
4
-6.0
31.5
7.7
0.003
4
-5.1
73.6
9.0
0.276
The comparisons of observed (minus background) and calculated concentrations
for Category 1A consider the effects of gravitational settling and dry depo-
sition, while the comparisons for Category IB do not consider these effects.
xxm
-------
before controls primarily because the roadway and storage pile emission
rates used in the model calculations for the period before controls were
too large, especially because of the significant rainfall and high
humidities.
All of the measures of model performance in Table IV indicate
that the use of ISCLT's gravitational settling/dry deposition option
improves the overall correspondence between observed (minus background) and
calculated concentrations. Also, for the period with fugitive dust con-
trols, the observed (minus background) and calculated "seasonal" average
concentrations agree within about a factor of two at ten of the thirteen
sites where total suspended particulate concentrations were measured, if
the effects of gravitational settling and dry deposition are included in
the ISCLT calculations, and at eight of the thirteen sites if these effects
are not included in the model calculations. If the gravitational
settling/dry deposition option is exercised in the calculations for the
second "season", ISCLT underestimates the average observed (minus
background) concentrations only at Site 2 (adjacent to a taxiway at the
municipal airport), Site 7 (the Armco Research Center in downtown
Middletown) and Site 12 (the Main Gate at the northern end of the Armco
Mill). We believe that these underpredictions are explained by the effects
of emissions from localized background particulate sources.
We concluded in the discussion of the results of the ISCST con-
centration calculations that, given "perfect" model inputs and a "perfect"
model, there should be an apparent bias toward overestimation attributable
to the uncertainties in the background particulate concentration estimates.
Additionally, the neglect of the effects of precipitation scavenging in the
model calculations should bias the model predictions toward overestimation,
especially at the more distant monitoring sites. These factors should also
result in a tendency for ISCLT to overpredict concentrations for the Armco
data set.
XXIV
-------
CONCLUSIONS
The principal conclusion of the study described in this report is
that the use of the ISC Model's gravitational settling/dry deposition
option yields calculated total suspended particulate concentrations that
are in better agreement with the corresponding observed (minus background)
concentrations than the concentrations calculated without consideration of
the effects of gravitational settling and dry deposition. Although the
differences in the measured average ambient particulate concentrations
between the sample days before and after the initiation of the fugitive
dust control program at the Armco Mill are relatively small, there are
significant differences in the concentrations calculated for the two
periods by the ISC Model computer programs ISCST and 1SCLT. The differ-
ences in the concentrations calculated for the two periods are primarily
explained by the different emission rates assigned to the two periods for
the storage pile and roadway sources. In retrospect, we believe that
emission rates for the uncontrolled storage pile and roadway sources
comparable to those used for the controlled storage pile and roadway
sources should also have been used in the model calculations for the period
before dust controls to account for the effects of rainfall and high
humidities during this period. Consequently, we believe that conclusions
about the accuracy of ISCST and ISCLT should be based primarily on the
results of the calculations for the period with dust controls.
Because of the uncertainties in the emissions data, the meteoro-
logical data and the observed (minus background) particulate concentrations
as well as other limitations such as sample size, it is not possible to
evaluate the absolute accuracy of the ISC Model as a particulate dispersion
model. However, the apparent bias toward overestimation found in the
results of both the ISCST and ISCLT concentration calculations is expected
because of the uncertainties in the observed (minus background) concen-
trations as well as the neglect in the model calculations of the effects of
precipitation . If allowance is made for the expected bias toward over-
estimation, we believe that the average overpredictions by ISCST and ISCLT
XXV
-------
are relatively small and, in combination with the results of tests of
ISCST's gravitational settling/dry deposition option described by Bowers
and Anderson (1981), support the use of the ISC Model as a particulate
dispersion model.
XXVi
-------
SECTION 1
INTRODUCTION
1.1 BACKGROUND AND PURPOSE
Armco, Inc. and the U. S. Environmental Protection Agency (EPA)
cooperated in a detailed air quality monitoring program for particulate
matter in the vicinity of the Armco Steel Mill at Middletown, Ohio during
the period March through October 1980. Measurements of particulate concen-
trations for particles in one or more size categories were made at fourteen
monitoring sites every third or sixth day throughout this period. Addition-
ally, wind and temperature measurements were made on a 100-meter meteoro-
logical tower located approximately 10 kilometers southwest of the Armco
Mill during the period May through October 1980. (The tower was struck by
lightning on 2 August 1980 and did not resume full operations until 3 Octo-
ber 1980.) On 27 July 1980, Armco initiated a new program to control fugi-
tive dust emissions from roadways and storage piles at the mill. Thus, the
period May through October 1980 may be divided into a 3-month period without
fugitive dust controls (1 May through 26 July 1980) and a 3-month period
with fugitive dust controls (27 July through 31 October 1980).
The Armco particulate emissions and air quality data for the
period May through October 1980 provide a data set that may be used to test
current dispersion modeling techniques for particulate matter. The dis-
persion models recommended for application to particulate sources in the
Guideline on Air Quality Models (EPA, 1978) do not consider the effects on
ambient particulate concentrations of gravitational settling and dry deposi-
tion. However, the Industrial Source Complex (ISC) Dispersion Model (EPA,
1979) is capable of considering these effects. The primary purpose of the
study described in this report was to use the Armco data set to determine
whether the gravitational settling/dry deposition option of the ISC Model
-------
improves the model's performance in calculating ambient particulate concen-
trations over that of similar modeling techniques that do not incorporate
these effects.
The ISC Model, which is briefly described in Appendix A, consists
of two computer programs, one for short-term impact analyses and one for
long-term impact analyses. The ISC Model short-term program ISCST is de-
signed to use sequential hourly meteorological inputs to calculate ground-
level concentration or dry deposition patterns for time periods ranging
from 1 hour to 1 year. The ISC Model long-term program ISCLT is designed
to use statistical summaries of wind speed and wind direction, classified
according to the Pasquill stability categories, to calculate seasonal and/
or annual concentration or dry deposition values. In the study described
in this report, the ISCST program was used to calculate the 24-hour average
particulate concentrations at the various air quality monitoring sites for
each sample day and the ISCLT program was used to calculate "seasonal aver-
age" concentrations at the various monitoring sites for the period before
fugitive dust controls and for the period with fugitive dust controls.
(The original ISCST program was modified to accept sequential hourly emis-
sion rates for use in this study.) The ISCST and ISCLT total suspended
particulate concentration calculations were performed both with and without
the use of the gravitational settling/dry deposition option as a test of
the performance of this unique ISC Model feature.
1.2 DESCRIPTION OF THE ARMCO AIR QUALITY MONITORING PROGRAM
Figure 1-1 is a topographic map of the area surrounding the Armco
Steel Mill at Middletown, Ohio. Middletown is located in the relatively
flat plain of the Miami River Valley, which has an approximate north-
northeast to south-southwest orientation. The numbered filled circles in
Figure 1-1 show the locations of the particulate air quality monitoring
sites used in the detailed air quality monitoring program. With the ex-
ception of Site 8, all of the air quality monitoring sites are at about
-------
4375
-M370
FIGURE 1-1.
Topographic map of the area surrounding the Armco Steel Mill
at Middletown, Ohio. Elevations are in feet above mean sea
level and the contour interval is 50 feet (15 meters). The
numbered filled circles show the locations of the particulate
air quality monitoring sites.
-------
the same elevation as the Armco Mill. Site 8 is approximately 45 meters
above the plant grade elevation of the Armco Mill.
The air quality monitoring sites shown in Figure 1-1 are identi-
fied by name and operator in Table 1-1, which also gives the Universal
Transverse Mercator (UTM) X (east-west) and Y (north-south) coordinates of
the sites. The site operators are: (1) Armco, (2) the Ohio Environmental
Protection Agency (Ohio EPA), (3) the U. S. Environmental Protection Agency
(U. S. EPA) and its contractor PEDCo Environmental, Inc., (4) the South-
western Ohio Air Pollution Control Agency (SWOAPCA), and (5) Butler County,
Ohio and its consultant Environmental Research and Technology, Inc. (ERT).
The following types of particulate air quality monitoring equipment were
used during the detailed air quality monitoring program:
• A standard high-volume (hi-vol) sampler, which collects
particles with diameters less than about 100 micrometers
• A size-selective-inlet hi-vol sampler, which collects
particles with diameters less than about 15 micrometers
• A dichotomous sampler, which collects particles with
diameters less than about 2.5 micrometers (fine mode) and
between about 2.5 and 15 micrometers (coarse mode)
Table 1-2 gives, for each monitoring site, the sampler type or
types and the normal sampling schedule (every third or sixth day). Size-
selective hi-vol samplers and dichotomous samplers were colocated at the
Yankee Road, Research Center, SW Ohio Steel and Lefferson Gate sites.
Thus, the sum of the fine and coarse particulate concentrations measured by
the dichotomous samplers at these sites provide estimates of particulate
concentrations for particle diameters less than 15 micrometers that are
independent of the corresponding estimates provided by the colocated size-
selective hi-vol samplers. The last sample day for the SWOAPCA hi-vol
-------
TABLE 1-1
UNIVERSAL TRANSVERSE MERCATOR (UTM) COORDINATES OF PARTICULATE MONITORING
SITES IN THE VICINITY OF THE ARMCO STEEL MILL
Site
No.
V
2
3 & 30*
4
5
6
7
8
;*'-
i-o.
11
12
13
14
Site Name (Site Operator)
Verity School (Ohio EPA)
Hook Field (Ohio EPA)
SREPCO (SWOAPCA & U. S. EPA/PEDCo)*
Coke Plant (U. S. EPA/PEDCo)
Yankee Road (U. S. EPA/PEDCo)
Oneida School (Butler Co./ERT)
Research Center (U. S. EPA/PEDCo)
Wilson School (U. S. EPA/PEDCo)
SW Ohio Steel (Armco)
Oxford Road (Armco)
Coil Paint (Armco)
Main Gate (Armco)
Reeds Yard (Armco)
Lefferson Gate (Armco)
Location
UTM X (km)
727.5
724.4
725.4
724.8
724.2
724.1
723.2
726.5
723.5
727.8
723.7
724.3
725.1
726.3
UTM Y (km)
4,374.7
4,378.7
4,374.5
4,372.9
4,374.4
4,373.7
4,375.8
4,376.5
4,371.7
4,372.6
4,374.1
4,375.5
4,374.2
4,374.2
* Colocated hi-vol samplers.
-------
TABLE 1-2
TYPES OF PARTICULATE AIR QUALITY MONITORING EQUIPMENT IN THE VICINITY
OF THE ARMCO STEEL MILL AND NORMAL SAMPLING SCHEDULES
Site
No.
1
2
3 & 30*
4
5
6
7
8
9
10
11
12
Site
Name
Verity School
Hook Field
SREPCO*
Coke Plant
Yankee Road
Oneida School
Research Center
Wilson School
SW Ohio Steel
Oxford Road
Coil Paint
Main Gate
13 ! Reeds Yard
14
Lefferson Gate
Monitoring Equipment
(Particle Diameter in ym)
Standard
Hi-Vol
X
X
X*
X
X
X
X
X
X
X
X
X
X
Size-Selective
Hi-Vol
«
X
X
X
X
X
X
X
X
Dichotomous
Sampler
(<2.5,
2.5-15, <15)
X
X
X
X
Normal Sampling
Schedule
Every
3rd Day
X
X
X
X
X
X
X
X
X
X
X
Every
6th Day
X
X
X
* Colocated hi-vol samplers.
-------
sampler at the SREPCO site was 12 September 1980, the first sample day for
the colocated U. S. EPA/PEDCo hi-vol sampler.
Our information on the physical surroundings of the particulate
air quality monitoring sites is based on a combination of the UTM coordin-
ates provided for the sites by Grantz (1981b), inspection of (1977) aerial
photographs and (1974 photorevised) USGS maps, and verbal descriptions of
the Armco-operated sites provided by Armco (Grantz and Steiner, 1981).
Brief descriptions of the sites during the detailed air quality monitoring
program are as follows:
1. Verity School - The standard hi-vol sampler was in an
open area about 1.4 kilometers north of the north-
eastern section of the Armco Mill. The monitoring site
was about 100 meters east of a major four-lane divided
highway and was within 200 meters of two other major
roads (north and south of the site).
2. Hook Field - The standard hi-vol sampler was located
about 3 kilometers north of the Armco Mill and was near
a taxiway at the municipal airport. The nearest major
road was 500 meters south of the monitoring site on the
other side of a canal. The site was also within about
100 meters of a minor road for local traffic.
3 (& 30). SREPCO - The standard and size-selective hi-vol
samplers were on the roof of a single story building
(electrical supply company) in an otherwise residential
area. The site was surrounded by trees, unpaved
streets and an unpaved parking lot. The nearest major
road was about 300 meters west of the site and the
nearest particulate sources at the Armco Mill were
about 400 meters west of the site.
-------
4. Coke Plant - The standard and size-selective hi-vol
samplers were located adjacent to an employee parking
lot at the southwest corner of the main section of the
Armco Mill. A plant access road was less than 50
meters east of the site.
5. Yankee Road - The standard and size-selective hi-vol
samplers and the dichotomous sampler were located in an
open area on the western edge of the northern section
of the Armco Mill. The site was less than 50 meters
from a plant road intersection.
6. Oneida School - The size-selective hi-vol sampler was
located in an open area adjacent to the school. The
monitoring site was about 50 meters west of a major.
road and about 200 meters west of several coke, coal
and pellet storage piles at the Armco Mill.
7. Research Center - The standard and size-selective
hi-vol samplers and the dichotomous sampler were
located on the roof of a large building in a commercial
complex. The monitoring site was about 1 kilometer
northwest of the Armco Mill. A major road was about
300 meters west of the site and several local-traffic
streets were within 100 meters of the site.
8. Wilson School - The standard and size-selective hi-vol
samplers were located on top of the school building.
The monitoring site was about 2 kilometers northeast of
the Armco Mill.
9. SW Ohio Steel - The standard and size-selective hi-vol
samplers and the dichotomous sampler were located in
the middle of a grassy field. No trees, buildings or
8
-------
roads were within 50 meters of the monitoring site,
which was about 2 kilometers southwest of the main
section of the Armco Mill and about 1 kilometer west of
a slag dump.
10. Oxford Road - The standard hi-vol sampler was in a
grassy field and was well away from trees. The moni-
toring site was about 50 meters north and about 200
meters west of a major road. The eastern section of
the Armco Mill was about 700 meters north-northeast of
the site.
11. Coil Paint - The standard hi-vol sampler was in a
slag-filled area that was partially covered by vegeta-
tion. The monitoring site was about 500 meters west of
the main section of the Armco Mill and about 200 meters
south of a paved plant road with truck traffic.
12. Main Gate - The standard hi-vol sampler was in the
middle of a grassy field at the northeast end of the
Armco Mill. Although no trees, buildings or roads were
within 50 meters of the monitoring site, the site was
about 150 meters south of a four-lane divided highway
and about 70 meters north of an employee parking lot.
13. Reeds Yard - The standard hi-vol sampler was located
about 5 meters above ground level at a monitoring site
that may not have been representative of ambient air
quality because: (a) the site was on the northeast
corner of a major intersection, (b) the site was
adjacent to a railroad track, (c) a truck terminal was
east of the site, and (d) a taxi dispatching station
was north of the site. The site was at the eastern
edge of the main section of the Armco Mill.
-------
1-4. Lefferson Gate - The standard and size-selective hi-vol
samplers and the dichotomous sampler were in the middle
of a grassy field near the north edge of the eastern
section of the Armco Mill. A major road was about 100
meters north of the monitoring site and several rail-
road tracks were located about 100 meters southeast of
the site. The site was about 100 meters east of a
plant access road and about 150 meters northeast of an
employee parking lot.
1.3 SUMMARY OF THE ARMCO AIR QUALITY DATA
Appendix B lists all of the valid 24-hour average particulate
concentration measurements made in the Middletown, Ohio area during the
period March through October 1980. The period considered in this report is
restricted to May through October 1980 because no valid meteorological data
are available from the meteorological tower prior to May 1980 and because
particulate emissions at the Armco Mill arising from activities such as the
construction of new roads and parking lots during March and April 1980 are
not known. As explained in Section 2.2.1., adequate meteorological data
for use in dispersion model calculations are available for 24 sample days
during the period before the initiation of the fugitive dust control
program at the Armco Mill and for 32 sample days with dust controls in
effect. The particulate air quality data for these 24-day and 32-day
periods are summarized in Tables 1-3 through 1-5. Each table gives, for
each monitoring site, the number of valid samples (observed 24-hour average
concentrations) during each period and the average (arithmetic mean)
concentration and standard deviation for the period. Table 1-3 summarizes
the hi-vol sampler measurements (total suspended particulates or particle
diameters less than about 100 micrometers), Table 1-4 summarizes the
size-selective hi-vol and dichotomous sampler measurements (particle
diameters less than about 15 micrometers) and Table 1-5 summarizes the
10
-------
CO
I
Id
iJ
05
in
,—1
t-i
4-J
O
O
4-1
0
1-1
OJ
1-1
o
it Contr
O
0
14-1
o)
*
"O C /-^
)-i Cro
T3 4J - —
e co 60
CO -H p.
4J > v-'
to oj
o
*
0
•H
OJ 4J ^
60 coo
M 4J """•-
OJ C 60
.< CJ ^
c
O
05
M-J 0)
0 i-l
O crj
*
•0 C x-s
M Oro
*O 4-1 ~~
C ro 60
cfl -H p
4J > ^
tO QJ
O
*
c
0
•H
OJ 4J ,— N
6C WO
Vi 4J -~^
OJ C 60
> OJ p
< o *— -
c
0
u
w
14-4 GJ
C .H
P.
O cfl
OJ QJ
« e
•H (t
tO Z
OJ •
4-1 O
to
1— •
t-H
00
CM
SO
»— 1
^
vO
CM
O
vO
-
CM co -? in
CO
VO
vO
CM
m
CM
^
-------
-3-
W
PQ
H
g
w
o
I-l
H
CJ •£
a o
CJ CJ
H <
rl W
P 33
CJ H
H
H H
(6 <
p-i W
PH O
fd Pi
CO
H
CO V)
§g
O w
S^
33 H
O M
H O
fa
< W
H
O <
T p
s z
[si W
H O
H fa
CJ K
[ri «
fil CO
CO ><
I <
K P
M
l-i K
CO i-J
P-,
O <
< CO
*£. E-
fa
ee
CO
cn
i-4
4-J
c
o
CJ
4J
cn
3
P
l-i
0)
1 i .
cn
i-4
o
4-1
C
0
CJ
4-1
cn
3
P
01
M
0
14-4
ni
UJ
*
T3 C /-v
M Cro
co *H E
""O 4-J ^*>
C CO 60
CB i-l 3.
CO CU
P
*
C
O
•H
01 4-1 /-v
oc ccro
COPE
t-l 4-1 -•»
CU C 60
> eu 3.
< 0 N-'
c
o
cj
en
U-i Q)
0 i-4
' E'
O tO
25 CO
*
*U C s~^
)4 00
cfl -i-i E
C co M
CO -H 3.
4-1 > ^
CO CU
P
•X
c
o
•H
0) 4-1 /-N
60 nro
CO (H E
01 C 6C
>
CO H
CO
CU 01
4-1 E
— i re
CO £2
0) •
4-J O
•H Z
CO
CM
•
O
CO
00
*
CO
vO
^
I— 1
1
1
1
,«4
O
>
1
•H
33
M
co
c
U
ft.
tc
f^
co
^M
1
1
1
1
1
1
cn
3
C
E
o
4-1
0
U
.,-4
P
CO - O
CM 1 CM CM
^o co o
• • •
v£> 1 •— I CM
CM CM CM
CO CO CM
• • •
00 1 ^ CO
r^- \o in
CM — I CM
CM 1 CM CM
cn co
3 3
i-l O i-l O
o E o E
> o > o
1 4J 1 4-1
•HO -HO
E .e 33 -c
0 0
c/: -H co -H
CO P CO P
4-1 01
cu c cu -o
O i-l CO
o d. co a:
^-i
-3- in
o
•
00 1
CM
m
•
CO 1
VO
CM
CM 1
vO
•
—i I
CM
CM
•
^ 1
\o
O*i
•— < 1
cn
3
i-l O
o E
> c
1 4-1
•H 0
33 X
O
CO -H
CO P
Cfl i— 1
-O O
•H O
0) .C
C 0
O c/o
v^
—4 in
* •
O CO
CM --I
o*» in
• •
CM CO
m O
1 4->
;H o
"" CJ
CO -H
CO P
f.
U >-
>-i 01
CO 4-1
01 C
X 0
0) CJ
cc
r*.
o
•
i— H
CM
m
•
f— i
m
00
CM
f— 1
•
o
CM
Ot
•
m
m
CM
CM
i— i
O
>
1
•H
33
c/:
CO
c
o
CO
1—
•I-l
3
oc
1
1
1
I
1
1
CO
3
0
E
o
4-1
0
CJ
•H
P
i — i
O
O
^
CJ
CO
1 — 1
•
^3"
CM
CO
•
m
1
•H
3=
CO
CO
O
•H
r"
O
3
Cfi
CN
m
•
r-H
i—4
^3"
•
r-H
CO
^o
CM
,—4
•
m
i— i
\D
•
0
\ 4J
•H O
33 SZ
CJ
W i-l
00 P
c
c
CO
VJ CU
CU 4J
V, CO
M-I c;
01
—
4-1
0)
cu
^M
CJ
I-l
Ct
C-
CU
>_
CO
CO
c
o
CO
1-
4J
C
01
CJ
C
o
12
-------
in
l
w
w
H
CO
Z
O
M iJ
H ,-J
O
CJ
O <
u
w
W p^
H H
•J H
CJ
l-l CO
H iJ
Pi O
< pa
PJ H
Z
pa o
w o
Pw H
CO Q
CO W
in
O t—1
H O
O S3
s; fa
CJ
(—1 p£
C5 W
H
O <
>• O
pa z
o <
CJ
H pa
< o
w
pr] pQ
t—I CO
CO ><
I <
K Q
CX, CO
CC
s
CO
c
•o o .-^
(j i-ro
T3 CO •»«.
e -H 60
CO ^ 3-
4J 41 ^
(ft CO Q
u
M
•u C
C 0
0 iH
CJ <1) 4J -^
60 nro
CO M 4-1 -»»
3 4) C 60
O > HI 3.
< u ^-"
l-i C
0) O
4J CJ
< en
H-l 01
O i-H
.§•
O CO
Z CO
c
T3 O r-»
t-i t-ro
^ ro ~~**
C -H 6C
W « > 3.
O CO Q
4J
C C
c o
CJ -H
4J GO fif1"}
3 ^ 4J ^
O 01 C 60
> (U 3.
Q) < O ^
VH C
O O
IM U
CO
I4J QJ
O i— 1
o-
C cc
Z CO
in O co cr*
CM CO CM — l
O O O O
CM CM CM CM
CM CT. in 00
•—I CM CM •—>
1— ( •— 4 i— 1 f-H
CM O J CJ
Jx! ca co 4J
d c dj c
TO Pi to o
>H I) CJ
D£
in r~~-
CM in r^ oo
CM ro *^
13
-------
dichotomous sampler measurements by size category (particle diameters less
than about 2.5 micrometers and between about 2.5 and 15 micrometers).
Tables 1-3 through 1-5 show that the average particulate concen-
trations during the period with fugitive dust controls at the Armco Mill
generally were less than the average particulate concentrations during the
period before dust controls. The exceptions are the Coke Plant for the
hi-vol sampler measurements, the Yankee Road site for the dichotomous
sampler measurements and the SW Ohio Steel site for the size-selective
hi-vol sampler measurements. Although the trend was toward reduced partic-
ulate concentrations after the addition of the dust control program, the
differences in average concentrations before and after the controls are not
significant if the sample sizes and standard deviations are considered.
However, the trend of particulate concentrations at the three permanent
monitoring sites has been decreasing since control measures for fugitive
emissions were initiated at the Armco Mill in mid-1978.
It is of interest to note that the average particulate concentra-
tions for particle diameters less than about 15 micrometers measured by
the size-selective hi-vol samplers exceeded the corresponding concentrations
measured by the colocated dichotomous samplers by percentages ranging
from 2 percent (Yankee road site after dust controls) to 49 percent
(Lefferson Gate after dust controls). Grantz (1981a) compared the 1980
paired measurements from the two Armco colocated size-selective hi-vol
and dichotomous samplers and found that the size-selective hi-vol sampler
concentrations exceeded the concurrent dichotomous sampler concentrations
91.6 percent of the time; the average difference was 10 to 11 micrograms
per cubic meter. Examination of the 24-hour average particulate concen-
trations for the sample days shows that, on occasion, the concentration
for particle diameters less than about 15 micrometers measured by a
size-selective hi-vol sampler exceeded the concurrent concentrations for
particle diameters less than about 100 micrometers measured by the
colocated standard hi-vol sampler. Thus, it is evident from an examination
of the air quality data that there are uncertainties in the data.
14
-------
Wilson School (Site 8 in Figure 1-1) is the only particulate air
quality monitoring site that is at an elevation higher than the plant grade
elevation of the Armco Mill. Because the prevailing winds during the
period May through October 1980 were from the south-southwest, Wilson
School was frequently downwind of the Armco Mill. If the maximum
short-term and long-term ground-level particulate concentrations
attributable to emissions from the Armco Mill are caused by buoyant stack
emissions impacting on elevated terrain, it follows that the particulate
concentrations at Wilson School should tend to be higher than the
particulate concentrations at other monitoring sites at similar distances
from the mill. However, the average particulate concentrations at the more
distance Hook Field site (Site 2 in Figure 1-1) before and after dust
controls were 15 and 22 percent higher than the corresponding average
concentrations at Wilson School. The highest and second-highest 24-hour
average concentrations at Hook Field also exceeded the highest and
second-highest 24-hour concentrations at Wilson School. The maximum
observed 24-hour and average particulate concentrations occurred internal
to the Armco Mill at the Reeds Yards (Site 13) and Coke Plant (Site 4)
monitor locations. Thus, the air quality data indicate that the maximum
particulate concentrations produced by Armco emissions occurred within and
adjacent to the Armco property. This result is consistent with the fact
that the majority of the mill's emissions were from low-level sources (see
Section 2.1).
1.4 FACTORS AFFECTING THE ACCURACY OF THE DISPERSION MODEL CALCULATIONS
The American Meteorological Society (AMS) Workshop on Dispersion
Model Performance (see Fox, 1981) identified the following factors which
limit the accuracy of air quality model predictions:
• The quality of the emissions data
15
-------
• The quality and relevance to the modeling problem of the
meteorological data
• The quality and representativeness of the air quality
measurements
• The capability of the model to represent the physical
processes that occur
Each of these factors is discussed in detail in the appropriate section of
this report. However, because the results of the study described in this
report will be used to assess the accuracy of the ISC Model as a particu-
late dispersion model, we believe that it is important to begin with a
review of these factors.
The particulate emissions inventory for the Armco Mill is perhaps
the most detailed and accurate particulate emissions inventory ever devel-
oped for a large industrial source complex. It is important to recognize,
however, that the majority of the particulate emissions from the Armco Mill
are from non-traditional (i.e., non-stack) sources. Particulate emissions
from non-traditional sources are difficult to quantify, especially on a
hour-by-hour rather than average basis. For example, the reliability of
the particulate emission rates for the Armco sources is estimated by PEDCo
Environmental, Inc. to be good or excellent for sources whose emissions
account for less than 10 percent of the total emissions, while the relia-
bility of the emissions estimates for the sources which account for 60 to
80 percent of the.total emissions is estimated to be fair or poor (see
Section 2.1). Also, although the magnitude of particulate emissions from a
non-traditional source is often affected by meteorological factors such as
wind speed and precipitation, the Armco emissions inventory does not
consider these factors. We have no basis for quantifying the uncertainties
in the calculated concentrations arising from the uncertainties in the
Armco emissions inventory.
16
-------
No onsite meteorological data are available for the Armco Mill
during the period of the detailed air quality monitoring program. The wind
data used in the ISC Model calculations described in this report were from
a tower located 10 kilometers southwest of the mill or from an airport lo-
cated 49 kilometers north-northeast of the mill, while the mixing height
estimates were derived from upper-air soundings made at the airport (see
Section 2.2.1). Although the long-term average wind directions at the
tower, the airport and the Armco Mill are likely to be very similar, com-
parison of concurrent wind data from the tower and the airport shows a gen-
erally poor hour-to-hour correspondence between the wind directions at the
two sites. Also, the average surface wind speeds at the tower and the air-
port typically differ by about a factor of two. Thunderstorms or rain
showers occurred in southwestern Ohio on many of the sample days. The
localized winds generated by a thunderstorm or rain shower in the vicinity
of the Armco Mill can differ significantly from the concurrent winds at the
airport or tower, and vice versa. It follows that the uncertainties about
the representativeness of the meteorological data used in the ISC Model
calculations contribute to unquantifiable uncertainties in the results of
the model calculations.
We have no basis for assessing the representativeness of the par-
ticulate air quality monitoring sites. The periodic performance audits of
the air quality monitoring equipment at the various sites that were conduc-
ted under contract to EPA showed all of the monitoring equipment to be op-
erating within acceptable limits. However, we point out that it is evident
from an examination of the air quality data that there are uncertainties in
the concentration measurements (see Section 1.3). For example, the particu-
late concentrations measured by the size-selective hi-vol samplers for par-
ticle diameters less than about 15 micrometers almost always exceed the
corresponding concentrations measured by the colocated dicotomous samplers.
The uncertainties in the particulate air quality data are small
in comparison with the uncertainties in our estimates of the "background"
particulate concentrations, the particulate concentrations attributable to
17
-------
emissions from all sources (natural and anthropogenic) not included in the
model calculations. The background particulate concentrations were sub-
tracted from the observed concentrations prior to performing the compar-
isons of calculated and observed concentrations (see Section 4.1).
Consequently, the apparent performance of the ISC Model is highly dependent
on the background concentration estimates. The only objective basis for
estimating the background concentration on each sample day was to assume
that the 24-hour average concentration at the monitoring site least likely
to have been affected by emissions from the Armco Mill during the day was
representative of a uniform background concentration over the entire Middle-
town area. Thus, the background concentration estimates do not account for
the actual spatial variations in background concentrations and, in many
cases, include the effects of some emissions from the Armco Mill.
The most straightforward manner in which to test the performance
of a particulate dispersion model is to apply the model to field experi-
ments involving the controlled release of particulates that are unique to
the source or sources to be modeled. This approach overcomes uncertainties
about particulate emission rates and background concentrations. Also,
detailed onsite meteorological measurements usually are made in conjunction
with field measurement programs. Bowers and Anderson (1981) applied the
ISC Model short-term program ISCST to three field experiments involving the
release of glass microspheres or spray droplets with diameters ranging from
about 10 to 200 micrometers and concluded that ISCST has an approximate
factor-of-two accuracy for particles with appreciable gravitational
settling velocities. Implicit in this conclusion is that the ISC Model's
assumption of steady-state meteorological conditions is satisfied. Winds
on some of the sample days were light and variable. Additionally, signifi-
cant precipitation occurred on a number of the sample days. Precipitation,
which currently is not considered by the ISC Model, affects ambient
particulate concentrations by removing particulates from the atmosphere and
by wetting fugitive dust sources.
18
-------
1.5 REPORT ORGANIZATION
In addition to the Introduction, this report consists of five
major sections and five appendices. The source and meteorological inputs
used in the ISC Model calculations are discussed in Section 2, which also
contains a discussion of the climatological representativeness of meteoro-
logical conditions during the detailed air quality monitoring program. The
calculation procedures are described in Section 3 and the results of the
comparisons of observed and calculated calculations are discussed in
Section 4. An evaluation of the adequacy of the air quality monitoring
network used during the detailed monitoring program is presented in Section
5. Section 6 gives the conclusions of the study and recommendations for
future model testing. Appendix A provides a brief description of the ISC
Model. All of the air quality data collected during the period March
through October 1980 are listed in Appendix B. Appendix C lists the hourly
meteorological inputs used in the ISCST concentration calculations and the
statistical wind summaries used in the ISCLT calculations. The particulate
emission rates used in the ISCLT concentration calculations are given in
Appendix D. The detailed results of the ISCST concentration calculations
and the comparisons of observed and calculated concentrations for the
individual sample days are contained in Appendix E.
19
-------
SECTION 2
SOURCE AND METEOROLOGICAL DATA
2.1 SOURCE INPUT PARAMETERS
This section presents the particulate emissions inventory for the
Armco Steel Mill in Middletown, Ohio in the format required for input to
the Industrial Source Complex (ISC) Dispersion Model computer codes ISCST
and ISCLT. The emissions inventory, which is for the period May through
October 1980, is based on emissions data, plant layout data and other infor-
mation provided to the H. E. Cramer Company by the U. S. Environmental
Protection Agency (EPA), Armco and PEDCo Environmental, Inc.
The EPA Source Receptor Analysis Branch provided the following
material for use in developing the Armco particulate emissions inventory:
• 21 December 1979 letter to Mr. Edward Burt of EPA from Mr.
Donald R. Perander of Armco
• 28 December 1979 letter to Mr. Burt from Mr. James A. Grantz
of Armco
• A list of significant roads and parking lots at the Armco
Mill
PEDCo provided emission factors, particle size distributions and
particle densities for the various particulate sources at the Armco Mill in
the following communications:
• 30 January 1981 letter from Mr. Gopal Annarcraju of PEDCo to
Mr. Harry Geary of the H. E. Cramer Company
• 6 February 1981 letter from Mr. Annamraju to Mr. Geary
21
-------
• 10 April 1981 letter from Mr. Annamraju to Mr. Robert C.
McCrillis of EPA
• 26 August 1981 letter from Mr. Fred Hall of PEDCo to Mr,
Henry Thomas of EPA
The locations of most of the particulate sources at the Armco
Mill were indicated on the 1977 aerial photographs of the mill that accom-
panied the 21 December 1979 letter from Mr. Perander. However, the new
fugitive dust control program at the Armco Mill includes changes in traffic
and parking at the mill, and there are significant differences between the
1977 and 1980 roads and parking lots. In a 24 February 1981 meeting at the
Armco Research Center in Middletown, Mr. Grantz and Mr. Bruce Steiner of
Armco identified on the aerial photographs the locations of the current
(1980) roads and parking lots. Mr. Grantz also forwarded to the H. E.
Cramer Company the daily fuel consumption and process rates for the combus-
tion and process sources.
The particulate sources at the Armco Mill were represented for
modeling purposes by 31 stack sources, 404 volume sources and 46 area
sources for a total of 481 sources. Table 2-1 identifies the model sources
by the source numbers used in the model calculations. At the request of
the EPA Project Officer, the sources in Table 2-1 were grouped into the
following categories:
• Point combustion and process sources (Sources 110 through
132)
• Non-point combustion and process sources (Sources 10510
through 11343)
• Storage pile sources (Sources 21101 through 26102)
22
-------
TABLE 2-1
IDENTIFICATION OF MODEL SOURCES BY SOURCE NUMBER
Source Number
Source Description
(a) Point Combustion and Process Sources
-110
2IX
410
510
6IX
710
720
810
910
920
930
940
1110
1120
1130
1140
No. 1 Boiler House; Boilers 15 and 16
No. 2 Boiler House; Boilers 1, 2, 3 and 4:
211 - Fueled by blast furnace gas
212 - Fueled by coke oven gas
213 - Fueled by natural gas
214 - Fueled by No. 6 fuel oil
Slab Scarfer
Sinter Plant Stack
No. 3 Boiler House; Boilers and Slab Reheat Furnaces
1, 2, 3 and 4:
611 - Fueled by coke oven gas
612 - Fueled by natural gas
613 - Fueled by No. 6 fuel oil
No. 2 Coke Plant Combustion Stack
No. 2 Coke Plant Quench Tower
No. 3 Blast Furnace Stove Stack
Basic Oxygen Furnace; Vessels 15 and 16
Basic Oxygen Furnace; Deslagging Stack
Basic Oxygen Furnace; Hot Metal Transfer Stack
Basic Oxygen Furnace; Desulfurizer Stack
No. 2 Open Hearth; Furnaces 9 and 10
No. 2 Open Hearth; Furnace 11
No. 2 Open Hearth; Furnaces 12 and 13
No. 2 Open Hearth; Furnace 14
23
-------
TABLE 2-1 (Continued)
Source Number
Source Description
121X
12 2X
123X
1310
1320
Soaking Pits; Batteries 1, 2 and 3:
1211 - Fueled by coke oven gas
1212 - Fueled by natural gas
1213 - Fueled by No. 6 fuel oil
Soaking Pits; Batteries 4, 5 and 6:
1221 - Fueled by coke oven gas
1222 - Fueled by natural gas
1223 - Fueled by No. 6 fuel oil
Soaking Pits; Batteries 7 and 8:
1231 - Fueled by coke oven gas
1232 - Fueled by natural gas
No. 3 Coke Plant Combustion Stack
No. 3 Coke Plant Quench Tower
(b) Non-Point Combustion and Process Sources
10510 & 10520
1071X & 1072X
10810
10820
10910 & 10920
11010
11110 - 11140
Sinter Plant Fugitive Emissions
No. 2 Coke Plant Fugitive Emissions:
107X1 - Doors and Topside
107X2 - Pushing
107X3 - Charging
No. 3 Blast Furnace; West Cast House
No. 3 Blast Furnace; East Cast House
Basic Oxygen Furnace Roof Monitor
McGraw Lancing
No. 2 Open Hearth Roof Monitor
24
-------
TABLE 2-1 (Continued)
Source Number
1131X - 1132X
1133X - 1134X
Source Description
No. 3 Coke Plant; West Battery Fugitive Emissions:
113X1 - Doors and Topside
113X2 - Charging
113X3 - Pushing
No. 3 Coke Plant; East Battery Fugitive Emissions:
113X1 - Doors and Topside
113X2 - Charging
113X3 - Pushing
(c) Storage Pile Sources
21101 & 21102
21201 - 21204
21301
22101 & 22102
24101 & 24102
25101
26101 & 26102
Metallurgical Coal Piles near the No. 3 Coke Plant
Coke Piles northwest of the No. 3 Coke Plant
Steam Coal Pile southeast of the Processing Area
Pellets Piles southwest of the No. 3 Blast Furnace
Pellets Piles northwest of the No. 3 Blast Furnace
Blast Furnace Slag Piles
Flue Dust and Sludge Piles near the Basic Oxygen
Furnaces
(d) Roadway Sources
40101 - 40118
40201 - 40207
40301 - 40308
40401 - 40407
40501 - 40502
40601 - 40603
40701 - 40730
40801 - 40807
Processing Loop Road (4-5-6-7-8-9)
South Stores Receiving Road (1-2-3-4)
Scale Pit Road (4-9-10)
General Services Road (2-12-11)
Soaking Pit Road (11 - 10)
OSR Gate Road (28-27-11)
Coke Plant Loop Road (28-29-30-31-29)
BOF/Coal Pile Road (20-24-26-27)
25
-------
TABLE 2-1 (Continued)
Source
40901 -
41001 -
41101 -
41201 -
41301 -
41401 -
41501 -
41601 -
41701 -
41801 -
41901 -
42001 -
42101 -
42201 -
42301 -
42401 -
42501 -
42601 -
42701 -
42801 -
Number
40903
41007
41118
41208
41306
41408
41511
41613
41715
41817
41905
42006
42107
42208
42327
42406
42518
42618
42705
42804
Source Description
Transportation Road (25 - 26)
Mold Yard Road (14-13-12)
Reeds Yard Road (16-15-14)
OH Loop Road (19-23-22-15)
EOF Loop Road (20-21-14)
Slag Hauler Road (BOF-24-Slag Dump)
Slag Dump Road (Slag Dump-28-24-25-13)
Old Lefferson Road (18-17-16)
Lefferson /Yankee Gate Road (18 - 36)
E-W Freeway (36-35-34-33-32)
Old Muzzy Road (17-33)
Old Stores Road (33-34)
BF/BOF Road (18-19-20)
No. 1 OH Road (37 - 35)
E. Processing Road (38-32-16)
W. Processing Road (38-37)
North Perimeter Road (16 - 1)
Millers Crossing Road (41 - 31)
Recycle Plant Loop Road (39 - 40 plus loop)
Recycle Plant/B.F. Road (28-39-20)
(e) Parking Lot Sources
50101 -
50201 -
50105
50211
HSM/CSM Parking Lot and Access Road
Maintenance Administration Parking Lot and Un-shared
Portion of the Access Road
26
-------
TABLE 2-1 (Continued)
Source Number
Source Description
50262 - 50273
50301 - 50319
50401 - 50404
50501 - 50525
50601 - 50607
50701 - 50706
50801 - 50805
50901 - 50905
51001 - 51012
51101
51201 - 51202
51301
Access Road Shared by the Maintenance Administration
and Lefferson/Shops Parking Lots
Main Gate Parking Lots and Access Road
Wicoff Gate Parking Lot and Access Road
Administration/Staff Parking Lot and Access Road
Lefferson/Shops Parking Lot
Coke Plant Parking Lot and Access Road
Contractor Parking Lot and Access Road
Yankee Gate Parking Lot and Access Road
Coil Paint Parking Lot and Access Road
Miscellaneous Parking in the South Area
Miscellaneous Parking in the Steel Area
Miscellaneous Parking in the North Area
27
-------
• Roadway sources (Sources 40101 through 42804)
• Parking lot sources (Sources 50101 through 51301)
Figure 2-1 is a map of the Armco Mill that divides the mill into
four areas:
• Area N - Northern section of the mill
i
• Area S - Southern section of the mill
• Area E - Eastern section of the mill
• Area C - Central section of the mill
These four sections of the mill are shown in detail in Figures 2-2 through
2-5, which indicate the locations of the sources identified in Table 2-1.
The numbers enclosed in squares in Figures 2-2 through 2-5 are the inter-
section numbers of the roadway sources and correspond to the numbers
enclosed by parentheses in Table 2-1(d).
ISCST Source Inputs
The source inputs for the point (stack) sources are given in
Tables 2-2, 2-3 and 2-4. Table 2-2 gives all of the source inputs except
the particulate emissions parameters such as the particulate size distribu-
tion. Table 2-3 gives the particulate emission factors for the various
sources that were used to calculate the actual total particulate emissions
from each source on each sample day. For each source and particle-size
category, Table 2-4 gives the mass fraction of the particulates in the
category, the gravitational settling velocity of the particulates in the
category and the surface reflection coefficient of the particulates in the
category. The gravitational settling velocities were obtained using the
28
-------
FIGURE 2-1. Map of the Armco Steel Mill in Middletown, Ohio showing the loca-
tions of Areas N, C, E and S within the mill. Note that areas
N, S and E all overlap Area C.
29
-------
4J7560O
Map of the northern section (Area N) of the Armco Mill showing
the locations of particulate sources identified in Table 2-1.
30
-------
1320®
I3IO®\ ®II33X
®II34X
4373500
4J7I5CO
726000
FIGURE 2-3. Map of the southern section (Area S) of the Armco Mill showing
the locations of particulate sources identified by source number
in Table 2-1.
31
-------
0 100 200 300 Meters
FIGURE 2-4. Map of the eastern section (Area E) of the Armco Mill showing the
locations of particulate sources identified by source number in
Table -2-1.
32
-------
CO
o
co
o
o
4-1
00 •
C --i
•H I
£ CM
O
f (U
03 iH
&
tH CO
rH H
•H
S C
•H
O
U M
(1)
O
»-l
a
o
en
H u
CO ,J2
(U
t-i 13
< 0)
C -H
O 4J
•H C
4-> a)
o -a
CD -H
03
03
iH CD
(0 U
>-4 VJ
j-j a
a o
a) CD
o
0)
01 4-1
o
o.
3
a
nj
C.
u-i
I
OS
3
C
33
-------
W
CSJ
I
csl
•< M
H S
CO
o
•H
CO
c
0)
00
c
pa
-a E
C E
QJ v-x
60 ^>
•H E
Q) s_/
o S ^
« « E
en Q
o
*J tH CO
•H (1) ""-.
£>£
W H
CO 4J
CO X
•H 60
E -H
W QJ
ca
c
o
o
o
0)
o .
l-i O
3 2
O
V3
OrHoooooo u^mmmooooooo o
O — 1 »-H i— t i-H t-H
OCMOOOOO OOOOOOOOOOO O
omooooo iirimuiiooooooocx3f^r^r~- ir^
in in^Hu-i>-(T-i cMcsicsicMcsicNjcsicgc^csicM cs
<— lOCNcMCMcsirororomoo
^-t CSI r-H CM •— 1 I— It— I
ooooooooooooooooooooo
co ro o"i ro ro ro ro co ro co co cO co CO co CO co co CO CO CO
ooooooooooooooooooooo
CO CO CM
-* CO i—I CM CO
i-t i-4 CM CM CN
II III
O——ICM
34
-------
TABLE 2-3
EMISSION FACTORS FOR THE COMBUSTION AND
PROCESS POINT SOURCES
-Source Number
Emission Factor
110
211
212
213
214
410
510
611
612
613
710
720
810
910
920
930
940
1110
1120
1130
1140
1211, 1221 & 1231
1212, 1222 & 1232
1213 & 1223
1310
1320
1338 g/ton
0.454 g/1000 cubic feet
2.72 g/1000 cubic feet
4.54 g/1000 cubic feet
5.31 g/gallon
1.45 g/sec
4.16 g/sec
2.72 g/1000 cubic feet
4.54 g/1000 cubic feet
5.13 g/gallon
77.2 g/ton of coal
136 g/ton of coal
21.8 g/ton of hot metal
14.1 g/ton of steel
11.4 g/ton of steel
5.90 g/ton of steel
1.36 g/ton of steel
42.4 g/ton of steel
21.2 g/ton of steel
42.4 g/ton of steel
21.2 g/ton of steel
2.72 g/1000 cubic feet
4.54 g/1000 cubic feet
5.31 g/gallon
77.2 g/ton of coal
109 g/ton of coal
35
-------
i
CM
w
to
w
M
H
l-l to
O W
O O
W ;=>
> O
to
o
a H
M a
i-J H
H O
H PU
W
CO [£)
3*
a oi
o o
M tu
< CO
H H
> W
O M
fn
* 1*4
CO W
3 O
O O
£3 O
« H
M H
Q W
pd
W
N W
O
O
O
CO
2
0)
O
VJ
3
O
to
OOOOOOOOOOOOOOOOOOOOO
OOOOOOOOOOOOOOOOOOOOO
OOOOOOOOOOOOOOOOOOOOO
OOOOOOOOOOOOOOOOOOOOO
OOOOOOOOOOOOOOOOOOOOO
0) O
pi -H
O 0)
n) O
M-l U
V-i
m "1 OOOOOOOOOOOOOOOOOOOOO
• S OOOOOOOOOOOOOOOOOOOOO
CM
i-l i-H l-l r-l t-l rH
V
P 6
o\ooooNOoooooooocriOoo
OOJ
E
CD I 3. nOOasOOOOC^OOOOOOOOC^OOO
to m ____.. ^_.___ _.^_ ^_
CM r-l
OOOOOOOOOOOOOOOOOOOOO
m cMOOpppoQpoopooopopoop
CM ~
OOOOOOOOOOOOOOOOOOOOO
Or-ICMcOs3-oOt-(csiroooOO
O
o •> CM
i— I i— I CM iH
-H r-l
CM CM
36
-------
techniques given by McDonald (1960) and the surface reflection coefficients
were obtained from Figure 2-8 in the ISC Model User's Guide.
If two or more stacks with identical stack heights, stack flow
rates and stack exit temperatures were located in close proximity, the
stacks were represented for modeling purposes by a single stack provided
that the separation of the stacks had a negligible effect on the concentra-
tions calculated at the distances from the stacks to the particulate air
quality monitors. A single model stack, used to represent multiple actual
stacks, was placed at the centroid of the actual stack locations and was
assigned a particulate emission rate equal to the sum of the emission rates
for the actual stacks. The model stack exit temperature and flow rate were
identical to the values for each of the actual stacks. That is, no allow-
ance was made for possible enhancement in buoyant plume rise due to the
close proximity of the multiple stacks.
Tables 2-5, 2-6 and 2-7 list the source inputs for the volume
sources used to represent the non-point combustion and process sources.
Table 2-5 gives all of the source inputs except the particulate emissions
parameters, Table 2-6 gives the particulate emission factors for the var-
ious sources, and Table 2-7 gives, for each source and particle-size cat-
egory, the mass fraction of particulates in the category, the gravitational
settling velocity of particulates in the category and the surface reflec-
tion coefficient of particulates in the category. The non-point combustion
and process sources consist of emissions from buildings, other structures
(for example, coke ovens) and roof monitors. These sources were treated as
line sources represented by either two or four separated volume sources
(see Figure 2-10 (b) in the ISC Model User's Guide). The initial vertical
dimension (o ) is given by the building or structure height divided by
z o
2.15 and the initial lateral dimension (a ) is given by the center-to-
yo
center spacing of the sub-volume sources divided by 2.15. The spacing
between the sub-volume sources does not exceed one-third of the distance to
the nearest particulate air quality monitoring site (see page 2-56 of the
ISC Model User's Guide). The effective emission height of each sub-volume
37
-------
TABLE 2-5
SOURCE INPUTS (EXCEPT PARTICULATE EMISSIONS PARAMETERS) FOR THE
NON-POINT COMBUSTION AND PROCESS SOURCES
Source
Number
10510
10520
10711-10713
10721-10723
10810
10820
10910
10920
11010
11110
11120
11130
11140
11311-11313
11321-11323
11331-11333
11341-11343
Coordinates (m)
UTM X
724,980
725,020
724,940
724,990
724,780
724,840
725,380
725,420
724,900
725,100
725,160
725,220
725,280
725,200
725,230
725,340
725,370
UTM Y
4,373,300
4,373,270
4,373,070
4,373,040
4,373,620
4,373,580
4,373,300
4,373,270
4,372,100
4,373,750
4,373,700
4,373,660
4,373,620
4,372,880
4,372,860
4,372,780
4,372,760
Emission
Height (m)
30.5
30.5
15.0
15.0
26.9
22.1
30.5
30.5
2.0
36.3
36.3
36.3
36.3
15.0
15.0
15.0
15.0
Initial Source Dimensions
Vertical
azo(m)
14.2
14.2
7.0
7.0
8.5
8.5
14.2
14.2
1.0
16.9
16.9
16.9
16.9
7.0
7.0
7.0
7.0
Lateral
ayo(m)
14.2
14.2
14.2
14.2
22.7
22.7
23.0
23.0
2.0
35.1
35.1
35.1
35.1
14.2
14.2
14.2
14.2
38
-------
TABLE 2-6
EMISSION FACTORS FOR THE NON-POINT COMBUSTION AND PROCESS SOURCES
Source Number
Emission Factor
10510 & 10520
10711 & 10721
10712 & 10722
10713 & 10723
10810 & 10820
10910 & 10920
11010
11110 - 11140
11311, 11321, 11331 & 11341
11312, 11322, 11332 & 11342
11313, 11323, 11333 & 11343
1.67 g/sec (each)
41.5 g/ton of coal (each)
114 g/ton of coal (each)
10.0 g/ton of coal (each)
136 g/ton of hot metal (each)
114 g/ton of steel (each)
999 g/ton processed
19.1 g/ton of steel (each)
27.7 g/ton of coal (each)
7.15 g/ton of coal (each)
3.18 g/ton of coal (each)
39
-------
I
CM
Ed
ca
U
co
o
p o
z en
CO CO
Ed Ed
M pi
C_) f\.
SP
Ed Z
O Z
Z O
iJ H
H co
W§
CO S
H O
i^C pii
H I
M Z
> O
O Ed
cts
- H
CO
Z Pi
O O
H En
H
pa H
Pi Ed
H H
P En
En
Ed Ed
M O
M O
CO
Ed O
P Id
O J
M En
H Ed
Pi Pi
c
o
•H
4J 4-1
o a
i-H >H
4H U
01 -H
PS M-4
4H
01 >
•H
a
o
!-t
£
c
,-H
4-1
4J
a>
en
C
o
•H
4-1
c8
)_l
En
CO
CO
:§
J
8 e
A
, a
^H o
CO
int
CN m
*— 1
"2 e
CM 3.
V
o e
co 3.
, 1
m
^H O
I 3
m 3.
•
i-H
m
0
O
o
CM
O
O
O
0
O
m
CM
0
o
in
i— i
0
0
o
i-H
0
o
o
U1
0
T— 1
CM
1 —
I-H
•Ji
^H
r^
O
O
0
CO
vO
0
0>
r^
O
0
0
CO
-*
r-^
CM
O
VD
CM
CO
o
0
o
vO
o
o
0
CM
0
0
O
O
o
00
1^
o
0
t-~
o
o
o
o
i-H
o
o
m
o
o
CM
CM
1^
i-H
«-a
CN)
r^-
0
o
o
CO
vD
O
Oi
r-
O
O
O
CO
O
0
CM
OO
^H
c«
O
00
0 0
o o
o o
m m
m m
o o
-a- <•
r^ t~-
o o
O 0
0 O
CO CO
r^ r^
O O
-H O
i-H
V^J O
i-H
O -H
ON
o
o
0
m
m
o
-*
r-
o
o
o
CO
r~
0
O
o>
c~-
o
o
o
r-{
CO
~*
r^
CM
O
VO
CM
CO
O
O
O
VO
o
o
o
CM
0
0
o
o
o
m
CM
o
0
m
i-H
0
o
o
i-H
o
0
o
in
o
«s
r-H <— i
CM
-H rH
co co
O
O
0
CO
VO
O
cr.
r~.
O
o
o
i-H
CO
-*
r-x
CM
O
vO
CN
CO
o
o
o
vO
o
o
o
CN
0
0
o
o
0
in
-*
o
o
i-H
o
o
o
i-H
i-H
o
o
CO
-3-
o
*,
CN CN
CM <•
CO CO
i-H i-H
^
f.
CM CN
co co
o
o
0
CO
VO
o
o>
r>.
o
o
o
1— 1
CO
-------
source was assumed to be the height of the building, structure or roof
monitor.
The emission parameters for the combustion and process sources
which vary with time are given in Tables 2-8 and 2-9. Table 2-8 gives the
daily average emission rates for all of the point and non-point combustion
and process sources (except the slab scarfer and the sinter plant) for each
of the 61 sample days in the period from May through October 1980. These
emission rates were determined by multiplying the daily process production
and fuel consumption totals provided by Armco by the emission factors given
in Tables 2-3 and 2-6. Table 2-9 gives the daily total number of hours of
operation for the slab scarfer (Source 410) and the sinter plant (Sources
510, 10510 and 10520). The periods of operation were assumed to be continu-
ous beginning at 0800 EST for days with not more than 16 hours of operation
and at midnight for days with more than 16 hours. These sources were
assumed to operate at full capacity (see Tables 2-3 and 2-6 for emission
rates) during the hours indicated.
With the exception of the gravitational settling parameters (par-
ticulate size distribution, gravitational settling velocities and surface
reflection coefficients), Table 2-10 gives the source inputs for the area
sources used to represent the storage piles. The gravitational settling
parameters for the storage piles are presented in Table 2-11. For modeling
purposes, each storage pile was divided into square areas with a combined
area approximately equal to the area of the corresponding storage pile (see
Figures 2-2, 2-3 and 2-5). The effective emission height of each area
source was set equal to the height of the storage pile.
The source inputs for the volume sources used to account for the
effects of particulate emissions from the roads shown in Figures 2-2
through 2-5 are given in Tables 2-12, 2-13 and 2-14. Table 2-12 gives all
of the source input parameters except the particulate emission rates, the
particulate size distributions, the gravitational settling velocities and
the surface reflection coefficients. The average particulate emission
41
-------
TABLE 2-8
DAILY AVERAGE EMISSION RATES FOR THE COMBUSTION AND PROCESS
SOURCES FOR THE 61 SAMPLINC DAYS
Source
110
211
212
213
214
611
612
613
710
720
810
910
920
930
940
1110 & 1130
1120 & 1140
1211
1212
1213
1221
1222
1223
1231
1232
1310
1320
107X1
107X2
107X3
108X0
109X0
11010
111X0
113X1
113X2
113X3
Emission Rate* (g/sec)
3 May
0
0.650
0.208
0.084
0
0.620
0
0.029
1.559
2.747
0.692
1.001
0.809
0.419
0.097
1.093
0.546
0.104
0
0
0.139
0
0
0.089
0
4.325
6.106
0.838
2.302
0.202
4.316
8.093
0
0.492
1.552
0.401
0.178
6 May
0
0.506
0.209
0
0
0.415
0
8.803
1.512
2.663
0.561
0.860
0.695
0.360
0.083
0.796
0.398
0.143
0
0.192
0.201
0
0
0.136
0
4.205
5.937
0.813
2.233
0.196
3.502
6.951
1.919
0.359
1.509
0.389
0.173
9 May
0
0.787
0.166
0
0.544
0.384
0
0.564
1.516
2.671
0.826
1.053
0.851
0.441
0.102
1.195
0.598
0.154
0
0.022
0.199
0
0
0.123
0
4.004
5.653
0.815
2.239
0.196
5.155
8.514
0
0.539
1.437
0.371
0.165
12 May
0
0.569
0.172
0
1.697
0.596
0.007
8.880
1.280
2.254
0.546
0.853
0.689
0.357
0.082
0.929
0.464
0.093
0
0.123
0.135
0
0
0.097
0
3.892
5.495
0.688
1.889
0.166
3.406
6.894
4.903
0.418
1.397
0.360
0.160
15 May
0
0.669
0.049
0
1.172
0.442
0.003
3.071
1.230
2.168
0.613
0.748
0.605
0.313
0.072
18 May
0
0.821
0.099
0
1.223
0.458
0
0.040
1.238
2.180
0.858
0.699
0.565
0.293
0.067
0.626 ;" 1.189
0.313 0.595
0.117 ! 0.108
0
0.100
0
0
i
1
0.229 ] 0.139
0 0
0 0
0.138 0.120
0 0
3.657 3.560
5.164 5.026
0.661 0.665
1.817 1.827
0.159 I 0.160
[
3.827 ; 5.353
6.046 5.653
0.913, 0
1
0.282
1.312
0.339
0.151
0.536
1.277
0.330
0.147
21 May
0
0.638
0
0
3.323
0.437
0.010
7.598
1.238
2.180
0.712
0.652
0.527
0.273
0.063
1.291
0.645
0.214
0
0.
0.211
0
0
0.118
0
3.631
5.127
0.665
1.827
0.160
4.439
5.274
0
0.581
1.303
0.336
0.150
24 May
0
0.662
0
0.066
0.894
0.658
0
0.857
1.241
2.186
0.701
0.574
0.464
0.240
0.055
1.098
0.549
0.122
0
0
0.170
0
0
0.125
0
3.364
4.750
0.667
1.833
0.161
4.374
4.640
0
0.495
1.207
0.312
0.139
27 May
0
0.586
0.005
0.068
6.623
0.505
0
5.258
1.253
2.207
0.583
0.640
0.518
0.268
0.062
1.051
0.525
0.156
0
0.151
0.166
0
0
0.143
0
2.954
4.171
0.673
1.850
0.162
3.638
5.178
6.637
0.473
1.060
0.274
0.122
Where several sources are indicated, the emission rate given is for
each source.
42
-------
TABLE 2-8 (Continued)
Source
110
211
212
213
214
611
612
613
710
720
810
910
920
930
940
1110 & 1130
1120 & 1140
1211
1212
1213
1221
1222
1223
1231
1232
1310
1320
107X1
107X2
107X3
108X0
109X0
11010
111X0
113X1
113X2
113X3
Emission Rate* (g/sec)
30 May
0
0.610
0
0.068
0.384
0.651
0.001
8.370
1.238
2.180
0.634
0.724
0.585
0.303
0.070
1.060
0.530
0.094
0
0.075
0.113
0
0
0.113
0
3.335
4.708
0.665
1.827
0.160
3.954
5.850
5.377
0.477
1.196
0.309
0.137
2 Jun
0
0.645
0
0.071
0.192
0.689
0
2.053
1.315
2.317
0.588
0.524
0.424
0.219
0.051
1.236
0.618
0.075
0
0.106
0.154
0
0
0.116
0
2.942
4.154
0.707
1.942
0.170
3.668
4.235
6.047
0.557
1.056
0.273
0.121
5 Jun
0
0.478
0
0.065
1.844
0.493
0
3.581
1.335
2.352
0.549
0.588
0.475
0.246
0.057
0.793
0.396
0.086
0
0.107
0.166
0
0
0.135
0
3.218
4.544
0.718
1.971
0.173
3.427
4.750
10.175
0.357
1.155
0.298
0.133
8 Jun
0
0.595
0
0.065
0.858
0.569
0.041
0.879
1.330
2.342
0.602
0.647
0.523
0.271
0.062
0.470
0.235
0.081
0
0.018
0.127
0
0
0.103
0
2.744
3.874
0.715
1.963
0.172
3.753
5.229
0
0.212
0.985
0.254
0.113
11 Jun
0
0.623
0
0.065
0.598
0.485
0.192
2.648
1.319
2.323
0.734
0.668
0.540
0.279
0.064
1.083
0.542
0.114
0
0.036
0.120
0
0
0.081
0
2.522
3.560
0.709
1.948
0.171
4.577
5.398
3.469
0.488
0.905
0.234
0.104
14 Jun
0
0.595
0
0.061
0.830
0.514
0
0.596
1.310
2.308
0.590
0.518
0.419
0.217
0.050
0.928
0.464
0.060
0
0
0.078
0
0
0.056
0
2.511
3.545
0.704
1.934
0.170
3.682
4.189
0
0.418
0.901
0.233
0.103
17 Jun
0
0.469
0
0.011
0.673
0.457
0.172
4.439
1.294
2.279
0.531
0.554
0.448
0.232
0.053
0.945
0.472
0.124
0
0.122
0.119
0
0
0.080
0
2.398
3.386
0.696
1.911
0.168
3.312
4.^78
8.903
0.426
0.860
0.222
0.099
20 Jun
0
0.364
0
0.064
1.282
0.492
0.065
1.283
1.299
2.289
0.260
0.485
0.392
0.203
0.047
0.935
0.467
0.144
0
0.151
0.168
0
0.106
0.084
0
2.575
3.636
0.698
1.918
0.168
1.623
3.919
5.920
0.421
0.924
0.238
0.106
23 Jun
0
0.529
0
0.068
0.769
0.470
0
3.908
1.283
2.260
0.450
0.462
0.374
0.193
0.045
0.613
0.307
0.115
0
0.154
0.110
0
0.430
0.024
0
2.448
3.457
0.690
1.895
0.166
2.807
3.738
6.463
0.276
0.878
0.227
0.101
K
Where several sources are indicated, the emission rate given is for
each source.
43
-------
TABLE 2-8 (Continued)
Source
110
211
212
213
214
611
612
613
710
720
810
910
920
930
940
1110 & 1130
1120 & 1140
1211
1212
1213
1221
1222
1223
1231
1232
1310
1320
107X1
107X2
107X3
108X0
109X0
11010
111X0
113X1
113X2
113X3
Emission Rate* (g/sec)
26 Jun
0
0.564
0
0.062
1.474
0.649
0
2.707
1.283
2.260
0.519
0.570
0.461
0.238
0.055
0.331
0.165
0.099
0
0.283
0.062
0
0.509
0.029
0
2.168
3.061
0.690
1.895
0.166
3.235
4.608
2.474
0.149
0.778
0.201
0.089
29 Jun
0
0.544
0
0
1.046
0.591
0
2.440
1.276
2.248
0.512
0.612
0.495
0.256
0.059
0.290
0.145
0.097
0
0.297
0.060
0
0.413
0.031
0
2.424
3.423
0.686
1.884
0.165
3.194
4.947
0
0.131
0.870
0.225
0.100
2 Jul
0
0.497
0
0.064
0.423
0.568
0
7.260
1.276
2.248
0.441
0.569
0.460
0.238
0.055
0.634
0.317
0.108
0
0.314
0.039
0
0.487
0.119
0
2.396
3.384
0.686
1.884
0.165
2.751
4.597
2.613
0.286
0.860
0.222
0.099
5 Jul
1.781
0.460
0.170
0.152
0.423
0.224
0
0.005
1.278
2.251
0.472
0.642
0.519
0.269
0.062
0.465
0.233
0.173
0
0
0.158
0
0
0.146
0
2.376
3.355
0.687
1.887
0.166
2.944
5.192
0
0.210
0.852
0.220
0.098
8 Jul
2.029
0.441
"0.064
0.291
0
0.528
0
5.025
1.231
2.169
0.563
0.579
0.468
0.242
0.056
0.636
0.318
0.212
0
0
0.158
0
0
0.161
0
2.362
3.336
0.662
1.818
0.159
3.512
4.679
3.850
0.286
0.848
0.219
0.097
11 Jul
0.991
0.502
0
0.265
0.019
0.512
0
7.195
1.276
2.248
0.570
0.636
0.514
0.266
0.061
0.442
0.221
0.108
0
0.282
0.146
0
0.154
0.140
0
2.519
3.556
0.686
1.884
0.165
3.556
5.145
0
0.199
0.904
0.233
0.104
14 Jul
0
0.602
0
0.277
0
0.583
0
6.101
1.264
2.227
0.554
0.595
0.481
0.249
0.057
0.758
0.379
0.090
17 Jul
0
0.579
0
0.025
0
0.651
0
6.070
1.275
2.246
0.601
0.737
0.596
0.309
0.071
0.597
0.298
0.072
0 0
0.187
0.066
0
20 Jul
0
0.623
0
0.065
1.281
0.624
0
6.025
1.285
2.264
0.620
0.679
0.549
0.284
0.065
0.613
0.307
0.101
0
0.0631 0.096
0.064
0
0.154! 0.139
0.063J 0.061
o ; o
2.549i 2.267
3.599! 3.201
0.153
0
0.047
0.092
0
2.420
3.416
0.680i 0.685 0.691
1.867 1.883 1.897
0.164
3.455
4.807
0
0.165
0.166
3.748 3.869
5.963 5.489
4.324 0
0.341 0.269 0.2761
0.915 0.813| 0.868
0.236 0.210
0.105 0.093
0.224
0.100
Where several sources are indicated, the emission rate given is for
each source.
-------
TABLE 2-8 (Continued)
Source
110
211
212
213
214
611
612
613
710
720
810
910
920
930
940
1110 & 1130
1120 & 1140
1211
1212
1213
1221
1222
1223
1231
1232
1310
1320
107X1
107X2
107X3
108X0
109X0
11010
111X0
113X1
113X2
113X3
Emission Rate* (g/sec)
23 Jul
0
0.528
0
0.038
0.961
0.503
0
7.459
1.250
2.202
0.575
0.636
0.514
0.266
0.061
0.642
0.321
0.156
0
0.081
0.077
0.029
0.513
0.129
0
2.209
3.119
0.672
1.846
0.162
3.586
5.142
3.122
0.289
0.793
0.205
0.091
26 Jul
0
0.419
0
0.169
0
0.639
0.054
0
1.294
2.279
0.790
0.664
0.537
0.278
0.064
0.478
0.239
0.110
0.017
0
0.151
0.082
0
0.102
0.017
2.305
3.255
0.696
1.911
0.168
4.928
5.370
0
0.215
0.827
0.214
0.095
29 Jul
0
0.437
0
0.091
0
0.685
0.826
0
1.230
2.168
0.804
0.719
0.581
0.301
0.069
0.435
0.218
0.019
0.043
0
0.201
0.033
0
0.050
0.041
2.434
3.437
0.661
1.817
0.159
5.018
5.812
0
0.196
0.873
0.225
0.100
1 Aug
0
0.437
0
0.090
0
0.625
0.894
0
1.312
2.311
0.773
0.711
0.575
0.298
0.069
0.453
0.227
0.124
0.035
0
0.142
0.053
0
0.039
0.044
2.404
3.394
0.705
1.937
0.170
4.821
5.750
5.839
0.204
0.862
0.223
0.099
4 Aug
0
0.518
0
0.156
0
0.490
0.554
0
1.285
2.264
0.745
0.737
0.596
0.308
0.071
0.468
0.234
0.114
0.047
0
0.136
0.075
0
0.059
0.034
2.340
3.304
0.691
1.897
0.166
4.647
5.960
3.712
0.211
0.840
0.217
0.096
7 Aug
0
0.449
0
0.367
0
0.450
0.890
0
1.276
2.248
0.666
0.588
0.475
0.246
0.057
0.293
0.147
0.092
10 Aug
0
0.601
0
0.180
0
0.539
0.068
0
1.303
2.295
0.742
0.322
0.261
0.135
0.031
0.594
0.297
0.051
0.035! 0.034
0
0.162
0
0
0.054
0.033
2.351
3.319
0.686
1.884
0.165
4.154
4.753
3.087
0.132
0.844
0.218
0.097
0
0.106
0.070
0
0.047
0.017
2.398
3.386
0.700
1.924
0.169
4.629
2.607
0
0.268
0.860
0.222
0.099
13 Aug
0
0.527
0
0.260
0
0.447
0.885
0
1.231
2.169
0.682
0.693
0.560
0.290
0.067
0.613
0.307
0.101
0
0
0.091
0.028
0
0.074
0.058
2.207
3.116
0.662
1.818
0.159
4.256
5.601
3.191
0.276
0.792
0.204
0.091
16 Aug
0
0.558
0
0.203
0
0.475
0.246
0
1.310
2.308
0.804
0.641
0.518
0.268
0.062
0.708
0.354
0.126
0
0
0.096
0.029
0
0.051
0.038
2.398
3.386
0.704
1.934
0.170
5.018
5.180
0
0.319
0.860
0.222
0.099
Where several sources are indicated, the emission rate given is for
each source.
45
-------
TABLE 2-8 (Continued)
Source
110
211
212
213
214
611
612
613
710
720
810
910
920
930
940
1110 & 1130
1120 & 1140
1211
1212
• 1213
1221
1222
1223
1231
1232
1310
1320
107X1
107X2
107X3
108X0
109X0
11010
111X0
113X1
113X2
113X3
Emission Rate* (g/sec)
19 Aug
0
0.586
0
0.195
0
0.485
0.822
0
1.312
2.311
0.806
0.597
0.483
0.250
0.058
0.732
0.366
0.129
0
0
0.041
0.056
0
0.085
0
2.588
3.654
0.705
1.937
0.170
5.026
4.827
3.145
0.330
0.928
0.240
0.107
22 Aug
0
0.323
0
0.483
0
0.511
0.869
0
1.290
2.273
0.373
0.577
0.466
0.241
0.056
0.746
0.373
0.123
0
0
0.087
0.014
0
0.096
0
2.489
3.515
0.694
1.905
0.167
2.325
4.663
3.839
0.336
0.893
0.231
0.103
25 Aug
0
0.332
0
0.458
0
0.507
0.617
0
1.283
2.260
0.323
0.608
0.492
0.255
0.059
0.637
0.319
28 Aug
0
0.543
0
0.195
0
0.555
0.702
0
1.299
2.289
0.701
0.715
0.578
0.299
0.069
0.796
0.398
0.068 0.084
0.044
0
0.065
0.020
0
0.067
0.052 • 0.023
0 0
0.082
0
0.087
0
2.588 2.Q28
3.654 2.864
0.690 0.698
1.895 1.918
0.166 0.168
2.013 4.371
4.918 5.784
4.001 1.653
0.287 0.359
0.928
0.240
0.107
31 Aug
0
0.666
0
0.205
0
0.573
0.140
0
1.319
2.323
0.765
0.675
0.546
0.282
0.065
0.740
0.370
0.073
o
0
0.082
0
0
0.048
0.056
2.575
3.636
0.709
1.948
0.171
4.774
3 Sept
0
0.583
0
0.194
0
0.562
0.583
0
1.310
2.308
0.670
0.754
0.610
0.316
0.073
0.937
0.468
0.101
0
0
0.076
0
0
0.047
0.041
2.387
3.371
0.704
1.934
0.170
4.182
5.456 6.098
0 1 2.301
0.333
0.728 0.924
0.188
0.084
0.238
0.106
0.422
0.857
0.221
0.098
6 Sept
0
0.577
0.065
0.152
0
0.424
0.469
0
1.305
2.298
0.769
0.486
0.393
0.203
0.047
0.757
0.379
0.097
0
0
0.108
0.025
0
0.049
0.043
2.282
3.222
0.701
1.926
0.169
4.799
3.927
0
0.341
9 Sept
0
0.606
0.080
0.123
0
0.423
0.816
0
1.272
2.241
0.758
0.715
0.578
0.299
0.069
0.637
0.319
0.107
0.002
0
0.050
12 Sept
0
0.571
0.100
0.230
0
0.346
0.972
0
1.244
2.191
0.541
0.660
0.534
0.276
0.064
0.782
0.391
0.031
0.124
0
0.128
0.028 0
o 1 o
0.029
0.009
0.048
0.032
2.555 2.529
3.608 3.570
0.684 0.669
1.879 1.837
0.165 0.161
4.732 3.373
5.780' 5.336
1.330 0.925
0.287 , 0.352
0.819 0.917 . 0.907
0.211
0.094
0.237J 0.234
0.105
0.104
i
Where several sources are indicated, the emission rate given is for
each source.
46
-------
TABLE 2-8 (Continued)
Source
110
211
212
213
214
611
612
613
710
720
810
910
920
930
940
1110 & 1130
1120 & 1140
1211
1212
1213
1221
1222
1223
1231
1232
1310
1320
107X1
107X2
107X3
108X0
109X0
11010
111X0
113X1
113X2
113X3
Emission Rate* (g/sec)
15 Sept
0
0.449
0.102
0.196
0
0.506
0.680
0
1.324
2.333
0.585
0.621
0.502
0.260
0.060
0.787
0.394
0.096
0.077
0
0.022
0.119
0
0.020
0.036
2.657
3.752
0.712
1.955
0.172
3.647
5.024
4.879
0.355
0.953
0.246
0.109
18 Sept
0
0.435
0.105
0.149
0
0.577
0.729
0
1.276
2.248
0.656
0.672
0.543
0.281
0.065
1.061
0.531
0.026
0.104
0
0.047
0.096
0
0
0.052
2.489
3.515
0.686
1.884
0.165
4.089
5.432
1.307
0.478
0.893
0.231
0.103
21 Sept
0
0.054
0.127
0.278
0
0.410
0.103
0
1.295
2.281
0
0.278
0.225
0.116
0.027
1.065
0.532
0.161
0.033
0
0.133
0.121
0
0
0.073
2.595
3.664
0.696
1.912
0.168
0
2.251
0
0.480
0.931
0.240
0.107
24 Sept
0
0.399
0.067
0.030
0
0.372
0.745
0
1.283
2.260
0.683
0.615
0.497
0.257
0.059
0.922
0.461
0.083
0.022
0
0.049
0.048
0
0
0.049
2.606
3.679
0.690
1.895
0.166
4.258
4.969
2.787
0.415
0.935
0.241
0.107
27 Sept
0
0.603
0.094
0.036
0
0.298
0.305
0
1.259
2.218
0.724
0.709
0.573
0.297
0.068
1.405
0.702
0.120
0.002
0
0.109
0.054
0
0.012
0.028
2.522
3.560
0.677
1.859
0.163
4.516
5.730
0
0.633
0.905
0.234
0.104
30 Sept
0
0.592
0.092
0.095
0
0.264
0.326
0
1.266
2.230
0.768
0.709
0.573
0.297
0.068
1.411
0.706
0.093
0.056
0
0.088
0.059
0
0.029
0.012
2.450
3.459
0.681
1.870
0.164
4.790
5.734
3.758
0.636
0.879
0.227
0.101
3 Qct
0
0.586
0.093
0.074
0
0.473
0.710
0
1.272
2.241
0.800
0.687
0.555
6 Oct
0
0.607
0.119
0.060
0
0.513
0.828
0
1.273
2.243
0.831
0.676
0.546
0.287 0.283
0.066 0.065
1.231
0.616
0.088
0.093
0
0.058
0.071
0
0.029
0
2.641
3.729
0.684
1.879
0.165
4.991
5.552
3.712
0.555
0.948
0.245
0.109
1.250
0.625
0.040
0.135
0
0.006
0.150
0
0.030
0
2.645
3.734
0.684
1.880
0.165
5.182
5.463
0
0.563
0.949
0.245
0.109
Where several sources are indicated, the emission rate given is for
each source.
47
-------
TABLE 2-8 (Continued)
Source
110
211
212
213
214
611
612
613
710
720
810
910
920
930
940
1110 & 1130
1120 & 1140
1211
1212
1213
1221
1222
1223
1231
1232
1310
1320
107X1
107X2
107X3
108X0
109X0
11010
111X0
113X1
113X2
113X3
Emission Rate* (g/sec)
9 Oct
0
0.780
0.120
0.075
0
0.306
0.745
0
1.315
2.317
0.963
1.034
0. 836
0.433
0.100
0.717
0.358
0.202
0
0
0.067
0.120
0
0.061
0.053
2.509
3.543
0.707
1.942
0.170
6.007
8.360
3.307
0.323
12 Oct
0
0.633
0.063
0.174
0
0.427
0.400
0
1.271
2.240
0.945
0.986
0.797
0.412
0.095
1.065
0.533
0
0.162
0
0.170
0.055
0
0.134
0
2.275
3.212
0.684
1.879
0.165
5.893
7.968
0
! 0.480
0.900 0.816
0.232 0.211
0.103
0.094
15 Oct
0
0.739
0.152
0.098
0
0.403
0.453
0
1.310
2.308
0.795
0.960
0.776
0.402
18 Oct
0
0.817
0.129
0.047
0
0.449
0.403
0
1.317
2.320
0.828
1.085
0.877
0.454
0.093 0.105
0.924 1.081
0.462 : 0.540
0.138 0.034
0.067 0.121
0 • 0
0.111 0.122
0.117 0.012
0 0
0.147 0.062
0 0
2.613 1.180
3.689 1.667
0.704
0.708
1.934 1.945
0.170 0.171
4.957 5.166
7.765 8.772
3.885
0
0.416 0.487
0.937
0.242
0.108
0.424
0.109
0.049
21 Oct
0
0.991
0.120
0.151
0
0.618
0.588
0
1.316
2.319
1.045
0.774
0.626
0.324
0.075
0.943
0.471
0.010
0.108
0
0.002
0.143
0
0.002
0.118
2.498
3.527
0.708
1.944
0.170
6.517
6.255
7.134
0.425
0.896
0.231
0.103
24 Oct
0
0.763
0.124
0.137
0
0.438
0.500
0
1.285
2.264
0.773
0.925
0.748
0.387
0.089
1.080
0.540
0.062
0.028
0
0.086
0.052
0
0.071
0.046
2.554
3.606
0.691
1.897
0.166
4.821
7.477
0
0.487
27 Oct
0
0.790
0.121
0.190
0
0.370
0.772
0
1.301
2.292
0.973
1.102
0.891
0.461
0.106
1.283
0.642
0.109
0.004
0
0.081
0.053
0
0.068
0.041
2.696
3.806
0.69?
1.921
0.169
30 Oct
0
0.655
0.097
0.131
0
0.390
0.982
0
1.301
2.292
0.962
0.982
0.794
0.411
0.095
1.116
0.558
0.142
0
0
0.077
0.064
0
0.056
0.057
2.409
3.401
0.699
1.921
0.169
6.068 ! 6.000
8.913 ' 7.938
2.243 0
0.578
0.916 0.967
0.237 ': 0.250
0.105
0.111
0.503
0.864
0.223
0.099
Where several sources are indicated, the emission rate given is for
each source.
48
-------
TABLE 2-9
DAILY NUMBER OF HOURS OF OPERATION FOR THE SLAB SCARFER AND
THE SINTER PLANT FOR THE 61 SAMPLE DAYS
1980
Sample
Day
3 May
6 May
9 May
12 May
15 May
18 May
21 May
24 May
27 May
30 May
2 Jun
5 Jun
8 Jun
11 Jun
14 Jun
17 Jun
20 Jun
23 Jun
26 Jun
29 Jun
2 Jul
5 Jul
8 Jul
11 Jul
14 Jul
17 Jul
20 Jul
23 Jul
26 Jul
29 Jul
1 Aug
4 Aug
7 Aug
10 Aug
13 Aug
16 Aug
19 Aug
22 Aug
25 Aug
28 Aug
Number of Hours of Operation
Source 410
10
16
9
9
16
9
16
9
16
8
8
9
9
9
0
0
9.5
8
8
8
8
8.25
8
8
2.5
0
8.75
8
8
8
8
8
8
8
8
8
8.25
8
7
9.25
Sources 510, 10510 & 10520
16
16
0
0
16
16
16
0
16
16
0
0
0
0
0
0
0
0
0
0
16
16
16
0
0
16
0
8
0
8
0
8
16
0
16
0
16
16
16
16
49
-------
TABLE 2-9 (Continued)
1980
Sample
Day
31 Aug
3 Sep
6 Sep
9 Sep
12 Sep
15 Sep
18 Sep
21 Sep
24 Sep
27 Sep
30 Sep
3 Oct
6 Oct
9 Oct
12 Oct
15 Oct
18 Oct
21 Oct
24 Oct
27 Oct
30 Oct
Number of Hours of Operation
Source 410
8
8
8
8
8
8
8.75
8
8
8
8
9
9
16
16
24
8
8
16
16
16
Sources 520, 10510 & 10520
0
16
16
16
0
0
16
8
0
0
0
0
0
16
16
16
0
16
0
24
24
50
-------
TABLE 2*-10
SOURCE INPUTS (EXCEPT GRAVITATIONAL SETTLING PARAMETERS) FOR
THE STORAGE PILE SOURCES
Source
Number
21101
21102
21201
21202
21203
21204
21301
22101
22102
24101
24102
25101
26101
26102
Emission Rate
(g/(sec«m2))
May-July
5.64xlO~5
5.64xlO~5
5.27xlO~5
5.27xlO~5
5.27xlO~5
5.27xlO~5
1.21xlO~4
1.21xlO~4
1.31xlO~4
2.23xlO~4
2.23xlO~4
4.55xlO~6
3.26xlO~4
3.26xlO~4
Aug-Oct
2.99xlO~5
2.99xlO~5
3.02xlO"5
3.02xlO~5
3.02xlO~5
3.02xlO"5
6.94xlO~5
6.91xlO~5
7.49xlO~5
1.28xlO~4
1.28xlO~4
2.61xlO~6
1.87xlO~4
1.87xlO~4
Coordinates (m)
UTM X
725,400
725,550
724,160
724,250
724,340
724,430
724,980
724,610
724,740
724,190
724,200
724,550
725,520
725,690
UTM Y
4,372,890
4,372,800
4,373,460
4,373,460
4,373,460
4,373,460
4,374,490
4,373,520
4,373,430
4,373,920
4,373,840
4,371,720
4,373,000
4,372,870
Emission
Height
(m)
15.0
15.0
10.0
10.0
10.0
10.0
5.0
20.0
20.0
12.0
12.0
5.0
5.0
5.0
Width (XQ) of
Square Area
Source (m)
145
145
90
90
90
90
60
125
125
65
65
600
100
100
51
-------
TABLE 2-11
PARTICULATE SIZE DISTRIBUTION, GRAVITATIONAL SETTLING
VELOCITIES AND SURFACE REFLECTION COEFFICIENTS FOR
THE STORAGE PILE SOURCES
Particle Size
Category
< 2.5 ym
2.5 to 15 ym
15 to 30 ym
> 30 ym
Mass
Fraction
0.150
0.560
0.290
0.000
Settling Velocity
(m/sec)
0.0002
0.0075
0.0407
—
Surface
Reflection
Coefficient
1.00
0.77
0.60
—
52
-------
TABLE 2-12
SOURCE INPUTS (EXCEPT PARTICULATE EMISSIONS
PARAMETERS) FOR THE ROADWAY SOURCES
Source
Number
Coordinates (m)
UTM X
UTM Y
Emission
Height
(m)
Initial Source Dimensions
Vertical
0 (m)
zo
Lateral
0 (m)
yo
(1) Processing Loop
40101
40102
40103
40104
40105
40106
40107
40108
40109
40110
40111
40112
40113
40114
40115
40116
40117
40118
726,930
727,070
727,210
727,360
727,510
727,660
727,720
727,640
727,570
727,570
727,570
727,480
727,320
727,160
727,090
727,090
727,095
726,950
4,373,850
4,373,850
4,373,780
4,373,770
4,373,770
4,373,770
4,373,690
4,373,620
4,373,550
4,373,390
4,373,230
4,373,140
4,373,140
4,373,140
4,373,225
4,373,385
4,373,545
4,373,560
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
69.8
69.8
69.8
69.8
69.8
69.8
65.1
65.1
74.4
74.4
74.4
74.4
74.4
74.4
74.4
74.4
74.4
74.4
(2) South Stores Receiving
40201
40202
40203
40204
40205
40206
40207
726,350
726,430
726,510
726,590
726,670
726,750
726,830
4,373,900
4,373,900
4,373,900
4,373,890
4,373,870
4,373,850
4,373,840
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
37.2
37.2
37.2
37.2
37.2
37.2
37.2
(3) Scale Pit
40301
40302
40303
40304
40305
726,875
726,870
726,865
726,860
726,855
4,373,780
4,373,630
4,373,480
4,373,330
4,373,180
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
69.8
69.8
69.8
69.8
69.8
53
-------
TABLE 2-12 .(Continued)
Source
Number
Coordinates (m)
UTM X
UTM Y
Emission
Height
(m)
Initial Source Dimensions
Vertical
rf f—\
o (.m;
zo
Lateral
yov
(3) Scale Pit (Continued)
40306
40307
40308
726,850
726,840
726,890
4,373,030
4,372,890
4,372,750
0.0
0.0
0.0
2.0
2.0
2.0
69.8
69.8
69.8
(4) General Services
40401
40402
40403
40404
40405
404Q6
40407
726,530
726,530
726,530
726,520
726,520
726,510
726,500
4,373,830
4,373,730
4,373,605
4,373,455
4,373,280
4,373,080
4,372,830
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
46.5
46.5
69.8
69.8
93.0
93.0
139.5
(5) Soaking Pit
40501
40502
726,585
726,785
4,372,670
4,372,670
0.0
0.0
2.0
2.0
93.0
93.0
(6) OSR Gate
40601
40602
40603
725,730
726,030
726,335
4,372,630
4,372,680
4,372,700
0.0
0.0
0.0
2.0
2.0
2.0
139.5
139.5
139.5
(7) Coke Plant Loop
40701
40702
40703
40704
40705
40706
40707
40708
40709
40710
725,650
725,500
725,360
725,230
725,140
725,070
725,030
724,980
724,950
724,930
4,372,710
4,372,710
4,372,670
4,372,660
4,372,700
4,372,750
4,372,790
4,372,820
4,372,840
4,372,850
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
69.8
69.8
69.8
46.5
46.5
34.9
23.3
23.3
11.6
11.6
54
-------
TABLE 2-12 (Continued)
Source
Number
Coordinates (m)
UTM X
UTM Y
Emission
Height
(m)
Initial Source Dimensions
Vertical
_ / \
a (m)
zov '
Lateral
** / \
0 (m)
yo v '
(7) Coke Plant Loop (Continued)
40711
40712
40713
40714
40715
40716
40717
40718
40719
40720
40721
40722
40723
40724
40725
40726
40727
40728
40729
40730
724,910
724,890
724,870
724,860
724,850
724,840
724,830
724,830
724,830
724,855
724,895
724,940
724,980
725,015
725,060
725,100
725,140
725,200
725,285
725,360
4,372,870
4,372,880
4,372,900
4,372,920
4,372,940
4,372,960
4,372,980
4,373,020
4,373,070
4,373,085
4,373,055
4,373,025
4,372,995
4,372,965
4,372,940
4,372,915
4,372,890
4,372,840
4,372,780
4,372,720
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
11.6
11.6
11.6
11.6
11.6
11.6
11.6
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
46.5
46.5
46.5
(8) BOF/Coal Pile
40801
40802
40803
40804
40805
40806
40807
725,305
725,435
725,560
725,690
725,815
725,950
726,100
4,373,190
4,373,100
4,373,005
4,372,910
4,372,820
4,372,760
4,372,730
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
69.8
69.8
69.8
69.8
69.8
69.8
69.8
(9) Transportation
40901
40902
40903
725,860
725,990
726,050
4,373,050
4,372,960
4,372,820
0.0
0.0
0.0
2.0
2.0
2.0
74.4
74.4
74.4
55
-------
TABLE 2-12 (Continued)
Source
Number
Coordinates (TO)
UTM X
UTM Y
Emission
Height
(m)
\"*/
Initial Source Dimensions
Vertical
a (m)
zo
Lateral
a (m)
yo >
(10) Mold Yard
41001
41002
41003
41004
41005
41006
41007
725,640
725,760
725,880
726,005
726,135
726,275
726,415
4,373,315
4,373,215
4,373,120
4,373.030
4,372,945
4,372,885
4,372,825
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
69.8
69.8
69.8
69.8
69.8
69.8
69.8
(11) Reeds Yard
41101
41102
41103
41104
41105
41106
41107
41108
41109
41110
41111
41112
41113
41114
41115
41116
41117
41118
725,100
725,115
725,130
725,140
725,150
725,165
725,180
725,195
725,220
725,260
725,295
725,330
725,380
725,450
725,515
725,560
725,560
725,565
4,374,170
4,374,155
4,374,135
4,374,115
4,374,095
4,374,070
4,374,050
4,374,035
4,374,005
4,373,970
4,373,940
4,373,900
4,373,845
4,373,775
4,373,700
4,373,610
4,373,510
4,373,415
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2,0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2,0
2.0
2.0
11.6
11.6
11.6
11.6
11.6
11.6
11.6
11.6
23.3
23.3
23.3
23.3
46.5
46.5
46.5
46.5
46.5
46.5
(12) OH Loop
41201
41202
41203
41204
41205
41206
41207
41208
724,910
725,000
725,050
725,100
725,170
725,230
725,300
725,400
4,373,790
4,373,790
4,373,870
4,373,910
4,373,880
4,373,830
4,373,780
4,373,760
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
46.5
46.5
34.9
34.9
34.9
34.9
46.5
46.5
56
-------
TABLE 2-12 (Continued)
Number
Coordinates (m)
UTM X
UTM Y
Emission
Height
(m)
Initial Source Dimensions
Vertical
Ozo(m)
Lateral
0 (m)
yo
(13) EOF Loop
41301
41302
41303
41304
41305
41306
725,225
725,135
725,145
725,255
725,395
725,520
41401
41402
41403
41404
41405
41406
41407
41408
725,600
725,850
725,750
725,610
725,780
725,590
725,350
725,150
4,373,305
4,373,430
4,373,480
4,373,485
4,373,505
4,373,415
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
74.4
74.4
74.4
74.4
74.4
74.4
(14) Slag Hauler
4,373,170
4,373,010
4,372,750
4,372,530
4,372,290
4,372,070
4,372,050
4,372,020
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
139.5
139.5
139.5
139.5
139.5
139.5
93.0
93.0
(15) Slag Dump
41501
41502
41503
41504
41505
41506
41507
41508
41509
41510
41511
724,620
724,770
724,940
725,100
725,290
725,520
725,730
725,720
725,670
725,850
725,810
4,371,710
4,371,730
4,371,730
4,371,800
4,371,850
4,371,940
4,372,160
4,372,430
4,372,680
4,372,850
4,373,060
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0,0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
69.8
69.8
93.0
93.0
93.0
139.5
139.5
139.5
116.3
116.3
116.3
(16) Old Lefferson
41601
41602
41603
41604
724,570
724,680
724,760
724,830
4,373,980
4,374,030
4,374,070
4,374,090
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
69.8
46.5
34.9
34.9
57
-------
TABLE 2-12 (Continued)
Qm i T*r* o
oourcc
Number
Coordinates (m)
UTM X
UTM Y
Emission
Height
(m)
V"**
Initial Source Dimensions
Vertical
0 (m)
zo
Lateral
Vm)
(16) Old Lefferson (Continued)
41605
41606
41607
41608
41609
41610
41611
41612
41613
724,890
724,940
724,980
725,015
725,040
725,065
725,090
725,115
725,140
4,374,110
4,374,130
4,374,150
4,374,170
4,374,170
4,374,180
4,374,180
4,374,190
4,374,190
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
23.3
23.3
23.3
11.6
11.6
11.6
11.6
11.6
11.6
(17) Lefferson/Yankee Gate
41701
41702
41703
41704
41705
41706
41707
41708
41709
41710
41711
41712
41713
41714
41715
724,460
724,360
724,280
724,240
724,210
724,190
724,170
724,160
724,160
724,160
724,160
724,160
724,160
724,160
724,160
4,373,950
4,373,960
4,374,020
4,374,100
4,374,160
4,374,220
4,374,270
4,374,300
4,374,325
4,374,350
4,374,375
4,374,400
4,374,425
4,374,450
4,374,475
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
46.5
46.5
46.5
34.9
34.9
23.3
23.3
11.6
13.6
11.6
11.6
11.6
11.6
11.6
11.6
(18) E-W Freeway
41801
41802
41803
41804
41805
41806
41807
41808
41809
724,175
724,200
724,225
724,250
724,280
724,330
724,380
724,430
724,480
4,374,480
4,374,480
4,374,480
4,374,480
4,374,480
4,374,470
4,374,460
4,374,470
4,374,480
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
11.6
11.6
11.6
11.6
23.3
23.3
23.3
23.3
23.3
58
-------
TABLE 2-12 (Continued)
Source
Number
Coordinates (m)
UTM X
UTM Y
Emission
Height
(m)
Initial Source Dimensions
Vertical
a (m)
zo '
Lateral
°yo(m)
(18) E-W Freeway (Continued)
41810
41811
41812
41813
41814
41815
41816
41817
724,530
724,590
724,670
724,740
724,810
724,890
724,960
725,040
4,374,490
4,374,490
4,374,490
4,374,490
4,374,490
4,374,490
4,374,490
4,374,480
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
23.3
34.9
34.9
34.9
34.9
34.9
34.9
34.9
(19) Old Muzzy
41901
41902
41903
41904
41905
724,800
724,760
724,720
724,710
724,710
4,374,130
4,374,190
4,374,250
4,374,340
4,374,440
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
34.9
34.9
34.9
46.5
46.5
(20) Old Stores
42001
42002
42003
42004
42005
42006
42101
42102
42103
42104
42105
42106
42107
724,730
724,690
• 724,620
724,530
724,460
724,460
4,374,530
4,374,590
4,374,610
4,374,620
4,374,600
4,374,520
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
34.9
34.9
34.9
46.5
34.9
34.9
(21) BF/BOF
724,560
724,700
724,830
724,930
724,950
725,020
725,120
4,373,930
4,373,880
4,373,810
4,373,700
4,373,560
4,373,430
4,373,290
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
69.8
69.8
69.8
69.8
69.8
69.8
93.0
59
-------
TABLE 2-12 (Continued)
Q/"i 1 1 T*r* c*
oourcc
Number
Coordinates (m)
UTM X
UTM Y
Emission
Height
(n)
\>u/
Initial Source Dimensions
Vertical
a (m)
zo
Lateral
a (m)
yo
(22) No. 1 OH
42201
42202
42203
42204
42205
42206
42207
42208
724,370
724,370
724,370
724,360
724,360
724,370
724,370
724,370
4,374,955
4,374,880
4,374,805
4,374,730
4,374,660
4,374,600
4,374,540
4,374,500
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2,0
34.9
34.9
34.9
34.9
34.9
23.3
23.3
23.3
(23) East Processing
42301
42302
42303
42304
42305
42306
42307
42308
42309
42310
42311
42312
42313
42314
42315
42316
42317
42318
42319
42320
42321
42322
42323
42324
42325
42326
42327
724,600
724,680
724,740
724,790
724,830
724,880
724,930
724,980
725,030
725,050
725,080
725,100
725,110
725,100
725,090
725,080
725,080
725,080
725,080
725,070
725,080
725,090
725,100
725,110
725,120
725,130
725,140
4,375,300
4,375,300
4,375,300
4,375,300
4,375,300
4,375,310
4,375,320
4,375,310
4,375,280
4,375,240
4,375,180
4,375,110
4,375,020
4,374,920
4,374,800
4,374,680
4,374,580
4,374,490
4,374,420
4,374,370
4,374,340
4,374,320
4,374,290
4,374,270
4,374,240
4,374,220
4,374,200
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
34.9
34.9
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
34.9
34.9
46.5
46.5
69.8
46.5
46.5
34.9
23.3
23.3
11.6
11.6
11.6
11.6
11.6
11.6
11.6
60
-------
TABLE 2-12 (Continued)
Source
Number
Coordinates (m)
UTM X
UTM Y
Emission
Height
Cm)
(24) West Processing
42401
42402
42403
42404
42405
42406
724,555
724,555
724,560
724,545
724,470
724,400
4,375,255
4,375,180
4,375,105
4,375,035
4,375,030
4,375,010
0.0
0.0
0.0
0.0
0.0
0.0
Initial Source Dimensions
Vertical
Ozo(m)
Lateral
ayo<»>
2.0
2.0
2.0
2.0
2.0
2.0
34.9
34.9
34.9
34.9
34.9
34.9
(25) North Perimeter
42501
42502
42503
42504
42505
42506
42507
42508
42509
42510
42511
42512
42513
42514
42515
42516
42517
42518
725,160
725,170
725,190
725,200
725,220
725,230
725,240
725,270
725,300
725,350
725,420
725,490
725,610
725,760
725,910
726,060
726,190
726,270
42601
42602
42603
42604
42605
42606
42607
42608
724,170
724,130
724,130
724,130
724,140
724,160
724,190
724,230
4,374,180
4,374,170
4,374,150
4,374,130
4,374,110
4,374,090
4,374,070
4,374,040
4,374,010
4,373,970
4,373,910
4,373,840
4,373,820
4,373,820
4,373,820
4,373,810
4,373,800
4,373,850
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
11.6
11.6
11.6
11.6
11.6
11.6
11.6
23.3
23.3
34.9
46.5
46.5
69.8
69.8
69.8
69.8
46.5
46.5
(26) Miller's Crossing
4,374,180
4,374,100
4,374,000
4,373,920
4,373,860
4,373,820
4,373,770
4,373,740
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
46.5
46.5
46.5
34.9
23.3
23.3
23.3
23.3
61
-------
TABLE 2-12 (Continued)
Number
Coordinates (m)
UTM X
UTM Y
Emission
Height
(m)
Initial Source Dimensions
Vertical
0 (m)
zo
Lateral
o (m)
yo
(26) Miller's Crossing (Continued)
42609
42610
42611
42612
42613
42614
42615
42616
42617
42618
724,270
724,310
724,340
724,410
724,510
724,630
724,720
724,790
724,800
724,810
4,373,710
4,373,680
4,373,650
4,373,610
4,373,540
4,373,450
4,373,360
4,373,290
4,373,210
4,373,140
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
23.3
23.3
23.3
46.5
69.8
69.8
46.5
46.5
34.9
34.9
(27) Recycle Plant Loop
42701
42702
42703
42704
42705
725,200
725,090
724,980
724,930
725,050
4,373,090
4,373,190
4,373,290
4,373,410
4,373,300
0.0
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2,0
2.0
69.8
69.8
69.8
93.0
69.8
(28) Recycle Plant/B.F.
42801
42802
42803
42804
725,580
725,430
725,330
725,230
4,372,780
4,372,890
4,372,990
4,373,120
0.0
0.0
0.0
0.0
2.0
2.0
2.0
2.0
93.0
69.8
69.8
69.8
62
-------
TABLE 2-13
AVERAGE PARTICULATE EMISSION RATES FOR THE ROADWAY SOURCES
Source
Number (s)
40101-40106
40107-40108
40109-40118
40201-40207
40301-40308
40401-40402
40403-40404
40405-40406
40407
40501-40502
40601-40603
40701-40703
40704-40705
40706
40407-40708
40709-40717
40718-40727
40728-40730
40801-40807
40901-40903
41001-41007
41101-41108
41109-41112
41113-41118
41201-41202
41203-41206
41207-41208
41301-41306
41401-41406
41407-41408
41501-41502
41503-41505
41506-41508
41509-41511
41601
41602
41603-41604
Emission Rate* (g/sec)
May - July
0.2935
0.2737
0.3128
0.6444
0.1254
0.1139
0.1709
0.2277
0.3416
0.1633
1.0225
0.1674
0.1115
0.0837
0.0559
0.0278
0.0559
0.1115
0.7229
0.0482
0.0997
0.0198
0.0397
0.0792
1.8659
1.4004
1.8659
1.3829
0.1522
0.1015
0.7458
0.9938
1.4906
1.2427
0.1387
0.0924
0.0693
August - October
0.0587
0.0547
0.0626
0.1289
0.0251
0.0228
0.0342
0.0455
0.0683
0.0327
0.2045
0.0335
0.0223
0.0167
0.0112
0.0056
0.0112
0.0223
0.1446
0.0096
0.0199
0.0040
0.0079
0.0158
0.3732
0.2801
0.3732
0.2766
0.0304
0.0203
0.1492
0.1988
0.2981
0.2485
0.0277
0.0185
0.0139
* Emission rates for each source.
63
-------
TABLE 2-13 (Continued)
Source
Number(s)
41605-41607
41608-41613
41701-41703
41704-41705
41706-41707
41708-41715
41801-41804
41805-41810
41811-41817
41901-41903
41904-41905
42001-42003
42004
42005-42006
42101-42106
42107
42201-42205
42206-42208
42301-42302
42303-42310
42311-42312
42313-42314
42315
42316-42317
42318
42319-42320
42321-42327
42401-42406
42501-42507
42508-42509
42510
42511-42512
42513-42516
42517-42518
42601-42603
42604
42605-42611
42612
42613-42614
42615-42616
Emission Rate* (g/sec)
May - July
0.0463
0.0231
0.0946
0.0710
0.0474
0.0236
0.0223
0.0449
0.0672
0.0403
0.0537
0.0389
0.0519
0.0389
1.3543
1.8044
0.0852
0.0569
0.1794
0.1197
0.1794
0.2390
0.3587
0.2390
0.1794
0.1197
0.0596
0.1027
0.0561
0.1127
0.1689
0.2250
0.3377
0.2250
0.0271
0.0203
0.0136
0.0271
0.0407
0.0271
August - October
0.0093
0.0046
0.0189
0.0142
0.0095
0.0047
0.0045
0.0090
0.0134
0.0081
0.0107
0.0078
0.0104
0.0078
0.2709
0.3609
0.0170
0.0114
0.0359
0.0239
0.0359
0.0478
0.0717
0.0478
0.0359
0.0239
0.0119
0.0205
0.0112
0.0225
0.0338
0.0450
0.0675
0.0450
0.0003
0.0002
0.0001
0.0003
0.0004
0.0003
* Emission rates for each source.
64
-------
TABLE 2-13 (Continued)
Source
Number(s)
Emission Rate* (g/sec)
May - July
August - October
42617-42618
42701-42703
42704
42705
42801
42802-42804
0.0203
0.3000
0.3998
0.3000
0.5720
0.4293
0.0002
0.0030
0.0040
0.0030
0.0057
0.0043
*Emission rates for each source.
65
-------
TABLE 2-14
PARTICLE SIZE DISTRIBUTION, GRAVITATIONAL SETTLING VELOCITIES
AND SURFACE REFLECTION COEFFICIENTS FOR THE
ROADWAY SOURCES
Particle Size
Category
Mass
Fraction
Settling Velocity
(m/sec)
Surface Reflection
Coefficient
(a) Paved Roadways (40101-42518)
< 2.5 ym
2.5 to 15 ym
15 to 30 ym
>30 ym
0.200
0.570
0.230
0.000
0.0002
0.0075
0.0407
-
1.00
0.77
0.60
-
(b) Unpaved Roadways (42601-42804)
< 2.5 ym
2.5 to 15 ym
15 to 30 ym
> 30 ym
0.200
0.480
0.320
0.000
0.0002
0.0075
0.0407
-
1.00
0.77
0.60
-
66
-------
rates are given in Table 2-13 and the remaining source input parameters are
given in Table 2-14. The roads were treated as line sources represented by
separated volume sources. A volume source was placed at the center of each
roadway segment with an initial lateral dimension equal to the length of
the segment divided by 2.15. In general, each roadway segment has a length
that is less than one-third the distance to the nearest particulate air
quality monitoring site. However, in order to keep the number of sources
at a reasonable level, the minimum length of the roadway segments is 25
meters. The initial vertical dimension of each roadway segment was set
equal to 2 meters, which is consistent with the empirically-based
assumptions of the HIWAY-2 model (Petersen, 1980). The effective emission
height was arbitrarily set equal to zero for all of the roadway segments.
Tables 2-15, 2-16 and 2-17 give the source inputs for the area
and volume sources used to represent the parking lots and parking lot
access roads. With the exception of the gravitational settling parameters,
all source inputs for the parking lots and the parking lot access roads are
shown in Tables 2-15 and 2-16, respectively. The gravitational settling
parameters for both the parking lots and the parking lot access roads are
given in Table 2-17. The parking lots were represented for modeling
purposes by area sources; the procedures used to develop the source inputs
for these sources were the same as those used to develop the source inputs
for the storage piles. The parking lot access roads were treated as line
sources represented by separated volume sources in the same manner as the
other roadway segments. We partitioned the total particulate emission rate
provided for each parking lot between the parking lot and the corresponding
parking lot access road according to the travel distance from the plant
entrance to the estimated median parking space. For example, if the travel
distance from the plant entrance to the entrance of a parking lot is 50
percent of the total travel distance from the plant entrance to the
estimated median parking space within the parking lot, the total
particulate emissions were equally divided between the parking lot access
road and the parking lot.
67
-------
TABLE 2-15
SOURCE INPUTS (EXCEPT GRAVITATIONAL SETTLING PARAMETERS)
FOR THE PARKING LOTS
Source
Number
50101
50102
50201
50202
50203
50301
50302
50401
50501
50502
50503
50601
50602
50603
50604
50605
50606
50607
50701
50702
50703
50801
50802
50901
50902
50903
51001
51002
51101
51201
51202
51301
Emission Rate
(g/(sec-m2))
May - Jul
4.23x10"^
4.23x10,
9.75x10"'
9.75x10 '
9.75x10 '
™ o
6.67x10 °
6.67x10 I
8.61x10°
2. 40x10
2.40x10"^
2.40x10 ,
4.86x10":'
4.86x10
4.86x10 ?
4.86x10 ,
4.86x10 ,
4.86x10";'
4.86x10 ;
7.39x10 ,
7.39x10 °
7.39x10"°
9.09x10 '
9.09x10
1.11x10 \
1.11x10 i?
1.11x10,
7.52x10"'
7.52x10 '
2.50x10"'
1.51x10 '
1.51x10 '
3.21x10
Aug - Oct
8.45x10"^
8.45x10,
1.95x10 ,
1.95x10 '
1.95x10"'
1.33x10":*
1.33x10";?
1.72x10,
4.79x10 '
4.79x10 '
4.79x10"'
9.73x10 '
9.73x10 '
9.73x10"'
9.73x10"'
9.73x10 '
9.73x10"'
9.73x10 '
1.48x10 I
1.48x10 ,
1.48x10",
1.82x10 '
1.82x10 '
2.22x10 !?
2.22x10 ,
2.22x10 ,
1.50x10 ,
1.50x10"'
5.00x10"^
3.03x10 "
3.03x10 r
6.42x10 B
Coordinates (m)
UTM X
726,860
726,860
726,450
726,450
726,450
724,400
724,300
724,560
724,210
724,230
724,300
726,050
726,050
726,110
726,170
726,150
726,420
726,490
724,775
724,775
724,775
727,100
727,200
724,000
724,000
724,000
723,470
723,540
726,500
724,100
724,850
724,200
UTM Y
4,373,100
4,373,325
4,373,310
4,373,370
4,373,430
4,375,120
4,375,130
4,375,370
4,375,010
4,374,930
4,374,930
4,373,840
4,373,940
4,373,940
4,373,940
4,373,920
4,373,910
4,373,910
4,372,815
4,372,770
4,372,725
4,372,600
4,372,600
4,374,290
4,374,350
4,374,410
4,374,080
4,374,080
4,372,700
4,373,400
4,372,600
4,374,500
Emission
Height
(m)
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Width (x0) of
Square Area
Source (m)
225
225
60
60
60
100
100
130
110
70
70
100
60
60
60
80
70
70
45
45
45
100
100
60
60
60
70
70
1200
1100
800
900
68
-------
CM f-l
td O
CO
§
•H
CO
e
•H
O
Source
Initial
i
•i
p
.^\
v^
en
^
0)
•P O
$ J*
M D
^ ^
cfl E
o v^
•H
l-i N
!> °
x-*\
3 e
D v-'
-\
1) 4-1
-< to
6 -rl
J 01
PC
>>
£
P
X!
25
s
4J
a
O
t
00
<
rH
O
•-J
1
>^
&
> M
j a>
•i ^>
M
5 BE
«fl
5
CO
01
o
3
4->
o
>J
S
ii
s
w
t-H
\-^
O OO CO
CO Q^ ^
0 O O
tN 04 CS
O O O
0 O 0
o o o
O\ CN r»
0\ OO \O
CM CS CM
I** f- r^
co co co
- r^
O CM CM
CM CM CM
O O O
0 O 0
OO vO vO
00 i-t rH
-3- i-l r-l
i-H r-< t-l
0 O 0
co -a- u~>
0 O O
T— i 1— « 1— 1
o o o
in iri m
13
a
£
CO
CO
c
t-l
s
X- S
CO
v^
vO vO vO vO ^O ^ vO ^O
OOOOOOOO
CMCMCMCMCMCMCMCM
OOOOOOOO
OOOOOOOO
oooomomo
moOOCM-*P^O\CM
avc^ooooo--!
coco«r~«r-.r^p^r-«t--
CO CO CO CO CO CO CO CO
-3--3-.r--r-.
CO CO CO CO CO CO CO CO
OOOOOOOO
OOOOOOOO
OOOOOOOO
vO vO vO vO vO ^O vO vO
r--t^r--r~-r~r~~^-r--
rHr-(iHrHl-|rHrHr-<
OOOOOOOO
OOOOOOOO
I
cMco-j-mvor^ooaN
^O \D VP \O ^O ^>O *«O vO
CMCMCMCMCMCNCMCM
OOOOOOOO
iriiniriiriiniri'Aio
69
-------
•a
ai
c
•H
4J
a
o
CM
1
nsions
0>
•H
O
al Source
Initi.
•t
I
S~\
e
N^»
U
0)
4-)
tfl
C
•H
•a
VJ
8
o
s~\
a
0)
CO
60
OJ
4-1
3
c
o
•H
CO
•H
W
C
(
:
c/
rH
2 'P
)-l C
Q) ^
4J O
3 to*
r-H ^
to B
O ^
•H
4J O
M N
Q) t)
/-%
55
/> 4-1
jj _£
-( 60
3 -H
J 0)
SB
>«
^
>
0
o
00
<
<-t
•n
1
^
1
Number
T3
«
&
0}
W
IV
U
%
73
(U
l-l
5
W
CO
a.
5
w
rt
§
en
M
M-l
•4-1 .
CU
*1
1
R
•rl
2
4J
CO
a
•H
1
a)
tenanc
•H
ro
v£> v£) \O \£>
• * * •
,_H ^_( r-< ^_(
O O O O
CN CM CS CV4
O O O 0
O O O O
ui o in o
sr r-» o\ cvj
—I i— 1 ,-H CM
•O'— 'co^ooO'-tio
cMcMcnc>llicorricncnrocn>io
ini/"iinu~>u"imir>u~>ininmu~>mminu">Ln
r~r~r~.r-»T^t^.r^r~-r-»r^-i~»i~»r--r*-r^r^rN»
ronrococororororoncorocococncoro
Or-icMn CO fO CO CO CO CO CO CO CO CO CO CO CO CO CO
OOOOOOOOOOOOOOOOO
inu">mu~>inu~>mminmLnu"iinu~iminm
70
-------
I
CM
W
§
•H
CO
n J3
•H 60
3 -H
q a)
P*
>H
|
X
H
P
4-1
O
0
1
60
3
r-l
3
"->
1
^
1
> M
J (U
11
"S
p3
CO
CO
0)
4J
3
cu
4J
«
O
14-1
<4-l
U
s
^*\
m
\-s
vO vJ3 \o
• • •
•-< >— I <— i
O O O
C>1 CM CM
O O O
O 0 (D
o o m
O CN -»
in m m
m m m
r- t-» r~
CO CO CO
-* -3- - p-
fS
o o o
o o o
o o o
CM CM CM
00 CO CO
i-t rH r-l
0 O O
0 O O
CM CO -*
-* -* -3"
O O 0
m in m
tfl
&
CO
m
u
3
a
U-l
m
C/5
C
O
u
CO
u
•H
C
•H
1
vO
v.^
c^ cy^ ON PO co co co \o \o ^o ^o \o vo vo \o *«o **o vo vo xD KO co
oooooooooooooooooooooo
CMCMCNCMCMCMCMCNCMCMCMCMCMCMCMCMCMCMCMCNCMCM
OOOOOOOOOOOOOOOOOOOOOO
oooooooooooooooooooooo
ooooooooinmoooinooinmoinoo
-nv£)i-ivooino\fo>3-invor^oooocj\O'-«r--r^r^pxr-»r~.r~.r-r-t~.r^r^.t-»r~
cocococococococococococococococococococococo
-*3-<->a-^«*
ooooommoooinoinoominooooo
ooNcocjxr-3->X)r^i^r--r~
-*COCOCMCMCMCMCMCMCMCOCOCOrO-d-^-o -HcMco^a-m
ininininininininminininininininminmininin
oooooooooooooooooooooo
ininininininininminininminmininiriininintn
71
-------
T3
cu
I
4J
C
O
U
vO
I
CN
w
•J
CO
C
o
•H
CO
§
a
O
Source
Initial
i
•T
C
3
/•^.
e
v^
0]
0)
4-1
nj
C
•H
*O
U
O
8
/-^.
o
cu
CO
60
cu
4-1
^
Pi
C
o
CO
CO
•H
e
W
0
C
)•
C/
** "e
M -^
V
4J O
JS &*
1-1 /-v
to e
0 >-'
1-1
>-i N
> D
^"s
§5
^
0 J-*
3S
a a>
SB
>H
X
&
4-1
O
O
1
60
3
.H
3
I
>>
JS
1) 1-1
; a)
J 43
s §
2 Z
•a
«J
o
(2
CO
CO
CU
o
o
<(
4-1
O
HJ
4J
Pi
rH
CU
??
0
CJ
/^s
r^
N— '
vD m ro
• • •
^H ro n
O 0 O
CN CM CM
o o o
o o o
O ir> m
^ r-» CM
r> vo v£>
* •» *
CM CM CM
r^ p~ r^
CO CO ro
-a- -•
^-v
o
\~s
vO vD
• •
r-4 1— t
O O
CM CM
0 O
O O
O 0
oo a\
c3
^~\
o
1-1
v-^
^OvOvOvOvOvOvOvOfiro
OOOOOOOOOO
CMCMCMCMCMCMCMC-4CMCM
OOOOOOOOOO
OOOOOOOOOO
OOOOOOOOOO
inminininininminm
•— li— li— li— Ir-li— (i— Ii-Hi— li— 1
r^r-^r— i — r~r^r-r^r^r~
CO CO CO CO CO CO CO CO CO CO
-3--d-«*»3'-*»*>3-
-------
TABLE 2-17
PARTICULATE SIZE DISTRIBUTION, GRAVITATIONAL SETTLING VELOCITIES
AND SURFACE REFLECTION COEFFICIENTS FOR THE PARKING LOT
AND PARKING LOT ACCESS ROAD SOURCES
Particle Size
Category
<2.5 pm
2.5 to 15 urn
15 to 30 Urn
>30 ym
Mass
Fraction
0.200
0.570
0.230
0.000
Settling Velocity
(m/sec)
0.0002
0. 0075
0.0407
—
Surface Reflection
Coefficient
1.00
0.77
0.60
—
73
-------
The particulate emission rates presented in Tables 2-13, 2-15 and
2-16 for the roadways, parking lots and parking lot access roads are aver-
ages for the periods before and after the initiation of the Armco dust con-
trol program. The usage of the roadways and parking lots varies according
to the time of day and the day of the week. We accounted for these varia-
tions in the ISCST calculations by multiplying the average emission rates
by the hourly emission rate scaling factors shown in Table 2-18. The
scaling factors in Table 2-18 were based on the following assumptions:
• Roadways - The traffic for the evening shift ("turn") is 60
percent of the traffic for the day shift, the traffic for
the midnight shift is 40 percent of the traffic for the day
shift and the traffic on weekends is 70 percent of the
corresponding traffic on weekdays
• Administration/Staff and Contractor Parking Lots - The
traffic for the evening and midnight shifts is 20 percent of
the traffic for the day shift and the traffic on weekends is
20 percent of the corresponding traffic on weekdays
• Coke Plant Parking Lot - The traffic does not vary with the
shift or the day of the week
• All Other Parking Lots - The traffic for the evening and
midnight shifts is 60 percent of the traffic for the day
shift and the traffic on weekends is 60 percent of the
corresponding traffic on weekdays
The two sources of particulate emissions at the Armco Mill that
were not considered in our emissions inventory are wind-blown dust from
open areas and dust entrained by vehicles traveling on minor roads.
Although sub-area sources covering the Armco Mill property could have been
used to account for the effects of these emissions, twelve of the air
quality monitoring sites are within or adjacent to the mill's property
74
-------
TABLE 2-18
HOURLY EMISSION RATE SCALING FACTORS FOR ROADWAY AND
PARKING LOT SOURCES
Source
4XXXX
501XX
502XX
503XX
504XX
505XX
506XX
508XX
509XX
510XX
511XX
512XX
513XX
Scaling Factor
Weekdays
0100-0800
EST
0.66
0.92
0.92
0.92
0.92
0.56
0.92
0.56
0.92
0.92
0.92
0.92
0.92
0900-1600
EST
1.64
1.54
1.54
1.54
1.54
2.78
1.54
2.78
1.54
1.54
1,54
1.54
1.54
1700-2400
EST
0.98
0.92
0.92
0.92
0.92
0.56
0.92
0.56
0.92
0.92
0.92
0.92
0.92
Weekends
0100-0800
EST
0.46
0.55
0.55
0.55
0.55
0.11
0.55
0.11
0.55
0.55
0.55
0.55
0.55
0900-1600
EST
1.15
0.92
0.92
0.92
0.92
0.56
0.92
0.56
0.92
0.92
0.92
0.92
0.92
1700-2400
EST
0.69
0.55
0.55
0.55
0.55
0.11
0.55
0.11
0.55
0.55
0.55
0.55
0.55
75
-------
boundary, and most of these sites would be within 100 meters of one or more
of the area sources used to represent the property. (The ISC Model cannot
calculate concentrations for source-receptor combinations that are separ-
ated by less than 100 meters.) Also, emissions from the open areas and
roads in the vicinity of the Armco Mill are likely to be similar to emis-
sions from the open areas and minor roads within the Armco property.
Consequently, we assumed that the background particulate concentrations
estimated for the vicinity of the mill also accounted for the effects of
emissions from open areas and minor roads within the mill's property.
ISCLT Source Inputs
With the exception of the particulate emission rates, the ISCLT
source inputs were the same as the ISCST source inputs. Because of the
hour-to-hour and day-to-day variations in emissions from the point and non-
point combustion and process sources, we used the ISCST hourly meteorolog-
ical inputs (see Section 2.2.1.) with the corresponding ISCST hourly source
inputs to calculate, for each point or non-point combustion or process
source, the average emission rates for each combination of wind-speed and
stability categories for the periods before and after the initiation of the
Armco dust control program. These emission rates are listed in Tables D-l
through D-40 in Appendix D. The emission rates used in the ISCST concentra-
tion calculations for the Coke Plant Parking Lot sources, which did not
vary by hour of the day or day of the year, were also used in the ISCLT
concentration calculations. For the roadway sources and the remaining
parking lot sources, the emission rate scaling factors in Table 2-18 were
used to generate "seasonal" average emission rate scaling factors for the
periods before and after the dust control program. Tables D-41 through
D-43 in Appendix D l^ist the average scaling factors for each of the joint
combinations of wind-speed and stability categories. The average scaling
factors in Appendix D were multiplied by the corresponding average emission
rates in Tables 2-13, 2-15 and 2-16 to obtain the emission rates for use in
the ISCLT calculations.
76
-------
Uncertainties in the Particulate Emission Rates
The particulate emission factors and emission rates provided by
PEDCo and used to develop the ISC Model source inputs were divided by PEDCo
into four categories of reliability which were defined as follows:
• Category A - The emission factor or rate was based on a
sound test methodology, and all test methodology and process
operation support data were presented in detail
• Category B- The emission factor or rate was based on a sound
test methodology, but process operation support data were
not presented in detail
• Category C - The emission factor or rate was based on
questionable or unreported test methodology
t Category D - The emission factor or rate was based on
calculations and/or experienced estimates
Table 2-19 gives the percentages of the total daily average
particulate emissions from the Armco Mill contained in each of the four
reliability categories for the periods May through July 1980 (before dust
controls) and August through October 1980 (after dust controls). For the
period prior to 1 August 1980, 78 percent of the emissions are contained in
Category D and 83 percent of the emissions are contained in Categories C
and D. For the period after 1 August 1980, 58 percent of the emissions are
contained in Category D and 63 percent of the emissions are contained in
Categories C and D. Thus, the uncertainties in the emissions for the
period May through July 1980 are larger than for the period August through
October 1980. However, for the purposes of model testing, the uncertain-
ties in the emissions for the entire period May through October 1980 are
significant.
77
-------
TABLE 2-19
PERCENTAGES OF TOTAL DAILY AVERAGE PARTICIPATE EMISSIONS FROM
THE ARMCO MILL CONTAINED IN FOUR RELIABILITY CATEGORIES
Reliability Category
A (Good to Excellent)
B (Fair to Good)
C (Unknown)
D (Fair to Poor)
Percentage of Total Emissions
May - Jul 1980
3.6
13.7
4.8
77.9
Aug - Oct 1980
7.5
29.6
4.6
58.3
78
-------
2.2 METEOROLOGICAL DATA
2.2.1 Meteorological Input Parameters
ISCST Meteorological Inputs
The principal source of meteorological data used to develop the
ISCST meteorological inputs for the period before fugitive dust controls at
the Armco Mill was an instrumented tower located in the middle of a corn
field about 10 kilometers southwest of the mill. Hourly values of wind
direction, wind speed and the standard deviation of the wind azimuth angle
were measured at heights above ground level of 10 and 100 meters. Addition-
ally, hourly values of the ambient air temperature at the 10-meter level
and the temperature difference between the 10-meter and 100-meter levels
were measured. Because of the distance from the tower to the Armco Mill
and the possibility that the winds at the 10-meter level of the tower may
have been affected by local influences, we believe that the winds from the
100-meter level of the tower provide the best available estimates of trans-
port wind directions and speeds at the mill. Our initial plan was to use
the 100-meter level wind directions and wind speeds in the model calcula-
tions, but to use the 10-meter level wind speeds in combination with the
concurrent Dayton, Ohio International Airport cloucl-cover observations to
assign the Pasquill stability category to each hour following the Turner
(1964) approach. (The Turner stability classification scheme is intended
for use with wind speeds measured at or near 10 meters.) We also planned
to use the tower temperature measurements in combination with the con-
current Dayton Airport upper-air soundings to calculate early morning and
afternoon mixing heights for the sample days following the Holzworth (1972)
procedure. Thus, our minimum data requirements for the 100-meter tower on
each sample day were: (1) complete wind-speed observations available for
the 10-meter level, (2) complete wind-speed and wind-direction observations
(without calms or variable wind directions) available for the 100-meter
level, and (3) complete temperature data available.
79
-------
A total of 26 of the 61 sample days satisfied our initial data
requirements. Start-up problems probably account for the tower's low data
recovery rates during early May 1980. Also, the tower was struck by light-
ning at 2300 EST on 2 August 1980, and all sensors on the tower were down
until 1300 EST on 16 August 1980 when measurements of wind speeds at the
10-meter and 100-meter levels resumed. The tower did not become fully
operational until 3 October 1980. Thus, no tower data are available for
the majority of the period with fugitive dust controls at the Armco Mill.
After consultation with the EPA Project Officer, it was decided to use the
tower meteorological data in the ISC Model calculations for the period
before dust controls at the Armco Mill and Dayton Airport meteorological
data in the calculations for the period after dust controls.
The procedures used to generate the hourly meteorological inputs
for the sample days during the period before fugitive dust controls at the
Armco Mill may be summarized as follows:
• The 100-meter level tower wind speeds were converted to
meters per second and the 100-meter wind directions were
reversed 180 degrees to conform to the directions toward
which the wind is flowing as required for input to ISCST;
for the days with wind directions reported to the nearest 5
degrees rather than to the nearest degree, random numbers in
0.5-degree increments between -2.5 and +2.5 degrees were
added to the reversed wind directions
• The 10-meter tower wind speeds and the concurrent Dayton
Airport cloud-cover observations were used with the meteoro-
logical preprocessor program for standardized short-term
dispersion models such as ISCST to calculate the stability
category during each hour of each sample day
• The tower temperature measurements in combination with the
Dayton Airport upper-air soundings were used to calculate
80
-------
Holzworth early morning and afternoon mixing heights for
each sample day; these mixing heights were used with the
meteorological preprocessor program to estimate hourly
mixing heights
• The tower temperature measurements were converted to degrees
Kelvin
• The ISCST default values for the wind-profile exponents and
vertical potential temperature gradients were used in the
model calculations
It was possible to follow the above procedures for 19 of the 29
sample days during the period before fugitive dust controls. The tower
data available for 3, 9, 12, 15 and 27 May 1980 were insufficient to deve-
lop model meteorological inputs. However, we increased the total number of
days for model testing from 19 to 24 by performing data substitutions as
follows:
• 6, 18 and 24 - Substituted the Dayton Airport surface air
May 1980 temperatures for the missing tower
temperatures
• 21 May 1980 - Substituted the 10-meter level wind speeds
and directions at 2300 and 2400 EST for the
100-meter level values (the 100-meter winds
were calm at 2300 and variable at 2400)
• 26 July 1980 - Substituted the 10-meter level wind speeds
and directions at 2300 and 2400 EST for the
100-meter level values (the 100-meter winds
were variable at 2300 and 2400)
81
-------
The hourly meteorological inputs for the 24 sample days during the period
before dust controls that were used in the ISCST calculations are listed in
Appendix C.
The Dayton Airport is located about 49 kilometers north-northeast
of the Armco Mill. Because of the distance from the Dayton Airport to the
mill and the relatively low airport wind measurement height (6.7 meters),
the representativeness of the Dayton Airport wind data of conditions at the
mill is unknown. We compared the Dayton Airport wind data with the 100-
meter tower wind data for nine days during July and October 1980 for which
wind data are available for both levels of the tower. The 24-hour average
wind directions at both tower levels and at the Dayton Airport agreed to
within 20 degrees except on a day with very light winds at both sites.
However, the hour-by-hour correspondence between the Dayton Airport and the
tower wind directions often was poor. The 24-hour average wind speeds at
the Dayton Airport were almost identical to the corresponding 24-hour
average wind speeds at the 100-meter level of the tower and about double
the wind speeds at the 10-meter level of the tower. We do not know of any
explanation for this difference in mean wind speeds. We also inspected the
Dayton Airport surface weather observations for the sample days during
August and September 1980 for which no tower wind data are available. Many
of these days were characterized by light and variable winds and localized
effects on the winds such as thunderstorms passing over the Dayton Airport.
Wind data from the Dayton Airport are least likely to be representative of
conditions at the Armco Mill during periods of light and variable winds or
during periods of showers. We conclude that the absence of onsite meteoro-
logical data adds an uncertainty to the results of the ISC Model calcula-
tions that is likely to be significant, but is impossible to quantify.
The procedures that we used to develop, from the Dayton Airport
surface and upper-air meteorological data, the hourly meteorological inputs
for the 32 sample days after the addition of fugitive dust controls at the
Armco Mill were the standard procedures for using airport surface and upper-
82
-------
air meteorological data to develop ISCST meteorological inputs. As in the
case of the ISCST calculations made using the tower wind and temperature
data, we used the ISCST default values for the wind-profile exponents and
vertical potential temperature gradients. The hourly meteorological inputs
for the sample days with fugitive dust controls that were used in the ISCST
calculations are contained in Appendix C.
Less than 50 percent of the area contained within a 3-kilometer
radius of the center of the main production area at the Armco Mill can be
assigned to Auer (1978) land use types II, 12, Cl, R2 and R3. Consequently,
the EPA Project Officer specified that the Rural Mode be used in the ISC
Model calculations for consistency with current EPA guidance on the use of
rural or urban dispersion coefficients and mixing heights in dispersion
model analyses. However, the ISC Model User's Guide (p. 2-4) notes that
"... the urban options may also be considered in modeling an industrial
source complex in a rural area if the source complex is large and contains
numerous tall buildings and/or large heat sources (for example, coke ovens)."
The Armco Mill resembles this description of an industrial source complex
in a rural area where the use of an urban mode might be appropriate. Addi-
tionally, Bowers and Anderson (1981) found that the ISC Model's performance
in Urban Mode 2 was superior to the model's performance in the Rural Mode
at a large industrial source complex that is in an area that would be de-
fined as "rural" following current EPA guidance. !*• would therefore be of
interest to perform the ISC Model calculations for the Armco data set in
both Urban Mode 2 and the Rural Mode. However, the use of an urban mode
was beyond the scope of this study. (A brief discussion of the ISC Model's
rural and urban options is given in Appendix A.)
ISCLT Meteorological Inputs
The principal meteorological inputs to the ISCLT program are STAR
summaries (statistical tabulations of the joint occurrence frequencies of
wind-speed and wind-direction categories, classified according to the
83
-------
Pasquill stability categories). We used the ISCST hourly meteorological
inputs for the 24 sample days during the period before fugitive dust con-
trols at the Armco Mill and for the 32 sample days during the period with
fugitive dust controls to generate STAR summaries for the two periods.
(The ISCST wind-direction inputs were converted to directions from which
the wind is blowing as required for input to ISCLT.) The two STAR sum-
maries are listed in Appendix C. Additionally, we used the hourly tempera-
ture observations to calculate the mean ambient air temperature for each
combination of sample period and Pasquill stability category. Table 2-20
lists the resulting mean temperatures. The ISC Model default values for
the wind-profile exponents and vertical potential temperature gradients
were used in the ISCLT calculations. We also determined the median rural
mixing height for each combination of wind-speed and stability categories
for each of the two sample periods. These mixing heights are listed in
Table 2-21. Because the ISC Model currently assumes unrestricted vertical
mixing during hours with the E or F stability category in rural areas, the
mixing heights for the E and F stability categories are defined as 10,000
meters in Table 2-21.
2.2.2 Climatological Representativeness of the Sample Period
Introduction
It is of interest to assess the climatological representativeness
of meteorological conditions in the Middletown, Ohio area during the period
May through October 1980 because certain meteorological parameters can in-
fluence ambient particulate concentrations. These meteorological parameters
include: (1) precipitation and relative humidity, (2) mean wind speed and
wind direction, and (3) atmospheric stability. This section first discusses
how these meteorological parameters can affect ambient particulate concen-
trations and then compares the values of these parameters during the period
May through October 1980 with the corresponding climatological averages.
The sources of climatological data used in this section are the monthly
precipitation summaries for the Miami Conservancy District,
84
-------
TABLE 2-20
MEAN AMBIENT AIR TEMPERATURES IN DEGREES KELVIN USED IN THE
ISCLT CONCENTRATION CALCULATIONS
Pasquill Stability
Category
A
B
C
D
E
F
Sample Period
Before Dust Controls
298.6
298.1
297.1
294.8
292.8
291.7
After Dust Controls
*
298.0
297.6
291.0
288.5
287.7
There were no occurrences of A stability during the sample days after
the initiation of fugitive dust controls.
85
-------
TABLE 2-21
MEDIAN RURAL MIXING HEIGHTS IN METERS USED IN THE ISCLT
CONCENTRATION CALCULATIONS
Pasquill Stability
Category
Wind Speed (m/sec)
0-1.5
1.6-3.0
3.1-5.1
5.2-8.2
(a) Before Fugitive Dust Controls
A
B
C
D
E
F
*
1,100
1,100
900
10,000
10,000
1,100
1,100
650
1,100
10,000
10,000
900
1,100
900
1,100
10,000
10,000
*
1,100
1,200
1,100
10,000
10,000
8.3-10.8
>10.8
*
*
525
700
10,000
10,000
*
*
*
1,100
10,000
10,000
(b) With Fugitive Dust Controls
A
B
C
D
E
F
*
*
900
*
10,000
10,000
*
1,350
1,350
1,000
10,000
10,000
*
1,350
1,100
1,100
10,000
10,000
*
*
*
1,350
10,000
10,000
*
*
*
1,500
10,000
10,000
*
*
*
*
10,000
10,000
* There were no occurrences during the sample days of the indicated combin-
ation of wind-speed and stability categories.
86
-------
which were provided by Armco (Grantz and Steiner, 1981), and the 1980
meteorological summary with comparative data for the Dayton, Ohio Inter-
national Airport (EDIS, 1981). The Dayton Airport is the site nearest the
Armco Mill for which detailed climatological data are available. The
period of record for the Dayton Airport climatological data used in this
section is 1941 through 1970.
Precipitation and Relative Humidity
Many studies (for example, Cramer, et a\_., 1976) have found that
the lowest 24-hour average total suspended particulate concentrations in a
region tend to occur on days with precipitation and/or with precipitation
on the previous day. This result can be explained by factors such as pre-
cipitation scavenging of airborne particulates, wetting of fugitive dust
sources such as open storage piles and cleansing of fugitive dust sources
such as paved roads. Theory indicates that the efficiency of precipitation
as a mechanism for removing particulates from the atmosphere increases as
the particle diameter increases (see Slade, 1968, p. 213). High relative
humidities can have effects on ambient particulate concentrations that are
similar to some of the effects of precipitation. For example, high rela-
tive humidities may tend to decrease fugitive dust emissions because wetted
fugitive dust sources do not dry as quickly with high humidities as with
low humidities. On the other hand, high relative humidities in the summer
months appear to be conducive to the formation of small secondary particu-
lates such as sulfates (for example, see Bornstein and Thompson, 1980).
The Miami Conservancy District records 24-hour precipitation
totals at more than 40 sites in the District, including Middletown. The
precipitation total listed for a given day is for the 24-hour period ending
at 0800 EST on that day. Table 2-22 lists the precipitation reported for
Middletown for the 24-hour period ending at 0800 EST on each sample day and
for the 24-hour period ending at 0800 EST on the day following each sample
day. (Table 2-22 includes the sample days that were deleted from the ISC
Model calculations because the available 100-meter tower meteorological
87
-------
TABLE 2-22
SUMMARY OF PRECIPITATION MEASURED AT MIDDLETOWN, OHIO FOR THE 24-HOUR
PERIOD ENDING AT 0800 EST ON EACH SAMPLE DAY AND THE 24-HOUR
PERIOD ENDING AT 0800 EST ON THE DAY FOLLOWING EACH
SAMPLE DAY
Sample Day
3 May 1980
6 May 1980
9 May 1980
12 May 1980
15 May 1980
18 May 1980
21 May 1980
24 May 1980
27 May 1980
30 May 1980
2 Jun 1980
5 Jun 1980
8 Jun 1980
11 Jun 1980
14 Jun 1980
17 Jun 1980
20 Jun 1980
23 Jun 1980
26 Jun 1980
29 Jun 1980
2 Jul 1980
5 Jul 1980
8 Jul 1980
11 Jul 1980
Precipitation (mm)
Period Ending at 0800 EST
on Sample Day
0.00
0.00
0.00
4.83
0.00
29.46
0.00
6.10
0.00
Trace
11.94
0.00
14.99
0.00
0.00
0.00
4.06
0.00
0.00
31.24
0.00
0.00
0.00
10.41
14 Jul 1980 1 0.00
17 Jul 1980
0.00
20 Jul 1980 0.00
23 Jul 1980 ; 3.30
26 Jul 1980 0.00
29 Jul 1980 ' 0.01
Period Ending at 0800 EST on
Day Following Sample Day
0.00
0.00
0.00
12.19
0.00
13.97
0.00
0.25
0.00
2.03
23.37
0.00
0.00
0.00
0.00
0.00
0.00
11.18
0.00
0.00
8.89
3.56
37.59
0.00
0.00
0.00
0.00
0.00
0.00
0.00
88
-------
TABLE 2-22 (Continued)
Sample Day
1 Aug 1980
4 Aug 1980
7 Aug 1980
10 Aug 1980
13 Aug 1980
16 Aug 1980
19 Aug 1980
22 Aug 1980
25 Aug 1980
28 Aug 1980
31 Aug 1980
3 Sep 1980
6 Sep 1980
9 Sep 1980
12 Sep 1980
15 Se'p 1980
18 Sep 1980
21 Sep 1980
24 Sep 1980
27 Sep 1980
30 Sep 1980
3 Oct 1980
6 Oct 1980
9 Oct 1980
12 Oct 1980
15 Oct 1980
18 Oct 1980
21 Oct 1980
24 Oct 1980
27 Oct 1980
30 Oct 1980
Precipitation (mm)
Period Ending at 0800 EST
on Sample Day
1.52
3.30
28.19
1.52
0.00
1.52
14.99
0.00
0.00
0.00
Trace
2.29
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Trace
0.00
0.00
0.00
0.00
43.18
0.00
Period Ending at 0800 EST on
Day Following Sample Day
3.30
0.00
0.00
1.52
0.25
1.78
0.00
0.00
0.00
0.00
0.00
0.00
0.00
3.81
0.00
0.00
0.00
0.00
0.00
0.00 !
0.00
0.76 j
0.00 j
0.00 !
0.00
0.00
0.00
0.00
0.00 I 23.62
0.00 j 7.37
0.00 1 0.00
89
-------
data were insufficient for modeling purposes.) It is of interest to note
that the maximum 24-hour rainfall ever recorded in the Miami River Basin of
179.1 millimeters occurred at Versailes (approximately 70 kilometers north-
west of Middletown) during the period ending at 0800 EST on 29 June 1980,
one of the sample days. However, the corresponding precipitation at
Middletown was only 31.2 millimeters.
Table 2-23 compares the monthly precipitation totals at Middle-
town during the period May through October 1980 with the corresponding
climatological average precipitation totals. Precipitation was normal or
near-normal during May and June 1980, well above normal during July and
August 1980, well below normal during September 1980 and above normal
during October 1980. For the sample period before fugitive dust controls
at the Armco Mill (May through July 1980), precipitation was 26 percent
above normal. Similarly, for the sample period after the addition of dust
controls (August through October 1980), precipitation was 7 percent above
normal in spite of the unusually dry September. Precipitation during the
entire period May through October 1980 was 18 percent above normal.
Table 2-24 compares the monthly average relative humidities at
the Dayton Airport during the period May through October 1980 with the
corresponding climatological average relative humidities. The 1980 monthly
average relative humidities at 0100, 0700, 1300 and 1900 EST are in close
agreement with the climatological averages. The relative humidities in
Table 2-24 also illustrate the high humidities that are characteristic of
southwestern Ohio.
Mean Wind Speed and Wind Direction
The effects of mean wind speed on ambient particulate concentra-
tions are complex. For example, although strong winds act to increase the
initial dilution of emissions from all particulate sources, strong winds
also tend to increase the wind-blown emissions from fugitive dust sources.
Also, the effects of gravitational settling on the mixing of airborne
90
-------
TABLE 2-23
COMPARISON OF MONTHLY PRECIPITATION TOTALS AT MIDDLETOWN, OHIO DURING THE
DETAILED AIR QUALITY MONITORING PROGRAM WITH THE CORRESPONDING
CLIMATOLOGICAL AVERAGE PRECIPITATION TOTALS
Month
May 1980
Jim 1980
Jul 1980
Aug 1980
Sep 1980
Oct 1980
Total Precipitation (mm)
1980
100.3
119.1
158.5
142.2
21.8
81.0
Climatological
Average
101.6
101.9
95.5
81.0
84.6
64.3
Ratio of 1980 and
Climatological Average
Precipitation Totals
0.99
1.17
1.66
1.76
0.26
1.26
91
-------
-3-
CN
I
CM
H
iJ
O
S5
M
OS O
p a
O H
og
M O
ss PLI
o to
w
" e5
Z OS
o o
H O CO
>H W
Z W
M M O
H OS
os a <
o
H S i-3
< >^ U
05 H M
W M CJ
5^9
< < i-3
P C
H M
Z <
O
S§
fa i-j
O M
e H
o w
CO Q
PM
§
u
•a
cfl
o
oo
o>
M-4
0
•l-l
4J
£
V
**
•^^
S-.
iJ
•a
1
slative t
PS
01
•H
JJ
•a
•H
e
£
0)
>
u
«
rH
(S
0)
DC
l-i
01
^
H
CO
W
O
O
r— 4
H
lfi>
K
O
0
^H
H
eft
H
O
0
0
H
to
H
0
0
O
J=
4J
1
H
W
W
O
c^
t— )
H
CO
H
0
m
i — i
H
CO
w
o
r-
O
H
CO
W
O
i— H
O
Average
o
00
ON
Average
o
oo
ON
*-4
Average
0
00
ON
i — i
Average
o
00
&
r-H
vO OO ro ON O 00
ON ON O O O ON
o o -* — < — < o
O NO CM O CN1 OO
O ON O •— i O ON
•—I O ' — 1 I-H i— 1 O
•-< vo o in m CM
O Ov O O O O
•— 1 O •— ' «-H F— 1 i — 1
ON O ON vO CN •— I
ON O ON O C O
O «— t O ' — i »— t i — i
r- oo ON -3- vo m
m in m vo vo vo
m r^ — i o vo -*
in *n vo i — • o vo
-a- <• m oo oo r~
m in m in in in
-a- CN vo -a- ON vo
m »n m NO in m
r^» ON •— * \o vo *~*
r^ rv oo oo oo oo
00 vD i— i O ON CO
r~- r^~ oo CTN oc oo
- ON CM .—i r~
r^ r^ r- oo oo r--
f*") r^ oo r^ ro oo
r^ i^. r~~ oo oo i^-
o o o o o o
OO 00 00 00 OO CO
ON ON ON ON ON ON
>N C iH 60 O. 4J
nj 3 3 s ai o
S *-3 ") < W O
92
-------
particulates to the surface increase as the wind speed decreases, and some
studies (for example, Cramer, et_ a\,, 1976) have found that the highest
total suspended particulate concentrations tend to occur on days with light
winds. The average wind direction during a sample period significantly
affects the particulate concentrations at monitoring sites that are near
major particulate sources because particulate concentrations downwind of a
major source are higher than particulate concentrations upwind of the
source. Table 2-25 compares the monthly average surface wind speeds at the
Dayton Airport during the period May through October 1980 with the corres-
ponding climatological average monthly wind speeds. Although the maximum
percentage difference between the 1980 and climatological monthly average
wind speeds is 24 percent, the maximum absolute difference is only 0.8
meters per second. The prevailing wind direction during each month of the
period May through October 1980 was south-southwest, the same as the
climatological prevailing wind direction.
Atmospheric Stability
Atmospheric stability affects the dispersion of all air pollu-
tants, including particulates. In the absence of direct turbulence measure-
ments, stability classification schemes generally are used to estimate the
intensities of turbulence in the surface mixing layer. For example, the
Turner (1964) scheme uses airport wind-speed and cloud-cover observations
to assign each hour to one of the Pasquill stability categories. In this
section, we use airport surface wind speeds in combination with observa-
tions of percent of possible sunshine and average daytime sky cover as
indices of atmospheric stability. Table 2-25 indicates that the monthly
average surface wind speeds during the period May through October 1980 were
very close to the climatological averages. The monthly percent of possible
sunshine and the monthly average daytime sky cover at the Dayton Airport
during the period May through October 1980 are compared with the correspond-
ing climatological averages in Tables 2-26 and 2-27, respectively. With
the exception of August 1980, Tables 2-26 and 2-27 indicate that solar
insolation during each month of the period was in good agreement with the
93
-------
TABLE 2-25
COMPARISON OF MONTHLY AVERAGE MEAN WIND SPEEDS AT DAYTON, OHIO DURING
THE DETAILED AIR QUALITY MONITORING PROGRAM WITH THE CORRESPONDING
CLIMATOLOGICAL AVERAGE MONTHLY WIND SPEEDS
Month
May 1980
Jun 1980
Jul 1980
Aug 1980
Sep 1980
Oct 1980
Monthly Average Wind Speed (m/sec)
1980
4.02
4.15
4.24
4.15
3.97
4.82
Climatological
Average
4.42
4.06
3.62
3.35
3.71
4.06
Ratio of 1980 and
Climatological Average
Monthly Mean Wind Speeds
0.91
1.02
1.17
1.24
1.07
1.19
94
-------
TABLE 2-26
COMPARISON OF THE PERCENT OF POSSIBLE SUNSHINE AT DAYTON, OHIO DURING
EACH MONTH OF THE DETAILED AIR QUALITY MONITORING PROGRAM WITH
THE CORRESPONDING CLIMATOLOGICAL AVERAGES
Month
May 1980
Jun 1980
Jul 1980
Aug 1980
Sep 1980
Oct 1980
Percent of
1980
54
72
68
55
72
55
Possible Sunshine
Climatological
Average
60
67
67
69
67
61
Ratio of 1980 and
Climatological Average
Possible Sunshine Percentage
0.90
1.07
1.01
0.80
1.07
0.90
TABLE 2-27
COMPARISON OF THE AVERAGE DAYTIME SKY COVER AT DAYTON, OHIO DURING EACH
MONTH OF THE DETAILED AIR QUALITY MONITORING PROGRAM WITH THE
CORRESPONDING CLIMATOLOGICAL AVERAGES
Month
May 1980
Jun 1980
Jul 1980
Aug 1980
Sep 1980
Oct 1980
Average Daytime Sky Cover (Tenths)
1980
6.3
5.8
5.6
6.2
5.7
5.7
Climatological
Average
6.6
6.2
5.8
5.6
5.5
5.4
Ratio of 1980 and
Climatological Average
Daytime Sky Cover
0.95
0.94
0.97
1.11
1.04
1.06
95
-------
corresponding climatological average insolation. Sunshine during August
1980 was only 80 percent of normal and the sky cover was about 11 percent
above normal. These results are consistent with the above normal precipi-
tation. We conclude from the monthly average surface wind speeds and the
monthly sunshine and sky-cover data that, with the possible exception of
August 1980, the average atmospheric stability closely corresponded to the
climatological norm during each month. The increased mean wind speed and
sky cover and the reduced sunshine during August 1980 suggest that the mean
stability during this month probably was closer to the neutral Pasquill D
stability category than is the climatological norm.
Summary
During the period May through October 1980, the meteorological
parameters other than precipitation that are most likely to affect ambient
particulate concentrations in the Middletown area did not differ signifi-
cantly from the corresponding climatological averages. The most signifi-
cant deviations of precipitation from the climatological averages occurred
during July and August 1980 when precipitation was well above normal and
during September 1980 when precipitation was well below normal. Also, pre-
cipitation was above normal for the 3-month sample period before the .addi-
tion of dust controls at the Armco Mill, for the 3-month sample period with
dust controls and for the entire 6-month sample period. Consequently, any
conclusions about the effectiveness of the Armco dust control program that
are based on a comparison of the particulate concentrations during the
3-month sample period before dust controls with the particulate concentra-
tions during the 3-month sample period after dust controls probably should
take into account the precipitation during the two sample periods.
96
-------
SECTION 3
CALCULATION PROCEDURES
The primary objective of the study described in this report was
to compare calculated and observed particulate concentrations for the
following particle size categories:
• Particle diameters less than 100 micrometers (total
suspended particulates)
• Particle diameters less than 2.5 micrometers
• Particle diameters between 2.5 and 15 micrometers
• Particle diameters less than 15 micrometers
It was therefore necessary to modify the Industrial Source Complex (ISC)
Dispersion Model computer codes ISCLT and ISCST to output concentrations by
user-specified combinations of particle-size categories. These modifica-
tions were performed by the authors of the two computer codes (J. R.
Bjorklund and C. S. Cheney). We compared the calculations of the modified
ISCLT computer code with the calculations of the unmodified version of the
code to ensure the equivalence of the concentrations calculated by the two
versions of ISCLT. Similarly, both we and Schewe (1981) verified that the
modified ISCST computer code gives results that are identical to the
unmodified version of the code.
Table 3-1 summarizes the ISC Model options and general input
parameters used in the concentration calculations. Terrain effects were
not included in the model calculations for two reasons. First, the deriva-
tion of the ISC Model's gravitational settling/dry deposition option
assumes flat terrain. Second, only one of the particulate air quality
97
-------
TABLE 3-1
SUMMARY OF ISC MODEL OPTIONS AND GENERAL INPUT PARAMETERS USED IN
THE CONCENTRATION CALCULATIONS
Option or Input Parameter
Option Exercised or Parameter Value
Terrain Effects
Rural/Urban Mode
Wind-Profile Exponents
Vertical Potential Temperature
Gradients
Distance-Dependent Plume Rise
Briggs (1973) Stack-Tip Downwash
Plume-Rise Entrainment Coefficients
Wind Measurement Height* (m)
Pollutant Decay
Building Wake Effects
Gravitational Settling/Dry Deposition**
No
Rural
Default Values
Default Values
Yes
No
Default
100, 6.7
No
Yes
Yes, No
**
The 100-meter wind data from the meteorological tower were used in the
model calculations for the period before fugitive dust controls and
the 6.7-meter wind data from the Dayton International Airport were
used in the model calculations for the period after fugitive dust
controls.
The model calculations of total suspended particulate concentrations
were performed both with and without the gravitational settling/dry
deposition option as a test of this unique ISC Model feature.
98
-------
monitoring sites is at an elevation above the elevation of the Armco Mill
(see Section 1.3). As discussed in Section 2.2.1, the Rural Mode was speci-
fied by the EPA Project Officer for use in the model calculations. In the
absence of sufficient information to develop site-specific wind-profile
exponents and vertical potential temperature gradients, the ISC Model's
default values for these parameters were used in the concentration calcula-
tions. Emissions from almost all of the point (stack) sources at the Armco
Mill are potentially subject to building wake effects, and the ISC Model's
building wake effects option was exercised in the concentration calculation
for these sources. Because the distance-dependent (transitional) plume
rise option is specifically intended for use with the building wake effects
option, distance-dependent plume rise was calculated for the stack sources.
Following the general guidance provided in the ISC Model User's Guide
(p. 2-53), the Briggs (1973) stack-tip downwash correction was not used in
the model calculations for the stack sources. In the absence of source-
specific information, the default values for the adiabatic and stable
plume-rise entrainment coefficients (Briggs, 1975) were used in the plume
rise calculations. The wind measurement height for the sample days before
fugitive dust controls at the Armco Mill was 100 meters and the wind
measurement height for the sample days with dust controls was 6.7 meters.
The concentration calculations assumed no removal (decay) or addition of
particulates by chemical transformations.
The total suspended particulate concentration calculations were
performed both with and without consideration of the effects of gravita-
tional settling and dry deposition. The particle-size categories for the
set of ISC Model calculations that considered these effects were: (1)
particle diameters less than 2.5 micrometers, (2) particle diameters
between 2.5 and 15 micrometers, (3) particle diameters between 15 and 30
micrometers, and (4) particle diameters above 30 micrometers. To minimize
computer costs, a fifth "size category" was used by ISCST to calculate
total suspended particulate concentrations without the effects of gravita-
tional settling and dry deposition. This fifth size category had a mass
fraction 0 of unity (100 percent of the emissions), a settling velocity
99
-------
V of zero and a surface reflection coefficient Y of unity (no dry
deposition).
Table 3-2 lists the sample days considered in the ISCST and ISCLT
concentration calculation. For each sample day, we used the modified ISCST
computer code with the source inputs given in Section 2.1 and the hourly
meteorological inputs given in Appendix C to calculate 24-hour average
particulate concentrations at the fourteen particulate air quality monitor-
ing sites shown in Figure 1-1 in Section 1.2. The Universal Transverse
Mercator (UTM) X and Y coordinates of these monitoring sites are listed in
Table 1-1. Similarly, for the period before fugitive dust controls and for
the period with fugitive dust controls, we used the modified version of the
ISCLT computer code with the source inputs discussed in Section 2.1 and the
"seasonal" meteorological inputs given in Section 2.2.1 and Appendix C to
calculate "seasonal" average particulate concentrations for the fourteen
air quality monitoring sites.
100
-------
TABLE 3-2
SUMMARY OF THE SAMPLE DAYS INCLUDED IN THE ISC MODEL CALCULATIONS
Sample Day
Before Dust Controls
6 May 1980
18 May 1980
21 May 1980
24 May 1980
30 May 1980
2 Jun 1980
5 Jun 1980
8 Jun 1980
11 Jun 1980
14 Jun 1980
17 Jun 1980
20 Jun 1980
23 Jun 1980
26 Jun 1980
29 Jun 1980
2 Jul 1980
5 Jul 1980
8 Jul 1980
11 Jul 1980
14 Jul 1980
17 Jul 1980
20 Jul 1980
23 Jul 1980
26 Jul 1980
After Dust Controls
29 Jul 1980
1 Aug 1980
4 Aug 1980
7 Aug 1980
10 Aug 1980
13 Aug 1980
16 Aug 1980
19 Aug 1980
22 Aug 1980
25 Aug 1980
28 Aug 1980
31 Aug 1980
3 Sep 1980
6 Sep 1980
9 Sep 1980
12 Sep 1980
15 Sep 1980
18 Sep 1980
21 Sep 1980
24 Sep 1980
27 Sep 1980
30 Sep 1980
3 Oct 1980
6 Oct 1980
9 Oct 1980
12 Oct 1980
15 Oct 1980
18 Oct 1980
21 Oct 1980
24 Oct 1980
27 Oct 1980
30 Oct 1980
101
-------
SECTION 4
COMPARISONS OF CALCULATED AND OBSERVED PARTICULATE CONCENTRATIONS
4.1 BACKGROUND PARTICULATE CONCENTRATION ESTIMATES
For the purpose of this report, we defined "background" particu-
late concentrations as ambient particulate concentrations attributable to
emissions from sources (natural and anthropogenic) not included in the ISC
Model calculations. Prior to performing the comparisons of calculated and
observed concentrations, it was therefore necessary to remove the effects
of the background concentrations from the observed concentrations. The
particulate air quality measurements provided the only objective basis for
estimating the background concentrations. In the ideal case, at least one
monitoring site was upwind of the Armco Mill throughout the sample day. In
the absence of more detailed air quality data, we were then forced to
assume that the particulate concentration measured at the upwind site
represented a uniform background concentration over the entire Middletown
area. Unfortunately, the ideal situation rarely occurred. Additionally,
it was necessary to specify background concentrations by particle-size
category, and many of the monitoring sites did not provide complete data by
particle size (see Section 1.2). Consequently, in order to obtain
estimates of which monitoring sites were least affected by emissions from
the Armco Mill during each sample day, we developed a computer program that
used the hourly wind data for the sample day and the corresponding hourly
source inputs to determine the "dosage" at each site atttributable to
particulate emissions from the Armco Mill and to rank the sites in
ascending order of "dosage".
We defined the "dosage" D, at the k air quality monitoring site
as
481 24
103
-------
where
. ,
i ,
the emission rate for the i source (total emission rate
for an area source)
the distance between the i source and the k monitoring
site
N
i.j.k
; k sampler is within a 90° sector downwind of
the i source during the j hour
0 ; k sampler is outside of a 90 sector downwind
of the i source during the j hour •*
(4-2)
We assumed that the site with the minimum "dosage" was most likely to be rep-
presentative of the 24-hour average background particulate concentration.
We observed the following rules in assigning a background concen-
tration estimate for each particle-size category:
1. If a monitoring site was upwind of the Armco Mill through-
out the sample day ("dosage" of zero), the concentration
observed at that site was defined as the background
concentration. If multiple sites were upwind throughout
the sample day, the average of the concentrations at the
sites was defined as the background concentration.
2. If no monitoring sites were upwind of the Armco Mill
throughout the sample day, the concentration at the site
least likely to have been affected by emissions from the
104
-------
Armco Mill (i.e., the site with the minimum "dosage") was
defined as the background concentration. If no sites
were upwind and two or more sites had approximately the
same "dosage", the average of the concentrations at the
sites was defined as the background.
The background particulate concentration estimates for the sample
days before and after the initiation of the fugitive dust control program
at the Armco Mill are listed in Tables 4-1 and 4-2, respectively. The
particle-size categories for the background concentrations in the two
tables are: (1) particle diameters less than about 100 micrometers as
determined by a standard hi-vol sampler, (2S) particle diameters less than
about 15 micrometers as determined by a size-selective hi-vol sampler, (2D)
particle diameters less than about 15 micrometers as determined by a dicho-
tomous sampler, (3) particle diameters less than about 2.5 micrometers as
determined by a dichotomous sampler, and (4) particle diameters between
about 2.5 and 15 micrometers as determined by a dichotomous sampler. Sep-
arate background concentration estimates were determined for the size-
selective hi-vol and dichotomous samplers (Size Categories 2S and 2D)
because of the differences in concentrations measured at the same site by
the two types of samplers (see Section 1.3). Tables 4-1 and 4-2 also give,
for each combination of sample day and particle-size category, the monitor-
ing site or sites used to estimate the background concentration. The moni-
toring sites are shown in Figure 1-1 in Section 1.2 and are identified by
name in Table 1-1. For each particle-size category, we averaged the back-
ground concentrations for the sample days before and after the initiation
of the Armco dust control program. The average background concentration
estimates for the two sets of sample days are given in Table 4-3.
It is important to emphasize that the particulate background
concentration estimates in Tables 4-1 through 4-3 are, at best, crude
approximations. First, the background concentrations were not uniform over
the Middletown area. For example, the Armco Research Center in downtown
Middletown (Site 7) on some sample days was upwind of the Armco Mill
105
-------
TABLE 4-1
BACKGROUND PARTICULATE CONCENTRATION ESTIMATES FOR THE SAMPLE
DAYS BEFORE FUGITIVE DUST CONTROLS
Sample
Day
6 May 1980
18 May 1980
21 May 1980
24 May 1980
30 May 1980
2 Jun 1980
5 Jun 1980
Observed Size
Category No.
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
Background Concentration
(yg/m )
122.8
84.7
93.2
28.4
64.8
41.4
35.4
29.6
18.1
11.5
77.0
69.4
60.3
40.5
19.9
55.3
41.0
27.1
18.4
8.7
63.0
64.1
42.0
29.0
13.0
63.0
53.3
39.3
22.4
16.9
57.0
38.0
28.0
16.0
12.0
Monitor (s) Upon Which
Background Is Based
7
7
7
7
7
7
7
7
7
7
2 & 10
8
7 & 9
7 & 9
7 & 9
2 & 7
7
7 & 9
7 & 9
7 & 9
9
7
9
9
9
9 & 2
7 & 8
7
7
7
1 & 9
9
9
9
9
106
-------
Table 4-1 (Continued)
Sample
Day
8 Jun 1980
11 Jun 1980
14 Jun 1980
17 Jun 1980
20 Jun 1980
23 Jun 1980
26 Jun 1980
Observed Size
Category No.
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
Background Concentration
(yg/m )
42.0
23.5
15.0
6.0
9.0
37.0
33.6
21.2
7.4
13.8
89.0
29.0
76.1
39.7
36.4
48.0
37.2
34.1
18.7
15.4
36.2
29.1
13.6
8.2
5.4
87.0
55.1
49.7
29.4
20.3
74.0
55.0
40.0
25.0
15.0
Monitor (s) Upon Which
Background Is Based
2
7
9
9
9
1 & 2
7
7
7
7
9
9
7
7
7
1 & 2
8
7
7
7
2, 7 & 12
7
7
7
7
2
8
7
7
7
2
9
9
9
9
,
107
-------
TABLE 4-1 (Continued)
Sample
Day
29 Jun 1980
2 Jul 1980
5 Jul 1980
8 Jul 1980
11 Jul 1980
14 Jul 1980
17 Jul 1980
Observed Size
Category No.
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
Background Concentration
(yg/m )
37.0
22.0
23.0
11.0
12.0
110.5
97.8
78.7
45.6
33.1
62.0
46.8
30.1
19.2
10.9
79.2
42.8
57.4
31.4
26.0
67.1
54.6
39.8
23.8
16.0
69.0
55.0
47.0
26.0
21.0
68.0
51.2
36.5
14.5
22.0
Monitor (s) Upon Which
Background Is Based
9
9
9
9
9
2 & 10
7 & 8
7
7
7
8
7
7
7
7
7
7
7
7
7
2 & 7 •
7
7
7
7
10
9
9
9
9
2
7
7
7
7
i
108
-------
TABLE 4-1(Continued)
Sample
Day
20 Jul 1980
23 Jul 1980
26 Jul 1980
Observed Size
Category No.
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
Background Concentration
(yg/m )
69.0
66.0
50.0
34.0
16,0
39.0
23.2
34.9
15.3
19.6
94.0
84.0
70.4
52.4
18.0
Monitor (s) Upon Which
Background Is Based
9
9
9
9
9
2, 7 & 8
8
7
7
7
2
7, 8 & 9
7 & 9
7 & 9
7 & 9
109
-------
TABLE 4-2
BACKGROUND PARTICULATE CONCENTRATION ESTIMATES FOR THE SAMPLE DAYS
AFTER FUGITIVE DUST CONTROLS
Sample
Day
29 Jul 1980
1 Aug 1980
4 Aug 1980
7 Aug 1980
10 Aug 1980
13 Aug 1980
16 Aug 1980
•
Observed Size
Category No.
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
Background Concentration
(yg/m )
68.9
47.0
38.0
29.0
9.0
57.2
40.0
37.7
21.1
16.6
86.0
54.0
36.0
25.0
11.0
50.0
50.0
19.0
13.0
6.0
68.0
57.0
55.6
33.6
22.0
61.0
41.6
51.5
27.1
24.4
31.0
24.9
20.8
12.0
8.8
Monitor (s) Upon Which
Background Is Based
7
9
9
9
9
7
9
7
7
7
2
9
9
9
9
9
9
9
9
9
2 & 9
9
7
7
7
2
8
7
7
7
1
8
7
7
7
110
-------
TABLE 4-2 (Continued)
Sample
Day
19 Aug 1980
22 Aug 1980
25 Aug 1980
28 Aug 1980
31 Aug 1980
3 Sep 1980
6 Sep 1980
Observed Size
Category No.
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
Background Concentration
(yg/m )
57.0
58.8
31.0
18.0
13.0
33.8
23.9
46.4
19.8
26.6
109.0
77.3
90.9
49.7
41.2
97.0
91.0
59.0
43.0
16.0
52.0
53.4
27.0
17.0
10.0
41.8
35.3
30.3
17.3
13.0
60.0
48.3
43.0
26.2
16.8
Monitor (s) Upon Which
Background Is Based
2 & 9
7
14*
14*
14*
2 & 8
8
5*
5*
5*
2 & 8
8
5*
5*
5*
9
9
9*
9*
9*
9
7
9
9
9
1 & 8
8
7
7
7
2
7 & 8
7
7
7
1
111
-------
TABLE 4-2 (Continued)
Sample
Day
9 Sep 1980
12 Sep 1980
15 Sep 1980
18 Sep 1980
21 Sep 1980
24 Sep 1980
27 Sep 1980
Observed Size
Category No.
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
Background Concentration
(yg/m )
65.1
61.5
48.8
28.8
20.0
65.0
56.0
42.0
23.0
19.0
41.0
38.1
34.5
18.3
16.2
33.0
37.2
25.9
11.9
14.0
37.0
31.0
29.0
18.0
11.0
41.0
36.2
22.0
13.0
9.0
50.8
36.0
26.0
10.0
16.0
Monitor (s) Upon Which
Background Is Based
7 & 9
7
7 & 9
7 & 9
7 & 9
9
9
9
9
9
1
7 & 14
7
7
7
10
7 & 8
7
7
7
9
9
9
9
9
10
8
14
14
14
7 & 9
7 & 9
9
9
9
112
-------
TABLE 4-2 (Continued)
Sample
Day
30 Sep 1980
3 Oct 1980
6 Oct 1980
9 Oct 1980
12 Oct 1980
15 Oct 1980
18 Oct 1980
Observed Size
Category No.
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
Background Concentration
(yg/m )
47.0
39.0
34.0
12.0
22.0
23.0
20.0
17.0
11.0
6.0
49.0
78.8
30.0
15.0
15.0
86.2
61.4
50.7
20.9
29.8
70.4
24.6
18.0
6.0
12.0
75.0
65.0
53.0
24.0
29.0
26.2
20.6
11.0
4.0
7.0
Monitor (s) Upon Which
Background Is Based
10
9
9
9
9
9
9
9
9
9
9
4
9
9
9
7
8
7
7
7
7 & 12
7
9
9
9
9
9
9
9
9
7 & 9
7 & 9
9
9
9
113
-------
TABLE 4-2 (Continued)
Sample
Day
21 Oct 1980
24 Oct 1980
27 Oct 1980
30 Oct 1980
Observed Size
Category No.
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
1
2S
2D
3
4
Background Concentration
(yg/m )
50.5
39.3
32.3
14.5
17.8
36.0
22.0
25.0
16.0
9.0
42.0
33.0
44.0
36.0
8.0
58.0
47.0
35.0
21.0
14.0
Monitor (s) Upon Which
Background Is Based
7
7
7
7
7
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
* Only dichotomous sampler with data available for the sample day.
114
-------
TABLE 4-3
BACKGROUND PARTICULATE CONCENTRATION ESTIMATES FOR THE SAMPLE PERIODS BEFORE
AND AFTER FUGITIVE DUST CONTROLS
Observed Size
Category No.
1
2S
2D
3
4
Before Dust Controls
No. of
Sample
Days
24
24
24
24
24
Average
Concentration
(yg/m3)
66.1
49.7
43.2
24.2
19.0
Standard
Deviation
(yg/m3)
23.2
20.4
20.7
12.1
12.1
After Dust Controls
No. of
Sample
Days
32
32
32
32
32
Average
Concentration
(yg/m3)
50.3
45.3
36.4
20.5
15.9
Standard
Deviation
(yg/m3)
20.4
17.7
15.7
10.1
7.9
115
-------
throughout the day and still had an observed concentration for particle
diameters less than 15 micrometers that was greater than or equal to the
corresponding concentrations observed at most of the other monitoring
sites. A second problem was that missing data and the lack of size-selec-
tive hi-vol and dichotomous samplers at all monitoring sites frequently
required that the background concentration be based on the concentration
measured at a site that was not the optimum site as indicated by the
"dosage". Because the background concentrations were subtracted from the
observed concentrations prior to performing the comparisons of calculated
and observed concentrations, we believe that the uncertainty in the back-
ground concentration estimates is one of the principal areas of uncertainty
in this study.
4.2 MEASURES OF MODEL PERFORMANCE
Background
A current major topic of discussion addresses the question of
what measures of performance should be applied to atmospheric dispersion
models. EPA has entered into a cooperative agreement with the American
Meteorological Society (AMS) to obtain guidance on the development and
application of dispersion models. To assist in the identification of
possible measures of model performance, the AMS conducted a Workshop on
Dispersion Model Performance at Woods Hole, Massachusetts on 8-11 September
1980. The summary report on the Workshop by Fox (1981) provides several
potential techniques for measuring model performance and notes that,
"Because there is so little experience with these techniques, there is a
great need for research to generate new data, analyze data, and develop and
test new performance measures." Consequently, EPA specified the measures
of model performance proposed by the AMS Workshop as the primary measures
of performance to be used in this study.
116
-------
Basic Definitions of Measures of Performance
The measures of dispersion model performance proposed by the AMS
Workshop are principally based on the differences between observed and
predicted concentrations. The first measure of performance is the bias
which is defined as
N
^
I -\/
(4-3)
where
AX
= Xo.-Xci (4-4)
the bias or average difference between observed and
calculated concentrations
N = the number of pairs of observed and calculated
concentrations used to compute AX
X . = the i observed concentration
oi
X . = the i calculated concentration
ci
The second measure of performance is a measure of the noise in the results
of the mode]
differences
2
of the model calculations and is provided by the variance cr of the
N
a2 = TT^T > (AX - Av-T (4-5)
117
-------
The third measure of performance Is a measure of the gross variability of
the differences and is given by the root mean square (RMS) error
RMS = f;
(4-6)
An additional measure of performance suggested by the AMS Workshop
is the linear correlation coefficient between observed and calculated con-
centrations, defined as
N _ _
E
-------
Definition of Terms Used in This Report
The AMS Workshop suggested that the proposed measures of perfor-
mance be applied to observed and calculated concentrations paired: (1) in
space and time, (2) in time only, and (3) in space only. For the purpose
of performing there comparisons, we define the following terms:
X {x. , t.} = the concentration calculated at the j monitor (loca-
c j i .«
tion defined by the vector x.) on the i day
(x., t. } = the concentration observed at the j monitor (location
defined by the vector x ) on the i day
X {x._., t.} = the maximum concentration calculated for any monitor on
J ^Vi
the i day; the vector x._, defines the location of the
calculated maximum concentration
X {x. ., t } = the observed concentration on the i day at the monitor
whose location is defined by the vector x.,
Xmo {x , t.} = the maximum concentration observed at any monitor on
the i day
X (x., t} = the maximum concentration calculated at the j monitor
J
(location defined by the vector x.) on any of the sample
days
XJJJQ {x.» t} = the maximum concentration observed at the j monitor
(location defined by the vector x.) on any of the sample
days
119
-------
Concentrations Paired in Space and Time
The first set of tests uses concentrations that are paired in
space and time (either for each sample day or for a set of sample days)
to evaluate the total fields of differences. For M monitors and D days,
the bias is given by
M
A* = M S
1-1 J-l
(4-8)
(4-9)
Similarly, the variance is given by
(DM-]
D M
]T (AX±j - AXV
1-1 j-1
(4-10)
and the RMS error is given by
RMS
D M
ffi E
1-1
(A
1/2
(4-11)
The correlation coefficient is given by Equation (4-7) with N equal to the
number of paired observations (N equals M monitors for each sample day and
DM for M monitors and D sample days).
120
-------
Maximum Concentrations Paired in Space and Time
The second set of tests focuses on maximum concentrations that
are paired in space and time, where the locations in space are given by
the locations of the maximum calculated concentrations. The bias is de-
fined as
AXJx, t)
(4-12)
and the variance is given by
(4-13)
- AX_{x,
(4-14)
while the RMS error is given by
RMS
D
1/2
(4-15)
The correlation coefficient is given by Equation (4-7) with N equal to the
D days with paired concentrations (X {x. ,, t.}, X (x. ,, t.}).
o j = l i me 3 = 1 i
We point out that the AMS Workshop suggested that the locations
in space of the maximum observed and calculated concentrations paired in
space and time should be based on the air quality observations rather than
on the model calculations. We did not follow the approach suggested by the
121
-------
AMS Workshop because of our concern with the Armco data set that localized
fugitive sources of particulate emissions not included in the model calcu-
lations may in part be responsible for many of the highest observed concen-
trations (see Section 4.1). Assuming this concern to be valid, the use of
the observed concentrations to define the locations of the maximum concen-
trations paired in space and time will lead to an apparent bias of the
model toward underestimation if the model predictions for the sources
included in the emissions inventory are accurate. For this reason, we be-
lieve that it is preferable in this study to use the results of the model
calculations rather than the air quality observations to define the lo-
cations of the maximum concentrations paired in space and time.
Maximum Concentrations Paired in Time Only
This set of tests is based on the hypothesis that the model can
predict the magnitude of the maximum concentration during any time period
with greater accuracy than it can predict the location of the maximum con-
centration. A major difficulty in applying this hypothesis to the Armco
data set is that the limited number of monitoring sites and receptors is
inadequate to determine the actual locations of the maximum observed and
calculated concentrations. The bias for maximum concentrations paired in
time only is
D
Xmo {V h} - Xmc
122
-------
The variance is given by
(D-l
(4-18)
and the RMS error is given by
RMS
J-l
1/2
(4-19)
The correlation coefficient is defined by Equation (4-7) with N equal to
the D days with paired concentrations (XmQ {x., t.^}, X^fx.j, t^).
Maximum Concentrations Paired in Space Only
This set of tests is based on the hypothesis that, for a long
period of record, the maximum concentration predicted by the model at a
specific location at any time during the period of record should equal the
maximum concentration observed at that location during the period of record.
This hypothesis assumes that the maximum observed and calculated concentra-
tions at a specific location are not paired in time because of factors such
as uncertainties in model inputs and the stochastic nature of atmospheric
turbulence. The principal problem with this hypothesis as applied to the
Armco data set is the limited period of record. The bias for maximum con-
centrations paired in space only is
M
TO
(4-20)
*m
(4-21)
123
-------
where M is the number of monitors. The variance is given by
M
(M-l) Z-r x -m j
and the RMS error is given by
1/2
{ - ^ (4-22)
RMS
M
r/s. i x 2.
(4-23)
Equation (4-7) gives the correlation coefficient with N equal to the M
paired concentrations (y {x., t}, y_ {£., t}).
Problems Associated with Model Evaluation Using the Armco Data Set
The Armco data set presents a number of problems that limit the
applicability to the data set of the proposed AMS measures of model perfor-
mance. First, there are unquantifiable uncertainties in the model source
inputs (see Section 2.1) and meteorological inputs (see Section 2.2.1.).
Also, the observed particulate concentrations attributable to Armco emis-
sions were obtained by subtracking relatively crude background concentra-
tion estimates from the total observed concentrations. The measures of
performance proposed by the AMS Workshop are intended for application to
hourly concentrations stratified into various combinations of meteorological
conditions (atmospheric stability, wind speed, wind direction, etc.).
Because of the 24-hour averaging time of the concentrations in the Armco
data set and the limited number of sample days, it was not possible to
divide the sample days into subsets corresponding to various combinations
of meteorological conditions. Finally, the monitoring network was not
adequate to define the observed concentration fields with sufficient resolu-
tion and the receptor array (identical to the monitoring network) was not
124
-------
adequate to define the calculated concentration fields with sufficient
resolution. The AMS Workshop noted that (Fox, 1981, p. 604):
In some cases, the quality and/or representativeness of
meteorological data and the number of available monitoring
sites may not allow a stratification of concentrations by
meteorological category. Workshop participants agreed
that it is difficult, if not impossible, to evaluate a
model under these circumstances.
4.3 RESULTS OF THE ISCST CALCULATIONS
The results of the ISCST calculations of 24-hour average particu-
late concentrations for the sample days before and after the initiation of
fugitive dust controls at the Armco Mill are presented in Section E.I of
Appendix E. Similarly, Section E.2 of Appendix E contains the statistical
analyses for the individual sample days of the total fields of differences
between observed (minus background) and calculated concentrations for the
set of sample days before dust controls and the set of sample days with
dust controls. The particle-size categories used in the ISCST calculations
and in the comparisons of observed (minus background) and calculated
concentrations are defined as follows:
• Category No. 1 - particle diameter less than about 100
micrometers (total suspended particulates)
• Category No. 2 - particle diameters less than 15 micrometers
• Category No. 3 - particle diameters less than 2.5
micrometers
• Category No. 4 - particle diameters between 2.5 and 15
micrometers
125
-------
The Observed ( - Background) Size Category No. 2 is separated
into Category No. 2S for concentrations measured by size-selective hi-vol
samplers and Category No. 2D for concentrations measured by dichotomous
samplers. Two sets of concentration calculations were made for Size Cate-
gory No. 1. The concentrations for Calculated Size Category No. 1A included
the effects of gravitational settling and dry deposition, while the concen-
trations for Category No. IB did not include these effects. (The concentra-
tions for the other Calculated Size Categories included the effects of
gravitational settling and dry deposition.) We point out that the concen-
trations for Calculated Size Category No. IB correspond to the concentra-
tions that would be calculated using the modeling techniques recommended
for application to particulate sources in the current Guideline on Air
Quality Models (EPA, 1978).
Table 4-4 gives, for the sample days before and after the initia-
tion of fugitive dust controls at the Armco Mill, the results of the
statistical analyses of total fields of differences between observed (minus
background) and calculated 24-hour average particulate concentrations. The
statistical measures of performance in Table 4-4, which are discussed in
detail in Section 4.2, are briefly defined as follows: (1) The bias is the
average difference between observed (minus background) and calculated con-
centrations; (2) The variance is a measure of the noise or scatter in the
differences; (3) The root mean square (RMS) error is a measure of the gross
variability of the differences; and (4) The linear correlation coefficient
is a measure of the degree to which changes in the magnitude of the model
predictions are linearly related to changes in the magnitude of the observed
(minus background) concentrations. The AMS Workshop on Dispersion Model
Performance (Fox, 1981) suggested that the use of confidence intervals for
measures of model performance such as the bias and variance might be appro-
priate in model evaluation studies. However, the Workshop also noted a
number of limitations on the use of confidence intervals and concluded that
"... the statistical significance of model performance measures will not be
established easily." We believe that the use of confidence intervals for
the Armco data set probably is not warranted because of the large
126
-------
TABLE 4-4
STATISTICAL ANALYSES OF TOTAL FIELDS OF DIFFERENCES BETWEEN OBSERVED
AND CALCULATED 24-HOUR AVERAGE PARTICIPATE CONCENTRATIONS
PAIRED IN SPACE AND TIME
Observed (- Background)
Size Category
Calculated Size Category
*
1A
IB
2S
2
2D
2
3
3
4
4
(a) Sample Days Before Dust Controls
No. of Paired Samples
Bias (yg/m )
2 6
Variance (yg /m )
RMS Error (yg/m3)
Correlation Coefficient
229
-60.7
9,799.3
115.9
0.558
229
-79.4
16,527.3
150.8
0.527
136
-46.1
6,137.3
90.6
0.420
81
-45.2
4,477.6
80.4
0.576
82
-17.9
897.4
34.8
0.379
81
-26.8
1,496.2
46.9
0.624
(b) Sample Days After Dust Controls
No. of Paired Samples
3
Bias (yg/m )
7 f\
Variance (yg /m )
RMS Error (yg/m3)
Correlation Coefficient
291
-11.6
1,202.6
36.5
0.576
291
-18.1
1,766.7
45.7
0.558
199
-10.4
637.4
27.2
0.516
83
-9.0
523.9
24.5
0.387
87
-4.7
131.5
12.3
0.328
83
-4.3
157.8
13.2
0.364
* The comparisons of observed (minus background) and calculated concentrations
for Category No. 1A include the effects of gravitational settling and dry de-
position , while the comparisons for Category Ho. IB do not include these
effects.
127
-------
uncertainties in the source inputs (Section 2.1), the meteorological inputs
(Section 2.2.1) and the observed (minus background) concentrations (Section
4.1) as well as the limited number of sample days and monitoring sites
(Section 1.2). Consequently, confidence intervals are not shown for the
comparisons of observed (minus background) and calculated concentrations
presented in this report.
The comparisons of observed (minus background) and calculated
total suspended particulate concentrations (Category No. 1 in Table 4-4)
are probably the best indicators of model performance because the number of
paired samples in this category is much larger than in the other size cate-
gories. The biases of the observed (minus background) and calculated total
suspended particulate concentrations for Size Category No. 1 in Table 4-4
are interpreted as follows:
• For the sample days before dust controls, ISCST overpredicts
the impact of Armco emissions by an average of 61 micrograms
per cubic meter if the effects of gravitational settling and
dry deposition are considered and by 79 micrograms per cubic
meter if these effects are not considered
• For the sample days with dust controls, ISCST overpredicts
the impact of Armco emissions by an average of 12 micrograms
per cubic meter if the effects of gravitational settling and
dry deposition are considered and by 18 micrograms per cubic
meter if these effects are not considered
Thus, the average model performance for the period with dust controls is
much better than for the period before controls. This results could pos-
sibly be explained by the use of meteorological data from different loca-
tions in the model calculations for the two periods. However, we believe
that the meteorological data from the nearby 100-meter tower used for the
period before dust controls are more likely to be representative of condi-
tions at the Armco Mill than the Dayton Airport meteorological data used
for the period with dust controls.
128
-------
To gain insight into the reasons for the differences in average
model performance for the sample days before and after the initiation of
fugitive dust controls, we examined the spatial variations in the differ-
ences between observed (minus background) and calculated total suspended
particulate concentrations. The average overprediction for the period
before dust controls is principally determined by large overpredictions at
monitoring sites internal or immediately adjacent to the Armco Mill (Sites
3, 4, 5, 12, 13 and 14 in Figure 1-1). Also, the large overpredictions for
the sample days before dust controls are primarily determined by the road-
way emissions. (The roadway emissions account for 52 percent of the total
emissions estimated for the Armco Mill before dust controls and for 20
percent of the emissions after dust controls.) The Armco dust control
program consists of wetting, when necessary, the fugitive dust sources
(unpaved roads, storage piles and paved roads). Thus, it is important to
recognize that significant rainfall occurred during the period before dust
controls (see Section 2.2.2). Additionally, the high relative humidities
(50 to 80 percent) and the cloudy skies (average cloud cover of 60 percent)
during this period precluded the rapid evaporation of moisture from the
fugitive dust sources during the intervals between the periods with precipi-
tation. We conclude that the apparent poor model performance during the
period before dust controls is primarily attributable to the use in the
model calculations of roadway and storage pile emission rates that were too
large, especially in view of the occurrence of rainfall and high humidities
during the period. This conclusion is supported by two additional facts.
First, the differences in the particulate concentrations measured before
and after the addition of dust controls are small and not statistically
significant (see Section 1.3). Second, because of the occurrence of rain-
fall and high humidities during the first month of the dust control program,
it frequently was not necessary for Armco to wet the fugitive dust sources
(Grantz and Steiner, 1981). It follows that the dust controls were in fact
present during the first sampling period because of the rainfall and high
humidities.
129
-------
If the effects of gravitational settling and dry deposition are
included in the ISCST calculations for the sample days after the initiation
of fugitive dust controls at the Armco Mill, the average overprediction for
the entire monitoring network of 12 micrograms per cubic meter is primarily
caused by average overpredictions of 19 to 30 micrograms per cubic meter at
monitoring sites where roadway emissions account for large fractions of the
calculated 24-hour average concentrations (Sites 3, 4, 13 and 14 in Figure
1-1). Emissions from non-point combustion and process sources and from
storage piles also contribute significantly to the highest 24-hour average
concentrations calculated at these monitoring sites. PEDCo Environmental,
Inc. based the emission factors for the roadways, storage piles and most of
the non-point combustion and process sources on calculations and/or exper-
ienced estimates (see Section 2.1). Thus, the relatively large average
overpredictions at Sites 3, 4, 13 and 14 during the set of sample days with
dust controls may in part be explained by the use of emission factors that
are too large. Other factors which may contribute to model overpredictions
in the vicinity of the mill include the use of unrealistically low disper-
sion coefficients in the model calculations. For example, because of the
effects of the surface roughness elements and heat sources at the Armco
Mill, we believe that ISC Urban Mode 2 might be more appropriate for use in
the model calculations than the Rural Mode (see Section 2.2.1). The use of
Urban Mode 2 in the model calculations instead of the Rural mode would act
to decrease the average overpredictions.
There is also an apparent bias of ISCST toward overestimation of
the impact of emissions from the Armco Mill arising from the method used to
obtain background particulate concentration estimates from monitor data.
In most cases, the monitor data used to estimate background concentrations
include some contributions from Armco emissions (see Section 4.1). These
background concentrations were subtracted from the observed concentrations
to obtain estimates of the concentrations attributable to Armco emissions.
Thus, the observed (minus background) concentrations for most of the sample
days probably underestimate the actual impact of Armco emissions. It fol-
lows that comparisons of the observed (minus background) concentrations
130
-------
with the corresponding concentrations calculated by a "perfect" model with
"perfect" source and meteorological inputs should indicate an apparent bias
toward overestimation. (We do not know of any objective method for correct-
ing the background concentration estimates to remove this bias.) Addition-
ally, because the effects of precipitation scavenging of particulates were
not included in the model calculations, the model should tend to overesti-
mate concentrations on sample days with significant precipitation.
For the total suspended particulate concentrations (Size Category
No. 1), the variances in Table 4-4 show that the use of ISCST's gravita-
tional settling/dry deposition option reduces the noise in the results and
the RMS errors show that this option improves the model's absolute accuracy.
The differences in the correlation coefficients, which simply reflect the
model's ability to predict trends, are not statistically significant.
Although the use of the gravitational settling/dry deposition option im-
proves the model's performance for both the sample days before dust con-
trols and the sample days with dust controls, this improvement is small in
comparison with the improvement in model performance between the sample
days before and after controls. As discussed above, we believe that the
improvement in model performance between the sample days before and after
controls is primarily explained by the differences in the emission rates
assumed for the roadways and storage piles.
The results of the ISCST concentration calculations summarized in
Table 4-4 for Size Categories No. 2, No. 3 and No. 4 are quantitatively the
same as the results for Size Category No. 1. In all cases, the biases,
variances and RMS errors for the sample days with fugitive dust controls
are much smaller than for the sample days before controls. For Size Cate-
gory No. 2, the average correspondence between observed (minus background)
and calculated concentrations appears to be about the same for the size-
selective hi-vol samplers (Category No. 2S) and the dichotomous samplers
(Category No. 2D). However, it should be remembered that different back-
ground concentrations were used in the comparisons for the two types of
samplers. It is also important to note that, although the lowest biases
131
-------
are found in Category No. 3 and Category No. 4, these categories also have
the lowest average observed concentrations.
The remaining comparisons of observed (minus background) and
calculated concentrations recommended by the AMS Workshop are presented in
Tables 4-5 through 4-7. Table 4-5 compares maximum observed (minus back-
ground) and calculated concentrations paired in space and time for the
sample days before and after the initiation of fugitive dust controls at
the Armco Mill, Table 4-6 compares maximum concentrations paired in time
only for the sample days before and after controls, and Table 4-7 compares
maximum concentrations paired in space only for the sample days before and
after controls. Because of the smaller number of monitoring sites and
sample days, the biases, variances and RMS errors in Tables 4-5 through 4-7
are larger than the corresponding values in Table 4-4. However, the quali-
tative results are the same.
The 24-hour average particulate concentrations that are of concern for
regulatory purposes are the highest or second-highest 24-hour average
concentrations measured or calculated during a year (EPA, 1978).
Consequently, it is of interest to compare the maximum observed (minus
background) and calculated 24-hour average concentrations paired in space
and time, paired in time only, paired in space only, and unpaired in either
space or time (see Fox, 1981, p. 608). These comparisons (biases) ate
shown in Table 4-8 for the periods before and after the initiation of
fugitive dust controls at the Armco Mill. (The calculated concentrations
were used to determine the locations of the maximum concentrations paired
in space and time or in space only in order to calculate the biases shown
in Table 4-8.) For each comparison, the use of ISCST's gravitational
settling/dry deposition option improves the model's performance in
calculating total suspended particulate concentrations by reducing the
bias. Also, the biases for the sample days with dust controls are 4 to 14
times lower than the corresponding biases for the sample days without dust
controls. The least rigorous of the four comparisons is the comparison of
maximum concentrations unpaired in space or time. As might be expected,
Table 4-8 shows that the smallest biases are found for this comparison.
132
-------
TABLE 4-5
STATISTICAL ANALYSES OF DIFFERENCES BETWEEN MAXIMUM OBSERVED AND
CALCULATED 24-HOUR AVERAGE PARTICIPATE CONCENTRATIONS
PAIRED IN SPACE AND TIME
Observed (- Background)
Size Category
Calculated Size Category
1*
1A
IB
2S
2
2D
2
3
3
4
4
(a) Sample Days before Dust Controls
No. of Paired Samples
3
Bias (yg/m )
7 (\
Variance (yg /m )
RMS Error (yg/m3)
Correlation Coefficient
24
-218.9
21,801.4
262.4
0.236
24
-275.0
35,840.2
331.6
0.242
24
-142.0
10,076.5
172.7
0.483
24
-102.2
7,344.4
132.2
0.520
24
-42.3
1,743.6
58.9
0.315
24
-59.5
2,270.4
75.6
0.571
(b) Sample Days after Dust Controls
No. of Paired Samples
3
Bias (yg/m )
7 f\
Variance (yg /m )
RMS Error (yg/m3)
Correlation Coefficient
32
-60.1
3,775.2
85.3
0.310
32
-80.6
4,917.2
106.1
0.269
32
-42.7
1,102.4
53.8
0.285
32
-23.5
802.7
36.5
0.331
32
-12.2
218.1
19.0
0.280
32
-11.4
249.5
19.3
0.276
* The comparisons of observed (minus background) and calculated concentrations
for Category No. 1A consider the effects of gravitational settling and dry
deposition, while the comparisons for Category No. IB do not consider these
effects.
133
-------
TABLE 4-6
STATISTICAL ANALYSES OF DIFFERENCES BETWEEN MAXIMUM OBSERVED AND
CALCULATED 24-HOUR AVERAGE PARTICIPATE CONCENTRATIONS
PAIRED IN TIME ONLY
Observed (- Background)
Size Category
Calculated Size Category
1*
1A
IB
2S
2
2D
2
(a) Sample Days before Dust Controls
No. of Paired Samples
3
Bias (yg/m )
9 £
Variance (yg /m )
RMS Error (yg/m )
Correlation Coefficient
24
-202.9
19,927.2
245.5
0.413
24
-256.8
34,018.3
313.9
0.377
24
-138.3
9,996.0
169.4
0.483
24
-98.8
7,578.9
130.5
0.480
3
3
4
4
24
-41.3
1,775.0
58.2
0.288
24
-57.0
2,423.1
74.7
0.463
(b) Sample Days after Dust Controls
No. of Paired Sample
3
Bias (yg/m )
Variance (yg /m )
RMS Error (yg/m3)
Correlation Coefficient
32
-46.8
3,928.7
77.4
0.280
32
-67.3
4,933.4
96.5
0.266
32
-39.6
1,102.2
51.4
0.292
32
-20.1
807.2
34.4
0.336
32
-10.7
197.7
17.5
0.369
32
-8.9
265.3
18.3
0.256
* The comparisons of observed (minus background) and calculated concentrations
for Category No. 1A consider the effects of gravitational settling and dry
deposition, while the comparions for Category No. IB do not consider these
effects.
134
-------
TABLE 4-7
STATISTICAL ANALYSES OF DIFFERENCES BETWEEN MAXIMUM OBSERVED AND
CALCULATED 24-HOUR AVERAGE PARTICIPATE CONCENTRATIONS
PAIRED IN SPACE ONLY
Observed (- Background)
Size Category
Calculated Size Category
1*
1A
IB
2S
2
2D
2
3
3
4
4
(a) Sample Days before Dust Controls
No. of Paired Samples
3
Bias (yg/m )
Variance (yg /m )
RMS Error (yg/m )
Correlation Coefficient
12
-199.3
26,559.7
253.2
0.931
12
-292.5
54,187.8
367.7
0.930
7
-168.4
15,473.7
204.0
0.543
4
-159.9
8,445.3
178.6
0.832
4
-83.8
3,964.4
100.0
0.674
4
-86.6
1,918.2
94.5
0.906
(b) Sample Days after Dust Controls
No. of Paired Samples
3
Bias (yg/m )
? £
Variance (yg /m )
RMS Error (yg/m3)
Correlation Coefficient
13
-25.4
1,444.7
44.5
0.831
13
-47.9
1,961.0
64.0
0.833
8
-32.0
949.3
43.0
0.583
4
-45.0
620.0
49.9
0.475
^
-24.6
141.5
26.7
0.535
1
4
-18.0
204.9
21.8
0.413
* The comparisons of observed (minus background) and calculated concentrations
for Category No. 1A consider the effects of gravitational settling and dry
deposition, while the comparions for Category No. IB do not consider these
effects.
135
-------
00
n
CO
2
O M
H M
H 06
CJ U
o <
CO
w
H pd
< W
rJ X
& H
CJ M
< M
PM
U U
O Prf
W PM
> Z
< t)
O
re
u u
H CJ
rJ PH
& CO
CJ
U
< z
o
Q
W H
CO
PQ Z
O M
U
h-1
H
Z
s cj
H <
W PM
PQ CO
CO Z
W M
O
W
<
PH
^.
CO
P
CM
CO
CM
*
f-H
X— x
*o
C
o
60 >-,
CJ O
CO 60
PC 01
4-1
1 cfl
v-' CJ
TD 0)
01 N
b. -H
M CO
O)
to
o
^c
CO
CN
CM
PQ
. — 1
^£
r— 1
r**>
Jj
O
60
4-1
cfl
CJ
cu
N
•H
CO
*o
01
4J
CO
rH
3
CJ
i— i
CO
CJ
CO
rH
0
(-1
4J
C
O
CJ
4-1
CO
3
Q
01
IH
O
UH
01
PQ
CO
rS
Cfl
a
01
rH
(X
E
cfl
CO
^— >,
cfl
OO OO -H rH
C7\ CTi 00 OO
in in co co
i— i i— i i— i i— i
I I I I
r-» r-- r— i i— i
O O oo oo
cy* o^ r^ r*-
r— i i— i i— t »— i
1 1 1 1
vo vO 00 00
oo oo o o
i— 1 r— 1 CTl O\
CO CO CM CM
1 1 1 1
.* *— t
LO lO CO CO
CO CO CO CO
1 1 1 1
GO ^Cf 00 00
• • • •
f"*» \,Q r_4 ^H
o ^ in m
o^ oc r** r*1*
i i i i
o*» in cy\ cTv
• • • •
oo r~- CM CM
r~ ~* CM CM
\c vo m in
1 1 1 1
^ CO
co E
E -^
•^ 60
60 -^ 3
3. ^> CO ^
^— x ro £
E ^ , rH O
C rH C
CO C O 01
O CJ
ai oi co
cj o) o o.
cfl E cfl co
a, -H o.
co H co C
•H
C C c
•H -H iH T3
0)
T3 »X3 t3 i-i
01 01 Ol -H
)-i I-l I-l CO
•H -H -H Q.
Cfl CO CO C
E E E E
X X X X
CO
i-H
O
^
4J
(^
O
CJ
4J
CO
•?
O
t-l
01
U-l
CO
>^
CO
o
a)
(X
E
cfl
CO
^•^
fi
>-'
CO CO CO CM
vC V.O C7t CTi
«* Sf CO r-H
1 1 1 1
CM CM .— I i— <
CO CO CT> CT»
m m -H ^-1
I I I I
CM CM r-. r--
rH rH x* «»
f*x ^s» ^f ^f
1 1 1 1
CO CO O O
oo 00 r-. r-
•-< i-l m in
rH ^ | |
1 1
^D ^D ^D f**«
• • • •
r** f**» oo r^>
CN CM ^d- r^
Csl CM i— t |
I 1 1
m tn m vo
• * • •
\O ^D (^** CO
00 00 O CO
rl ~H rH |
1 1 1
^ CO
co E
E -^
~^ 60
60 ^^3.
3. S-* C<1 •*-"
v-' co 6
E ~^ cu
eg — 6C E
E 60 3 -H
•H 3. >- ' H
r . ^._^
^ r4
T3 f^ rH O
C rH C
CC C O 0)
O 0
oi ai co
CJ 0) CJ CX
co E co co
CX -H (X
co H co c
•H
c c c
•H -H -H "C
01
•a TS -o i-i
oi ai oi -H
M l-i l-i CO
•H -H iH p.
CO CO CO C
PL, put pui t>
E E E E
X X X X
co j:
C *
o
•H •
*j a
co o
M -H
4-1 4-1
C -H
0) (0
CJ O
C Pu
O 01
CJ TD
S >,
3 rJ
E T3
•H
!>< 13
CO C
B CO
T3 60 •
01 C CO
4-1 -H 4J
CO rH CJ
rH 4J 01
3 4J U-l
O 0) <*H
rH CO 01
CO
O rH 0)
CO CQ
-O C -H
o co tc
t-l rJ C
60 60 O
^i o
U U-l
cfl O 4J
CO
CO
4-1
3OO
C CU •«
•H U-l
E u-i PQ
T3 0) •
01 f, O
> 4-1 Z
01 »H f^
CO Ol U
f^ *O O
O -H 60
CO 01
01 O cfl
cu O cj
& .
4J -i Ol
s: o j=
H U-l 4J
c
01
136
-------
The estimation of background particulate concentrations criti-
cally affects the results of not only this study, but also of dispersion
model analyses performed for regulatory purposes because the concentrations
of concern for regulatory purposes are obtained by adding the calculated
and background concentrations. Although the same bias is obtained if the
background concentration is subtracted from the observed concentration or
added to the calculated concentration, the ratios of calculated to observed
(minus background) and calculated (plus background) to observed concentra-
tions differ. For example, Table 4-9 compares the maximum observed (minus
background) and calculated 24-hour average total suspended particulate
concentrations paired in space and time, paired in time only, paired in
space only, and unpaired in either space or time. (The calculated concen-
trations in Table 4-9 consider the effects of gravitational settling and
dry deposition.) The maximum 24-hour average concentration calculated for
the sample days with dust controls is 254 micrograms per cubic meter (Site
4 on 15 September 1980). If the background concentration estimated for 15
September 1980 of 41 micrograms per cubic meter is added to this maximum
calculated concentration, the resulting concentration of 295 micrograms per
cubic meter is almost identical to the maximum observed concentration of
292 micrograms per cubic meter (Site 13 on 15 October 1980). However, the
ratio of _he maximum calculated concentration to the maximum observed (minus
background) concentration of 217 micrograms per cubic meter (Site 13 on 15
October 1980) is 1.17.
4.4 RESULTS OF THE ISCLT CALCULATIONS
There are several features of the ISCLT "seasonal" average concen-
tration calculations that should be considered in interpreting the compar-
isons of observed (minus background) and calculated concentrations for the
two "seasons" defined by the sample days before and after the initiation of
fugitive dust controls at the Armco Mill. ISCLT divides the area surround-
ing a point source into sectors of equal angular width corresponding to the
sixteen standard 22.5-degree wind-direction sectors and partitions the
137
-------
TABLE 4-9
COMPARISON OF MAXIMUM OBSERVED (MINUS BACKGROUND) AND CALCULATED
24-HOUR AVERAGE TOTAL SUSPENDED PARTICIPATE CONCENTRATIONS*
Concentration
Pairing
Bias
(yg/m )
Ratio of Calculated to
Observed (minus Background)
Concentrations
Ratio of Calculated
(plus Background) to
Observed Concentrations
(a) Sample Days Before Dust Controls
Space and Time
Time Only
Space Only
Unpaired
-678.9
-617.5
-522.9
-522.9
34.95
8.59
3.97
3.97
8.00
4.90
3.25
3.25
(b) Sample Days After Dust Controls
Space and Time
Time Only
Space Only
Unpaired
-186.5
-186.5
-107.5
-36.6
3.78
3.78
1.74
1.17
2.73
2.73
1.27
1.01
The calculated concentrations consider the effects of gravitational
settling and dry deposition.
138
-------
seasonal emissions according to the wind-direction frequencies. If there
are sufficient occurrences of each combination of wind-direction, wind-
speed and stability categories to generate a representative joint frequency
distribution and if all wind directions within each wind-direction sector
are assumed to be equally probable, it can be shown that the horizontal
distribution of emissions within each sector is uniform (Calder, 1971).
For sixteen wind-direction sectors, six wind-speed categories and six stabil-
ity categories, there are 576 combinations of wind-direction, wind-speed
and stability categories. Although some of these combinations do not occur,
the number of potential combinations is exactly equal to the number of
hours in the set of sample days before dust controls. Thus, the number of
hours in the first "season" may be too small to derive a representative
joint frequency distribution of wind-speed and wind-direction categories,
classified by stability categories. The same reasoning applies to the 768
hours which define the "season" with dust controls. We have no basis for
determining the representativeness of the wind frequency distributions used
in the 1SCLT calculations.
The "seasonal" average particulate concentrations calculated by
ISCLT for the two "seasons" comprised of the sample days before and after
the initiation of fugitive dust controls at the Armco Mill are compared
with the corresponding "seasonal" average observed (minus background) con-
centrations in Tables 4-10 and 4-11, respectively. Table 1-1 in Section
1.2 identifies the monitoring sites and Figure 1-1 shows their locations.
The particle-size categories used in this section are identical to the
particle-size categories defined in Section 4.3.
For the "season" before dust controls, Table 4-10 shows that
ISCLT significantly overpredicts the average concentrations for all
combinations of particle-size category and monitoring site except Category
No. 1A (total suspended particulate concentrations calculated using ISCLT's
gravitational settling/dry deposition option) at Site 2, which is adjacent
to a taxiway at the municipal airport and is probably affected by emissions
from localized background particulate sources. For the "season" after the
139
-------
o
I—I
I
PQ
HH
O
cj
OH
CJ C/3
2 O
w r*i
> H
O
= O
O CJ
sg«
w w
w >
H
w e>
< fa
o w
u on
d P
•^4 H^l
CJ W
PQ
O
5!
/-N CO
Q <
C3 W
O :
o w
cj H
o
to fa
CJ
M 55
> E-
K Z
W W
W CJ
PQ Z
O O
CJ
fa
O
Z
o
C/3
O
CJ
•O
01
Obse
•o
01
in
c
o
•H
4-1
Cfl
IH
4-1
01
CJ
O
CJ
T3
§
O
J?
o
CO
m
i
cfl
cd
to
a.
C
o
C
01
o
o
CJ
ID
0)
o
c
o
60
U
PQ
I
•O
0)
01
CD
X)
O
co
en
P
CM
CN
CN
*
PQ
*
PQ
01
CN
OI •
4J C
oo co
S3- OO
1 1
oo r~-
-* m
1 1 1 1 8 1 1 8 1 1 1 1
in .a-
CN
1 1 1 1 1 1 8 1 1 1 1
o m oo
O CO C7\ OO OO O OO
co vc «* co -H r» oo
IIS 8 I I I I
m vo vo co o- m
i-H
OO CO i-4 *^ CO tT" O *^~ CTs CO *3"
mcMvococo vom CM oomco
I . . 8 •! ...
r^- in -H
CM
r^-3"Cy\cNco mo ^ CM^JCM
vo»-H^~'rovo som oo OI^.CN
• . . i . 8 -I ...
OJ ^H ^O Lf^ ^^ (sj i™H CO *«^ lO O^
m \o
i-H
O-3-CNOOOOOOOOOOCNOOO-^Cn
I-H
i-H in ^^ "sf CO *~~l *-H I-H r~H CN CM f-"3 »^
STOCMCyiOOCN>-H"—iCTvCOi—1|— OS i—I
i-H f-H t-H t—I CM i-H
CMCO*-H^H i—ii—i-3-CN
1 1
O
O
ON
O
O
CN
CM
O
-
O
O
CN
1 1
O
O
^H O
1 1
m o
1 1
CJ
02
vO vO in CM CN O
OO vO vO OO vO O
I I
O
00
vO CN O CO CJ5 OO OO
• . . . C/J . .
C O
CO 0)
UH
60 UH
C 0)
•H
iH 0)
4-1 CO
4j a)
0) 4=
CO 4-1
*-•
cd O
M C
60
O
yn -a
o
PQ
CD i-H
4-1
y «
01 o
01
o
60
01
M CO
01 cj
•a
•H IH
CD O
C H-l
o
o -a
O CJ
Z i-t
CO
;*, o
M
O CD
60 C
01 o
4-1 H
)H 4-1
O C
>4H 01
U
•o c
0) O
4J (J
CO
—I 0)
3 .C
U 4-1
i— I
cfl 0)
CJ i-H.
•H
CD f
C >
o
•H •
•*-> C
CO O
t-l -H
4-1 4-1
C i-l
01 CD
CJ O
c a.
o 01
o -a
0) >-,
.£ M
H T3
140
-------
V)
O =
-
01
Observ
•O
C
CO
•O
01
4J
CO
t-H
U
CO
0
o
CO
•H
4J
e
\*s
e
o
4J
CO
ncent
u
U
rations
oncent
CJ
/-N
C
o
00
u
PQ
1
13
4J
CO
3
CJ
T-t
rt
•o
3
O
00
CO
ca
i
o>
f,
Ol
o
4-1 O
CO
3
CO
Q
CM
CM
CO
CM
i— (
PQ
i-H
i— 1
i— 1
^
CO
CM
i-H
<
•— 4
-T
«n
0
CM
vO CO O
CT* CT» CM
1 1 1 1 1 * t 8 I 1 1 1
o o m
O\ 00
^ o
1 1 1 1 1 1 8 1 1 1 1 8
CM »-)
r- 0
1 t 1 1 1 • t 8 1 1 1 1 8
t— i f— i
^ i-*. r— •— t 10 co o
^^ ^D ^3 ^^ CT^ C*J 00
*— 4 •— 1 CM *— 4 O «—* *4"
^5 00 CO ^^ ^^ *^ O^ f"^ ^^ ^O ^^ PO ^~^
fH O i-H ^-1 t— 1 O ^™t *^" CS C*J r— t i— I CO
c>i o co I*** co co 10 co CM -^ r^» i—4 ^3
CM CM CMi ^^ ^^ | ^D ^^ ^D ^^ GO f*"* *^ ^D
r—4 ^^ r~^ ^^ ^~4 I^J ^^ OO CM »*H ^^ F"H CM
^O^-HCMinr-^r-1-.c^inoo^
CM -H •— l i— I CO i— l
^ONvOvoo^^in-ooo^c^vo
.— l CM i— l i— 1 CM i— 1
o^o-cNvocsvooo^^oor-r-
^^co^rcQcM^^^^S^Jnco
^^^(NrOCMCMoOCMr-Or-O-*
S^S^O^^^^^SSSS
co^^oo O ^" ' CM CO «^~
i— i t— < r-^ i— t *—*
(Q
*a AJ
c o
CO 0)
c
•H
rH Wl
CO c
CO O
•U U
•H
> «
n o
^ c
00
o
U-l T3
CO ^
•U
o •
CJ O
<*-! B
M-l
t)0
Q) CJ
T3
•H U
CO O
C «M
O
cj 73
3
U
O
oc
4J
CO
O
M
O
3
U
CO
o
CO
t-l
4-1
C
01
O
c
o
O
01
f.
C
O
CO
O
a
oi
13
141
-------
initiation of dust controls, comparison of Tables 4-10 and 4-11 shows that
the correspondence between observed (minus background) and calculated
concentrations is significantly improved. For example, the observed (minus
background) and calculated "seasonal" average total suspended particulate
concentrations agree within about a factor of two at ten of the thirteen
monitoring sites if the effects of gravitational settling and dry
deposition are considered and at eight of the thirteen monitoring sites if
these effects are not considered. If ISCLT's gravitational settling/dry
deposition option is used in the total suspended particulate concentration
calculations for the second "season", ISCLT underestimates the observed
(minus background) concentrations only at Site 2, Site 7 (the Armco
Research Center in downtown Middletown) and Site 12 (the Armco Main Gate on
the major highway into Middletown). We believe that these underpredictions
are explained by the effects of emissions from localized background
particulate sources.
Table 4-12 summarizes, for the "seasons" before and after the
initiation of fugitive dust controls, the statistical analyses of total
fields of differences between observed (minus background) and calculated
"seasonal" average particulate concentrations. The results of the ISCLT
concentration calculations for all particle-size categories are qualita-
tively the same as the results of the corresponding ISCST concentration
calculations. For example, the biases of the observed (minus background)
and calculated total suspended particulate concentrations in Table 4-12
show that: (1) For the "season" before dust controls, ISCLT overpredicts
the impact of Armco emissions by an average of 80 micrograms per cubic
meter if the effects of gravitational settling and dry deposition are con-
sidered and by 142 micrograms per cubic meter if these effects are not
considered; and (2) For the "season" with dust controls, ISCLT overpredicts
the impact of Armco emissions by an average of 4 micrograms per cubic meter
if the effects of gravitational settling and dry deposition are considered
and by 12 micrograms per cubic meter if these effects are not considered.
As explained in the discussion of the ISCST results in Section 4.3, we
142
-------
TABLE 4-12
STATISTICAL ANALYSES OF TOTAL FIELDS OF DIFFERENCES BETWEEN OBSERVED AND
CALCULATED "SEASONAL" AVERAGE PARTICIPATE CONCENTRATIONS PAIRED IN
SPACE AND TIME
Observed (- Background)
Size Category
Calculated Size Category
*
1A
IB
2S
2
2D
2
3
3
4
4
(a) "Season" Before Dust Controls
No. of Paired Samples
3
Bias (yg/m )
7 ft
Variance (yg /m )
RMS Error (yg/m3)
Correlation Coefficient
12
-79.8
5,290.4
105.9
0.930
12
-141.5
16,758.2
188.1
0.941
7
-70.7
2,373.6
83.9
0.820
4
-65.5
2616.8
79.1
0.760
4
-27.8
404.2
32.8
0.052
4
-37.8
973.9
46.5
0.725
(b) "Season" After Dust Controls
No. of Paired Samples
3
Bias (yg/m )
2 6
Variance (yg /m )
RMS Error (yg/m3)
Correlation Coefficient
13
-4.4
109.4
11.0
0.870
13
-12.1
229.4
18.9
0.869
8
-11.1
74.6
13.7
0.804
4
-11.2
196.8
16.5
0.189
4
-6.0
31.5
7.7
0.003
4
-5.1
73.6
9.0
0.276
The comparisons of observed (minus background) and calculated concentrations
for Category No. 1A consider the effects of gravitational settling and dry
deposition, while the comparisons for Category No. IB do not consider these
effects.
143
-------
believe that the average ISCLT performance for the period with dust con-
trols is better than for the period before controls primarily because the
roadway and storage pile emission rates used in the model calculations for
the period before controls were too large, especially because of the signi-
ficant rainfall and high humidities. Table 4-12 also shows that the use of
ISCLT1s gravitational settling/dry deposition option improves the overall
correspondence between observed (minus background) and calculated total
suspended particulate concentrations by reducing the bias, the variance and
the RMS error.
We concluded in the discussion of the results of the ISCST concen-
tration calculations that, given "perfect" model inputs and a "perfect"
model, there should be an apparent bias toward overestimation attributable
to the method used to obtain the background particulate concentration esti-
mates. Additionally, the neglect of the effects of precipitation
scavenging in the model calculations should bias the model predictions
toward overestimation, especially at the more distant monitoring sites.
These factors should also result in a tendency for ISCLT to overpredict
concentrations for the Armco data set.
144
-------
SECTION 5
EVALUATION OF THE AIR QUALITY MONITORING NETWORK
The numbered ® symbols in Figure 5-1 show the locations of the
particulate air quality monitoring sites used during the joint Armco/EPA
monitoring program. One of the objectives of the study described in this
report was to assess the adequacy of the monitoring network shown in Figure
5-1 for measuring the air quality impact of Armco emissions. In the follow-
ing evaluation of the air quality monitoring network, we assume that the
particulate concentrations of concern are the concentrations attributable
to emissions from the Armco Mill that occur at or beyond the boundaries of
the Armco property.
The results of the ISC Model (ISCST and ISCLT) 24-hour and "sea-
sonal" average particulate concentration calculations discussed in Sections
4.3 and 4.4 as well as the air quality data summarized for the sample days
in Section 1.3 provide the first basis for evaluating the air quality moni-
toring network shown in Figure 5-1. The results of the model calculations
and the air quality data indicate that, both with and without the use of
fugitive dust controls, the maximum air quality impact of Armco emissions
occurs internal to the Armco property. Additionally, the results of the
model calculations indicate that this maximum impact is primarily deter-
mined by emissions from fugitive dust sources and low-level emissions from
non-point process and combustion sources. The highest calculated and
observed (minus background) 24-hour and "seasonal" average concentrations
during the period May through October 1980 occur at monitoring sites near
the northeast and southwest boundaries of the mill (Sites 3, 4, 6 and 13).
Thus, the highest concentrations that occur at or beyond the boundaries of
the mill during the months of May through October probably are located
along the northeast and southwest property boundaries. We point out that
it is not possible from the results of this study to determine whether the
highest concentrations throughout the entire year are also located along
the northeast and southwest property boundaries. However, Schewe (1981)
145
-------
RECEPTORS NE AND SW
OF MILL WITH HIGHEST
CALCULATED TSP CONCENTRATIONS
- 4JTS
- «370
FIGURE 5-1. Map of the area surrounding the Armco Steel Mill at Middletown,
Ohio. The numbered ® symbols show the locations of particu-
late air quality monitoring sites and the filled circles show
the locations of the receptors used in the ISCLT calculations
to evaluate the representativeness of the monitoring sites.
146
-------
used the ISC Model short-term program ISCST with an Armco particulate emis-
sions inventory based on the emissions inventory used in this study and
sequential hourly meteorological inputs developed from 1977 Dayton Inter-
national Airport surface weather observations and concurrent Wright-
Patterson Air Force Base mixing height data to calculate 24-hour and annual
average particulate concentrations for the monitoring sites shown in Figure
5-1. Schewe's results show that, if the entire year is considered, the
highest calculated 24-hour and annual average concentrations occur at the
monitoring sites near the northeast and southwest boundaries of the mill.
The comparisons of observed (minus background) and calculated
particulate concentrations discussed in Sections 4.3 and 4.4 indicate that,
for the available data, the most accurate particulate concentration calcu-
lation that can be made for the Armco Mill is the calculation of "seasonal"
average concentrations for the period following the addition of the Armco
dust control program. To evaluate the air quality monitoring network shown
in Figure 5-1, we therefore repeated the ISC Model long-term program ISCLT
calculations for the "season" with dust controls using the 90-receptor
array shown by the filled circles in Figure 5-1. This receptor array covers
the areas northeast and southwest of the Armco Mill where both the calcu-
lated and observed concentrations for the existing monitoring network indi-
cate that the highest concentrations attributable to Armco emissions are
likely to occur. The maximum receptor separation in Figure 5-1 is about
200 meters.
Table 5-1 gives, for the areas northeast and southwest of the
Armco Mill, the magnitudes and locations of the highest "seasonal" average
particulate concentrations calculated for the "season" following the initia-
tion of fugitive dust controls at the mill. With the exception of concen-
trations for particle diameters less than 2.5 micrometers, the maximum
concentrations calculated for all particle-size categories occur at the
northeast boundary of the Armco property at a point that is about 465 meters
south-southwest of the SREPCO site (Site 3) and about 220 meters southeast
of the Reeds Yard site (Site 13). The concentrations calculated at this
147
-------
TABLE 5-1
HIGHEST "SEASONAL" AVERAGE PARTICIPATE CONCENTRATIONS CALCULATED
AT OR BEYOND THE BOUNDARIES OF THE ARMCO PROPERTY DURING
THE "SEASON" WITH FUGITIVE DUST CONTROLS
Particle
Diameter
(um)
Concent rat ion*
C**g/m3)
Location
UTM X (m)
UTM Y (m)
(a) Area Northeast of the Arraco Mill
<2.5
2.5 - 15
<15
<100
31.7
43.4
75.1
91.2
725,320
725,320
725,320
725,320
4,374,040
4,374,040
4,374,040
4,374,040
(b) Area Southwest of the Armco Mill
<2.5
2.5 - 15
<15
<100
37.3
26.1
53.8
68.9
725,000
724,740
725,000
724,740
4,372,400
4,373,340
4,372,400
4,373,340
* Calculated concentrations include the effects of gravitational settling
and dry deposition.
148
-------
point for the various particle-size categories are about double the concen-
trations calculated for the SREPCO site and about 30 percent higher than
the concentrations calculated for the Reeds Yard site. The maximum concen-
tration calculated for particle diameters less than 2.5 micrometers, which
occurs on the southwest boundary of the Armco property about 540 meters
south-southeast of the Coke Plant site (Site 4), exceeds the corresponding
concentration calculated for the Coke Plant by about 65 percent. The
maximum concentration calculated at or beyond the southwest boundary of the
Armco property for particle diameters less than 15 micrometers also occurs
at this point. The maximum concentrations calculated at or beyond the
southwest boundary of the Armco property for the other particle-size cate-
gories occur at a point that is about 445 meters north of the Coke Plant
site and about 734 meters east- southeast of the Oneida School site (Site
6). The concentrations calculated at this point for the various particle-
size categories are about double the concentrations calculated for the
Oneida School site and 25 to 35 percent higher than the concentrations
calculated for the Coke Plant site.
In summary, the results of the ISCLT concentration calculations
made using a detailed receptor array with a 200-meter spacing of receptors
indicate that the concentrations measured at the Reeds Yard site (Site 13)
provide the best available estimates of the highest concentrations attribu-
table to emissions from the Armco Mill in the area northeast of the mill.
Similarly, the results of the ISCLT calculations indicate that the concen-
trations measured at the Coke Plant site (Site No. 4) provide the best
available estimates of the highest concentrations attributable to emissions
from the Armco Mill in the area southwest of the mill. However, both the
Reeds Yard and Coke Plant sites are on Armco property. The concentrations
measured at the SREPCO site (Site 3) and the Oneida School site (Site 6),
which are not on Armco property, provide the second-best available esti-
mates of the highest concentrations attributable to emissions from the
Armco Mill in the areas northeast and southwest of the mill.
149
-------
SECTION 6
CONCLUSIONS AND RECOMMENDATIONS
6.1 CONCLUSIONS
The principal conclusion of the study described in this report is
that the use of the ISC Model's gravitational settling/dry deposition option
yields calculated total suspended particulate concentrations that are in
better agreement with the corresponding observed (minus background) concen-
trations than the concentrations calculated without consideration of the
effects of gravitational settling and dry deposition. Although the differ-
ences in the measured average ambient particulate concentrations between
the sample days before and after the initiation of the fugitive dust con-
trol program at the Armco Mill are relatively small, there are significant
differences in the concentrations calculated for the two periods by the ISC
Model computer programs ISCST and ISCLT. The differences in the concentra-
tions calculated for the two periods are primarily explained by the differ-
ent emission rates assigned to the two periods for the storage pile and
roadway sources. In retrospect, we believe that emission rates for the
uncontrolled storage pile and roadway sources comparable to those used for
the controlled storage pile and roadway sources should also have been used
in the model calculations for the period before dust controls to account
for the effects of rainfall and high humidities during this period. Conse-
quently, we believe that conclusions about the accuracy of ISCST and ISCLT
should be based primarily on the results of the calculations for the period
with dust controls.
Because of the uncertainties in the emissions data, the meteor-
ological data and the observed (minus background) particulate concentra-
tions as well as other limitations such as sample size, it is not possible
to evaluate the absolute accuracy of the ISC Model as a particulate disper-
sion model. However, the apparent bias toward overestimation found in the
results of both the ISCST and ISCLT concentration calculations is expected
151
-------
because of the uncertainties in the observed (minus background) concentra-
tions as well as the neglect in the model calculations of the effects of
precipitation. If allowance is made for the expected bias toward overesti-
mation, we believe that the average overpredictions by ISCST and ISCLT are
relatively small and, in combination with the results of tests of ISCST's
gravitational settling/dry deposition option described by Bowers and
Anderson (1981), support the use of the ISC Model as a particulate disper-
sion model.
6.2 RECOMMENDATIONS
The comparisons of observed (minus background) and calculated
24-hour and "seasonal" average particulate concentrations discussed in this
report indicate that the particulate emission factors used to develop the
emission rates for the roadways critically affect the concentrations calcu-
lated for the Armco Mill. Consequently, our principal recommendation is
that the roadway emission factors used in this report be reviewed to deter-
mine if they are the most appropriate of the available emission factors.
Also, we believe that the roadway emission factors for the period before
the initiation of the fugitive dust control program at the Armco Mill should
account, insofar as possible, for the effects of the significant rainfall
and high relative humidities during this period. If a review of the roadway
emission factors indicates that more appropriate emission factors are avail-
able for the period before dust controls and/or for the period with dust
controls, we recommend that the ISC Model calculations described in this
report be repeated using the updated emission factors.
As discussed in Section 2.2.1, the principal source of the meteor-
ological inputs used in the ISC Model calculations for the sample days
before the addition of fugitive dust controls at the Armco Mill was a
100-meter meteorological tower and the principal source of the meteorolog-
ical inputs used in the model calculations for the sample days with dust
controls was the National Weather Service (NWS) station at the Dayton
152
-------
International Airport. It was not possible to use the 100-meter tower
meteorological data in the model calculations for the sample days with dust
controls because of the damage sustained by the tower when it was struck by
lightning on 2 August 1980. The use of meteorological data from two differ-
ent locations in the model calculations for the two sets of sample days
complicates any comparisons of the concentrations calculated before and
after the initiation of the dust control program. We therefore recommend
that the ISC Model concentration calculations for the sample days before
fugitive dust controls be repeated using meteorological inputs based on
Dayton Airport meteorological data.
153
-------
REFERENCES
Auer, A. H., 1978: Correlation of land use and cover with meteorological
anomalies. Journal of Applied Meteorology, 17, 636-643.
Bernstein, R. D. and W. T. Thompson, 1980: Analysis of SURE aircraft sul-
fate data. Preprint Volume for the Second Joint Conference on
Applications of Air Pollution Meteorology, American Meteoro-
logical Society, Boston, Mass.
Bowers, J. F. and A. J. Anderson, 1981: An evaluation study for the
Industrial Source Complex (ISC) Dispersion Model. EPA Report No.
EPA-450/4-81-002 (NTIS Accession No. PB81-1765391), U. S.
Environmental Protection Agency, Research Triangle Park, North
Carolina.
Briggs, G. A., 1971: Some recent analyses of plume rise observations. In
Proceedings of the Second International Clean Air Congress,
Academic Press, New York.
Briggs, G. A., 1973: Diffusion estimates for small emissions. ATDL
Contribution File No. (Draft) 79, Air Resources Atmospheric
Turbulence and Diffusion Laboratories, Oak Ridge, Tennesse.
Briggs, G. A., 1975: Plume rise predictions. Lectures on Air Pollution
and Environmental Impact Analyses, American Meteorological
Society, Boston, Mass.
Calder, K. L., 1971: A climatological model for multiple source urban air
pollution. Proc. 2nd Meeting of the Expert Panel on Air Pollu-
tion Modeling, NATO Committee on the Challenges of Modern Soci-
ety, Paris, France, July 1971, 33.
Cramer, H. E., et^ _a_l., 1976: Assessment and updating of particulate
emissions data for the Southwest Pennsylvania Intrastate Air
Quality Control Region. H. E. Cramer Company, Inc. Technical
Report TR-76-104-01, prepared for U. S. Environmental Protection
Ageny, Region III, Philadelphia, Pennsylvania.
Environmental Data and Information Service, 1981: Local climatological
data: Annual summary with comparative data - 1980 Dayton, Ohio.
Climatological Summary, National Climatic Center, Asheville,
North Carolina.
Environmental Protection Agency, 1978: Guideline on air quality models.
EPA Report No. EPA-450/2-78-027, OAQPS No. 1.2-080. U. S.
Environmental Protection Agency, Research Triangle Park, North
Carolina.
155
-------
Environmental Protection Agency, 1979: Industrial Source Complex (ISC)
Dispersion Model user's guide. EPA Reports EPA-450/4-79-030 and
EPA-450/4-79-031 (NTIS Accession Numbers PB80-133044 and
PB80-133051), U. S. Environmental Protection Agency, Research
Triangle Park, North Carolina.
Fox, D. G., 1981: Judging air quality model performance: A summary of the
AMS Workshop on Dispersion Model Performance. Bulletin American
Meteorological Society, 6>2_(5) , 599-609.
Grantz, J. A., 1981a: Inhalable particulate matter in the vicinity of an
integrated iron and steelmaking complex. APCA Paper No. 81-5.4,
presented at the 74th Annual Meeting of the Air Pollution Control
Association, Philadelphia, Pennsylvania, June 21-26, 1981.
Grantz, J. A., 1981b: Private communication (10 February 1981 letter to
A. Anderson, H. E. Cramer Company, Inc.).
Grantz, J. A. and B. A. Steiner, 1981: Private communication (24 February
1981 meeting in Middletown, Ohio with J. F. Bowers and A. J.
Anderson, H. E. Cramer Company, Inc.).
Holzworth, G. C., 1972: Mixing heights, wind speeds and potential for
urban air pollution throughout the contiguous United States.
Publication No. AP-101, U. S. Environmental Protection Agency,
Research Triangle Park, North Carolina.
Huber, A. H. and W. H. Snyder, 1976: Building wake effects on short stack
effluents. Preprint Volume for the Third Symposium on Atmo-
spheric Turbulence, Diffusion and Air Quality, American Meteoro-
logical Society, Boston, Mass.,
Huber, A. H., 1977: Incorporating building/terrain wake effects on stack
effluents. Preprint Volume for the Joint Conference on Appli-
cations of Air Pollution Meteorology, American Meteorological
Society, Boston, Mass.
Petersen, W. B., 1980: User's guide for HIWAY-2: A highway air pollution
model. EPA Report No. EPA-600/8-80-018, U. S. Environmental
Protection Agency, Research Triangle Park, North Carolina.
Schewe, G. J., 1981: NAAQS impact assessment of the steel industry: Dis-
persion modeling. PEDCo Environmental, Inc. Report No.
PN 3480-14 (Draft), prepared for U. S. Environmental Protection
Agency, Research Triangle Park, North Carolina.
Slade, D. H., (ed.) 1968: Meteorology and Atomic Energy. Prepared by Air
Resources Laboratories, ESSA, for U. S. Atomic Energy Commission,
445.
156
-------
Turner, D. B., 1964: A diffusion model for an urban area. J. Appl.
Meteor., 3(1), 83-91.
Turner, D. B., 1970: Workbook of Atmospheric Dispersion Estimates. PHS
Publication No. 999-AP-26, U. S. Department of Health, Education
and Welfare, National Air Pollution Control Administration,
Cincinnati, Ohio.
157
-------
APPENDIX A
DESCRIPTION OF THE INDUSTRIAL SOURCE
COMPLEX (ISC) DISPERSION MODEL
The Industrial Source Complex (ISC) Dispersion Model was developed
by the H. E. Cramer Company, Inc. under Contracts 68-02-2507 and 68-02-3323
with the Source Receptor Analysis Branch of the U. S. Environmental Protec-
tion Agency. In the ISC Model, various EPA dispersion model algorithms are
combined and updated in two computer programs that can be used to assess
the air quality impact of emissions from the wide variety of sources associ-
ated with an industrial source complex. The ISC Model short-term program
ISCST, an extended version of the EPA Single Source (CRSTER) Model, is
designed to calculate concentration or dry deposition values for time peri-
ods of 1, 2, 3, 4, 6, 8, 12 and 24 hours. If used with a year of sequential
hourly meteorological data, ISCST can also calculate annual average concen-
tration or total deposition values. The ISC Model long-term program ISCLT
is a sector-averaged model that updates and combines basic features of the
Air Quality Display Model (AQDM) and the Climatological Dispersion Model
(CDM). The long-term model uses STAR summaries for either five or six
Pasquill stability categories to calculate seasonal and/or annual ground-
level concentration or deposition values.
Table A-l lists the major features of the ISC Model. The ISC
Model programs accept the following source types: stack, area and volume.
Multiple volume sources are also used to simulate line sources. The steady-
state Gaussian plume equation for a continuous source is used to calculate
ground-level concentrations for stack and volume sources. The area source
equation in the ISC Model programs is based on the equation for a continuous
and finite crosswind line source. The generalized Briggs (1971, 1975)
plume-rise equations, including the momentum terms, are used to calculate
plume rise as a function of downwind distance. The Huber and Snyder (1976)
and Huber (1977) procedures are used to evaluate the effects of the aero-
dynamic wakes and eddies formed by buildings and other structures on plume
dispersion. A wind-profile exponent law is used to adjust the observed
A-l
-------
TABLE A-l
MAJOR FEATURES OF THE ISC MODEL
Polar or Cartesian coordinate systems
Plume rise due to momentum and buoyancy as a function of downwind
distance (Briggs, 1975)
Huber and Snyder (1976) and Huber (1977) procedures for evaluating
building wake effects
Separation of multiple point sources
Consideration of the effects of gravitational settling and dry depo-
sition for particulates
Capability of simulating stack, volume, line and area sources
Capability to calculate dry deposition
Variation with height of wind speed (wind-profile exponent law)
Concentration estimtes for 1-hour to annual average
Single Source (CRSTER) Model terrain-adjustment procedures for complex
terrain
Consideration of time-dependent exponential decay of pollutants
A-2
-------
mean wind speed from the wind measurement height to the emission height for
the plume rise and concentration or deposition calculations. The Single
Source (CRSTER) Model procedures are used to account for variations in
terrain height over the receptor grid. The Pasquill-Gifford Curves (Turner,
1970) are used to calculate lateral (o ) and vertical (o ) plume spread.
The ISC Model has a rural and two urban options. In the Rural Mode, rural
mixing heights and the 0 and a values for the indicated stability category
are used in the calculations. In Urban Mode 1, the stable E and F stability
categories are redefined as neutral D stability following the Single Source
(CRSTER) Model procedures. In Urban Mode 2, the E and F stability cate-
gories are combined and the o and a values for the stability category one
step more unstable than the indicated stability category (except A) are
used in the calculations. Table A-2 shows the dispersion coefficients used
in the three modes.
As indicated in Table A-l, the ISC Model programs use either a
polar or a Cartesian coordinate system as specified by the user. Discrete
points corresponding to the locations of air quality monitors or other
points of interest can be assigned with either coordinate system. Separate
locations for each source within an industrial source complex can be speci-
fied by the user. Source locations are specified in the Cartesian coordin-
ates with respect to the origin of the Cartesian or polar coordinate system.
Table A-3 lists the source input parameters required by the ISC
Model. For a stack, these parameters include the pollutant emission rate,
stack location, stack height, stack inner diameter, stack exit velocity and
stack exit temperature. If concentration or deposition values are to be
calculated for particulates with appreciable gravitational settling veloci-
ties, source input requirements include the particle-size distribution, the
settling velocity for particulates in each settling-velocity category and
the surface reflection coefficient for particulates in each settling-
velocity category. If building wake effects are possible, the dimensions
of the tallest building adjacent to the stack are required. Source input
requirements for a volume or area source are similar to those for a stack
A-3
-------
TABLE A-2
PASQUILL-GIFFORD DISPERSION COEFFICIENTS
USED BY THE ISC MODEL IN THE RURAL
AND URBAN MODES
Actual Pasquill
Stability Category
A
B
C
D
E
F
Pasquill Stability Category for the a /a
Values Used in ISC Model Calculations2
Rural Mode
A
B
C
D
E
F
Urban Mode 1
A
B
C
D
D
D
Urban Mode 2
A
A
B
C
D
D
A-4
-------
TABLE A-3
SOURCE INPUTS REQUIRED BY THE ISC MODEL
Parameter
Definition
Stacks
X,Y
Z
s
h
V
s
d
T
sn
\
W
Pollutant emission rate for concentration calculations
(mass per unit time)
Total pollutant emissions during the time period
for which deposition is calculated (mass)
X and Y coordinates of the stack (meters)
Elevation of base of stack (meters above mean sea
level)
Stack height (meters)
Stack exit velocity (meters per second)
Stack inner diameter (meters)
Stack exit temperature (degrees Kelvin)
Mass fraction of particulates in the n settling-
velocity category
Gravitational settling velocity for particulates in
the n settling-velocity category (meters per sec-
ond)
Surface reflection coefficient for particulates in the
n settling-velocity category
Height of tallest building adjacent to the stack
(meters)
Width of tallest building adjacent to the stack
(meters)
Length of the tallest building adjacent to the stack
(meters)
Pollutant decay coefficient (seconds )
A-5
-------
TABLE A-3 (Continued)
Parameter
Definition
Volume
Source
X,Y
H
3
yo
3
zo
n
V
sn
Yn
Same definition as for stacks
Same definition as for stacks
X and Y coordinates of the center of the volume source
(meters)
Elevation of the ground surface at the point of the
center of the volume source (meters above mean sea
level)
Height of the center of the volume source above the
ground surface (meters)
Initial horizontal dimension (meters)
Initial vertical dimension (meters)
Same definition as for stacks
Same definition as for stacks
Same definition as for stacks
Same definition as for stacks
Area
Source
X,Y
Pollutant emission rate for concentration calculations
(mass per unit time per unit area)
Total pollutant emission during the time period
for which deposition is calculated (mass per unit
area)
X and Y coordinates of the southwest corner of the
area source (meters)
A-6
-------
TABLE A-3 (Continued)
Parameter
H
n
sn
n
Definition
Elevation of the area source (meters above mean sea
level)
Effective emission height of the area source (meters)
Width of area source (meters)
Same definition as for stacks
Same definition as for stacks
Same definition as for stacks
Same definition as for stacks
A-7
-------
except that the dimensions of the volume or area source are used in place
of the stack diameter, exit velocity and exit temperature.
Table A-4 lists the meteorological inputs required by the ISCST
program. In general, these inputs are developed by the preprocessor for
standardized short-term dispersion models such as the EPA Single Source
(CRSTER) Model using hourly surface weather observations made at the nearest
representative National Weather Service (NWS) surface station and concur-
rent twice-daily mixing height estimates for the nearest representative NWS
upper-air station. The preprocessor program uses objective Turner (1964)
criteria to assign the Pasquill stability category to each hour. Because
airport wind directions are reported to the nearest 10-degree sector, a
random number generator modifies each reported wind direction and assigns
the wind direction to the nearest degree within the 10-degree sector for
use in the model calculations. This procedure is followed in an attempt to
account for the actual variability of the wind direction. An interpolation
scheme is used to assign hourly rural or urban mixing heights on the basis
of the early morning and afternoon mixing heights calculated using the
Holzworth (1972) procedures. The ISCST hourly meteorological inputs may
also be developed from onsite meteorological measurements or from a combin-
ation of concurrent onsite and NWS meteorological measurements.
Table A-5 lists the meteorological inputs required by the ISCLT
program. The principal meteorological inputs are seasonal or annual STAR
summaries (statistical tabulations of the joint frequency of occurrence of
wind-speed and wind-direction categories, classified according to the Pas-
quill stability categories). The ISCLT program accepts either six stability
categories (A through F) or five stability categories (A through E with the
E and F categories combined). The ISCLT Program cannot use the STAR sum-
maries used by the CDM. The CDM STAR summaries combine the E and F stabil-
ity categories and divide the D stability category into day and night cate-
gories. The other ISCLT meteorological inputs for which no default values
are provided are climatological values of the ambient air temperature and
A-8
-------
TABLE A-4
HOURLY METEOROLOGICAL INPUTS REQUIED BY THE
ISC MODEL SHORT-TERM PROGRAM ISCST
Parameter
Definition
AFVR
m
Stability
11
9z
Mean wind speed in meters per second (m/sec) at
height z1
Wind system measurement height (default value is 10
meters)
Average random flow vector (direction toward which
the wind is blowing)
Wind-profile exponent (default values assigned on the
basis of stability)
Ambient air temperature in degrees Kelvin ( K)
Depth of surface mixing layer (meters), developed from
twice-daily mixing height estimates by the CRSTER
preprocessor program
Pasquill stability category (1=A, 2=B, etc.)
Vertical potential temperaure gradient in degrees
Kelvin per meter (default values assigned on the
basis of stability)
A-9
-------
TABLE A-5
METEOROLOGICAL INPUTS REQUIRED BY THE ISC
MODEL LONG-TERM PROGRAM ISCLT
Parameter
Definition
"
1,1
Frequency of occurence of wind-speed and wind-
direction categories by stability for the S,
season (STAR summary)
Mean wind speed in meters per second (m/sec) at
height z. for each wind-speed category (default
values based on STAR wind-speed categories)
Height at which wind-frequency distributions were
obtained (default value is 10 meters)
Wind-profile exponent for each combination of wind-
speed and stability categories (default values are
assigned on the basis of stability)
Ambient air temperature for the k stability cate-
gory and the fc season (degrees Kelvin)
Vertical potential temperature gradient (degrees
Kelvin per meter) for each combination of wind-speed
and the stability categories (default values are
assigned on the basis of stability)
Mixing height for the i .wind-speed category, k
stability category and i season (meters)
i,k
a;k,£
36
32
l.k
A-10
-------
the mixing height. The sources of the meteorological data used to develop
the 1SCLT meteorological inputs usually are the same as the sources of the
meteorological data used to develop the ISCST meteorological inputs.
A-ll
-------
APPENDIX B
OBSERVED PARTICULATE AIR QUALITY DATA
This appendix contains the 24-hour average particulate concen-
trations measured at the various air quality monitoring sites during the
entire detailed air quality monitoring program (March through October
1980). Section 1.2 in the main body of the report describes each monitor-
ing site and the particulate concentration monitoring equipment at the
site. The site identification numbers used in this appendix and in the
main body of the report are defined as follows:
• Site No. 1 - Verity School
• Site No. 2 - Hook Field
• Site No. 3 - SREPCO (SWOAPCA); colocated with Site No. 30
• Site No. 30 - SREPCO (U. S. EPA/PEDCo); colocated with Site
No. 3
• Site No. 4 - Coke Plant
• Site No. 5 - Yankee Road
• Site No. 6 - Oneida School
• Site No. 7 - Research Center
• Site No. 8 - Wilson School
• Site No. 9 - SW Ohio Steel
• Site No. 10 - Oxford Road
B-l
-------
•. Site No. 11 - Coil Paint
t Site No. 12 - Main Gate
• Site No. 13 - Reeds Yard
• Site No. 14 - Lefferson Gate
The particle-size categories for the observed 24-hour average
particulate concentrations presented in this appendix are defined as
follows:
• Category No. 1 -
particulate concentrations for particle
diameters less than 100 micrometers as
determined by standard hi-vol samplers
• Category No. 2S -
Particulate concentrations for particle
diameters less than 15 micrometers as
determined by size-selective hi-vol
samplers
• Category No. 2D -
particulate concentrations for particle
diameters less than 15 micrometers as
determined by dichotomous samplers
• Category No. 3 -
particulate concentrations for particle
diameters less than 2.5 micrometers as
determined by dichotomous samplers
• Category No. 4 -
particulate concentrations for particle
diameters between 2.5 and 15 micrometers
as determined by dichotomous samplers
B-2
-------
A particulate concentration of "999.9" tnicrograms per cubic meter indicates
a missing observation. For the 56 sample days during the period May
through October 1980 that were used in the ISC Model calculations, this
appendix also gives the observed concentrations minus the corresponding
background concentration estimates. Section 4.1 in the main body of the
report discusses the estimation of the background concentrations.
B-3
-------
ro
o
«c >•
CD DC O
O ftf
ae i-
•-• «
K O
U> Ul
X H-
O Ul
x => —
Ul O
o
x CK oe
o ui «
ac at ae
woo
to o u.
CM
O Ul
Ul ISI
>• M
Of W>
Ul
CO «•
>• o
DC O A
W Ul 01
(0 t—
Itt «
LU CO
IM
CM
CM
O
is.
OS
CM
CM
r>
ea
— CM
o o
o
10
-• CM
CM
OS
n
B-4
-------
o
ac • n
o o
*: z
u
»
tt OC O
O o
•-• «c
(0
Vt Ul
z «-
O Ul
« o o
AC M «0
H- OD OS
X => —
111 O
(J
X OC (K
O Ul «
O O. X
O (A O
Ul K ^
>• • M
oe to
z ro
O V
ui on
> o
d* o A
Ul Ul (M
CO I—
ca a
^J ^J
Ul tO
IM «M
CO
o ui
»- t-
»M l-«
Z CO
Os
Os
OS
OS
OS
OS
OS
OS
Os
OS
OS
OS OS OS
os
OS
OS
Os
Os
Os
OS
OS
OS
0s
Os
Os
OS Os
OS
OS
OS
OS
OS
OS
OS
OS
Os
OS
Os
OS OS
•
OS
OS
Os
OS
OS
OS
OS
OS
OS
OS
OS
OS
0»
os
Os
Os
OS
OS
OS
Os
OS
OS
OS
Os
Os
OS
OS OS
OS OS
OS OS
OS
OS
OS
OS
OS
OS
OS
Os
OS
OS
Os OS
Os os
OS os
O
OS
O
CM
fsl
O
ro
O> OS
OS os
OS OS
OS
-------
o
at
U
-
« CK
o
u> o
o to
to ui
X t-
O 111
UI M
> ••«
C* CO
•-> K (O —
»- CK
« O O O
OC M CO
UI 0
U
x ac ae
o ui «
o a. JE
o w> to
ot ac at
ui u o
co o u.
CD oe o
o • o
a: o
ui ui
c» »-
en a
ui co
IM t
o o o
OS
OS
OS
OS
I!
^
OS
K.
co
o
OS
IS.
*.
CO «• O
«« o •*•
ir> <•• r>
(^ ^B_
*» ^B
-• t*
CO IF)
O
m
(M
2
(SI
CO
10
10
r>
10
»s
OS
OS
Os
OS
Os
Os
OS
OS
OS
OS
ID ff*
OS Os
eg os
OS
CM OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
os
OS
OS
OS
OS
OS
OS
OS
OS OS OS OS OS
O> OS OS OS OS
r> so
o o
~« csj ro
B-6
-------
OC
to uJ
o
oc • n
U O
!* Z
O
•
a at A
O (SJ
i-i XT
K U (A
(SI
fit Ul
Ul N
u>
»- DC
« o o o
a: >-• oo
Ul O
O
z a: oe
o ui «
o a. x:
o to so
oe oe oe
ui u o
o u.
-
v^ UK
> O
OS O O
U U (SI
w »-
ou ^x
o o
CM
o ui
»- *-
M« i-i
z ce
0*
OS
OS
OS
OS
0>
OS
0s
er«
OS
V>
OS
OS as 0s
0s 0s 0s 0s «» OS
9> 0s OS «s os os
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS Os Os
•
OS Os
OS »>
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
Os
Os
OS
.
OS OS OS.
Os OS OS
OS
Os
Os
OS
OS
OS
Os
OS
OS
OS
OS
OS
Os
OS
Os
OS
.
Os Os OS
OS Os OS
in
V*
(SI
os
os
OS
OS
OS
OS
Os
ro
«s r>
Os \0
(SI
n
CSJ
rs. os
rs. os rs. os
* os * 0s
O OS
sO OS
Os
— (SJ
o o
OS
OS
OS
OS
OS
OS
OS
Os
OS
OS
OS
OS
OS
O
rs. csj
os os r^
OS OS IS.
OS OS (SJ
«-« (SJ
B-7
-------
co ui
X H"
0 Ul
H»
4» ^y ^^
•«•
0
•^
=>
0
tv • ro
0 O
*c x
o
OS ^*
CD Of 0
O (SJ
•a ui
X 0 CO
(4
Ul Nl
>• •-•
OC W>
Ul
tO W4
Fl
o
0s
0s
0S
0s
0s
»
0S
0s
0s
^*
w*
0S
0s
0s
0S
0s
0S
0s
0s
•
0S
0s
0s
Mh
w*
0S
0S
0S
0s
0s
0s
0s
0s
0s
0S
0s
•
0S
0s
0s
0S
0s
0S
0s
0s 0s
0S 0S
0S 0S
0S 0S
0s
0s
0s
0s
0s
•
0s
0s
0s
••t
^^
0S
0s
0s
Mfe
w»
9*
0s
0S 0S 0\ 0S 0S
0s 0s 0s 0s 0s
0S 0S 0S 0S 0S
oi. •-> to
>- flo 0s
x r» «-«
Ul O
CO
0 0. C
O CO 0S
Ul JC «••
> • 0
oe o o
Ul Ul (SJ
CO t-
ea
* 0s in 0s s» co
WO 0s x Ofs O —
«" " «^ - -
— csj ro o » «
«> ^^ ••> »« <» e>
0s
•
0s
0S
0S
0s
0S
0s
0S
0s
•
0s
0s
0S
CO 0s
.
« 0s
Ift ^^
0s
0S
0s
0s
0S
v* N.
«k •»
0s
•
0s
0S
0s
0s
•
0s
0s
0s
0s
0s
0s
0S
¥> 0s
.
\O 0s
fO 0s
0S
0S 0S 0S
0S 0S ffS
0S 0S 0S
0s 0s «s
CD 0s O
0s
•
0S
0S
0s
0S
•
0S
0S
0s
0s
0s
0S
0s
0S
0s
0s
0S
0S 0S 0S 0S
0s 0s 0s 0s
0S 0S 0S 0S
0s 0s 0s 0s
— c* i»» »
B-8
-------
o
ce.
w
•
o
a ui
to
CO Ul
X »-
O UJ
—
Ul O
o
•X. Of. OC.
o ui a
o a. c
o to cst
Ul JC (SI
>• «
ac of oe
ui o o
cx> o u.
CD Of
o o
Ul IM
Ul
(O —
01
o >•
Ul 0£
>• O
oe ts o
ui ui CM
V) t-
00
OS
OS
OS
•
•
Os
OS
OS
Os
OS
Os
Os
Os
OS
fs
OS
•
•
Os
OS
OS
OS
OS
Os
OS
OS
OS
OS
OS
OS
OS
ON
OS
CO
o
OS
rs.
CM
OS
Os
(Si CX
OS ID
(SI s»
(sj n
o o
n
(SI
*» * n* OS
O> (SJ Is. OS
OS W (NJ OS
OS OS
(M ir> * *• C*l fO fO Os OS
OS os
tf>
CSI
OS
OS
OS
o «•
n e-
10
is. CO Os
— (NJ
B-9
-------
o
z
CD Of &
o rg
Ul
(SJ
ot
V» Ul
Ul
«
ac e* ae
woo
crt O U.
CO 0*
O Ul
Ul Isj
> «-l
a: v>
ui
OT
at
e
o
z
O V
Ul Of
»• o
Of 19
Ul Ul
00 <(
O O
Ul «rt
IM (SJ
(A
ae
o
Z (0
o
0s
0S
0s
0s
0s
0s
0s
0S
0s
0s
0S
0S
0S 0S 0S
0S 0S 0S 0S 0S 0S
0s 0s «s 0s 0s 0s
0s
0s
0S
0s
0S
0s
0s
0s
0s
0S
0s
0s
0S 0S 0S
4Vt MS
V* V
0s
0s
0s
0s
0s
0s
0S
0s
0s
0s
0S
0s
0S
0S
•
0s
0S
0s
0S
0s
0S
0s
0S
0S
0S
0S
0S
0S
0S
0S
0S
o
SO
0s m
0s
0s
CSJ
CSJ
csi
CSJ
0S
^
CSJ
-
ro
*
•«•
0s
CO
CO
**
0S
0S
0S
0S
0s
0S
0s
0s
0s
0s
0s
0S
rs. 0s
vO 0S
CSJ 0S
0s
fO 0s
fj 0s
0S
0S
0S
0S
0s
0S
0s
0s
0s
0S
0S
0s
0S
0S
0s
0S
0S
0S 0S 0S 0S 0S
0S 0S 0S gS 0S
0S 0S 0S 0S «S
« en
o •»
o *
r>
o
»•« Ol
B-10
-------
ae
CO LJ
UJ
« o o
OC « 00
UI O
o
z ac of
o u a
ac at ot
ui o o
co o u,
Ol CK
o o
o
Of •
C9 O
00 OC
O
in o
x o co
(SI
A UI
UI IM
ui
v>
ec
co
Ul O£
>• o
(K O A
UI UI (M
- »- o
t-4 ** •-•
X CO
OS
9*
os
««
0s
as
«•>
as
as
as
as
as
•
•
as
as
as
as
as
as
as
as
as
*
as
as
as
* » »
as as
as
as
as
as
as
as
as
as
as
as
as
as
as
•
o> as as as
as
a*
as
as
as
as
as
as
as
as
as
as
OS
as
a» as
in
o
oo
(SJ
as
V>
(M
CO
csj
rO 4^
•>• f)
^r m
OS x 9.
n. p> as
OB
csj
-* CM
o o
OS OS OS 00
OS OS 00 CO
OS Os M
in
o
in
m
SP N.
O O
o
OS
00
o
ro
O O O Os
is. is. CO as
vo vO 0s
Os
*"* CSJ fj
O
CM
CM
O
SO
O>
Os
n
B-ll
-------
oe •
C9 O
o
ec.
(A
o 01
to ui
X t-
« o o
oe M o>
t- OB 0S
X 3 -«
UJ O
U
Z QC OC
O U O U.
00 Of
ui
UI isl
at tn
x ro
UI Of
>• o
ae o o
UI UI (S|
to >-
oa
OS
OS
«s os
OS OS
OS 0s
OS
OS
OS
OS
OS
OS
OS
OS
Os
OS
OS
OS
OS
OS
0\ os Os Os Os
OS OS 0s os OS
OS OS OS Os OS
CM
O
o
csi
oo
csi
O
ro
N.
csj
csj
OS 0S
o
CSJ
— CSJ
« 0s
r> os
OS
to
-• «*»
to csj
w o
ID V)
r>
r>
CSJ
0s
CSJ
OS
CSJ
OS 0s OS
os 0s os
OS 0s 0s
OS
OS
0s
0s
0s
0s 0s 0s
0s OS as
OS OS 9>
OS OS os
rw oo 0s o
o o o «
— csj n *•
B-12
-------
at
to ui
O
at •
13 O
ea oe o
O tM
«o o
9 UI
X O «O
(4
O Ul
UI N
O UI UI
M IB to -4
t- cr.
« u o o
oe ix oo
«- co 9>
x a <-«
ui o
u
x at at
O UI 0.
u a. «
O (ft to
ui x o *
oe oe oe
woo
to o u.
co oe o
oo x to
fcx
K O V
w ui at
> o
oe o o
UI Ui CM
CO »-
CO
CO
ON
ON
•S (A ON JK ON
N.
01
00
OB
O
O
cv
(M
o
*•
n
in
ON
«« o is. r- •«•
ON •>« 00 00 N.
m
r>
o
00
10
n r>
ON IS.
ct «
is. rs.
to
ON
o —
cvr>>o<«>ir><*rs.c00NO»«
o o w oooooo^**^
B-13
-------
at
co ui
X *-
O Ul
-i CO
*- CO OS
3C =» —
Ul W
o
ac oe «*
o ui a.
o a. a:
a to v0
oe oe oc
u» o o
CO O U.
OB oe.
»
c>
a
C3
D£ • W
C9 0
-
00 & O
O ftf
l» 19
9 Ul
X >-
M< —
or.
O
OS
OS
OS
OS
OS
OS
OS
OS
OS
•
os
OS
OS
OS os OS
OS OS OS OS OS OS
OS «s os OS OS OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
*
Os
Os
OS
OS os
.
OS
«N
«s
OS
OS
OS
OS
OS
Os
OS
•
Os
OS
OS
OS OS
.
OS OS
es
OS
es
OS
OS
OS
OS
Os
OS
•
OS
OS
OS
OS
OS
OS
OS
.
Os OS OS OS OS
a 5-
Ul Ul «SJ
CO I-
O>
rsi (si
(A
O Ul
>- >-
•^ KM
X (O
(M
CSl
os
OS
o
o
o
r>
r>
- n os
eo *- os
(SJ
o
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
CsJ
N-
VD
O
r>
is. O>
O O
«• OS OS
•» OS »•
OS OS
OS
Os
OS
•** (SJ
B-14
-------
O
ae •
o o
o csj
V5
M IU
o u
ui rsi
ae to
ui
i— 01
0s
Z 9 M
UI O
X CK Of
O UI flL
U O. • o
Of O O
ui ui (si
ui o>
ISI W -•
csi n esi n •«•
o
in
ro
in
(SJ
o
o
0s
0s
(M
O
IT) <•
IS. CD 0S
O O O
•o — csi r»>
B-15
-------
CO Ul
O
oc • ro
C9 O
>£ X
O
o
o
1C O «
«SJ
O UI
UI IM
•^ _-
oe (si
ui x — *•
at ot oc
UI O O
co o u.
OC QC O
o u x n
»M
»: o >-
^ UI OC
> o
O» O A
UI UI (SJ
CO f-
oo
0s
0s
0s
0s
0s
a.
0s
OS
OS
0s
ON
0s
0s
0s
0s
0«
OS
0s
OS
OS
0s
OS
OS
0s
OH
0s
o
0s
oo
(SI
OP. 0s *•
0s 0s OO
0s O 0s
0s
0s
0*
ro o •* r> v* N.
0S
0s
0S
0s
(SJ
0S
O
0S 0S
0s
0s
0s
0S 0S
0s 0s
0s 0s
0s
0S
0s
0s
0s
0s
ON
0S
10
r>
B-16
-------
ro
a >>
a ot
O
U> C9
=» Ul
ot
(A Ul
« u o
C* *H 00
»- OB OS
z a —
Ul O
o
x oc of
o ui a.
o a. M
ae a>
UJ
«o •-
ui ac
>• o
ae o o
ui ui 01
at t-
oa
ro
cs)
o
«•
ro
CM o ro
ro «« «s
«M
o
o — «sj ro
B-17
-------
OB ae o
o «sj
u> to
V- 01 0s
z s> •*•
UJ U
u
z on at
o uj a.
o a. «
o «/> eo
oc.ac.tc
ui o o
co o u.
OD ae
o w
V)
(SI
ae co
ui
to -
Ul Of
>• o
oe i* o
UJ UJ (SJ
to >-
OB CC
ui co
ISl (SJ
CO
O UI
*- I-
»« l-<
Z V)
OS
0s
0s
O>
0s
0s
OS
OS
OS
OS
0s
OS
0S
OS
OS
0s Os! OS*
0S OS OS
OS OS 0s OS OS
OS OS 0S OS 0S
0S OS 0S 0S OS
os
*s
0s
0s
0s
0s
0s
0s
0s
0s
0S
0s
•
OS 0s
OS
0s
0v
0»
0s
0V
0s
0V
0v
0s
0*
0v
0s
0V
0s
•
OS 0s
0s 0v
OS OS
OS 0s
0s
0s
0s
0V
0S
OS
0s
0s
OS
0S
OS
OS
0s
0s
OS
OS
OS OS OS OS OS
OS OS OS OS 0s
OS OS 0s 0s 0s
Csl
rsj
fs-
(SJ
(SJ
(SJ
(SJ
as r^ ct> x
os — co oo
os •-
00
OB
0V
0s
0s
0V
(SJ 0S
(SJ
0s
OS
0s
(M
O
0s
0s t» so
OS ts. N
OS os 0s 0s •*
ve os os os -•
OS 0s 0s M
OS
0s
OS
0s
0V
0s
OS
OS
0s
OS
0S
OV
OS
tr>
oo
o
Ov
O
-« (SJ
B-18
-------
CO UI
— •
Ul X 01
> «
at ac ac
Ul C9 O
M O U.
aa at
o w
C9 O
U
•
CD QC Ct
O CM
3 Ul
Z »—
»-• l-l
co *«
er.
x ro
ui at
>• o
ae o o
u ui en
CO *-
da «
ui
O Ul
fc- *-
fcM •*•
3E CO
cr.
0V
ON
OS
Os
OS
OS
«s
OS
CM
r>
O>
OS
OS
•s
OS
O>
O
o
r>
to 10
~ in
vO is.
o> ro
-« CM n o *
o o o w o
CM
OS
O -<
B-19
-------
o
oc • n
v> o
i,C Z
O
«c >-
OB DC o
O • •-
z »- oc co
O Ul Ul
•" K «A
»- on
-« CO
»- co 0s
x => —
ui o
o
Z OC Of
O Ul ft.
o a, a:
d to *•
ui K 01
> a
at at us
ui o o
co o u.
at at
o
z ro
at o A
UJ Ul CM
co t-
Ul CO
fM M
co
CA
OS
OS
ON
0s
OS
ON
0s
0s
0s
«N
* 9. 0s" ON* os* OS*
0S 0S OS ON 0S OS
Cf>
0s
0s
0s
0s
0S
0s
0s
0s
•
•
0s
OS
0s
OS
OS
ON
ON
0s
OS
0S
0s
0s
•
0s 0s 0s os
0s
OS
0s
OS
0s
0s
0«
0S
0s
0s
0*
9*
0s 0s 0s
in
vO
10
o
co
rsi
ON
OS
ON
ON
OS
0s
00
— CM
o o
OS O
OS O
ON •-•
o
10
CNJ
ro
OS
»* »•• OS
r> * 0s
O
O
ON
O
0%
0s
OS
0S
OS
OS
ON
OS
CO
O
01
0S
0S
0N
0S
0S
0S
-« CNJ
B-20
-------
«0
X
o
« o o
ae « co
f- CO OS
x => ~
ui o
o
x of oe
o ui o.
oo.cc
o to r»
ui x csi
>
o
ae • ro
UP o
o> u a
o M
•-•
oe «o
ui
co <-
en
o
ro
•>• o
t* 13 O
Ul Ul «SJ
CO >-
oo OS
* «S1
OS O
ro os
•« ro
* OS OS
OS
OS
OS CO CO OS OS W OS
os is. m os is.
o o o » o o
os o o os ON os o
CNJ ^ m os os 0s m
os tr> «« os os os rs.
os os os
—i CSJ
csj
B-21
-------
n
o
•
CO 0£ O
e fsj
(O O
9 UI
z v-
K« —
ui o
o
a: ot a:
o ui a.
o a.
«
UI
v>
«K
O
W
o
z to
o >•
UI Of
•> o
UI UI
isi ft
••^
(A
e ui •
t- t- G>
******
z to
0s
0s
OS
0s
0s
0s
0s
0S
0s
0S
0s
0S 0S 0» 0S
0s 0» 0s 0s
0s 0s 0s 0s 0* 0s
0s
0s
0S
0s
0s
0s
0s
0s
0s
0S
0s
0s
0s
0S
•
0s
0s
OS
0s
9>
0s
0s
0s
0s
0s
0S
0S
0S
0s
0s 0s
0s 0s
•
0s 0s
0s 0s
0s
0S
0S
0S
0S
0s
0s
0S
0S
0S
0S
0s
0s
0S
0s
0* 0S 0S 0S 0S
0s 0s 0s 0> 0s
O
(SI
is.
«*
n
CO
o
0s
O
to
o
O
IV.
r>
0S 0S
CO 0S
ro
o
« 0s
ftft 9*
9*
V> O
V) If)
0s
0s
0s
0s
0s
0s
0s
•* esi
o o
n
o
0S
0S
0S
0s <• (M
0s f<> CO
0s •«
n
to
B-22
-------
Of
«/> UJ
O UJ
« o o
ee •* to
H- co at
z => -•
Ul O
o
X C* >•
O Ul «
cj a. x
Ul K O
> «
« dc at
Ul C9 O
(O O U.
00 Of
o o
ti) O
kC X
U
«c >•
at oe
o
«A 13
=> Ul
A Ul
Ul N
ro
> o
Of 19
ui co
N CM
en
CO
at
at
at
at
at
at
.
•
at
as
at
at
at
at
at
at
at
at
•
•
as
as
as
as
at
at
at
at
at
at
at
at
at
at
at
at
at
at
at
at
at
at
at
at
at
at
.
(V
CO
10
W» 00
FS. r>
CK
o
o
IS.
n
CO
o
<*
CM
00
0*
at
at
f«- r> as as as ro
ft. o en a» as en
ft M 0S «S «
n
o
t^ 9*~
co at
CO
o
eo
co
o o o
i*> «• a>
at •• ot
— cti
B-23
-------
o
z
9
O
o
z
CD C*
O
Ul
u
V> Ul
z »-
O Ul
UJ
Ul
CM
co
H» at
•
o in ae
o u
O 9-
Ul QC
>• O
dt A
Ul Ul O*
«0 H-
dD 00
0s CM
CM CM CM en en en
tn «• rv en en vo
CM o o o en
\e <* en
en
-------
CA Ul
X t-
O Ul
•
a> ac o
e a*
w» tf
=3 Ul
Ul IM
>• •-
CK «0
Ul
CA •«
« o o o
I— CD OS
Ul U
u
X (K >•
O Ul «
O &• «C
£^ O) 0^
Ul X O
> o
at u A
Ul Ul (SJ
(A t-
a> «
o e->
ui e»
IM (SJ
(A
O Ul
»- I-
»4 »4
X (O
OS
OS
OS
OS
OS
OS
OS OS OS OS OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS OS
* OS
O OS
Ift 0^
OS
SB OS
os
«« OS
OS
o^ ro ^i m os
os n os (sj os
Os 00 00 00 OS
OS OS
O O OS O is. OS
ro SB os (sj m o«
(sj n o * tr> s* is.
^^ 4^ V) ^^ ^^ ^^ O
OS
OS
OS
OS
OS
OS
OS
00
o
Os
Os
OS
OS
o
OS
OS
OS
OS
os
«»
OS
OS
OS
O
OS
OS
O
OS
OS
OS
os
OS
OS
os
os
os
OS
OS
OS
os
OS
OS
OS
o o o o o
^^ 00 SB SB OS
fo (si n os tf)
o « N n »
B-25
-------
n
19 O
•X Tf.
O
e
(O O
CO
(M
V> Ul
x t-
o u
•
O HI «
O CL K
A «rt CM
Ul JC ~«
> • •-•
oe co
cr.
o
o >•
Ul Of
>• o
DC O O
Ul Ul CM
CO t-
oa «
o o
ui to
Nl CM
OS
ON
OS
(ft
OS
OS
os
OS
OS
OS
•s
OS
OS
OS
OS
•s
•s
OS os
ro
rv.
CM
CM
O
O
f»
OS M
o
0»
OS
os
« -. OS
•* CSI
O O
n
«SJ •«•
rs. K.
in
o
o
N.
r>
to
OS OS OS
«s OS OS
OS
tfi
CM
s*
O
B-26
-------
O (M
*0 (9
a w
«0
to ui
X l~
O UI
« w o
oc M o»
t- OB 0->
x s» ••
tu o
o
X tt >-
o ui a
o a. K
oc at at
UI 19 O
«A o u.
CD Of
O UI
UI ISI
> i-i
to *•
eo
o
UI «
> o
ae \s ct
ui ui
o
•M
(M
O
m
o
ro
(M M 0S
•s CM »N
w * ID o
e» 0s
o o
rs. oo
B-27
-------
CK
CO Ul
O Ul
OL • ro
o o
U
« >•
CD CK O
O CM
*"* ^E
i
O Ul
<^J to
(K W
Ul
-
O Ul «
u a. c
O 09
oe oe dt
Ul (9 O
CO O U.
o o
to
o >»
Ul tt
>• o
oe o
O Ul
o»
ro
ro ro o
ON
ON
09 ON
M ON
O (M
Ch ON ON ON
ON ON ON ON
ON
ON
ON
O O O
r> v> CD
*•* CNI W
ON ON
ro
— 0>
* m
ro
m
o
CO
n
(M
n
ON
ON
OS O
ON CD
ON «O
ON
O
10
09 ON O -<
-------
at • n
19 O
CD BC
O
W> O
9 Ul
0>
Ul
at
v> itt
ac »-
o ut
X DC >•
O III «
O O. K
Cl (ft •*
Ul K fNJ
> •
Ul CK
> O
oe o o
Ul Ul (M
CO »-
CO ON CSJ
OS ON
o» «r os
OS «- OS
OS CO OS
OS
as
OS
OS
os
o o
os o
fs. O
CSJ CSJ
t\-
10
o
CO
o
ro
ON
ON
O O O OS «• ON
« r»- VB os co ON
r. CD o os n o»
O OS
oo
ON —
ON O •« tM
roooooo
B-29
-------
=>
o
of • ro
CJ O
CJ
-> CO
I- CO OS
Z S> ~«
Ul O
o
z oe >-
O Ul •
v ui a.
»• o
a« o A
Ul Ul CM
ui co
IM CM
CO
Z CO
o
CM
in
%0
VP
01
OS OS CO
os rs. co
IS.
n
N. IS. OS
*> O^ OS
•* OS
OS
OS
OS
OS
rs. o
ro
OS
OS
N.
in
«s OS Os
n>
o
ro
os
OS
OS
en
CM OS
ro os
so OS
OS
CD
CD
CM
CM
o
OS
OS
OS
ro «
N. *0
o> o «-i CM ro *
O ,4 M ^ ^1 ^
B-30
-------
=>
o
ta o
u
•-«
x »- at to
o ui ui
MX CO *•
t- CO
« o o o
at »-> co
•- CD o
at «J
ui ui
co t-
a> «
e u
ui
CO
X CO
«s «s cr> CN
x CO
^o r*>
•*• CD
m ««
O (M
(ft O CM ft- CM «N
•S — «N
•> tr» N. rw os
«•> »-• « *° 5s
0
CM
0
CM
O
ro
n
(0
(ft
«s
•>
«s
(ft
•s
*»
(ft
O (ft » IO ft-
(ft CM
«*CMf<>o*-ir>i**-a»(ftO<~
-------
o
«
C9
a«
o
a ui
ro
CM
to
CM
C« UI tsl
(ft UI > 1-1
XI- 0( CO
10 »«
«c o o
K M CO
»- to os
K 3 ->
u o
o
X QC >-
O UJ a
ac ac at
u u o
«o o u.
ao ae
o o
o >•
ui oe
> o
« o o
ui ui m
w i—
UI (A
IM CM
(O
CM
ID
o
00
rs.
CM
o o
rw r>
ro N.
CM «*•
m CM
10 •»
OS •«
o
in
CM
CM
CM
ro
CM
Os CM
os is.
CO
00
0
fO
ts.
OS
00 —
CM *•
o
ro
OS
CM
CM
ro o»
os os
in os
o>
o o
Os O
•- CM
O O
CM in
09
IS.
ro
o
o
to
in
fsl
ro
IS.
ro
o
OS
OS
OS
ro o
CM ~>
•* CM ro •«•
B-32
-------
9
O
VI
to
o CM
O CO
CM
co ui
x t-
O UI
UI
CO
CM
UI
at oc oc
UI O O
CO O U.
to at o
oo x ro
M
K O >-
v UI Cfc
•>• o
Of O O
UI UI CM
co i~
CO r>
o o o
.* n *•
OS OS
CSJ
OS
o
OS
OS
OS
os
OS
OS
os
OS
OS
OS
OS
OS
O
*
CM
o
o
«M
o o
* <0
« K.
«s
OS
OS
OS
f^
OS
•s
•s
vM
«#•
CM
v4
*•
*
csi •*•
O CO
«o r>
OS Mf
OS «•«
OS CO
OS
*•
CM
CM
ro
OS
ro
* CM
CM *•
r> m
CO
o
0
CM
O
*
ro
ro os
CM OS
H> OS
OS
01 A
W ^^
x O
o
vo
CM
O
ro
n
o
CO
10
•t fl< Mk Mk 4^
W> W> V* wS ^>
OS OS OS os Is.
OS os OS as
<«• tO
O O
o •*
csj
ro
B-33
-------
~
fi(
(0 m
•^ •»
0 UI
JU *
A u> n
U) K O
> a
01 Ot Of
UI U O
CO 0 U.
co te.
C5 ^J
M
p»
w
A
Z
9
0
or • ro
C9 0
X X
-
m Of o
O (SJ
3 UI
Z t—
•-• •
UI Of
>• o
oe v» o
UI UI (SI
*o t~
O LJ
UI (A
Nl (SJ
1-4
o>
_
CK
0 UI •
»- »- o
fc* w *»
Z CO
0
E
0s
(SI
CO
tr
K.
•
(SI
T
0s (SI CO
. • .
0s so n
0s ro «•
flft
dMk , <^ JK JHk fkl Uh
V^ ^V ^7 O^ Ol V7
•* (SJ 0S V> <0
0S
0s
•
0s
10
CO
o
ro
N.
•
o
rs.
0s (SI CO
.
0s * ro
0s r- co
**
so ro r> 0s 0s (si
* - -
« o* ro o * «
o o o ro o o
n ** /
ca o o
ro •*•
no o
.
(SI •*•
*••
is. 0 0
.
n ce
(SJ
0> (SI CO O O
• * . * •
o ro <« co
* (SJ
N 0s 0s 0s 0s
«S 0S 0S 0S
01 0 O
.
n oj
-------
o
z
ro
co ui
se »-
«
oc oc a:
ui o o
«o o u.
o> CK
o u
o
Z
CD oc o
o en
o» u
3 UI
Z I-
•-1 «
x u o>
CM
O UI
UI rM
9- «-«
ac co
ui
co «<
CD
A >-
UI Ot
»• o
oe c9 o
UI UI CM
UI CO
IM Ol
Z CO
CO
CO
OS
OS
OS
OS
OS
OS
OS
OS — O
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
O
OS
OS
OS
OS
OS
OS
OS
OS
es
o
to
OS
OS
CM
— CM
CO
IV-
SO
tw
*•
to
CM
os r- os
OS CM OS
OS OS
0 OS x CM
(M OS (M OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS ID
OS n
OS CM
OS
OS
OS
OS
OS
0
OS
O
NO
o
r>
*• o
«« OS
CM «*
OS O OS OS O
«o r» os os K
t*t CM OS Os if)
OS OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
OS
•s
OS
O
1*)
CsJ
O 0
Ift CO
tr> «•
ro
o
r>
o
OB OS
o —
(M
B-35
-------
ae
v» u
z •-
o w
> cc
BC CK oc
u o o
10 O U.
GO Ct
o o
o
cc • ro
u o
K Z
O CSJ
K O CO
CSJ
O UI
UI M
V> •«
OB
O
Ck 9-
Ul 0£
>• o
CK •_** o
111 UJ CM
4rt J™*
CO
m
CM
m
IS.
CM
» •« (SI O OS O
SO N. Is. ON
OS
ro co ro *» o os
OS O
os m os os os is.
os os os os
os os os os
o
n
o»
«o
00
CSJ
o
SO
O
is.
O
O
tr>
O
CM
OJ
m o>
00 OS
ID
0^
OS (M OS OS os
OS T O> OS OS
OS OS OS ON
o
<#•
CM
O
in
ro
to *» is. o> os o •••
o o o o o — «••
ro
B-36
-------
o
at • ro
19 O
X Z
CJ
•
no oe a
O 01
O) O
=> UI
01
v> ui
x »-
O UJ
UI IM
> >-
ce CA
UI
»- 00
o u.
tt Of
o o
o >•
Ul IX
a- o
Of O O
Ul Ul (M
w »-
00
10
V)
01 «-
* o
NO
» 09
01 —
ON
•S
ON
n
01
01
r>
O
OS
99
Ol
OS
ON
•^ Ol
O Ol
»* Ol
o
w
to
ON 00 PO
ON ON O
ON -«
ON \o n
ON — Ol
o
oo
01
V0 O
ON
01
r>
09
fN,
ON
OS
SO
ON
OJ M
o
N.
ON M M M
ro
ON o -« 01 n *
O «4 V «« M ««
B-37
-------
o
oe • to
u o
tt Z
o
« >-
to oc a
O CM
to o
U CO
CM
UJ
Ctf
co ui
se h-
o w
ui
oe co
ui
M C «0 •*
t- 00
« o o o
ae >-> co
i- to as
Z 3 ««
UI U
O UI
z oe z
O UI 3
U OU ^
UI C M
> a
oc oc oe
til C9 O
co o u.
co oe o
o o z
M
C Ct >•
UI UI CM
00
os * w ro »<
o o
o o
OS
OS
eo
CM
o
(M
r>
tv.
ao
r>
CM
OS 00
00
n
o
OB
O
a>
ID
CM ••
M CM
O O
%0 fs.
o o
OS
OS
OS
OS
ON
OS
OS
OS
OS
OS OS OS
OS OS OS
OS os «s
CM
B-38
-------
e>
3
o
at • ro
u
• M
« u o
CK M O)
I- 00 OS
X 3 *4
UI O
O UI
Z Of X
O UI 3
O O. -»
o to o
UI X CM
> «
at at at
ui u o
co o u.
o o
o
z ro
A >-
UI CK
>• o
(K O O
UI UI CM
o u
UI CO
ISI CM
(A
CO
CM
CM
ro r*.
ro
O «•
CM
CM
O O O
CD CD
ro
CO
tr>
Cft
n
CM
CM
CD
ro
ro
o
r.
o
0
o
f-
ro
CM
•« •> Cf« <•• cr>
ro «r» CM CM —t
O O •» <• CD
<* r>- o o «s
m oo
** CM ro o
o o o ro
ro ce
f> IO
CM
CM
ro w»
o «
0
CM
O O
m ro
ro *
B-39
-------
ro
C9
a*
o
0 CM
O CA
(A IU
Z I-
O IU
•
ui ae
> o
Ul Ul (SI
to »-
as a
Ul (A
M (SI
(A
CA
M
ro
n
00 »>
ro (si
N. ON
V (SI
(SI
(SI
ro
o
ID
r>
r>
in
ro
ro
o
(SJ
(si
O O O ON
V) Oft 0& ^
OB
(SI
ro
IS.
O
09
m
(SI
o
ro
— o
in o»
r> 00
O O O
ON o -< (si ro "»•
O —• ••* «* <•< «*
B-40
-------
*^
oe
CO UI
3C I11*
O UI
M X
»-
«c u o
oe •-• «
ui u
u ui
ac oe x
e ui 9
o o. -»
o to «o
UI K CM
> «t
ac oe oe
UI C9 O
CO 0 U,
at oe.
o u
!-•
K
^^
«f
td
9
O
^£ • fA
&•» • i^*
C9 O
*: z
o
« >»
oa o* A
0 CM
to o
3Ltl
UJ
X t-
^^ 4C
K W CO
CM
a ui
UI IM
> M
DC CO
UI
to -«
CD
O
»
.
O
z ro
4% 9^*
UI K
>• o
ac o o
UI UI (M
to t—
CD r> *
•N * ro
ON
o o o ON ro <**
O O •» «• OB
— 10 ON ID »
ON
U)
•
«•
ro
<«
•
ON
CM
CM
•*
UB
». — ro
jHk. ^\. MK
V* ^ V^
^ O CO
9k »4
O^% rf* lV^ >A M«|
0^ *7 W» f*J «^
^r ^* «• 0s o c^
Wl^_ ^% aV^ M% Ai
^^ %^ w^ ^f W^f
•^ ON •- —
•« CM to o
o o o ro o ^x
ro o o
OB ro
ON o o
V* ^
CM 0 0
. .
0 N.
•* V*
9* 9> CM O O
. • . • •
ON ON ro **
ON ON »* CM
ON ON
•T* f> O 4^ flN 4^ O
o <» r> ON ro o OB
•« ON CM OB ro
ON
ro o o
.
ro ir> OB
«M — CM
ON O O
S0 V) ON
CM CM (M
CM O O
• .
o o rw
10 •* »
ON ON (M O O
.
ON ON OB V) ^*
ON ON v» let OB
ON ON
•Vk Mt rf% 4% «Vh 4"± ^^ ^%
W* I*J ^^ *W V* ^? ^^ *^
^roBioesOKrw^cM
»•> cA ^ tfBk MrV M% ^^
-T** *i» F^ W* O' Wf *^
ON -. —
^r*.oBONo»4CMro
-------
ro
C9 O
at oc
o
CO C9
=3 UI
X t—
co
CM
CO UI
z i-
o ui
~ co
«- CD os
z => —
ui o
o ui
z c* z
O UI 3
O O. -a
UI C CM
> a
« ac ae
UI t3 O
co o u.
n «
o u
UI Nl
CO —
CD
O
O 9-
ui ae
>• o
oc o o
UI UI CM
co >—
ui co
rvi CM
CO
Of
o
CO
o
CO
CM
ro
n
N. O
M CM
OS *
M CM
o rs.
CM CM
O ON
w
ID
CM
«s C>- CD
CM
O
(M
O W ^
* CO W
CM » «»
«SI «•«
r>- CM
m iw
«• o r-
*» w> w
-• CM
B-42
-------
sc i-
O UI
«C 0 O
UI O
o >•
O UI 9
o a. -a
(A (M
ac ac ac
ui ta o
co o u.
OB oe
o u
at • ro
o o
* z
o
a >•
m ac A
o w
u» o
JC O CA
M
ac «o
UI
ac o
ui ui
«> fc-
ca «
e u
ui o>
IM N
o ui
**
o
ro ON ^
•> ON
•» «»
ON o ON
es ON
ON ON
ON ON
ON in
r»
to
o
o
0»
ON lf» ON O>
ON fO ON O
IO
O ON ON *•
o ON ON is.
«• ON r*. o
** ON v^ «••
ON ON »o ON
n
ro
ro o o o o o
B-43
-------
(O U
x i-
•
X Ot -I
O Ul 9
O 0. -»
O£ • P)
C9 O
M Z
o
n
Ul C O *•
> a
at K K
ui o e
(0 O U.
CD DC 0
o u x ro
^^
C 0 >•
v ui e*
>• o
oc o tf&
Ul Ul CM
OB «
o cj
Ul (O
NfU
ra
CO
OK
O Ul •
»- 1— O
Z CO
OS
•
CO
N
*
•
o
fit
n
.
OS
OS OS O
•
OS OS f
OS OS
o os o os os n
W OS M os OS «
Ub ^Bk gMk -M^ ^L .A
W W W ^W V *»
OS OS OS
•- (sj M o <«• r>
M 0
OS
•
o
«4
(SJ
•
OS
^«
^4
.
•«>
IO CO
• •
lo *^
OS
OS
OS
«* N.
O^BL
"9
OS
•
OS
OS
OS
OS
.
OS
OS
OS
OS
•
OS
^^
OS
0 OS
•
m os
OS
O OS OS OS
(SJ OS OS OS
CO OS O «•«
OS
.
OS
OS
OS
OS
.
OS
os
OS
OS
.
OS
OS
OS
o
•
^
OS 0 0
ffi CO »
CM ro ^h
B-44
-------
«*
OC
to ui
Z »-
O Ul
>- 1C
t»
X ^^
Ul O
O 9-
z at «J
e ui s
O O. ->
O CO CD
Ul X 0
> -
tt> ^^ ^^
m ^t ca
O CM
(O til
9 Ul
* «
X 0 CO
CM
O Ul
Ul ISI
t> ••*
at co
Ul
CO ««
CD
O
o
at o o
Ul Ul CM
CO t-
0 0
Ul CO
rsi CM
CO
—
at
O Ul •
1- *- 0
Z CO
0S
.
0s
en
0t
CD CD CD 0s
....
CM N. o rs
-• «• 0s
0s
0S
.
^^
w*
O 0 O 0s
CM rs. O 0s
0s CD CM
O 0s CM
M 0S ON
O 0s is.
O O O O
0s 0S
0s 0s
tfr\ AS
V W
0s 0s
0s 0s
0S 0S
0S 0S
0s 0s
0s 0s
0S 0s
0S 0S
0s 0s
•«• *• CM
. .
M) 0S M9
CM CIS «•
0S
CM 0s CD 0s CD 0s CD
.......
0S 0S
0S 0S
0S 0S
.
0S 0S
0S 01
0s 0v
0S 0S
.
0s as
0S 0S
flpk glk
V ^^
CM 0s o
. .
0S 0S *.
0S
<•• 0S O 0s O 0s 0
ro 0s CM cr> n 0s M
CD 0s 0> 0s CD 0s CM
0s 0s 9k M
CD 0s O -« CM ro <*
B-45
-------
">
QC
O> UJ
9C fc*
O UJ
^** |C
h-
-
CO Ot O
O CM
(0 Ol
3 Ul
Z *—
E U CO
CM
O Ul
OC CO
Ul
co ««
CD
O
*.
•
o
Z W
0 >•
i, I ^^
Ui CK
> o
DC U O
Ul Ul CM
CO t-
CO. 4E
o o
Ul CO
rsi CM
£
O Ul •
H- »-> O
M f-l M
X CO
o
*
o
01
«4
.
«o
• • • «
0S «4 CO IS.
0S «•»
0S
• • . • .
CM
0s
.
r>
v>
—
0S V> V 0S
• ...
0> to n -<
1^ IO CO ^» 00 0H
^ CM n o «• n s»
o o o w o o o
u /. /:
O 0 0
CM
O O O
00 CM
M
o o o •*
... .
O O 0s OS 0s OS as
r> 0s 0s os CM
•4 0S 0S 0S SO
0s 0s «s
O O O
«• — CO
*4 ^ 94
00 0 0
• .
f> O fO
CM CM CM
CO O O
• • •
0s «• ~«
n f> *
V* » O 0
... .
f co ro
irt ^ ^^ m
-------
Ct
z
C9 O
f»
«J
o
x a *«
Ul fc>
o >•
z oe -i
e ui a
o o. -»
ae oe oc
ui o o
(A O U.
•« «e
f. 0 (O
CM
o Ul
ui »M
> «-•
oe co
ui
(ft -«
at
o
ac ro
o >•
ui at
•> o
dC (3 A
Ul Ul (M
0> »-
IB
o
ro
f>
o
09
O O
os
o o
(O •*
(M
ON
«•»
ON
O
(M
14
o
CM
•A
CM
tO
CM
O
CM
ON O
ON r>
o
CM
O
V>
O
CM ON O ON
«• » « r- CD os
o o o o o o
•< CM ro *
B-47
-------
o
at •
ta o
k£ Z
O
O
(O 19
9 Ul
U CO
M
Ul
o u
M K
I-
M
oe to
u
«>
OD
O
o
x i-o
Ul Ul
ui co
O Ul
H- I-
M« M
Z «0
O 01
01
in o o o%
is. W
•~ 9>
•*
c* c* o *•
«S «s «V
«N «> «H
9. »4
ro
eg
(M
in
-• l«- K.
fw r> «• r>
o o o «•
V) 0& tt) ^»
«>«•««»
O O
t* ON ex e» c* t*
«S ON ON «N «N
vcMroo^in^rwcAaNOv-icMro^
OOOP>OOOOOOwM««
-------
a
o
at
»o
u
•
CO CK A
O N
W> O
3 Ul
CO
at
tO Ul
X I-
O Ul
Ul U
O >•
Z Of -J
O Ul =9
o a. -«
A W> O
Ul K «
> a
at oc at
Ul Ul O
to o u.
CO K
o u
O Ul
Ul IM
> »-i
CK O
Ot V9 O
Ul Ul CM
Cft »-
CD «
O U
Ul CO
CO
01
to
O
ro
o o
ON CD r* o 0* •*
CM
o
ro
ro
OB
«0
CN <• CO
«• n
*. CO
o
<*•
ro
o
o
tn
o
•»
CM
rs.
ro
in
«o
O
r>
w
01
CO
^ CO
cr> n
SO ^
•»* 01 ro •*
B-49
-------
ID
O
O O
«e >•
« at A
O CM
(A O
9 Ul
CO
(O Ul
Z f
o ui
Ul IM
oe
ui
»- CD
O O
D£ •-« CD
t- A OS
O 9-
z oe _i
O IU 3
o a, -»
o «> w
Ul f CSI
at oe ae
111 t» O
CO O
CD DC
O O
O
X
I »-
at o o
Ul Ul CM
co »-.
CD «
O O
CM
DC
O Ul
0
JC
to
OS CD CD
OS
OS
OS
CD
CM
(^ €^
OS tO
OS
OS
«r>
to
OS
OS
OS
OS
ro
V) O
n is-
o
CD
OS CM O
os n •»
OS CM CM
OS
OS
O OS
CM
» O
f> ^*>
•* CM
O O
«s •« 0
O T r>
sD CD OS OS
CM CM OS OS
CM
ro
O O
B-50
-------
CO Ul
Z I-
o ui
co
t*" tiO O^
z a «••
Ul O
o >-
O Ul 3
o a. ->
^^ OV NO
Ul K CM
Ul C3 O
CO O U.
O U
o
oe •
o o
tt Z
U
*-
« O
to
is.
to
CO
ro
M> O O
N* IO f>
ON O
M» ON ON
CO ON CO
ON
v>
o
r^ co ON o
o o o ~
B-51
-------
o
oe • ro
« O
* >-
tt fi£
•-> •
z oe _i
o ui a
o o. -»
A > o
AC O O
Ul Ul C«
to »-
o> «
o o
Ul (O
rsi c<
*••
(A
O Ul
^- ^>
k« •><
X CA
CM
00
r>
ro
ro
ON * o
os di «r
M>
ON
O
«4
ro
ON fs.
ro
oa
to
ro
o
ON
OS
CS1
00
ro
or*
0S
ON
ON
ON
••
ON
o
00
— csj
o o
o
o
to
r>
rs.
««
rs.
^ n
00
00
rs. oo
o o
OS
ON
OS
OS
OS
OS
o »•
«sj
ro »
B-52
-------
C9 O
U
K O CO
CM
«*. O Ul
CK Ul rsl
<0 Ul > •-«
X h- «* CO
O Ul Ul
M «: co «-
»- CO
a. • a
Ul O O
(/> O U.
CD OC
o o
ro
Ul Of
> o
(X \J> C»
Ul Ul (M
CO »-
CD co «* co w
o o
r<- rw co
v« «O CO
CM
CM ro o
o o ro
[> ON O OS O O O
Kcr>ON — ONOONCO
«sr>»sv»ONCO'«ls.
as «s -<
IT>VBN.COONOo«CM
OOOOO^^Mv*
B-53
-------
(A Ul
X H-
O Ul
o
OC
at tx. et
O (Sj
M> o
•~>
as •«• o
OS
OS
os
o
CSI
o
(SI
Os
OS
f> CO
CSJ M
n>
(SJ
V)
0*
CM
Ctf
OS
OS
as
as
OS
o
(SJ
o
(SJ
o
r>
OS OS
Os OS
Ul SC O
3* 4!
ae t* at
Ul 0 0
(A O U.
ea ae
O 0
i_>
j^
>^
•
o
z ro
0 5-
ui a:
>• 0
oc o o
Ul Ul OJ
CA t—
m
os *• is-
OS
o o o os «s n
^ s0 SO OS OS CD
fs» CO O OS Os (0
— OS OS
ts.
(SI
(SJ
»H
•
0
^
00
•
CSI
<0
\P N
•
(SJ fw
r>- rs.
CO
ro
00
0
J,
«^
O
-
y>
fl> OS
CO OS OS OS
^ OS OS OS
OS OS OS
as
OS
OS
OS
OS
•
OS
OS
OS
OS
•
OS
OS
OS
O
•
^
**•
ON O O
OS O •*
OS VO OS
OS —
-«
r>
r>. co
o o
OS
O
O x
CM
M
B-54
-------
U
V-
Ul
«E O O O
IK M CO
»- CO 0>
Ul U
e u a
o a. o:
O (A tv
u c o
>• -
«>V ^^
IK> m
O CM
W> (9
9 Ul
X *—
•-« •-•
•£ CO
Ul
to -<
en
o
*•
,J
«i
0s 0s fO
0S 0S
ro
0
»->
y>
•
ro
*•
CO
ro
CM
IO 0s ^
• • •
f 0> *
0S «*
0s
O »-
<• «M
0 0s
0s
0S
0s
O 0>
• •
0s
0S
0s
O 0«
0S
0S
0s
O 0
• •
IS.
CM
09 0» ro •*• co
». 0s <•• ro rf»
0S -«
X ro
Ul Of
> o
at o o
Ul Ul CM
CO »-
CD «
O U
Ul CO
M CM
••4
CO
Of
O Ul
35 (ft
CM
rv.
ro
CM
CM
*
ro
v>
OS IT> OS
0s s» N.
0S
o 0s 0s ro
•* 0S 0
so
CM
CO
CM
o
0s
0S
0s
0S
0S
0t
0S
ro
o
o co
if> se
ro *•
0s to
CM ro
CO
o
B-55
-------
to
U
«e >-
a DC o
o CM
(A 19
3 Ul
U>
CM
V> Ul
3C I-
O Ul
ui rvi
oc tn
ui
i- ec
X OC O
O Ul 3
O 0. •
Ul Of
»• o
DC «J A
ui ui r>i
CO »-
Ul (A
N C*
CO
Ul
I—
I £04
to
en
ON
in
r> o o>
CO
o o o PS «s co v* o o «s ON en o
»•< »«"* ^* V* ^^ 0^ I
^N tf*» V^ ^* 4
CM
ro
O
CO
o
CM
VO
ro
m
oo ro
^*t* ^O
ON
ON
ON
ON
r- o
fl-
r>
o
CO
000
iw rs. co
o o o
*• r>
«• o
o 10
0»
o cr> ON os o
r> «s os *s
O ^« CM
B-56
-------
ro
at « o
o 01
1-1 Ul > »•«
JC F- Of M)
O Ul Ul
>-i K 0» »*
t- rr.
« o o o
oe •*> co
I- O> *>
X 3 —
Ul U
O
x at is
O Ul =J
o a. CO f*>
ee of a:
Ul L? O
tf> O U.
O u> -C fO
•—»
x c* >-
o
O
UJ
CU
o
UJ (A
ISJ (M
ac
o ut
•- »- o
X CA
o o o
O O
Csi
o o o
ro 0s eo
0S
0S
0s
0»
0>
r>
•M
n
v» o^
M 0S
W 0s
O
o
o
0s
CM
ro
o
o «• o *•
(M
~ (SJ
o o
CO 0s 0s 0s
0S 0S 0S
m o
0 0
00
r>
O O O
-------
a
at • f>
O O
CO Ot A
O M
(A 19
CO
(M
O UI
UI IM
Ctt
co ui »• «-•
xi- at co
o ui ui
M K CO —
»- tr.
« « o o
I- SB 0"»
UI U
u
X K US
O UI 9
O O. > o
Of o A
UI UI (M
CO >-
CQ
ui co
rsi rj
to
ui
co
os
OS
OS
(SI
O
OS
o -«
m o • o os
ro os os
OS OS
OS OS
to
(SJ
OS OS O
00
CO
o
(SJ
CO
o
CSI
OS OS
o*
OS
OS
OS
o o o
OS OS CSI OS *• OS)
os (sj r> os (si ro
Os OS
-« OS O OS OS (SJ
ro (sj «r os os «•
OS OS
OS OS O
+• r> o os «r> os csj
ro cv «• os os os n
OS OS OS
OS
OS
CO
CSJ
ro
-< CSJ
O O
ro o •»
o to «•
r>
SB IS, CO OS
o o o o
o -« c* ro
B-58
-------
/*
DC
V) UI
X !•»
O UI
•t C
>-
o
13 O
K Z
O
•
CD OC O
o CM
«/> o
9 UI
z »—
•1 •-•
OC Cft
UI
(0 *><
at
o
0s 0s
ON 0S
0S OS
0S 0>
....
IO f*O SO 0s 0s
•x 0s 0s
0S 0S
OS
Os
OS
0s
flS
v
OS
OS
0s
.
0*
OS
OS
OS
*i
0s
0t
.
0s
0s
OS
Os
0s
Os
0s
•
OS
0s
0s
.
0s
0s
OS
* O
00
S0*
r-
Os
Os
OS
0s
0t
OS
OS
0s
•
0s
0s
0s
0 0s
0s
OS
0s
.
0
'
0
»
O
.
W OS 0s 0s O
9\ 0k 0S ££
5fc
ui a o
m o u.
aa oe
o w
•M
K
\^
*
•
O
z ro
A >-
UI Of
> 0
AC O fit
UI UI CSJ
V) t—
to ^
O <_)
UI (O
IM
OS
Os
OS
OS
OS
OS
*
*
SO
Os
Os
Os
Os
Os
•
OS
Os
Os
OS
.
0s
0s
0s
O OS
. .
O 0s
V) 0s
OS
OS O> O
«• *• o
in 10 so
o
ro
o
CO
*-l
0
.
_l
to
0
.
r>
H)
0» 0s 0s o
0s 0s 0s N.
OS OS 0S S0
OS OS OS
— «SJ
O O
ro
o
so
o
OS
o
~ CM
B-59
-------
=>
o
DC
U
IB ft: a
C< (V
•-! c c; tn
'•s e> ui
C£ Ul M
W UI 3>- >-i
x >- o; «•
o ui ui
—
Ut O
U>
z & o
o uj o
o a. :
S- o:
ct tf. a
o
o u.
CO
o
x:
UI Of.
•>• o
ui ui
cv c-j Cii e-»
h- ir> oo «s
in
CO IS.
to «
ro oo
~ o «•«
00 IO
OS O
CN
OS
OS
OS
OS
OS
OS
OS
(SJ
ni
ro
OJ
•*
(SI
1*0
o
«»
«o
CM
00
OS
*;
V0
s»
*
V0
OS
OS
os
OS
OS
OS
Os
OS
Os
OS
0 O
ro os r>-
Os OS OS Os !•>
OS os Os OS
r>
0
CD
O
OS
B-60
-------
n
a o
a o£
O
U) CD
co
(M
«O Ul
ac H-
o ui
CO
CO —
tr.
o o
*•• o>
Ul U
o
z « ta
o ui 9
o a. «
a v> n
ui x N
> «
ac at at
Ul C9 O
co o u.
oc at
o o
ui DC
> o
woo
Ul Ul N
CO •-
at «
o u
ui co
(O
at
o ui
OS
OS
O» O> CD O M CD
o os r> o
n «s «• co
ti is.
(O is-
OS
OS
O
co
o o
es n f<-
«s OS «s
(O
o
os M
— OS
o »-
B-61
-------
*
g
3
O
cv • w
C9 O
v: z
« >•
CD ec o
O (NJ
vt o
a ui
z t-
i- O UI
Of UI IM
(O UI > •—
Z 1- OC OT
O UI UI
>-• C CO -•
1- IT.
« 0 O O
ON
OS
OS
ON
ON
ON
ON
ON
•
ON
ON
ON
ON ro <**
ON ON •*
ON « <«•
ON
« o o ON o m
......
in IN. « ON «« vo
I*) PQ ON \o IN.
ON
ON
«s
ON
ON
ON
ON
ON
ON
ON
.
ON
ON
ON
M >0
.
IN. 09
ro -«
ON
.
ON
ON
ON
n
.
tNJ
«^
**
.
00
^«
oc >-> to
»- eo «s
o
z oc o
O Ul 9
o a. N.
ON
ON
ON
ON
ON
«s
ON
ON
ON
ON
ON
M 1*
00 ON
(NJ O
N.
-« ON
ON
ON
OS
OS
OS
OS
ON
ON
ON
ON
ON
OS
o
00
0
«N ON ON O O
ON ON v< r> (NJ
ON ON ON V) (SJ
ON ON ON »•« —
o
ro
o >i
(M
B-62
-------
to
a> oc
o
IA O
9 UI
* I-
»1 o
oe o e>
ui ui
0S
ON - 0S
0S (SJ 0s
os m 0s
0S OS
0 0S ^4 0S
*A 0s s« 0s
eo os so 0s
OS 0s
*•>
»
V4
IS.
(Sl
o
*
•*• *
(M fO
IO V)
»
(SJ
o
o
01
o
fs.
0«
0
IS.
(SJ
o* OS
0S «N
* OS
0s
« O O 0s
* (M f> 0S
IO IO IO 0^
0s
o
•4
o
n
o
«<
<*•
o
»
0 O 0
co 0s n
so o so
«4
n o
o i«>
r>
o
O —
(SJ
B-63
-------
^s
DC
(O Ul
Z t-
O Ul
M 1C
t-
« 0 O
DC •— CO
H> CD 0s
X«* ^—
••^ *"•
Ul O
o »-
z oe OL
O Ul Ul
o o. co
o> co ro
Ul K 0
> -
CD oe a
O CM
(0 O
9 Ul
z »-
i-i •-
at to
Ul
CO ^<
K
O
<••
•
o
z ro
A >•
ui oe
» o
0! O Ck
Ul Ul CM
CO 1-
OD «
0 U
Ul CO
ISl (M
>-«
to
v«
oe
e ui •
1- 1- Ct
•-I •-< •»
Z (0
o
in
CM
o
•
o
•
OS — OS os
. . . .
OS .* OS os
OS CM OS 0S
Os OS 0s
CM OS N OS 0S OS
••••••
(ft M OS OS 0S
Os OS Os 0s
r>
•
w>
4M
CM
•
V0
N.
•
W4
CM
OS <•• OS os
. . * •
OS O» 0« OS
os v> os os
0s OS Os
O«*S ^ offSi rih 0^
V* ^^ »^ »* **
(M OS CO OS OS OS
» 0» W OS OS OS
O« OS OS os
— CM tO 0 * 10 M>
o o o to o o 9
T> £ /.
O O OS
OS
OS
OS
O 0 O
.
O O 0s
. .
0s
OS
0s
0s O N. f.
.
tw ^» ^y ^^
N.«4co»ososroto
v«»osososn«
Os OS OS
.is-cootOMCMro^
OOOMM»«MM
-------
O
at •
o o
«c >-
CB 0< O
O N
W> O
O U>
UI N
> M
Ot CO
UI
«A UI
X »-
O UI
•"• K
^>
« O O
DC «- W
t- O) <*
X 3 ••«
UI U
O »-
x « a.
o ui ui
o a. M
o u> o
ac o o
ui ui (si
w *-
oo «
O CJ
ui to
M (SI
•M4
(A
ae
o ui
o o
OS »»
O 10
o o
o o o o o
o o
OS OS O V>
M O
^ <*•
r> v>
0000
-. CSI
•000
o o o o o
B-65
-------
o
ac • r->
ta o
•*. z
u
•
an a: o
o o*
01
"> O UI
oe ui IM
co ui »• M
z H- oe co
o ui ui
•-• a: to —
h- CD
*3£ fj oj^ Q
oe M o>
I- CD OS
z => *••
ui u
O !••
3£ O£ O»
O UI UI
o a. to
ui x o *
oe ae oe
UI C9 O
CO O U.
to at o
o u z ro
»•«
K O >•
^ ui oe
>• o
UI UI (M
CO »-
at ~>
co
O UI
OS
os
OS
ON
as
as
n
in
os
o
as
co
10
OS
«s
O
CSJ
(SJ
O
as o
OS O
OS O
as *4
OS SO
M» —
(SJ
es
as
OS
OS
OS
OS
OS
OS
O
CO
r» o
Csl
ro ««
r*. co
N- OS
m as
OS
os
OS
OS
OS —
r«- OB «s
o o o
~4 (SJ
B-66
-------
»N
CK
(O U
Z t~
0 Ul
t_
-»
Ul O
o t-
Z CK 0-
0 Ul Ul
U O. CO
0 CO CM
Ul C -•
CK CK CK
Ul (J9 O
V> O U.
OCi CK
O O
M
y;
^
M-
a
z
o
CK • W
CJ O
K Z
« >-
CD Of A
0 CM
t/t ^S
=9 Ul
Z »—
JC 0 CO
CM
O Ul
Ul IM
>• •-•
CK CO
Ul
CO »«
cr.
o
^
•
0
o >-
U QC
»• O
CK 0 O
Ul Ul CM
CO >-
CD 4C
O <-J
Ul CO
ISJ CM
»^
CO
v^
CK
O Ul •
1- I- O
Z CO
e
X
s0 0 0
0 tf>
f*9 v"4
O -« O
SO
OS — 0
. . .
is. -•
CM CM
* 0 ~> 0 CM O O
......
— CM «* W »«
CM W -« CM »<
°. °. *^ T * **! *^ . *!
O O * ••« CM V) CM
CM ro ro *o CM «•
^
SO O O
.
^N ^^ ^N
^* ^() «««
fO ** O
...
O C^ F*k
CM CM CM
Os ^ 0
• * .
as ro CM
so so *•
•*• 0 — 0 rs. ^ so
r^ CD rs. is. so 10 10
^* IO IO OS so N. O f*> IO N.
sOCOCfsOSCMCO CDSOS0S0
-*sprs.coaso
T* /" "»
as
as
as
0
*
as
•
as
as
as
as
.
as
as
as
.
CO IO O •*
•M 1-1 ro
as
•
as
as
as
0
IS.
CM
as
.
as
as
as
as
as
as
as
ro o 10 so
co co as so
*~ CM fO •«•
-------
V» UJ
X I-
O Ul
-
CD ac a
o CM
Ui i3
JB O CO
CM
A UI
UI rsi
> •-•
ee to
ui
(A -«
cr.
o
z ro
o >•
UI Of.
>• o
oe o d
UI Ul Ol
CO >-
ui to
M Cd
O UI
as
as
as
as
as
es
as
as
os
os
CD 0r* ID 9* *4 0*
S0 flN 0S
ON ^^
CO •M Os « OS
Os OS O
r»o to cr> ••- os os os *•
«O CM «« os OS OS OS
OS OS OS OS
OS
OS
OS
OS
Os
OS
OS
OS
Os
OS
o se
~4 «-l
ts. •«•
CM
SO
CO
in
*
fO
(M
OS *
OS
O
o
OS
OS
OS
CO M
v* OS
•*• as
«• as
OS
m
as
Os
OS
CM
O
o
(M
as
as
os
as
OS
OS
OS
OS
OS
OS
O
in
OS
as
OS
OS
OS
OS
Os
OS
OS
•O
o
o
N.
— CM
o o
r>
~* CM
B-68
-------
i O
z
CD IK O
O ui > •-
ae H- CK u>
o ui ui
I- CD
-
v ui at
> o
Of O
UI UI
(O I—
CD is.
(SJ
OS
torn
csioooc-ooo
M »- —• (SJ CM »->
(M
<*
OS
r>
c* f> rsj ro o M
« fw * w m os
o
rsj
o
ro
CO
OS
Os
CS
OS OS OS ID *>
•» «s o> «• os
0s OS os
(Sj
(M fs|
ID r>
to tM ro m v*
rr> in m •v
» O O O «• ^ O
B-69
-------
ro
MJ
t-
ut
C9 O
^) ^9
=3 UI
X t-
•-« v> —
ui K O
(ft O U.
at ae
o o
o >•
UJ IK
>• O
DC O O
UJ UI OJ
CO •-
an
••• r>
o o
0s
0s
0* OS
0s (SJ
0s ~
— fsl
B-70
-------
oc
UJ
o w
« u o
K «i <0
H- A «S
z => —
w u
o *-
z as a.
O UJ UJ
O ft. «O
UI K (SJ
> «
o: Of at
ui u o
i/> o u.
CD oe
o v>
o
X
oe • w
(9 O
* z
o
-
a> oe o
O (SJ
i— M
oe to
ro
tu
9-
oe
> o
oe o
Ui UI
Cft V-
oo «
o «_>
CA
oe
o ui
X V)
o
ro
CM
00
00 «*
(SJ
ID
(SJ
0s 0s 0s VB n
0S 0S 0S «• t«
0S 0S 0.
(SJ
r>
(SJ
0s
0s
0s
0S
0S
0s
0s
0s
0s
0
0s
0s
0s
o
(SI
o
ve
0
—
O 00
00
O 0 O O
- r
o
o
•
o
O O 0
— 00
0s
0S
0S
0s
0s 0s 0s
0S 0S 04
00 O
N- *•
OS O
O *
(SJ
0s
0s
0S
0s
0s
0S
0S
0S
csi
0S
0S
0s
(SJ
o
0s
o
0s
(M
r>
oo -«
10 *
(SI
r>
(M
ro
-« (SJ
B-71
-------
v> ui
z *-
O Ul
o
ac • n
C9 O
•
a ex a
O CM
V) O
3 UJ
Z »-
>-• M
OC V*
Ul
r~ CO
« u o o
C£ »-> CO
Z => —
Ul U
O I-
z be o.
O Ul Ul
O 0.
UJ X CM
> 00
•» * «
«« ro ^
o o
|w O
— n
•*• CM
Cll
10
r>
CO
ro
n
0s
0s
as r»
OS 0S
OS 0S
ID *•
-------
(A Ul
X »-
O Ul
« u o
z a —
ui u
o i-
z oc o.
O Ul Ul
l> Q, CO
a to o
Ul K W
> cc
ae (X At
Ul U O
to o u.
cc oe
o o
o
X
o
ae •
ts o
ea ae o
o CM
§^ m
=9 Ul
X »-
x u to
O Ul
Ul ISJ
x ro
e >•
ui ae
> o
oe o o
Ul UJ (M
at - 9> v* o
r> ro «
«• n •>
^rooooooo
CM f^ *^ CM
W
r>
CO
•s «\
«- CM
ro
n
ro
to
ro N. ».
CM Is.
o 10
o n
r>
(M
O
in
«>i
ro
*M ts.
o
CO
o
ro
n
CM
o
OS
CM
«* CM
IS. fO
(O 9>
O O
B-73
-------
at
CA Ul
Ul
« o o
Q£ *•* OO
h- CD OS
z =a -«
Ul O
o
Z DC >-
O Ul O
o a. o
o to ro
ui x o
> M
(A
(A
e
A >•
ui ac.
>• o
oe. t-
oa so CO O 00 W
•- e» o
O» 9- O O
0« «s OS
so
N-
O»
O
rs.
o
fv.
<* •*•
O OO N O
W
-------
co ui
x *-
O UI
« u o
at *-• o>
ui u
o
X 0* »-
O UI O
u a. o
o «o ««
UI K O
UI U O
10 O U.
CB IK
O O
O
ac • to
o o
oc, o
O »-•
CK (O
UI
to -«
O >•
ui at
> o
« O A
UI UI di
V> t-
03 «
e <->
ui co
N (M
OC
O UI
r>
in
or
o
in
o o
00
to
•«
ON 9>
•» — o
o •* **
« eo «•
o o w ^ ° ° °
o o^
B-75
-------
o
at
r>
at
to ui
z *»
O Ul
•
CO Ot A
O 01
to o
S> Ul
z t-
»- •-«
at co
ui
to «••
er,
o
z ro
A >•
ui at
>• o
at o A
ui ui o*
co *-
ui co
Nl (M
CO
O Ul
r~
(M
n
w
ro
*• o o o <••
o •* w «4 M O
•_><«<«
9k «S •» tft
tft O
CO
««
(M
eft
o
CO
CM
o
M
O
CM
ro
*> tft •»
tft «N tft
tft tft tft
N ro
«• «*
««
« *
(M (M
ro tft
CM
eo —
M
CM
(O
ro
O !•»
O «
B-76
-------
o
at • r->
C9 O
CD oc o
o w
10 19
cv
«0 Ul
XI-
O Ul
•IK
I-
« o o
CK M CO
Ul 0
o
JE a
« tt «
Ul U O
«0 O U.
« ac
o o
Ul
CO <•»
«
o
U OC
>• o
« O
Ul Ul
(O I-
Ul C0
IM (M
r>
oo
n
ID
in
»k O O
o o o
in
tO «N
r>
•» O IO fw V»
r^ «• co o «•
tO IO N « o «• N.
\« to to *
to r> N «o
o «* h-
to o o o o
r>
V-
o
B-77
-------
o o
O CNJ
CO
at
« eo
ui o
o
X CK *-
O Ul O
o a. o
Ct W Ift
UJ X — T
at at at
ui is o
w» o u.
co ac o
o u x t*
K Ck »•
w Ul OC
» O
ae u o
Ul UJ CNJ
co i-
ui co
IM CNJ
CO
in
ON «
«•• CO 00 •»
C«J CNJ n
ON ON ID
ON ON rv
SO
n
"- « O
ID N
v>
CNJ
0 0
••« W
CN1
V)
(•k glJL ^^ ^^ ^^
W W *^ ^^ ^^
CNJ
O
CNJ
to
ID
IN. •< CNJ O
n fo %o ID
o
CNJ
O
CO
CNJ
O
O
rw
— n
ID ON ON
N. ON •>
ON ON
CNJ ID
*• VH
NO IN. CO OH O —
o o o o •• ~
B-78
-------
«A ui
z i-
o ui
UI CJ
o
z oe *-
o ui o
o a. o
o to a»
oc PC «
UI 19 O
co o u,
OD at
o u
o
at • ro
a o
K z
O •->
a* «o
da
o
o
z
>•
UJ UJ CM
CO »-
at «
ui to
IM CM
(O
z to
ro
OS
OS
OS
OS
os
«s
OS
OS OS
CM
<*• o
n
ro
o ea o
SO
OS OS Os
OS OS OS
a
»
ro
ro
os
OS
OS
OS
OS
OS
v>
(SI
os 5p ro rs. to is. eo
CM CM CM CM CM CM
-------
at • ro
cs o
o
00 OC A
O CM
K W CO
CM
O Ul
Ul N
at co
ui
K
to ui
co -«
Ui K CM
Ul U O
CO O U.
CD oc o
o w z n
••4
K A >•
xx ui oe
:> o
oc o o
Ul Ul N
co »-
CO «
O U
IM
CK
O Ul
CO
CM
vO
IO
CO O 00 0^
IS. N. |»- ON
«•• o
o o
ON
9i W> V>
^^ PO ft
CM
CM
ca
CM
CM
r>
o
CM
m
n
CM
o
SO
CM
n n o
OH ON n
o
rw
00
o
ID
B-£
-------
r»
o
is o
co ac
O
u> o
9 UI
a: >-
tO UI
«-<
at co
ui
co «-
ui CK
> o
IK O A
ill UI CM
CO >-
CD «
O *J
UI CO
N CM
o «*>
CM a-
ro
(M
ts. »«•••«*
r> *
c* o
0s o r> w»
M CM ro
CM
C*
CM
o
in
CM
— .» 4k ^%
|S_ «^ ••« *• O* W »»
M- in
CM CO «
r> co co
CM
CM
«s O O
o
«0
ro
.4 CM n o
o o o i*
N. CO
v* in «7 w» 9> CM
!•> T ¥> V> VO M-
^ CM
ooo
B-81
-------
QC
co bi
« o o
CK •» CD
bJ u
u
x at *-
o hi o
O» ON ON ON ONON«ONONON«rO
Mroo«w«*rN.««No««
-------
HUS BACKGROUND
TEGORY NO.
20 3 4
«>
CM
at
a* ui
o ui
z s> «
Ul U
u
z at i-
O Ul O
u o. e
d to o
ui K ro
>• «
CK K at
Ul C9 O
CO O U.
CO «
O 4J
M
K
v
O Ul
Ul ISI
at to
Ul
to
CO
o
z n
o >•
ui at
>• o
ac o o
Ul Ul CM
to H.
CB cr> «s <**
O O O O
fo a^ 10 c^
••
cf» to «r
w» r> N.
000
O O IS.
— CM W O
« « o ra
ID
M> Iw O>
O O O
o «
B-83
-------
APPENDIX C
SUPPLEMENTARY METEOROLOGICAL DATA
The hourly meteorological inputs used in the ISC Model short-term
program ISCST 24-hour average particulate concentration calculations for
the periods before and after the initiation of fugitive dust controls at
the Armco Mill are listed in Tables C-l and C-2, respectively. The hourly
meteorological inputs for the periods before and after dust controls were
used to develop "seasonal" joint frequency distributions of wind-speed,
wind-direction and Pasquill stability categories for use in the ISC Model
long-term program ISCLT "seasonal" average particulate concentration
calculations. The wind distributions for the "seasons" before and after
dust controls are listed in Tables C-3 and C-4, respectively. The wind
measurement height for the period before dust controls (see Tables C-l and
C-3) was 100 meters and the wind measurement height for the period with
dust controls (see Tables C-2 and C-4) was 6.7 meters.
C-l
-------
Ul Z I
-I Ul
•-• X I
o a.
at x
a. ui
>• >-
*- «
•x O
co
M Ul I
« « I
t- U
CO I
oooooooooooo
oooooooooooo
ooooooooooo
ooooooooooo
tt
o
u.
Ul
OB
CO
O Of
a. o
ac
M ui
• Ul
19 K
Ul
Ul
a.
oooooooooooooooooooooooo
oooomooooooooooooooooooo
OONOItONOOOOOOOOOOOOOMOIMOO
oooooooooooooooooooooooo
o o
ooooooooooooo
o o
a.
x
Ul
•p* co
i
o ui
«-«
I
>•
I «C
I
I
I
I
I
I
I
I
I
o »«
Ol
-------
I
t~
Ul X t
-J Ul
•» X \
u. o
o a. i
Q£ X
0. Ul I
ooooooooooooooooooooo
ooooooooooooooooooooo
oo
oo
>• >- i
»- AC
U
C *- J»C bl
w a: i- i
H- ui -in
•-•OKI
• O ul
*• « o at i
o oe v ui
a. - e»
+* lij m i ^*
9 a. x i
to «^
i
i
^s
(O I
OC Ul
a o ui i
o >- oe
_i o u i
U. Ul Ul
C-3
-------
-J Ul
•* ZC I
u. o
o a. «
at x
oooooooooooooooooooooooo
oooooooooooooooooooooooo
oooooommonommmooinoonoooo
>• >• i
»- Ot
M O I
-I (9
«•• Ut I
DO »-
« o x: i
o
a.
~«
Z Ul CO I
«— ui a.
9 a. c i
to
a» o ui i
o »- cc
u. ui ui
oooonir>onooooinoir>or>ior>or>v>inm
o «
Ot Ol
01
C-4
-------
I
t~
ui at i
-J UI OOOOOOOOOOOOOOO^OOOOOOOO
o •-• at i oooooooooooooooooooooooo
aiocx
a. ui i
i
v >» i
i- tt
M O I
at H-
« « i
m i
i
. <«.
A. CK I
K »- tt UI OOOOOOOOOOOOOOOOOOOOOOOO
ui ac »- i ooooooooooooooooooooooom
I- UI • UI OOOOOOOOOOOOOOOOOOOOCJOWP*
• o ui
»- « a CK I ^OOOOOOOOOOOOOOOOOOO O
o ae « w
a. t9 a. i
i
**• ^»
O * *rf f
S> K • I
X sx I «N
O «••
O I
w ^* >•
ca i- to i Otf X* ^^ ^i ^^ ^> 1O ^1 ^^ 40 Irt PO **^ ^^ \9 ^^ ^* ^^ ^0* ^^ ^^ fV ^ft ^^ ^1 ^fr
i ^"^ OJ ^J I flD f^ IO ^f TO ^1 *^ 0^ 00 ^^ ^0 ^f ^9 CNI ^t ^1 (SI Oi Of VO ^^ ** ^0 ^f
*•* UJ ^^ i CSI ^** v4 v^ *^ ^^ v4 v4 *^ *4 v4 ^4 V<4 w*4 ^4 ^^ ^*4 ^^ ^* ** ** W v4 v4 v4
UI K X K
J -^ I
a>
oooor>tr>r>«r>o>r>oir>r>r>r>r>r>oo
OI-CK
%^
i
i
DC
X I
C-5
-------
U Z I
-J UJ
O M Z I
x u. o
MOO. i
S OC X
OOOOOOOOOOOOOOOOOO^OOOOO
OOOOOOOOOOOOOOOOOOOOOOOO
>• >• I
« « i
«- o
w i
at i
t- *c ui
o oe v ui
a. o o.
<*<*inr<>t<>i<'>n><+r<»
o
O
u
UI
_l
CD
O CO
v I «s
•"• >•
»- W I «
•-I UJ Ul
K X K
1U CO
i
i
S"»
O> 1
111
u i
o»-oe
-j o o
u. ui tu
C-6
-------
I
»-
UI Z I
—j t^ ^% ^y 4^ ^^ 4^ 4^ ^^ 4^ 4^ ^^ 4)* 4^ 4^ 4^ 4^ 4^ 4)^ 49^ ^^ 4^ ^^ 4^ 4)* 4^
OMZ I OOOOOOOO^OOOOOOOOOOOOOOO
90CX
O. UI I
I
>• >- 1
I- Ot
•x e i
-4 U
00 t-
«
o *•
O I
u» i~ m i z
•-• UI UI I W —
UI K Z X
-1 vx I
IB
« I
I-
I
•xuia.
I
X'h
CO I
Of lit
a»oui i oooooooooooooooooooooooo
o >- oe
XX
I
Of
X I
C-7
-------
UJ 3E I
.J UI
O M ac I
x u. o
•» o a. i
9QC3C
& tot i
OOOOOOOOOOOOOOOOOOOOOOOO
OOOOOOOOOOOOOOOOOOOO-OOOO
oooooomooooioinooo«ir>r>ooooo
>- >• I
-I O
MUI i
a. ac i
KH-KUI
uix »- i
H- ui • ui
ix O K I
• O 01
KttAaei
O « v UI
B. u a. i
OOOOOOOOOOOOOOOOOOOOOOOO
oooo
ooo
ui
e
O
v- I
UI
> CO
ui to i
«> ^
i
i
t*
CO I
« UI
a* o ui i
o i- oe
-j o ui i
u. ui ui
>A i
w
I
I
at
a i
O
X I
C-8
-------
UI
ooooooooooooooooooooo
ooooooooooooooooooooo
ooo
ooo
fr- O
a.
K
UI
H-
»x 19
• O al
H-CO
e te -•
ui
o. i
I
OOOOOOOOOOOOOOOOOOOOO
OOOOOOOOOOOOOOOOOOOOITi
OOO
lOlOO
OOOOOOOOOOOO
O O OOOOOOOO
H- UJ I
o
^ I
I
">
CJ) »- «0 I
zz«
MOUI i
o
UI
~ ui a.
CO I
at ui
O UI I
OOOOOOOOOOOOOOOOOO'OOOOOO
u. ui ui
a:
a
o
C-9
-------
UI Z I
_j ui oooooooooooooooooooooooo
MZ I OOOOOOOOOOOOOOOOOOOOOOOO
oc x
a. ui i
I
>- >• I
•- o i
-I C9
nnv>ir>ir>oooooooooooooov>ir>inir>
•-« O K I OOOOOOOOOOOOOOO
• O ul
»<- *»<»W. o* ~ o ff> ^ o ro « r» «& co OK «» o-o—«-« o ^ c» N.
e
CJ
O h- 0> I 3
zzae -»rwttakMN«409«Mao, •4-MN.^t- •*.* ^-•<«-'«-<«-xao
>-i o ui i c*c4C«iMrofoo<»*«fo, «-•> «*,•»«•*,•*• •*•*•* •« r> r>
XK«»~ »«ir>ir>inir>in^n>v««o9O^nir>ir)irfmioif>in *>•«» in
UI K Z K
-I w I
I
I
o"> <*cB
ui «o i ro m
«o ^
i
*\
(O I
UI
UI I OOOOOOOOOOOOOOOOOOOOOOOO
UJ UI
> O I
O
z
C-10
-------
I
k-
ui x i
-j ui
«-• X I
u. o
es a. i
• >- I
>- at
•-» o i
I
I
u»
ui •
M o
Ul
t-
ui
0. 13
Ut
O.
o
Ul
oooooooooooooooooooooooo
oooooooooo^ooooooooooooo
oooooooir>oir>ootr>ooir>oinir>ir>ir>ooo
oo^oooooooooooo
ooooooooo
UlU
e
o
C9 »- (O I =k
x x ae -»
•— C9 Ul I i
X *-> H* «•
»• Ul Ul I ««
tft V*
I ««•*•«•
I
I
I
ae ui
o ui i
»~ w
Ul Ul
>• A I
w
I
I
a i
•ooooooooooo-ooooooooooooo
OI OI
C-ll
-------
Ul Z I
_J UI OOOOOOOOOOOOOOOOOOOOOOOO
t-i Z I OOOOOOOOOOOOOOOOOOOOOOOO
u.o oooooooir>oir>ooooir>ir>ov»ooir>ooo
o a. i r«>i»>i«>i>of»>r<>wc««rofo
at x
a. uj i
>- >- i
i- at
M O I
t- o
CO I
I
. A
a. at i
Kl-ttUI OOOOOOOOOOOOOOOOOOOOOOOO
• O UI
I- « Ck a£ I -OOOOOOOOOOOOOO
49 Of ^" 1*1
a. 19 a. i
K • I ........................
ut ca o <* » ro N w *• « ir> co w> o -« w tn'w »•*.«*• » n ** o w oo
^ UI I QO CO, CO CO CO CO CO CO CO CO ' CN CN tfN OK 0N 'ON ON 19* ON • ON ON ON 00 11QO
O ONNCilCdOIOIMOINCMCNiCtlCilCMOICMMNNMOIINiNCiiril
v I ««
I UI
~ X
O •- 00 I 9
zxac -»ir»ir>'M*r^orov«ori»«*oNiy>aN*>»>*NON**wo<4»
Mcauii M«Ao. n to w » w w •»• •» «* o
x «-«
UI to I O«^i«<««4>AC4>->OOininv»t4 C*
ui a.
o. K i ^rnMfMroMCtfra^r^rorooioir^
CO w
co i
ot ui
O UI I
»- CK
o o i
I cMCMMMnroraNCMc^MoicKMncy n n n n n n
i
i
t ~-'C*r«>'««-r>«*r^co«r»o»-o*f>'<»-r>««f^«ooso-«o*r'>*-
O
z i
C-12
-------
o
Ul
Ul
i-
Ul X
_l Ul
A •-• X
X U. O
HP.
T \
K STABILITY Pf
ER> CATEGORY E>
Ul X »—
K Ul • Ul
• A Ul
i- « A ac
e oc w ui
a. o o.
a.
UJ C9
O
vx
19 *- CO
X X IK
X •-• *-
XXX
» a. K
00 ^
CO
at ui
a o ui
_l 0 0
U. Ul Ul
at
9
e
X
i
i
i
i
i
-9 v» 00 ON ** ro
O 00 0» CO CO 00
1
1
1
r> «s w> w»
1
i »-• c* ro •*• m
i
vovt«rocMM«*<«
%.»r4oooto^in^ros.fto
CNI fJ *^h ^* C^ ^^ ^^ CNi FO vt Ift *^ ^P ^Ct* ^O w* C^ ^* ^»
CMCMCMOiCMCMCMCMCMCMCMCMCMCMCMCMCMCMCM
v»romi<-ocM*-toooaDO»ooo»ooooaoaoa»«o
•MOorooNo»o>cMrwcMCMCMCMr>ooroaN«soNO»
ooooooooooooooooooo
*^ CM »•• ^> ^* r^» CM ON oo ro oo ro ro o ^ o <^ r^ ••<
r^io^rH»fwoioiooooooftoooNr^«A^Aro«<«ro
\«r>.oD9>o>«cMro«-m««rwoo0No^«CMCMCMCMCM
C-13
-------
ui x i
_l UI OOOOOOOOOOOOOOOOOOOOOOOO
M Z I OOOOOOOOOOOOOOOOOOOOOOOO
u. o ooooooononomnmonminnnninmin
OO. I Wf«>l«>l«>fOf«>»or«l«M^i-*««-iNCilCMW• >- i
M O I
~i o
MUI i
CD I-
ioior>«r>oooooooooooooooooo
ui • ui OMromttoMooooooooooooooooo
M O K I OOOOOOOOOO«OOOOO«»OOOOO«
*- »• » * o w « « o «y ^ o1
»- UI I «0«»^««0609««^»>W«'OOO'O
A W»WCSIWNNW«MCMW W H> W dl W C* M Ol «* Ol
^ zxat
i MOUI
o x •-• »-
**UIUft
UI K Z K
ui m i wwooir»w»«vo«*ooo*»M»>«l»«*ow« •» m
CO I
at ui
atoui i oooooooooooooooooooooo
o »- ae
^ u o i *ww<*^-ir>w»oo»-<«*»«-»ro «* »• o a> r«> m n
U.WUI o^o9»oo
-------
Ul Z I
_j Ul OOOOOOOOOOOOOOOOOOOOOOOO
O »• Z I OOOOOOOOOOOOOOOOOOOOOOOO
:• at x
a. ui i
i
3- >- I
H- at
M O I
-I (9
flB I-
-------
O *- X I
z u. o
M o a. i
:» CK X
a. ui i
i
>• >• i
K- at
•-• e i
-i o
oooooooooooooooooooooooo
oooooooooooooooooooooooo
nv)r>r>tnnoir>inir>ir>inoinoii>inir>ininoooo
•- o
CO I
I
• A
a. oe i
K »- K UI
ma: H- i
»- UJ • U)
•-•UK I
• O 01
t-«
OOOOOOOOOOOOOOOOOOOOOOOO
A
ui
U
UI
-I
CO
ui en
i
i
**
CO I
oe ui
a o ui i
OH»
-------
Ul X I
_j w oo- >- I
»- M
M O I
j (9
« « I
k- U
OT I
I
i ^S
o. at i
Ul X »- I
t- Ul • Ul
i- C9 K I
• O Ul
I- « O (K I
O M v Ul
A. «0 O. I
I
oooooooooooooooooooooooo
«tnw>*>wooo^ooooooooooooooo
WMMWWOIOOOOOOOO'OOOOdO^OOO
o o o o o o oooooooo ^ o o o o o o o o o
Itl
9
X
O
O
i
o
Ul
Ul U
I- Ul I
>*• O Ul I ^
X •- >-
•-> Ul Ul I ui a.
» o ui i
o i- «
-i O (9 I
U. Ul Ul
> O I
V
I
ro
en w to
C-17
-------
-J Ul
C> ~< X t
x u. o
M O 0. I
a» of x
o. ui i
o
o
oo
oo
oo
oo
ooooooo
oo
oo
ooo
ooo
ooo
ooo
o
o
ui ar
>• >• I
i- at
«•• o i
-I C9
•- Ul I
CD i-
>- o
to i
I
^
at i
ui
»- i
ui
K I
Ul
=>
a
u
I
Ul
_l
OB
«
lo
f>
o
oo
mo
f>OI
oo
I- « O OC I
o oe v ui
a. o o. i
i
• M i
a.
K • i
UJ C9
»— Ul I
o
^ I
I
«s
C9 t- 0» I
X X ttl
«-• C9 Ul I
X •- >-
1-4 Ul Ul I
XXX
•w I
I
I
X Ul 0> I
a* a. K i
I
I
CO I
at ui
at o ui i
o i- at
_j o ca i
u. ui ui
i
i
at
a i
o
X I
o o
o o
o o
O •
oo
oo
oooooo
oo
oo
OO
oo
oo
oo
OO
oo
oo
oo
ooo
ooo
ooo
onn
o
o
o o o o
oooooooooooooooooooooooo
f> V4
C-18
-------
Ct •- X
ae u. o
oe x
O. Ul I
oooooooooooooooooooooooo
>- >• I
»- OC
M O I
-I O
f« Ul I
00 t-
CB I
*- M Ul
oooooooooooooooooooooo-oo
o
a.
o
Ul
a
M til JC I
O Ul
« O Oe I
ae K* ui
u o, i
i
^»
• M i
oooooooooooooooooooooooo
......•••• -*^. •••-••••••
ooooooooooooooooooo
X Ul (O
*- ui a.
(ft +*
en
atoui
O «- Of
u. ui ui
oooooooooooooooooooooooo
C-19
-------
o a. i
at x
a. ui i
oooooooooooooooo
oooooooooooooooo
ooooooomoiftoooonm
rowwrorofOrotMCSiw.*-*-*'.-'
O O O O O
o o o o o
o o
o o
f*>rO
>•
at
o
Ul
lo^-roca.- — -••«*h'>r<>'«rir>»***
v- u
CO I
i
f\
at i
t- *: LU
x i- i
ui • ui
•-> o x i
o at >- ui
a. u a.
oooooooooooooooooooooooo
v>tf>v>v>inv>o«oooooooooooooomiD
rorororoforoNooooooo^oooooocaroro
oooooooooooooooooooooooo
oooooooooooooo
Ul U
i cocftescfeasertcNtfsc^ostftoo
o
CJ
I
o
u i~ co
M U Ul
X "• t-
»• uu ui
OOOOOOOOO9k0^9>
I =>
-3'«-'«-ir>m>»rco'oooooooN.roooi
i cco-«*«eoff> — w^rr>r>-«»«r>»»«r»»«»4»>»\o-"O
ui co
ui a.
a. x
en +*
co
ae ui
o ui
i- ae
u o
ui ui
> A
««N-rw(«irwr<.««w>tn«oin^ro
« rw «• •• w
nnoooonmoo
a i
o
X I
O* C4 tt CM
-------
I
»—
ui x i
-I UI
O •-* X I
x u. o
*~ o a. i
a oc x.
O. UI I
oooooooooooooooooooooooo
>• >• I
oooooooooooooooooooooooo
oooooooooooooooooooooooo
ooooooooooooo
ui
O
O
I
u
UI
00
ir>oioir>«oooir>ooioooiotf>io
^* fo r*> ^*
o r> o
fw «f, OS
C-21
-------
»•• 3E I
u. o
OOOOOOOOOOOOOOOOOOOOOOOO
OOOOOOOOOOOOOOOOOOOOOOOO
ooor>or>oomu>moooov>ooioooooo
v >• i
t» at
M O I
_i o
•-> u< i
CD »-
« I
I
*s
oc i
H UI
o ae w ui
a. o o. i
OOOOOOOOOOOOOOOOOOOOOOOO
»ir>ooooooooooooooooooM>ir>«>in
mroNOoioooooooooooooooirarQraro
OOOOOOOOOOOOOOOOOOOOOOOO
........................
O OOOOOOOOOOOOOO
ui
ui o
ID •»
ae
o
U
UI
v i —
I >•
*•• -i
(9 »- «O I =>
x •— •-
*^ UI til
*
i
i
v>ov>or>ir>v>ir>nooH>H>ooooooinov>or>
C-22
-------
I
I-
ui x i
_J Ul
"-. 3S I
u. o
o a. i
Of X
a. ui i
i
>- s- i
oooooooooooooooooooooooo
oooooooooooooooooooooooo
4M. ^v .At. J*. rfV ^V ^k IrfSi ^t. ^k •«* MS u^ ftA UN W\ ^^ ^ U\ ^^ ^^ ^ ^^ ^fc
« « I
t** &)
«O I
I
. /•*
a. oe i
K»~MUI OOOOOOOOOOOOOOOOOOOOOOO'O
mUK I OOOOOOOOOO^OOOOOOOOOOOOO
• O Ul
o ae v ui
a. (A a. i
i
^^ <^
O -XI
=» K • I
Z w I *•
O
O I >-
w *** «J
€9 K" CO I 2>
Ul X X E
-J *s I
aa
-------
UJ Z I
.J UI OOOOOOOOOOOOOOOOOOOOOOOO
O «•« Z I OOOOOOOOOOOOOOOOOOOOOOOO
z u. o oooooor>ov>v>oir>iftr>oir>oooif>oooo
«-• o - >• i
»- AC
••» O I
-i ta
Mbi i sfttotAtatAin^m'r'rracvtMfvra'rroKim^-invftse'*
an t-
« v>v>»>v>ooooooooooooooooir>ir>tf>
t- ui -ui roror^roroNooooooooooooooMroraro
•-•OK I OOOOOOOOOOOOOOOOOOOOOOOO
• o ui •
»- « O Q£ I OOOOOOOOOOOOOO
o oe v ui
o. o a. i
i
s*. **
A -Ml
ui a. oor><*»o»OB'*OM*>ir)oov«c*C4C4'*if>f«.^oir>
3 K • I •
X U«U ONM~«OO0>avON<»V)««*.rk.ts.tt«»«*.«Af««l«- A •'tMtVNCVCgOINCMMOICVCilNmNNfVNaiMCiMCiJN
Z v I v«
O
O I >-
w ^ J
O t- o«AMK.i^>r9«s«A(M««ir»«40»cDMaDir>»«w>cx««*«s4»c*t».to»v'»o^^»<«»-«»-««c<»ir>«sr^w
o XM»> r»a>a> <«4to«o««wo«»w»<>w?«>»^t<)K4?<>»«>^rto
MUIUIIM M^4M«.mMM«>MV*^«v«v4
ui «c z jc
_J w I -,
a
« i
i-
i
oo<^ >»ir>
z ui to i di«»M«»'«>^rir>tr>oirtowc4»>wf»c*twcaio»rw«--«
•xuia. •
» a. z i <*«ir>«-^r^rfon^rm«ir>«««in<«u»«««v>m<«fwi«-
(0 *^
I
I
^«
«0 t
ae u
aoui i ov>oo»mtr>v>mv>ooov>oir»oir)oir>*>ir»ino
o»~ot
-j u u i oa»oin0kO4r»e»c**»a>f«in««o«-*««««^o»^a>«k«k«o«»a»a>o»««9«««
^ O I v« v4 v^ «^ V* *4 qp« «M «^ *^ v« v« «* v4 «« v« «4 ^« W v« w« V4 W 4^
\*
I
I
oe
C-24
-------
U X
.J Ul
O M X
u.
o
at
>- >-
ooooooooooo
ooooooooooo
oo
oo
oo
oo
oo
oo
ooo
ooo
oo
oo
oo
oo
oooooooooooooooooooooooo
v>ir»w>mooooooooooooooooir>ir>ir>i»
wwrowcrfooooooooooooooftirorowro
oooooooooooooooooooooooo
i
o
o
z
>-<
>c
x «>:
o ui
z ui
a.
«»
DC Ul
3» O Ul I
O t- Of
-1 O t» I
U. Ul Ul
>• O I
«rf
I
I
AC
9 I
C-25
-------
I
I—
Ul Z I
.J Ul
O » Z I
x u. o
M O O. I
a* oe »c
>• >- I
-I t- Otf
O >i O I
OC _l «9
x to t-
O « o oe v ui
>• a. *a o. i
«c
O I
^N
Ul -Ml
tt. K • I
I %* I w«
U U,
X I >-
Ul H- •«• J
_i o i- I
O. I
X
M I
« MUIO.
19 I
o
-J I
o ^
ae mi
o oc ui
ui aiouii oooooooooooooooooooooooo
I- Ot-IX
> O I
>• •w
I
oc
a i
o of
x i
C-26
-------
oooooooooooooooooooooooo
oooooooooooooooooooooooo
minonmmininoininoooooirtoiftoonon
i
>- >- i
t- at
M O I
-J O
M Ul I
to i-
v>^r«««>«>n««>nnran>r4'ir>'<«>
xe t- M ui
ui x i- i
t— ui • ui
»-i o x: i
• A 01
»- « & oe i
o os w ui
O. O A. I
oooooooooooooooooooooooo
oooooooooooooooooooooooo
oooiooooooooooooooooniruocdo
oooooooooooooooooooooooo
o o
oooooooooooooooo
u
e
o
I
U
Ul
I
• at i
C • I
K-UI I O>«»«N««9>9»9>9>«S0t«OOOOOO
OOOOCTv«N9>«>
r*j ^9 ^o ^o 01 ^i wi c*i
U) *- V>
Z X IK
»-« O UJ I
X «- »-
M Ul Ul I
(O
=>
o
•«inr>ir>s«^t*v«<*f*.r>-pwr>-N.»^N.r».»-»»is.'*<»
rs.co«\o»-»*l»>sO<«^O^«»»««N.«'-''-«»«««<-i«»«<«-«
•«-«ir<>««r4w»<«>m^>wfta><*N.ir>trt»'«-<«-ir>ir>«Air»io
Of Ul
O Ul I
t- oe
O O I
Ul Ul
> e i
i
i
oc
9 I
O
X I
00000000000000000000000
N C4 N (M N
C-27
-------
t
t-
Ul X I
_l UI OOOOOOOOOOOOOOOOOOOOOOOO
Ck »i X I OOOOOOOOOOOOOOOOOOOOOOOO
3» OC X
O. UI I
I
>• >• I
l» OC
•-« O I
»-ui i M»«o*»«»ir>*^r^r'r»*-i»>'»'»*i«>wi«»'*^''«-*<«-'»
oa s-
K • I
X ^ I —
o
O I t—
U) h- CO I S»
UI K X K *
—I X* I
OB
-------
Ul Z I
_l Ul OOOOOOOOOOOOOOOOOOOOOOOO
AMX I OOOOOOOOOOOOOOOOOOOOOOOO
xu.0 r>r>ioif>ioir>r>r>or>oooir>ooir>ioir>ooooo
atoex
o. ui i
>• 9~ I
I- OK
•i O I
O> »-
« x • i
z ui o o«*v>nnir>inin*-0'k<>4i'«>«>«rir>>0inir>r<>r<>-«o9>v>0l»
X V I ^4
O
0 I »-
C9 I- W I 9
Ul C X K *.
l • ^/ I
CD
«C < I
CO +s
I
tt Ul
a o ui i oooooooooooooooooooooooo
ot-oc
> A I » "* ~
I
oe
X I
C-29
-------
Ui Z I
_j Ul
• >• I
OOOOOOOOOOOOOOOOOOOOOOOO
oooooooooooooooooooooooo
•- ui i
I- O
(O I
a. cc i
ac v- ^ ui
Ui ae u- i
•-• « K i
• o ui
>- « o oc i
o ae v ui
tt. C9 O. I
OOOOOOOOOOOOOOOOOOOOOOOO
oooioioooooooooooooooooooo
OCXCJWroCMOOOOOOOOOOOOOC^OOOW
OOOOOOOOOOOOOOOOOOOOOOOO
........................
O OOOOOOOOOOOOO OOO
ui
a.
«:•
utu
z
O
O
ui
acxoc
I CO
3
C3
— c*
CO I
ae ui
o H.
C-30
-------
Ul Z I
-J Ul OOOOOOOOOOOOOOOOOOOOOOOO
O *-> X I OOOOOOOOOOOOOOOOOOOOO-OOO
z u. o •oooooor>oiooir>oooonooioir>ir>v>ir>ir>
MOO. i rr>r«>rorofOfor<»cj — cMcoioiciir«i(MCM
a§ oc x
a. ui t
i
>• >- i
K. ac
«-• o i
_j «
MUII »**»<*<*««»»*t»)Cuw*-WH>WH>e«itor«)^r*'*^'*«-
a i-
« « i
i~ o
«o i
i
^s
a. oe i
K t- *e ui oooooooooooooo-oooooooooo
uiae »-i nm»>r>ir>ooooooo«>oo>oooooo»oo
t- ui • ui nronrorooioooooooooooooooO'OO
1-1 O X I ^-OOOOOOOOOOOOOOOOOOOOOOO
• e ui
»-«AOCI OOOOOO«OOOOOOOOOOO
o at v ui
O. U O. I
I
oioo'
» a:-io
x ui u »M>ir>ir»tr»
*•* t*** b^ I 4^ Q^ ff^ ^^ 9* 0^ (^ ^^ ^^ ^^ 0^ ^^ 4^ 4^ ^^ ^^ ^^ ^^ ^^ ^^ (f* tf^ t^ ^^ 4^
»- o •MC«NMr<4cri>r>>fororotor<>C4c^.<*'*-r<>r>»>'*«»w<*--r>»'*>»>«Nff>ff<«-ros»o)oo
i •-• o ui i «c«)«D«««rr>.roir>a>orav>a>o«K
o se — •- * «• to r> 10 «»e»a»aaoa^»>.««o»
•^UIUIIP>««'«<^«M«M «Mv««^«'««J»<^«a'M«ao<«v«<«««o«««0«M
•» ui a.
9 o. x: i (xr4NNi<>Mr<>ri>n«-io^r^-tr<«-«ynro'«-
i o o i <«M>ir>ir>r>4<«o»c>io»Obir>««Min(V(Mrwir>ooxC4^(v«A
u. ui ui «-fw«-«^>«OMMO« **«s — K-io*-fwcJWfoo»>^r
> o i (MCMMotnoinrora — CM eg r* to
V.'
I
I
ae
a i ~cwf>*r><*rv.eD»vo,-'»
O M«<^«4MM«M
-------
I
I-
Ul Z I
_i UJ OOOOOOOOOOOOOOOOOOOOOOOO
O *-• Z I OOOOOOOOOOOOOOOOOOOOOOOO
a ae x
a. ui i
i
>- >• i
v- t*
M O I
-J O
M ui i *-*-in''*-'«>*-''Vr<>'V>«-'«>r<'>'*-'«-«><«-<»<«>«'in«''«-*-<«-
OB t~
"
U I CO
« /% 3
C3 >- CO I til
•-" ItJ UJ 1 V® «*«***«-I«H«>4«>t*>4**«v« 1-«
Ul K Z C
I
I
to i
at ui
2IOUI I OOOOOOOOOOOOOOOOOOOOOOOO
CD »- 0£
I
ee
X I
C-32
-------
Ul Z I
_J Ul OOOOOOOOOOOOOOOOOOOOOOOO
A M Z I OOOOOOOOOOOOOOOOOOOOOOOO
z u. o nir>inooe-iftv>inir>oooir>v>oinininooooo
MOO. i (SicMoiroforocsicMcsioicticMctiottMCVCMfUMropofomr'')
:» U X
>• »- i
I- OC
•- e i
-i o
«-" Ul I
(0
o. at i
C »~ M Ul OOOOOOOOOOOOOOOOOOOOOOOO
ui x !•• i o o o o m o o o o o o ooooooooo IP o 10 o
Ul
i-
o *-
w I
o o o
I CO
I
o
(9 Ul I
M k-
Ul Ul I
w I
I
I
ui eo i
ui a.
a. z i
co ««x
t
o ui i
o o
Ul Ul
ro
C-33
-------
Ul
=>
z
z
o
G>
O
Ul
a.
Ul
i~
o
0.
z
1-4
1-
X
Ul
ac
o
0
Ul
1—
Ul
z
0
a.
Ul
s-
V.
o
Ul
o
oe
Ul
t—
Ul
ac
Ul
A.
ot
Ul
X
(0
o,
o>
Ul
Ul
o
Ul
Of
z»
z
1
1
1
1
1
1
1
1
\
t
I
1
1
1
1
t
1
1
'
1
1
1
1
1
1
1
1
1
1
1
1
1
SSSSSSSSSSESSSSSSSSSSSSS
......*......*...«.. —
fr~
C0
9
0
04
oooooooooooooooooooooooo
SS£?££«££«»««£«£::a2«£«£o
C-34
-------
Ul Z I
_) U)
O »« Z I
z u. o
MOO. i
a IK x
O. Ul I
I
>• >- I
i- at
M e i
_» (9
OOOOOOOOOOOOO-OOOO-OOOOOOO
OOOOOOOOOOO-OOOOOOOOOOOOO
ooo>oooor>oir>»ir>omir>oooir>oooo>o
rorot»>r«)Mror»jt»t>i«»-»-ca— -"cgtMcowrorororoM
(A
I
i
x%
a. at i
x: i- ^ u
ut x »- t
i_ui .u
»iOE I
• O kit
»- « a K i
O fit s^ Ul
o. ui a. i
I
oooooooooooooooooooooooo
v>tf>v>ir>v>v>oooooooO'OOO'Ooov>mv>4>
ror^ronrorooJoooooooooooocMrororooi
OOOOOOOOOOOOOOOOOOOOOOOO
........................
oooooooooooo
a.
jc
ui
O
O
•«• S9
i- tn i c»
ttl
_J
00
zuieo i
i-« UJ a.
3» o. x i
to «^
ooooo«mr>iooor>r>ir>ir>or>«r>oooO'iovo
(0 I
»OUI I
O I- <£
C-35
-------
I
I-
Ul Z I
_JUI OOOOOOOOOOOOOOOOOOOOOOOO
A •-• Z I OOOOOOOOOOOOOOOOOOOOOOOO
aoex
CL Ul I
I
>• >• I
»- «
•-> O I
ID K-
-------
X U. O
*•* O tt> I
<• Q£ 5C
a. ui i
\
OOOOOOOOOOOOOOOOOOOOOOOO
oooooooooooooooooooooooo
nooooonoomtnoonoinH^inmnoooo
-I 49
I- O
oooooooooooooooooooooooo
oooooooooooooooooooooooo
o oooooooooooooo
^^ ^^ ^^ 09 CB 09 ^*
^> ^^ tf* fl^ ^* flN ^N
to to C* f4 ft ft t*
I
^%
I »- W» I
: x ot
> U UI I
X >-i
UI
UI Ul
X K
0< N ^ O O
O A ~
z ui a. K
Of UI
a o ui i
o i- oc
-I O U I
U, UI UI
oooooooooooooooooooooooo
to
— MMM 10 -•
tO
oc
3
O
C-37
-------
I
I-
Ul Z I
-J Ul
O *-• Z I
z u. o
M O ft. I
3» 0£ X
o. ui i
ooooooo
oooo-ooo
oo
oo
o
o
oooo
oooo
oooooo
oooooo
>• >-
M Ul I
IB •-
to
>- M tu
x i-
ui • ui
o oe
a. o
I
I
I
»- I
Ul
B: i
ae i
ui
o. i
i
ooooooooooooooooooooooo
oooo^oooooo o
o
\*/
Ctf
I
o
Ul
_l
OD
z ui «A i
«-• ui a.
s a. c i
i
i
^v
(O I
at u
a* o ui i
o »- os
-toot
u. ui ui
> O I
I
to »
C-38
-------
o
u
(M
I
ID
_l UI OOOOOOOOOOOOOOOOOOOOOOOO
O •-> 3E I OOOOOOOOOOOOOO€>OOOOOOOOO
a» oc x
a. ui i
I
>• >• I
M O I
da »-
« « i
»- o
*s
at i
»- K UI OOOOOOOOOOOOOOOOOOOOOOOO
MUK I OOOOOOOOOOOOOOOOOOOOOOOO
>f^^| ••••••».••••••.•••••••••
' ae v ui
o a. i
*s I Of
UI
I 0
**• sc
t» »- CO I UI
K X K
v I <*
I
I
MUIO.
I
I
Cft I
Of UI
» O UI I OOOOOOOOOOOOOOOOOOOOOOOO
o >- ae
+*
i
ae
O M^«*^MMMM
-------
•-i o a. i
ai at x
a. ui i
o
•o
oo
oo
oo
oo
oo
oo
oo
ooo
ooo
ooo
ooo
ooo
ooo
a.
K
U
>• >• i
i- at
M O I
j ^j
•x Ul I
da t-
•
I
W
U)
ca
x
(9 t- (O I Ul
Z X OC t- **
M o ui i o. r«.
»C « I- Ut W
M Ul Ul I MX
XXX
w I V>
I
I
Ul
a. x i
«> ^
o o
o o
o o
o o
o o
o o
o o o
o o o
o o o
o o
o o o
o o o
to i
oc ui
a» o ui i
o *- ot
_j o o i
U. Ul Ul
oooooooooooooooooooooooo-
I
I
oc
9 I
O
X I
C-40
-------
oc x
a. ui i
i
>• >• i
I- dc
«-• O I
o
oooooooooooooooooooooooo
oooooooooooooooooooooooo
ooooooonooooooonoiooooooo
o ac
O. UI
at i
M UJ
»-i
• ui
C9K I
Ml
OdCI
w ui
O. I
I
OOOOOOOOOOOOOOOOOOOOOOOO
mir>ioooir>oooooooooooooor>ooo
wt«»MCMCMrocjooooooooooocg«ro• o i
>^
i
i
at
3 i
e
x i
OOOOOO-OOOOOOOOOOOOOOOOOO
r>' «M
-------
I
Ul Z I
_J Ul OOOOOOOOOOOOOOOOOOOOOOOO
O I-H Z I OOOOOOOOOOOOOOOOOOOOOOOO
9 CK x
O. Ul I
I
^ >• I
t- oe
<-> O I
_l 19
•-ui i *«r'«''r'»sr*- i o o o o oooooooooooooooooooo
I-UI -Ul OOOOOOOOOOOOOOOOOOOOOOOO
i-iUJC I OOOOOOOOOOOOOOOOOOOOOOOO
• O III
»- « Ck OC I OOOOOOOOOOOOOOOOOOOOOOOO
O QC v Ul
a. 19 cv. i
I
S X • I CO
z uio e»»ooo»wooct>ff>«-roro'*-rorofooie>iooooo
••• |nv ^^ I V«A ^^ Q\ (^ Q(^ CO CO CO CO CO CO C^ Q^ C^ C^ C^ ^^ Q^ C^ Q^ C^ ^^ C^ G^ C^
Z ^ ^ i tv
O Ul
U I CO
v *» x:
*3 t- CO I Ul
o X*-*. uiir>*»*'»w»^r<>w«yw«Mw*«*«-< — «^»- — -«-«o»
>-> Ul UJ I (O>H«<4«MV44^<«41Hv4««<-fl«4l-4«M
Ul XXX
CD «•»
f-
I
C> O -^ «sOC4*'«r'*«««**'*'<«1C4«MCil
zuico i o«* — -«<-'»-«««»-«-»»-«-»»-«-*o««'*»r»^\«'*o-«v»
MUIO.
CO v^
I
I
CO I
OC Ul
9OU1 I OOOOOOOOOOOOOOOOOOOO-OOOO
OH-ae
i
i
ac
o ,*.4.«M«*~iMMMM
-------
m
••* X I
u. o
O O. I
OC X
a. ui i
i
V >• I
oooooooooooooooooooooooo
OOOOOOOOOOOOO'OOOOOOOOOOO
ir>inv>inooooooooooooov>mnov>
t— ui • ui loroiorororooiO'OoooooooooNrororociiro
•MC9K I OOOOOOOOOOOOOO^OOOOOOOOO
• A til
K- « o ac i ooooooooooo
o at w» ui
a. o o. i
*N ^<>
o • *e i
ui a- oa»*.«M(Mir><*tt«»Qp>aD*eK.*-a>mror'>ir>aD4»
s» «:•!»
x in is Vk<+nr<>to«Actfr>>'<»*.e>0»<*i«>in'«-«><«>*F~«<*v>O9>0>
IMC K*UII*^titkttOdtf9CO 4d QD 40 ODOO 4D 9* Oft Cd
f- O (MOIICi»ICa
z via
O UI
o i at
w •"• K
O H- OT I UI
c« zzne »-ir><*v«r^a>ff»».o«c^M*«r»<*v*'A^»*MC4 — ooso1-
i Mouiio.r>»'Min9>mis.M<*-in.eo9>oo'OOO9>««r>>onN.«r««*nt9u>«owoxi>nminir>^>ce
»*UIUII«0«-l«'>««-<«*«'««* «M ^rwN'Osv«9>^4r«rv.9>s*r^r^i«>r^^>N>9>r<-r>>-t>-asor^
ac ui oooooir>ooinir>inioir>inoinv>inor><>ctfN.rk.Mio^9t9>ino9>v4r>><*
b. ui ui com — -»o«s»-'»r>ir>^M^'rr»>rr>if>r>iO'*'oi ^<<-«c*(Mc^- ion
vx
I
I
M
a t »-<^roi»iov»r>~CD»»O'->wro<'ir>ver^oo»>O'«cjrn'*
O MM^>M«4M^Mx^(M(MNC^C4
x i
C-43
-------
OB
<£
A.
X
Ul
t-
O
ft,
0
»-
3C
Ul
at
C9
ts
z
X
1-4
K
a
u.
Ul
»•*
u.
0
a.
_j
eo
*-
CO
•
a.
X
Ul
X
a.
ac
0
o
Ul
z
Ul
z
o
a.
Ul
«9
*-
at
Ul
»—
Ul
Ul
0.
CO
Ui
»-
Ul
x:
(O
Ul
Ul
Ul
ac
z
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
t
i
i
i
i
i
i
i
oooonmnmmiomminminiftmmoooin
PO PO ro fo ^i ^i ^i c^ CNI ^t c^ c^i ^i ^ii CJ M ^i ^i PO PO PO ^i
ae
Ul
OB
K
Ul
OOO^OOOOOOOOOOOOOOOOOO
to t*>
*4 Oi PO ^ ID %A fv 00 9* O **^ ^1 f ^ tft ^1 ^" OB fl^ O *•* C<4
CM
O
O
IS.
CM
SO
•«•
CO
SO
o
00
CM
to
If)
CM
O
o
v*
(M
SO
o
SO
CM
•*
C-44
-------
Ul
_J Ul
»i ac
u. o
O O.
- >•
M Ul
at »-
"• X
C9 K- CO I Ul
^^ Ul UJ I M* *** *"* ^"* *^ **' *^ *^ *"* ^^* *^^ "^ *** ^^ *^ ^^ *^ ^^ **"* ^^
XXX
w I »
I
I
I
I
CO I
fit Ul
9OU1 I OOOOOOOOOOOOOOOOOOOOOOOO-
Ob.M
vx
I
I
a:
C-45
-------
»—
UJ Z I
.J Ul OOOOOOOOOOOOOOOOOOOOOOOO
O »-< Z I OOOOOOOOOOOOOOOOOOOOOOOO
z u. o oooooooioor>mr>tr>oc>r>r>ir>r>oooir>in
1-4 o a. i rorororoi-ororotMcxi^v-i — ~• >• i
i— at
M O I
-i u
i— ui i m»*»*<*v*»«r>-»-r«>rgcr»>'»'«-'*'»-ir>r>r»'*'*
CD t-
«c . *±
O -Ml
ui a. o*>»i>otv.(MiNC>i-»-r<>owiou>fN.«D'j9Oo
3 Z • I 00
Z UIU W»O«sW^OJCOCOOOO-<»-N.^.-"(Mr«»^rOJfsjO«HO9rv.tv.^N-
•— >- id i •^»N.tv.r^fs.f>-CDepa>oo*'ff«o>ff»»>*>o»oooooooooo
»- o c4dic4*c<*roTir)<«<0*«uB<0aao<>imr<.rs.
-«*- uif>H»ror«>'»'«»- >«>mK.0s>Mroinv>irmir>invou)a>a>r^«*-«-rvis.\**.oo'»'*'*«>^O'*cMioiooO''*so-< — *<-<<*«•-< — »^»-io
MUIO.
a»CLZ i r»oiMi*>cr<>wr)r>ir>v*wir>'r'*-*'rw
«> vx
i
i
r*
to i
t* Ul
9 O Ul I OOOOOOOOOOOOOOOOOOOOOOOO
o »— at
_i u o i «*off*oi^- — ff»tv.r>oc*»ocr»**oo9o»'*-<«»>»*r^.oo»»
u. ui ui r>^«»*»^.r-*»r^««ri>ir> — cwc«iwr> »- •*• c* c* •» •* c«»
> o i «^«-^««<».c*ro
»^
i
i
at
3 i «*cvr«>'«-in<0r».aB9>o~4««r'>«'in<«r>-aD«r«'O-«f4ro
-------
Ut X I
»-i &; i
u. o
oex
• >- I
*~ Of
«-« O I
-I «9
M UI I
BO »-
« « I
I- U
OOOOOOOO^OOOOOOOOOOOOOOOO
oooooooooooooooooooooooo
o. at i
act— MUI
UtZ •- I
t- ui -ui
MUKI
• e» oi
i- * * «r
XUIU) I
>-> ui a.
i
i
^*
CO I
tie ui
O UI I
u, ui ui
OOOOOOOOOOOOOOOOOOOOO-9OO
C-47
-------
UJ Z I
_J U
a »* z i
x u. o
M o a. i
ai u x
a. ui i
i
>• >- i
*- «
M O I
U *»
M UI I
OB t»
«
X «-« «-
M UI UI
a a **
z ui to
>-• ui a.
•» a. K
t» vx
CO
Of UI
a o ui
o *- of
-i o u
u. ui ui
>• o
vx
I O • •
I
Of
I UI
I O
I
I
I
ooooooooooo-ooooooooooooo
C-48
-------
I
Ul Z I
-I UI
O •-> Z I
Z U. O
M o a. i
a oc x
a. tu i
i
>• >• i
i- at
•-« o i
at-
r^rv.N.
O ^(MCilN
w> I
ac
I Ul
^\ pn
»- CA » O
x oe t-*0*-cv
i oir>rx.«r»
»«c«i'«'ir>w»«oo««w-«0i»»»«>i«»««o— •
z ui to i
•_ ui (L
*> XX
I
(O I
a o ui i
O t- «
—j o ^s i
U. Ul Ul
>• O I
\s
\
I
ac
a i
o
Z I
oooooooooo
<0v>r^a»v«ni^'i«>a>^r
^r «> r> m n t* n ^•
t»> ro
oooooooooooooo
C-49
-------
oooooooooooooooooooooooo
oooooooooooooooooooooooo
rotorororotocMCMCMcsicMCMCMCMCMCMCMCMrororotororo
ir>
n
«r *•<+*» in m
10 10
Ul Z I
-4 Ul
O *x Z I
z u. o
M O 0. t
» e* x
a. ut i
i
>• >• i
R- M
M O I
-J (9
IM Ul I
n *-
« r«.u>ir>9>rwir)rw^i-v»rk.cBCAro«0s0r«.rwcMwBM)-ro
=> z-io
z bu c? co r^ lo 10 ro ^f *^ o ro NO co o csi ro ^ ^ ^* C4 r^ ro ro ro &t o ^^
•- V-UllCr0»CDCOCOCOCOCOCh9.C^9>9>94(MCMCMCMVMCMCMr>ICMNr4C- CO I O
i >-r<-cM<»»v>orM««oo«tCDN.M><«*ecB«tO'*«r>«>»»*O(«*Wi«N.twlVO»rw
•-•UlUlli1***^*^***"* «*
ui z z x «r>
_l vx t
IB
a I
t-
I
AO^^ toN-oocMooto>ANoroto^«^r^rr>-r-^rrorocM^«iCM9>
MUJO.
CO ^>
I
I
co i
at ui
9 O Ul I OOOOOOOOOOOOOOOOOOOOOOOO
O I- DC
> O I **«*CMCM~4CMCMCMCMCMCMCMCMCMCMCMCMCM«MCMCMCM
I
t
at
eft i **• CM ro ^ 10 MX t^ CD c^ o *^ CM ro ^ 10 ^x r^« co cv* ^^ ^^ CM ro ^
O —«^,«,<-.^.-« — »-CXCM«MCMCM
Z I
C-50
-------
UI Z I
O f* Z I
z u. o
M o a. i
:» oc x
a. ui i
oooooooooooooooooooooooo
oooooooooooooooooooooooo
Irt Irt lf> Ift IO Irt Irt IO IO IO IO IO IO IO tf) IO IO V) Ift O O O 4^ 4^
^1 ^1 ^4 ^4 INI (M ^4 IM tSI dl ^J (Nl Ol CM CM, CM CM CM CM to fi f) fO fO
•» * « r> «• io «o
>- >• i
I- CtC
M O I
J til
MUII •^•••'r^'r^r***'******'
ca H-
« Z-IO
M i-uii9>cococo«ocorwivN.coa>eocococococo(OcocoiN.r«.K.i\.N.
o at
Cl I UI
^ ^^ db
U I— Cft I O
CM zactx >-ir>O'«-«N'*-eDMrwtM^—i\ooinir>r>ir)'*ve<»otM<«>ir)
UI X X K CM
J ** t -•
CO
« I
»-
o a ^ r».w^'*fvir>ONN.os'*'*'c<<'
MUIO.
a»o.x i vB«ioiDN.invBir>rs.<«v»co<«so««v»vBir>v>rof>>r<>n>r<>
CO v^
I
I
CO I
Of UI
O UI I
•- ae
O O I
UI UI
> a i
oooooooooooooooooooooooo
ro m CM CM m
r>-«eNo-«*-
V4MMCMCMCMCMCM
C-51
-------
Ul X I
-I Ul
«a -« ae i
x u. o
M O O. I
s» oe x
O. Ul I
oooooooooooooooooooooooo
ooo-ooooo-oooooooooooooooo
a.
K >-
>• >- I
I- OC
M O I
-I 19
M Ul I
9ft »-
•-• O K I
• O 0)
fr- *'****'r'rw'*'«»''r**''*sor-tr>iO'«-r*>-«oco»0s
I Q^COOOdOCDCOODCOOOtt>C^C^(1'^O^^9><^C^d^O^O^O^dOQ940
I
oe
1 Ul
A
I O
i—o»«'«-r^ — •*ts.oro«*or^-*».*>ffx*«t^roo^c\ioB'*
i cjv«»s — Mv*a>or<>r>r«.o«v«va>««jor)*ca — ^>^
OK»rr;^r'r*^-mioir)«r>«*»«jvO'ji>u>v6vo«AU5«j>sou>ir>r>
m
i •—
i
i
A o ~
z ui to i
•-> ui a.
» a. x i
0) V*
I
*FN.*-'*'rk-N.*'e\AOrs.?iiooc4^'r
io<«ioir)«*«*r>'*vOMOvjov*v«r>iN.««'«-r<>r<5'*-'*'rinr)
Of ui
9OUI I OOOOOOOOOOOOOOOOOOOOOOOO
Ol-tt
> O I
I
at
» I —
o
N N N C4 M
C-52
-------
O
o
CM
I
CD
o
a i-
« Ul • Ul
« 0 X
o ae v ui
a. u o.
Ul O
I- Ul
o
0 t- CO
x x ae
••* O Ul
JC »- 1-
_ ui Ul
XXX
x ui m
a a. x
to w
CO
ae ui
a» o ui
-100
U. Ul Ul
>> A
ae
a
x
i
t
1 Nco«ooo
1 O
i *-«
(0
1 '-I
1
1
1
1
1 OOOOOOOOOOOOOOOOOOOOOOOO
1 ^> 4^ ^* ^* C^l ^^ ^9 ^^ 4^ ^^ ^^ \0 00 Irt *** ^t *^ flD 00 4^ ^^ *^ 00 fl&
%A ^* ^* {^ f) Orf 4^ ^9* ^f ^* ^^ ^^ ^> ^* ^A ^^ ^^ ^ft Irt Irt ^h ^O ^O Oft
1
1
t vt ^i to ^^ vy ^& ^> eft (^ 4^ *^ ^i PO ^^ 10 ^*v ^^ 00* o^ 4^ **4 ^i ^0 ^^*
1
C-53
-------
ui z
_i ui
a •-• z
z u. o
•" o a.
:» DC X
oooooooooooooooooooooooo
oooooooooooooooooooooo-oo
>• >-
»- oc
M O
_1 O
•— UI
CD t~
<: ui
o
u
CM
I
o
UI
_l
ea
M U K
O 01
« A oe
K v^> UI
L3 O.
UJ U
t- UI
C9 t- CO
z x ae
<-• C9 UI
X «— •-
Z UI CO
«-> ui a.
a a. x
tn ^
OOOOOOOOOOO'OOOOOOOOOOOOO
oooooooooooooo«>o«oinooooo
tWCJOICWO-OOOOOOOOOOOOWMtNCM'OOlOU
oooooooooooooooooooooooo
ooooooooooooo «•
til
00
— oo
oe ui
9 o ui
o *- w
—1 O O
U. Ui UI
•> o
I OOOOOOOOOOOOOOOOOOOOOOOO
Of
a
o
CM
C-54
-------
I
I-
Ut X I
.J Ul OOOOOOOOOOOOOOOOOOOOOOOO
O>->X I OOOOOOOOOOOOOOOOOOOOOOOO
•-> O O, I CMCMWroroCMWCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCM
» at x
a. ui i
>- >• i
*- Ot
_J O
«•« ui i »«»»r>»r»»ir»^r«'**'^r»<«'*»**r»<»'»'«r*>'r
co »-
« « i
i- o
CO I
O. Ot I
KH-XUI OOOOOOOOOOOOOOOOOOOOOOOO
MIX I- I OOOOOOOOOOOOOOOOOOOOOOOO
H- Ul . Ul OO Ul
O- O O. I
I
^s ^%
A -Ml
3 *-IO
x ui o ceoooc»>c*scf>e*ioror>r-»^r»^-^'*-'*-'rr«>MMroroo
H- A •»O\0CMCMCMCMCMCMCMCMCMCMCO
o xi-«t- omr>m«r>*-«-^r -.-.CMr*»w*'«-^-»r*M-*«r'**ts.
M Ul Ul I
Ul K X K •«•
U N/ I CM
CD
& * N.M>focM<«>ri>-'wrwo<«--'«>^rv>ocMor^.r<>ncMO
MUIO.
a» a. x i v»tr>^r«-ir>M>inw»rwior>-<*w»ir>ir>vois.fs.r>M»-«-^r'«-rw
i
CO I
or ui
»OUI I OOOOOOOOOOOOOOOOOOOOOOOO
o •— ac
u. ui ui •«ocf«CMCM>MCMNr<)«rir>in<->oir>««
i
i
O MMM^^MMM^MCMCMCMCMCM
C-55
-------
Ul
C» »-• X I
ac u- o
M o a. i
:» cc x
a. ui i
i
>- >- i
oooooooooooooooooooooooo
oooooooooooooooooooooooo
ooooooommtotninmu-jminminmininrjinm
M ui i «r»w»ir>tr>if>>r'«»-'»<»>*'»i*-'»*'«'<»>*'*<»N«MttOINOilOOOOOOOOOO'OOOOOOO
MV9X: I OOOO'OOOOOOOOOOOO'O'OO-OOOOO
• o ui •
H-OCCkOtl OOOOOOOOOOOOO^OOO
o a: ^ ui
A. «J Ou I
I
» «:-io
z uiu ao9«oo9<9>9s9«.«o^v>9<9>9>^eors-^r<-aar«-iS'
M t-uii9«^N.N.v»i«w»vor>-rv.r^r^rs.rwfs.^r^N-fs.r^rv.i«.f^N.N.
Z *-> I
o ot
O ( Ul
^x ^** CD
ta t— m i o
(si x x a: t-^ior^.^^r?r>io*s«*«-.rs.r^i^i^r^K.f^o
M Ul Ul I -«
ui K x c r-
_l V, I W
CO
.eNrw».i«.c.w'r
xuito i ir»if»ir>oir»oir>-«-«-«'««-«^»-«'«M»-<«-»*C(i-««4-«-«
» ui a.
•o \x
i
• .
«K Ul
atoui i oo-oooo^ooooooooooooooooo
o i- oe
s^
I
I
OC
O — ^^^^^.« — -.^«Mc*f^cWfX
X I
C-56
-------
Ul
O <-i X
X U. O
9 Ot X
a. ui i
oooooooooooooooooooooooo
oooooooooooooooooooooooo
i- oc
M O I
•*« Ul I
OB »-
*'rr>»«if>«*
*»
CO
o.
X t-
Ul X
v: ui
t-
• ui
M O X
• o
I- ^>totr>r)«r»r>oooooooooo«oioo«r>o
wiowrowwwwoiooodoooooiwoiwoiwoi
oooooooooooooooooooooooo
o«oooooo
t^ ^ ^9
Ul
en
»>
ooo — ooNo^^rjoo^o—!*>!«>«•-<
r^r^r^r<.rwso^r~r>-r^rwa»a»
(v. r» ro.
01 01 01
t»- N. N.
01 o* 01
Ol
I
o
ca K-
i
ot
I Ul
CD
I O
I O
O
I
oir'>r>ir>rk
«ir>N.(r>ooooooooooo
oa
mv>oor>mioom»«oi
— oo«o*«o-«
01
W T
(O I
Ot Ul
a» o ui i
o »- ae
o u o i
oooooooooooooooooooo
o o o
•s ir»
-------
w
i
^
Ul
_J
oa
.
o
X
u.
r>
o
Ul
oc
U.
1—
X
*«
o
•9
BEFORE DUST CONTROLS
z
0
(A
(A «
« >
1
Ul s
X
H"
<
O i
X t
•-• e
ac «
23
A
1
>• 3
t- •
i
aa
Ul
»-
« i
t?
i
i
i
0 00
s. 1
• W
jt •
m
u
u
u <•«
L. •
o if>
i
^«
» •
C PO
•4
»
o
.
n
i
^
.
r+
n
.
«^
i
^^
•
o
c
o ~
•* ac
F- O
u »-
ktl O
K U.
M (A
a w
i
i
i
i
i
i
i
i
i
>
i
t
i
i
i
i
i
i
i
i
i
i
i
t
i
i
i
i
i
i
Is.
O
O
O
o
•o
•0
0
o
0
«••
•o
0
^
^
1^
0
z
is. r. o
•o o o
O 0 0
O O 0
0 O O
o o o
o •* o
O fs. O
•o o o
•o o o
0 0
w^ O O
o o
o o o
o o o
o o o
•o o o
o o o
o o o
Ul Ul
Z Ul Z
Z Z Ul
oa x o
<•• c* o
W ID O
o o o
O O 0
0 O 0
000
fv is. O
000
o o o
0
** w o
o
o o o
o o o
o o o
o o o
o o o
o o o
Ul
(A 111
Ul Ul «0
N.
o
0
o
o
0
o
0
o
«i4
o
0
o
o
o
o
o
Ul
CA
10
•«• «o rs. eo *-nois.
is. <- «r *• is. cr\ o •«•
^^foww ^Hsaof*>
oooo oooo
0
oooo oooo
oooo oooo
oooo oooo
rs. rs. ^- rs. N-c^oo
oooo oooo
oooo oooo
o o
o«»o»^ ow^oro
o o o o
oooo oooo
oooo oooo
oooo oooo
o o o o o o o o
oooo oooo
oooo oooo
99 99
CA 9 «A Z 3» Z
(A
-------
o v> «r
CO T •>-
CM O
V) 09 •*• CD
CA
_l
O
(X. W>
I
fO ^> |4J
O •-• H-
»
UI O
_l U. Z
OD o M
« «
ui >-
oc h-
o •-•
O CD
U.
O
U. U.
=>
at v>
UI UI
at M
U. (K
O
^- o
Z UI
(v.*> FN» * K. O*
i— &* i i
U •* CO — 00 O * * 0
1 IOOOO OOOO
a • 1 . • • • ....
r f» o o o o
-« i
i
O •»O»|S. T»OO
n <•» o w< w »-« »* o o
. i .... ....
— o o o
1
,
n oooo oooo
•IOOOO OOOO
-» oooo oooo
1 IOOOO OOOO
o oooo oooo
. 1 .... ....
o oooo oooo
1
C 1
3 *v
^ O UI UI UI UI
j i- i x ui x co ui to
tlfc» Z Z Z UI UIUIWW
1C U. 1
- tf>
a V 1
0v»«^as 9iteiv.>e
«* oi o r> <0eo«->o»
OON.V fw rs. O O
o o o o
* — o>* ^r^-ON.
Ct^ ^^ ^0 ^w ^^ ON ^^ ^^
oooo oooo
0
O*-<
o r> o « to o ^ ir>
o o o
o » o o oooo
o r>- o o oooo
o«*oo oooo
oooo oooo
oooo oooo
o o o oooo
a a» a* »
1 O
N
i n
o
i
i
i
0
1 00
N.
1 ft!
0
1
1
1
^
1 N-
«*
1 O
o
1
1
1
1 «
1 fr-
o
1 »-
1
C-59
-------
z z
O 0
u >-
Ul CO
Of =9
•i A
O
I Ul
O CK
»! O
i-> u.
a» ui
•cc z
o
0. U>
CO -
I
ro A lu
i z x
O «-« h-
a»
Ul V9
_l U. Z
ca o —
a c¥
»- IX) =7
O A
an
lu >•
*- •">
M CO
o
<3 CO
Ul Ul
OK «
U. OK
O
K- O
Z Ul
M I-
O «E
-9 CJ
«J
«
h-
0
09
•
O
«*
/^
09
.
O
v*
1
to
•
09
(S,
.
oa
i
CM
ID
V*
.
n
i
••4
to
o
to
1
\*
^4
in
.
W4
1
O
.
o
at
o
i*
u
u.
CO
w
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
t
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
ts.
to
fv-
«*
O
o
CM
<*
O
V4
O
•H
CM
m
o
o
«4
O
0
0
o
^
o
0
z
ON
sO
CO
O
o
o
N.
v«
O
o
1*4
CM
r>
o
o
M
0
0
o
o
o
o
Ul
z
z
CO CO
•r •*•
to to
o o
o o
o o
o o
o o
0 0
o o
o o
o o
o o
o o
0 0
o o
«* — •
^ ^
N. N.
V* «•«
0 0
o o
Ul
Ul Z
Z Ul
OOSCMOS os — <«• to oo^ros
OM0VOCO OO •«• O <-> O>N-«O
ooomto to o v* CM to^oo
0 0 0 0 0000 00 0
o
OOOO 0 0 O O O 0 O
ON.*»OS OS •*• co ON ONO.M>
o -< to so M>toto<« <* «-> ro
OOOO OO-.0 OOO
o
o o r-
or>to>« to «* <* ro so o »•
o o
o *-< vo r> ~«otoo OO-O twOOO OOO
o o — o ^ o o o ooo
OOOO OOOO OOO
OOOO OOOO OOO
o o o o o o o o o
ui ui at at at
co ui to coatco za»
uiuitoto co to to a» ata»x
IS.
in
CM
CM
to
CM
in
o
(M
in
0
OS
SO
<4-
M
0
O
O
o
0
0
at
z
X
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
•
i
i
i
i
i
1
i
i
i
1
i
i
i
i
i
ON
m
o
CO
.
IS.
^
l*>
O
o
•
to
4f
o
•Mt
o
•
o
•
•<*•
in
ON
CM
O
SO
ON
*o
o
0
•
_l
^
»-
o
^
C-60
-------
w
I
Ul
_J
CO
•
z
u.
=»
fgf
ttl
Ctf
u.
1—
z
••4
o
•->
_J
0
Ul
QC
O
u.
Ul
oa
*
z
o
CO
,
o
«•«
XN
CD
.
0
1
•
CO
Ul *•» •
W> (A CO
V. 1
E CM
Ul N'
x m
»—
A
O Ul
X Ul —
•-• 0. •
M mm
a i
0 •
>- Z l*>
»— «-l
••a 31
- i
M
CO
<£
^»
*A
A
o
(4.
CA
Ul
M
Of
O
o
.
n
i
M>
M
n
.
«^
i
o
•
o
**
Ul
t- z
« O *N
U M CK
H- 0
u t-
Ul U
ot u.
O w
CO CM CM o o in
| O M M O
o o o o
1 ....
1
1
1 ....
1
1
o •*• o o
1
1
i •«• •* o rv.
fj p£ ^> ^"*
1 . . . .
o
1
1
r*. IN. N. o
I * * * O
0 O 0 O
o
1
1
-» -« r> o
o
i
i
O 0 *• fw
1 O 0 IN. <«•
O O — M
1 O O O «>
O O 0 O
1 ....
O O
1
1
1
Ul Ul
1 Z Ul Z
Z Z Z Ui
1
1
co NO NO M tf>
•^ *^ «J> •*• f> f«> ON
<«• CO M CO •«• * -*
!**• Uft CXI MO PN. 0N CV
— «« w
o
O O O 0 0 0 O
o o o o o o o
o o 0 o o o
•o o o o o o o
Ul Ul »
CA Ul (A CA 3»
Ul Ul W> iO iA (A (A
IN.
CO
vfi
^»
o
in
CM
CO
vfi
CO
CO
u>
CD
o
0
o
o
0
o
at
CA
at
«*
in
9>
CM
0
^^
•s
^
^f
n
0
o
0
o
o
0
o
o
0
•o
at
f>
\ff
m
«M
o
CO
CO
^>
fN.
«
0
o
o
9
o
o
3k
z
a»
m
o
vA
CM
O
n
CM
v«
CO
CO
^
IN.
•^
<*•
rs.
m
o
o
o
o
o
o
31
Z
CM
NO
in
••-«
o
o
0
IN.
l«>
O
CM
m
IN.
^
ro
tN.
•r
»
o
o
o
o
o
o
at
z
z
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
1
i
t
i
i
i
i
i
i
i
i
CO
•
**
CM
m
o
o
•
vO
m
in
in
0
•
n
CM
^
»•
•
•<
IN.
IN.
*&
O
•
CM
*£
CO
<«•
O
•
«*
CM
r>
0
o
•
_J
~
C-61
-------
(O
-1
o
M
t-
X Z
o o
>-> CJ
I—
o »-
Ul (0
ft! =>
M e,
o
I Ul
Ck OL
y. o
^4 ^^
9 Ul
do
O
X -
«E Z
O
O 10
Ul «
U> ui
O. *0
(0 »
I
ro A ui
i z x
ui
Ctt O M
OC •-•
r>
O
u. 10
o
Ul U.
Ul Ul
at M
u. a:
o
I- O
X Ul
>- »-
o «
•
PS
0
S.
E
*J
JU
X.
%
r
"4
X
o
N4
h-
u
Ul
•c
h*
_J
«
t-
o
t-
00
•
o
«H
yx
CD
O
v*
1
r*>
.
CO
(Si
.
09
1
(SI
in
••«
VI
.
K>
O
f*>
•
.
«*
m
.
««
i
o
•
o
«•"
oe
o
^
u
Ui
S
i
i
i
i
t
i
i
i
i
i
i
i
i
i
i
i
i
1
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
t
,
,
(SI
CM
O
0
o
o
i>»
•V
«4
0
0
o
o
0
ps»
V*
o
o
o
0
o
o
*
•I
CM
o
o
o
o
o
^F
•4
•»
0
o
o
o
0
0
o
^
•o
^
Ul
3"
z
0
o
o
o
o
0
o
o
o
o
o
o
o
o
o
0
o
o
^
o
0
Ul
•x.
o
0
o
•o
o
0
o
o
o
0
0
o
0
^>
o
o
o
o
0
o
o
Ul
z
Ul
o r> •* to
O ON rs- vs
o o- o &
o
0 O O 0
0 O O 0
0 0 O O
o
o o o —
O 0 0 0
o o o o
o o o
Ul Ul
(0 ui m
Ul Ul W> (0
* t>- c » ^ •* (M csi w ^
oooo> o- o o o
O 0 O O 0
^r OJ tiO 10 rw^v^
«^o«ot4 ro ^ ^ w
o<*r^o o^N-t^-
0000 0000
o o o
0 00 0 O
o o o o ooo 9 99
(0 9 (0 Z 9 Z
«A(0«09 99XX
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
,
1
OD
*
(SI
o
«•*
•
o
o
o
o
o
•
o
•0
00
o
o
•
9>
o
(SI
\ft
o
ft
o
(S»
o
•
IS-
•••
(SI
0
.
oo
^
f>
o
o
•
^
-------
o o
ui
i ui
A Of
X O
00
o
T. -
«C X
Ul «C
<&> Ul
O. *f>
V) -
I
o ui
x x
ui
o
o
Ul
OO O «-t
o o
X
Ul >•
Of t-
O •—
f.J DO
O -
U. \f>
o
u ac
o
u
Ul
o O
CO O « — 0
• 1 ....
in o
•
1
• i ve CM it. ON
in co in — «*
• i ....
n
i
i
f> fO ^* O ^^
• 1 ....
-« O 0
1
1
n o » ••• o
• i o *. r*. o
— 0 ~ ~ 0
1 1 O 0 O 0
o o o o o
• 1 ....
00 0
1
1
at i
o ui ui
H- 1 Z Ul Z
0 X X X Ul
Ul 1
•0
v 1
*•* CO ^^ O^ ^f ^0 W^ QD f^ O^ IO fO *^*
•••N-OOCO oo CM oo >.o ^reo« focN****^ to to ro ro CM
i
i
to m to to oooo otonin <*•
. . . . . . . . . . . . |
OOOO O
1
1
oo^r>- a* «
(A Ul (O COatCO X9ZIH-
ui ui to (« «oo>w>ai aoakxz o
i *-
i
C-63
-------
o t-
t- o
o u
« (O
O 9
I O
o
x ac
ar
4K
O (A
cx>
I
Ul O
-i u. as
OD O M
t^ (&
9E
IU 9-
O£ H>
Ox: »•
3 -i
o >^
«j oa
u ot
ac o
Ul U.
=>
o w>
Ul Ul
»- o
Z Ul
«J
0
00
.
0
w«
^\
00
0
94
1
f>
.
09
• e*
«• .
ft 09
S. 1
C
u
iJ --
a. •
» in
i
v«
C fO
M4
O
.
f>
1
uO
.
*+
n
.
**
i
o
o
X
D ^>
M DC
- O
u t»
DC Ul
ix m
O NS
IOOOO OOOO OOOO
, °. . . . °. .
1
,
1 .... .... ....
1
IOOOO OOOO OOOO
1 .... .... ....
1
1
1 .... .... ....
1
1
1 .... .... ....
1
1
1 .... .... ....
1
1
OOOO OOOO OOOO
IOOOO OOOO OOOO
OOOO OOOO OOOO
IOOOO OOOO OOOO
OOOO OOOO OOOO
OOOO OOOO OOOO
1
ui ui ui ui a» »
i zuiae muicn o»a»
-------
u
_J
o
Ul
Of h-
•"• CO
A 9
1 A
O
X X
A
TK *
to co
CO - VI
1 K 01
O tu v
z z n
•-« t—
O Ul
U. Z UJ •*
O M 0. •
oc *o r>
Ul => 1
U O i-
X O •
u >• Z f>
Of 1- «
at -. a»
9 —I
O »•<
« ) OB
O
1 ....
1
1
o o o o
1 O O O O
O O 0 0
1 0 0 O O
o o o o
O 0 O O
1
1
1
Ul Ul
1 Z Ul Z
Z Z Z Ul
1
1
•* f» r»> * O 9O ^O ^^ ^O CO W C^ ^*1 ^^ ^V 1
-< o -* -« OOIN-W* ro<»oo
. 1
0 O O O
1
1
OOOO OOOO OOOO
OOOO OOOO OOOOI
OOOO OOOO OOOO
OOOO OOOO OOOOI
OOOO OOOO OOOO
1
OOOO OOOO OOOO
1
1
1
ui ui » a» » at
couico co a» co zaizi
uiuiw>u> to «o u> ai aia»zz
1
1
o
M
ITy
o
o
o
o
o
o
'•
o
o
^
o
0
o
o
o
•
0
(W
^4
(M
Ol
o
.
fO
C?i
»•
o
•
o
o
o
4>
o
.
o
_J
«
^_
o
t-
C-65
-------
O I-
>- X
t- o
o o
Ul
or t-
ix CO
O 9
I O
« at
>- to =>
•s.
CM
n
co r>
i
to
10
i
r>
o
o
•- oe
t- o
o »-
Ul O
ac u.
>•• (O
o ~
1
1
'
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
I
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
CM
«
OS
O
o
o
^
o
•*
Os
to
o
to
v«
•0
ft
OS
to
o
o
o
0
o
o
X
O O -rt
os »o co
to (M Is.
o o o
o o o
o o o
o o to
O 0 —
0 0
o o <-<
CM O fO
0
000
«•« CM CM
o o o
o o o
o o o
0 O 0
o o o
o o o
Ul Ul
X Ul X
X X Ul
o o — « -« " CM CM ro 10 CM o o o
*•> CM «* M n os os
SO IT> as 0s MMtO0s 0s v* TO IO
oooo oooo oooo
u» so ro to o os rs. to os ro so ro
CM CM ~« CM o tO «• «* to — CM *•
o
^1 CSi IO IO Irt SO 0s to PO CM *^ ^^
OOOO — 0 O « OOOO
— 0 CM -• SO •» CM r> ^ CM 0 —
oooo oooo oooo
oooo oooo oooo
oooo oooo oooo
oooo oooo oooo
oooo oooo oooo
oooo oooo oooo
ui ui » 3» at »
BQ Ul CO CO 9 CO X A X
uiuicoco wcocoa* » » x x
1
1
1
1
1
1
1
1
1
1
1
1
1
•
,
1
1
1
1
1
1
1
1
1
,
1
1
1
1
1
1
1
1
1
1
1
1
1
1
IS.
M
»0
•«•
V4
•
0
o
0
o
0
•
o
o
o
o
.
o
tr»
AO
0
»
.
es
to
SB
*g
o
-
to
*»•
-------
Z Of
O H-
•1 Z
»- o
Ul
at »-
I-* co
O 9
I O
Ul Ul
a. v>
m -
i
V A 1U
I Z Z
J U. Z
03 O »
I- Ul =D
C.) Ok
O «
U. VA
O
o at
z o
Ul U.
r>
o to
ui ui
ac •-.
U. Otf
o
I- O
-> O
_J
«
^w
o
fc-
00
.
O
«
0>
.
o
1
o>
CM
•* .
a co
S. 1
E CM
in
*i
jj •«
X. •
fl V)
1
a •
•4
10
i
SO
v*
V)
•
- QC
- O
J t-
tt o
K U-
•« CO
a w
CM IO f> M^
1 O «O »• «
to r> CM ON
1 M « CM •»
O O O O
1 ....
1
1
1 ....
,
1 ....
1
1
-• CM (M •«
1 « *• l>- •»
SO O •* IO
1 O — •* O
'
,
« 0 — M
1 ^ fO 0& ^>
1 ....
1
1
CM IO CM IO
1 ....
1
1
O O O O
1 O O O O
o o o o
i O O O O
0 0 0 O
o o o o
1
1
Ul Ul
1 Z Ul Z
Z Z Z Ul
1
1
^^ fO M^ IO m 0^ *•* BO n> sO ^0* CM
|s*«^€>«>« O^Of^*^ •>4|>O«^ *0
«• CM SO CM 0%CMrs.O CM O CM m
CMCMCMCM CMmsOm CM •*- C« «
O O O 0 OOOO OOOO
•M CM •< •» rosofvm <\i M- fo »«
CMOin^ ON >• (M CO !»>-.««—•
m to so ON so i^ M)» po ^o* CM m ON
OMOO - CO ^v CO *M \o ON ^ 09 m ON ON
so o r^ in to^cMso OONCMCM
0 0
OOOO OOOO OOOO
OOOO OOOO OOOO
^y ^> o o 4^ ^^ ^^ o o o ^^ ^^
OOOO OOOO OOOO
OOOO OOOO OOOO
OOOO OOOO OOOO
Ul Ul a 9 99
co ui co matco z»z
Ul UJ O) sO sft 4A (O ^k ^k 3i Z 3E
^
1 ••»
sO
1 sO
^
1
1
1
o
1 0
o
1 O
1
0
1
*•*
1 00
1 O
1
'
1
CM
1 0
is.
i in
CM
1 .
1
1
IO
1 fO
ON
1 10
1
1 •••
<•«
1 SO
o
1 •
1
1
0
I O
o
1 O
o
o
1
1
-------
*r
1
o
Ul
_J
CD
«
t—
Z
o
1-4
•—
0
Ul
ae
•"
i
o
3T
-^
O
3^
.
^j
z
Ul
rrs
o
Ul
ae
u.
0
•»
T CONTROLS
• 3
*•» *
«-« :
_i
i-t
00
>.
:
u
u
k.
0
a
c
•*
k
E
9
M*
K
LJ
1*1
K
M«
O
«••
.
\fl
|
^
-< -4 O O
Vfi -< S0 <^
10 — 00
O O O O
1
1
1
1
10000
1 ....
1
,
1 O TO FO O
0 -* — 0
1 ....
o o
1
1
-« •* 0 0
1 OPk O) «0 O
ro ^-* 01 ^^
1 ....
o
1
1
1 M* *O *O fO
Ol OJ Ol -•
1 ....
1
1
o o o o
1 0 O 0 O
o o o o
1 O O O O
0 O 0 O
1 ....
0000
1
1
1
Ul Ul
1 Z Ul Z
Z Z Z Ul
1
1
*+ o ro oj ^ oi fo ro oio*^** ^
ooojooi* o oj r> ro orooiooioh
r»inotn «ooeN« ro « r> ts. o
ooca<« OJMM— ^oooioo
oooo o o o o oooo —
1
1
1
1
. . . . . . . . . . . . 1
'
1
OOtOO OliAOION OOOOI*1*
oooio moimro oooo 01
o o o oooo
i
i
OON>« OJOJOJ— OO»--< *
<4>*« 04tQ«oo* ivrooroio
m 01 ^w ON ^. 01 01 ro ** *^ ^^ *** m
. . . . . . . . . . . . t
i
OOOO OOOO OOOOIO
oooo oooo oooo o
OOOO OOOO OOOOIO
oooo oooo oooo o
. . . . . . . . . . . . 1
oooo oooo oooo o
1
1
1 _J
ui ui ak ak ak ak «
(A Ul (A CftakCO ,Z3»ZIH-
uiuiwto u» u> w> ak akakzz o
i i—
i
C-68
-------
U)
_J
o
X Ot
O I-
••• X
*- o
o o
Ul
ac >-
M (0
ft S>
X
Ui «
O< UJ
a. to
«o •
i
«• A ui
I X X
U *"• H-
a>
Ul (0
_) U. X
aa o "-
at
o
o
u ae
or o»
ui ui
u. ee
4C 1 (*t CO f^» 00
*- co r- -• r-
, . . . .
i
i
i ....
i
CD 1
i
,
.y • 1 ....
1
a
«i i
-* i
i
O f«> «-t N. 0>
ro oo N. « r>-
• i • • • •
i
m oooo
• 1 OOOO
— oooo
1 1 O O O O
o oooo
o oooo
1
E 1
- UL \
J t- 1 X Ul X
it 0 X X X Ul
K U. 1
a ^ i
•s oo •»••••« \o to ON o> <•• r> «• 00 i IP
. . . . . . . . . . . . i
i
i
. . . . . . . . . . . . i
i
i
i
i
. i
i
i
. . . . . . , . . . . . i
i
i
** ^* rtl ** fO ^* ^O ^* Ol ••* *•* *•* IO
^^ CO ^ *^ M) IO 4^ 0) <^ IO ^ 0) 1 Ift
. . . . . . . . . . . . |
1
1
OOOO OOOO OOOO O
OOOO OOOO OOOOIO
OOOO OOOO OOOO 0
OOOO OOOO OOOOIO
oooo oooo oooo o
. . . . . . . . . . . . 1
oooo oooo oooo o
1
1
touito «n 2» on x » x i h-
ui tu m 9 :• » x x o
1 »-
1
C-69
-------
APPENDIX D
PARTICULATE EMISSION RATES USED IN THE ISCLT CALCULATIONS
Tables D-l through D-40 list, for each combination of wind-speed
and Pasquill stability categories, the average particulate emission rates
for the point and non-point combustion and process sources during the
sample days before and after the initiation of fugitive dust controls at
the Armco Mill. Similarly, Tables D-41 through D-43 list, for each combin-
ation of wind-speed and stability categories, the average emission rate
scaling factors for the roadway, parking lot and parking lot access road
sources. The average emission rate scaling factors were multiplied by the
average emission rates (see Tables 2-13, 2-15 and 2-16 in Section 2.1) to
obtain the emission rates for use in the ISCLT concentration calculations.
D-l
-------
09
O
O O O O O O
O O O O O O
******
U) UI UI ^J Ul If1
O O O v* O O
o o o r^ o o
O O O O O O
O O O O O O
******
UI UJ UJ UJ UI UI
O O O O O O
O O O O O O
O O O O O O
00
o
«M
I
o o o o o •*
O O O O O- O
******
|«i |«l laf • •! 1«< l*f
o o oo ey s0 o
o o f> ff> r» tf>
o o ro •* to --
o o o o <* o
o o o o o o
******
ui m uj ui uj ui
o o o o o o
• n ui
o oc
UI O
to U. O «•* O O O- »•*
"X. •» UI OOOO-OO
K • IS *!***!
^ ID UI UI UI UJ UI IU
I &) d^ *^ ^ c*i &*> ^h
>•*< >- *. v» * o *« N.
UI
w
V
UI
<*> o
UI
o
•
O.
K
UI
X
ae
o
u.
o o o o o o •—
o o o o o o
****** at
O ^ O *& O* ^
o o ^> ^> o o
UI UI UI UI UI UI
O O O O O O
<»««.<'<»<>«•
O O O O O O
o o o <» o o
«» o o o o o
******
o o
o o
o o
-I O
M U
Cb UI
-------
09
O
o o o o o o
O O- O O O O
o o r» N. o o
O O »» N O O
o o r> to o o
o o o o o o
o o o o o o
** + * + »
bl bl bl '•' bl bl
o o o o o o
000000
o o o o o o
w
o
o o o o o o
o o o o o o
*•**•* + *
bl bl bl bl tij bl
O O -« 09 «•««•«
o o N. m M r>
o o v> H> n «•
o o o o o o
o o o o o o
+ •*• * •**•«•
bl bl bl bJ bl til
000*00
o «» o w o o
000*00
n ui
o cc
bl O
CO U.
N. «•« bl
o
o
o
O
o
o o
o o
tal
«
*•
n
>» r> r>
o
o
bl
»-
«C
bl bl
ui
jo
— V.
-. O
o o
o o
bl bl bl
« O O
r>
Ul
n to
n
r>
a
^
N.
OS
bl
CO
U)
«
«A n o
o o o o o o
o o o o o o
•*••*• + + + <«•
bl bl bl bl bl bl
O f) OS ~4 O O
o *« * os ro o
o « « ID « o
o o o o o o
o o o o o o
bl bl bl bl bl bl
o -« >» r»- 10 «M
r>
r> «« r>
o o o o o o
o o o o o o
* * * 4- + *
bl U bl bl bl bl
o o o o o o
o o o o o o
o o o o o o
an u
« t-
CD U O bl U.
-------
00
o
O O O »•• O O
O O O O O O
* * * I * *
I.I llf l.f HI ^1 III
O ^ O F4 O O
O O O >M O O
O O O (SI O O
o o o o o o
o o o o o o
o o o o o o
o o o o o o
O O O O 9 O
o
*4
I
o o «-i «* •-« o
o «• o o o «•
+ * I I 1 *
111 Ul Ul Ul Ul Ul
o o is. CM o r-
Ul
cx> o
— QC
x: a
Ul O
«A
Ul
o
•
w> o» o «•» o
•-• ro rw o r» o
t«> 04 V* «>* »< O
o
»•* O O O O O O
_) • I * I * I
Ow iji i«f 1*1 i«i lii iif
X IT> N* O ON O 09
« r> w o » o •*•
(A — 01 O -* O —
m •» o
06
Ul
CA
9-
O O O
* I I
O O
O
o
O» CO O
•* l«w «
w r> rv.
* >
ui b
o *
o o
I I
oo r>
<» o o •••
o •*•'*•»
o o
m
o>
r»
Ul
X
»- o o o o
o «• o o
oe * * * *
u. o- o o o
o o o o
o o o o
Ul
X
o o
o •»
**
o o
o «•
o o o o
o o o o
****
ui ui ui ui
o o o o
o o o o
o o o •»
_l O
— o
a» ui
Ul U.
-------
co
o
V*
/s
O O •* O O O
o o o o o o
+ * I * * +
[ti tai m 1^1 m 1^1
o o o 0* o o
o o « «-• o o
o o «* — o o
o o o- o o o
o o o o o o
o o o o o
o o o o o
o o o o o
CO
o
9*
t
00
O O -« O « O
o o o o o o
* * I * I *
Ul Ul UJ llJ W Ul
o o o » * r«i
o o «* o *»• <*;
o o ON — r> «M
o o o o o o
o o o o o o
•«•*•++ + +
Ul Ui UJ UU Ul Ul
o o o r> o o
O O O (M O O
O O O CM O O
o
u.
M
CO
ID
o o
* I
Ul Ul
O «r«
o o
O «*
o o
I I
Ul Ul
•> CO OB
o o o
o o eo a> ID o
• « >- *• «
o ~* ~« -^
o o
I I
Ul Ul
vA m
r> eo
r«. co
co o
e
(A
o n o
Ul U
t- x
•
O
o in o <« IA o
O «* -
o o o o >* ro r>
* + * * 4- *
^ Hj ^ |»i iff |ti
o o o o o o
o o o o o o
o o o o o o
K. >•
M ac
-I O
1-4 O
OB Ul
« Ok O Ck Ul U.
-------
CO
o o o o o o
o o o o o o
+ + *•***
l«l \ftl t£j Itl ^«l 111
o o co r> o o
•o o m w -o «•
O O 0> v* O O
o o o o o o
o o o o o o
+ *•* + + •«•
•^i ijj ui ui Ul Ul
o o o o o o
o o o o o o
o o o c> o o
09
o
I
n
•o o o o o o
o o o o- o o
+ + + + + +
Ul Ul Ul Ul 111 Ul
o o r«> ve CD —•
•» O * O> CN -«
o o v* ro r* ol
o
O
o o
O O
tij uj ui m m i&i
o o o o o o
o o
o o o o o o
o o
«0
I
>. A n ui
at ui o
at «o u.
ui v to
I « >• -H >-
A on *• oe • «
M on o
Ul Ul
o •-•
o
+
Ul
«•
o
ui -»
— o
o
*
Ul
««•
!>•-
10
Ul
_» O
CD —
a (A
v- to
ui
o
Ul
ui
Ul (J
*- Z
->
Ul <*
Ul •
o « o •-<
o o o o
A A J. A
UJ Ul Ul Ul
Ul Ul
til
— •» z
00 «
«« «M tn »^ — «
co »< ON *+ **
CD
•*
to
e.
JC
«
to
o o o o o o
+ + + •*•••••«•
Ul Ul Ul UA Ul Ul
o •» o o o o
o o
o o •» o o c>
o o o o o o
o o o o o o
+ * ••• * * •«•
Ul Ul Ul Ul Ul Ul
o o o o o o
o o o o o o
o o «• o o o
o o o o o o
O O O O O O
* * •«• •*• * *
UJ Ul Ul Ul Ul Ul
o o o o o o
O •?* O O O
A
X
O -H *•• O -• -*
ui
rn
-c»
+
ui
f>
o
O)
ir>
ui
o
o <> <> o o o
o o o o o o
•»••«••«• + •»••«•
Ul Ul Ul Ul Ul Ul
o o o o o o
o o o o o o
o o o o o o
V- >•
•-« ot
tt Ul
« >-
»- ui u.
D-6
-------
OOMOOO OOOOOO
CO OOOOOO OOOOOO
4-4.4-4.4-4- 4- 4- 4 4- 4- 4-
O bl bl bl bl bl bl bl bl bl bl bl bl
— oomooo oooooo
A O O <•• O O O OOOOOO
o o — o o o oooooo
» 00 — 000 O O O — O O
oooooo oooooo
O *• 4- 4- •*• * 4- * + + * + *
^* bl bl bl bl bl bl bl bl bl UJ bl Ud
I O O — ff» O O O O O IO O O
»9 O O CM «•« O O O O O «» O O
o o -- •* o o oooooo
«o ...... ......
>•
_l O — O O O O >- OOMOOO
(M 9 OOOOOO -I OOOOOO
OC • -9 4- + + 4- 4- 4- S> <*> + + + + +
O 10 blblblblblbl *^ blblbfblblbl
b. i f- o-*>vtoir>o oor»rw««o
M (^o>^r«rcv «*. o o •* * f>- o
«> • O — CD •*• — O »- O
>• ^ww ...... ......
« O OC M
oe ui o bi
O£ U> b. O — O O O O »- O 4-
& bl ^ID blblblblblbl bl-bl bl bl bl bj
t- i to v» o o w o m •—•-« — o
i • <*• o o •*• o o >• o «• -« 10 o o
o oto oe • <•• o o -i -• O O —
« «/>bl - OOOOOO
H- COO O • _l 4.4.4.^4.4. ^ J. i * a. * i
*->ae to a. blblblblblbl o. blblblblblbl
K3 o ' K r. •> « «> o o K omN.n<0«i
bio bi «* o ii «/>«•» c*
co bi- a> o» — a> •*• o o u> o -« — cw — ITJ
bl tt. — ...... ......
ae o H- o — — o o o •- oooooo
bi x oooooo oooooo
> MIO OC 4- 4- 4- 4- 4- 4- AC 4- 4- 4> 4- 4- 4-
4T 9* o blblblblblbl O i^j >«i t.i •»! 1.1 IPI
-< u. or>ioN.ir>«- b. oooooo
i o«m
-------
o o o «- o o
o o o o o o
******
ui ui ui ui ui ui
o o o »» o o
o o o fo o o
o o o *• o o
o o o o o o
o o o o o o
******
Ul Ul Ul Ul UJ UJ
o o o o o o
o o o o o o
o o o o o o
09
O
•M
I
VK
e»
I
w
in
O
o
*
1^1
o
o
O
O » ro o
•« •< CO «VJ
o
o
*
o
*
O>
er>
-• ~« o
o o o
***
ill ^«l i«f
N. ¥> OS
o »> ^« »- •«
Ul
(O
o at o ot
— o ro
ui x w o
_l O U
eo •-< >—
tfi
NO
ro
vft *+
00
o o
**
UI UI
i^- o
N. O
CM O
-9
is.
U.
«
ro <>
o o »- « r>- o
O O O O O O
******
Ul Ul Ul Ul UJ Ul
o a. oo •* o o
o v«> co cxj in o
o CM CM ~« -« o
o
o
UI
co
ro ~« ~«
o
o
*
UI
•*
os
r>
o o
o o
UI UI
o •»•
o
o
O O
Ul
Ul Ul Ul Ul UJ Ul
o «a ro ro oo *•
<> CM n ro in CM
o to c«i ••<««-
oe
ui
>-> in
O 0* •"« *» »•««•<
o o •» o -o o
******
UI UI UI UI UI UI
o
-------
00
o
o o o o o o
o o o o o o
* * * + + •*
I.I ^1 ^«> l«l 111 yl
o o ff\ ro o o
o o »o te o o
o o r> r> o o
o o o o o o
o o o o o o
•*•<«••*••«••*••*•
Ul *** *•* ^| >*i m
o o o o o o
o o o o o o
o- o o o o o
CO
o
I
10
o o o o o o
o o o o o o
+• + + + + +
Ul Uf U 111 Ul Ul
o o n ro rw »p — f>-
o o w w « ro
<* O O O O O
€> O O O O O
* * *• + •*• +
U) Ul Ul Ul Ul Ul
o o o ~< o o
000*00
000*00
at
o
,JI v> ir> 10
>. O O O O O O
_l O O O O O O
a * + * * + +
-9 Ul Ul Ul Ul Ul Ul
« O (V f>> 0
r^ - -<
U.
Ul
CO
V)
>•
r>
H- O O O O O O
U. O O O O O O
«c •*••*••*><*>•*••
Ul Ul Ul Ul Ul UJ
to o cfc in * in o
>• O O> v* <* O* O
a o * » •* * o
— o ro
UJ Z SO Ul
JO Ul
co «— t->
« (A Ul « O
t- to u o •
o
o
Ul
K S3
Ul O
co
ae
ui
Ul 00
a. M
u»
«c
to
Ul
X
r> r>
o o
o o
Ul Ul
M •«
N m
m in
o o o o o o
o o o o o o
CL ui ui ui ui ui ui
ac o M «•
r> «••**
Ul
o
o
o o
o o
*•»•
O N. N. 00 0» »>
O M M ff» * «x
o * * r> ir> w
>- o o o o o o
o o o o o o
CK •»•*•* + + +
e ui ui ui ui ui ui
U. O O O O O O
o o o o o o
O ^- o O O O
I— >•
M Ot
-I O
M U
CD Ul
<£ O> O
Ul U.
Ul U.
D-9
-------
CO
o
O O « — O O
o o o o o o
•*••*•! !•*•<*•
ui uj tjj ui *nj m
o o o r- o o
o o -* w o o
o o •*• -« o o
o o o o o o
o o o o o o
* + + **• +
t«t tjj UJ UI UJ U)
•o o o o o o
o o o o o o
o o o o o o
o o •*«*•• o
o o o o o o
•»• + I I I *
UI UI UJ UJ UJ UI
o o tn w o o
o o o o ro o
o o w ^ — o
O O O O O O
o o o o o o
*••*•«.*•*• +
UJ UI UI UI UI UI
O O O tf> O O
o o o ^- o <>
o o o r*> o o
at
O CO
U, I
(K
oe
• «
UN K •
M O to
ac so u
O UI
v r>
i
ft.
CM
UI
oe
u,
UI
(O
co ui
w w
•- ot
x a
ut o
«/>
UI
4
o o o o o o
•*• I I I I I
UI UI UI UJ UI UJ
O f*» ff* O O f*>
o o o o o o
I I I t I I
UI UI UI UJ UI UI
r«. M eo oo o ON
oo * sp r*» 10 »>
«^a w w -« CM ro
M O
o o o
• lilt'
UI UI UI UI UI UI
* 0s 10
(A
UI
10 •* -<
>» O O O O O n ro o
o o •» o
o o o o o o
o o o o o o
+ + + ** +
UI UI UI UI U UI
O 1^. •» 9* *O O
o m o art m o
o r> f>- « r> o
u
5K 0000
•x O O O O
a.
K
a
(O
UJ UI U UI UI
o rx. »> «N -*
o. « •«•
oc
ut
K- o -< »— o-
oo o e>
-------
CO
o
O O
o o
O «• o O
o o o o
o r< oo o o
o *. (^ o o
o «> ca o o
o o o o o o
o o o o o o
•*••*> + + + +
ui ui *i* ui *j** uj
o o o o o o
o o o o o o
o o o o o o
I
n
O O •-* -*<••-«
O 0> N m
o o n n •* f<
Ill lAl W (Jhl W lil
o o o o o o
o o o o
ee
o
u.
V)
O UJ
V4 >-»
I «C
A O£ W
v«
ui at M>
-1 O
o> «-•
*- to o
UI
ut
O O O 9 O O
_I O O O O O O
+ + + + + a> + + + 4> + 4.
«•• « 00 ** «- O O O O O O
CVi flfr -
at
o
o
« o
o •
ee
ui
V4V4»«V4N>«0 CO OOOOOO
cr> v* 10 o o o >- o o o o o o
^fon^fo^ »(M c
co w^^fro^'W co
. (3 ijj i^ uj i^ ijj m
•4 U. OOOfOtOO
4^ Cv 41^ ^^ €t ^v
o o o o ^ ^
UI UJ UJ UJ UJ UI
O > ^> ^^ ^>
o o o o o o
•-• oe
-i e
«-• o
oa o e» ui u.
ui u.
CA U
D-ll
-------
CO
O
O O -< -• O O
O O O O O O
o o ro oo o o
o o r>
o o »- •* o o
o o o o o o
o o o o o o
+ + + + + •+
UI UI UI UI UI *•'
o o o o o o
o o o o «• o
o o o -o o o
CD
o
I
n
O O «•• ••« -4 .*
o o o o o o
+ ** + * +
UI UI UI UI UI UI
o o o n
O ^ l*> M W M
O O ••« •-« »•• «-•
O O O »-•
o o o o o o
O ^ O W ^ *
o o o -« o o
>•
«
CK
UI X
_l O
co
co
ut
o
UI O
UI
u
9- « >-
OC • 4
(O UI
X
a •-
M m at
a» • o
— u.
o
o o o o o o
+ ** + + *
••I ^J t»i t.t ta^ i>l
O 9> l>> OB tX 9<
o tv
O *< *• •< *< ~*
o o o o o o
4. 4. 4. 4. 4. 4.
UI UI UI UI UI UI
O O O 9> ff»
n ro ro re r« N
o o o o o
* * 4- * +
1. 1 1.1 t*f iji i.i
GO
01
o *^ *•« «•* -*^ *^
o •» o o o o
+ * + * * +
UI UI UI UI UI UI
O * V «• •* 1^.
o rw 9v O
O C4 (ta CM «M O
o •«•>*-••••« o
O O O O O O
* A A + A A
UI UI UI UI UI UI
o o o o o o
o o o o o o
4- * * * * +
UI UI UI UI UI UI
o o « o o o
o o o o o <*
o o o o o o
t- >•
K- Of
_l O
M O
at ui
•a CD o o ui u.
«E at o
-------
o o -« ~» o o
o o o o o o
o o o o o o
000000
o o
o o
O O
*• in o o
r>> (si o o
M CM O O
o o o o
o o o o
o o o o
o
o
o
0>
o
o o »•• ~* ~* «*<
o o o o o o
+ + + + + +
Itl m lil 1*1 lil lil
O O 0> <>« N. «•
O O «\i f*> «SI (SI
O O CSl W (SJ N
o o o -• o o
o o o o o o
+ + + ** +
IA! 111 III 111 toJ lil
o o o eo o o
til til lil 1*1 lil lil
o tt «? n to «&
^^ A* ^S> ^^ rt ft*
O CSI 01 W W Ol
>• o o •« —• ~« o
_l O O O O O O
s> *** + * +
-> Ul Ul Ul Ul Ul Ul
O O N. » O O
CSl O O (SI tsl CSJ O
(sj ui
x t-
I w o
u. o o o o o o
« ••••*••»• * 4- *•
Ul Ul Ul Ul Ul Ul
tO O «0 M) 09 *. O
>- O (SI (S| fSi CSl O
Ul •
Ul O. —i
^9 (A
a.
3C
«>
Ul
o
*
Ul
o o
* •*•
Ul Ul
\a IP
fta W
en (si
u
-. -. « X
o o o »•«
Ul Ul
r> *•
ra
O O O O O O
•«• * •«• * * *
Ul Ul Ul Ul Ul Ul
o i« rv. a> K. ««
-------
o
«4
/s
o o o o o o
o o o o o o
*•»••»••»••»• +
UJ '•* UJ UJ UJ UJ
OOfMO^OO
o o o to o o
o « <* in o o
o o o o o o
o o o o o o
+ + + + + +
ui ui ui ui '•' uj
o o o o o o
o o- o o o o
o o o o o o
09
o
•«
t
o o o o o o
o o o o o o
•*• + + * * *
ui *'* ui ui ui uj
O O 00 OS 0N 00
O O «* <* *• —«
o o to « * n
o o o o o ^^ *tf ** 00 1^ W
CNf ^^ ^^ 4^ ^^ 0^ *"
>• o o o o o o
_l O O O O O O
=> + + **• + •«•
—> UI UI UJ UJ UJ UI
o o n o «H o
i^» n> ^> ft in o o
W O O is. IW •*>•*•<*•
UI UI UI UI UI UJ
CO N. O ID *P * O
1^* ^f '^ ^^ fj ^^ 0^
U
to
UI
O O O O O O
O O O O O O
d. * A 4> •». *
UI UJ UI UI UI UI
ro 10 NA «• vo « « «>>
n « r» ^ m m
o o o o o o
o o o o o o
•*• + + + + *
1^ UJ UI UJ UJ UJ
O «SJ (S| 0> K. M
O o
o r>> i<> t^. r. o
z o o o o o o
«-« o o o o o o
O
o
O
IM
Ot 09
o o o o
bJ but ijJ fcftJ W bJ
o o o o o o
o o o o o o
o o o o o o
O
a
UI
UJ U.
-------
00
o
o o o o o o
o o o o o o
******
bl bl bl bl bl bl
o o r» p> o o
O O •< r- •<
o o 10 w n «
o o o o o o
o o o o o o
******
bl bl bl bl bl bl
O O O «* O O
•o o o -* o o
o o o <« o o
at
oe
I
a
U
_l
m
v ID
AC • «
o w A
o o o o o
o o o o o
*****
bl bl bl bl bl
o oo -• r> o
o o o o o o
o o o o o o
******
bl bl bl bl bl bl
O O « »• l«» O
«v « v> «A fw o
O O W» t». 4» O
o
o
*
bl
o o
o o
**
bl bl
w es
o «i
o
*
bl
v»
w
r>
o
o
*
bl
oc
bl
i- o
O
bJ
O
Of
o o o
Ul «>>
bl •
a. •<
bl
»•»
r>
M m ac
a» • o
(A
w
x
t- o
*
o
* 4» A A
bl bl bl bl
M m
vo r>
o
o
bl
«
CO
W» O
o o> 9> ra n o
O <* « is. K. O
o o o o o o
o o o o o o
4- A * * * *
^»l tjl Ipl lil »«t ^1
O * » « M 10
o r« a) w» o »>
O U» -
to u
D-15
-------
CO
o
«*
x\
000000
o o o o o o
******
l»l I.I llf •!> ^J !«•
o o f> r> o o
o o N «r> o o
o o n *• o o
o o o o o o
o o o o o o
* * * * * *
•jl ^J QJ ^| Iff ^|
o o o o o o
o o o o o o
o o o o o o
o>
o
o o o o o
o o o o o
*****
o o
o o
o o o
o o o
i o o o o o
o o r> •*•<»••* o o o <*• o o
co
^
-J O O O O O O >- O O O O O O
(V a o o o o o o _i o o o o o o
oc • -» ****** sa ******
C^ CO UI UI UI UI UI UI *T> UI UI UI UI UI UI
u. i rw o <« oo «•• u» «• o o to to m o
$0 • o v) ID irt ITJ to fti o o 10 *A irt o
>. **» *> ui
u. o o o o o o t- o o o o o o
•OE V.M UI OOOOOO U. O O O O O O
K- o> ****** «e ******
10 ui ^^ m ui ui ui ui ui ui ui ui ui ui ui ui
•M i— i CA oo r- -« •«• r*> m w> or-nr^cwo
i ^c ^* ^^ ^* 09 oo o ^»» o3 co )~ o m ^0 ON «* o
o Ofo of • n m v» o
w ew A o
-J O UI 49 U
00 M »- X OOOOOO X OOOOOO
«C <0UI r> m in in
UI O. «
U «A UI UI
« X X
at A t- oooooo »- oooooo
UI X OOOOOO OOOOOO
>• Mm oc ****** ae ******
tf 3k* O UIUIUIUIUIUI O UI UI UI UI UI UI
«* u, o»t-^---«'.ACs; u. oooooo
I O(M>X*40NO OOOOOO
o o r> r> •*•««• m oooooo
OC
O
U
UI
ui u.
D-16
-------
o o o o o o
•o o o o o o
******
kl Ul Ul Ul 111 Ul
O O -* v« O O
o o N. r> o o
O O CM CM O O
o o o o o o
o o o o o o
******
•>• ^.1 1*1 ^ <«t 1*1
o o o o o o
o o o o o o
o o o o o o
O O O A O O
o o o o o o
******
UJ iff U! UJ Ul UJ
o o «r> to »« «*
<» o •>>. r> •*• r>
O O M W «M M
o o o o o <»
o o o o o o
******
IU Ul Ul Ul Ul Ul
o o o *. o o
o o o m o o
O O O CM O O
*
Ul
O
O
O
o
*
CM CM
iw
• -«
at o fit •
» o w
x «« o
o ui
M I-
co ui « o
CO U « •
« as M
K 3 O I
Ul O Ul »O
CO Ul •
ui a, ~*
o u>
. « r>
o
o
*
Ul
ro
m
(M
CD
CO
o
X
a. ui ui ui
CM
o
o
*
Ul
«
r>
(M
o
*
-
or
19
co
Ul
ae
o
u.
CM CM CM (M CM
m
(M
•O O O 9 O O
O O O O O «
******
UJ Ul Ul Ul Ul Ul
O O «•« CO CM O
o o » ^ W »» N. O
o tn «s o *4 o
O CM CM tO M O
o o o o o o
o o o o o o
4. A * ««. A *
Ul UJ Ul Ul Ul Ul
O «• 00 IS. «• M
<» rv. *. rs. w 9>
O CM CM CM CM CM
o o o o o o
o o o o o o
******
^.1 1*^ ^^ ^»l ^«t ^
o o o o o o
o o o o o o
o o o o o o
M at
_l O
•-> ta
-------
O O T~ ~« O O
o o o o o o
+ + I I * *
111 l^| l«l ]fl L|J ial
o o o o o o
o o cv «F» o o
O O SA IO O O
o o o o o o
o o o o o o
******
UJ UJ UJ UJ UJ Ul
o o o o o o
o o o o o o
o o o o o o
* *
o o o o
I I I I
o o o «••
o o o o
o o
o o
O O O >« Ml O
o o w •*» r> *»
o o <« 10 in «
UJ UJ lAl UJ 111 lAl
0001000
O O O *> O O
o o o r> o o
9-
_| O «« «< v« «« «4 ». |