United States Office of Air Quality EPA-450 4-81 002
Environmental Protection Planning and Standards January 1981
Agency Research Triangle Park NC 27711
Air
c/EPA An Evaluation Study for
the Industrial Source
Complex (ISC)
Dispersion Model
-------
f 3?- 222
EPA-450/4-81-002
An Evaluation Study for the
Industrial Source Complex (ISC)
Dispersion Model
by
J.F. Bowers and A.J. Anderson
H.E. Cramer Company, Inc.
University of Utah Research Park
Post Office Box 8049
Salt Lake City, Utah 84108
Contract No. 68-02-3323
EPA Project Officer: Sharon R. Kraft
Work Assignment No. 5
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air Quality Planning and Standards
Source Receptor Analysis Branch
Research Triangle Park, North Carolina 27711
January 1981
U.S. Environmental Protection Agency
-------
This document is issued by the Environmental Protection Agency to
report technical data of interest to a limited number of readers.
Copies are available free of charge to Federal employees, current
contractors and grantees, and nonprofit organizations - in limited
quantities - from the Library Services Office (MD 35), U.S.
Environmental Protection Agency, Research Triangle Park, NC 27711;
or, for a fee, from the National Technical Information Service,
5285 Port Royal Road, Springfield, VA 22161.
This report was furnished to the Environmental Protection Agency
by H.E. Cramer Company, Inc., University of Utevh Research Park,
P.O. Box 8049, Salt Lake City, Utah 84108, in fulfillment of Contract
No. 68-02-3323. The contents ot this report are reproduced herein
as received from H.E. Cramer Company, Inc. The opinions, findings,
and conclusions expressed are those of the author and not necessarily
those of the Environmental Protection Agency.
Publication No. KI'A 450/4-81 002
U,S. Environmental Protection Agency
ii
-------
ACKNOWLEDGEMENTS
We are indebted to the staff of the Source Receptor Analysis
Branch, U. S. Environmental Protection Agency, Research Triangle Park,
North Carolina for their assistance and technical guidance in carrying
out the. evaluation study for the Industrial Source Complex (ISC) Dis-
persion Model described in thJs report. Mr. George Schewe served as
the EPA Project Officer for Work Assignment No. 5 of KPA Contract No.
68-02-3323 until May 1980 when he accepted a position with private industry.
Following Mr. Schewe's departure from government service, Mr. Joseph
Tikvart acted as EPA Project Officer for Work Assignment No. 5. Mr.
Alan Huber assisted us in obtaining and evaluating some of the data
sets used to test the ISC Model. Mr. Schewe, Mr. Tikvart, Mr. Huber
and Mr. James Dicke all provided helpful technical guidance and suggestions
during the study. We also wish to thank Mrs. Sharon Kraft, the current
Project Officer for EPA Contract No. 68-02-3323, for her assistance in
coordinating the review and publication of this report. Additionally,
we are indebted to Dr. Warren B. Johnson of SRI International for
providing a copy of the SKI report on the tracer study at the- Millstone
Nuclear Power Station and to Dr. R. A. Caska and Mr. J. M. Brown of
the DOW Chemical USA Michigan Division for supplying data I'rom the DOW
plant in Midland, Michigan.
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. Craig Cheney and
Mr. William Hargraves for their assitance in performing the computer
calculations. All technical illustrations in this report were prepared
by Mr. Kay Memmott. The report was typed by Ms. Sarah Barlow, Ms. Cherin
Christensen, Ms. Lori Siedenstrang and Ms. Bonnie Swanson.
11
-------
EXECUTIVE SUMMARY
INTRODUCTION
This report summarizes the results of an evaluation study for
the new Industrial Source Complex (ISC) Dispersion Model (EPA, 1979).
The ISC Model, which is designed to assess the air quality impact of
emissions from large industrial source complexes, contains several fea-
*
tures which arc not contained in any of the dispersion models recommended
in the current Guideline on Air Quality Models (EPA, 1978). The most
important of these new features arc the gravitational settling/dry depo-
sition option and the building wake effects option. (The ISC Model's
distance-dependent plume rise option is considered in this report to be
an integral part of the building wake effects option.) Because many of
the basic components of the ISC Model have been previously tested by EPA
and others, the focus of the ISC Model evaluation study was on testing
the key features which distinguish the model from the other generally
available dispersion models. Three sets of published deposition experi-
ments, three sets of published diffusion experiments and the source,
meteorological and air quality data for a large industrial source complex
wen.- used to Devaluate the performance of UIOBC key ISC Model features.
ll is Important to note that the available data .sets were not sufficient
in number or detail to validate the new features of the' ISC Model in a
strict statistical sense. However, it was possible to compare the per-
formance of the ISC Model with that of current models. For example,
the ISC Model short-term program 1SCST in its default mode corresponds
to the Single Source (CRSTER) Model for a single source and to the MPTER
Model for multiple sources. The results of the ISC Model evaluation
study are briefly summarized in the following paragraphs. All of the
concentration or deposition calculations described in this report were
made using the ISCST program.
iv
-------
RATIONALE FOR MODEL TESTING
Insofar as possible, consistent procedures were used throughout
the ISC Model evaluation study. However, the model tests made using the
various data sets may at times appear to be inconsistent because the data
sets themselves are inconsistent in the information available. Addition-
ally, some of the calculations that were performed were in part determined
by the available source-specific information such as visual observations
of plume behavior. because the observed concentration or deposition values
were reported in a variety of normalized, crosswind-tntegrated and/or
non-dimensional forms, we selected the following parameters as the primary
measures of model performance: (1) the mean ratio (MR) of concurrent
calculated to observed concentration or deposition values, (2) the root
mean square error (RMSE) as defined by Draxler (1980), and (3) the per-
centage error bands. The MR is defined as
N
MR
where X . Ls the calculated concentration or deposition for the i trial
and X . is the corresponding observed concentration or deposition. Be-
cause air quality observations usually span more than one order of magni-
tude, Draxler (1980) uses a geometric form of the RMSE:
RMSE = exp
N
X
ci
(II)
The percentage error bands, as used In this study, are simply the percent
of the trials with calculated concentration or deposition values within
a factor of 2 of the corresponding observed values.
-------
lii (lit .-ibseiu-o of ;my systematic errors in a dispersion model,
the model. Inputs, or Che observed concentration or deposition values, the
MR should tend toward unity for a large sample size. Although inspec-
tion of Equation (II) shows that the RMSE also equals unity for a perfect
model, uncertainties on a trial-by-trial basis in the model inputs as
well as in the concentration or deposition measurements may significantly
affect the RMSE. That is, the RMSE for a given data set is in part
determined by the accuracy of the model and is in part determined by the
quality of the model inputs and the concentration or deposition measure-
ments. The percentage error bands are also significantly affected by
the quality of the model inputs .and the concentration or deposition
measurements. However, the RMSE and the percentage error bands are con-
venient indicators of relative model performance for a given data set.
THE GRAVITATIONAL SETTLING/DRY. DEPOSITION OPTION
The three sets of deposition experiments used to test the gravi-
tational settling/dry deposition option of the ISC Model were the continu-
ous, elevated releases of glass microspheres with diameters of 50 to 200
micrometers described by Walker (1965) and Stewart (1968) and the line-
source aerial spray releases with droplet diameters ranging from 15 to
180 micrometers described by Boyle, £t al. (1975). The premise of these
tests was that, if the ISC Model can match observed deposition patterns,
mass continuity requires that the model also match ambient particulate
concentrations with gravitational settling and dry deposition occurring.
The results of the ISC Model calculations for the Walker and Stewart
data sets provide an indication of the model's accuracy when used to
calculate ambient particulate concentration or dry deposition patterns
attributable to stack or fugitive emissions comprised of particulates
witli gravitational settling velocities ranging from about 0.1 to 1.5
meters per second. Similarly, the results of the ISC Model calculations
for the Boyle, et a_^. data set provide an indication of the model's
accuracy when applied to particulate emissions with gravitational set-
tling velocities ranging from about 0.01 to 0.5 meters per second.
vi
-------
Additionally, the results of the ISC Model calculations for the Boyle,
et^ ad^. trials test the model's representation of a line source by mul-
tiple volume sources.
Table I summarizes the comparisons of concurrent calculated
and observed maximum deposition values for the Walker (1965), Stewart
(1968) and Boyle, £t al. (1975) data sets. For the Walker and Stewart
data sets, calculated and observed crosswind-integrated deposition
(CWID) values are compared in Table I in order to remove from the depos-
ition values the effects of differences in the duration of emissions
for the various trials. Because the maximum deposition for the Boyle,
ct_ a-U trials occurred near the center of the line source where edge
effects are unimportant, crosswind integration of the calculated and
observed deposition values was not necessary. The maximum observed
deposition for all of the Boyle, ot a1. trials and the maximum observed
CWID values for some of the Walker trials occurred at downwind dis-
tances of less than 100 meters. The ISC Model does not make deposi-
tion calculations at distances less than 100 meters. Consequently,
Table I compares calculated and observed deposition values at 100
meters for the Boyle, et al. trials and calculated and observed
CWTD values at 100 meters for the Walker trials in which the observed
maximum CWN) values occurred at distances- of Jess than 100 meters.
SLi-wnrt >;1vrs only the maximum CWID values, and Table I considers
cnly the Stcwnrt trials with maximum observed (JWM) vn I IMVS at or beyond
100 meters. As Indicated by the MR values In Table I, the TSC Model,
on the average, predicts the maximum deposition to within 30 percent
for all three data sets. Also, the RMSE values are less than 2.0
for all three data sets. The concurrent calculated and observed max-
imum CWTD values are within a factor of 2 for all of the Stewart trials
and for all but two of the Walker trials. Similarly, the concurrent
calculated and observed deposition values at 100 meters for four of
the five Bovle. et al. trials are within a factor of 2.
The good performance of the ISC Model when used to calculate
maximum deposition values for the Walker, Stewart and Boyle, et_ al. data
vii
-------
TABLE I
SUMMARY OF THE RESULTS OF THE ISC MODEL CALCULATIONS
FOR THE WALKER (1965), STEWART (1968) AND BOYLE,
ET AL. (1975) DEPOSITION EXPERIMENTS
Data
Set
Walker (1965)
Stewart (1968)
Boyle, et al. (1975)
No. of
Trials
12
6
5
MR of Maximum
Deposition
Values*
0.71
1.12
0.91
RMSE of Maximum
Deposition
Values*
1.57
1.51
1.89
% Within a
Factor of 2
83
100
80
*Crosswind-integrated deposition (CWID) for the Walker (1965) and Stewart
(1968) trials and maximum deposition at 100 meters for the Boyle, et al.
(1975) trials.
vlil
-------
sets is in part explained by the relatively large gravitational settling
velocities of the glass microspheres used by Walker and Stewart and of
the larger droplets used by Boyle, e£ _aj_. That is, the effects of atmo-
spheric turbulence on the mixing of the microspheres and larger droplets
to the surface were small in comparison with the effects of gravitational
settling. For example, Figure I compares the calculated and observed
deposition profiles for Boyle, et^ al. Trial 1-5. There is a close
correspondence between the calculated and observed deposition profiles
within about the first 800 meters where deposition was principally
determined by the gravitational settling of the larger droplets. At
longer downwind distances, the only droplets available for deposition
were the smaller droplets which were affected by both gravitational
settling and atmospheric turbulence. The calculated and observed depos-
ition profiles show a secondary maximum between about 4 and 8 kilometers
that is explained by the downward reflection of the smaller droplets at
the top of the surface mixing layer. Although the correspondence between
the calculated and observed magnitudes and locations of the secondary
maximum is not exact, the calculated deposition generally is within a
factor of 2 of the observed deposition at all downwind distances. A
similar correspondence between calculated and observed deposition pro-
files was obtained for the four other Boyle, et_ a^. trials.
THE BUILD INC WAKE EFFECTS OPTION
The three sets of diffusion experiments used to test the
building wake effects option of the ISC Model were the tracer releases
at Millstone Nuclear Power Station (Johnson, et_ al^. , 1975), the tracer
releases at the Materials Test Reactor Engineering Test Reactor (MTR-
ETR) Complex (Islitzer, 1965) and the tracer releases at the CANDU
Nuclear Power Generating Station (Munn and Cole, 1967). Additionally,
the DOW Midland, Michigan plant's source, meteorological and air quality
data, which were provided to the U. S. Environmental Protection Agency
(EPA) for use in this study by DOW Chemical USA, were used to test the
building wake effects option. The calculations for each data set were
ix
-------
OBSERVED
CALCULATED
2 468 |Q3 2 4
DISTANCE FROM LINE SOURCE (m)
6 8 ,04
FIGURE I. Observed and calculated deposition profiles for Boyle, et al.
(1975) Trial 1-5.
-------
first performed without using the building wake effects or stack-tip
downwash options to provide a reference for judging the effectiveness of
these ISC Model features. The ISC Model calculations were then repeated
using the building wake effects option. A third set of ISC Model calcula-
tions for the buoyant stack emissions combined the building wake effects
and stack-tip downwash options.
The ISC Model building wake effects option enhances the initial
rate of vertical dispersion if the plume height to building height ratio
exceeds 1.2, but is less than 2.5, at a downwind distance of two building
heights, where the plume rise used to determine the plume height is the
rise due to momentum alone. If the plume height to building height
ratio is less than or equal to 1.2, the ISC Model enhances both the
initial rate of vertical dispersion and the initial rate of lateral
dispersion. The dispersion coefficients calculated within the building
wake region (three to ten building heights for a squat building and
three to ten building widths for a tall building) are functions of the
building dimensions and are independent of atmospheric stability. Thus,
the ground-level concentrations calculated within the wake region are
independent of stability. Table IT summarizes the four data sets used
to test the building wake effects option of the ISC Model. As shown by
Table II, the results of the ISC Model calculations provide insight into
the accuracy of the ISC Model when applied to either buoyant or non-
buoyant industrial stack emissions subject to building wake effects and
to non-buoyant low-level emissions subject to building wake effects.
The results of the tests of the building wake effects option
of the ISC Model revealed that It is at times difficult to specify in
advance the dimensions of buildings or groups of buildings affecting
initial plume dispersion. The lateral dispersion coefficients (o )
measured in the building wake region for the Islitzer (1965) experiments
were used to infer which buildings produced wakes affecting initial
dispersion. The comparison of calculated and observed concentrations
XI
-------
W
r-3
O
O
pQ ( 1
W
W O
PC O
H S
H CJ
C/3 CO
W M
H
W
O H3
Q P-i
W O
CO
P3 f,
O
CO M
H H
W FL,
CO O
•T, CO
H H
O W
W [X,
U4
C rH
O rt
•H Mi en
en cn CO
)-J 4-1 4->
CO C rt
fX CU fj
U) -rt
•H O
O -H
MH rH
13 M-I rt
cu cu u
•H O -H
4-1 CJ 4->
•H M
13 CU
£ >
13
-K
r-l
cu rtl
A3CM H
0 en cc
3 > 1 H
?•< tfl
H
CO - — -
13 00
0 II
r-l
pr j <^
fa
14-1 CO
O en
rt
CU 0)
O. rH
>. CO
H Pi
•K
em
C
•H
Td
r-H
*rj
£3
CO
t4_1
0
0)
p.
H
CO
O
^-4
rj
o
c/i
rt
4_)
rt
Q
cu
tH
en
CO
EH
o
oo
A
4-1
O
rt
*i r^t
13 O
CU 3
4J pq
rt
j> j>^
CU rH
rH 4J
W A3
t>0
•H
CO
^-^
in
vl
A3 CN
^__ •
** rH
A3
V 1
V 1
o
^H ,C
**^s **^_
A3
4-1
rt A3
3 4-1
CT* *iH
CO S
m
CTv
1
r^
•
^H
c«
^>
0)
»>
c
o
efl
^
o
cn
CO
EH
en
cu
^H
<3^
1 — ^.
^
•—(
0) -U
> C!
QJ ca
r-H ^
I O
-a 3
3 i
0 ti
M O
O ?;
A3
4-1
•H
— ^
4-J
o
3 m
cr1 ti
cn A
I 1*"^
r-l A3 A3
<1> A3 ~~-~
§"^5 ^
CO A3
- '—'
^— x
in
ON
rH
t-l
CO
C*4
4-J
•H
rH
en
M
en
cu
£H
en
cu
^
•> 4-1
•0 (3
co rt
4-1 »^>
rt o
> 3
CU M
rH |
W (3
O
X— ^
in
v 1
,n r-j
.1 .' •
rH
*"^
.r:' v|
V| J3
rH \
~**s (~;
4-1 A3
rt 4-1
3 -H
CO
p
CTi
rH
*^s
CO
rH
O
a
T3
G
rt
§
J3
en
01
cn
EH
•k
•K
0
r*^
A
A
t3 4-1
cu c
4J rt
rt P^
r> O
CO 3
rH PQ
in
m CN
V | V
A^1 ,n
A3 ,a
t* ,i i
r?
V
V|
r-H •
^-^ rH
4J A3
rt 4-i
3 -r-l
cr S
CO
4-J
C
rt
rH
P-l
13
(3
rt
rH
U
•H
^1
t ,,
g
M
C
•H
13
_j
f_^
•H
3
^
13
C
•rt
en
CO
O
t-t
O
CO
>
•H
4-1
0
cu
14-1
IM
CO
0)
A3
4-1
cn
•H
A3'
s~ s
rH
^.— ^
• •
CO
J^
o
rH
rH
O
14-1
CO
rt
13
cu
c
•H
MH
0)
T3
cu
rt
CO
rH
rt
4-1
CU
A3
4->
C
•H
13
a)
en
3
CO
}-f
0)
4-1
0)
1
rt
f\j
at
,r;
H
C
r>
^
2.
CU
""O
rt
J^j
00
4-1
13
rt
rH
a,
cu
>
o
r\
rt
4J
A3
OO
•H
CU
A3
*4
CJ
rt
J4->
en
rH
rt
CJ
•H
W
^»
A3
a
cu
A3
en
•H
A3
S~~\
("*")
*^S
f\
4-1
A3
60
•H
CO
A3
00
f*
•H
13
rH
•H
3
A3
CU
A3
4-1
en
•H
J.
A3
S~ N
CN
„
,.c2
4-1
13
•H
*
ni
6
CD
4-1
4J
•H
cu
y
O
rt
4-1
en
ai
A3
4->
en
•H
C
M
/"-^x
O
^^
M
>%
^ j
'>
rt
VJ
00
o
4J
01
3
*O
c
o
•H
4-1
rt
}-l
0)
rH
cu
o
o
rt
co
A3
4-1
cn
•H
oo
in
^^
*.
>.
4-1
•H
O
O
rH
CU
>
4J
•H
9
3
^
O
rt
4-1
cn
eu
r^
4J
en
'H
•
J-t
CO
4->
cu
s
Ort
•H
13
y
a
rt
4J
en
cu
A3
4-J
cn
•H
13
^^
oo
^— s
•a
C
rt
«*
cu
3
4J
rt
M
CO
P-»
0
cu
4J
M
•H
rt
4J
cu
•H
,JO
^
rt
cu
A3
4-1
cn
•r(
rt
H
s- s
1"-*,
^^
aT
3
4J
rt
M
cu
CX
C3
O
•H
4-J
rt
•H
rt
^!
3
0
A3
1
0
4-J
1
(-4
3
O
43
CU
A3
4-1
U-t
O
cu
cn
3
rt
a>
*
r-l
CU
>
CU
£j
o
A3
• »
O
•
0^1
13
CU
0)
O
X
cu
U)
,Si^
ej
rt
4J
I/I
rH
, — |
rt
^1
o
i^_j
en
u
cu
A3
R
3
c
cu
•o
3
o
n
13
ra
o
rH
1
•-H
i — 1
U-l
CU
1^
H
•K
•K
•
4-1
CU
cn
CO
4~l
«
13
*".*
§
0)
A3
4J
A3
•H
*^
j^i
3
CJ
O
O
rH
0
cn
rt
j2
o
rH
en
rt
en
rJ
cu
rO
R
P
(3
CU
a;
3
0
UH
,
cn
CO
•H
4-1
•H
O
O
•— 1
CU
>
4J
•r-t
01
l^f.
o
rt
4-1
CO
0)
r!
4-J
r.
-, i
xii
-------
given below for Islltzer (1965) data set Is based on the inferred
building dimensions. With this exception, building dimensions were
estimated following the general recommendations given in the ISC Model
User's Guide.
Millstone Diffusion Experiments
Johnson, et al. (1975) summarize the results of tracer diffusion
experiments conducted at Millstone Nuclear Power Station. The tracer
SF, was released from the reactor building main vent and the tracer
Freon-12B2 was simultaneously released from three vents on the adjacent
turbine building. Sampling arcs were located at downwind distances of
350, 800, and 1500 meters. (The 350-meter sampling arc was within the
building wake region.) The first set of ISC Model calculations that we
performed assumed no building wake effects, which is equivalent to the
Single Source (CRSTER) Model for the SF, trials and to the MPTER Model
for the Freon trials. We then repeated the concentration calculations
for the following combinations of ISC Model options:
• Case A - Distance-dependent plume rise and building
wake effects options
• Case B - Distance-dependent plume rise, building wake
effects and stack-tip downwash options
Additionally, concentrations were calculated under the assumption of
surface-based emissions affected by building wakes (Case C) because of
visual observations by Johnson, et^ al. that the plume often was inter-
mittently entrained into the building's cavity zone and effectively
acted as a ground-level source during these periods. The possible
effects of intermittent entrainment into the cavity zone include
enhanced vertical dispersion and reduced average plume heights.
xiii
-------
Table III summarizes the comparisons of concurrent calculated
and observed maximum centerline concentrations for the Millstone dif-
fusion experiments. Tf the building wake effects option is not exercised
(equivalent to current modeling techniques), the MR indicates that the
model, on the average, underestimates the concentrations observed in
the building wake region (i.e. , at the 350-meter sampling arc) by an
order of magnitude or more. The model's performance, as indicated by
both the MR and the RMSE, is significantly improved if building wake
effects are included in the model calculations for Cases A and B (the
elevated releases). However, the MR indicates that the model has a
systematic tendency to underestimate concentrations at all three sampling
arcs for Cases A and R. The assumption of surface-based emissions af-
fected by building wakes (Case C) yields the best overall model perform-
ance.
The systematic tendency of the ISC Model to underpredict
concentrations near the source for the Millstone diffusion experiments
could be a result of errors in the lateral dispersion coefficients,
errors in the vertical dispersion coefficients, errors in the effective
plume heights or a combination of these factors. A comparison of the
lateral dispersion coefficients calculated with building wake effects
included and the corresponding coefficients estimated from the concen-
tration measurements by Johnson, et a1. showed the calculated lateral
dispersion coefficients to be less than or equal to the observed values.
It follows that the ISC Model's underpredictions of concentrations near
the source for the Millstone diffusion experiments are most likely
caused by errors in the plume heights, vertical dispersion coefficients
or both. Although no vertical concentration profiles are available for
the Millstone experiments, the visual observations of plume behavior
made by Johnson, et al during the experiments indicate that the hourly
average effective emission height during most of the trials was inter-
mediate between the release height and the ground surface. Also, the
results of the model calculations for Case C suggest that the intermittent
xiv
-------
TABLE 1:I
SUMMARY OF THE RESULTS OF THE LSC MODEL CALCULATIONS FOR THE
MLLLSTONE DIFFUSION EXPERIMENTS (JOHNSON, ET AL., 1975)
Sampling Distance (ra)
350
800 1500
(a) Mean Ratio (MR) of C.i leu la ted to Observed Concentrations
SV
No Buildings
Buildings: Case A
Buildings: Case B
Buildings: Case C
Freon-12B2**
No Buildings
Buildings: Case A
Buildings: Case B
Buildings: Case C
0.00
0.33
0.33
0.66
0.14
0.47
0.48
0.74
0.18
0.46
0.46
0.74
0.62
0.55
0.55
0.72
0.62
0.73
0.74
1.09
0.84
0.68
0.68
0.83
(b) Root Mean Square Error (WISE)
SF6*
No Buildings
Buildings: Case A
Buildings: Case B
Buildings: Case C
Freon-12B2**
No Buildings
Buildings: Case A
Buildings: Case B
Buildings: Case C
259.90
3.17
3.09
2.06
19.76
2.20
2.12
1.53
9.38
2.31
2.28
1.83
5.07
1.89
1.88
1.64
2.88
1.79
1.78
1.88
1.94
1.69
1.68
1.64
(c) Percent Within a Factor of 2
SF6*
No Buildings
Buildings: Case A
Buildings: Case B
Build I ngs : Ca s e C
Freon-12B2**
No Buildings
Buildings: Case A
Buildings: Case B
Buildings: Case C
0
31
28
56
0
38
38
85
a
42
42
86
11
7'$
73
88
67
83
78
83
81
88
88
88
*The sample size is 36.
**The sample size is 26.
xy
-------
entrainment into the cavity zone, which currently is not fully accounted
for by the ISC Model, may be the primary cause of the model's tendency
to underestimate concentrations near the source for the Millstone dif-
fusion experiments.
We point out that the majority of the Millstone diffusion
experiments were conducted under very similar meteorological conditions
(moderate or strong south-southwest winds in combination with the neutral
Pasquill D stability category). Also, the upwind fetch was over water.
Thus, the Millstone data set may represent a unique combination of
source and meteorological factors. Additionally, the turbulent intensities,
especially the vertical turbulent intensities, measured during the
Millstone experiments generally were smaller than the median vertical
turbulent intensities implicit in the Pasquill-Gifford dispersion coef-
ficients used by the ISC Model. If these smaller vertical turbulent
intensities apply, the vertical expansion is reduced and this may account
for the model's tendency to underestimate centerline concentrations even
when surface-based emissions are assumed (Case C).
MTR-ETR Complex Tracer Releases
Lsllt/i-r (1965)) provides the re.siil.ts of ground-level non-
buoyant tracer ri1 Lenses m;idf downwind from the MTK-KTK Complex. Samplers
were placed at downwind distances of 118, 350, 550 and 850 meters from
the release point. The meteorological data available for the Islitzer
tests are not sufficient to use the Turner (1964) stability classifi-
cation scheme, the stability classification scheme routinely used with
the ISC Model. Consequently, the Nuclear Regulatory Commission's ver-
tical temperature difference (AT) and wind-direction standard deviation
(a ) stability classification schemes were used to estimate the stability
during each test. The Pasquill stability categories indicated for a
given test by the two schemes differed by as many as six categories.
However, with the exception of Test 7, one or both of the schemes indicated
the very unstable Pasquill A stability category during all of the tests.
xvi
-------
Table IV summarizes the comparisons of concurrent calculated
and observed centerline concentrations for the Islitzer (1965) tests
assuming that the Pasquill A stability category existed during all of
the tests. The 118-meter sampling distance is within the wake region of
the MTR-ETR Complex (i.e., between three and ten building heights downwind)
where the calculated concentrations are entirely determined by building
wake effects. As shown by the MR in Table IV, the inclusion of building
wake effects in the model calculations improves the average correspondence
between calculated and observed concentrations by almost a factor of 2.
Also, the RMSE is decreased from 2.7 to 2.0, and the number of calculated
concentrations within a factor of 2 of the corresponding observed concen-
trations is almost doubled. Under very unstable meteorological conditions
(the Pasquill A stability category), the calculated effects of atmospheric
turbulence on plume dispersion become far more important than the calcu-
lated effects of building wakes within a relatively short downwind
distance. Thus, the concentrations calculated by the ISC Model at the
350-, 550- and 850-meter sampling distances are primarily determined by
atmospheric turbulence and are not significantly affected by building
wakes.
CANDU Nuclear Power Station Tracer Releases
Eleven tracer releases are described by Munn and Cole (1967),
but only four of the tests were suitable for use in evaluating the ISC
Model. The single sampling arc for six of the eleven elevated, non-
buoyant tracer releases was within three building heights downwind
(i.e., within the building's cavity zone) where the ISC Model does not
calculate ground-level concentrations. Additionally, there is strong
circumstantial evidence that the plume centerline missed the sampling
arc during one of the five remaining trials in which the samplers were
placed in the building wake region between three and ten building heights
downwind.
XVJ 1
-------
TABLE IV
SUMMARY OF THE RESULTS OF THE ISC MODEL CALCULATIONS
FOR THE ISL1TZER (1965) TESTS ASSUMLNG
A STABILITY FOR ALL TESTS
Downwind
Distance
(m)
118
350
550
850
No. of Valid
Samples
LI
10
10
9
MR of Centerline
Concentrations
No Hldgs
1 . 94
1.87
L.4I
0.78
Bldjvs
L.09
1 .87
I .2b
0.72
RMSE of (Interline
Concentrations
No Bld&s
2.74
2.34
i. 94
1.86
Bldgs
1.97
2.34
1.82
1.90
% Within
Factor of 2
No Bldgs
36
40
50
67
Bldgs
b4
40
60
50
XV L I
-------
Table V summarizes the results of the comparisons of the
calculated center!ine concentraU ons and Hie corresponding maximum
observed concentrations for the- four Munn and Colo (1967) trials we
determined were suitable for use for the reasons given above. (Because
of the separation between samplers, the highest observed concentrations
were not necessarily measured at the plume centerline.) The results
presented in Table V illustrate the sensitivity of concentrations calcu-
lated in the wake region to different assumptions about the buildings
with wakes affecting initial plume dispersion. However, whether or not
the reactor building or all buildings are assumed to have affected
initial plume dispersion, the overall model performance is significantly
improved by the inclusion of building wake effects in the model calcula-
tions .
DOW Chemical Midland, Michigan Plant
The DOW Chemical USA Michigan Division plant in Midland,
Michigan includes two power houses with large sulfur dioxide (SO~)
emissions. The DOW Midland plant currently uses a Supplementary Control
System (SCS) to maintain the National Ambient Air Quality Standards
(NAAQS) for SO,-,. Because the plant's SCS curtails emissions whenever
high ground-level S02 concentrations are anticipated or observed, the
S0? emissions from the two power houses are highly variable. The air
quality impact of the plant is measured by nine continuous SO,., monitors.
The DOW data set used to test the building wake effects option of the
ISC Model consisted of the source, meteorological and air quality data
for 90 hours which were selected for model testing because of the occur-
rence of high ground-level SO,., concentrations in combination with mete-
orological conditions conducive to adverse building wake effects on
plume dispersion. We point out that uncertainties in the source, mete-
orological and a1 r quality data for the DOW data set tend to be larger
than for the three sets of published diffusion experiments discussed
above because the DOW data wore compiled from routine records of plant
xix
-------
TABLE V
SUMMARY OF THE RESULTS OF THE ISC MODEL CALCULATIONS
FOR THE MUNN AND COLE (1967) TRIALS*
Case
No Buildings
ReacLor Building
All Buildings
Mean Ratio of
Calculated to
Observed
Concentrations
(MR)
0.01
1.79
0.99
Root Mean
Square
Error
(RMSE)
176.84
2./»7
1.79
% of Centerline
Concentrations within a
Factor of 2 of Highest
Observed Concentrations
0
50
75
*Sample size is 4.
xx
-------
operations. Additionally, uncertainties about the locations of the
plume centerlines arising from the uncertainties about the mean trans-
port wind directions, which are not present in the data from the dif-
fusion experiments because the locations of the plume centerlines were
defined by the monitoring networks, cannot be removed from the DOW data
set. Thus, the RMSE values for the DOW data set are expected to be much
larger than for the diffusion experiments. Similarly, the percent of
calculated concentrations within a factor of 2 of the corresponding
observed concentrations is expected to be much smaller for the DOW data
set than for the diffusion experiments.
Table VI summarizes the results of comparisons of concurrent
calculated and observed 1-hour S0~ concentrations for the 143 observed
concentrations greater than or equal to 0.10 parts per million (ppm)
during the 90 hours selected for model testing. As shown by the table,
six combinations of ISC Model options were applied to the DOW data set.
On the basis of a site survey of the DOW Midland plant, Urban Mode 2 was
included in the model calculations (Combinations 4, 5 and 6 in Table VI)
because the presence of numerous large roughness elements (buildings) at
the plant indicated the possibility of enhanced turbulent intensities in
the vicinity of the plant. Whether or not building wake effects are
included in the ISC Model calculations, the correspondence between
calculated and observed concentrations obtained in Urban Mode 2 i.s su-
perior to that obtained In the Rural Mode. This result supports the
hypothesis of enhanced turbulence at the DOW plant. Although the building
wakt' effects and stack-Lip downwash options Lmprove. the model's perform-
ance tn both the Rural Mode and Urban Mode 2, the MR values suggest
that there may be errors in the model inputs or a bias in the ISC Model
towards underestimating concentrations for the DOW data set.
We are not certain that the MR, RMSE and percentage error bands
are the most appropriate measures of model performance for the DOW data
set because the density of the monitoring network is inadequate to deter-
xxi
-------
M
>
W
hJ
PQ
53
H
H
W
C/)
<:
H
<
Q
g
O
Q
U *
£ M
p^ ^-.4
0 H
FH PH
O
CO
S3 1-1
o w
M Q
H O
< 51
£o
u to
3M
U fo
o
nJ
W C/3
Q 5^
0 0
S M
H
U <
co a
M M
w§
ffi o
H O
sa
CO
w
Hrv/
l-M
ri °
t> (^
to
s
H
P-'
f-H
tt.
o
b
PS
?.i
ft-,
!_>
to
CM
UJ
O
01
O M
C 0
nj w
s o
w cd
O ti<
U-l
0) -r-l
0) ^
13
O &^5
a
M-j
o w
C/3
f A »F-<
UJ ^>.
QJ pg
V)
3
to
cd
0)
rt
CD
03 W
^£ a
0)
W) M-I
T3 W
rH
PQ
^— s
CO
r~^ O*
(Tv -H J2
rH H Cfl
^-^ / | C(J
cn o d
GC Ct) S
M) *J O
•H CO O
PQ
0)
Q a)
s T3
o
u S
to
M
o a
•H O
•P -H V-i
a. 4J Q)
o cd ^3
C 9
r-H -H 3
01 ^3 V.
"^O Q
S o
« 0
co cr\ ^f ^^o oo cj*\
f— i CM CO ^3"
~*
CM CM CM
r-H r-l rH C 0 C
crj c3 cij n3 d] oj
Vi V^ V-^ pO ,o >Q
3 a 3 M M M
pci CA prf C3 t3 O
• — i CM co -
-------
mine the locations of the plume centerlines. The current Guideline on Air
Quality Models (EPA, 1978) suggests that, as an alternative to direct
comparisons of concurrent calculated and observed concentrations, short-
term dispersion models may be evaluated by comparing the upper percentiles
of the calculated and observed concentration frequency distributions.
Additionally, Turner (1979) observes that a comparison of the maximum
concentration observed at any point with the maximum concentration cal-
culated at any point as of interest for regulatory purposes. We therefore
used the results of the concentration calculation for model option
Combinations L and 6 in Table VI to generate the cumulative frequency
distributions of calculated and observed hourly SO™ concentrations for
the DOW data set shown in Figure Ti. We point out that the concentrations
calculated by the TSCST program for model option Combination 1 are the
same as calculated by conventional modeling techniques (for example,
the MPTER Model). As shown by the figure, the upper end of the observed
concentration frequency distribution is much more closely matched by
ISCST for Combination 6 than by ISCST for Combination 1. It should be
emphasized that the hourly concentrations used to form the observed
concentration frequency distribution shown in Figure II were restricted
to concentrations greater than or equal to 0.10 ppm. Thus, the
observed concentration frequency distribution reflects the presence at
the monitors of emissions from one or both of the DOW power houses
during every hour. However, if the wind directions used as input to
the model are assumed to be representative of the winds defining
plume trajectories during every hour, the wind directions imply that
emissions from one or both of the DOW power houses were not present
at the monitors during every hour. Thus, it is not unexpected that the
ISC Model underpredicts the lower end of the observed concentration
frequency distribution in Figure II.
xxiii
-------
FT CURE I i .
(uudd) NOIlVdlNaONOO 2OS
Comp.-ir Icon of t-.-i Iculnl ccl .-incl obst-rvrd rtiiiui l.i I i ve 1-hour S02 concrn-
trntlcn f rt-qiicncy cl JsL r LbuflotKS lor the DOW i-t.
xxiv
-------
CONCLUSIONS AND SUGGESTIONS FOR MODEL IMPROVEMENT
We conclude from the tests described in this report of the key
features which distinguish the ISC Model from other generally available
dispersion models that:
1. The gravitational settling/dry deposition option adds
capabilities lacking in most current nodels, and the
accuracy of this option for particulates with appre-
ciable gravitationsl settling velocities appears to
correspond to the approximate factor of 2 accuracy
generally attributed to the results of short-term
dispersion model calculations in the absence of com-
plicating factors (AMS, 1978).
2. For plumes subject to building wake effects, the build-
ing wake effects option significantly improves the
performance of the ISC Model over that of the corres-
ponding models (CRSTF.R and MPTEk) that do not consider
building wake effects when used to calculate concen-
trations near the source.
However, the results of the model calculations for the Millstone data
set indicate thnt it may be desimble to modify the model to consider
a reduction in plume height due to initial entrainment into the cavity
zone for stacks on or adjacent to squat buildings with stack height to
building height ratios less than 1.2
The best overall model performance for the Millstone data
set was obtained for Case C (see Table III) , and essentially the same
performance can be achieved by adding to the ISC Model the following
definition of the effective emission height for the insignificantly
buoyant plume from a stack on or adjacent to a squat building with a
stack height to building ratio less than 1.2:
xxv
-------
H{x} =
h + Ah{x> ; h + Ah {2hu} > 1.2 h,
m b — b
0 ; h + Ah {2h } < 1.2 h,
mo b
(III)
where
H{x} = effective emission height at downwind distance x
h = physical stack height
Ah{x} = plume rise due to momentum and/or buoyancy at downwind
distance x, calculated by Equation (2-4) or Equation
(2-7) in the ISC Model User's Guide
Ah {2h. }
ra b
plume rise due to momentum alone at a downwind distance
of 2h, , where h, is the building height
As explained above, we believe that the ISC Model's tendency to under-
estimate concentrations for Case C, as indicated by the MR, may be
attributed to vertical turbulent intensities that are smaller than the
median values associated with the Pasquill-Gifford vertical dispersion
coefficients used by the model. It should be noted that the addition
of Equation (III) to the LSC Model will have no effects on the model's
performance tor the DOW data set because the stack height to building
height ratios exceed 1.2.
The results of the ISC Model calculations given in Table VI
for the DOW data set indicate that the model's building wake effects
option may have a systematic tendency to underestimate the concentrations
produced near the source by buoyant stack emissions subject to building
wake effects. As part of the ISC Mode] evaluation study, we added the
Cramer, et al. (1975) stack-tip downwash correction and the Scire and
xxvi
-------
Schulman (1980) "downwash radius" correction to special versions of the
I SCSI program and repeated the concentration calculations for the DOW
data set. Both corrections improved the model's performance as indicated
by the MR, RMSE and percentage error bands. However, as noted above,
the MR, RMSE and percentage error bands may not be the most appropriate
measures of model performances for the DOW data set because the uncer-
tainties about the locations of the plume centerlines arising from the
uncertainties in the mean transport wind directions cannot be removed
from the DOW data set, as was done in the diffusion experiments, because
the density of the DOW monitoring network is inadequate for fixing the
locations of the plume centerlines. It is not possible, from a com-
parison of the upper percentiles of calculated and observed concentration
frequency distributions, to determine whether or not either of the
two corrections significantly improves the model's performance for the
DOW data set.
xxvii
-------
Section
TABLE OF CONTENTS
Title
Page
4
5
DISCLAIMER ii
ACKNOWLEDGEMENTS iii
EXECUTIVE SUMMARY iv
INTRODUCTION l-l
1.1 Background and Purpose 1-1
1.2 Review, Evaluation and Selection of Candidate
Data Sets 1-2
1.3 Rationale for Model Testing 1-3
1.4 Report Organization 1-6
TESTS OF THE GRAVITATIONAL SETTLING AND DRY
DEPOSITION OPTION 2-1
2.1 Walker (1965) Depositon Experiments 2-1
2.2 Stewart (1968) Deposition Experiments 2-15
2.3 Boyle, et^ al. (1975) Aerial Spray Experiments 2-22
TESTS OF THE AERODYNAMIC WAKE EFFECTS OPTION 3-1
3.1 Millstone Nuclear Power Station Tracer
Experiments (Johnson, e_t^ al., 1975) 3-2
3.2 Islitzer (1965) Tracer Releases 3-33
3.3 Munn and Cole (1967) Tracer Releases 3-50
3.4 Dow Midland, Michigan Plant Data 3-56
SUGGESTIONS FOR MODEL IMPROVEMENT 4-1
CONCLUSIONS 5-1
REFERENCES 6-1
Appendix
A
CALCULATED AND OBSERVED DEPOSITION PROFILES FOR
THE WALKER (1965) AND STEWART (1968) EXPERIMENTS
A-l
xxviil
-------
SECTION 1
INTRODUCTION
1.1 BACKGROUND AND PURPOSE
The Industrial Source Complex (ISC) Dispersion Model (EPA, 1979)
was developed by the H. E. Cramer Company, Inc. under Contracts 68-02-2547
and 68-02-3323 with the Source Receptor Analysis Branch (SRAB) of the U. S.
Environmental Protection Agency (EPA). The ISC Model, which is designed
to evaluate the air quality impact of emissions from large industrial
source complexes, 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, an advanced version of the Single Source (CRSTER) Model
(EPA, 1977), is designed to use sequential hourly meteorological data 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, which combines and updates basic features of the Air Quality Dis-
play Model (EPA, 1969) and the Climatological Dispersion Model (Busse and
Zimmerman, 1973), is designed to use statistical summaries of wind-speed
and wind-direction categories, classified according to the Pasquill sta-
bility categories, to calculate seasonal and annual concentration or dry
deposition values.
The ISC Model programs contain the following special features:
(1) physical separation of multiple sources; (2) plume rise due to momen-
tum and buoyancy as a function of downwind distance; (3) effects of aero-
dynamic building wakes on plume dispersion; (4) effects of stack-tip
downwash on plume dispersion; (5) emissions from stack, area, volume and
line sources; (6) effects of gravitational settling and dry deposition;
(7) effects of variations in terrain height on ground-level concentrations
("full terrain adjustment") for receptors with elevations less than the
stack top elevation; and (8) time-dependent exponential decay of pollutants.
1-1
-------
Many of the basic components of the ISC Model have been previously tested
by EPA (for example, Turner et^ al^., 1972 and Lee, e_t_ _al_. , 1975) and others.
However, the ISC Model includes several dispersion model options (the build-
ing wake effects and gravitational settling/dry deposition options) that
are not contained in any of the models currently recommended in the Guideline
on Air Quality Models (EPA, 1978). Consequently, it is necessary to
test these new features. The purpose of this report is to describe the
results of an evaluation study in which these key TSC Model features
were tested. This report assumes that the reader is familiar with the
detailed technical description of the ISC Model given in Section 2 of the
Industrial Source Complex (ISC) Dispersion Model User's Guide (EPA3 1979).
1.2 REVIEW, EVALUATION AND SELECTION OF CANDIDATE DATA SETS
A primary consideration in the review and evaluation of candidate
data sets for use in the ISC Model evaluation study was to select data sets
that would provide a test of one or more of the key features of the ISC
Model. During the summer of .1978, an extensive review of published data
sets was conducted to determine which data sets might be suitable for use
in the ISC Model evaluation study. The selection criteria were as follows:
• Basic requirements — simple terrain, absence of extra-
neous sources in the vicinity (unless their effects can
be reliably quantified), reliable and sufficient air
quality data (including background), reliable emissions
data and representative meteorological data
• TSC Model requirements — the source must exhibit one
or more of the key ISC Model features listed in Section
1.1 above, adequate input data must exist for consider-
ing the feature (s) in the model calculations (for example,
particulate size data, building dimensions, etc.) and air
quality data must be sufficient to reflect, the effect(s)
of the complicating feature(s)
1-2
-------
The review of the published data sets revealed that, in many
cases, the published source, meteorological and concentration or deposition
data lacked the detail required for model testing. Also, several excellent
data sets collected at nuclear power plants (for example, Halitsky and
Woodard, 1974 and Start, et^ al., 1977) reflected meteorological conditions
that masked building wake effects so that the data were not suitable for
testing the ISC Model building wake effects option. Table 1-1 lists the
data sets that were found to be suitable for use in the ISC Model evalua-
tion study. In addition to published data, Table 1-1 includes the data
that were provided for use in this study by DOW Chemical USA for the
Michigan Division plant in Midland, Michigan. As shown by Table l-l,
each unique ISC Model feature discussed In Section 1.1 is included in at
least one of the selected data sets. Direct measurements of plume rise
and stack-tip downwash are not available for any of the data sets.
However, comparisons of calculated and observed concentrations for some
of the data sets provide indirect tests of these ISC Model features.
It is important to note that the data sets listed in Table 1-1
are not sufficient in number or detail to validate the new features of the
ISC Model in a strict statistical sense. However, the data sets provide
a basis to evaluate the performance of these new features. Also, the data
sets enable a comparison of the performnnce of the TSC Model with that of
current models. For example, tin; ISC Model short-term program JSCST In
its default mode corresponds to the Single Source (CKSTHR) Model for a
single .source, and to the MPTKK Model (Pierre arid Turner, J 980) for
multiple sources.
1.3 RATIONALE FOR MODEL TESTING
Insofar as possible, consistent procedures were used throughout
the ISC Model evaluation study. However, the model tests made using the
various data sets may at times appear to be inconsistent because the data
sets themselves are inconsistent in the information available. Additionally,
1-3
-------
TABLE 1-1
DATA SETS SELECTED FOR USE IN THE ISC
MODEL EVALUATION STUDY
Data Source
ISC Model Feature(s)
Walker, 1965
Stewart, 1968
Boyle, at al., 1975
Islizter, 1965
Munn and Cole, 1967
Johnson, et al., 1975
DOW Chemical USA
Michigan Division
Gravitational settling and dry deposition
Gravitational settling and dry deposition
Gravitational settling and dry deposition;
volume source representation of a line source
Aerodynamic wake effects
Aerodynamic wake effects
Aerodynamic wake effects; stack-tip downwash;
plume rise as a function of downwind distance;
and separation of multiple sources
Aerodynamic wake effects; stack-tip downwash;
plume rise as a function of downwind distance;
and separation of multiple sources
1-4
-------
some of the calculations that were performed were in part determined
by the available source-specific information such as visual observations
of plume behavior. Because the observed concentration or deposition values
were reported in a variety of normalized, crosswind-integrated and/or
non-dimensional forms, we selected the following parameters as the primary
measures of model performance: (1) the mean ratio (MR) of concurrent
calculated to observed concentration or deposition .values, (2) the root
mean square error (RMSE) as defined by Draxler (1980), and (3) the per-
centage error bands. The MR is defined as
N
2~t Xci
MR = —-1 (1-1)
N
1=1
where xci is the calculated concentration or deposition for the i trial
and x j is the corresponding observed concentration or deposition. Be-
cause air quality observations usually span more than one order of magni-
tude, Draxler (1980) uses a geometric form of the RMSE:
RMSE =•= exp
(£n xci - *n Xoi)2/N
(1-2)
The percentage error bands, as used in this study, are simply the percent
of the trials with calculated concentration or deposition values within
a factor of 2 of the corresponding observed values.
In the absence of any systematic errors in a dispersion model,
the model inputs, or the observed concentration or deposition values, the
MR should tend toward unity for a large sample size. Although inspection
1-5
-------
of Equation (1-2) shows that the RMSE also equals unity for a perfect
model, uncertainties on a trial-by-trial basis in the model inputs as
well as in the concentration or deposition measurements may significantly
affect the RMSE. That is, the RMSE for a given data set is in part
determined by the accuracy of the model and is in part determined by the
quality of the model inputs and the concentration or deposition measure-
ments. The percentage error bands are also significantly affected by
the quality of the model inputs and the concentration or deposition
measurements. However, the RMSE and the percentage error bands are con-
venient indicators of relative model performance for a given data set.
1.4 REPORT ORGANIZATION
The unique features of the ISC Model may be grouped into two
general categories: (1) the gravitational settling/dry deposition option,
and (2) the building wake effects option. The results of tests of the
gravitational settling and dry deposition option, made using the Walker
(1965), Stewart (1968) and Boyle, £t al. (1975) data sets, are described
in Section 2. The Boyle, et^ al. data also provide a test of the ISC Model
representation of an elevated line source with initial source dimensions.
The results of teats of the building wake effects option, made using the
Johnncm, ej^ aJL (1975), IwHtzer (196!)), Munn and Cole (1967) and DOW data
sets, are discussed In Section 3. The .lohuson, eU <'ii. nnd DOW data also
include the effects of the physical separation of multiple sources and
are used to illustrate the effects on calculated ground-level concentrations
of the generalized plume rise equations and the stack-tip downwash option.
Section 4 discusses possible improvements to the ISC Model. The study
conclusions are given in Section 5. The deposition profiLes for the
Walker (1965) and Stewart (1968) experiments are presented in Appendix A.
1-6
-------
SECTION 2
TESTS OF THE GRAVITATIONAL SETTLING AND
DRY DEPOSITION OPTION
Three sets of deposition experiments (Walker, 1965; Stewart,
1968; and Boyle, £t al., 1975) were used to test the gravitational set-
tling and dry deposition option of the ISC Model. The premise of the
tests was that, if the ISC Model can match observed deposition patterns,
the ISC Model can also be expected to match ambient concentrations with
gravitational settling and dry deposition occurring because of mass con-
tinuity. The tests made using the Walker and Stewart data sets, both of
which involved continuous, elevated releases of glass microspheres, are
discussed in Sections 2.1 and 2.2, respectively. Section 2.3 discusses
the use of the deposition data from the aerial spray releases reported
by Boyle, et al. to test both the ISC Model gravitational settling and
dry deposition option and the ISC Model line source representation.
2.1 WALKER (1965) DEPOSITION EXPERIMENTS
Walker (1965) summarizes the results of twelve 30-minute to 60-
minute continuous, elevated releases of solid glass spheres. The experi-
ments were conducted at the Suffield Experimental Station in Ralston,
Alberta, Canada. The emission heights were 15 and 7.42 meters, and the
diameters of the glass spheres ranged from 56 to 107 micrometers. Samples
were collected on sticky paper at radial distances of 27.4, 45.7, 73.2,
86.9, 100.6, 114.3, 128.0, 146.3, 201.2, 274.3, 402.3 and 804.6 meters.
The angular separation between sampling locations was 3.6 degrees. (No
illustration of the experimental layout is provided by Walker.) For each
trial and downwind distance, Walker gives the standard deviation of the
crosswind deposition pattern a in radians and the crosswind-integrated
deposition (CWID), divided by the total emissions Q . The source data
2-1
-------
for the Walker experiments are discussed in Section 2.1.1, the meteoro-
logical data are discussed in Section 2.1.2, and the ISC Model calculation
procedures and results are given in Section 2.1.3.
2.1.1 Source Data
During the deposition experiments described by Walker (1965),
solid glass spheres were emitted from a continuous elevated point source
with no buoyancy and minimal momentum. Three sizes of glass spheres were
used. For Trials A through F, the mass mean diameter of the glass spheres
was 107 micrometers and the density was 2.4 grains per cubic centimeter.
For Trials 0, H, I and L, the mass mean diameter was 56 micrometers and
the density was 2.2 grams per cubic centimeter. For Trials J and K, the
mass mean diameter was 49 micrometers and the density was 2.0 grams per
cubic centimeter. Walker gives terminal fall velocities V of 0.58, 0.19
and 0.14 meters per second, respectively, for the 107-, 56- and 49-microm-
eter spheres. We computed terminal fall velocities for the three sphere
sizes using the methods of McDonald (1960) and found very good agreement
with the values given by Walker. We estimated the surface reflection
coefficient y f°r the three sphere sizes from Figure 2-8 of the ISC Model
User's Guide. During Trials A through F, the glass spheres were emitted
at a height of 15 meters above the ground. The height of emission from
Trials G through L was 7.42 meters. The source parameters for all twelve
trials nre given in Table 2-1. For each trial, these parameters include
the date, the start time, the duration of the trial, the total weight of
the glass spheres emitted, the settling velocity and the surface reflec-
tion coefficient.
2.1.2 Meteorological Data
Walker (1965) gives the mean wind speed, mean wind direction and
ambient air temperature for each of the twelve deposition experiments.
Wind speeds were measured on a tower at heights of 0.5, 1.0, 2.0, 4.0 and
2-2
-------
(/I
H
w
ON
11
CM
W
t£i rr*
3 H
H O
co
c c
O 0)
WO c
CJ -H >-
0) MH
rH <»-<
U-l Q)
. O
C w ^ W
-O
4-1 0) /-<.
3 4J v-'
0 -H
IfloT
c
O O ^O "^ ^J NO CO if} CO *>O
O rH o o ( — m in (TI w r*~
in in CM CO CM r-| rH rH ro O
ON OO rH in rH Q\ CT^ *^ CO "vj" Q^
u~i vO 10 in to in \o so ^ tn
rOUCCCU>^>C4J
GJflJCd(OQ)OQ)OCQO
feg'n^MSspwe'no
vo in ON ON rH r~* r^» CM co in
CM rH CM CM rH rH rH
00 •*-*
3 o
< O
O OJ
CM
$4 tJ
-------
8.0 meters during all trials and at 16.0 meters during Trials A through
F. For each trial, Walker gives the mean wind speed at each tower level
and for the layer between the emission height and the surface. We used
a logarithmic least-squares regression technique with the tower wind
speed measurements to calculate a wind-profile exponent p for each
trial. The Pasquill stability category was estimated for each trial
using the Turner (1964) criteria, which are based on wind speed and
solar radiation (insolation). Insolation indices were estimated on the
basis of the dates and times of the various trials. The 9-meter wind
speeds were used in determining the stability category because this
level is closest to both the airport wind measurement height used by the
Turner (1964) stability classification scheme and the 10-meter measure-
ment height used by the original Pasquill (1961) scheme. Because of the
low emission heights and relatively short distances to the sampling
arcs, the depth of the surface mixing layer did not significantly affect
the results of any of the trials. Consequently, unrestricted vertical
mixing was assumed for all of the Walker trials.
Table 2-2 lists, for each of the twelve trials, the meteorolog-
ical input parameters required to apply the ISC Model short-term computer
program ISCST to the Walker trials. The ambient air temperature and verti-
cal potential temperature gradient are used by the ISC Model to calculate
plume rise. Because the experiments described by Walker involved non-
buoyant omissions with essentially no initial vertical momentum and no
plume rise, estimates of the vertical potential temperature gradients and
measurements of the ambient air temperatures for the Walker trials are
not given in Table 2-2. All of the trials were conducted under neutral
(D stability) conditions. As explained in Section 2.1.3, the ISC Model
deposition calculations were performed in two ways. First, the mean
wind speed al the measurement height nearest the emission height (shown
as uOv) in Table 2-2) was used as the reference wind speed and was
adjusted to the emission height using the observed wind-profile exponent.
2-4
-------
TABLE 2-2
METEOROLOGICAL PARAMETERS FOR THE WALKER (1965) EXPERIMENTS
Trial
A
B
C
D
E
F
G
H
I
J
K
L
4*
Average Flow
Vector (Deg) *
070
125
010
010
130
005
078
289
025
155
065
060
Wind Speed
UL**
5.53
6.19
6.06
5.52
8.20
4.87
2.43
4.96
2.78
5.98
5.58
7.37
u{h}***
7.65
7.64
7.39
6.96
9.14
5.65
2.56
6.29
4.09
7.13
6.54
8.89
Wind Profile
Exponent
0.34
0.22
0.23
0.26
0.16
0.20
0.32
0.21
0.32
0.20
0.18
0.18
Pasquill
Stability
Category
D
D
D
D
D
D
D
D
D
D
D
D
*Direction toward which the wind is blowing.
**Mean wind speed for the layer from the surface to the emission height.
***Mean wind speed at the measurement height nearest the emission height.
2-5
-------
Second, the mean Layer wind speed (shown as u in Table 2-2) was used
as the reference wind speed and the wind-profile exponent was set equal
to zero.
2.1.3 Calculation Procedures and Results
The source data given in Section 2.1.1 and the meteorological
data given in Section 2.1.2 were used with the ISC Model short-term com-
puter program ISCST in the Rural Mode to calculate, for each trial, the
total deposition along the plume centerline at the sampling distances
and at several intermediate distances to aid in defining the distribution
of the deposited glass spheres. For each trial, the calculated centerline
deposition values were then multiplied by /2ir cr , where 0 is the
lateral dispersion coefficient, and divided by the amount of material
emitted Q to obtain a profile of the calculated normalized crosswind-
integrated deposition CWID/Q for comparison with the observed values
reported by Walker (1965). Model a values for the appropriate Pasquill
stability category were used in the normalized CWID calculations. (It
is important to note that the crosswind integration removes from the
observed deposition values the effects of differences in the duration
of emissions for the various trials.)
The mean wind speed that is appropriate for use in dispersion
model calculations for particulates with appreciable gravitational settling
velocities is the average wind speed through the layer between the plume
stabilization height and the ground surface. The ISC Model approximates
this wind speed by the wind speed at the emission height, an approxima-
tion that is appropriate for buoyant stack emissions. However, the trials
summarized by Walker involved non-buoyant emissions of glass spheres
with relatively large gravitational settling velocities. Consequently,
the ISC Model deposition calculations for the Walker trials were performed
in two ways. First, the mean wind speed at the measurement height nearest
Che emission height was adjusted to the emission height using the wind-profile
2-6
-------
exponent calculated from the wind measurements. Second, the wind-profile
exponent was set equal to zero (wind speed independent of height above
the surface) and the average wind speed given by Walker for the layer
between the emission height and the surface was used in the deposition
calculations.
Figures A-l through A-12 in Appendix A compare the calculated
and observed normalized CWID profiles for the twelve trials. In each
case, the normalized CWID profile calculated using the mean layer wind
speed u. is a better approximation to the observed normalized CWID
profile than is the profile calculated using the mean wind speed at the
emission height u{h}. The shapes of the normalized CWID profiles
calculated using the mean layer wind speed u are generally in good
Li
agreement with the shapes of the corresponding observed profiles. That
is, the differences between the calculated and observed CWID values at
various downwind distances for a given trial tend to be systematic. An
uncertainty about the amount of material emitted during a trial or a
systematic error in the estimates of the observed CWTD could account for
this result.
Table 2-3 compares the calculated and observed distances to
the maximum normalized CWID values for the Walker (1965) trials. The
distances to the maximum calculated and observed normalized CWID values
were obtained by inspection of the plotted CWID profiles. As shown by
Table 2-3, the maximum observed CWID values occurred at downwind distances
of less than 100 meters during about half of the trials. Because the ISC
Model cannot make deposition calculations at downwind distances less
than 100 meters, direct comparisons of calculated and observed distances
to maximum CWID values are not possible for about half of the trials. For
the trials with calculated and observed distances to maximum CWID values
greater than or equal to 100 meters, the overall performance of the ISC
Model is about the same If the mean wind speed at the emission height ii{h}
or the mean wind speed for the layer between the emission height and the
ground UL is used In the calculations. Because u{h} always exceeds
2-7
-------
TABLE 2-3
COMPARISON OF CALCULATED AND OBSERVED
DISTANCES TO MAXIMUM NORMALIZED
CWID VALUES FOR THE WALKER
(1965) EXPERIMENTS
Trial
A
B
C
D
E
F
G
H
I
J
K
L
Distance to Maximum CWID/Q Value (m)
Observed
128
135
120
115
145
90
80
100
80
95
100
1.30
ISC (u{h»
135
135
135
130
150
115
A
*
A
*
*
100
ISC (UL)
115
120
120
115
145
A
*
A
A
A
A
100
Mean K<-H io (MR)
Ratio of Predicted to Observed
u{h)
1.05
1.00
1.13
1.13
1.03
1.28
A
A
A
*
A
0. 77
1 .04
"L
0.90
0.89
1.00
1.00
1.00
A
A
A
A
A
A
0.77
0.92
*The ISC Model cannot make deposition calculations at distances less than 100
meters. Inspection of the calculated CWID profile indicates that CWID values
higher than the value calculated at 100 meters may have occurred at distances
less than 100 meters.
2-8
-------
u,, the distance to the maximum CWID value is always greater if u{h}
Li
is used in the calculations that if UT is used in the calculations.
L
Table 2-4 compares the calculated and observed maximum normalized
CWID values for the Walker trials, obtained by inspection of the plotted
CWID profiles. As explained above, the ISC Model cannot make deposition
calculations at downwind distances less than 100 meters. Consequently,
for each trial with the maximum observed CWID located within 100 meters
of the source (see Table 2-3), Table 2-4 compares the calculated and
observed CWID values at 100 meters. The mean ratio (MR) of calculated to
observed maximum CWID values is 0.56 if the mean wind speed at the
emission height u{h} is used in the calculations and 0.71 if the mean
wind speed between the emission height and the surface u is used in
ij
the calculations. There is more than a factor of 2 difference between
calculated and observed maximum CWID values for four of the twelve
trials if u{h} is used in the calculations and for two of the twelve
trials if u is used in the calculations.
LJ
It is important to note that the values of the surface reflection
coefficient y suggested for use in the ISC Model User's Guide are
principally based on experiments involving spray droplets (see Section
2.3) and may not be applicable to glass spheres. The maximum possible
deposition is calculated if the reflection coefficient y is set equal
to zero. Because Trials G through L have non-zero reflection coefficients
(see Table 2-1), we repeated the calculations for Trials G through L
with the reflection coefficient y s£t equal to zero. Figures A-13
through A-18 in Appendix A show the observed normalized CWID profiles
for Trials G through L and the corresponding profiles calculated with the
reflection coefficient set equal to zero. Also, Table 2-5 compares the
calculated and observed maximum CWID values with the reflection coef-
ficient set equal to zero for all trials. (If the maximum observed CWID
is located within 100 meters of the source, Table 2-5 compares the
calculated and observed CWID values at 100 meters.) The MR Is 0.63
if the mean wind speed at the emission height u{h) is used in the calculations
2-9
-------
TABLE 2-4
COMPARISON OF CALCULATED AND OBSERVED MAXIMUM
NORMALIZED CWID VALUES FOR THE WALKER
(1965) EXPERIMENTS*
Trial
A
B
C
1)
E
F
G
H
I
J
K
1.
Maximum CWID/Q (mg/(g-m))
Observed
20.6
10.8
6.0
7.4
6.0
9.6
8.8
3.9
9.4
2.7
2.9
3. 1
ISC (u{h})
5.4
5.2
5.5
6.0
4.2
8.2
4.3
2.8
3.7
1.7
1.8
2.2
ISC (UL)
7.9
8.0
7.2
8.3
4.7
10.0
4.6
3.4
4.5
1.9
2.0
2.5
Moan Kilt Jo (MR)
Ratio of Predicted to Observed
u{h}
0.26
0.48
0.92
0.81
0.70
0.85
0.49
0.72
0.39
0.63
0.62
0.71
0.56
"L
0.38
0.74
1.20
1.12
0.78
1.04
0.52
0.87
0.48
0.70
0.69
0.81
0.71
*lf the maximum observed CWID for a trial occurs at a distance of less than
100 meters, the calculated and observed CWID values in the table are the
values at 100 meters because the ISC Model cannot make deposition calcula-
tions at distances less than 100 meters.
2-10
-------
TABLE 2-5
COMPARISON OF CALCULATED AND OBSERVED MAXIMUM NORMALIZED
CWID VALUES FOR THE WALKER (1965) EXPERIMENTS
WITH ZERO REFLECTION COEFFICIENTS FOR
ALL TRIALS*
Trial
A
B
C
D
E
F
G
H
I
J
K
L
Maximum CWID/Q (mg/(g-m))
Observed
20.6
10.8
6.0
7.4
6.0
9.6
8.8
3.9
9.4
2.7
2.9
3.1
ISC (u{h}>
5.4
5.2
5.5
6.0
4.2
8.2
5.9
3.5
5.1
2.9
3.0
3.0
ISC (u"L)
7.9
8.0
7.2
8.3
4.7
10.0
6.3
4.1
6.1
3.2
3.3
3.4
Mean Ratio (MR)
Ratio of Predicted to Observed
u{h}
0.26
0.48
0.92
0.81
0.70
0.85
0.67
0.90
0.54
1.07
1.03
0.97
0.63
GL
0.38
0.74
1.20
1.12
0.78
1.04
0.72
1.05
0.65
1.19
1.14
1.10
0.79
*If the maximum observed CWID for a trial occurs at a distance of less than
100 meters, the calculated and observed CWID values in the table are the
values at 100 meters because the ISC Model cannot make deposition calcula-
tions at distances less than 100 meters.
2-11
-------
and 0.79 11" the mean wind speed between the emission height and the
surface u is used In the calculations. There is more than a factor of
2 difference between calculated and observed CWID values for two of the
twelve trials if u{h}is used In the calculations and for one of the
twelve trials if UT is used in the calculations. Consequently, the dis-
4->
crepancies between calculated and observed maximum CWID values for
Trials G through L may in part reflect uncertainties about the appropri-
ate value of the reflection coefficient for glass spheres.
Table 2-6 summarizes the comparisons of concurrent calculated
and observed maximum CWID values for the Walker data set. The best
overall model performance is obtained if the mean layer wind speed u,
is used in the deposition calculations with the surface reflection coef-
ficient set equal to zero for all trials. For this combination of model
inputs, the MR indicates that the ISC Model, on the average, predicts
the maximum CWID to within about 20 percent for the Walker trials.
Also, the minimum root mean square error (RMSE) is obtained for this
combination of inputs, and the concurrent calculated and observed maximum
CWID values are within a factor of 2 for all but one of the trials.
Walker also gives, for each trial and sampling distance, the
standard deviation of the crosswind deposition pattern 0 in radians.
We multiplied each O value in radians by the appropriate downwind
distance in meters to obtain the corresponding a in meters. Table
2-7 gives the ratios of calculated (Pasquill-Gifford) to observed a
values. In general, the calculated to observed a ratio is approxi-
mately the same at all sampling distances for an individual trial. The
departures from unity of the calculated to observed 0 ratios probably
reflect the effects of using discrete stability categories in the ISC
Model.
2-12
-------
TABLE 2-6
SUMMARY OF THE RESULTS OF THE ISC MODEL CALCULATIONS FOR
THE WALKER (1965) DEPOSITION EXPERIMENTS*
ISC Model Inputs
Wind Speed
u(h>
\
U {h}
"L
Reflection
Coefficients
From Table 2-1
From Table 2*1
0 for all trials
0 for all trials
— . . „,
MR of
Maximum
CWID** Values
0.56
0.71
0.63
0.79
RMSE of
Maximum
CWID** Values
1.86
1.57
1.66
1.42
% Within
a Factor
of 2
67
83
83
92
*The sample size is 12.
**Crosswind-integrated deposition.
2-13
-------
TABLE 2-7
RATIOS OF CALCULATED TO OBSERVED STANDARD
DEVIATIONS OF THE CROSSWIND DEPOSITION
PATTERN c FOR THE WALKER (1965)
EXPERIMENTS
Trial
A
B
C
D
E
F
G
H
I
J
K
L
Mean
Ratio**
Downwind Distance (in)
100.6
1.47
0.95
0.79
1.19
0.50
0.62
0.42
0.81
1.12
0.47
0.41
0.45
0.64
114.3
-
0.95
-
-
-
-
-
-
-
-
-
-
0.95
128.0
1.52
0.97
0.73
1.14
0.50
0.63
0.42
0.86
1.23
0.49
0.39
0.46
0.64
146.3
-
0.83
-
-
-
-
-
-
-
-
-
-
0.83
201.2
1.49*
1.15
0.82
1.12
0.44
0.56
0.56
0.83
0.62
0.49
0.33
0.46
0.58
274.3
-
0.99*
-
-
0.45
0.58*
0.81*
0.76
0.90*
0.48
0.29
0.45
0.45
402.3
-
-
0.85*
0.39*
0.42*
0.61*
2.43*
0.83*
0.97*
0.42
0.27
0.33
0.33
804.6
-
-
0.95*
0.33*
0.58*
-
-
1.59*
-
0.39*
0.23*
0.24*
-
1097.3
-
-
-
-
-
-
-
-
-
0.55*
0.26*
-
-
Mean
Ratio**
1.49
0.98
0.78
1.14
0.46
0.60
0.47
0.80
0.84
0.46
0.31
0.46
0.54
*0bserved a value reported as unreliable.
**Mean ratios exclude unreliable measurements.
2-14
-------
2.2 ST1WAKT (1968) DEPOSITION EXPERIMENTS
Stewart (1968) summarizes the results of seven continuous,
elevated releases of glass spheres. Five of the experiments (the B
series) were conducted at the Bale du Dore Research Station of the Great
Lakes Institute, University of Toronto, Ontario, Canada. The release
height for the B series was 18.6 meters, and the sphere diameters ranged
from 100 to 200 micrometers. The two other experiments (the S series)
were conducted at the Suffield Experimental Station, Ralston, Alberta,
Canada. The release heights for the S series were 30.5 and 92.4 meters,
and the sphere diameters ranged from 50 to 200 micrometers. For both
series, sticky-tape samplers were placed at 1.3- or 2-degree intervals
at radial distances of 2, 3, 5, 7, 9, 11 and 13 stack heights. (No
illustration of the experimental layout at either site is given by
Stewart.) The results of the experiments presented by Stewart for each
trial are in non-dimensional form and are restricted to: (1) the maximum
crosswind-integrated deposition (CWID), multiplied by the release height
h and divided by the total amount of material emitted Q ; and (2) the
distance to the maximum CWID, divided by the release height h. Thus,
the results given by Stewart are far less detailed than the results
given by Walker (1965). The source data for the Stewart experiments are
discussed in Section 2.2.1, the meteorological data are discussed in
Section 2.2.2, and the ISC Model calculation procedures and results are
given in Section 2.2.3.
2.2.1 Source Data
The deposition experiments described by Stewart (1968) consisted
of the continuous, elevated release of solid glass spheres over a period
of JO to 43 minutes. The spheres were emitted with no buoyancy and
minimal momentum. In all but one of the seven trials, two sizes of
glass spheres were emitted and the deposition of each size of sphere was
evaluated separately, resulting in the equivalent of 13 separate trials.
2-15
-------
For Trials Bl, B2, B3, B4 and S2, glass spheres with diameters of 100
and 200 micrometers were used. For Trial B5, only 100-micrometer
spheres were used. Spheres with diameters of 50 and 100 micrometers
were used in Trial SI. The gravitational settling velocities of the
three sizes of glass spheres given by Stewart are in good agreement with
the corresponding settling velocities that we computed using the methods
of McDonald (1960). We determined the surface reflection coefficient y
for each trial from Figure 2-8 of the ISC Model User's Guide. The
source parameters used in the ISC Model calculations for the Stewart
trials are given in Table 2-8. For each trial, these parameters include
the emission height, total amount of material emitted, gravitational
settling velocity and surface reflection coefficient. Additionally,
Table 2-8 gives the date, start time and duration of each trial.
2.2.2 Meteorological Data
The only meteorological inputs to the ISC Model short-term
computer program ISCST that are given l>y Stewart (1968) are wind speed
and wind direction. Wind measurements were made at heights of 2, 4, 8,
12 and 18 meters for Trials Bl through B5 and at heights of 2, 8, 16,
48, 64, 80 and 92 meters for Trials SI and S2. However, Stewart gives
only one wind speed and wind direction for each trial. We believe that
Che reported wind speeds and directions probably are for the layer
between the emission height and the ground surface. As in the case of
the Walker (1965) trials discussed in Section 2.1, the mean layer wind
speed UT is probably appropriate for use in ISC Model calculations for
i_i
the type of experiments summarized by Stewart. We therefore used the
wind speeds reported by Stewart with the wind-profile exponent set equal
to zero (wind speed independent of height above the surface). Although
Stewart does not give any mixing height data for the seven trials, it is
extremely unlikely that the mixing height affected the maximum deposition
for these trials because of the low emission height and the relatively
large gravitational settling velocities of the glass spheres. Consequently,
2-16
-------
cn
H
00
vO
CM
U
4-t
c c
o a)
•H -H
•U O C
0 -H >-
0) <4-l
U-l , O
C *J CO
•H -H CO
rH O ""*^
w ° 3,
t®
4J 01 ^-N
§4-1 00
JJ V.X
JTa 0^
M
o /_v
7J .
(rj C
3 5
0)
H
tia
W r-4
ctt v^
CO
eu
4-1
a
9"
S
01
s
•H
— .
rH
H
O O 0 O
^3 vO v^ ^O
CO CO CO CO
CM CM CM CM
m m o o
rH rH rH rH
o o m m
CM CM CO CO
o o o o
-t -» Sf s*
vO vO vO vO
rH rH rH rH
3333
r^ r** oo oo
rH rH CM CM
O O O O
O O O O
rH CM rH CM
rH rH CM CM
PQ PQ PQ PQ
O
VO
00
CM
m
o
o
o
CM
0
CO
rH
m
o
00
sf
vO
rH
3
00
CM
o
o
rH
CO
PQ
O
vD
co
CM
-*
rH
CO
rH
rH
O
co
rH
m
o
00
^
vD
rH
•3
oo
CM
O
o
CM
CO
PQ
O
vo
oo
CM
m
0
0
o
CM
m
vO
rH
0
rH
vO
^
vD
00
O
rH
O
O
rH
^j.
PQ
O
vO
00
CM
m
o
o
0
CM
O
^-
m
CM
CM
rH
^
vD
00
3
f^
CM
O
O
rH
m
PQ
0
vO
00
CM
«T
rH
O
O
rH
0
-a-
CM
CM
rH
^
vO
00
r^
CM
0
O
CM
in
PQ
in
CM
CO
O
m
o
r-
rH
o
rH
VO
CO
rH
CM
O
rH
oo
o
rH
rH
in
vo
rH
3
-t
rH
0
in
r-l
CO
o
in
o
CM
m
o
rH
vo
CO
rH
CM
O
rH
00
o
rH
rH
in
vO
rH
3
.J.
rH
O
O
rH
rH
CO
o
<3-
CM
cy*
CM
tn
o
CM
O
ON
VO
CM
rH
rH
m
rH
in
vO
rH
3
,3.
rH
O
O
rH
CM
CO
O
-------
we assumed unrestricted vertical mixing in the ISC Model calculations
for the Stewart trials.
Table 2-9 gives the meteorological inputs used in the ISC
Model calculations for the deposition experiments reported by Stewart.
For each trial, these inputs include the average flow vector, the mean
wind speed and the Pasquill stability category as determined using the
Turner (1964) criteria. All but two of the trials were conducted under
slightly unstable (C stability) conditions. One trial was conducted
under neutral (D stability) conditions and one of the trials was conducted
under slightly stable (E stability) conditions. As explained above, the
wind-profile exponent was set equal to zero for all trials. The ambient
air temperature and vertical potential temperature gradient, which are
used by the ISC Model to calculate plume rise, were not required as
model inputs because the experiments described by Stewart involved non-
buoyant emissions.
2.2.3 Calculation Procedures and Results
The source data discussed in Section 2.2.1 and the meteorolog-
ical data discussed in Section 2.2.2 were used with the ISC Model short-
term computer program ISCST in the Rural Mode to calculate, for each
trial, the total deposition along the plume centerline at the sampling
distances and at several other distances to aid in defining the downwind
distribution of the deposited spheres. These centerline deposition
values were then multiplied by /2'/r a h, where h is the emission
height, and divided by the total amount of material emitted Q to
obtain the calculated non-dimensional crosswind-integrated deposition
(CWID) values for comparison with the observed maximum values given by
Stewart (1968). (The crosswind integration removes from the observed
deposition values the effects of differences in the duration of emissions
for the various trials.)
2-18
-------
TABLE 2-9
METEOROLOGICAL PARAMETERS FOR THE
STEWART (1968) EXPERIMENTS
Experiment
Bl
B2
B3
B4
B5
SI
S2
Average
Flow Vector*
(deg)
034
046
012
032
007
040
099
Wind
Speed
(rn/sec)
6.3
5.6
4.6
2.5
4.9
5.4
4.8
Pa squill
Stability
Category
C
C
E
D
C
C
C
*Direction toward which the wind is blowing.
2-19
-------
The observed maximum non-dimensional CWID and the calculated
non-dimensional CWID profile for each combination of trial and sphere
diameter (except Trial B3 for 200-micrometer spheres) are presented in
Figures A-19 through A-28 in Appendix A. Trial B3/200 is omitted because
the calculated nondimensional CWID at 100 meters is five orders of
magnitude lower than the maximum observed value at 56 meters. The
maximum observed CWID values were found at distances less than 100
meters from the source for five combinations of trial and sphere diam-
eter. Although the ISC Model cannot be applied at distances less than
100 meters from a source, subjective extrapolations of the calculated
CWID profiles for these cases show a good qualitative agreement between
the observed and calculated values of maximum non-dimensional CWID (see
Figures A-19, A-20, A-22 and A-24). It is important to note that the
non-dimensional CWID values calculated by the ISC Model at 100 meters
for these trials cannot be compared with the observed non-dimensional
CWID values because Stewart gives only the maximum non-dimensional CWID
values.
Table 2-10 compares the calculated and observed values of the
magnitude and location of the maximum non-dimensional CWID for each of the
Stewart trials. For the six trials with maximum observed CWID values
located more than 100 meters from the source, the MR values indicate that,
on the average, the ISC Model predicts the magnitude and location of the
maximum CWID to within about 10 pert-.ent Tor the Stewart trials. All of
the concurrent calculated and observed maximum CWID values for these trials
are within a factor of 2, and the RMSE is only 1.51. Tims, the performance
of the ISC Model for the Stewart deposition experiments is equivalent to the
model's performance for the Walker (1965) deposition experiments discussed
in Section 2.1.
2-20
-------
w
>—H I
g§
a
rt
0)
"^ 3
o
f4
u
0
a
3
""U
0)
&
Q)
CO
O
}
5
4
•)
a
>
a
j
3
>
(N <& rH vO OO r— 1
"*^*K »^"^ »*^ ••••
r-« rH rH O O rH
O m in o O >o
* -K in* -K <-)•*: r^os» * * CO* coir»»OiA
• • * • * •
O O rH rH rH rH
^j. ^j. ^ pr, (^ 0
* " f^ * * C5 CS| ON rH rH
rH rH O O rH VO
d d d d o c>
o5rHoincN^nrHSiSo
in^cooocOrHroooosr
OOOrHOOOOOOO
ooooooo poo
OQOOOOO OOOO
^C^rHC^^HrH^inHrHC^
nmmmMmM MMMCO
CXI
o
r— (
/*N
g
O
•H
«
P^
a
cu
J2
CN
rH
i-l
O
s
v— '
o
•r-f
4J
CO
C
cd
cu
SS
co
^
cu
4_l
I
O
o
CO
CO
HI
to
co
cu
CJ
§
4-1
co
nH
T3
T3
C
4-1
cd
CO
c
o
•H
4-1
CTJ
rH
U
rH
C8
O
O
•H
4-1
•H
CO
O
a.
cu
•o
cu
A!
to
e
§
g
CJ
cu
T3
O
oo
2-21
-------
2.3 BOYLE, ET AL. (1975) AERIAL SPRAY EXPERIMENTS
Boyle, et al. (1975) present the results of five releases of
oil spray from a low-flying DC-7B aircraft. The experiments were con-
ducted at the U. S. Army Dugway Proving Ground in Utah for the Food and
Agriculture Organization of the United Nations. The aircraft flew at
heights of 27 to 88 meters above the surface. However, the effects of
aircraft wake vortices in combination with atmospheric turbulence de-
termined the effective emission height and initial dimensions of the
source. Spray was released along a pre-marked flight path over dis-
tances ranging from 7.4 to 14.1 kilometers. For three of the five
trials (Trials 1-5, 1-6 and 1-7), the aircraft released oil spray along
a line nearly perpendicular to the mean wind direction. The sampling
grid for these trials is shown in Figure 2-1. During the two remaining
trials (Trials 2-2R and 2-3), the aircraft flew approximately into the
mean wind. The sampling grid for the two alongwind releases Is shown in
Figure 2-2. The drop sizes for the oil spray ranged from about 15 to
180 micrometers. Deposition samples were collected on Printflex cards
placed directly on the ground at the sampling locations shown in Figures
2-1 and 2-2. Boyle, ej^ al. give the total deposition determined at each
sampling location by drop-size category. The data are also summarized
in graphical form. The source data for the Boyle, et al. experiments
are discussed in Section 2.3.1, the meteorological data are discussed in
Section 2.3.2, and the ISC Model calculation procedures and results are
given In Section 2.3.3.
2.3.1 Source Data
The Boyle, elt oJL. (1975) experiments consisted of five aerial
releases from a DC-7B aircraft of an oil spray at a height of 27 to 88
meters above the surface. The aircraft flew along a straight line and
released oil spray over distances ranging from 7.4 to 14.1 kilometers.
The spray, which was released just above the wing of the aircraft, was
2-22
-------
140°
M
i
r 230 0
1
170 O
I6O O
1523m
i
ISO O
I4O O
I3O O
I2O O
no o
IOO A To war
9O O
80 O
F -»x-x A%
6O O
5O O
4O O
/
True North
402 m spacing continuing downwind
i i
40? m
o-«
161 m spacing to station 23
°T
161 m
O-'
152m
ol
0
O
O
O
00 0
O
O (J,
o tr
O
o
3O O O 1
20ji— 183 m— ig»— 183 m-ngsO.S m
Lin* A Line 8 Line C *
FIGURE 2-1. Sampling locations for the crosswind spray releases reported
by Boyle, et al. (1975).
2-23
-------
10
20 •
30
40*
60*
70
80*
90*
100
1097 m-
1097m
True North
I46m
600 Meters
FIGURE 2-2. Sampling locations for the alongwind spray releases reported
by Boyle, e^ al_. (1975).
2-24
-------
vigorously mixed with ambient air by the aircraft's wake vortices. The
wake vortices sank toward the ground as they dissipated, resulting in
long, narrow volume sources with effective emission heights less than or
equal to the actual release heights.
The ISC Model represents a long, narrow volume source (i.e.,
a line source) by a series of adjacent volume sources. The individual
volume sources have initial horizontal dimensions (0 ) equal to the
width of the line source divided by 2.15. Boyle, et_ al_. estimated the
effective width of the spray plume to be equal to three wingspans, or 84
meters. Therefore, for all five trials, the initial lateral dispersion
coefficient o~ for an individual volume source was set equal to 84
yo
meters divided by 2.15, or 39 meters. The number of volume sources was
determined for each trial by dividing the length of the spray release
path by 84 meters. The effective emission heights of the volume sources
and the initial vertical dispersion coefficients 0 were estimated by
Boyle, et al. primarily on the basis of vertical measurements of spray
dosage made on a tower downwind of the release line. The height of the
peak dosage was defined as the effective emission height and the distance
from the peak dosage to the top of the cloud was divided by 2.15 to obtain
O , Table 2-11 lists, for each trial, the effective emission height,
the effective source strength (amount disseminated multiplied by the
dissemination efficiency and divided by the number of sub-sources), the
length of the dissemination path, and the number of sub-sources used to
simulate the spray releases.
The spray drops released in the trials described by Boyle, ej^
aj^. varied in size from about 15 to 180 micrometers. The droplet size
distributions and corresponding settling velocities were determined
experimentally and are given In the report by Boyle, et al. These data,
which were used directly as ISC Model inputs, are reproduced in Table 2-12.
Also given In Table 2-12 are the surface reflection coefficients Y
n
and the mass fractions 4> that correspond to the various drop-size
2-25
-------
CN
W
W
rH
Cfl
•H
(^
CO
I
CN
CN
1
CN
r*^
I
rH
^O
1
rH
m
i
rH
V4
0)
4-1
0)
a
cfl
t-i
cfl
PU
V
O
^|
£j
O
en
r-» m
CN rH
vo in
^3* rH
00
m m
rH CO
oo
*
m m
rH CO
CO
f
m CN
rH CN
c
o
,_J
o
co
>-l
eu
euo
co a
•H s— '
c o
o o
•H rH N
CO CO t>
co o
•rl -rl 4J
0 /-* *J fl
w a w y
> 4J -rl
•rl J2 rH «4H
4-1 GO CO «W
O -rl -rl CU
(U 0) WO
<4-l ffl -HO
14-4 fj
W H
CO
O
m
CN
rH
CN
rH
CO
CO
O
O
ON
CN
m
o
CN
oo
"""^
•K
4-1
00
a
0)
)_l
4-1
(U
o
M
0
en
o>
•H
4-)
O
co
14-1
PL)
CO
O
t<
CN
vO
\O
00
o
1
r^
^
00
CN
O
rH
CO
O
rH
^
00
OO
•
CO
o
rH
X
vO
p-^
CO
^
o
rH
^
oo
O
•rl
T3
C
cO
U
C
(U
•rl
O
C
o
•H
0)
CO
03
•H
0)
0)
•H
•r-l
4J
rH
a
"
CU
W
§
i .
cO to
CU
rH O
Cfl >-i
4-> 3
O O
H CO
2-26
-------
TABLE 2-12
DEPOSITION PARAMETERS FOR THE BOYLE,
ET AL. (1975) EXPERIMENTS
Mean Drop
Diameter
(ym)
Settling
Velocity
(m/sec)
Mass
Fraction
*n
Reflection
Coefficient
Yn
Trial 1-5
17.0
29.7
46.7
64.2
82.1
100.4
119.1
138.3
7.25x10"!?
2.21X10 y
5.47x10"^
1.04x10":
1.69x10 ,
_ i
2.32x10 I"
2.81x10":
3.49x10"
8.00xlO~?
— 1
1.65x10
3.02x10"
3.96x10":
»7
8.20x10 ,
—2
3.33xlO_^
9 . 70x10
3 . 40xlO~3
0.80
0.71
0.61
0.49
0.33
0.17
0.045
0.0
Trial 1-6
25.2
37.4
53.3
69.2
85.1
100.9
116.7
132.4
148.1
163.6
179.0
_2
1.65x10 *
3.67x10"::
—7
7.46x10
1.26x10":
1.90x10":
2.41xlO~:
— 1
2.84x10 :•
3.38x10 :
4.07x10
4.78x10 7
_1
5.45x10
_3
7.50x10 ;?
-*7
9.95x10 ,
_1
2.02x10
2.72x10"
1.56x10"
1.14x10"^
7.69x10"^
3.63x10"^
—7
2.83x10 ,
5.90xio~;:
1.30x10
0.74
0.66
0.56
0.44
0.27
0.145
0.035
0.0
0.0
0.0
0.0
Trial 1-7
25.2
37.4
53.3
64.2
85.1
100.9
116.7
132.4
148.1
1.67x10 2.
— 7
3.67x10 ,
— 7
7.46x10
1.26x10"
1.90x10":-
2.41x10",
2.84xlO~r
3.38x10":
4.07x10"
-2
1.03x10 ,
1.17x10",
— 1
3.33x10 :•
2.74x10":
1.39x10"^
7.47x10 ,
—7
2.74x10
7.90x10";:
1.56x10
0.74
0.66
0.56
0.44
0.27
0.145
0.035
0.0
0.0
2-27
-------
TABLE 2-12 (Continued)
Mean Drop
Diameter
(ym)
Settling
Velocity
(m/sec)
Mass
Fraction
*n
Reflection
Coefficient
^n
Trial 2-2R
17.0
29.7
46.7
64.2
82.1
100.4
119.1
138.3
7.47xlO~^
2.28x10 ,
—.9
5.63x10 f
T
1.06x10
1.74x10 T
2.37x10
2.87x10 .
_i
3.58x10
9.80x10"^
9.55x10*^
3.36x10"::
_i
3.35x10 I"
1.22xlO~^
6.75x10
2.57x10"^
_x
9.70x10
0.81
0.70
0.61
0.485
0.315
0.155
0.030
0.0
Trial 2-3
17.0
29.7
46.7
64.2
82.1
100.4
7.34x10 ^
2.24xlO~^
5.54x10 ^
1.05x10 \
1.71x10,
— 1
2.34x10
1.50x10"^
1.82xlO~T
4.09x10"}
3.19x10",
4.38x10"^
3.12x10
0.81
0.71
0.61
0.50
0.295
0.165
-------
categories. The reflection coefficients were determined from Figure 2-8
of the ISC Model User's Guide.
2.3.2 Meteorological Data
Extensive meteorological data were collected during the trials
reported by Boyle, ej^ al. (1975). Wind speed and wind direction were
measured at six levels on a 98-meter tower for Trials 1-5, 1-6, and 1-7
and at three levels on a 32-meter tower for Trials 2-2R and 2-3. PIBAL
and rawinsonde soundings were made for each trial to determine the depth
of the mixing layer and the vertical profiles of wind speed and wind
direction. Observations of cloud cover and ambient air temperature were
also made for each trial. With the exception of the Pasquill stability
category, all of the meteorological input parameters required by the ISC
Model short-term computer program ISCST are provided in the report by
Boyle, et al.
Table 2-13 lists the meteorological inputs for the trials
described by Boyle, et: al. The cloud cover and wind data given by Boyle,
et al. were used to estimate the Pasquill stability category for each
trial following the Turner (1964) procedures. Because the wind speeds
in Table 2-13 are at the effective emission heights, the wind-profile
exponent was set equal to zero for all trials. The ambient air tempera-
tures and vertical potential temperature gradients during the various
trials are not shown in Table 2-13 because the ISC Model does not require
these parameters to calculate deposition for volume sources.
2.3.3 Calculation Procedures and Results
The source data given in Section 2.3.1 and the meteorological
data given in Section 2.3.2 were used with the ISC Model short-term com-
puter program ISCST in the Rural Mode to calculate, for each trial,
values of the total deposition at various distances normal to the line
2-29
-------
TABLE 2-13
METEOROLOGICAL PARAMETERS FOR THE BOYLE,
ET AL. (1975) KXPERIMENTS
Meteorological
Parameter
Average Flow Vector *
(deg)
Mean Wind Speed**
(m/sec)
Mixing Height
(m)
Pasquill Stability
Category
Trial
1-5
150
3.0
850
B
1-6
120
1.0
150
B
1-7
125
1.8
400
C
2-2R
135
4.0
350
C
2-3
140
4.2
500
C
*Direction toward which the wind is blowing.
**Mean wind speed at the effective emission height.
2-30
-------
source. For each trial, calculations were made for receptors along
three parallel Lines that corresponded to the sampling lines (see Figures
2-1 and 2-2). The calculated deposition values at points equidistant
from the line source were averaged to obtain representative deposition
values (the observed data were averaged by Boyle, e_t^ al. in a similar
manner). We then plotted the calculated total deposition values for
each trial versus distance 1'rom the line source for comparison with the
observed deposition profile. The calculated and observed deposition
profiles for the. five trials are shown in Figures 2-3 through 2-7. In
general, the profiles generated by the ISC Model are in good agreement
with the observed profiles. Specific differences between the calculated
and observed profiles are discussed below.
Figure 2-3 compares the calculated and observed deposition
profiles for Trial 1-5. In general, the calculated and observed deposi-
tion values are within a factor of 2 at all downwind distances. The
highest calculated and observed.deposition values occur at the first
sampling distance. There is a secondary maximum in the observed deposit-
Ion profile between about 4 and 8 kilometers that is attributable to
the downward reflection of smaller droplets at the top of the surface
mixing layer. This effect is also reproduced by the ISC Model, although
there are differences in the calculated and observed magnitudes and
locations of the secondary maximum. (All droplets that mix to the top
of the surface mixing layer are assumed by. the ISC Model to be reflected
downward, but droplets that mix to the ground surface may or may not be
reflected upward, depending on the value of the surface reflection
coeffient.)
Figure 2-4 compares the calculated and observed deposition
profiles for Trial 1-6. The ISC Model generates a deposition profile
which approximately matches the shape of the observed profile. However,
the overall correspondence between the calculated and observed deposition
profiles is not as good as in the case of Trial 1-5. It should be noted
that the ability of the ISC Model to duplicate the observed deposition
2-31
-------
-i
r-r
T 1 1 1—I—TT
102
OBSERVED
CALCULATED-
2 4 6 8 |Q3 2 4
DISTANCE FROM LINE SOURCE (m)
6 8
I04
FIGURE 2-3. Observed and calculated deposition profiles for Boyle, e_t al.
(l()7rj) Trial 1-5.
-------
I02
CM
I
o>
V)
o
Q.
LJ
O
10'
8
10°
8
6
10
-I
I I
I III
|0*
OBSERVED
CALCULATED
4 6 8 |0S 2 468 |Q4
DISTANCE FROM LINE SOURCE (m)
FIGURE 2-4. Observed and calculated deposition profiles for Boyle, et al.
(1975) Trial 1-6.
2-33
-------
profile for Trial 1-6 is hampered by the fact that there were signific-
ant shifts in the wind direction during the period when the oil spray
was affecting the sampling network.
I '
Figure 2-5 shows the calculated and observed deposition pro-
files for Trial 1-7. Following the Turner (1964) definitions of the
Pasquill stability categories, the slightly unstable C category existed
during Trial 1-7. However, the standard deviation of the wind azimuth
angle o. during Trial 1-7 indicates that meteorological conditions
during Trial 1-7 probably corresponded to the unstable B 'category.
Consequently, ISC Model deposition profiles for both stability categories
are shown in Figure 2-5. At downwind distances less than about 1 kilom-
eter, the calculated deposition profiles are principally determined by
the gravitational settling of large droplets and are minimally affected
by atmospheric turbulence. Thus, there is a generally good agreement
within the first kilometer between the observed deposition profile and
the two calculated deposition profiles. If the C stability category is
assumed to have existed during the trial, the ISC Model calculates a
secondary maximum in the deposition profile that is farther downwind and
lower in magnitude than the observed secondary maximum. If the B stability
category is assumed to have existed during the trial, the agreement
between the ISC Model deposition profile and the observed profile is
significnntly improved at the longer downwind distances where the only
droplets still available for deposition are the smaller droplets which
are significantly affected by both gravitational settling and atmospheric
turbulence. The calculated and observed deposition values generally
agree to within a factor of 2 at all downwind distances if the B stab-
ility category is assumed to have occurred during Trial 1-7.
Figure 2-6 compares the calculated and observed deposition
profiles for Trial 2-2R. The DC-7B aircraft flew approximately into the
mean wind during Trial 2-2R, and deposition was measured on both sides
of the flight path. For the area to the right of the flight path, the
ISC Model deposition profile is within a factor of 2 of the observed
2-34
-------
M
e
**—»
2
O
cn
O
a.
CALCULATED:
C Stability
2 4 6 0 (03 Z 4 6 8 |Q4
DISTANCE FROM LINE SOURCE (m)
FIGURE 2-5. Observed and calculated deposition profiles for Boyle, £t al.
(1975) Trial 1-7.
2-35
-------
N
e
e
10'
8h
4 -
2 -
10° ~
8
6
O
CL
LJ
Q
^-2
1 till
OBSERVED
CALCULATED-
-PROFILES RIGHT OF FLIGHT PATH
|- PROFILES LEFT
I I I 1 I
4 68 |Q2 2 468 |Q3 2 468 |Q4
DISTANCE FROM LINE SOURCE (m)
FIGURE 2-6. Observed and calculated deposition profiles for Boyle, et al
(1975) Trial 2-2R.
2-'36
-------
profile at distances of more than about 200 meters normal to the flight
path. The ISC Model overestimates the deposition within about 200
meters. For the area to the left of the flight path, the ISC Model
calculates deposition values that are higher than the observed values at
all distances normal to the flight path.
Figure 2-7 compares the calculated and observed deposition pro-
files for Trial 2-3. As in the case of Trial 2-2R, the DC-7B aircraft
flew almost directly into the mean wind, and the deposition was measured
on both sides of the flight path. For the area to the right of the
flight path, the ISC Model yields calculated deposition values beyond
about 200 meters that are within a factor of 2 of the observed values.
However, the ISC Model overpredicts the deposition in the area to the
left of the flight path. In both Trials 2-3 and 2-2R, the component of
the mean wind normal to the flight path was blowing from left to right,
causing the highest calculated and observed deposition values to occur
to the right of the flight path. The performance of the ISC Model for
these two trials is good for the area to the right of the flight path
and poor for the area to the left of the flight path. The ISC Model does
not accurately account for deposition in the area upwind of the line
source because deposition in this area is principally determined by the
complex interaction of the wind and the aircraft wake vortices.
Table 2-14 Hummurl/.t-H Llie comparisons of concurrent calculated
and observed deposition values at 100 and 500 meters from the release
line for the Boyle, et^ ai. experiments. We point out that the calculated
and observed deposition values are near the center of the line source where
edge effects are unimportant. Also, Table 2-14 considers only the depo-
sition to the right of the flight path for Trials 2-2R and 2-3 because
the ISC Model cannot account for deposition upwind of the flight path
for these trials. The MR and RMSE values in Table 2-14 show that the un-
certainties in the model calculations tend to increase with distance from
the line source. This result is explained by the fact that deposition
near the source was principally determined by the gravitational settling
2- •}
-------
10'
8
6
10°
CSJ
'e
o>
2
O
o
flL
LU
O
\-z
"I—I—I "I I
1 1 \ 1 1—I—TT
1 1 1—i—i i r
r PROFILES LEFT
OBSERVED-
CALCULATED-
PROFILES RIGHT OF FLIGHT PATH
4 6 8 |02 2 4 6 8 |03 2 468
DISTANCE FROM LINE SOURCE (m)
FIGURE 2-7. Observed and calculated deposition profiles for Boyle, et al.
(1975) Trial 2-3.
2-38
-------
TABLE 2-14
SUMMARY OF THE RESULTS OF THE ISC MODEL CALCULATIONS
FOR THE BOYLE, ET AL. (1975) DEPOSITION EXPERIMENTS*
Distance from
Line Source (m)
100m
500m
MR of
Deposition Values
0.91 (0.90)**
1.78 (1.92)**
RMSE of
Deposition Values
1.89 (1.90)**
2.33 (2.13)**
% Within a
Factor of 2
80 (80)**
60 (80)**
*The sample size is 5.
**The numbers enclosed by parentheses assume that B stability rather
than C stability existed during Trial 1-7.
2-39
-------
of the larger droplets. At longer distances from the source, deposition
was determined by smaller droplets which were affected by both gravita-
tional settling and atmospheric turbulence.
'.>.-t\0
-------
SECTION 3
TESTS OF THE AERODYNAMIC WAKE
EFFECTS OPTION
The published results of three diffusion experiments and the
source, meteorological and air quality data for an existing industrial
source complex were used to test the aerodynamic building wake effects
option of the ISC Model. The calculations for each data set were first
performed without using the building wake effects, distance-dependent
plume rise and stack-tip downwash options to provide a reference for
judging the effectiveness of these ISC Model features. The distance-
dependent plume rise option was then used in combination with the build-
ing wake effects option In the ISC Model calculations for buoyant stack
emissions because: (1) The building wake effects option was specifically
designed for use in combination with distance-dependent plume rise equa-
tions, and (2) Sensitivity analyses performed during the development of
the ISC Model showed that the use of the final plume rise option in com-
bination with the building wake effects option often yielded zero or
effectively zero calculated concentrations within the building wake
region (i.e., within ten building heights or widths downwind, depending
on the critical length scale). The third set of TSC Model calculations
for buoyant stack emissions involved the- combination oT the distanre-
plume rise, building waki- effc'C'ts and slack- Lip downwash options.
The published diffusion experiments used to test the building
wake effects option of the ISC Model consisted of the elevated, slightly
buoyant tracer releases at Millstone Nuclear Power Station described by
Johnson, e£ al . (1975), the ground- level tracer releases downwind of the
MTR-ETR Complex at the National Reactor Testing Station described by Islitzer
(1965), and the elevated, non-buoyant tracer releases at CANDU Nuclear
Power Generating Station described by Munn and Cole (1967). Additionally,
DOW Chemical USA provided to EPA for use in this study concurrent source,
meteorological and air quality data for the DOW Michigan Division plant
-------
in Midland, Michigan (Brown, 1979a and 1979b and Abrams, 1978 and 1979).
The three published data sets and the DOW data provide direct tests of
the building wake effects option of the ISC Model. The Millstone and
DOW data sets also provide direct tests of the ISC Model's representation
of the physical separation of sources and indirect tests of the model's
generalized (distance-dependent) plume rise equations and the stack-tip
downwash option. The application of the ISC Model to the Johnson, et_
al. (1975), Islitzer (1965), Munn and Cole (1967), and DOW data sets is
discussed in Sections 3.1, 3.2, 3.3 and 3.4, respectively.
3.1 MILLSTONE NUCLEAR POWER STATION TRACER EXPERIMENTS (JOHNSON,
ET AL., 1975)
Johnson, et al. (1975) present the results of gaseous tracer
experiments that were conducted at the Millstone Nuclear Power Station
in southeastern Connecticut during the fall of 1974. Figure 3-1 is a
map of the area surrounding the Millstone Station, which is located in
Waterford, Connecticut on the tip of a small peninsula that extends into
Long Island Sound. Sulfur hexafluoride (SF,) gas was released from the
reactor building main vent and dibromodifluoromethane (Freon-12B2) gas
was released from three of the ten vents on the adjacent turbine building.
Air samples were taken with bag samplers along arcs at nominal downwind
distances of 350, 800 and 1500 meters from the reactor building main
vent. The angular spacing between samplers was approximately 6 degrees.
The sampler locations are shown by the + symbols in Figure 3-1. The re-
sults of 36 releases of SF, and 26 releases of Freon-12B2 are presented
in the report by Johnson, et^ al. For each experiment, a non-linear least-
squares regression technique was used to fit a Gaussian-shaped curve to
the concentrations observed along each sampling arc. An estimate of
the centerline concentration for each experiment at each distance was
then derived from the fitted curves. These estimates are presented by
Johnson, et^ al. in normalized form XU/Q» where x is the centerline
concentration In micrograms per cubic meter, Q is the source emission
-------
N
MILLSTONE
NUCLEAR
POWER
STATION
238.5
LEGEND*
AIR SAMPLERS
A NUSCO 137-m METEOROLOGICAL
TOWER
150.8 deg
FIGURE 3-1. Map of the area surrounding the Millstone Nuclear Power Station.
The -f symbols show the sampling locations used in the experiments
described by Johnson, e£ a_l . (1975) and the A symbol shows the
location of the 137-meter meteorological tower.
3-3
-------
rate in grams per second, and u is the hourly mean wind speed at the
release height in meters per second. The source data for the Millstone
experiments are given in Section 3.1.1, the meteorological data are
discussed in Section 3.1.2, and the ISC Model calculation procedures and
results are given in Section 3.1.3
3.1. 1 Source Data
Figure 3-2 is a scale drawing of the Millstone Nuclear Power
Station; building heights in meters above ground level (AGL) are also
shown in the figure. During the diffusion experiments described by
Johnson, ££ al^ (1975), SF, and Freon-12B2 gases were injected into four
ventilation ducts at the Millstone Nuclear Power Station. Thirty-six
1-hour releases were conducted in which SF, was injected into the reactor
building main vent. Also, during 26 of the 36 experiments, Freon-12B2
was simultaneously injected into three of the ten vents on the turbine
building. The exit velocities and exit temperatures for the vents,
which varied from trial to trial, are given in Table 3-1. The physical
vent locations and dimensions are given in Table 3-2. For the rectangular
vent on the reactor building, Table 3-2 gives an effective stack diameter
that is the diameter of a circular opening with the same area as the
vent. It should be noted that, in the ISC Model calculations described
in Section 3.1.3, the SF, experiments were modeled separately, and
different coordinate systems were used in the two sets of calculations.
The origin of the coordinate system was placed at the location of the
reactor building vent for the SF, trials, while the origin of the coordi-
nate system was placed at the location of turbine building Vent E for
the Freon-12B2 trials.
The location of the Millstone Nuclear Power Station at the
southern end of a small peninsula restricted the Millstone diffusion
experiments to wind directions from south-southeast through south-
southwest, and the majority of the experiments were conducted with
south-southwest winds. Thus, the turbine building was generally upwind
3-4
-------
OJ
o
~ _J -J
-;=; co
LU
cc
_
o
<
E
O
d
(S>
ID
E
tO
in
O
E
CM
CM
£
O
<
E
CVl
cvi
O
E
OJ
CM
po
W*1
0)
o-i
tlj
O
z
d
s
o
UJ
IT
Z
K-UJ
ID
IT
I-
O
=>m
cc
Cfl
u
c aj
•H M
O rt
a,
01
c -w
o x:
•H 60
en -H
w ai
•H &
S oo
c
u-i 3
O K3
U3
C •
O /""v
•H LO
4J r--
to av
O rH
O ^^
00 0)|
c«
•H •-
S C
o o
x: w
co c
j:
c o
o ^
•H
w >,
IT) ^D •
4J ^^
W) tJ i-J
MOO)
cfl i— (
a) en
i-l 4J T3
o c c
3 0) 3
ss e o
•H )-i
a> l-i c*
C CD
o a. a»
w x >
to cu o
—i j&
•H C «
•H O
S ••-< en
en »-i
0) 3 Q)
,C MH i->
•u u-i a)
o c
OJ -H
3 w
O
cfl o -H
J u-i &C
W
«
^J
O
-------
TABLE 3-1
BUILDING VENT EXIT TEMPERATURES AND VELOCITIES FOR
THE MILLSTONE DIFFUSION EXPERIMENTS
Trial
Number
1
2
3
4
5
6
8
9
10
11
12
13
14
15
16
17
18
28
29
30
31
32
36
45
46
47
48
49
50
51
52
53
54
57
58
59
Reactor Building Main Vent
Exit
Temperature
<°K)
294.2
293.7
292.7
293.0
291.1
291.5
294.5
294.8
294.0
295.9
295.8
296.3
295.4
295.4
296.7
295.9
295.8
295.7
296.0
296.0
29S.9
295.7
294.9
294.0
294.1
294.1
294.1
294.2
294.4
294.9
294.5
294. 1
293.2
293.5
293.5
293.4
Exit
Velocity
(m/sec)
8.7
8.7
8.7
8.7
8.3
8.3
8.1
8.1
8.1
8.7
8.7
8.7
8.7
8.7
8.7
8.6
8.7
6.2
4.6
4.6
4.6
4.6
6.3
8.5
8.5
8.4
8.5
8.4
8.5
8.4
8.6
8.7
8.7
8.6 '
8.5
8.6
Turbine Building Vents
Exit
Temperature
(°K)
293.2
293.7
293.6
293.5
306.1
293.6
293.8
293.5
296.6
296.0
295.3
294.9
294.7
293.9
291.7
291.6
291.6
291.7
291.6
292.6
293.1
293.6
293.1
293.5
292.7
292.7
Exit
Velocity
(m/sec)
10.5
10.5
10.5
10.5
10.5
10.5
10.5
10.5
10.5
10.5
10.5
10.5
10.5
10.5
10.5
10.5
10.5
10.5
10.5
10.5
10.5
10.5
10.5
10.5
10.5
10.5
3-6
-------
TABLE 3-2
PHYSICAL BUILDING VENT PARAMETERS FOR THE
MILLSTONE EXPERIMENTS
Vent
Reactor Building
Turbine Building
Vent A
Vent E
Vent J
Height of Vent
(m AGL)
48.3
29.1
29.1
29.1
Effective Stack
Diameter (m)
2.12
1.40
1.40
1.40
X-Coord *
(m)
0
-10.9
0
13.1
Y-Coord.*
(m)
0
37.5
0
-45.1
*The origin of the calculation grid for the SFfc releases from the reactor
building main vent is at the location of the vent, while the origin of
the calculation grid for the Freon-12B2 releases from the turbine build-
ing vents is at the location of Vent E (see Figure 3-2).
3-7
-------
of the reactor building, which was the dominant building at the Millstone
station during the experiments described by Johnson, ct al. The reactor
building is 44.7 meters high and 47 meters square in the horizontal.
These building dimensions were used in the ISC Model calculations for
all trials, including the Freon-12B2 releases from the turbine building.
It is important to note that the emission height to building
height ratio is less than 1.2 for all four of the vents used in the
Millstone diffusion experiments. Also, the building width to building
height ratio is greater than 1 and less than 5. Thus, the Millstone
experiments provide a test of the building wake effects option for the
special case of a squat building with an emission height to building
height ratio less than 1.2. The modified vertical dispersion coefficient a'
is a function of the building height and the modified lateral dispersion
coefficient a' is a function of the building height and crosswind
width. Because of the horizontal symmetry of the reactor building, the
use by the ISC Model of a single effective building width (defined by
the diameter of a circle with equal horizontal area) does not significantly
affect the accuracy of the model calculations for any trial (see page 2-39
of the ISC Model User's Guide).
3.1.2 Meteorological Data
.lolmson, et a 1. (1975) give the mean wind speed, mean wind
direction, ambient air temperature, and potential temperature gradient
for each of the 36 diffusion trials. Wind speed and direction measurements
are reported for the 10.0-, 43.3-, 114- and 136-meter levels of an onsite
meteorological tower (see Figure 3-1). The winds measured at the 43.3-
meter level were used in the ISC Model calculations as the reference
level mean wind speeds and flow vectors. We also used the wind-speed
measurements at the four tower levels with a logarithmic least-squares
regression technique to calculate a wind-profile exponent p for each
trial. The values of the vertical potential temperature gradient and
3-8
-------
ambient air temperature given by Johnson, et al. were used directly as
ISC Model inputs. (The ambient air temperatures were measured at the 19.5-
meter tower level, while the vertical potential temperature gradients were
based on the temperature differences between the 136- and 10-meter tower
levels.)
We estimated the Pasquill stability category for each trial
using the Turner (1964) criteria, which are based on wind speed and
solar radiation (insolation). Insolation indices were determined for
each trial on the basis of the date and time of the trial and the cloud
cover data reported by Johnson, et al. Cloud cover amount Is given by
Johnson, et al. in tenths of the total sky covered, while cloud height
is reported as low, middle or high. We. assumed low clouds to be at
heights less than 7,000 feet (2,133 meters) and middle and high clouds
to be above 7,000 feet. Wo used the mean wind speed at the 10-meter
tower level during each trial with the corresponding insolation index to
determine the Pasquill stability category for the trial. The 10-meter
wind speeds were used because airport wind measurements, which are used
in the Turner (1964) stability classification scheme, are at that approxi-
mate height. Also, the wind measurement height used in the original
Pasquill (1961) scheme is 10 meters. Table 3-3 gives all of the parameters
that were used to determine the Pasquill stability categories along with
the corresponding stability categories estimated for the various trials.
The neutral D stability category existed during 31 of the '36 trials,
slightly unstable (C stability) conditions existed during one trial, and
stable (E stability) conditions existed during four trials.
The hourly meteorological inputs to the ISC Model short-term
computer program ISCST are listed in Table 3-4 for each of the Millstone
trials. These parameters were developed following the procedures discus-
sed above. The wind speeds in Table 3-4 are the wind speeds at the
43.3-meter tower level, the wind measurement height nearest the emission
height. Consequently, the wind system measurement height Z, was set
3-9
-------
TABLE 3-3
PARAMETERS USED TO DETERMINE PASQUILL STABILITY
CATEGORIES FOR THE MILLSTONE TRIALS
Trial
Number
1
2
3
4
5
6
8
9
10
11
12
13
14
15
16
17
18
28
29
30
31
32
36
45
46
47
48
49
50
51
52
53
54
57
58
59
Date
12 Oct
12 Oct
12 Oct
12 Oct
14 Oct
14 Oct
15 Oct
15 Oct
15 Oct
25 Oct
25 Oct
25 Oct
25 Oct
25 Oct
25 Oct
25 Oct
25 Oct
31 Oct
31 Oct
31 Oct
31 Oct
31 Oct
4 Nov
14 Nov
14 Nov
14 Nov
14 Nov
14 Nov
14 Nov
14 Nov
14 Nov
14 Nov
14 Nov
20 Nov
20 Nov
20 Nov
Time
(EST)
0800
0900
1000
1100
0900
1000
0800
0900
1000
0700
0800
0900
1000
1100
1200
1300
1400
1700
1800
1900
2000
2100
0700
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
0200
0300
0400
Cloud Cover
Amount
(Tenths)
10
10
10
10
7
8
10
10
10
5
5
5
10
10
10
10
10
0
0
5
5
5
10
0
7
7
7
8
8
8
4
4
4
10
10
10
Height*
Low
Low
Low
Low
High
Low
High
High
High
_
-
-
Low
Low
Low
Low
Low
-
-
_
-
-
Low
-
Middle
Middle
Middle
High
High
High
-
-
-
Middle
Middle
Low
Net Radiation
Index
0
0
0
0
2
1
1
1
1
1
1
2
0
0
0
0
0
1
-2
-1
-1
_ i
1
-2
_l
-1
-1
-1
1
2
2
2
2
-1
-1
0
10-Meter
Mean Wind
Speed
(m/sec)
7.2
6.6
6.4
6.9
4.8
4.5
6.8
5.2
6.2
7.9
8.5
8.3
9.5
10J3
10.1
10.2
9.9
3.0
3.7
2.9
3.1
3.1
3.7
7.2
8.1
9.5
9.2
9.5
9.8
10.3
11.2
10.9
10.6
3.6
3.6
3.4
Pasquill
Stability
Category
D
D
D
D
C
D
D
D
D
D
D
D
D
D
D
D
D
D
E
E
E
F.
n
D
D
D
D
u
D
n
D
D
I)
D
D
D
*Heights were not reported for amounts less than or equal to five tenths.
3-10
-------
TABLE 3-4
METEOROLOGICAL INPUT PARAMETERS FOR THE MILLSTONE TRIALS
Trial
Number
1
2
3
4
5
6
8
9
10
11
12
13
14
15
16
17
18
28
29
30
31
32
36
45
46
47
48
49
50
51
52
53
54
57
58
59
-i
Average
Flow
Vector*
(deg)
025
029
028
034
327
329
031
049
052
016
Oil
017
022
024
017
023
027
022
034
036
043
047
048
018
008
023
016
Oil
012
009
006
004
003
355
332
330
43.3-Meter
Mean Wind
Speed
(m/sec)
7.9
7.3
7.2
7.9
6.5
6.0
9.3
7.6
8.7
8.6
9.2
9.2
10.9
11.8
11.5
11.8
11.5
3.8
4.6
3.4
3.5
3.7
5.2
7.7
8.7
10.4
10.2
10.4
10.8
11.4
12.6
12.5
12.5
4.4
4.1
4.1
Ambient
Air
Temperature
(°K)
287
288
288
289
285
286
290
291
291
284
284
285
286
286
286
286
286
287
286
286
286
286
287
282
283
283
284
284
284
285
285
285
285
284
284
284
Wind
Profile
Exponent
0.053
0.063
0.064
0.074
0.130
0.099
0.246
0.284
0.279
0.052
0.057
0.048
0.067
0.067
0.062
0.078
0.090
0.201
0.188
0.124
0.171
0.234
0.289
0.050
0.056
0.050
0.049
0.044
0.051
0.057
0.058
0.062
0.076
0.142
0.107
0.151
Potential
Temperature
Gradient
(°K/m)
0.000
0.000
0.000
0.000
0.000
0.000
0.007
0.008
0.010
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.014
0.009
0 . 003
0.006
0.008
0.003
0.000
0.000
0.000
0.000
0.000
0.000
0 . 000
0.000
0.000
0.000
0.000
0.000
0.000
1
st
c
1
i
Pasf'ui1i
Stability
Category
D
D
D
D
C
D
I)
D
1)
I)
D
D
D
D
D
D
D
D
E
E
E
E
I)
I)
D
D
I)
I)
I)
D
D
D
D
D
D
D
*Direction toward which the wind is blowing.
3-11
-------
equal to 43.3 meters in the ISC Model calculations. Table 3-4 does not
include any mixing height estimates. Because of the low emission heights,
the relatively short distances to the sampling arcs and the meteorological
conditions during the various trials, it is unlikely that the restriction
on vertical mixing at the top of the surface mixing layer significantly
affected the observed concentrations during any of the trials. We
therefore assumed unrestricted vertical mixing in the ISC Model calculations
for all of the trials.
3.1.3 Calculation Procedures and Results
The source data discussed in Section 3.1.1 and the meteorologi-
cal data discussed in Section 3.1.2 were used with the TSC Model short-
term computer program TSCST in the Rural Mode to calculate, for each
Millstone diffusion experiment, the centerline tracer concentration x divided
by the emission rate Q at the downwind distance of each sampling arc.
Concentrations were calculated without any building wake effects or dis-
tance-dependent plume rise and with building wake effects for the fol-
lowing combinations of ISC Model options:
• Case A — Distance-dependent plume rise and building
wake effects options
• Case B — Distance-dependent plume rise, building
wake effects and stack-tip downv/ash options
Additionally, concentrations were calculated under the assumption of surface-
based emissions affected by building wakes (Case C) because of visual obser-
vations by Johnson, et al. (1975) that the plume often was intermittently
entrained into the building's cavity zone and effectively acted as a ground-
level source during these periods. All of the calculated centerline concen-
trations were multiplied by the mean wind speed u at 43.3 meters to obtain
normalized centerline concentrations Xu/Q, the form in which the results of
the Millstone experiments are presented by Johnson, et al.
3-12
-------
The calculated and observed normalized centerline concentra-
tions for the SF, trials at downwind distances of 350, 800 and 1500
o
meters are listed in Tables 3-5, 3-6 and 3-7, respectively. Similarly,
the calculated and observed normalized centerline concentrations for tho
Freon trials at the three sampling distances are listed in Tables 3-8,
3-9 and 3-10. We point out that our comparisons of the normalized
centerline concentrations given in Tables 6 and 7 of the report by
Johnson, et^ a^. with the detailed results presented in Appendix C of
their report revealed several typographical errors in their Table 6.
Thus, the observed concentrations in Tables 3-5 through 3-7 of this
report differ In some cases from the corresponding concentrations given
in Table 6 of the Johnson, et^ _al_. report.
The majority of the Millstone experiments were conducted under
very similar meteorological conditions (moderate or strong south-southwest
winds in combination with neutral stability). Thus, the TSC Model used
unmodified (no building wake effects) or modified (with building wake
effects) Pasquill-Gifford dispersion coefficients for the neutral 1)
Pasquill stability category in the majority of the concentration calcu-
lations for the Millstone experiments. Because each of the calculated
centerline concentrations was multiplied by the mean wind speed and
divided by the emission rate for comparison with the corresponding
observed normalized centerline concentration given by Johnson, e_t al. ,
the calculated normalized concentrations arc essentially the same for
the majority of trials. The turbulent intensities associated with the
Pasquill-Gifford dispersion coefficients are the median values for the
various Pasquill stability categories and do not necessarily correspond
to the actual turbulent intensities during any of the Millstone ex-
periments. Consequently, the trial-to-trial differences in observed
normalized concentrations are larger than the trial-to-trial differences
In calculated normalized concentrations.
3-13
-------
o
Pn
to
O
i
H
W
I
W O
S! Q
M
W W
H H
•Z W
o
O 10
td ro
SI
M H
i-l d 13
CU X
N
•rl
iH
O
r-t
Cfl
•H
H
C
•r-
T
•r
PC
rC
15
5
•o
Calculatt
u
Q)
"e
3
CJ
at
en
co
CJ
:
i PQ
1 (U
, (A
) CO
) O
1
0)
CO
0
00
c
•H
•a
5 g
o
a)
to
cfl
00 O
c
•H
g d)
PQ co
>C CJ
CU
u
No
Building
Observed
T, oo sr N oo r, co ~, oo m^c^cN^^^^oo^
OOO>— ir-H-— lOOO OOOOOOOO'-^'— '
^ss^ss^^^ ^ssssssss
OOOOOOOOO OOOOOOOOOO
55^2^5:^5^ s^s^s^ss^^
OOOOOOOOO OOOOOOOOOO
OOOOOOOOOO OOOOOOOOOO
OOOOC^OOOO OOOOOOOOOO
OOOOOOOOO OOOOOOOOOO
OOOcN4f—
-------
13
<1)
C
O
o
I
ro
w
•d
0)
E:
CU
CO
J3
O CO
pj
T) O
CJ •rl
<4 4-J
rt
•O to
CD 4->
w «3
« 0)
iH O
3 C
o o
rH 0
Cfl
o
U-l
0
o
•H
4J
5
•H
4J
rt
Vl
4-1
(3
CU
o
c
o
O s~*
CN
VI 1
CD O
4J ^-1
d ^
r^^.
Normalized (
xs/c
T-t
(0
•rl
M
H
c
•r
a
r-
•r
p
j:
^
D
1
•a
CU
4J
rt
rH
jj
^1
id
o
Number
3
3
^
3
J
H
E
*>
O r
z •
(
(
(
T
r-
•r
p
4
4
£
1
Observed
y
H
d
H
H
«
W
-1
3
H
3
-1
•4
J
^
C
o
5
u
0)
en
n)
O
PQ
•— i 1 — in O> u~> LT> r-( vO v£> CT\ O
r^-Nfr^-oro^-i— i(Ncsicvi oicscMcNroinm
OOO-HOOOOOO OOOO'-I'-ICN
OOOOOOOOOO OOOOOOO
OOOOOOOOOO OOOOOOO
OOOOOOOOOO OOOOOOO
CMCNCNfNCNCMCNCSJCMCN CSICSICNCS--ICN-H
uio>r^cNo>in— <^-ICSICNI v£)
cs^oc^r-»CT\CT\forn
(1)
(0
x;
ca
6
a>
C
•H
C
(D
O
01
&
3-15
-------
vO
I
o")
23
PH
•o
CJ
u
U-l
o
o
4-1
ration
2 Concent:
-2)
c s
•H
rHvO
0) O
4-1 r-l
r
z •
(
1
r
•
P
J
4
t
•a
(U
cu
to
a
H
a
H
3
y
3
3
^
q
*
J
£
25
CJ
CO
tfl
U
PQ
CD
cfl
U
CU
co
(4
O
CJ
01
W
CJ
PQ
0)
CO
CO
CJ
W
Building
VOOOCNOO'— ICTV'— l'— IOO
OOr— IO-— I »— 1 CM CO •— t OOOOOOOOc^JO
S^OCTJ^Ioo^-cNOO^i f^^i^imcoro^^oo-l
OOOOO-— ir- {--to OOOOOOOOr-nO
CO 00 CD ^O LO fO ^>O CD ^O ^D 00 CO I*^» ^D CD O*^ ^^ ^O fO
•sj iO O^ ^O CO "s^1 *™H CD ^O fO CO CO CO CO CO C^J -^f1 C5 CO
OOOOOi-ti-ltNO OOOOOOOO--HO
2^^SSS2^^ ^2^222^S^g
ooooooooo oooooooooo
oooooooo
-------
0)
3
c
o
I
CO
w
"O
cu
0)
M
o co
c
T3 O
C -H
el 4-1
CU 4J
jj g
(S (U
rH O
3 C
O O
r-{ O
U
U-l
o
o
•H
4J
§
•H
4J
rt
i Concent
-2)
rl 1
§•§
i
60
1
Z -rl
3
•H
-o
r-4
•H
M
J3
4J
S
O
TJ
CO
0)
o
CJ
0)
CO
CJ
PQ
0)
en
o
0)
co
ct>
CJ
0)
co
rt
CJ
<1>
CO
rt
CJ
CU
CO
.3
Building
oo«—isti-^\oONCr>cor~-'— t co CT> •— i •— i t~» vD O
_, r- 1 O — IOOOOOO OOOO-HrHCO
in ^O co r^ co co co co co co ^l* co co co ^^ O *~^
OOOOOOOOOO O O O O O •— ( CN
cNO-oococMcnoo^ ZSZZSSS
OOOOOOOOOO OOOOOr-tCM
OOOOOOOOOO OOOOOOO
OOOoooooooooooooo oooooooooooor-^
i — 1 ' — 1 i-H
S3SSSSSRKK SSRS^SS
co *j" in co ^o io ^n m in in m tn un in in in *^"
OOO— i— imi~-.oooooo cr>o-— ic^vncom
-k
SSSS3SS35S ^SKSKSS
0
5
o
VO
0
00
1 — t
o
\
\
o
c
OJ
j
1
•H
IU
M)
a.
E
O
c
V
-------
W
U C/)
to
hJ
w X
en
§
o
u
*a
I
cu
co
,g
O CO
•o o
s *
OjJ V*
•O >-i
CU i->
AJ d
ctf 0)
iH O
3rt
M
3
0
o
•H
pjj
G
O
•H
4J
cd
4J
a
d)
u
c
o
o ^
41 V
Jj3
M 1
41 O
0 ^
•3-x
N
•H
^
a
{4
O
co
>r4
H
£
C
•p
T
•r
PC
j:
^
5
5
2
n)
*"H
3
U
fH
U
M
CU
*
iD
3
*
J
4
4.
3
3
*
j
•1
E
60
C
•H
T3
Ol
1^
5 *H
3
«
W)
C
•H
•o
rH
•H
3
PQ
.c
4J
*
O
ss
cu
^
0)
CO
0
cu
CO
(0
o
pq
cu
en
.3
•^
O ' — 'O O <— t ' — 1' — *OO" — IO' — ICN
I
1
SoS^tCStrlScSS ^OToS^ScSSoO
C? O '•H O *~^ cNjasO£^p-)-icsn --i^-o-ivOLnr^-ir-vc-J-
oomrocoiriiriinaNvi ioooot^.-*-d-oovooo
OO-— (Or-H-.-tOOO OO'- c*o "*^ ^^ **^* oo ^^ r^*- *^* ^^ oo ^^ ^o *"~i
OO^nOrHcOOOO OO--HOOOOOOO
"•d" ~^r-r^ loorornf-ncnromior^
rOrocoro^-ifOrNjrviCs) roromcorommrorvioj
cOrococ'S^NrO'— ( *— ICN) rOooonrornrororO'— *
•K
*— H p**- iO >^1 '"^ 00 c\l ^^ CNJ o^ ^«O CNJ "v^ co c^i i— ^ O^ -^ vO
*sf ^O CN| ^^ ^~^ LO CO **O vO CO f*^ **cT r^ r^* *«^" ^-^ csi CNJ
_H ^_1r-H^Hr_l,— (^-Jr-Jr-HCNJC^)
c
o
CO
c
CU
CD
(fi
d)
c
•H
r-l
cu
o
CU
0
•H
a
a
to
a
•H
<*-!
•H
U5
CO
3
•a
4)
cu
CO
cn
co
B
I
cu
J-J
0)
c
o
•H
4-t
cO -^^
a co
o -J-
CO
•x
J-18
-------
-a
0)
C!
•H
I
m
w
•o
C
T3 O
C -H
M
alated
acentr.
0 O
rH CJ
c
CO
Building
r— inoomr~.r— oooOoOO rv i-^ p- co -cf m — i
CM-.-.OJOOOOO- OOOO^OOVO
C) co c**4 I-O LO *"""' *«o co c? CM f~*"* ^^ r"""*1 f~~^ co QO ^"^
C~) r*1^ O^ <~ H i-O iO vO ^O ^D CO ^O ^O LO ^O C3 *^" T~~*
T— 4 (^ C^ r-n C5 ^3 C^ ^3 d3 C^ ^^ CD C5 C3 *~^ CM ^J"
00 CNI o^ r-H oO C5 -^" ••— i oo CT* LO r*N. LO O^ rn 00 ro
OOOt-*OOOOOO OOOOr-HCM-^-
ssssssasss sssssss
OOOOOOOOOO OOOOO-— ICN
3S33S3SS33 3333333
SSSSSSSSSS SSSSSSS
CO 00 CD r"~| "Nt" r*^ CO CO < — I CN •— t t— i csl CN| O ^D CO
ro co co co co co co ro co co CM csj CNJ
•K
CN| CO CO •""! >O **iD LO l^ LO "
cn
B
i
•H
X
03
e
c
o
•H
4-J
cd
O /—N
O OO
M DH
cu
a. m
B r-
Cd ON
C/: rH
•x
3-19
-------
oo
I
ro
W
PC!
g
CO
z
o
M
H
2
H
U 3
Z Z
o &
CJ O
u
Z co
M 35
r-J W
H
@
w o
o m
CO
P
W H
N <3
M
,-4 CO
^S
W O g
J Z M
PP «
Z
U O
d ^
< fK!
CJ C
Q W
r2j Z
-^ o
H
a co
,_)
> !-J
8£
CO
§
fn
O
§
CO
O
CJ
T3
u
cu
O CO
•a o
C -H
ft u
n)
CU 4J
•W C!
to a)
r-< CJ
3 C
o
<4H
o
o
J-l
G
o
•rl
0)
4-)
C
a)
CJ
C
o
O '—>
CM
IJf
cu o
4J r-l
o o-
Normalized
Xu/
00
•rH
H
C
C
•r
T:
^"
T
ff
rC
•r
5
•U
CD
4-1
cd
rH
3
U
tfl
O
cu
1
uU
^
3
^
3
}
^
j
j
r> r
z •
(
t
!
T
r
•f
P
4
4
i
Observed
60
C
rl
a
-H
H
q
?
.j
3
H
H
3
q
j
H
E
O
z;
u
0)
CO
n)
pq
cu
en
n)
u
0)
CO
cfl
0
a)
CO
rt
o
pa
Q)
to
CO
o
cu
in
Building
o-iaoaNi^oc^ini^mo ^vnoo^mcNr-oom^
OOOOOOOOOO •— ir— it— lOOOOOOO
3£S-S~-P!$S 3S£3SS3£SS
OOOOOOOOOO OOOOOOOOOO
c^ .0 c* CN ^ ,H ^ c* <* 5££32S3;?;S£
OOOOOOOOOO OOOOOOOOOO
OOOOOOOOOO OOOOOOOOOO
o
-------
•d
v
a
•r!
4-1
C
O
u
OO
I
CO
w
•d
at
*>
H
01
J3
O M
C
•d o
C "'"i
(4 4J
Cl)
u_l • .
^O w
01 4-1
ri) 0)
•H O
3 C
"3 "
U
4-(
O
0
•H
PQ
c
O
•H
4J
ctt
>-l
4J
c
01
o
a
0
U /-s
CN
01 1
a s
•H
HvD
U 1
01 O
a C
3a<
•sS
N
•H
rH
(rt
g
W
o
55
<0
•H
V-l
H
a
q
•r-
"C
r—
*f
PC
f
4-
*r"
-d
a)
4-1
n)
M
u
tH
<^* "™H
ro ro 0s* f^ c-J oo
o^ oo r*- oo ^o oo
0 0 O O O O
co "-"i r*^ 'O c^J co
&\ CQ r^- cQ \o oo
o o o o o o
vO p"^ vD CJN CO v^
CO CO CO CO *"H C)
o o o o o o
CTI oo oo oo oo <—t
I— t t— 1 .-H r— 1 r-t CN
«^ O co
jo in uo uo
•*-t
'"^*
o
00
-^
O
r--
-*
O
N^~
i-H
O
/^^s
S
S_ X
o
•H
-U
&4
EJ
cfl
0)
3-21
-------
0>
en
O CO
C
TJ O
§ t!
OS
0> 4-)
4J p>
n) cu
rH O
o o
rH O
M-l
O
O
i-1
O
•rl
4J
CO
i Concenti
-2)
aj1
01 O
4J r-H
cu
tl l~*
Normalized (
Xu/(
rH
CO
•rl
H
c
(.
•r
t
r-
*•
pZ
,£
•r
3
;
Calculate
IH
CU
2
u
CU
CO
rt
U
] «
•1 0)
< CO
' cfl
5 0
4
1
CU
10
It)
o
oo
C
•H
T3
O rH
3 lH
O
0)
CO
CO
60 O
rH RQ
3 (1)
PQ en
cfl
X 0
4J
Sc
01
w
No
Building
'O
o
>
>-i
0)
CO
^1
O
0 oo i^ ^ o oio CMO vo CN oo oo rn r^ in c*
OOOOOOOOr- 10 i— I i— i i— IOOOOOO--H
S^2^^3S^2S S£S£££3S3£
OOOOOOOOrlO OOOOOOOOOO
oo f~H *v3~ r~4 f-n csj cr\ o*j o^ ^D ^.O o^ r~^- ~^T ^o LO o^ f^ o^ co
i ^ vrf ^O ^O **^ ••^ 0*4 tO ^~^ ^O V*O LO *^O ' O lO LO ~^ ^O -^ O*^
OOOOOOOOrHO OOOOOOOOOO
^Sin°S^5^5o §oS2°PSc2SS
OOOOOOOOOO OOOOOOOOO--H
oooooooocxDoooooooocTi c^cno\c»oooooooooOcxi
SSSSSSSSSS SSK3SSSSSS
SS2SSSSSISS SSR3SSSSSS
oo ^^ *~^ fO r*** I-O t^"\ ~^ ^^ "**^" c^ "^ c*^ *»si" LO «~^ *"~i co to *-o
00 O^ ^^ O^ 00 O*^ O^ O^ CO t~~^ CO O^ O^ O"^ O^ O^
•K
•r-t
VJ
W)
(50
d
•H
r-H
(X
to
0)
U-J
H-l
O
c
ai
ai
c
c
0)
o
at
0
rH
Ol
-------
•n
ai
o
o
i
m
W
J
5
H
T3
0)
b»
CU
CO
rQ
O W
C
13 O
C *ri
«l -*J
T1 LJ
W w
CU 4J
4J fj
cd cu
i-H O
3—4
c
o
o
•H
4J
frt
p2
e
o
•H
4-1
«)
>-i
4J
C
CU
CJ
c
o
O /-^
c-vi
13
CU X
N
•H
r-l
3
o
,—f
ct)
•H
M
.H
0
e
•i-
t
r—
PC
4:
^-
5
X)
cu
4-1
CTJ
M
3
o
*-{
m
cu
CO
U3
O
" I
•—I ON iA CM O •— <
OO in m vO ON -H
O O O O O
ON r^ vO r^« oo oo
00000--!
O «— i ^ O r-- oo
oo oo oo oo r^ r^
ON '— ^ CN "—4 in r*^ r^« *-o vo
oo o •— i o m
-------
W
&
O
-T3
01
t
0)
CO
O w
C
T3 O
C -rl
-< i
a) o
4J r-H
C! ^
A ~
lormalized (
Xu/C
;o
•H
<0
•H
t-l
H
t
C
•r
i:
-r
f
£
*r
5
t-
T3
Calculate
Number
1
f
3
•1
3
4
^
E
0 r
B •
(
1
T
•f
P
rl
4
C
|
Observed
00
a
H
O
^
H
3
4
2P
3
i
•4
J
I
O
z
c_>
a)
a>
00
O
w
d)
w
rt
O
<
0)
0)
cd
0
o
01
w
rt
o
pq
(U
(0
A
U
<
01
U)
.3
Building
OrOvOOfO^HOOOOOOOO cMOCT\r^rOvC>OOO<^00
voo>cMvooov£)u~)00
OO--1OOOOOCN.-I OOOOOOOOOO
ooroi— inmr-cNioofO^o oocr>-^vOir\\Ov£iLrii^
OO-HOOOOOCN'-H OOOOOOOOOO
tnoocT\r--OO'JDr^moo
i^t-HLor^OLOr^cMCTsvo r— i(>j~a'r^-'— <
O_l,-iO'-iOO'— ICMO OOOOOOOOOi— i
CsirO-*oOi— iocx)Oooaooo
romro-d-^i-Lri'-or-~sOior~-<±
^-H
,— (CMro-^u^i^^oooOOs o--
o
a>
e
a
rt
a
•H
<4-l
•H
oo
•H
05
rt
13
0)
>
(-4
a)
CO
-3
X
0) ^
J2 OO
S -*
C •
o P.
0)
o
O
1-1 t-l
CX
-------
T)
cu
3
C
C
O
U
O
•—I
I
w
0)
fc
0)
CO
O co
a
•o o
C -H
nj 4-1
rt
01 4->
w £
rt X
N
•H
.-1
3
£j
O
r-j
n)
•H
n
H
b
C
•r-
•C
I— •
•i-
j:
4.
•r-
3
t-
*t3
a>
4-1
n)
rH
3
O
a!
0
a>
rO
0
^
U
0)
CO
CO
0
0
•
«
J
4 -H •— 1
cs en ro ro ro ro
*^ ^s] ^T ~-J ^J
co ^O (^ O ON
LO u~> u"^ uO 10 in to in
00
o
00
*•£}
-------
Table 3-11 summarizes the comparisons of the concurrent calcu-
lated and observed maximum centerline concentrations for the Millstone
diffusion experiments. Because the equation for the root mean square
error (RMSE) given in Section 1.3 cannot be applied to zero concentrations,
the normalized centerline concentration was set equal to 1x10 per
square meter for the cases with zero calculated concentrations in order
to estimate the RMSE. As shown by Table 3-11, if the building wake
effects option is not exercised (equivalent to current modeling techniques) ,
the mean ratio (MR) of calculated to observed concentrations indicates that
the model, on the average, underestimates the concentrations observed in
the building wake region (i.e., at the 350-meter sampling arc) by an order
of magnitude or more. The model's performance, as indicated by both the
MR and the RMSE, is significantly improved if building wake effects are
included in the model calculations for Cases A and B (the elevated releases).
However, the MR indicates that the model has a systematic tendency to
underestimate concentrations at all three sampling arcs for Cases A and B.
The assumption of surface-based emissions affected by building wakes (Case C)
yields the best overall model performance.
In addition to emission rate, wind speed and mixing height, the
basic components of a Gaussian plume model are the effective emission height
and the lateral and vertical dispersion coefficients. The emission rates
and wind speeds arc known with a relatively high degree of confluence for
Millstone diffusion experiments, and the restriction on vertical mixing at
the top of the surface mixing layer is unlikely to have affected the ob-
served concentrations. Thus, the fact that the ISC Model tends to under-
predict concentrations near the source for Cases A and B may be attributed
to one or more of the following factors:
• The ISC Model may not accurately describe lateral
plume spread (o )
• The ISC Model may not accurately describe vertical
plume spread (a )
z
5-26
-------
TABLE 3-11
SUMMARY OF THE RESULTS OF THE ISC MODEL CALCULATIONS FOR THE
MILLSTONE DIFFUSION EXPERIMENTS (JOHNSON, ET AL., 1975)
Sampling Distance (m)
350
800 1500
(a) Mean Ratio (MR) of Calculated to Observed Concentrations
SF *
D
No Buildings
Buildings: Case A
Buildings: Case B
Buildings: Case C
Freon-12B2**
No Buildings
Buildings: Case A
Buildings: Case B
Buildings : Case C
0.00
0.33
0.33
0.66
0.14
0.47
0.48
0.74
0.18
0.46
0.46
0.74
0.62
0.55
0.55
0.72
0.62
0.73
0.74
1.09
0.84
0.68
0.68
0.83
(b) Root Mean Square Error (RMSE)
SF6*
No Buildings
Buildings: Case A
Buildings: Case B
Buildings: Case C
Freon-12B2**
No Buildings
Buildings: Case A
Buildings: Case B
Ru lid ings: Case C
259.90
3.17
3.09
2.06
19.76
2.20
2.12
1 .53
9.38
2.31
2.28
1.83
5.07
1.89
1.88
1 .64
2.88
1.79
1.78
1.88
1.94
1.69
1.68
1.64
(<•) Percent Within a Factor of 2
No Buildings
Buildings: Case A
Buildings: Case B
Buildings: Case C
Freon-12B2**
No Buildings
Buildings: Case A
Buildings: Case B
Buildings: Case C
0
31
28
56
0
38
38
85
8
42
42
86
77
73
73
88
r "
67
83
78
83
81
88
88
88
. L
*The sample size is 36.
**The sample size is 26.
3-27
-------
• The ISC Model may not accurately predict the effective
emission height (H)
Of the three possible sources of error, lateral plume spread is the sim-
plest to evaluate because o may be estimated from the data given by
Johnson, et al. The other two possible sources of error are far more
difficult to evaluate because no vertical concentration profiles are
available for the Millstone experiments. Additionally, because the H/a
z
ratio appears in the Vertical Term of the concentration equation, no
direct procedure for the estimation of 0 is known if H is unknown, and
z
vice versa.
Johnson, et a1. used a nonlinear least-square regression tech-
nique to fit a Gaussian-shaped curve to the observed concentrations at
each of the three sampling arcs for each trial. Although the fitted
curves are shown in Appendix C of the report by Johnson, et al., the
fitted a values are not listed in their report. We contacted Johnson
y
to obtain the fitted 0 values. However, the fitted a values had not
y y
been located at the time of the preparation of this report. Consequently,
we graphically estimated the a value at each sampling distance for each
SF, trial using the definition for a Gaussian distribution that the
distance between the locations on a Gaussian curve of concentrations
equal to 10 percent of the centerline concentration is 4.3 a . (We did
not attempt to estimate O values for the Freon trials because these
trials involved simultaneous releases from multiple vents.)
Table 3-12 compares the calculated a values for the neutral D
and slightly unstable C stability categories with the mean observed
(fitted) a values for the trials conducted under conditions of D stab-
ility, the stability category associated with about 89 percent of all
trials. We point out that the o values calculated for C stability in
the Rural Mode in Table 3-12 also correspond to the a values calculated
for D stability in Urban Mode 2. Also, the calculated a values at 350
y
meters are the same for the D and C stability categories because the
3-28
-------
TABLE 3-12
COMPARISONS OF CALCULATED Oy VALUES AND OBSERVED MEAN CTy VALUES
FOR THE SF6 TRIALS CONDUCTED UNDER NEUTRAL
(D STABILITY) CONDITIONS
Downwind
Distance
(m)
350
800
1500
Mean
Observed
ay
(m)
44.5
87.5
139.5
Calculated a (m)
D
Stability
33.0
63.6
106.7
C
Stability
33.0
84.2
149.8
Calculated ay
Observed a
D
Stability
0.74
0.73
0.76
C
Stability
0.74
0.96
1.07
3-29
-------
modified dispersion coefficients calculated by the ISC Model within the
wake region are independent of stability. For the effective crosswind
building width h of 53 meters used in the ISC Model calculations, the
model underestimates lateral plume spread by about 25 percent at all
downwind distances if turbulent intensities corresponding to the D
stability category are assumed to have occurred during the trials. If
turbulent intensities corresponding to the C stability category are
assumed to have occurred during the trials, the model underestimates o
by about 25 percent at 350 meters and closely matches the observed mean
O values at 800 and 1500 meters. It Is also possible to solve for the
value of h that forces the calculated (J in the Rural Mode under D
W y
stability to match the observed mean O value at each sampling arc.
This value ranges from about 86 meters at 350 meters downwind to about
153 meters at 1500 meters downwind.
Our comparison of calculated and observed a values for the
y
SF, trials may be summarized as follows:
Because the O values used in the model calculations
are about 25 percent lower than the corresponding mean
observed O values at the three sampling arcs, the
ISC Model's tendency to underestimate concentrations
near the source cannot be attributed to overestima-
tiona of lateral plume spread
The calculated a values can be forced to match the
observed a values by varying the effective crosswind
building width h , but no single value of h yields
calculated a values that match the mean observed a
y y
values at all downwind distances
We conclude from these results that the discrepancies between calculated
and observed centerline concentrations near the source are primarily
3-30
-------
attributable to errors in the calculated plume heights H and/or in the
calculated vertical dispersion coefficients a •
2
Figure 3-3 illustrates the visual observations of plume behavior
reported by Johnson, ejt aj.. for the Millstone diffusion experiments.
When the exit velocity V to mean wind speed u ratio was less than or
s
equal to about 0.9 (Mode I), complete entrainment of the plume into the
building's cavity zone appeared to occur with the plume effectively
acting as a ground-level source. On the other hand, when the V /u
ratio was greater than or equal to about 5 (Mode II), the plume remained
elevated and appeared to be unaffected by downwash into the cavity zone.
However, most of the time the plume appeared to be in a condition of
partial entrainment (Mode III) in which portions of the plume would
intermittently become entrained into the cavity zone. The prevalent
mode during the Millstone experiments was Mode III. It follows from
Figure 3-3 that the hourly average effective emission height during most
of the trials probably was intermediate between the release height and
the ground surface.
As discussed above, the majority of the Millstone diffusion experi-
ments were conducted under very similar meteorological conditions (moderate
or strong south-southwest winds in combination with the neutral Pasquill
D stability category). Also, the upwind fetch was over water. Thus, the
Millstone data set may represent a unique combination of source and meteoro-
logical parameters. Additionally, the turbulent intensities, especially
the vertical turbulent intensities, measured during the Millstone experi-
ments generally were smaller than the median turbulent intensities implicit
in the Pasquill-Gifford dispersion coefficients used by the ISC Model. Be-
cause vertical plume expansion is reduced if these smaller vertical turbu-
lent intensities apply, site-specific turbulent intensities may account
for the ISC Model's tendency to underestimate centerline concentrations
when surface-based emissions are assumed (Case C).
3-31
-------
1
Z U4
UJ U
2 IT
_ uj o b
sill
UJ
K- O
2 CC
UJ Z>
2 O
2 UJ >
h-
z
5 2 WO
c
o
en
c
.c
o
13
OJ
1-1
QJ
0)
fl)
£
cu
e
r-t -H
a. M
o x
at
tn
01 C
a. o
>. •--*
-U «)
3
o
•3-12
-------
3.2 TSLITZER (1965) TRACER RELEASES
Islitzer (1965) presents the results of tracer tests conducted
at the Materials Test Reactor Engineering Test Reactor (MTR-ETR) Complex
of the National Reactor Testing Station in Idaho. Uranine dye in solution,
which was used as the tracer, was released from near ground-level at the
downwind edge of the main reactor building, which is shown in Figure 3-4.
High volume (hi-vol) air samplers were placed at four downwind distances
ranging from the wake region (nominal downwind distance of 118 meters) to
850 meters from the tracer source. The small filled circles in Figure
3-4 show the locations of the hi-vol samplers and the filled square
shows the release point. The results of the tracer tests given by
Islitzer include estimates of the normalized centerline concentration x/Q
for each test at each sampling distance and the lateral plume spread a
at each of the first three sampling distances. No a estimates are
given for the 850-meter sampling distance because the 850-meter sampler
array was too narrow to allow accurate estimates of plume width. The
source data for the Islitzer tracer releases are discussed in Section
3.2.1, the meteorological data are given in Section 3.2.2, and the ISC
Model calculation procedures and results are described in Section 3.2.3.
3.2.1 Source Data
The tracer source for the Islitzer (1965) tests was a non-
buoyant, ground-level point source with only minimal vertical momentum.
Excluding building dimensions, the only ISC Model source input parameter
required for a surface-based, non-buoyant source is the emission rate.
However, it is not necessary to know the actual emission rates for the
Islitzer tests because the tracer concentrations are reported in the
normalized form X/Q» where X is the estimated centerline concentration
In grams per cubic meter and Q is the tracer emission rate in grams per
second.
3-33
-------
350 m«
•3
•4
N
• 850m
118m
D
L_nJ
• METEOROLOGICAL STATION
• RELEASE POINT
• HI-VOLUME SAMPLER
• 550m
100 50 0 100 200 M
3 1
FIGURE 3-4. Experimental layout for the IslJtzer (1965) diffusion experiments.
-------
AM shown by Figure 3-4, the tracer release point was immedi-
ately downwind of the main MTR building, which is 24.4 meters high and
45 meters square in the horizontal. The remainder of the MTR-ETR Complex
consists of a number of smaller buildings with heights ranging from 6 to
18 meters. (No elevations are given by Islitzer for the individual
buildings other than the MTR building.) If the aerodynamic wakes and
eddies generated by the MTR building are assumed for modeling purposes
to have had the primary effect on the initial dispersion of the tracer,
the ISC Model will modify both the vertical and lateral dispersion
coefficients for the Islitzer tests. The modified vertical dispersion
coefficient a' is a function of the MTR building height and the modified
Z
lateral dispersion coefficient 0' is a function of the MTR building
height and crosswind width. On the other hand, if the aerodynamic wakes
and eddies generated by the entire MTR-ETR Complex are assumed to have
affected the initial dispersion of the tracer, Figure 3-4 indicates
that the effective crosswind width h is more than 5 times the MTR
w
building height h . The ISC Model uses either Equation (2-31) or Equation
(2-33) in the ISC Model User's Guide to calculate the modified lateral
dispersion coefficient when the h /h, ratio exceeds 5. As discussed on
page 2-38 of the ISC Model User's Guide, Equation (2-33) generally is
considered to be more appropriate than Equation (2-31). Consequently,
we used Equation (2-33) in the set of ISC Model calculations which
considered the effects of the building wakes of the entire MTR-ETR
Complex. The modified lateral dispersion coefficient o' defined by
Equation (2-33) is a function of the building height, but not of the
building width.
3.2.2 Meteorological Data
Wind speed and wind direction were measured during the Islitzer
(1965) tests at the four locations downwind from the release point shown
by the large filled circles in Figure 3-4. In addition to wind speed
and direction, the standard deviations of horizontal (a.) and vertical
A
-------
(o ) wind (11 ructions were measured. No measurements were made of the
h
undisturbed winds upwind of the MTR-ETR Complex.
Measurements of mixing height are not reported by Islitzer.
However, because of the relatively short sampling distances and because
of the presence of the MTR-ETR Complex upwind of the test area, we
assumed that the surface mixing layer was sufficiently deep that reflection
from the top of the mixing layer did not significantly affect the measured
ground-level tracer concentrations.
The meteorological data given by Islitzer are not sufficient
to determine the Pasquill stability categories for the tests using the
Turner (1964) method, the stability classification scheme routinely used
with the ISC Model. Specifically, insolation data, the times of the
tests and the cloud cover data for the tests are not provided. However,
measurements of two other meteorological parameters, the vertical tempera-
ture gradient (AT) and the standard deviation of the wind direction
angle (aA), can be used to estimate the Pasquill stability categories
A
following criteria established by the Atomic Energy Commission (1972).
The AEC criteria for determining stab:
measurements are given in Table 3-13.
The AEC criteria for determining stability categories from AT and a.
A
The meteorological inputs for the Islitzer tracer tests are
given in Table 3-14. As shown by the table, the Pasquill stability
categories determined for a given test by the AT and a. methods differ
A
by as many as six stability categories. Therefore, ISC Model calculations
were performed for both sets of stability categories. We point out that,
with the exception of Test 7, the very unstable A Pasquill stability
category is indicated for all tests by one or both of the stability
classification schemes. Thus, it is entirely possible that A stability
existed during all tests except Test 7. The mean wind speed at the
emission height during each test is also given in Table 3-14. The other
ISC Model meteorological input parameters are not required to model the
Islitzer tests. For example, the ambient air temperature and the vertical
3-36
-------
TABLE 3-13
ATOMIC ENERGY COMMISSION (1972) PASQUILL
STABILITY CLASSIFICATION CRITERIA
Pasquill
Stability
Category
A
B
C
D
E
F
G
Standard Deviation
of Horizontal Wind Direction*
% (deg)
25.0
20.0
15.0
10.0
5.0
2.5
1.7
Temperature Change
with Height
AT (°C/100ra)
<-1.9
-1.9 to -1.7
-1.7 to -1.5
-1.5 to -0.5
-0.5 to 1.5
1.5 to 4.0
> 4.0
*Values are averages for each category.
TABLE 3-14
METEOROLOGICAL DATA FOR THE
ISLITZER (1965) TESTS
Test
Number
3
4
5
6
7
8
9
10
11
12
n
Pnsquill Stability Category
AEC: AT
A
A
C
A
G
A
A
G
A
A
A
ARC: aA
A
A
A
C
D
C
C
A
C
C
I)
Mean Wind Speed
at Rclense Height
(m/sec.)
3.1
3.1
3.1
2.7
4.5
3.1
3.1
2.2
2.2
2.7
I .8
3-37
-------
potential temperature gradient, which are used to compute plume rise,
are not needed because the tracer source was non-buoyant. Similarly,
the wind-profile exponent, which is used to adjust the wind speed from
the measurement height to the release height, is not required because
the emission height, the wind sensors and the hi-vol samplers were all
at approximately the same elevation.
3.2.3 Calculation Procedures and Results
The source data presented in Section 3.2.1 and the meteorologi-
cal data given in Section 3.2.2 were used with the ISC Model short-term
computer program ISCST in the Rural Mode to calculate 1-hour average
ground-level centerline tracer concentrations for comparison with the
centerline concentrations estimated by Islitzer (1965) from the observed
tracer concentrations. Values of lateral plume spread a were also
calculated for comparison with the measured values of a • The ISC Model
calculations were performed under the assumption of no building wake
effects and under the assumption that building wakes affected initial
tracer dispersion. As explained below, the results of the lateral plume
spread calculations within the building wake region (i.e., at the 118-
meter sampling distance) were used to estimate the appropriate building
dimensions for use in the ISC Model calculations for the Islitzer tests.
The first sampling distance of 118 meters Is within Lhe building
wake region of both the MTR building alone and the entire MTR-ETR Complex,
Table 3-15 compares the observed values of the lateral dispersion coeffi-
cient a at 118 meters for the Islitzer tests with the corresponding
coefficients calculated in the absence of any building wake effects and
with the MTR building and the MTR-ETR Complex defining the effects of
building wakes on lateral plume spread. The calculated o values in
the absence of any building wake effects are based on the stability
categories defined by the AT and OA stability classification schemes.
A
Also, for the reasons given in Section 3.2.2 above, a third set of a
3-38
-------
TABLE 3-15
COMPARISON OF CALCULATED ANT) OBSERVED Oy VALUES AT 118 METERS
FOR THE ISLITZER (1965) TESTS TOR THE AT AND OA STABILITY
CLASSIFICATION SCHEMES AND FOR A STABILITY
Test
No.
3
4
5
6
7
8
9
10
11
12
13
Observed
a
y
(m)
54.5
58.2
55.1
47.4
24.8
55.7
43.4
50.0
44.5
54.4
49.5
Calculated a (m)
No. Bldgs.
AT
31.2
31.2
14.5
31.2
4.7
31.2
31.2
4.7
31.2
31.2
31.2
°A
31.2
31.2
31.2
14.5
9.6
14.5
14.5
31.2
14.5
14.5
21.2
A Stability
31.2
31.2
31.2
31.2
31.2
31.2
31.2
31.2
31.2
31.2
31.2
MTR
Bldg.
20.8
20.8
20.8
20.8
20.8
20.8
20.8
20.8
20.8
20.8
20.8
MTR-ETR
Complex
45.7
45.7
45.7
45.7
45.7
45.7
45.7
45.7
45.7
45.7
45.7
3-39
-------
values calculated without any building wake effects assumes that A
stability existed during all of the tests (with the possible exception
of Test 7). The ct values calculated by the ISC Model within the wake-
region are independent of atmospheric stability if building wake effects
are included in the calculations. As shown by Table 3-15, the modified
0 values attributable to the effects of the aerodynamic wake of the MTR
building alone are smaller than the point source a values at 118 meters
for the A and B stability categories. However, the modified a values
attributable to the effects of the aerodynamic wake of the entire MTR-ETR
Complex exceed the point source a values for A and B stabilities and
are in good agreement with the observed values for all tests except Test 7.
We therefore assumed that the building wakes affecting tracer dispersion
during the Islitzer tests were determined by the entire MTR-ETR Complex.
Table 3-16 compares the calculated and observed normalized
centerline concentrations at the 118-meter sampling distance for the
Islitzer tests. In the absence of building wake effects, the best cor-
respondence between calculated and observed concentrations is obtained
if A stability is assumed to have existed during all of the tests. How-
ever, the ISC Model overpredicts the observed concentrations by an
average factor of 1.94 for A stability with no building wake effects.
If the effects of the building wakes for the entire MTR-ETR Complex are
included in the calculations, the model overpredicts the observed center-
line concentrations by an average factor of only 1.09.
The observed a values at 350 and 550 meters for the Islitzer
tests are compared with the corresponding values calculated using the AT
and a stability classification schemes in Tables 3-17 and 3-18, respect-
ively. (It is important to note that the a values calculated with
building wake effects included are not Independent of atmospheric stabil-
ity at 350 and 550 meters because these distances are beyond the building
wake region.) The largest differences between calculated and observed a
values in Tables 3-17 and 3-18 occur when a stability category other
3-40
-------
vD
4 1
|
ro
W
,-J
CQ
<3
H
H
<
to
a
o
) — f
H
z
w
u
2
O
"S
£3
?s
I H
g C/3
§^
O p M
E r-J
W 3 M
25 ^ pq
M r. C a
05 O
W pj M
5S w y
M N •<
°£§
^^ S
<*H rn t"1
< M ^
cn
P u r! i-4
H P o
5g
3-
0 rt
r W
S E-1
O g
55
O oo
^^
3"
s
o
o
T)
>
M
0)
(fl
J3
o
T3
C
cfl
T3
0)
alculat
u
<4-\
O
•H
4-J
cd
&
CO
0
o
(U
Emtration X/Q (10~6 s
o
c
o
0
13
0)
•rl
r-l
O
a
•ations
Dncenti
u
Calculated
.u
to
-.
4J
•H
1 — vOC^vDt-~CTi
CNcsrO-HO'H'— iio-^-oocr>r-c-i
mu^-— iroCT>Or--CTNt--3-
r— i «— i i— icsii— ii— i i— (CNcNcslro
oo ^
aooooooor^r-^r^oiooo1^
o\cT>rofOi— ir-ioO"— iror~-
< — (i — If— 1 O"» *OOOOCN »«(T*vO
. — 1 »-H
CM O\
ooooi — ooooooomooocxi
O>ONr— i— t ^H ^ojcMim
uO r-i
, — t
tTv^fvOOOOO-— 1 OO O •— 1
r~r^.CNir^.rJP*.r^-a>vO
r^r^-vDcsjr-~cMoo
i— 1 CO i— 1 •— 1 i— 1
co-O'— ICNO")
• — i i — i < — i i — i
OS
0
i— t
rH
-•t
i-n
i — i
/-N
g
v-^
0
•H
4-)
cO
Pi
C
CO
S
3-41
-------
00
w
I
w
OS
o
fc co
co S
Pd hH*
H
b
o
m o
m !?:
«J o
Q-I
o co
m w
H O
0
i— 1 >i
(0 t)
o
<4-l
o
o
•H
4J
5
T3
a
•H
>
«s
e
o
u-i
LO
townwind
e
o
ui
m
u
a
0
0
m
in
e
o
m
ro
T3
Q)
•M s-*-
(T) 0
r-( ^-^
3
a >
rH t)
«
t3
0) .-
> I
0)
W
f£) t
O
T3
Calculate
av (m)
£
01
03
,0 t
r <2-
1 .
> o
< z
«S X
H (U
W rH
1 P*
H O
S U
03
00
a
•H
Si°i
•H
PQ
PS X
H 0)
W rH
I a,
H o
fe u
03
00
e
•H
0 TJ
•H
3
pa
Pi X
N a)
1 PL
K! 0
EH O
fe O
03
W)
e
0 tJ
55 ,-H
•H
a
«
-\
3
>n
5
pi X
H a)
W i-l
F*5 0
H o
y< o
03
OC
0
•H
& rH
•H
^
-\
3
>.
>
rocyi-J'OOONOO r— sr
oooo^cjN'— io>tocN
OOOO-— lOOi— 1
cnCT\i-HQoo\oooo>*
OOOOOOONrHCTlr-HCSJ
OOOO— lOO— '
LOCOCNi^a-ro— i(T>c^
CTicriO>cscNrg
O O O >-H i— 4 i — 1 O >-" 1
in«yi^o
criCTiiricNCvj-H t— H r-H r«l i— 1 )-H
OOO— (CNOc-JOO
O^CJOrou-iroinr-ico
^tmC^cslOcNi^-ios
rornior^iroror-^ro
cNicsivj^r^cMf^r-^rsi
OOOOvDOOcOOOuncX3
mroinnrorocr>ro
CNCNIT'CMCVlC-JrvlCNJ
cXJOOflOOOOaO'-HOO
rOC^Or'IOi/1'— 100
vOoo--iv£ir^aooom
OOCX3f^^Dv£>u^u~lvD
ro<±u~i^DOOCriOc\l
i — t rH
in
oo
O
00
i>
o
0
•—I
1— I
00
OS
d
»r™^
B
^^x
0
•H
4-1
a
$
&
3-42
-------
w
PC!
e> '
o
m o
m a >
to
W
S3 H H M
H EH
gggd
MMH
' to PH
tn
to
OHO
W co w
> W H
OS H •<
pa m u
o >x> t-3
ON
iH
3 >
JH °
co
u
o
o
•H
4J
3
••a
c
•H
>
O
o
0
0
to
"I
a
•H
5
£
0
O
in
ro
i.
u
a
f-
0
O
m
m
0
o
m
-------
than the very unstable A Pasquill stability category is assumed in the
model calculations. Because It is entirely possible that A stability
occurred during all of the Islitzer tests with the possible exception of
Test 7 for which no observed 0 values are available at 350 and 550
meters, Table 3-19 compares the observed a values at 350 and 550
meters with the corresponding values calculated for A stability. Inspection
of Table 3-19 shows that the calculated effects of building wakes on
lateral dispersion at 350 and 550 meters are negligible in comparison
with the calculated effects of atmospheric turbulence if A stability is
assumed. The overall correspondence between calculated and observed a
values at 350 and 550 meters is about the same for the AT and a. stability
classification schemes and if A stability is assumed for all tests, both
with and without the inclusion of building wake effects in the model
calculations.
The observed normalized centerline concentrations at 350, 550
and 850 meters for the Islitzer tests are compared with the corresponding
concentrations calculated using the AT and a. stability classification
schemes in Tables 3-20 and 3-21, respectively. Additionally, Table 3-22
compares the observed concentrations at 350, 550 and 850 meters with the
corresponding concentrations calculated under the assumption that all of
the tests were conducted during periods of A stability. Inspection of
Tables 3-20 through 3-22 shows that the best overall model performance
at 350, 550 and 850 meters is obtained if it is assumed that A stability
existed during all trials. At these distances, the calculated effects
of atmospheric turbulence for A stability are far more important than
the calculated effects of building wakes. Thus, the inclusion of building
wake effects in the calculations does not significantly improve the ISC
Model's performance for the Islitzer tests except within the building
wake region.
Table 3-23 summarizes the comparisons of concurrent calculated and
observed centerline concentrations for the Islitzer tests assuming that the
3-44
-------
TABLE 3-19
COMPARISON OF OBSERVED Oy VALUES AT 350 AND 550 METERS FOR
THE ISLIZTER (1965) TESTS WITH THE CORRESPONDING CTy
VALUES CALCULATED UNDER THE ASSUMPTION OF
A STABILITY DURING ALL TESTS
Test
No.
3
4
5
6
8
9
10
12
Lateral Plume Spread a (m)
350m
Observed
86.3
88.9
71.0
66.3
67.0
58.5
58.1
63.8
Calculated*
82.3
82.3
82.3
82.3
82.3
82.3
82.3
82.3
550m
Observed
149.0
138.0
193.0
125.1
103.2
125.0
111.2
98.8
Calculated*
123.0
123.0
123.0
123.0
123.0
123.0
123.0
123.0
Mean Ratio (MR)
Ratio of
Calculated and
Observed 0y
Values
350m
0.95
0.93
1.16
1.24
1.23
1.41
1.42
1.29
1.18
550 m
0.83
0.89
0.64
0.98
1.19
0.98
1.11
1.24
0.94
*Calculated CTy values are the same both wttti and without building wake
effects.
3-45
-------
f
o
in
co
H
W
CO 32
Z H
O
H H W
25 M £
Z CO O
W H Z
U CO H
O H P
U 'W
M *n M M
Z vO H EB
rH OV •< O
w o z
H cut h-J O
Z W <£ M
CM S H Pd 1— I
pi W O M
W H Z <<
CO W M H
pa Sc o co
o z
O O H
ffi oo co
H W
Q |X
0^0
z o
o
O
e
o
CJ
•a
01
N
•rl
rH
i
o
s
O
00
e
o
m
in
S
0
m
CO
0
m
00
Q
0
un
m
S
0
m
en
C/D
n)
rH
a
rH
CO
u
rl
C
g
cu
rH
CX
O
CO
60
•O
rH
PQ
X
OJ
D,
g
U
M
60
•O
rH
PQ
CU
rH
cx
e
o
O
CO
•o
rH
PQ
rncH
W iH
1 0.
pcJ 0
S O
CO
M
O T)
*
in
Q
D
H cu
W rH
1 CX
pS 0
H O
fe O
6£
O TJ
53 rH
PQ
J3
D
D
H Q)
W rH
s rH O"t CM oj
* — i co *~H f~~^ r^» r^> o^ i^> v*o czj
CMr— I^Ht— lOOOOr— 1 O CM
-H M
CM
-^•OvOOOOOCTiinO'— I^D
COCMf~-r-ii—(i— IOOCO-HCM
SSS2§2^8-S
COCMOO— <•— IrH^COrHCM
-1 ^
vOMOrHc^vOvDr^coa^oo
i — \ r-H !<(• r-H rH rH CO CM ' — ) CM
rH V£)
oooo^ooooo^moo
C5 IS *™^ ^^ ^O CO t^J CvJ l^ CO
00 00 *^O r^~* oo oo 0^1 CNI r~> c?
i_T") Lf*^ LO ^D L/"l i-O •vCt" OO "^O CD
CM l^> rH
^^vominm^cMincM
vo^tor-^vo^^r-.-H
S2SS^?IS2§S
r—{
CM^rHCOCNCMinoOrO^-
i— it-H
-------
O
in
ro
H
CO 33
*^ T \
H H W
55 W O
U CO M
25 Cd CO
0 H D W
pr^ LO [l3 PC
5S ''•O E— i U
pq cj O
H & J M
25 W 3 H
^ g S ° 3
csj v-> M CJ
1 h-l CO H
' n iJ z fn
^ W CO O M
w N M M CO
n H H co
g | HNU
igg£
S CO 8 H
J> P5 PQ
W H Z H
CO W M CO
PQ 5u Q
0 1Z <
0 0 D
33 00 CO
H U
o << o
u
Z 0
o in
co m
1
T3
>
CU
CO
0
C en
cfl C
0
•a -H
0) 4J
4J n)
O Q)
cd C
o o
u
o
•H
4J
cd
Pi
ro
O
CU
CO
1
, — |
Cf
0
•H
icentral
o
u
13
N
•H
•— i
0
o
z
0
o
m
oo
0
m
m
0
ro
0
O
m
0
o
LO
0
m
ti
0)
FH
Pi
H
W
Pi
O
z
pi
H
W
0
pi
H
W
Pi
ilculate
10
U
ri
•o
01
cd
rH
3
rH
cd
c
4-J
vO LO "^^
CM* 'ooo^vooo^^^
2 2 2 ,n- S ° S 2 2 !^
-J" LO fO l~~* ^ ^^ ^J ^^ I**"* 00
cs^cs^vomcM^m^
^•j- ^^ *^- vO ^J" OO CN| LO CNJ CNJ
^O fxj LO "*3~ CNJ "^" ^^ C^l LO CNJ
(TICNOOCNlCNCNCOvDCN-a-
vO C^J LO *~~l ^O C*J f^ t^ ^«D
rO CNJ C*O vO LO ^O fO "^O to
^^vDCxl^^oo^ONin
r— 4 r— i i~H **O "J O^ O^ C^J
r-^ , 1 , 1 ,— 1 ,— J r-H
OJ CM c-J r- * -o o^> r^.
r | r I o] CO OJ CN
iO LO LO CO **O *^D CN| -^ ^«o LO oo oo o^ oo LO .^^
v^ i— I CO *^O LO C\l O^ «"H CO CT\
cM-*cvjmr-ooco o r—i c-o co
, — i , — i , — i , — f
i — i
CM
-*
cr.
in
*sO
*\o
on
m
VO
ro
O
ro
--*
0
^>
^^
0
•H
cfl
C
c3
3-47
-------
s
o
r^i
^H fj
O PH
00 6
<3 p
0 H
in --<
OO MM
W P _]
U 55 53 M
>J M O CP
<} S CO H
H W y CO
W O
CJ O Pn
O
Q M
W S 53
M M
rJ g H
O co
H H
W /"-N
co in
PQ vO
O CTv
W
PL, NI
0 H
O co
O
CJ
-o
tO
CO V4
M 4J
3 (3
fH O
0 0
o
0
•H
2
ro
B
a
01
CO
o
i — i
&•
X
c
4->
t-l
4->
0)
O
c
o
a)
N
•H
i
,9
o
in
CO
o
in
o
CO
Q
O
m
00
o
m
0
O
in
4-1
CO
0)
,H
p:
t-
b.
^
a
£
PL
o
ff!
W
E-
4-1
ca
3
U
rH
CO
rl
13
0)
rH
3
CO
c
t3
0)
rH
3
M
CO
4
O
X
at
rH
B
o
o
CO
60
•O
pa
X
rH
H
O
CJ
M
*"O
pa
X
o>
M
Oi
O
CJ
CO
60
T>
pa
W rH
EH S
60
0 TJ
pa
J3
Q
H 0)
W rH
ti
0 M
£3 pQ
/>
p
H 0)
fU rH
& 0
£3
60
53 rH
oa
/)
D
3
oo ^J *~^ o^ o** r^-. r*^> ^" ~xf O*^ CO CO
M OOOOCNOOO
^1 SSfnlnSoc^c^
CN OOOOCMMOO
t— ( co LO »~H r**- r*^» ^^ o^ ^«D c^
CNJ r— 1 C^l i — H C? C^ CXJ *~H CO CV|
^SSPISS^SS^
CNr-ICNr-IOOCNCNlOO-J
^o^cogajgo^,.
^^OOM^rHCOOOMCN
\O CNI tO «~*^ C^ ^-4 ^^ CD C^4 ~>^
CO CSl CO *~~^ *~~* *~~^ CO CO »~H Cs|
vOvOvOCTN^^C-l^ONCO
1— ti— li-4rHr-ti-4C>JCNi— ICNl
aooocooooooinmoo
rHr-Hr-ICN^-l^-JCNlCNCNrO
So^oScxjSPi^CN
OSMooooroOCMinoo
oooooor-oOoocNCNr-o
r-H
^in^ininmcNcNLn^
^so^r^vovooc^r--
vO O^ G5 O*\ cO CT1^ *\O cO ^"^ '^O
^D r-H CO ViD LO O^J O^ ^-* < ^^ O^
cN^cNinr-ooco^o^
CMCMCMO-1CNICVJOOCOCO-0-
ScN?gcN^^c?I^cN^
O.CNCM^CNCNOOOOOO^
c^c^rl^rlSSS^^
s^ss^ssg
»— 1 i— I r-t i— 1
CM
0
CO
o
CN
i— H
1 — 1
r-H
OO
1 — 1
oo
-'
0
•H
4-J
to
ti
CO
Q)
s
3-48
-------
TABLE 3-23
SUMMARY OF THE RESULTS OF THE TSC MODEL CALCULATIONS
FOR THE ISLITZER (1965) TESTS ASSUMING
A STABILITY FOR ALL TESTS
Downwind
Distance
(m)
118
350
550
850
No. of Valid
Samples
11
10
10
9
MR of Centerlinc
Concentrations
No lUdgs
1.94
1.87
1.41
0.78
Hl
-------
Pasquill A stability category existed during all of the tests. Under
this assumption, the concentrations calculated at the 350-, 550- and 850-
meter sampling distances are primarily determined by atmospheric turbulence
and are not significantly affected by building wake effects. However, as
noted above, the 118-meter sampling distance is within the wake region
of the MTR-ETR Complex where the calculated concentrations are entirely
determined by building wake effects. As shown by the MR in Table 3-23,
the inclusion of building wake effects in the model calculations improves
the average correspondence between calculated and observed concentrations
by almost a factor of 2. Also, the RMSE is decreased from 2.7 to 2.0, and
the number of calculated concentrations within a factor of 2 of the cor-
responding observed concentrations is almost doubled.
3.3 MUNN AND COLE (1967) TRACER RELEASES
Munn and Cole (1967) present the results of eleven tracer
releases conducted at the CANDU Nuclear Power Generating Station at
Douglas Point, Ontario. Uranine dye, which was used as the tracer, was
released from the top of a 46-meter stack while the station was under
construction. Several large buildings which were located near the stack
might be expected to affect the dispersion of the tracer. Five to seven
portable samplers were placed in a single arc at downwind distances that
differed from trial to trial. For six of the eleven trials, the samplers
were placed within three building heights of one or more of the large
buildings of the complex. These trials are not included in this analysis
because the ISC Model does not estimate concentrations within three
building heights of a building (i.e., within the cavity zone). Munn and
Cole give only the maximum observed normalized concentration for each
trial. The sampling array was too small to estimate either centerline
concentrations or plume widths. The source data for the Munn and Cole
(1967) tracer releases are discussed in Section 3.3.1, the meteorological
data are given in Section 3.3.2, and the ISC Model calculation procedures
and results are described in Section 3.3.3.
3-50
-------
3.3.1 Source Data
Figure 3-5 shows the experimental layout for the tracer
releases at the CANDU Nuclear Generating Station described by Munn and
Cole (1967). The heights above ground level of the buildings at the
CANDU Nuclear Generating Station are labeled in meters in Figure 3-5.
The tracer source for each of the Munn and Cole trials was a non-buoyant
release from the top of the 46-meter stack (chimney) shown in Figure 3-5.
Although Munn and Cole do not give the tracer emission rates for the
various trials, the emission rates are not required for use in testing
the ISC Model because the observed concentrations are given in the
normalized form X/Q» where X is the maximum observed concentration in
grams per cubic meter and Q is the tracer emission rate in grams per
second. The tallest building at the CANDU Station was the reactor
building, which is 44 meters high and 43 meters in diameter. The tracer
release point was about 45 meters from the reactor building, but was not
either directly upwind or directly downwind of the reactor building
during any of the five trials analyzed in this section. The next largest
building was the turbine building, which is 28 meters high, 55 meters
long and 46 meters wide. The tracer release point was about 70 meters
from the turbine building and was downwind of the turbine building
during two of the five trials. The stack used as the tracer release
platform is located on the end of the third large building at the CANDU
Station, which is 12 meters high, 65 meters long and 40 meters wide.
Two sets of building dimensions were used in the ISC Model
calculations for the Munn and Cole trials. The first set of building
dimensions corresponded to the reactor building and consisted of a
building height h, of 44 meters and an effective building width h of 43
D W
meters. The second set of building dimensions assumed that the building
wakes of the entire complex affected the dispersion of the tracer. The
building height h, for the second set of building dimensions was set
equal to the height of the tallest building (the 44-meter reactor building)
3-51
-------
LAKE
HURON
N
I
150m
(a) Munn and Cole (1967) Trials 2, 3 and 12
LAKE
HURON
CHIMNEY
150m
(b) Munn and Cole (L967) Trials 39 and 40
FIGURE 3-5. Experimental layout for (a) Trials 2, 3 and 12 (west-northwest
winds) and (b) Trials 39 and 40 (north-northeast winds)
reported by Munn and Cole (1967). Building elevations are
shown in meters above ground level (AGL). The filled circles
show the sampler locations and the chimney shows the emission
point.
3-52
-------
and the effective crosswind width h of the source complex was estimated
w
from Figure 3-5 to be about 100 meters for all five trials. Because
the stack height to building height ratio is only about 1.05 for both
sets of building dimensions, the ISC Model modifies both the vertical
and lateral dispersion coefficients. The modified vertical dispersion
coefficient a1 for the first set of building dimensions is a function of
z
both h, and h , while the modified vertical dispersion coefficient for
the second set of building dimensions is a function of h, alone. Similarly,
the modified lateral dispersion coefficient a' for the first set of
building dimensions is a function of h alone, while the modified lateral
dispersion coefficient for the second set of building dimensions is a
function of both h and h, .
w D
3.3.2 Meteorological Data
The only meteorological data given by Munn and Cole (1967) are
the mean wind speeds and wind directions measured on a 24-meter tower
located near the source. (The wind speeds are listed in Table 3-24 in
Section 3.3.3.) It was therefore necessary to estimate the remaining
ISC Model meteorological input parameters. Because the trials were
conducted on summer days with wind speeds greater than or equal to 3.6
meters per second, we assumed that the mixing heights during the trials
were sufficiently deep so that the restriction on vertical growth at the
top of the surface mixing layer did not affect the observed concentrations.
The ambient air temperature and vertical potential temperature gradient,
which are used by the ISC Model to calculate plume rise, were not required
as inputs because the tracer emissions were non-buoyant with essentially
no initial vertical momentum. All of the samplers were located within
the building wake, and the ISC Model yields concentration estimates that
are independent of stability within the wake. Consequently, it was not
necessary to know the Pasquill stability category for the two sets of
calculations that included building wake effects. The neutral D stability
category was arbitrarily assumed to apply during all trials in the set
3-53
-------
of calculations without wake effects. The default value of the wind-
profile exponent for D stability (0.25) was used in all of the model
calculations.
3.3.3 Calculation Procedures and Results
The source data presented in Section 3.3.1 and the meteorologi-
cal data discussed in Section 3.3.2 were used with the ISC Model short-
term computer program ISCST in the Rural Mode to calculate ground-level
centerline concentrations at the distances of the highest observed con-
centrations under the following assumptions:
• Building wakes did not affect tracer dispersion
• The aerodynamic wakes and eddies formed by the reactor
building alone affected tracer dispersion
• The aerodynamic wakes and eddies formed by all of the
buildings of the CANDU Station affected tracer dispersion
The results of these calculations are summarized in Table 3-24. For
each trial, the table gives the highest observed normalized concentration,
the corresponding normalized centerline concentrations calculated under
the three assumptions about building wake effects listed above, and the
ratios of the calculated and observed concentrations.
It is important to note that the ISC Model was used to calculate
centerline concentrations for each trial, while the highest observed
concentrations in Table 3-24 were not necessarily measured at the plume
centerline. Thus, the calculated concentrations in Table 3-24 should be
greater than or equal to the highest observed concentrations if the ISC
Model is accurately simulating the actual dispersion of the tracer.
Trial 12 is probably an example of a case in which the maximum observed
3-54
-------
TABLE 3-24
COMPARISON OF THE MAXIMUM OBSERVED CONCENTRATIONS FOR THE
MUNN AND COLE (1967) TRIALS WITH THE CORRESPONDING
CENTERLINE CONCENTRATIONS CALCULATED UNDER
THREE DIFFERENT ASSUMPTIONS ABOUT
BUILDING WAKE EFFECTS
Trial
Number
2
3
12
39
40
24-Meter
Wind
Speed
(m/sec)
4.9
3.6
6.7
8.9
8.5
Highest Normalized Concentration
X/Q (10~6 sec/m3)
Observed
15.6
11.9
0.2
5.4
21.8
Calculated
No
Bldg
0.30
0.36
0.00
0.00
0.04
Reactor
Bldg
25.9
35.5
23.1
20.1
16.4
All
Bldgs
15.2
20.7
12.2
9.3
9.2
Mean Ratio (MR)*
Ratio of Calculated
and
Observed Concentrations
No
Bldg
0.02
0.03
0.00
0.00
0.00
0.01
Reactor
Bldg
1.66
2.98
115.50
3.72
0.75
1.79
All
Bldgs
0.97
1.74
61.00
1.72
0.42
0.99
*The MR values exclude Trial 12 for the reasons given in the text.
3-55
-------
concentration was measured some distance from the actual plume centerline.
For this trial, the highest observed concentration is about two orders
of magnitude lower than both of the calculated centerline concentrations
with building wake effects included. Also, the highest observed concen-
trations for Trials 2 and 3, which were conducted under meteorological
conditions similar to those of Trial 12, are about two orders of magnitude
larger than the highest observed concentration for Trial 12. Because of
this strong implication that the highest observed concentration for
Trial 12 is about two orders of magnitude lower than the actual plume
centerline concentration, the Trial 12 results are excluded from the MR
given at the bottom of Table 3-24.
Table 3-25 summarizes the results of the comparisons of the cal-
culated centerline concentrations and the corresponding maximum observed
concentrations for the four Munn and Cole trials we determined were suitable
for use for the reasons given above. The results presented in Table 3-25
illustrate the sensitivity of concentrations calculated in the wake region
to different assumptions about the buildings with wakes affecting initial
plume dispersion. However, whether or not the reactor building or all
buildings are assumed to have affected initial plume dispersion, the overall
model performance is significantly improved by the inclusion of building
wake effects in the model calculations.
3.4 DOW MIDLAND, MICHIGAN PLANT DATA
DOW Chemical USA currently uses a Supplementary Control System
(SCS) at its Michigan Division plant in Midland, Michigan to maintain
the National Ambient Air Quality Standards (NAAQS) for sulfur dioxide
(S09). (A Supplementary Control System curtails S0~ emissions whenever
meteorological conditions conducive to high ground-level concentrations
occur or are anticipated.) Figure 3-6, a map of the Midland area,
shows the locations of DOW's West and South Power Houses, which are the
3-56
-------
TABLE 3-25
SUMMARY OF THE RESULTS OF THE ISC MODEL CALCULATIONS
FOR THE MUNN AND COLE (1967) TRIALS*
Case
No Buildings
Reactor Building
All Buildings
Mean Ratio of
Calculated to
Observed
Concentrations
(MR)
0.01
1.79
0.99
Root Mean
Square
Error
(RMSE)
176.84
2.47
1.79
% of Center line
Concentrations within a
Factor of 2 of Highest
Observed Concentrations
0
50
75
*Sample size is 4.
3-57
-------
FIGURE 3-6. Map of the area surrounding the West and South Power Houses
at the DOW Midland, Michigan pJant. The filled circles show
the locations of the continuous S02 air quality monitors.
3-58
-------
primary sources of S0? emissions in the Midland area. The filled circles
in the figure show the locations of the nine continuous SO^ air quality
monitors in the vicinity of the DOW plant. The stack height to building
height ratios at both power houses are less than 1.8 and the stack exit
velocities are relatively low. Consequently, DOW and the Michigan
Department of Natural Resources (DNR) believe that the aerodynamic wakes
and eddies formed by the power house buildings have an adverse effect on
the dispersion of the plumes from the power house stacks. Because DOW
operates an SCS at the Midland plant, detailed and concurrent emissions,
meteorological and air quality data are available for the plant. The
DOW Michigan Division provided these data to EPA for use in testing the
ISC Model building wake effects option. The DOW source data are discussed
in Section 3.4.1, the meteorological inputs for 90 selected cases are
given in Section 3.4.2, and the ISC Model calculation procedures and
results are described in Section 3.4.3.
3.4.1 Source Data
DOW's meteorological consultant, Environmental Research and
Technology, Inc. (ERT), provided magnetic tapes containing hourly emis-
sions, meteorological and air quality data for the Midland plant during
the period 10 November 1975 through 27 September 1976 (Abrams, 1978 and
1979). We analyzed the meteorological and air quality data to isolate
hours when meteorological conditions were favorable for adverse building
wake effects. The selection criteria were as follows:
• A 1-hour SO™ concentration greater than or equal to
0.25 parts per million (ppm) at one or more of the
S0» monitors shown in Figure 3-6
• A 10-meter wind speed greater than or equal to 6.17
meters per second (12 knots)
3-59
-------
The basis for the first selection criterion was that we wished to model
the hours with the highest observed SO- concentrations. The two reasons
for the second selection criterion were: (1) Adverse building wake effects
are most likely during periods of moderate or strong winds, and (2) Following
the Turner (1964) definitions of the Pasquill stability categories, the
neutral D stability category is almost always associated with 10-meter wind
speeds greater than or equal to 6.17 meters per second (no cloud cover
data were available to estimate stability using the Turner approach).
Table 3-26 gives the dates, Julian Days and hours of the 90
cases selected for use in testing the building wake effects option of
the ISC Model. It is important to note that, in addition to the West
and South Power Houses, the DOW Midland plant has two package boilers
that are normally used as peak load units. One of the package boilers
was in operation during 25 of the 90 cases (Cases 5, 6, 7, 8, 9, 10, 14,
15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 29, 30, 31, 32 and
33) and both package boilers were in operation during Case 13. The
package boiler S0~ emissions are small in comparison with the SO™
emissions from the power houses. For example, the maximum contribution
of the package boiler emissions to the total S0« emissions from the DOW
plant for the 90 cases listed in Table 3-26 is 6.5 percent (Case 13).
Consequently, the air quality impact of emissions from the package
boilers was assumed in this study to be negligible in comparison with
the impact of emissions from the power houses.
Scale drawings of the West and South Power Houses are shown in
Figures 3-7 and 3-8, respectively. These figures, which are based on
DOW plant layouts dated 23 February 1979, indicate structure elevations
in meters above ground level (AGL) . The stack height for Units 13, 19
and 20 is 56 meters, while the height of the single stack serving Units
9 and 12 Ls 48 meters. The stack height for all stacks at the South
Power House is 54 meters. The building dimensions derived from Figures
3-7 and 3-8 and assumed in the ISC Model calculations are listed in
Table 3-27. The stack height to building height ratio for each stack at
3-60
-------
TABLE 3-26
HOURS SELECTED FOR TESTING THE ISC MODEL AT THE
DOW MIDLAND PLANT
Case Number
1 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Date
10 November 1975
12 November 1975
20 November 1975
21 November 1975
22 November 1975
1 December 1975
16 December 1975
17 December 1975
18 December 1975
20 December 1975
3 January 1976
5 January 1976
8 January 1976
14 January 1976
Julian Day
314
Hour*
0100
0200
0500
1300
2300
| 2400
316
324
325
326
335
350
351
352
354
003
005
008
014
1700
1200
1300
0800
1500
1600
0400
1500
1900
2000
2100
0400
0700
1000
1100
1300
1200
1300
1400
1600
1700
2400
0100
1100
1300
1600
1700
0300
0400
0500
0600
0700
0800
2100
3-61
-------
TABLE 3-26 (Continued)
Case Number
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
i
Date
14 January 1976
28 January 1976
7 February 1976
8 February 1976
11 February 1976
12 February 1976
15 February 1976
18 February 1976
21 February 1976
5 March 1976
6 March 1976
7 March 1976
13 March 1976
14 March 1976
24 March 1976
3 April 1976
4 April 1976
5 April 1976
15 April 1976
Julian Day
014
028
038
039
042
043
046
049
052
065
066
067
'073
074
084
094
095
096
106
Hour*
2200
1600
1700
2400
0100
0200
1200
0900
1700
1800
1900
1700
1800
1900
1300
1500
1600
1000
1100
1500
2400
0200
1400
1500
1700
1200
1300
1900
2000
2400
0200
0300
0400
1800
2000
2300
0200
1600
1700
0200
3-62
-------
TABLE 3-26 (Continued)
Case Number
81
82
83
84
85
86
87
88
89
90
Date
16 April 1976
18 April 1976
22 April 1976
1 May 1976
3 May 1976
5 May 1976
12 June 1976
7 July 1976
Julian Day
107
109
113
122
124
126
164
189
Hour*
1200
1000
0600
0700
0800
2200
1200
0200
2400
1800
^Depending on the day of the year, times are in Eastern Standard Time (EST)
or Eastern Daylight Time (EDT) as used by the DOW Midland Plant's Supple-
mentary Control System (SCS).
3-63
-------
>"9I
T
e
t^
to
E
8
UJ
CO
O
I
(T
Ld
O
Q.
O
tr>
o
UJ
4
<
E
(\J
to
j
CO 60
0)
3 O
o .n
M to
0) VJ
? 0)
O 4J
PM Q)
e
OJ O
f. A
ffl
4J
3 w
o c
>> o
Cfl 'rt
I
w
o
3-64
-------
13m AGL
TANKS
35m
10
STACK (unit 19)
35m AGL
32 m AGL
25 m AGL
STACK (unit 13)
—o
-STACK (unit 20)
32 m AGL-*
29 m AGL
25 m AGL 32 m AGL*
STACK (units 9,12)
16 m AGL-
WEST POWER HOUSE
20 m AGL
23m AGL
9m AGL
4m AGL
4 m AGL «•
26 m AGL
-29 m AGL.
20 m AGL
17 m AGL
<
6
in
is
<
e
10 20 METERS
— ~-^r ------------ 1
13 m AGL
FIGURE 3-8. Layout of the West Power House at the DOW Midland, Michigan
plant. Structure elevations are shown in meters above
ground .1 eve 1 (AGL) .
3-65
-------
TABLE 3-27
BUILDING DIMENSIONS ASSUMED FOR THE SOUTH
AND WEST POWERHOUSES
Power
House
South
West
Building Dimensions
Height (m)
33
32
Length (m)
90
70
Width (m)
25
70
3-66
-------
each power house exceeds 1.2. Also, the building width to building
height ratio is between 1 and 5 for each power house. Thus, the ISC
Model building wake effects option will modify only the vertical disper-
sion coefficient 0 , and the critical building scale for each power
z
house is the height scale.
The stack and full-load source data, excluding S0~ emission
rates, are listed for the South and West Power Houses in Tables 3-28
and 3-29, respectively. These data were developed from the plant layouts
and additional information provided by DOW (Brown, 1979a and 1979b).
The DOW emissions data tape (Abrams, 1979) included the following parameters
for each stack for each hour:
• HHV — fuel heat content in British Thermal Units
(BTU) per pound
• SPC — fuel sulfur content in percent
• LOAD — load in thousands of pounds of steam per
hour
Based on the information provided by Abrams (1979), the hourly S02 emission
rate in grams per second is given by
n/- / ^ 0.86 x Cl x LOAD x SPC
Q(g/sec) = EFF x HHV
where Cl is a constant, EFF is the boiler efficiency in percent and 0.86 is
a units conversion factor. Table 3-30 lists the values of Cl and EFF
provided by Abrams (1979). The hourly stack exit velocity in meters per
3-67
-------
TABLE 3-28
STACK AND FULL-LOAD SOURCE DATA FOR THE SOUTH POWER HOUSE
Parameter
Stack Height (m)
Stack Inner Diameter (m)
Stack Exit Temperature (°K)
Stack Exit Velocity at Full
Load (m/sec)
Full load (103 Ib steam/hr)
UTM X Coordinate (m)
UTM Y Coordiante (m)
Parameter Value
Units 14 & 15
53.5
4.57
408.0
9.32
720
724,590
4,830,620
Units 16 & 17
53.5
4.57
408.0
9.32
720
724,620
4,830,620
Unit 18
53.9
3.28
419.4
10.05
400
724,550
4,830,660
TABLE 3-29
STACK AND FULL-LOAD SOURCE DATA FOR THE WEST POWER HOUSE
Parameter
Stack Height (m)
Stack Inner Diameter (m)
Stack Exit Temperature (°K)
Stack Exit Velocity at Full
Load (m/sec)
Full Load (10 Ib steam/hr)
UTM X Coordinate (m)
UTM Y Coordinate (m)
Parameter Value
Unit 19
55.5
3.79
411.1
10.96
625
722,620
4,831,070
Unit 13
56.0
2.47
416.7
14.18
360
722,630
4,831,070
Unit 20
55.5
2.43
422.2
18.01
400
722,650
4,831,070
Units
9 & 12
47.9
4.42
416.7
5.17
400
722,640
4,831,040
3-68
-------
TABLE 3-30
PARAMETERS USED TO CALCULATE S02 EMISSION RATES
Unit(s)
13
18
19
20
9 & 12
14 & 15
16 & 17
Cl
(Watts/103 Ib steam)
342, 810
342, 810
342, 810
342, 810
342, 810
342, 810
342, 810
EFF
(%)
90
89
89
90
85
91
91
3-69
-------
second is given by Brown (1979b) as
Vs(Wsec) = Vs{Hax} (3-2)
where V {Max} is the exit velocity in meters per second at full load
s
(LOAD {Max}). Tables 3-28 and 3-29 give the values of V {Max} and LOAD{Max}
S
for the various stacks at the South and West Power Houses.
The stack exit velocities and SO™ emission rates calculated
for the various stacks at the South and West Power Houses during the
selected cases identified in Table 3-26 are presented in Tables 3-31
and 3-32. Inspection of Tables 3-31 and 3-32 shows that there is no
case chosen for model testing in which all of the boilers at the South
and West Power Houses were in simultaneous operation. The emissions
data tape gives non-zero values of HHV and LOAD for Unit 18, but zero
values of SPC (fuel sulfur content). Consequently, Unit 18 was assumed
to have had zero SO,, emissions.
3.4.2 Meteorological Data
The source of the meteorological data used by the DOW Midland
plant's SCS is a 91.5-meter meteorological tower located near S02 moni-
tor D03H in Figure 3-6. The meteorological tower, which is adjacent to
a nuclear power plant currently under construction, is owned by a power
company rather than by DOW or its meteorological consultant. Insofar as
possible, the meteorological inputs used in the ISC Model calculations
for the DOW plant were developed from the tower meteorological measure-
ments.
3-70
-------
TABLE 3-31
HOURLY STACK EXIT VELOCITIES AND S02 EMISSION RATES FOR THE SOUTH
POWER HOUSE DURING THE SELECTED DOW CASES
Case
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Units 14 and 15
Exit
Velocity
(m/sec)
5.9
5.2
5.7
8.4
6.5
6.5
8.3
8.3
8.3
8.4
8.3
8.3
4.4
8.6
8.6
8.6
8.6
8.6
8.8
8.7
8.3
8.6
8.1
8.1
8.3
8.3
6.9
4.3
8.0
7.9
3.9
0
0
4.4
4.4
S02
Emission
Rate
(g/sec)
491.4
429.8
478.4
697.6
540.0
540.0
695.5
690.1
695.5
700.9
694.4
694.4
368.3
718.1
718.1
718.1
718.1
718.1
734.3
722.5
695.5
718.1
671.7
671.7
691.1
691. I
572.4
359.6
665.2
660.9
324.0
0
0
370.4
370.4
Units 16 and 17
Exit
Velocity
(m/sec)
4.2
3.8
4.2
2.5
3.7
3.7
4.2
0
0
0
3.9
3.9
3.5
4.3
4.3
4.3
4.3
4.3
4.2
4.2
4.6
4.6
2.8
4.1
3.9
3.9
3.9
4.5
4.0
4.0
8.2
8.1
8.1
8.1
8.1
S02
Emission
Rate
(g/sec)
348.8
319.7
346.7
33.6
49.5
49.5
349.9
0
0
0
53.0
53.0
47.2
357.5
357.5
357.5
357.5
357.5
351.0
349.9
61.3
61.3
230.0
338.0
324.0
324.0
52.3
60.8
53.3
54.3
680.3
672.8
672.8
671.7
671.7
Unit 18
Exit
Velocity
(m/sec)
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
S02
Emission
Rate
(g/sec)
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
3-71
-------
TABLE 3-31 (Continued)
Case
No.
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
Units 14 and 15
Exit
Velocity
(m/sec)
4.4
4.5
4.5
4.5
4.4
4.4
4.4
4.4
8.8
8.8
8.8
8.2
9.0
8.6
8.6
8.0
8.4
6.9
6.9
8.4
6.5
5.7
8.0
8.2
8.2
5.0
5.0
4.1
2.6
2.2
8.7
8.7
8.6
8.6
8.7
8.9
8.9
8.9
8.5
SO 2
Emission
Rate
(g/sec)
370.4
375.8
375.8
375.8
369.3
368.3
369.3
369.3
408.6
408.6
408.6
382.2
418.8
397.2
400.2
369.0
388.8
321.6
321.6
391.2
65.7
58.0
668.5
687.9
685.7
413.6
413.6
345.6
216.0
185.7
723.5
723.5
721.4
719.2
722.5
740.8
740.8
740.8
711.7
Units 16 and 17
Exit
Velocity
(m/sec)
8.1
8.2
8.2
8.2
8.8
8.8
4.1
4.1
4.4
4.4
4.4
4.1
4.3
3.8
3.8
4.3
4.1
4.6
4.6
4.6
4.6
4.5
4.1
4.3
4.4
8.3
8.3
8.8
8.8
8.9
7.8
8.2
8.2
8.2
8.2
8.5
8.5
8.0
4.4
S02
Emission
Rate
(g/sec)
671.7
685.7
685.7
680.3
730.0
735.4
339.1
54.7
59.2
59.2
59.2
54.5
58.4
51.7
51.7
57.5
55.4
61.5
61.5
61.8
61.3
61.0
345.6
57.5
59.2
694.4
112.0
118.4
118.4
119.3
104.5
109.7
110.3
110.8
111.0
712.7
115.0
108.0
365.0
Unit 18
Exit
Velocity
(m/sec)
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
S02
Emission
Rate
(g/sec)
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
3-72
-------
TABLE 3-31 (Continued)
Case
No.
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
Units 14 and 15
Exit
Velocity
(m/sec)
8.5
8.5
8.4
8.5
8.5
8.8
8.5
8.3
4.5
4.5
4.5
4.5
4.3
3.8
4.2
8.9
S02
Emission
Rate
(g/sec)
711.7
706.3
697.6
708.4
708.4
737.6
708.4
695.5
374.7
374.7
374.7
378.0
356.4
314.3
346.7
745.1
Units 16 and 17
Exit
Velocity
(m/sec)
4.4
4.5
4.3
4.2
4.2
4.3
4.2
4.2
8.6
8.6
8.6
8.3
8.5
8.1
4.5
3.1
SO 2
Emission
Rate
(g/sec)
365.0
378.0
58.5
352.1
352.1
58.5
203.8
56.4
716.0
716.0
716.0
691.1
114.1
108.3
371.5
41.8
Unit 18
Exit
Velocity
(m/sec)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
S02
Emission
Rate
(g/sec)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3-73
-------
CO
W
W
IS
w
co c/>
W
H Q
CJ [d
O CO
w o
> re
H pi
H W
W (5
PM
<;
H
I-J
O
33
CN
••a
c
rt
CT\
CO
4-1
•rH
c
C3
o
CM
4J
•H
5
m
1 — 1
4-1
•H
5
cr>
1 — 1
4-1
•H
c
t3
c
c
o
•H
en /--s
cn o
B 4-1 en
W rt -^
« 00
CN s_*
o
C/l
^v
4J /-x
4-1 -H a
•H O Q)
W rH ~^
^
c
o
•rH
CO s~*>
en a
a 4-1 w
W (fl -^
P5 W)
CN ^^
o
e/3
p^
4J /-N
4-1 -H O
X o co
cu 0
> ^
c
o
•r(
CO /— ^
CO U
e 4J co
w a -~^.
(^ 60
CN — '
o
05
>>
4J s-^.
4J -H O
X O co
eu 0
> ^
S02 Emission
Rate
(g/sec)
Exit
Velocity
(m/sec)
D
n •
-d o
J 25
vo r~- i~- oo r~- O *~ * vo vo *o co
OOOOOOLOCNCNjr-^O\ vo vo CN CM •— i cO
in O O --t 00
OOOOOOcSrOrOvO- vO O •— iOin~a"cOoooOcnvciO\ ON r^ r--. r-» r~
criooomvrmo~3-LnLou-im
^o-3-r-Hu->moomooa\cNcNt-^vom ooooo
cnro-a-rorocNrOfnco^a-cococoro
CN<}-cyvr--OOOOLOOOcOu'>invOOvO vOvOvO\Ov£>
u-i<3-v£> 00 C30 00 -O
ooooooooooronm'-'f^oooocsir--
cococo-a-v±-a--j-oo~d'rMCNOOOO
OOOOOOOOOOCMCNr-.cxlc-JCMtNCNirororOOOmcO'-*'— i
oooooooooooooooooooooooooo
oooooooooooooooooooooooooo
i— icsro-a'LO^or— oooi>O'— icNPO-^i/^vDr^ooc^O'— itNroi)
^I^H^I^-I^H^-lr-l^Hi— l^-ICMCNlCSICMCvlCMCM
3-74
-------
CN
.—4
1
ON
tn
•H
o
•H /—.
cn o
cn cu cu
•H 4-) cn
pj K 00
4-1 -H O
•H CJ QJ
X O CO
W ,-t -^
CO CM
i — o>
U1 CO
co •— i
"O
0)
3
c
•H
4-J
c
o
o
CM
ro
W
4-1
•H
4-J
•H
o
•H .-v
W O
K 0) 0)
•H -U 03
00)-^.
W Pi 60
^x
CM
o
C/3
CTifOOOOOOOOOOOOOOOOOOOOOOOOOOO
^
en
"^
C)0
4-1 -H O
•H O CU
X O cn
cu 0
cu
en •
cd o
3-75
-------
01
3
C
•H
4-J
C
o
o
CM
CO
I
ro
W
r-<
r^
3
Units
o
IN
4J
•H
£
&
co
i — i
4-1
•H
a
P
a\
t— H
4->
•rl
C
O
C
o
•H /-s
cn o
•H 4-1 05
p <5 •^-~
W pi 00
v«^
CM
o
cn
>>
•M x^
•U «H O
•HUH)
W rH ">
Q) 6
> ^
SO- Emission
Rate
(g/sec)
Exit
Velocity
(m/sec)
o
•H s-*.
cn u
•H W CO
S «q -^
W « 60
x_x
CM
O
CO
>,
•M /-v
4J -H O
X O CO
01 0
> ^
O
•H -~N
CO U
•H -M cn
0 « -^,
W pi 60
.
•U /-v
•U -H O
•rl O 0)
W 'H >
0) 0
> ^
0)
CO •
n) o
u n
Otricn-i^r^-
ONCs|rocNcnrOvO'-rsio\r^si-ror-~mooooma\oncTiro ooro
rMCNcsicM ^HinfOLOtommm^-^forornirivi-m-^-^i-ro^ cn00'^O^CNUlLOCNlvO^U~irOOOC^CTiCTvCNirO-<3-CTv'y^Cri CNO"*
^^ONC^oocor^r^£ivO>£i^£>vOOOvOvC
lOvOr^COCT\O'— (CNrO-sf-LOVDr-^OOCTiO'— 1 CM rO
-------
-a
cu
H
•H
O
O
CO
W
H O
•H CJ CD
X o co
W "H -^
01 E
I> ^^
C
o
*H x— s
to o
CO 0) CD
•H 4-1 CO
6 CO \
WWW)
CN
0
to
>>
4-j -H a
•rl a 0)
X O CO
W -H --.
cu e
> -—-
c
o
*H /*~^
CO CJ
co a) cu
•H 4-1 CO
W (2 "«>
CN
O
00
4-1 ^-s
4-1 -H O
•H O Q)
« O to
W rH -v.
CU 6
E> ^^
a>
to •
cfl O
o o o o o o o
o o o o o o o
m o o> in ' — '
• • • • •
O O 00 CN 00 CN CO
CO CT» f*- VD (T»
ro ro co ro ro
O CN vO ON CN
c^ co "^j" ^o in *^* **o
I — 1 1-H r— 1 I — I 1— »
O O O O O O O
o o o o o o o
CN CN
• •
vO vD O O O O O
ro ro
O> CTt
• •
v£> vO O O O O O
-* in \£> r-~ oo o> o
<» oo oo oo oo oo cr>
3-77
-------
Table 3-33 lists the hourly meteorological Inputs for the 90
DOW cases selected for testing the ISC Model. Although the 60-meter
level of the meteorological tower is the wind measurement height nearest
to the stack height, we selected the wind data from the 91.5-meter tower
level for use in the model calculations because we believe that the
91.5-meter wind directions are more likely to be representative of wind
directions at plume height than the 60-meter wind directions. (The ISC
Model uses a wind-profile power law expression to account for vertical
wind-speed shear, but does not account for vertical wind-direction
shear.) The tower 91.5-meter hourly wind directions are reversed 180
degrees in Table 3-33 to obtain the average flow vectors required for
input to the ISCST program, the tower 91.5-meter wind speeds are converted
to meters per second from miles per hour and the 91.5-meter ambient air
temperatures are converted to degrees Kelvin from degrees Celsius.
Following the Turner (1964) definitions of the Pasquill stability categories,
the D category existed during all of the hours selected for model testing
with the possible exception of Case 87 when the C category might have
occurred. Because only three tower levels were available for calculating
wind-profile exponents, the ISC Model default values were used in the
model calculations. Additionally, the ISC Model default values for the
vertical potential temperature gradient were used in the calculations.
In the absence of onsite mixing height measurements, the average of the
early morning and afternoon mean annual mixing heights given by Holzworth
(1972) for Flint, Michigan was assigned to all hours.
3.4.3 Calculation Procedures and Results
The ISC Model short-term program ISCST was used with the source
data discussed in Section 3.4.1 and the meteorological inputs listed in
Table 3-33 to calculate 1-hour average ground-level S02 concentrations
at the locations of the S0« monitors shown in Figure 3-6 for the 90 cases
identified in Table 3-26. The Universal Transverse Mercator (UTM)
coordinates of the SO- monitors are listed in Table 3-34, The ISC Model
3-78
-------
TABLE 3-33
HOURLY METEOROLOGICAL INPUTS FOR THE SELECTED DOW CASES
Case
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
Average Flow
Vector *(deg)
346
350
344
065
078
077
073
328
330
031
087
086
082
044
042
033
031
112
105
111
111
115
247
245
234
235
230
194
081
067
033
084
078
107
108
118
114
114
91.5-Meter Mean
Wind Speed
(m/sec)
10.94
11.38
11.96
19.69
10.40
10.31
9.15
9.29
10.00
8.53
8.21
8.21
10.45
8.35
8.44
9.46
9.15
8.62
10.13
7.41
8.97
9.11
10.36
11.65
12.54
11.92
11.56
9.60
8.79
9.15
9.02
11.74
10.58
9.20
8.71
10.94
8.93
9.42
Mixing
Height
(m)
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
Ambient Air
Temperature
(°K)
290
291
289
280
281
280
279
286
287
276
275
275
268
272
272
272
272
270
266
265
262
262
272
271
270
270
270
268
272
271
263
262
262
271
271
270
269
268
Pasquill
Stability
Category
D
D
1)
D
I)
])
1)
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
1)
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
*Direction toward which the wind is blowing.
3-79
-------
TABLE 3-33 (Continued)
Case
No.
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
Average Flow
Vector* (deg)
107
099
106
028
032
073
076
077
109
031
071
073
078
064
074
077
234
226
229
066
068
071
091
106
102
109
121
067
079
079
074
108
028
029
027
226
231
230
228
91.5-Meter Mean
Wind Speed
(m/sec)
8.97
9.02
9.42
10.36
10.09
11.07
11.70
11.74
11.52
10.80
12.41
12.99
12.54
11.88
13.17
13.84
10.89
9.51
11.07
17.50
18.57
18.30
10.18
12.14
12.23
12.86
12.46
11.16
10.13
10.18
10.31
11.34
11.12
11.65
11.29
15.80
14.38
13.75
13.84
Mixing
Height
(m)
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
937
Ambient Air
Temperature
(°K)
268
266
266
267
268
271
272
272
273
275
284
284
284
279
276
275
274
274
274
272
272
272
273
271
271
271
271
275
274
276
275
274
281
280
281
278
277
276
277
Pasquill
Stability
Category
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
•D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
*Direction toward which the wind is blowing.
3-80
-------
TABLE 3-33 (Continued)
Case
No.
78
79
80
81
82
83
84
85
86
87
88
89
90
Average Flow
Vector* (deg)
094
104
029
066
027
062
071
070
103
109
028
325
090
91.5-Meter Mean
Wind Speed
(m/sec)
8.62
8.93
11.65
10.49
9.24
11.34
10.76
9.91
8.08
8.39
13.04
9.55
9.33
Mixing
Height
(m)
937
937
937
937
937
937
937
937
937
937
937
937
937
Ambient Air
Temperature
(°K)
285
285
292
294
294
283
282
282
287
277
284
295
300
Pasquill
Stability
Category
D
D
D
D
D
D
D
D
D
D
D
D
D
*Direction toward which the wind is blowing.
3-81
-------
TABLE 3-34
UNIVERSAL TRANSVERSE MERCATOR (UTM) COORDINATES OF
THE S02 MONITORS IN THE VICINITY OF THE
DOW MIDLAND PLANT
Monitor
ID No.
D23H
D33H
D63H
D53H
D73H
D83H
D93H
D03H
D13H
Monitor Name
Abbott Road
East lawn
Austin St.
Bullock Creek
Ellsworth St.
Kent St.
Waldo Rd.
Nuclear Site
14SB (State Monitor)
UTM X Coordinate
(m)
725,790
724,010
724,140
722,270
722,230
725,920
726,960
723,310
725,550
UTM Y Coordinate
(m)
4,832,660
4,833,190
4,831,680
4,829,100
4,832,690
4,831,230
4,830,380
4,829,190
4,830,290
3-82
-------
calculations were performed using the six combinations of model options
outlined in Table 3-35. On the basis of a site survey of the DOW Midland
plant, Urban Mode 2 was included in the model calculations (Combinations
4, 5 and 6 in Table 3-35) because the presence of numerous large roughness
elements (buildings) at the plant indicated the possibility of enhanced
turbulent intensities in the vicinity of the plant.
We point out that, because the DOW data were compiled from
routine records of plant operations, the uncertainties in the source,
meteorological and air quality data for the DOW data set tend to be
larger than for the three sets of published diffusion experiments dis-
cussed in Sections 3.1, 3.2 and 3.3. Additionally, uncertainties about
the mean transport wind directions, which are not present in the data
from the diffusion experiments because the locations of the plume center-
lines were defined by the monitoring networks, cannot be removed from
the DOW data set. (See Ellis and Liu (1980) for a discussion of the sen-
sitivity of short-term concentrations calculated at fixed monitor locations
to slight uncertainties about the transport wind direction.) Thus, the
KMSE values for the DOW data set are expected to be much larger than for the
diffusion experiments. Similarly, the percent of calculated concentrations
within a factor of 2 of the corresponding observed concentrations is
expected to be much smaller for the DOW data set than for the diffusion
experiments.
Table 3-36 compares the calculated and observed 1-hour ground-
level S02 concentrations for Combination 1 in Table 3-35. Because of
the uncertainties about the SO- background and the effects of the package
boiler emissions, Table 3-36 considers only monitors with observed 1-hour
S02 concentrations greater than or equal to 0.10 ppm. Similarly, the
observed concentrations greater than or equal to 0.10 ppm are compared
with the corresponding concentrations calculated by the ISC Model for
Combinations 2 through 6 in Table 3-37 through 3-41.
3-83
-------
?
CO
a
i
O H
9 is
JZ
cti
|
P.
•H
H
1
r*
O
a
4J
co
CO
0
5
0)
^
00
•H
•d
•H
3
CD
CO
•H
PS
cu
1
i
CO
U
to
H
O
CO
CD
&
CO
01
cu d
u cu
C3 *O
2 S
co a
a o
H
d
tn
CN
0
(0
M
H
at
A
d
o
•H
I 1
•H
XX XX
X X
X X
XX XX
XX XX
X X
XXX
XXX
t— 1 CN CO
-------
TABLE 3-36
COMPARISON OF CALCULATED AND OBSERVED 1-HOUR S02 CONCENTRATIONS
FOR THE DOW CASES FOR COMBINATION 1 IN TABLE 3-35
Case
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
IS
16
17
18
19
20
21
22
23
24
Year/
Julian
Day
75/314
75/314
75/314
75/314
75/314
75/314
75/316
75/324
75/324
75/325
75/326
75/326
75/335
75/350
75/350
75/350
75/350
75/351
75/351
75/351
75/352
75/352
75/354
75/354
Hour
0100
0200
0500
1300
2300
2400
1700
1200
1300
0800
1500
1600
0400
1500
1900
2000
2100
0400
0700
1000
1100
1300
1200
1300
Monitor
D33H
D63H
D73H
D33H
D73H
D63H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D63H
D23H
D63H
D63H
D83H
D63H
D83H
D23H
D33H
D23H
D23H
D23H
D33H
D13H
D93H
D13H
D93H
D13H
D13H
D83H
D13H
D53H
D03H
D53H
Observed
Concentration
(ppm)
0.36
0.11
0.12
0.26
0.13
0.49
0.14
0.27
0.28
0.19
0.30
0.15
0.12
0.33
0.39
0.55
0.27
0.27
0.30
0.13
0.30
0.12
0.38
0.12
0.30
0.27
0.29
0.10
0.28
0.23
0.14
0.25
0.35
0.25
0.11
0.31
0.25
0.36
0.10
Calculated
Concentration
(ppm)
0.20
0.00
0.05
0.15
0.03
0.01
0.06
0.09
0.00
0.03
0.00
0.00
0.01
0,00
0.00
0.00
0.03
0.00
0.00
0.17
0.00
0.04
0.00
0.00
0.00
0.08
0.10
0.07
0.01
0.00
0.09
0.02
0.02
0.00
0.00
0.00
0.00
0.00
0.01
Ratio of
Calculated
and Observed
Concentrations
0.56
0.00
0.42
0.58
0.23
0.02
0.43
0.33
0.00
0.16
0.00
0.00
0.08
0.00
0.00
0.00
0.11
0.00
0.00
1.31
0.00
0.33
0.00
0.00
0.00
0.30
0.34
0.70
0.04
0.00
0.64
0.08
0.06
0.00
0.00
0.00
0.00
0.00
0.10
3-85
-------
TABLE 3-36 (Continued)
Case
No.
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
Year/
Julian
Day
75/354
75/354
75/354
75/354
76/003
76/003
76/005
76/008
76/008
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/028
76/028
76/038
76/039
76/039
76/042
76/043
76/046
Hour
1400
1600
1700
2400
0100
1100
1300
1600
1700
0300
0400
0500
0600
0700
0800
2100
2200
1600
1700
2400
0100
0200
1200
0900
1700
Monitor
D03H
D03H
D03H
D53H
D63H
D83H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D13H
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D93H
D93H
D13H
D93H
D13H
D93H
D23H
D23H
D33H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D93H
D33H
D23H
D63H
Observed
Concentration
(ppm)
0.65
0.49
0.44
0.35
0.26
0.23
0.28
0.20
0.30
0.27
0.14
0.30
0.10
0.37
0.20
0.26
0.21
0.33
0.18
0.50
0.59
0.30
0.34
0.21
0.48
0.33
0.21
0.32
0.44
0.19
0.47
0.15
0.45
0.18
0.47
0.16
0.40
0.16
0.26
0.22
0.29
Calculated
Concentration
(ppm)
0.00
0.00
0.06
0.08
0.00
0.02
0.03
0.07
0.07
0.00
0.11
0.00
0.01
0.16
0.01
0.12
0.01
0.00
0.00
0.01
0.01
0.00
0.01
0.04
0.26
0.18
0.03
0.13
0.08
0.10
0.02
0.00
0.00
0.00
0.00
0.00
0.11
0.00
0.10
0.02
0.10
Ratio of
Calculated
and Observed
Concentrations
0.00
0.00
0.14
0.23
0.00
0.09
0.11
0.35
0.23
0.00
0.79
0.00
0.10
0.43
0.05
0.46
0.05
0.00
0.00
0.02
0.02
0.00
0.03
0.19
0.54
0.55
0.14
0.41
0.18
0.53
0.04
0.00
0.00
0.00
0.00
0.00
0.28
0.00
0.38
0.09
0.34
3-86
-------
TABLE 3-36 (Continued)
Case
No.
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
Year/
Julian
Day
76/046
76/046
76/049
76/049
76/049
76/052
76/052
76/052
76/065
76/065
76/065
76/066
76/067
76/067
76/067
76/067
76/073
76/074
76/074
76/074
76/074
76/084
76/084
76/084
76/094
76/094
76/094
Hour
1800
1900
1700
1800
1900
1300
1500
1600
1000
1100
1500
2400
0200
1400
1500
1700
0200
1300
1900
2000
2400
0200
0300
0400
1800
2000
2300
Monitor
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D03H
D03H
D03H
D63H
D83H
D83H
D63H
D83H
D13H
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D93H
D23H
D23H
D23H
D03H
D03H
D03H
Observed
Concentration
(ppm)
0.81
0.23
0.42
0.18
0.40
0.15
0.29
0.15
0.28
0.16
0.39
0.28
0.29
0.11
0.34
0.29
0.12
0.26
0.12
0.31
0.27
0.15
0.34
0.11
0.32
0.29
0.26
0.12
0.25
0.10
0.29
0.17
0.28
0.11
0.16
0.25
0.30
0.35
0.30
0.26
0.27
0.59
Calculated
Concentration
(ppm)
0.03
0.00
0.00
0.00
0.04
0.02
0.03
0.00
0.01
0.00
0.00
0.01
0.00
0.09
0.13
0.06
0.07
0.02
0.00
0.06
0.13
0.02
0.11
0.12
0.10
0.00
0.06
0.01
0.00
0.01
0.00
0.01
0.02
0.00
0.08
0.00
0.12
0.08
0.05
0.11
0.01
0.02
Ratio of
Calculated
and Observed
Concentrations
0.04
0.00
0.00
0.00
0.10
0.13
0.10
0.00
0.04
0.00
0.00
0.04
0.00
0.82
0.38
0.21
0.58
0.08
0.00
0.19
0.48
0.13
0.32
1.09
0.31
0.00
0.23
0.08
0.00
0.10
0.00
0.06
0.07
0.00
0.50
0.00
0.40
0.23
0.17
0.42
0.04
0.03
3-87
-------
TABLE 3-36 (Continued)
Case
No.
77
78
79
80
81
82
83
84
85
86
87
88
89
90
Year/
Julian
Day
76/095
76/096
76/096
76/106
76/107
76/109
76/113
76/113
76/113
76/122
76/124
76/126
76/164
76/189
Hour
0200
1600
1700
0200
1200
1000
0600
0700
0800
2200
1200
0200
2400
1800
Monitor
D03H
D13H
D93H
D13H
D93H
D23H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D13H
D23H
D63H
D13H
D93H
Observed
Concentration
(pptn)
0.27
0.30
0.16
0.60
0.16
0.64
0.10
0.26
0.35
0.15
0.36
0.21
0.31
0.25
0.31
0.25
0.27
0.29
0.28
0.30
0.10
Calculated
Concentration
(ppm)
0.02
0.00
0.14
0.11
0.05
0.10
0.05
0.04
0.05
0.01
0.04
0.04
0.02
0.04
0.02
0.07
0.04
0.09
0.00
0.00
0.02
Mean Ratio (MR)
Ratio of
Calculated
and Observed
Concentrations
0.07
0.00
0.88
0.18
0.31
0.16
0.50
0.15
0.14
0.07
0.11
0.19
0.06
0.16
0.06
0.28
0.15
0.31
0.00
0.00
0.20
0.15
3-88
-------
TABLE 3-37
COMPARISON OF CALCULATED AND OBSERVED 1-HOUR SO,
CONCENTRATIONS FOR THE DOW CASES FOR '
COMBINATION 2 IN TABLE 3-35
Case
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Year/
Julian
Day
75/314
75/314
75/314
75/314
75/314
75/314
75/316
75/324
75/324
75/325
75/326
75/326
75/335
75/350
75/350
75/350
75/350
75/351
75/351
75/351
75/352
75/352
75/354
Hour
0100
0200
0500
1300
2300
2400
1700
1200
1300
0800
1500
1600
0400
1500
1900
2000
2100
0400
0700
1000
1100
1300
1200
Monitor
D33H
D63H
D73H
D33H
D73H
D63H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D63H
D23H
D63H
D63H
D83H
D63H
D83H
D23H
D33H
D23H
D23H
D23H
D33H
D13H
D93H
D13H
D93H
D13H
D13H
D83H
D13H
D53H
Observed
Concentration
(ppm)
0.36
0.11
0.12
0.26
0.13
0.49
0.14
0.27
0.28
0.19
0.30
0.15
0.12
0.33
0.39
0.55
0.27
0.27
0.30
0.13
0.30
0.12
0.38
0.12
0.30
0.27
0.29
0.10
0.28
0.23
0.14
0.25
0.35
0.25
0.11
0.31
0.25
Calculated
Concentration
(ppm)
0.28
0.01
0.14
0.20
0.07
0.05
0.11
0.24
0.00
0.00
0.01
0.00
0.05
0.02
0.00
0.02
0.09
0.00
0.00
0.23
0.00
0.05
0.00
0.00
0.00
0.17
0.22
0.11
0.11
0.00
0.03
0.03
0.09
0.04
0.00
0.02
0.00
Ratio of
Calculated
and Observed
Concentrations
0.78
0.09
0.17
0.77
0.54
0.10
0.79
0.89
0.00
0.00
0.03
0.00
0.42
0.06
0.00
0.04
0.33
0.00
0.00
1.77
0.00
0.42
0.00
0.00
0.00
0.63
0.76
1.10
0.39
0.00
0.21
0.12
0.26
0.16
0.00
0.06
0.00
3-89
-------
TABLE 3-37 (Continued)
Case
No.
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
Year/
Julian
Day
75/354
75/354
75/354
75/354
75/354
76/003
/6/003
76/005
76/008
76/008
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/028
76/028
76/038
76/039
76/039
76/042
76/043
Hour
1300
1400
1600
1700
2400
0100
1100
1300
1600
1700
0300
0400
0500
0600
0700
0800
2100
2200
1600
1700
2400
0100
0200
1200
0900
Monitor
D03H
D53H
D03H
D03H
D03H
D53H
D63H
U83H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D13H
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D93H
D93H
D13H
D93H
D13H
D93H
D23H
D23H
D33H
D63H
D83H
D63H
1)8311
D63H
D83H
D13H
D93H
D33H
Observed
Concentration
(ppm)
0.36
0.10
0.65
0.49
0.44
0.35
0.26
0.23
0.28
0.20
0.30
0.27
0.14
0.30
0.10
0.37
0.20
0.26
0.21
0.33
0.18
0.50
0.59
0.30
0.34
0.21
0.48
0.33
0.21
0.32
0.44
0.19
0.47
0.15
0.45
0.18
0.47
0.16
0.40
0.16
0.26
Calculated
Concentration
(ppm)
0.00
0.02
0.00
0.00
0.02
0.18
0.00
0.03
0.13
0.09
0.15
0.00
0.13
0.01
0.01
0.35
0.02
0.30
0.01
0.02
0.00
0.07
0.08
0.00
0.02
0.06
0.36
0.35
0.03
0.22
0.14
0.17
0.07
0.01
0.02
0.00
0.01
0.00
0.22
0.00
0.16
Ratio of
Calculated
and Observed
Concentrations
0.00
0.20
0.00
0.00
0.05
0.51
0.00
0.13
0.46
0.45
0.50
0.00
0.93
0.03
0.10
0.95
0.10
1.15
0.05
0.06
0.00
0.14
0.14
0.00
0.06
0.29
0.75
1.06
0.14
0.69
0.32
0.89
0.15
0.07
0.04
0.00
0.02
0.00
0.55
0.00
0.62
3-90
-------
TABLE 3-37 (Continued)
Case
No.
49
50
51
52
53
54
55
56
57
58
'59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
Year/
Julian
Day
76/046
76/046
76/046
76/049
76/049
76/049
76/052
76/052
76/052
76/065
76/065
76/065
76/066
76/067
76/067
76/067
76/067
76/073
76/074
76/074
76/074
76/074
76/084
76/084
76/084
Hour
1700
1800
1900
1700
1800
1900
1300
1500
1600
1000
1100
1500
2400
0200
1400
1500
1700
0200
1300
1900
2000
2400
0200
0300
0400
Monitor
D23H
D63H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D03H
D03H
D03H
D63H
D83H
. D83H
D63H
1)8 3H
D13II
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D63H
1)83H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
1)9311
1)2 3H
1)2 3H
D23H
Observed
Concentration
(ppm)
0.22
0.29
0.81
0.23
0.42
0.18
0.40
0.15
0.29
0.15
0.28
0.16
0.39
0.28
0.29
0.11
0.34
0.29
0.12
0.26
0.12
0.31
0.27
0.15
0.34
0.11
0.32
0.29
0.26
0.12
0.25
0.10
0.29
0.17
0.28
0.11
0.16
0.25
0.30
0.35
0.30
Calculated
Concentration
(ppm)
0.02
0.25
0.09
0.01
0.01
0.00
0.12
0.10
0.06
0.01
0.01
0.00
0.00
0.02
0.01
0.15
0.36
0.18
0.09
0.07
0.00
0.12
0.27
0.02
0.17
0.14
0.27
0.00
0.17
0.13
0.00
0.01
0.00
0.01
0.05
0.01
0.20
0.00
0.25
0.17
0. 1]
Ratio of
Calculated
and Observed
Concentrations
0.09
0.86
0.11
0.04
0.02
0.00
0.30
0.67
0.21
0.07
0.04
0.00
0.00
0.07
0.03
1.36
1.06
0.62
0.75
0.27
0.00
0.39
1.00
0.13
0.50
1.27
0.84
0.00
0.65
1.08
0.00
0.10
0.00
0.06
0.18
0.09
1.25
0.00
0.83
0.49
0.37
3-91
-------
TABLE 3-37 (Continued)
Case
No.
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
Year/
Julian
Day
76/094
76/094
76/094
76.095
76/096
76/096
76/106
76/107
76/109
76/113
76/113
76/113
76/122
76/124
76/126
76/164
76/189
Hour
1800
2000
2300
0200
1600
1700
0200
1200
1000
0600
0700
0800
2200
1200
0200
2400
1800
Monitor
D03H
D03H
D03H
D03H
D13H
D93H
D13H
D93H
D23H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D13H
D23H
D63H
D13H
D93H
Observed
Concentration
(ppm)
0.26
0.27
0.59
0.27
0.30
0.16
0.60
0.16
0.64
0.10
0.26
0.35
0.15
0.36
0.21
0.31
0.25
0.31
0.25
0.27
0.29
0.28
0.30
0.10
Calculated
Concentration
(ppm)
0.20
0.02
0.03
0.05
0.00
0.26
0.22
0.06
0.18
0.13
0.24
0.11
0.04
0.21
0.12
0.08
0.13
0.11
0.14
0.16
0.13
0.00
0.00
0.05
Mean Ratio (MR)
Ratio of
Calculated
and Observed
Concentrations
0.77
0.07
0.05
0.19
0.00
1.63
0.37
0.38
0.28
1.30
0.92
0.31
0.27
0.58
0.57
0.26
0.52
0.35
0.56
0.59
0.45
0.00
0.00
0.50
0.31
3-92
-------
TABLE 3-38
COMPARISON OF CALCULATED AND OBSERVED 1-HOUR SO,
CONCENTRATIONS FOR THE DOW CASES FOR
COMBINATION 3 IN TABLE 3-35
Case
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Year/
Julian
Day
75/314
75/31.4
75/314
75/314
75/314
75/314
75/316
75/324
75/324
75/325
75/326
75/326
75/335
75/350
75/350
75/350
75/350
75/351
75/351
75/351
75/352
75/352
Hour
0100
0200
0500
1300
2300
2400
1700
1200
1300
0800
1500
1600
0400
1500
1900
2000
2100
0400
0700
1000
1100
1300
Monitor
D33H
D63H
D73H
D33H
D73H
D63H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D63H
D23H
D63H
D63H
D83H
D63H
D83H
D23H
D33H
D23H
D23H
D23H
D33H
D13H
D93H
D13H
D93H
D13H
D13H
D83H
D13H
Observed
Concentration
(ppm)
0.36
0.11
0.12
0.26
0.13
0.49
0.14
0.27
0.28
0.19
0.30
0.15
0.12
0.33
0.39
0.55
0.27
0.27
0.30
0.13
0.30
0.12
0.38
0.12
0.30
0.27
0.29
0.10
0.28
0.23
0.14
0.25
0.35
0.25
0.11
0.31
Calculated
Concentration
(ppm)
0.34
0.02
0.14
0.24
0.07
0.07
0.12
0.31
0.00
0.00
0.01
0.00
0.05
0.03
0.01
0.02
0.10
0.00
0.00
0.23
0.00
0.05
0.00
0.00
0.00
0.20
0.26
0.11
0.16
0.00
0.25
0.03
0.12
0.06
0.00
0.02
Ratio of
Calculated
and Observed
Concentrations
0.94
0.18
0.17
0.92
0.54
0.14
0.86
1.15
0.00
0.00
0.03
0.00
0.42
0.09
0.03
0.04
0.37
0.00
0.00
1.77
0.00
0.42
0.00
0.00
0.00
0.74
0.90
1.10
0.57
0.00
1.79
0.12
0.34
0.24
0.00
0.06
3-93
-------
TABLE 3-38 (Continued)
Case
No.
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
Year/
Julian
Day
75/354
75/354
75/354
75/354
75/354
75/354
76/003
76/003
76/005
76/008
76/008
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/028
76/028
76/038
76/039
76/039
76/042
Hour
1200
1300
1400
1600
1700
2400
0100
1100
1300
1600
1700
0300
0400
0500
0600
0700
0800
2100
2200
1600
1700
2400
0100
0200
1200
Monitor
D53H
D03H
D53H
D03H
D03H
D03H
D53H
D63H
D83H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D13H
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D93H
D93H
D13H
D93H
D13H
D93H
D23H
D23H
D33H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D93H
Observed
Concentration
(ppm)
0.25
0.36
0.10
0.65
0.49
0.44
0.35
0.26
0.23
0.28
0.20
0.30
0.27
0.14
0.30
0.10
0.37
0.20
0.26
0.21
0.33
0.18
0.50
0.59
0.30
0.34
0.21
0.48
0.33
0.21
0.32
0.44
0.19
0.47
0.15
0.45
0.18
0.47
0.16
0.40
0.16
Calculated
Concentration
(ppm)
0.00
0.00
0.02
0.00
0.00
0.02
0.18
0.00
0.03
0.15
0.11
0.18
0.00
0.14
0.01
0.01
0.41
0.02
0.36
0.01
0.02
0.00
0.10
0.11
0.00
0.02
0.07
0.39
0.40
0.03
0.26
0.17
0.18
0.08
0.01
0.02
0.00
0.01
0.00
0.26
0.00
Ratio of
Calculated
and Observed
Concentrations
0.00
0.00
0.20
0.00
0.00
0.05
0.51
0.00
0.13
0.54
0.55
0.60
0.00
1.00
0.03
0.10
1.11
0.10
1.38
0.05
0.06
0.00
0.20
0.19
0.00
0.06
0.33
0.81
1.21
0.14
0.81
0.39
0.95
0.17
0.07
0.04
0.00
0.02
0.00
0.65
0.00
3-94
-------
TABLE 3-38 (Continued)
Case
No.
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
Year/
Julian
Day
76/043
76/046
76/046
76/046
76/049
76/049
76/049
76/052
76/052
76/052
76/065
76/065
76/065
76/066
76/067
76/067
76/067
76/067
76/073
76/074
76/074
76/074
76/074
76/084
Hour
0900
1700
1800
1900
1700
1800
1900
1300
1500
1600
1000
1100
1500
2400
0200
1400
1500
1700
0200
1300
1900
2000
2400
0200
Monitor
D33H
D23H
D63H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D03H
D03H
D03H
D63H
D83H
D83H
D63H
D83H
D13H
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D93H
D23H
Observed
Concentration
(ppm)
0.26
0.22
0.29
0.81
0.23
0.42
0.18
0.40
0.15
0.29
0.15
0.28
0.16
0.39
0.28
0.29
0.11
0.34
0.29
0.12
0.26
0.12
0.31
0.27
0.15
0.34
0.11
0.32
0.29
0.26
0.12
0.25
0.10
0.29
0.17
0.28
0.11
0.16
0.25
0.30
Calculated
Concentration
(ppm)
0.17
0.02
0.29
0.11
0.02
0.01
0.01
0.14
0.13
0.07
0.01
0.01
0.00
o.oo
0.03\
0.01
0.17
0.47
0.24
0.11
0.10
0.00
0.14
0.34
0.02
0.20
0.15
0.35
0.00
0.19
0.16
0.00
0.01
0.00
0.01
0.06
0.01
0.24
0.00
0.29
Ratio of
Calculated
and Observed
Concentrations
0.65
0.09
1.00
0.14
0.09
0.02
0.06
0.35
0.87
0.24
0.07
0.04
0.00
0.00
0.11
0.03
1.55
1.38
0.83
0.92
0.38
0.00
0.45
1.26
0.13
0.59
1.36
1.09
0.00
0.73
1.33
0.00
0.10
0.00
0.06
0.21
0.09
1.50
0.00
0.97
3-95
-------
TABLE 3-38 (Continued)
Case
No.
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
Year/
Julian
Day
76/084
76/084
76/094
76/094
76/094
76/095
76/096
76/096
76/106
76/107
76/109
76/113
76/113
76/113
76/122
76/124
76/126
76/164
76/189
Hour
0300
0400
1800
2000
2300
0200
1600
1700
0200
1200
1000
0600
0700
0800
2200
1200
0200
2400
1800
Monitor
D23H
D23H
D03H
D03H
D03H
D03H
D13H
D93H
D13H
D93H
D23H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D13H
D23H
D63H
D13H
D93H
Observed
Concentration
(ppm)
0.35
0.30
0.26
0.27
0.59
0.27
0.30
0.16
0.60
0.16
0.64
0.10
0.26
0.35
0.15
0.36
0.21
0.31
0.25
0.31
0.25
0.27
0.29
0.28
0.30
0.10
Calculated
Concentration
(ppm)
0.19
0.13
0.25
0.02
0.04
0.06
0.00
0.30
0.26
0.07
0.21
0.13
0.30
0.12
0.05
0.28
0.14
0.11
0.15
0.14
0.15
0.21
0.16
0.01
0.00
0.06
Mean Ratio (MR)
Ratio of
Calculated
and Observed
Concentrations
0.54
0.43
0.96
0.07
0.07
0.22
0.00
1.88
0.43
0.44
0.33
1.30
1.15
0.34
0.33
0.78
0.67
0.35
0.60
0.45
0.60
0.78
0.55
0.04
0.00
0.60
0.37
3-96
-------
TABLE 3-39
COMPARISON OF CALCULATED AND OBSERVED 1-HOUR SO CONCENTRATIONS
FOR THE DOW CASES FOR COMBINATION 4 IN TABLE 3-35
Case
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
J6
17
18
19
20
21
22
23
24
Year/
Julian
Day
75/314
75/314
75/314
75/314
75/314
75/314
75/316
75/324
75/324
75/325
75/326
75/326
75/335
75/350
75/350
75/350
75/350
75/351
75/351
75/351
75/352
75/352
75/354
75/354
Hour
0100
0200
0500
1300
2300
2400
1700
1200
1300
0800
1500
1600
0400
1500
1900
2000
2100
0400
0700
1000
1100
1300
1200
1300
Monitor
D33H
D63H
D73H
D33H
D73H
D63H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D63H
D23H
D63H
D63H
D83H
D63H
D83H
D23H
D33H
D23H
D23H
D23H
D33H
D13H
D93H
D13H
D93H
D13H
D13H
D83H
D13H
D53H
D03H
D53H
Observed
Concentration
(ppm)
0.36
0.11
0.12
0.26
0.13
0.49
0.14
0.27
0.28
0.19
0.30
0.15
0.12
0.33
0.39
0.55
0.27
0.27
0.30
0.13
0.30
0.12
0.38
0.12
0.30
0.27
0.29
0.10
0.28
0.23
0.14
0.25
0.35
0.25
0.11
0.31
0.25
0.36
0.10
Calculated
Concentration
(ppm)
0.22
0.10
0.16
0.17
0.11
0.16
0.10
0.22
0.03
0.03
0.04
0.04
0.12
0.15
0.05
0.09
0.19
0.00
0.00
0.18
0.01
0.06
0.01
0.03
0.03
0.27
0.31
0.10
0.21
0.01
0.27
0.10
0.18
0.10
0.00
0.07
0.04
0.00
0.07
Ratio of
Calculated
and Observed
Concentrations
0.61
0.91
1.33
0.65
0.85
0.33
0.71
0.81
0.11
0.16
0.13
0.27
1.00
0.45
0.13
0.16
0.70
0.00
0.00
1.38
0.03
0.50
0.03
0.25
0.10
1.00
1.07
1.00
0.75
0.04
1.93
0.40
0.51
0.40
0.00
0.23
0.16
0.00
0.70
3-97
-------
TABLE 3-39 (Continued)
Case
No.
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
Year/
Julian
Day
75/354
75/354
75/354
75/354
76/003
76/003
76/005
76/008
76/008
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/028
76/028
76/038
76/039
76/039
76/042
76/043
76/046
Hour
1400
1600
1700
2400
0100
1100
1300
1600
1700
0300
0400
0500
0600
0700
0800
2100
2200
1600
1700
2400
0100
0200
1200
0900
1700
Monitor
D03H
D03H
D03H
D53H
D63H
D83H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D13H
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D93H
D93H
D13H
D93H
D13H
D93H
D23H
D23H
D33H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D93H
D33H
D23H
D63H
Observed
Concentration
(pptn)
0.65
0.49
0.44
0.35
0.26
0.23
0.28
0.20
0.30
0.27
0.14
0.30
0.10
0.37
0.20
0.26
0.21
0.33
0.18
0.50
0.59
0.30
0.34
0.21
0.48
0.33
0.21
0.32
0.44
0.19
0.47
0.15
0.45
0.18
0.47
0.16
0.40
0.16
0.26
0.22
0.29
Calculated
Concentration
(ppm)
0.03
0.02
0.07
0.25
0.01
0.07
0.21
0.21
0.25
0.00
0.11
0.07
0.05
0.41
0.08
0.39
0.05
0.10
0.00
0.20
0.21
0.00
0.08
0.16
0.39
0.39
0.10
0.21
0.13
0.19
0.15
0.06
0.08
0.04
0.07
0.03
0.25
0.02
0.16
0.05
0.31
Ratio of
Calculated
and Observed
Concentrations
0.05
0.04
0.16
0.71
0.04
0.30
0.75
1.05
0.83
0.00
0.79
0.23
0.50
1.11
0.40
1.50
0.24
0.30
0.00
0.40
0.36
0.00
0.24
0.76
0.81
1.18
0.48
0.66
0.30
1.00
0.32
0.40
0.18
0.22
0.15
0.19
0.63
0.13
0.62
0.23
1.07
3-98
-------
TABLE 3-39 (Continued)
Case
No.
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
Year/
vTulian
Day
76/046
76/046
76/049
76/049
76/049
76/052
76/052
76/052
76/065
76/065
76/065
76/066
76/067
76/067
76/067
76/067
76/073
76/074
76/074
76/074
76/074
76/084
76/084
76/084
76/094
76/094
76/094
Hour
1800
1900
1700
1800
1900
1300
1500
1600
1000
1100
1500
2400
0200
1400
1500
1700
0200
1300
1900
2000
2400
0200
0300
0400
1800
2000
2300
Monitor
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D03H
D03H
D03H
D63H
D83H
D83H
D63H
D83H
D13H
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D93H
D23H
D23H
D23H
D03H
D03H
D03H
Observed
Concentration
(ppm)
0.81
0.23
0.42
0.18
0.40
0.15
0.29
0.15
0.28
0.16
0.39
0.28
0.29
0.11
0.34
0.29
0.12
0.26
0.12
0.31
0.27
0.15
0.34
0.11
0.32
0.29
0.26
0.12
0.25
0.10
0.29
0.17
0.28
0.11
0.16
0.25
0.30
0.35
0.30
0.26
0.27
0.59
Calculated
Concentration
(ppm)
0.16
0.07
0.04
0.03
0.19
0.15
0.12
0.05
0.05
0.03
0.01
0.04
0.02
0.11
0.34
0.21
0.08
0.14
0.01
0.24
0.25
0.04
0.20
0.13
0.25
0.02
0.21
0.24
0.03
0.04
0.03
0.04
0.12
0.08
0.23
0.03
0.35
0.21
0.18
0.23
0.09
0.12
Ratio of
Calculated
and Observed
Concentrations
0.20
0.30
0.10
0.17
0.48
1.00
0.41
0.33
0.18
0.19
0.03
0.14
0.07
1.00
1.00
0.72
0.67
0.54
0.08
0.77
0.93
0.27
0.59
1.18
0.78
0.07
0.81
2.00
0.12
0.40
0.10
0.24
0.43
0.73
1.44
0.12
1.17
0.60
0.60
0.88
0.33
0.20
3-99
-------
TABLE 3-39 (Continued)
Case
No.
77
78
79
80
81
82
83
84
85
86
87
88
89
90
Year/
Julian
Day
76/095
76/096
76/096
76/106
76/107
76/109
76/113
76/113
76/113
76/122
76/124
76/126
76/164
76/189
Hour
0200
1600
1700
0200
1200
1000
0600
0700
0800
2200
1200
0200
2400
1800
Monitor
D03H
D13H
D93H
D13H
D93H
D23H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D13H
D23H
D63H
D13H
D93H
Observed
Concentration
(ppm)
0.27
0.30
0.16
0.60
0.16
0.64
0.10
0.26
0.35
0.15
0.36
0.21
0.31
0.25
0.31
0.25
0.27
0.29
0.28
0.30
0.10
Calculated
Concentration
(ppm)
0.12
0.02
0.35
0.28
0.15
0.20
0.15
0.35
0.18
0.10
0.35
0.16
0.24
0.18
0.27
0.21
0.22
0.10
0.06
0.00
0.14
Mean Ratio (MR)
Ratio of
Calculated
and Observed
Concentrations
0.44
0.07
2.19
0.47
0.94
0.31
1.50
1.35
0.51
0.67
0.97
0.76
0.77
0.72
0.87
0.84
0.81
0.34
0.21
0.00
1.40
0.48
3-100
-------
TABLE 3-40
COMPARISON OF CALCULATED AND OBSERVED 1-HOUR S02 CONCENTRATIONS
FOR THE DOW CASES FOR COMBINATION 5 IN TABLE 3-35
Case
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Year/
Julian
Day
75/314
75/314
75/314
75/314
75/314
75/314
75/316
75/324
75/324
75/325
75/326
75/326
75/335
75/350
75/350
75/350
75/350
75/351
75/351
75/351
75/352
75/352
75/354
75/354
Hour
0100
0200
0500
1300
2300
2400
1700
1200
1300
0800
1500
1600
0400
1500
1900
2000
2100
0400
0700
1000
1100
1300
1200
1300
Monitor
D33H
D63H
D73H
D33H
D73H
D63H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D63H
D23H
D63H
D63H
D83H
D63H
D83H
D23H
D33H
D23H
D23H
D23H
D33H
D13H
D93H
D13H
D93H
D13H
D13H
D83H
D13H
D53H
D03H
D53H
Observed
Concentration
(ppm)
0.36
0.11
0.12
0.26
0.13
0.49
0.14
0.27
0.28
0.19
0.30
0.15
0.12
0.33
0.39
0.55
0.27
0.27
0.30
0.13
0.30
0.12
0.38
0.12
0.30
0.27
0.29
0.10
0.28
0.23
0.14
0.25
0.35
0.25
0.11
0.31
0.25
0.36
0.10
Calculated
Concentration
(ppm)
0.21
0.12
0.16
0.16
0.11
0.20
0.09
0.22
0.03
0.04
0.04
0.04
0.12
0.18
0.08
0.13
0.20
0.00
0.00
0.18
0.01
0.06
0.01
0.03
0.03
0.27
0.31
0.10
0.39
0.01
0.41
0.09
0.35
0.21
0.00
0.15
0.03
0.00
0.07
Ratio of
Calculated
and Observed
Concentrations
0.58
1.09
1.33
0.62
0.85
0.41
0.64
0.81
0.11
0.21
0.13
0.27
1.00
0.55
0.21
0.24
0.74
0.00
0.00
1.38
0.03
0.50
0.03
0.25
0.10
1.00
1.07
1.00
1.39
0.04
2.93
0.36
1.00
0.84
0.00
0.48
0.12
0.00
0.70
3-101
-------
TABLE 3-40 (Continued)
Case
No.
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
Year/
Julian
Day
75/354
75/354
75/354
75/354
76/003
76/003
76/005
76/008
76/008
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/028
76/028
76/038
76/039
76/039
76/042
76/043
76/046
76/046
Hour
1400
1600
1700
2400
0100
1100
1300
1600
1700
0300
0400
0500
0600
0700
0800
2100
2200
1600
1700
2400
0100
0200
1200
0900
1700
1800
Monitor
D03H
D03H
D03H
D53H
D63H
D83H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D13H
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D93H
D93H
D13H
D93H
D13H
D93H
D23H
D23H
D33H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D93H
D33H
D23H
D63H
D63H
D83H
Observed
Concentration
(ppm)
0.65
0.49
0.44
0.35
0.26
0.23
0.28
0.20
0.30
0.27
0.14
0.30
0.10
0.37
0.20
0.26
0.21
0.33
0.18
0.50
0.59
0.30
0.34
0.21
0.48
0.33
0.21
0.32
0.44
0.19
0.47
0.15
0.45
0.18
0.47
0.16
0.40
0.16
0.26
0.22
0.29
0.81
0.23
Calculated
Concentration
(ppm)
0.03
0.02
0.07
0.24
0.02
0.06
0.23
0.26
0.25
0.00
0.11
0.08
0.05
0.59
0.07
0.58
0.05
0.16
0.00
0.34
0.35
0.00
0.07
0.20
0.39
0.55
0.10
0.20
0.12
0.18
0.16
0.07
0.09
0.04
0.07
0.04
0.31
0.02
0.16
0.04
0.31
0.16
0.07
Ratio of
Calculated
and Observed
Concentrations
0.05
0.04
0.16
0.69
0.08
0.26
0.82
1.30
0.83
0.00
0.79
0.27
0.50
1.59
0.35
2.23
0.24
0.48
0.00
0.68
0.59
0.00
0.21
0.95
0.81
1.67
0.48
0.63
0.27
0.95
0.34
0.47
0.20
0.22
0.15
0.25
0.78
0.13
0.62
0.18
1.07
0.20
0.30
3-102
-------
TABLE 3-40 (Continued)
Case
No.
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
Year/
Julian
Day
76/046
76/049
76/049
76/049
76/052
76/052
76/052
76/065
76/065
76/065
76/066
76/067
76/067
76/067
76/067
76/073
76/074
76/074
76/074
76/074
76/084
76/084
7 6 /OH 4
76/094
76/094
76/094
76/095
Hour
1900
1700
1800
1900
1300
1500
1600
1000
1100
1500
2400
0200
1400
1500
1700
0200
1300
1900
2000
2400
0200
0300
0400
1800
2000
2300
0200
Monitor
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D03H
D03H
D03H
D63H
D83H
D83H
D63H
D83H
D13H
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D93H
D23H
D23H
D23H
D03H
D03H
D03H
D03U
Observed
Concentration
(ppm)
0.42
0.18
0.40
0.15
0.29
0.15
0.28
0.16
0.39
0.28
0.29
0.11
0.34
0.29
0.12
0.26
0.12
0.31
0.27
0.15
0.34
0.11
0.32
0.29
0.26
0.12
0.25
0.10
0.29
0.17
0.28
0.11
0.16
0.25
0.30
0.35
0.30
0.26
0.27
0.59
0.27
Calculated
Concentration
(ppm)
0.04
0.03
0.20
0.17
0.12
0.05
0.05
0.03
0.01
0.04
0.02
0.11
0.34
0.21
0.07
0.14
0.01
0.23
0.32
0.04
0.23
0.12
0.29
0.02
0.22
0.29
0.03
0.05
0.03
0.04
0.12
0.10
0.36
0.02
0.34
0.20
0.18
0.23
0.08
0.12
0.12
Ratio of
Calculated
and Observed
Concentrations
0.10
0.17
0.50
1.13
0.41
0.33
0.18
0.19
0.03
0.14
0.07
1.00
1.00
0.72
0.58
0.54
0.08
0.74
1.19
0.27
0.68
1.09
0.91
0.07
0.85
2.42
0.12
0.50
0.10
0.24
0.43
0.91
2.25
0.08
1.13
0.57
0.60
0.88
0.30
0.20
0.44
3-103
-------
TABLE 3-40 (Continued)
Case
No.
78
79
80
81
82
83
84
85
86
87
88
89
90
Year/
Julian
Day
76/096
76/096
76/106
76/107
76/109
76/113
76/113
76/113
76/122
76/124
76/126
76/164
76/189
Hour
1600
1700
0200
1200
1000
0600
0700
0800
2200
1200
0200
2400
1800
Monitor
D13H
D93H
D13H
D93H
D23H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D13H
D23H
D63H
D13H
D93H
Observed
Concentration
(ppm)
0.30
0.16
0.60
0.16
0.64
0.10
0.26
0.35
0.15
0.36
0.21
0.31
0.25
0.31
0.25
0.27
0.29
0.28
0.30
0. 10
Calculated
Concentration
(ppm)
0.03
0.35
0.42
0.15
0.20
0.16
0.39
0.18
0.10
0.39
0.17
0.27
0.19
0.32
0.33
0.32
0.10
0.07
0.00
0.14
Mean Ratio (MR)
Ratio of
Calculated
and Observed
Concentrations
0.10
2.19
0.70
0.94
0.31
1.60
1.50
0.51
0.67
1.08
0.81
0.87
0.76
1.03
1.32
1.19
0.34
0.25
0.00
1.40
0.54
3-104
-------
TABLE 3-41
COMPARISON OF CALCULATED AND OBSERVED 1-HOUR S02
CONCENTRATIONS FOR THE DOW CASES FOR
COMBINATION 6 in TABLE 3-35
Case
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Year/
Julian
Day
75/314
75/314
75/314
75/314
75/314
75/314
75/316
75/324
75/324
75/325
75/326
75/326
75/335
75/350
75/350
75/350
75/350
75/351
75/351
75/351
75/352
75/352
75/354
Hour
0100
0200
0500
1300
2300
2400
1700
1200
1300
0800
1500
1600
0400
1500
1900
2000
2100
0400
0700
1000
1100
1300
1200
Monitor
D33H
D63H
D73H
D33H
D73H
D63H
D63H
D83H
D83H
D83H
D63H
D83H
D63H
D83H
D63H
D63H
D23H
D63H
D63H
D83H
D63H
1)8 3H
1)2 3H
D33H
D23H
D23H
D23H
D33H
D13H
D93H
D13H
D93H
D13H
D13H
D83H
D13H
D53H
Observed
Concentration
(ppm)
0.36
0.11
0.12
0.26
0.13
0.49
0.14
0.27
0.28
0.19
0.30
0.15
0.12
0.33
0.39
0.55
0.27
0.27
0.30
0.13
0.30
0.12
0.38
0.12
0.30
0.27
0.29
0.10
0.28
0.23
0.14
0.25
0.35
0.25
0.11
0.31
0.25
Calculated
Concentration
(ppm)
0.21
0.14
0.16
0.16
0.11
0.22
0.10
0.24
0.03
0.04
0.04
0.04
0.12
0.19
0.09
0.15
0.20
0.00
0.00
0.18
0.01
0.06
0.01
0.03
0.03
0.28
0.32
0.10
0.44
0.01
0.46
0.10
0.38
0.24
0.00
0.16
0.04
Ratio of
Calculated
and Observed
Concentrations
0.58
1.27
1.33
0.62
0.85
0.45
0.71
0.89
0.11
0.21
0.13
0.27
1.00
0.58
0.23
0.27
0.74
0.00
0.00
1.38
0.03
0.50
0.03
0.25
0.10
1.04
1.10
1.00
1.57
0.04
3.29
0.40
1.09
0.96
0.00
0.52
0.16
5-105
-------
TABLE 3-41 (Continued)
Case
No.
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
Year/
Julian
Day
75/354
75/354
75/354
75/354
75/354
76/003
76/003
76/005
76/008
76/008
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/028
76/028
76/038
76/039
76/039
76/042
76/043
Hour
1300
1400
1600
1700
2400
0100
1100
1300
1600
1700
0300
0400
0500
0600
0700
0800
2100
2200
1600
1700
2400
0100
0200
1200
0900
Monitor
D03H
D53H
D03H
D03H
D03H
D53H
D63H
D83H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D13H
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D93H
D93H
D13H
D93H
D13H
D93H
D23H
D23H
D33H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D93H
D33H
Observed
Concentration
(ppm)
0.36
0.10
0.65
0.49
0.44
0.35
0.26
0.23
0.28
0.20
0.30
0.27
0.14
0.30
0.10
0.37
0.20
0.26
0.21
0.33
0.18
0.50
0.59
0.30
0.34
0.21
0.48
0.33
0.21
0.32
0.44
0.19
0.47
0.15
0.45
0.18
0.47
0.16
0.40
0.16
0.26
Calculated
Concentration
(ppm)
0.00
0.07
0.03
0.02
0.07
0.24
0.02
0.07
0.24
0.28
0.26
0.00
0.11
0.08
0.06
0.65
0.08
0.64
0.05
0.19
0.00
0.38
0.40
0.00
0.08
0.21
0.40
0.60
0. 10
0.21
0.]3
0.18
0.16
0.08
0.09
0.04
0.07
0.04
0.34
0.02
0.16
Ratio of
Calculated
and Observed
Concentrations
0.00
0.70
0.05
0.04
0.16
0.69
0.08
0.30
0.86
1.40 1
i
0.87
0.00
0.79
0.27
0.60
1.76
0.40
2.46
0.24
0.58
0.00
0.76
0.68
0.00
0.24
1.00
0.83
1.82
0.48
0.66
0.30
0.95
0.34
0.53
0.20
0.22
0.15
0.25
0.85
0.13
0.62
3-106
-------
TABLE 3-41 (Continued)
Case
No.
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
(>4
6r>
66
67
68
69
70
71
72
73
Year/
Julian
Day
76/046
76/046
76/046
76/049
76/049
76/049
76/052
76/052
76/052
76/065
76/065
76/065
76/066
76/067
76/06 /
76/067
76/007
76/073
76/074
76/074
76/074
76/074
76/084
76/084
76/084
Hour
1700
1800
1900
1700
1800
1900
1300
1500
1600
1000
1100
1500
2400
0200
1 400
1 500
1700
0200
1300
1900
2000
2400
0200
0300
0400
Monitor
D23H
D63H
D63H
D83H
D63H
D83H
D63H
118311
D63H
D83H
D63H
D83H
T)03H
D03H
D03H
D63H
D83H
D83H
D63H
D83H
D13H
D93H
D13H
D93II
D13II
09311
DI3II
1)1 ill
06511
D83II
D63II
D831I
D63H
D83H
D6311
D83H
D13H
D93H
D23H
D23H
D231I
Observed
Concentration
(ppm)
0.22
0.29
0.81
0.23
0.42
0.18
0.40
0. 15
0.29
0.15
0.28
0.16
0.39
0.28
0.29
0.11
0.34
0.29
0.12
0.26
0.12
0.31
0.27
0. 15
0. 34
0. 1 1
0.37
0.29
0.26
0. 12
0.25
0.10
0.29
0. 17
0.28
0.11
0. 16
0.25
0.30
0.35
0.30
Calculated
Concentration
(ppm)
0.05
0.32
0.17
0.08
0.04
0.03
0.21
0.18
0.12
0.06
0.05
0.03
0.01
0.04
0.02
0.11
0.38
0.23
0.08
0.16
0.01
0.24
0.36
0.04
0.26
0. 13
0.3?
0.03
0.23
0.31
0.04
0.05
0.03
0.05
0.13
0.10
0.40
0.03
0.35
0.21
0.18
Ratio of
Calculated
and Observed
Concentrations
0.23
1.10
0.21
0.35
0. 10
0.17
0.53
1.20
0.41
0.40
0.18
0.19
0.03
0.14
0.07
1.00
1.12
0.79
0.67
0.62
0.08
0.77
1.33
0.27
0.76
1 .18
1 . 00
0. 10
0.88
2.58
0.16
0.50
0.10
0.29
0.46
0.91
2.50
0.12
1.17
0.60
0.60
3-107
-------
TABLE 3-41 (Continued)
Case
No.
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
Year/
Julian
Day
76/094
76/094
76/094
76/095
76/096
76/096
76/106
76/107
76/109
76/113
76/113
76/113
76/122
76/124
76/126
76/164
76/189
Hour
1800
2000
2300
0200
1600
1700
0200
1200
1000
0600
0700
0800
2200
1200
0200
2400
1800
Monitor
D03H
D03H
D03H
D03H
D13H
D93H
D13H
D93H
D23H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D13H
D23H
D63H
D13H
D93H
Observed
Concentration
(ppm)
0.26
0.27
0.59
0.27
0.30
0.16
0.60
0.16
0.64
0.10
0.26
0.35
0.15
0.36
0.21
0.31
0.25
0.31
0.25
0.27
0.29
0.28
0.30
0.10
Calculated
Concentration
(ppm)
0.24
0.09
0.12
0.12
0.03
0.36
0.47
0.15
0.20
0.16
0.42
0.19
0.11
0.43
0.18
0.29
0.20
0.34
0.36
0.36
0.10
0.08
0.00
0.14
Mean Ratio (MR)
Ratio of
Calculated
and Observed
Concentrations
0.92
0.33
0.20
0.44
0.10 j
2.25
0.78
0.94
0.31
1.60
1.62
0.54
0.73
1.19
0.86
0.94
0.80
1.10
1.44
1.33
0.34
0.29
0.00
1.40
0.58
3-108
-------
Table 3-42 summarizes the results of comparisons of concurrent
calculated and observed 1-hour SO concentraLtons for the 143 observed
concentrations greater than or equal to 0.10 ppm during the 90 hours
selected for model testing. Because the equation for the RMSE cannot be
applied to zero concentrations, the concentration was set equal to 0.01
ppm for all cases with zero calculated concentrations. (The threshold
and accuracy of the S0« monitors are about 0.01 ppm.) Whether or not
building wake effects are included in the ISC Model calculations, the
correspondence between calculated and observed concentrations obtained
in Urban Mode 2 is superior to that obtained in the Rural Mode. This
result supports the hypothesis of enhanced turbulence at the DOW plant.
Although the building wake effects and stack-tip downwash options improve
the model's performance in both the Rural Mode and Urban Mode 2, the MR
values suggest that there may be errors in the model inputs or a bias in
the model towards underestimating concentrations for the DOW data set.
We are not certain that the MR, RMSE and percentage error bands
are the most appropriate measures of model performance for the DOW data
set because the density of the DOW monitoring network is inadequate
to determine the locations of the plume centerlines. The current Guide-
line on Air Quality Models (EPA, 1978) suggests that, as an alternative
to direct comparisons of concurrent calculated and observed concentrations,
short-term dispersion models may be evaluated by comparing the upper per-
centilos of the calculated and observed concentration frequency distrib-
utions. Additionally, Turner (1979) observes that a comparison of the
maximum concentration observed at any point with the maximum concentration
calculated at any point is of Interest for regulatory purposes. We
therefore- used Tables 3-36 and 3-41 to generate the cumulative frequency
distributions of calculated and observed hourly SO concentrations for
the DOW data set shown fn Figure 3-9. We point out that the concentra-
tions calculated by the ISC Model for model option Combination 1 are the
same as calculated by conventional modeling techniques (for example
the MPTER Model). As shown by Figure 3-9, the upper end of the observed
3-109
-------
W
H
0)
o
c
cd
g
O
cu
rH
cu
o
g
O
CO
3
CO
3
S
3
n)
*
00
'O
PQ
CO
CT^ *r
rH E-
v-' 1
^
CO C.
M n
00 4-
•H C/
M
pq
rH
cu
o
^
u
M
~ 1
e
•r
^1
t3 <_
O
C
U -r
C/5 4-
M e
C
(Nl
'•M
O
VJ
o
4J
a
cd
•H
•H
&-S
w
C/D
§
rt
CO
4_l
a
cu
M-4
4-1
H
X
* JS
< CO
' §
! c
s &
J O
3 O
0)
T3
o
J
> S
> 3
?;
C
t o
I -H
-* 4-J
cd
O O> -^ vO OO CTv
i — 1 CN CO -^ <}- si- m co CT\
co r*^- co in ^d~ r—i
CM oo oo ~3-
3-110
-------
8?
ai
o>
cri
o>
op
o>
o>
8
8
$
UJ
O 8
~ h-
10
CVJ
o
O
b
00
(ujdd) NOIlVdlN30NOO 2OS
FIGURE 3-9. Comparison of calculated and observed cumulative 1-hour S02 concen-
tration frequency distributions for the DOW data set.
3-111
-------
concentration frequency distribution is much more closely matched by
the ISC Model for Combination 6 than for Combination 1. It should be
emphasized that the hourly concentrations used to form the observed
concentration frequency distribution in Figure 3-9 were restricted to
concentrations greater than or equal to 0.10 ppm. Thus, the observed
concentration frequency distribution reflects the presence at the monitor
of emissions from one or both of the DOW power houses during every hour.
However, if the wind directions used as input to the model are assumed
to be representative of the winds defining plume trajectories during
every hour, the wind directions imply that emissions from one or both of
the DOW power houses were not present at the monitors during every hour.
Thus, it is not unexpected that the ISC Model underpredicts the lower end
of the observed concentration frequency distribution in Figure 3-9.
3-112
-------
SECTION 4
SUGGESTIONS FOR MODEL IMPROVEMENT
The results of the ISC Model calculations described in Section 3
for the DOW and Millstone (Johnson, et al., 1975) data sets indicate
that the model's building wake effects option may have a systematic
tendency to underestimate the concentrations produced in a building's
wake region by buoyant or slightly buoyant stack emissions subject to
building wake effects. Tn addition to the emission rate, the wind speed
and the mixing height, the basic components of a Gaussian plume model
arc the effective emission height and the lateral and vertical dis-
persion coefficients. Although the uncertainties in the DOW emissions
are unknown, the emissions for the Millstone data set are estimated by
Johnson, e^ al. to be accurate to within about 5 percent. Also, the
wind speeds are known with a relatively high degree of confidence for
both the Millstone and DOW data sets, and the restriction on vertical
mixing at the top of the surface mixing layer is unlikely to have affected
the observed concentrations for either data set. Additionally, the
Islitzer (1965) and Millstone data sets do not reveal any systematic
errors in the lateral dispersion coefficients calculated by the ISC
Model in the wake region. It follows that the apparent systematic ten-
dency toward underpredictions in the wake region results from errors in
the effective emission heights, the vertical dispersion coefficients or
both. Consequently, modifications that might improve the ISC Model's
performance Include adjustments In the effective emission heights to
account for the effects of downwash and the use of different vertical
dispersion coefficients.
Of the various data sets used to test^the ISC Model, the fol-
lowing features are unique to the Millstone data set: (1) The Millstone
experiments involved slightly buoyant releases from a squat building with
a stack height to building height ratio less than 1.2, and (2) Visual obser-
vations indicated that the plume was in a condition of partial entrainment in
4-1
-------
which portions of the plume were intermittently entrained into the
building's cavity zone and effectively acted as a ground-level source.
Thus, we suggest the following change in the way that the ISC model
calculates effective emission heights for insignificantly buoyant emis-
sions from stacks located on or adjacent to squat buildings with stack
height to building height ratios less than 1.2:
H{x)
where
h + Ah{x} ; h + Ah {2h, } > 1.2 h,
' m b - b
; h.
1.2h
(4-1)
H(x>
h
Ah(x}
Ahm(2hb}
effective emission height at downwind distance x
physical stack height
plume rise due to momentum and/or buoyancy at downwind
distance x, calculated by Equation (2-4) or Equation
(2-7) in the ISC Model User's Guide
plume rise due to momentum alone at a downwind distance
of 2h, , where h, is the building height
We point out that the addition of Equation f4-l) to the ISC Model will have
no effects on the model's performance for the DOW data set because the stack
height to building height ratios exceed 1.2. However, the addition of
Equation (4-1) to the ISC Model will yield essentially the same model per-
formance for the Millstone data set as obtained for Case C (see Table
3-11 in Section 3.1.3). As explained above, we believe that the model's
tendency to underestimate concentrations for Case C. as indicated by the
mean ratio (MR) of calculated to observed concentrations, may be attributed
to vertical turbulent intensities that are smaller than the median values
nssociated with Pusquill-Gifford vortical dispersion coefficients.
Briggs (1969), Fay, et aJU » (1970) and other developers of plume
rise equations have expressed the belief that their plume rise equations
4-2
-------
may not be applicable when the ratio of the stack exit velocity to the
mean wind speed is less than about 1.5 because of downwash of the plume
into the low pressure region in the lee of the stack. Additionally,
Briggs (1969, p. 6) points out that these adverse effects may be abated
for stacks with low Froude numbers on the order of unity. Bjorklund
and Bowers (1979) conclude from an analysis of the Bringfelt (1968) plume
rise data that the effects of stack-tip downwash on buoyant plume rise
are insignificant for stacks with Froude numbers (as defined by Briggs,
1969) less than 1.0 and, depending on the stack, may or may not be im-
portant for stacks with Froude numbers between 1.0 and 3.0. For stacks
with Froude numbers greater than 3.0, Bjorklund and Bowers conclude that
stack-tip downwash significantly affects buoyant plume rise when the
ratio of the stack exit velocity to the mean wind speed at stack height
is less than about 1.5.
To account for the effects of stack-tip downwash, Cramer, et_ ad.
(1975) multiplied the unmodified plume rise Ah by the semi-empirical cor-
rection factor
; u{h} < V /I.5
™~ s
; V /I.5 < u{h} < V
(4-2)
0 ; u{h} >^ V
where V is the stack exit velocity and u{h} is the mean wind speed at
S
stack height. As shown by Equation (4-2), plume rise is precluded when
the wind speed at stack height equals or exceeds the stack exit velocity.
The development of Equation (4-2) is discussed in Appendix G of the report
by Bjorklund and Bowers. Because the full-load Froude numbers for the DOW
stacks exceed 3.0, the substitution in the ISC Model of Equation (4-2)
L'or the; Briggs (1973) stack-tip downwash correction currently used by the
4-3
-------
model might improve the model's performance for the DOW data set if
stack-tip downwash is the cause of the ISC Model's apparent tendency to
underpredict concentrations for the data set.
Scire and Schulman (1980) propose a "downwash radius" correction
to the Briggs plume rise equations to account for the effects of building
wakes on initial entrainment and hence on buoyant plume rise. If momentum
effects are considered, the generalized solution to Equation (12) in the
paper by Scire and Schulman is of the form (Dumbauld, et al., 1973)
Ah{x) =
M{x] + B{x}
R
3
1/3
R
(4-3)
where M{x} is the momentum term at downwind distance x, B{x) is the
buoyancy term, R is the "downwash radius" and 3 is the entrainment
coefficient. (See Equations (2-4) and (2-7) in the TSC Model User's
Guide for definitions of M{x} and B{x}). On the basis of the results of
wind tunnel experiments (Huber and Snyder, 1976), Scire and Schulman
hypothesize that R is approximately equal to the building height h, .
The addition of Equation (4-3) to the TSC Model might also improve the
performance of the model for the DOW data set if the effects of building
wakes on buoyant plume rise are the cause of the model's apparent tendency
to underestimate concentrations for the data set. However, it is import-
ant to note that Equation (4-3) is not applicable to free standing
stacks and thus cannot account for the effects on buoyant plume rise of
stack-tip downwash.
As a test of the Cramer, et al. (1975) stack-tip downwash cor-
rection, we substituted Equation (4-2) for the Briggs (1973) stack-tip
downwash correction in the TSCST program and repeated the ISC Model calcu-
lations for the DOW data set. Concentrations were calculated in both the
Rural Mode and Urban Mode 2 using the distance-dependent plume rise and
building wake effects options. Because of the uncertainties about whether
or not Equation (4-2) should be applied to stacks with Frotide numbers less
4-4
-------
than 3.0, Equation (4-2) was first applied to all stacks during each hour
and then was applied only to stacks with Froude numbers greater than or
equal to 3.0 during each hour. (The Froude numbers for the various stacks
varied significantly from hour to hour.) Similarly, as a. test of the
Scire and Schulman (1980) "downwash radius" correction, we converted the
plume rise equations in the ISCST program to the form of Equation (4-3)
and repeated the ISC Model calculations for the DOW data set. The concen-
tration calculations for the Scire and Schulman "downwash radius" cor-
rection were also made in both the Rural Mode and Urban Mode 2 using
the distance-dependent plume rise and building wake effects options.
The results of the ISC Model calculations for the Cramer, et^ aJL. stack-
tip downwash correction are given in Tables 4-1 through 4-4 and the results
of the calculations for the Scire and Schulman "downwash radius" cor-
rection are given in Tables 4-5 and 4-6.
Table 4-7 summarizes the results of the various ISC Model cal-
culations for the DOW data set. In both the Rural Mode and Urban Mode 2,
the overall performance of the ISC Model - as indicated by the mean
ratio (MR) of calculated to observed concentrations, the root mean
square error (RMSE) and the percentage error bands - is significantly
Improved by both the Cramer, et a1. stack-tip downwash correction and
the Scire and Schulman "downwash radius" correction. However, as
discussed in Section 3.4.3, the correspondence between the upper per-
centiles of the calculated and observed concentration frequency distrib-
utions may be a better measure of model performance for the DOW data set
than the MR, RMSE and percentage error bands because the density of the
DOW monitoring network is inadequate to determine the locations of the
plume centerlines.
Figures 4-1 and 4-2 compare the observed 1-hour SO,, concentration
cumulative frequency distribution for the DOW data set with the concen-
tration frequency distributions calculated by the ISC Model using the
Cramer, et al. stack-tip downwash correction for all stacks and for
4-5
-------
TABLE 4-1
COMPARISON OF CALCULATED AND OBSERVED 1-HOUR SO CONCENTRATIONS FOR THE
DOW CASES FOR COMBINATION 2 IN TABLE 3-35 AND THE CRAMER, ET AL.
(1975) STACK-TIP DOWNWASH CORRECTION FOR ALL STACKS
Case
No.
1
Year/
Julian
Day
75/314
2 ; 75/314
3
4
5
6
75/314
75/314
75/314
75/314
\
f
7 75/316
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
75/324
75/324
75/325
75/326
75/326
75/335
75/350
75/350
75/350
75/350
75/351
75/351
75/351
75/352
75/352
75/354
Hour
0100
0200
0500
1300
2300
2400
1700
1200
1300
0800
1500
1600
0400
1500
1900
2000
2100
0400
0700
1000
1100
Monitor
D33H
D63H
D73H
D33H
Observed
Concentration
(ppm)
0.36
0.11
0.12
0.26
D73H 0.13
D63H
Calculated
Concentration
(ppm)
0.65
0.07
0.14
0.41
0.08
0.49 0.21
D63H 0.14 0.20
D83H 0.27 0.60
D63H 0.28
D83H 0.19
D63H j 0.30
D83H
D63H
D83H
D63H
D63H
D23H
D63H
D63H
D83H
D63H
D83H
D23H
D33H
D23H
D23H
D23H
D33H
D13H
D93H
D13H
D93H
D13H
D13H
D83H
1300 • D13H
1200
D53H
0.15
0.12
0.33
0.39
0.55
0.27
0.27
0.30
0.13
0.30
0.12
0.38
0.12
0.30
0.27
0.29
0.10
0.28
0.23
0. 14
0.25
0.35
0.25
0.11
0. 31
0.25
0.00
0.01
0.01
0.01
0.06
Ratio of
Calculated
and Observed
Concentrations
1.81
0.64
1.17
1.58
0.62
0.43
1.43
2.22
0.00
0.05
0.03
0.07
0.50
0.19 0.58
0.09 0.23
0.27 0.49
0.49
0.00
0.00
1.81
o.oo ;
0.00
0.26 2.00
0.00 0.00
0.06
0.00
0.00
0.00
0.78
0.96
0.11
1.29
0.00
1.51
0.06
1 .10
1.19
0.00
0.37
0.01
0.50
0.00
0.00
0.00
2.89
3.31
1.10
4.61
0.00
10.78
0.24
3. 14
4.76
0.00
1.19
0.04
4-6
-------
TABLE 4-1 (Continued)
Case
No.
24
25
26
27
28
29
Year/
Julian
Day
75/354
75/354
75/354
75/354
75/354
76/003
I
30 76/003
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
76/005
76/008
76/008
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/028
76/028
76/038
76/039
76/039
76/042
76/043
Hour
1300
1400
1600
1700
2400
0100
1100
1300
1600
1700
0300
0400
0500
0600
0700
0800
2100
2200
1600
1700
2400
0100
0200
1200
0900
i
Monitor
D03H
D53H
D03H
D03H
D03H
D53H
D63H
D83H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D13H
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D93H
D93H
D13H
D93H
D13H
D93H
Observed
Concentration
(ppm)
0.36
0.10
0.65
0.49
0.44
0.35
0.26
0.23
0.28
0.20
0.30
0.27
0.14
0.30
0.10
0.37
0.20
0.26
0.21
0.33
0.18
0.50
0.59
0.30
0.34
0.21
0.48
0.33
0.21
D23H 0. 32
D23H
D33H
0.44
0.19
D63H 0.47
D83H 0.15
D63H
D83II
D63H
D83H
D13H
D93H
D33H
0.45
0.18
0.47
0.16
0.40
0.16
0.26
Calculated
Concentration
(ppm)
0.00
0.05
0.00
0.00
0.06
0.20
0.00
0.07
0.76
1.15
0.73
0.00
0.29
0.03
0.02
2.54
0.03
2.32
0.01
0.19
0.00
1.11
1.30
0.00
2.31
0.18
0.98
1.94
0.06
0.57
0.38
0.50
0.31
0.08
0.08
0.01
0.04
0.01
1.12
0.00
0.44
Ratio of
Calculated
and Observed
Concentrations
0.00
0.50 .
0.00
0.00
0.14
0.57
0.00
0.30
2.71
5.75
2.43
0.00
2.07
0.10
0.20
6.86
0.15
8.92
0.05
0.58
0.00
2.22
2.20
0.00
6.79
0.86
2.04
5.88
0.29
1.78
0.86
2.63
0.66
0.53
0.18
0.06
0.09
0.06
2.80
0.00
1.69
4-7
-------
TABLE 4-1 (Continued)
Cast
No.
49
1
Year/
i Julian
i T^ TTT
Day
76/046
j
Hour
1700
50 76/046 | 1800
i
51 76/046
i
1900
52 76/049 j 1700
1
53
54
55
56
57
58
59
60
61
62
i i
!
63
64 j
65
66
67
68
69
1
70
71
72
73
76/049
1 76/049
76/052
76/052
76/052
76/065
76/065
76/065
76/066
76/067
76/067
76/067
76/067
76/073
76/074
76/074
76/074
76/074
76/084
76/084
76/084
Monitoi
i
D23H
D63H
j D63H
D83H
D63H
1 D83H
1 D63II
!
j D83H
1800 1 D63H
1900
1300
1500
I 1600
| 1000
i D83H
D63H
D83H
D03H
D03H
D03H
D63H
| D83H
1100 I D83H
1500 | D63H
[ D83H
2400 j D13H
: D93H
0200 i D13H ]
1400
1500
1700
0200
1300
i
1900
2000
2400
0200
0300
0400
D93H
D13H !
D93H
D13H
D13H
D63H
D83H
D63H
D83H i
D63H I
i
D83H I
D63H
D83H
D13H
D93H
D23H
D23H
D23H
Observed
r Concentratior
Calculated
i ! Concentratior
(ppm) ; (ppm) i
Ratio of
Calculated
(
i
and Observed j
^FF ' | VFF ' Concentrations i
0.22
0.29
0.81
0.23
0.42
0.18
0.40
i
!
Of) A i i i r*. i
n i o >
. \J *-r | \J . -JU u [
0.69 ! 2.38 |
| 0.28 i n T; i
j 0.07
0.01
0.01
i 0.42
0.15 0.57
0.29 0.15
0.15 0.03
i 0.28 0.02
j 0.16
0.39
0.28
0.29
: 0.11
0.34
0.29
0.12 ' L».it ! J.I/
\ 0.30
Orv>
• \)£
0.06
1 05
J- • \J _J
3.80
0 S?
\J . -> £-
0.20
n n7
. -- | \J 9 \J 1
0.10 | 0.06
0.00 j 0.00
0.10 j 0.36
0.02 ' 0.07
0.24 i ? 10
0.92
j 0.51
0.14
1 <-. 1.U
2.71
1.76
1 1 7
0.26 1 0.21 1 0.81
0.12 0.00 i n nn
0.31
0.27
0.15
0.34
0. 11
0.32
0.29
0.26
0.12
0.25
0.10
0.29
0.17
0.28
0.11
0.16
0.25
0.30
0.35
0.30
0.46 | 1 . 48
0.98 ! 3.63
0.03 { 0.20
0.47 i ' i«
0.27
0.80
0.01
0.68
1.09
0.01
0.02 |
0.01
0.02
0.19
0.09
1.78
0.01
1.09
0.68 1
0.48
(
2.45
2.50
0.03
0 A?
•t. • \J iL.
9.08
0 04
W • \J^T
0.20
0 03
\J • \J J
0.12
0.68
0.82
11 13
4. J- • A -J
0.04
3.63
1.94
1.60
4-8
-------
TABLE 4-1 (Continued)
Case
No.
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
.
Year/
-luJ ian
Day
76/094
! 76/094
! 76/094
; 76/095
76/096
76/096
I
76/106
; 76/107
76/109
76/113
76/113
76/113
76/122
76/124
76/126
76/164
76/189
. Hour
1800
2000
; 2300
0200
1600
1700
0200
1200
1000
0600
0700
0800
2200
1200
0200
2400
1800
Monitor
D03H
DO 3 El
D03H
D03H
D13H
D93H
D13H
D93H
D23H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D13H
D23H
D63H
D13H
D93I1
Observed
Concentration
(ppm)
0.26
0.27
0.59
0.27
0.30
0.16
0.60
0.16
0.64
0.10
0.26
0.35
0.15
0.36
0.21
0.31
0.25
0.31
0.25
0.27
0.29
0.28
0.30
0.10
Men
i
Calculated
Concentration
(ppm)
0.47
0.05
0.09
0.17
0.00
0.89
1.17
0.16
0.63
0.13
1.39
0.48
0.14
1.15
0.42
0.48
0.54
0.76
0.55
1.12
0.26
0.02
0.00
0.18
n Ratio (MK)
Ratio of
| Calculated
and Observed
Concentrations
1.81
0.19
0.15
0.63
0.00
5.56
1.95
1.00
0.98
1.30
5.35
1.37
0.93
3.19
2.00
1.55
2.16
2.45
2.20
4.15
0.90
0.07
0.00
1.80
1.42
4-9
-------
TABLE 4-2
COMPARISON OF CALCULATED AND OBSERVED 1-HOUR S02 CONCENTRATIONS
FOR THE DOW CASES FOR COMBINATION 5 IN TABLE 3-35 AND THE
CRAMER, _ET AL. (1975) STACK-TIP DOWNWASH CORRECTION
FOR ALL STACKS
Case
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Year/
Julian
Day
75/314
75/314
75/314
75/314
75/314
75/314
75/316
75/324
75/324
75/325
75/326
75/326
75/335
75/350
75/350
75/350
75/350
75/351
75/351
75/351
75/352
75/352
75/354
Hour
0100
0200
0500
1300
2300
2400
1700
1200
1300
0800
1500
1600
0400
1500
1900
2000
2100
0400
0700
1000
1100
1300
1200
Monitor
D33H
D63H
D73H
D33H
D73H
D63H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D63H
D23H
D63H
D63H
D83H
D63H
D83H
D23H
D33H
D23H
D23H
D23H
D33H
D13H
D93H
D13H
D93H
D13H
D13H
D83H
D13H
D53H
Observed
Concentration
(ppm)
0.36
0.11
0.12
0.26
0.13
0.49
0.14
0.27
0.28
0.19
0.30
0.15
0.12
0.33
0.39
0.55
0.27
0.27
0.30
0.13
0.30
0.12
0.38
0.12
0.30
0.27
0.29
0.10
0.28
0.23
0.14
0.25
0.35
0.25
0.11
0.31
0.25
Calculated
Concentration
(ppm)
0.24
0.23
0.16
0.18
0.11
0.34
0.11
0.30
0.03
0.05
0.04
0.07
0.13
0.36
0.25
0.39
0.28
0.00
0.00
0.18
0.01
0.06
0.02
0.03
0.04
0.38
0.43
0.10
1.19
0.01
1.22
0.12
0.95
0.93
0.00
0.58
0.04
Ratio of
Calculated
and Observed
Concentrations
0.67
2.09
1.33
0.69
0.85
0.69
0.79
1.11
0.11
0.26
0.13
0.47
1.08
1.09
0.64
0.71
1.04
0.00
0.00
1.38
0.03
0.50
0.05
0.25
0.13
1.41
1.48
1.00
4.25
0.04
8.71
0.48
2.71
3.72
0.00
1.87
0.16
4-10
-------
TABLE 4-2 (Continued)
Case
No.
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
Year/
Julian
Day
75/354
75/354
75/354
75/354
75/354
76/003
76/003
76/005
76/008
76/008
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/028
76/028
76/038
76/039
76/039
76/042
76/043
Hour
1300
1400
1600
1700
2400
0100
1100
1300
1600
1700
0300
0400
0500
0600
0700
0800
2100
2200
1600
1700
2400
0100
0200
1200
0900
Monitor
D03H
D53H
D03H
D03H
D03H
D53H
D63H
D83H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D13H
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D93H
D93H
D13H
D93H
D13H
D93H
D23H
D23H
D33H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D93I1
D33H
Observed
Concentration
(ppm)
0.36
0.10
0.65
0.49
0.44
0.35
0.26
0.23
0.28
0.20
0.30
0.27
0.14
0.30
0.10
0.37
0.20
0.26
0.21
0.33
0.18
0.50
0.59
0.30
0.34
0.21
0.48
0.33
0.21
0.32
0.44
0.19
0.47
0.15
0.45
0.18
0.47
0.16
0.40
0.16
0.26
Calculated
Concentration
(ppm)
o.oo
0.08
0.04
0.03
0.10
0.25
0.03
0.08
0.39
0.61
0.37
0.01
0.12
0.12
0.08
1.65
0.09
1.63
0.07
0.46
0.00
1.15
1.19
0.00
0.09
0.42
0.51
J.43
0.13
0.25
0.16
0.23
0.24
0.14
0.14
0.07
0.11
0.06
0.66
0.02
0.19
Ratio of
Calculated
and Observed
Concentrations
0.00
0.80
0.06
0.06
0.23
0.71
0.12
0.35
1.39
3.05
1.23
0.04
0.86
0.40
0.80
4.46
0.45
6.27
0.33
1.39
0.00
2.30
2.02
0.00
0.26
2.00
1.06
4.33
0.62
0.78
0.36
1.21
0.51
0.93
0.31
0.39
0.23
0.38
1.65
0.13
0.73
4-11
-------
TABLE 4-2 (Continued)
Case
No.
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
Year/
Julian
Day
76/046
76/046
76/046
76/049
76/049
76/049
76/052
76/052
76/052
76/065
76/065
76/065
76/066
76/067
76/067
76/067
76/067
76/073
76/074
76/074
76/074
76/074
76/084
76/084
76/084
Hour
1700
1800
1900
1700
1800
1900
1300
1500
1600
1000
1100
1500
2400
0200
1400
1500
1700
0200
1300
1900
2000
2400
0200
0300
0400
Monitor
D23H
D63H
D63H
D83H
D63H
D83H
D63H
D83H
U63H
D83H
D63H
D83H
D03H
D03H
D03H
D63H
D83H
D83H
D63H
D83H
D13H
D93H
D13H
D93H
U13H
D93H
D13H
D13H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D93H
D23H
D23H
D23H
Observed
Concentration
(ppm)
0.22
0.29
0.81
0.23
0.42
0.18
0.40
0.15
0.29
0.15
0.28
0.16
0.39
0.28
0.29
0.11
0.34
0.29
0.12
0.26
0.12
0.31
0.27
0.15
0.34
0.11
0.32
0.29
0.26
0.12
0.25
0.10
0.29
0.17
0.28
0.11
0.16
0.25
0.30
0.35
0.30
Calculated
Concentration
(ppm)
0.05
0.42
0.22
0.12
0.05
0.04
0.29
0.30
0.15
0.08
0.06
0.04
0.02
0.06
0.03
0.13
0.47
0.29
0.08
0.20
0.01
0.31
0.60
0.05
0.38
0.14
0.46
0.05
0.34
0.58
0.06
0.07
0.05
0.07
0.19
0.21
1.05
0.03
0.47
0.28
0.24
Ratio of
Calculated
and Observed
Concentrations
0.23
1.45
0.27
0.52
0.12
0.22
0.73
2.00
0.52
0.53
0.21
0.25
0.05
0.21
0.10
1.18
1.38
1.00
0.67
0.77
0.08
1.00
2.22
0.33
1.12
1.27
1.44
0.17
1..31
4.83
0.24
0.70
0.17
0.41
0.68
1.91
6.56
0.12
1.57
0.80
0.80
4-12
-------
TABLE 4-2 (Continued)
Case
No.
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
Year/
Julian
Day
76/094
76/094
76/094
76/095
76/096
76/096
76/106
76/107
76/109
76/113
76/113
76/113
76/122
76/124
76/126
76/164
76/189
Hour
1800
2000
2300
0200
1600
1700
0200
1200
1000
0600
0700
0800
2200
1200
0200
2400
1800
Monitor
D03H
D03H
D03H
D03H
D13H
D93H
D13H
D93H
D23H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D13H
D23H
D63H
D13H
D93H
Observed
Concentration
(ppm)
0.26
0.27
0.59
0.27
0.30
0.16
0.60
0.16
0.64
0.10
0.26
0.35
0.15
0.36
0.21
0.31
0.25
0.31
0.25
0.27
0.29
0.28
0.30
0.10
Calculated
Concentration
(ppm)
0.28
0.11
0.15
0.16
0.06
0.46
1.08
0.19
0.26
0.16
0.70
0.26
0.15
0.69
0.24
0.49
0.28
0.63
0.82
0.76
0.12
0.13
0.01
0.19
Mean Ratio (MR)
Ratio of
Calculated
and Observed
Concentrations
1.08
0.41
0.25
0.59
0.20
2.88
1.80
1.19
0.41
1.60
2.69
0.74
1.00
1.92
1.14
1.58
1.12
2.03
3.28
2.81
0.41
0.46
0.03
1.90
1.02
4-13
-------
TABLE 4-3
COMPARISON OF CALCULATED AND OBSERVED 1-HOUR S02 CONCENTRATIONS FOR THE
DOW CASES FOR COMBINATION 2 IN TABLE 3-35 AND THE CRAMER, ET AL.(1975)
STACK-TIP DOWNWASH CORRECTION FOR STACKS WITH FROUDE NUMBERS ABOVE 3.0
Case
No.
1
Year/
.' Julian
Day
i 75/314
1
2 75/314
i
3
4
5
6
7
8
9
75/314
75/314
75/314
75/314
75/316
75/324
75/324
10 | 75/325
11
12
13
14
15
16
17
18
19
20
21
22
23
75/326
75/326
75/335
75/3W
75/350
75/350
75/350
75/351
75/351
75/351
75/352
75/352
75/354
Hour
0100
0200
0500
1300
2300
2400
1700
1200
1300
0800
1500
1600
0400
1500
1900
2000
2100
0400
0700
1000
1100
1300
1200
Monitor
D33H
D63H
D73H
D33H
D73H
D63H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D63H
D23H
D63H
D63H
D83H
D63H
D83H
D23H
D33H
D23H
D23H
D23H
D33H
D13H
D93H
D13H
D93H
D13H
D13H
D83H
D13H
D53H
, .. , -. _, .« ..._.•..,., .- -.
Observed
Concentration
(ppm)
0.36
0.11
0.12
0.26
0.13
0.49
0.14
0.27
0.28
0.19
0.30
0.15
0.12
0.33
0.39
0.55
0.27
0.27
0.30
0.13
0.30
0.12
0.38
0.12
0.30
0.27
0.29
0.10
0.28
0.23
0.14
0.25
0.35
0.25
0.11
0.31
0.25
Calculated
Concentration
(ppm)
0.28
0.01
0.14
0.20
0.08
0.05
0.20
0.59
0.00
0.00
0.01
0.00
0.05
0.14
0.09
0.27
0.49
0.00
0.00
0.23
0.00
0.05
0.00
0.00
0.00
0.61
0.71
0.11
0.57
0.00
1.17
0.05
0.13
0.04
0.00
0.02
0.00
Ratio of
Calculated
and Observed
Concentrations
0.78
0.09
1.17
0.77
0.62
0.10
1.43
2.19
0.00
0.00
0.03
0.00
0.42
0.42
0.23
0.49
1.81
0.00
0.00
1.77
0.00
0.42
0.00
0.00
0.00
2.26
2.45
1.10
2.04
0.00
8.36
0.20
0.37
0.16
0.00
0.06
0.00
4-14
-------
TABLE 4-3 (Continued)
Case
No.
24
?5
26
27
28
29
30
31
32
33
34
35
36
37
38
19
40
41
42
43
44
45
46
47
48
Year/
1 Julian
Day
r
75/354
75/354
75/354
75/354
75/J54
76/003
76/003
76/005
76/008
76/008
76/014
76/014
76/014
76/014
76/014
70/014
76/014
76/014
76/028
76/028
76/038
76/039
76/039
76/042
76/043
Hour
1300
1400
1600
1700
2400
0100
1100
1300
1600
J700
0300
0400
0500
0600
0700
0800
2 1 00
2200
1600
1700
2400
0100
0200
1200
0900
Monitor
D03H
D531T
1)03 11
1)0311
1)0311
1)5311
D63H
D83H
1)6311
D83H
D23H
D63H
D83H
D63H
D83H
1)1311
D93H
D13H
D93H
D13H
D93H
1)1 3 H
Di3ll
1)9311
1)9311
1)1311
1)9311
DL3II
D93II
D23I1
D23H
D33H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D93H
D33H
i Observed
Concentration
(ppm)
0.36
0.10
0.65
0.49
0.44
0.35
0.26
0.23
0.28
0.20
0.30
0.27
0.14
0.30
0.10
0.37
0.20
0.26
0.21
0.33
0.18
0.50
0.59
0.30
0.34
0.21
0.48
0.33
0.21
0.32
0.44
0.19
0.47
0.15
0.45
0.18
0.47
0.16
0.40
0.16
0.26
Calculated
Concentration
(ppm)
0.00
0.02
0.00
0.00
0.02
0.18
0.00
0.07
0.74
0.09
0.15
0.00
0.28
0.03
0.02
0.54
0.02
0.42
0.01
0.02
0.00
0.08
0.09
0 . 00
0.02
0. 14
0.79
) . 35
0.06
0.22
0.14
0.46
0.30
0.08
0.08
0.01
0.04
0.01
1.04
0.00
0.43
Ratio of
Calculated
and Observed
Concentrations
0.00
0.20
0.00
0.00
0.05
0.51
0 . 00
0. 30
2.64
0.45
0.50
0.00
2.00
0.10
0.20
1.46
0.10
1.62
0.05
0.06
0.00
0.16
0.15
0 . 00
0.06
0.67
1.65
4.09
0.29
0.69
0.32
2.42
0.64
0.53
0.18
0.06
0.09
0.06
2.60
0.00
L.65
-------
TABLE 4-3 (Continued)
Case
No.
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
Year/
Julian
Day
76/046
76/046
76/046
76/049
76/049
76/049
76/052
76/052
76/052
76/065
76/065
76/065
76/066
76/067
76/067
76/067
76/067
76/073
76/074
76/074
76/074
76/074
76/084
76/084
76/084
Hour
1700
1800
1900
1700
1800
1900
1300
1500
1600
1000
1100
1500
2400
0200
1400
1500
1700
0200
1300
1900
2000
2400
0200
0300
0400
Monitor
D23H
D63H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D03H
D03H
D03H
D63H
D83H
D83H
D63H
D83H
D13H
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D93H
D23H
D23H
D23H
Observed
Concentration
(ppm)
0.22
0.29
0.81
0.23
0.42
0.18
0.40
0.15
0.29
0.15
0.28
0.16
0.39
0.28
0.29
0.11
0.34
0.29
0.12
0.26
0.12
0.31
0.27
0.15
0.34
0.11
0.32
0.29
0.26
0.12
0.25
0.10
0.29
0.17
0.28
0.11
0.16
0.25
0.30
0.35
0.30
Calculated
Concentration
(ppm)
0.04
0.69
0.28
0.07
0.01
0.01
0.41
0.52
0.07
0.01
0.01
0.00
0.00
0.02
0.01
0.17
0.36
0.49
0.11
0.21
0.00
0.35
0.41
0.02
0.36
0.23
0.57
0.00
0.67
0.98
0.01
0.02
0.01
0.02
0.05
0.09
1.72
0.01
1.09
0.68
0.48
Ratio of
Calculated
and Observed
Concentrations
0.18
2.38
0.35
0.30
0.02
0.06
1.03
3.47
0.24
0.07
0.04
0.00
0.00
0.07
0.03
1.55
1.06
1.69
0.92
0.81
0.00
1.13
1.52
0.13
1.06
2.09
1.78
0.00
2.58
8.17
0.04
0.20
0.03
0.12
0.18
0.82
10.75
0.04
3.63
1.94
1.60
4-16
-------
TABLE 4-3 (Continued)
Case
No.
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
Year/
Julian
Day
76/094
76/094
76/094
76/095
76/096
76/096
76/106
76/107
76/109
76/113
76/113
76/113
76/122
76/124
76/126
76/164
76/189
Hour
1800
2000
2300
0200
1600
1700
0200
1200
1000
0600
0700
0800
2200
1200
0200
2400
1800
Monitor
D03H
D03H
D03H
D03H
D13H
D93H
D13H
D93H
D23H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D13H
D23H
D63H
D13H
D93H
Observed
Concentration
(ppm)
0.26
0.27
0.59
0.27
0.30
0.16
0.60
0.16
0.64
0.10
0.26
0.35
0.15
0.36
0.21
0.31
0.25
0.31
0.25
0.27
0.29
0.28
0.30
0.10
Calculated
Concentration
(ppm)
0.39
0.04
0.07
0.16
0.00
0.61
0.66
0.10
0.60
0.13
1.18
0.45
0.04
0.90
0.12
0.35
0.13
0.54
0.26
0.25
0.18
0.00
0.00
0.17
Mean Ratio (MR)
Ratio of
Calculated
and Observed
Concentrations
1.50
0.15
0.12
0.59
0.00
3.81
1.10
0.63
0.94
1.30
4.54
1.29
0.27
2.50
0.57
1.13
0.52
1.74
1.04
0.93
0.62
0.00
0.00
1.70
0.83
4-17
-------
TABLE 4-4
COMPARISON OF CALCULATED AND OBSERVED 1-HOUR S02 CONCENTRATIONS
FOR THE DOW CASES FOR COMBINATION 5 IN TABLE 3-35 AND THE
CRAMER ET AL. (1975) STACK-TIP DOWNWASH CORRECTION
FOR STACKS WITH FROUDE NUMBERS ABOVE 3.0
Case
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Year/
Julian
Day
75/314
75/314
75/314
75/314
75/314
75/314
75/316
75/324
75/324
75/325
75/326
75/326
75/335
75/350
75/350
75/350
75/350
75/351
75/351
75/351
75/352
75/352
75/354
Hour
0100
0200
0500
1300
2300
2400
1700
1200
1300
0800
1500
1600
1400
1500
1900
2000
2100
0400
0700
1000
1100
1300
1200
Monitor
D33H
D63H
D73H
D33H
D73H
D63H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D63H
D23H
D63H
D63H
D83H
D63H
D83H
D23H
D33H
D23H
D23H
D23H
D33H
D13H
D93H
D13H
D93I1
D13H
D13H
D83H
D13H
D53H
Observed
Concentration
(ppm)
0.36
0.11
0.12
0.26
0.13
0.49
0.14
0.27
0.28
0.19
0.30
0.15
0.12
0.33
0.39
0.55
0.27
0.27
0.30
0.13
0.30
0.12
0.38
0.12
0.30
0.27
0.29
0.10
0.28
0.23
0.14
0.25
0.35
0.25
0.11
0.31
0.25
Calculated
Concentration
(ppm)
0.21
0.12
0.16
0.16
0.11
0.20
0.11
0.30
0.03
0.04
0.04
0.04
0.12
0.31
0.25
0.39
0.28
0.00
0.00
0.18
0.01
0.06
0.02
0.03
0.04
0.36
0.40
0.10
0.85
0.01
1.02
0.12
0.48
0.21
0.00
0.15
0.03
Ratio of
Calculated
and Observed
Concentrations
0.58
1.09
1.33
0.62
0.85
0.41
0.79
1.11
0.11
0.21
0.13
0.27
1.00
0.94
0.64
0.71
1.04
0.00
0.00
1.38
0.03
0.50
0.05
0.25
0.13
1.33
1.38
1.00
3.04
0.04
7.29
0.48
1.37
0.84
0.00
0.48
0.12
4-18
-------
TABLE 4-4 (Continued)
Case
No.
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
Year/
Julian
Day
75/354
75/354
75/354
75/354
75/354
76/003
76/003
76/005
76/008
76/008
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/028
76/028
76/038
76/039
76/039
76/042
76/043
Hour
1300
1400
1600
1700
2400
0100
1100
1300
1600
1700
3000
0400
0500
0600
0700
0800
2100
2200
1600
1700
2400
0100
0200
1200
0900
Monitor
D03H
D53H
D03H
D03H
D03H
D53H
D63H
D83H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D93H
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D93H
D93H
D131I
D93H
D13H
D93H
D23H
D23H
D33H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D93H
D33H
Observed
Concentration
(ppm)
0.36
0.10
0.65
0.49
0.44
0.35
0.26
0.23
0.28
0.20
0.30
0.27
0.14
0.30
0.10
0.37
0.20
0.26
0.21
0.33
0.18
0.50
0.59
0.30
0.34
0.21
0.48
0.33
0.21
0.32
0.44
0.19
0.47
0.15
0.45
0.18
0.47
0.16
0.40
0.16
0.26
Calculated
Concentration
(ppm)
0.00
0.07
0.04
0.03
0.07
0.24
0.03
0.07
0.39
0.26
0.25
0.01
0.12
0.12
0.06
0.61
0.08
0.60
0.05
0.16
0.00
0.35
0.36
0.00
0.08
0.33
0.48
1.16
0.12
0.20
0.12
0.22
0.24
0.14
0.13
0.07
0.11
0.06
0.63
0.02
0.19
Ratio of
Calculated
and Observed
Concentrations
0.00
0.70
0.06
0.06
0.16
0.69
0.12
0.30
1.39
1.30
0.83
0.04
0.86
0.40
0.60
1.65
0.40
2.31
0.24
0.48
0.00
0.70
0.61
0.00
0.24
1.57
1.00
3.52
0.57
0.63
0.27
1.16
0.51
0.93
0.29
0.39
0.23
0.38
1.58
0.13
0.73
4-19
-------
TABLE 4-4 (Continued)
Case
No.
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
Year/
Julian
Day
76/046
76/046
76/046
76/049
76/049
76/049
76/052
76/052
76/052
76/065
76/065
76/065
76/066
76/067
76/067
76/067
76/067
76/073
76/074
76/074
76/074
76/074
76/084
76/084
76/084
Hour
1700
1800
1900
1700
1800
1900
1300
1500
1600
1000
1100
1500
2400
0200
1400
1500
1700
0200
1300
1900
2000
2400
0200
0300
0400
Monitor
D23H
D63H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D03H
D03H
D03H
D63H
D83H
D83H
D63H
D83H
D13H
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D93H
D23H
D23H
D23H
Observed
Concentration
(ppm)
0.22
0.29
0.81
0.23
0.42
0.18
0.40
0.15
0.29
0.15
0.28
0.16
0.39
0.28
0.29
0.11
0.34
0.29
0.12
0.26
0.12
0.31
0.27
0.15
0.34
0.11
0.32
0.29
0.26
0.12
0.25
0.10
0.29
0.17
0.28
0.11
0.16
0.25
0.30
0.35
0.30
Calculated
Concentration
(ppm)
0.05
0.42
0.22
0.11
0.04
0.04
0.29
0.29
0.12
0.05
0.05
0.03
0.02
0.04
0.02
0.11
0.34
0.29
0.07
0.20
0.01
0.29
0.39
0.04
0.29
0.13
0.38
0.04
0.34
0.54
0.06
0.07
0.05
0.07
0.12
0.21
1.04
0.03
0.47
0.28
0.24
Ratio of
Calculated
and Observed
Concentrations
0.23
1.45
0.27
0.48
0.10
0.22
0.73
1.93
0.41
0.33
0.18
0.19
0.05
0.14
0.07
1.00
1.00
1.00
0.58
0.77
0.08
0.94
1.44
0.27
0.85
1.18
1.19
0.14
1.31
4.50
0.24
0.70
0.17
0.41
0.43
1.91
6.50
0.12
1.57
0.80
0.80
4-20
-------
TABLE 4-4 (Continued)
Case
No.
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
Year/
Julian
Day
76/094
76/094
76/094
76/095
76/096
76/096
76/106
76/107
76/109
76/113
76/113
76/113
76/122
76/124
76/126
76/164
76/189
Hour
1800
2000
2300
0200
1600
1700
0200
1200
1000
0600
0700
0800
2200
1200
0200
2400
1800
Monitor
D03H
D03H
D03H
D03H
D13H
D93H
D13H
D93H
D23H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D13H
D23H
D63H
D13H
D93H
Observed
Concentration
(ppm)
0.26
0.27
0.59
0.27
0.30
0.16
0.60
0.16
0.64
0.10
0.26
0.35
0.15
0.36
0.21
0.31
0.25
0.31
0.25
0.27
0.29
0.28
0.30
0.10
Calculated
Concentration
(ppm)
0.27
0.10
0.14
0.16
0.05
0.43
0.85
0.17
0.25
0.16
0.65
0.25
0.10
0.62
0.17
0.43
0.19
0.54
0.58
0.40
0.10
0.07
0.01
0.18
Mean Ratic (MR)
Ratio of
Calculated
and Observed
Concentrations
1.04
0.37
0.24
0.59
0.17
2.69
1.42
1.06
0.39
1.60
2.50
0.71
0.67
1.72
0.81
1.39
0.76
1.74
2.32
1.48
0.34
0.25
0.03
1.80
0.76
4-21
-------
TABLE 4-5
COMPARISON OF CALCULATED AND OBSERVED 1-HOUR SO CONCENTRATIONS
FOR THE DOW CASES FOR COMBINATION 2 IN TABLE 3-35 AND THE
SCIRE AND SCHULMAN (1980) "DOWNWASH RADIUS" CORRECTION
Case
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Year/
Julian
Day
75/314
75/314
75/314
75/314
75/314
75/314
75/316
75/324
75/324
75/325
75/326
75/326
75/335
75/350
75/350
75/350
75/350
75/351
75/351
75/351
75/352
75/352
75/354
Hour
0100
0200
0500
1300
2300
2400
1700
1200
1300
0800
1500
1600
0400
1500
1900
2000
2100
0400
0700
1000
1100
1300
1200
Monitor
D33H
D63H
D73H
D33H
D73H
D63H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D63H
D23H
D63H
D63H
D83H
D63H
D83H
D23H
D33H
D23H
D23H
D23H
D33H
D13H
D93H
D13H
D93H
D13H
D13H
D83H
D13H
D53H
Observed
Concentration
(ppm)
0.36
0.11
0.12
0.26
0.13
0.49
0.14
0.27
0.28
0.19
0.30
0.15
0.12
0.33
0.39
0.55
0.27
0.27
0.30
0.13
0.30
0.12
0.38
0.12
0.30
0.27
0.29
0.10
0.28
0.23
0.14
0.25
0.35
0.25
0.11
0.31
0.25
Calculated
Concentration
(ppm)
0.55
0.00
0.37
0.36
0.17
0.16
0.19
0.53
0.01
0.01
0.01
0.01
0.15
0.10
0.03
0.12
0.35
0.00
0.00
0.46
0.00
0.09
0.00
0.00
0.00
0.49
0.65
0.23
0.73
0.00
0.79
0.06
0.63
0.41
0.00
0.15
0.01
Ratio of
Calculated
and Observed
Concentrations
1.53
0.00
3.08
1.38
1.31
0.33
1.36
1.96
0.04
0.05
0.03
0.07
1.25
0.30
0.08
0.22
1.30
0.00
0.00
3.54
0.00
0.75
0.00
0.00
0.00
1.81
2.24
2.30
2.61
0.00
5.64
0.24
1.80
1.64
0.00
0.48
0.04
4-22
-------
TABLE 4-5 (Continued)
Case
No.
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
Year/
Julian
Day
75/354
75/354
75/354
75/354
75/354
76/003
76/003
76/005
76/008
76/008
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/028
76/028
76/038
76/039
76/039
76/042
76/043
Hour
1300
1400
1600
1700
2400
0100
1100
1300
1600
1700
0300
0400
0500
0600
0700
0800
2100
2200
1600
1700
2400
0100
0200
1200
0900
Monitor
D03H
D53H
D03H
D03H
D03H
D53H
D63H
D83H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D13H
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D93H
D93H
D13H
D93H
D13H
D93H
D23H
D23H
D33H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D93H
D33H
Observed
Concentration
(ppm)
0.36
0.10
0.65
0.49
0.44
0.35
0.26
0.23
0.28
0.20
0.30
0.27
0.14
0.30
0.10
0.37
0.20
0.26
0.21
0.33
0.18
0.50
0.59
0.30
0.34
0.21
0.48
0.33
0.21
0.32
0.44
0.19
0.47
0.15
0.45
0.18
0.47
0.16
0.40
0.16
0.26
Calculated
Concentration
(ppm)
0.00
0.04
0.00
0.00
0.04
0.45
0.00
0.06
0.46
0.46
0.44
0.00
0.23
0.02
0.01
1.32
0.03
1.27
0.01
0.09
0.00
0.47
0.49
0.00
0.03
0.16
0.81
1.18
0.06
0.47
0.30
0.41
0.21
0.04
0.05
0.01
0.03
0.01
0.70
0.00
0.35
Ratio of
Calculated
and Observed
Concentrations
0.00
0.40
0.00
0.00
0.09
1.29
0.00
0.26
1.64
2.30
1.47
0.00
1.64
0.07
0.10
3.57
0.15
4.88
0.05
0.27
0.00
0.94
0.83
0.00
0.09
0.76
1.69
3.58
0.29
1.47
0.68
2.16
0.45
0.27
0.11
0.06
0.06
0.06
1.75
0.00
1.35
4-23
-------
TABLE 4-5 (Continued)
Case
No.
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
Year/
Julian
Day
76/046
76/046
76/046
76/049
76/049
76/049
76/052
76/052
76/052
76/065
76/065
76/065
76/066
76/067
76/067
76/067
76/067
76/073
76/074
76/074
76/074
76/074
76/084
76/084
76/084
Hour
1700
1800
1900
1700
1800
1900
1300
1500
1600
1000
1100
1500
2400
0200
1400
1500
1700
0200
1300
1900
2000
2400
0200
0300
0400 1
Monitor
D23H
D63H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D03H
D03H
D03H
D63H
D83H
D83H
D63H
D83H
D13H
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D93H
D23H
D23H
D23H
Observed
Concentration
(ppm)
0.22
0.29
0.81
0.23
0.42
0.18
0.40
0.15
0.29
0.15
0.28
0.16
0.39
0.28
0.29
0.11
0.34
0.29
0.12
0.26
0.12
0.31
0.27
0.15
0.34
0.11
0.32
0.29
0.26
0.12
0.25
0.10
0.29
0.17
0.28
0.11
0.16
0.25
0.30
0.35
0.30
Calculated
Concentration
(ppm)
0.03
0.61
0.23
0.05
0.01
0.01
0.33
0.36
0.13
0.02
0.02
0.01
0.00
0.06
0.02
0.23
0.80
0.44
0.13
0.18
0.00
0.30
0.74
0.03
0.37
0.23
0.63
0.00
0.48
0.57
0.01
0.02
0.01
0.02
0.14
0.04
0.86
0.01
0.71
0.45
0.32
Ratio of
Calculated
and Observed
Concentrations
0.14
2.10
0.28
0.22
0.02
0.06
0.83
2.40
0.45
0.13
0.07
0.06
0.00
0.21
0.07
2.09
2.35
1.52
1.08
0.69
0.00
0.97
2.74
0.20
1.09
2.09
1.97
0.00
1.85
4.75
0.04
0.20
0.03
0.12
0.50
0.36
5.38
0.04
2.37
1.29
1.07
4-24
-------
TABLE 4-5 (Continued)
Case
No.
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
Year/
Julian
Day
76/094
76/094
76/094
76/095
76/096
76/096
76/106
76/107
76/109
76/113
76/113
76/113
76/122
76/124
76/126
76/164
76/189
Hour
1800
2000
2300
0200
1600
1700
0200
1200
1000
0600
0700
0800
2200
1200
0200
2400
1800
Monitor
D03H
D03H
D03H
D03H
D13H
D93H
D13H
D93H
D23H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D13H
D23H
D63H
D13H
D93H
Observed
Concentration
(ppm)
0.26
0.27
0.59
0.27
0.30
0.16
0.60
0.16
0.64
0.10
0.26
0.35
0.15
0.36
0.21
0.31
0.25
0.31
0.25
0.27
0.29
0.28
0.30
0.10
Calculated
Concentration
(ppm)
0.41
0.04
0.08
0.13
0.00
0.69
0.74
0.13
0.46
0.34
0.85
0.31
0.11
0.73
0.32
0.30
0.39
0.42
0.48
0.73
0.23
0.02
0.00
0.14
Mean Ratio (MR)
Ratio of
Calculated
and Observed
Concentrations
1.58
0.15
0.14
0.48
0.00
4.31
1.23
0.81
0.72
3.40
3.27
0.89
0.73
2.03
1.52
0.97
1.56
1.35
1.92
2.70
0.79
0,07
0.00
1.40
0.92
4-25
-------
TABLE 4-6
COMPARISON OF CALCULATED AND OBSERVED 1-HOUR SO CONCENTRATIONS FOR THE
DOW CASES FOR COMBINATION 5 IN TABLE 3-35 AND THE SCIRE AND
SCHULMAN (1980) "DOWNWASH RADIUS" CORRECTION
Case
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Year/
Julian
Day
1
75/314
75/314
75/314
75/314
75/314
75/314
75/316
75/324
75/324
75/325
75/326
75/326
75/335
75/350
75/350
75/350
75/350
75/351
75/351
75/351
75/352
75/352
75/354
Hour
0100
0200
0500
1300
2300
2400
1700
1200
1300
0800
1500
1600
0400
1500
1900
2000
2100
0400
0700
1000
1100
1300
1200
Monitor
D33H
D63H
D73H
D33H
D73H
D63H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D63H
D23H
D63H
D63H
D83H
D63H
D83H
D23H
D33H
D23H
D23H
D23H
D33H
D13H
D93H
D13H
D93H
D13H
D13H
D83H
D13H
D53H
Observed
Concentration
(ppm)
0.36
0.11
0.12
0.26
0.13
0.49
0.14
0.27
0.28
0.19
0.30
0.15
0.12
0.33
0.39
0.55
0.27
0.27
0.30
0.13
0.30
0.12
0.38
0.12
0.30
0.27
0.29
0.10
0.28
0.23
0.14
0.25
0.35
0.25
0.11
0.31
0.25
Calculated
Concentration
(ppm)
0.24
0.20
0.21
0.18
0.14
0.31
0.11
0.29
0.04
0.05
0.05
0.06
0.17
0.29
0.17
0.28
0.26
0.00
0.00
0.20
0.01
0.06
0.02
0.03
0.04
0.35
0.40
0.11
0.93
0.01
0.85
0.11
0.91
0.58
0.00
0.39
0.04
Ratio of
Calculated
and Observed
Concentrations
0.67
1.82
1.75
0.69
1.08
0.63
0.79
1.07
0.14
0.26
0.17
0.40
1.42
0.88
0.44
0.51
0.96
0.00
0.00
1.54
0.03
0.50
0.05
0.25
0.13
1.30
1.37
1.10
3.32
0.04
6.07
0.44
2.60
2.32
0.00
1.26
0.16
4-26
-------
TABLE 4-6 (Continued)
Case
No.
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
Year/
Julian
Day
75/354
i
75/354
75/354
75/354
75/354
76/003
76/003
76/005
76/008
76/008
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/014
76/028
76/028
76/038
76/039
76/039
76/042
76/043
Hour
1300
1400
1600
1700
2400
0100
1100
1300
1600
1700
0300
0400
0500
0600
0700
0800
2100
2200
1600
1700
2400
0100
0200
1200
0900
Monitor
D03H
D53H
D03H
D03H
D03H
D53H
D63H
D83H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D13H
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D93H
D93H
D13H
D93H
D13H
D93H
D23H
D23H
D33H
D63H
D83H
D63H
D83H
D63H
D83H
Observed
Concentration
(ppm)
0.36
0.10
0.65
0.49
0.44
0.35
0.26
0.23
0.28
0.20
0.30
0.27
0.14
0.30
0.10
0.37
0.20
0.26
0.21
0.33
0.18
0.50
0.59
0.30
0.34
0.21
0.48
0.33
0.21
0.32
0.44
0.19
0.47
0.15
0.45
0.18
0.47
0.16
D13H ! 0.40
D93H 1 0.16
D33H
0.26
Calculated
Concentration
(ppm)
0.00
0.08
0.04
0.03
Ratio of
Calculated
and Observed
Concentrations
i
| 0.00
0.80
0.06
0.06
0.09 i 0.20 j
0.31
0.03
0.08
0.34
0.45
0.33
0.01
0.12
0.11
0.07
1.16
0.09
1.19
0.06
0.34
0.00
0.79
0.78
0.00
0.09
0.34
0.48
1.08
0.12
0.24
0.15
0.22
0.22
0.11
0.12
0.06
0.10
0.05
0.53
0.02
0.18
0.89
0.12
0.35
1.21
2.25
1.10
0.04
0.86
0.37
0.70
3.14
0.45
4.58
0.29
1.03
0.00
1.58
1.32
0.00
0.26
1.62
1.00
3.27
0.57
0.75
0.34
1.16
0.47
0.73
0.27
0.33
0.21
0.31
1.33
0.13
0.69
4-27
-------
TABLE 4-6 (Continued)
Case
No.
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
Year/
Julian
Day
76/046
76/046
76/046
76/049
76/049
76/049
76/052
76/052
76/052
76/065
76/065
76/065
76/066
76/067
76/067
76/067
76/067
76/073
76/074
76/074
76/074
76/074
76/084
76/084
76/084
Hour
1700
1800
1900
1700
1800
1900
1300
1500
1600
1000
1100
1500
2400
0200
1400
1500
1700
0200
1300
1900
2000
2400
0200
0300
0400
Monitor
D23H
D63H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D03H
D03H
D03H
D63H
D83H
D83H
D63H
D83H
D13H
D93H
D13H
D93H
D13H
D93H
D13H
D13H
D63H
D83H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D93H
D23H
D23H
D23H
Observed
Concentration
(ppm)
0.22
0.29
0.81
0.23
0.42
0.18
0.40
0.15
0.29
0.15
0.28
0.16
0.39
0.28
0.29
0.11
0.34
0.29
0. 12
0.26
0.12
0.31
0.27
0.15
0.34
0.11
0.32
0.29
0.26
0.12
0.25
0.10
0.29
0.17
0.28
0.11
0.16
0.25
0.30
0.35
0.30
Calculated
Concentration
(ppm)
0.05
0.41
0.21
0.10
0.05
0.04
0.27
0.26
0.15
0.07
0.06
0.04
0.02
0.05
0.03
0.12
0.45
0.28
0.08
0.19
0.01
0.29
0.52
0.05
0.34
0.14
0.41
0.04
0.30
0.47
0.05
0.06
0.05
0.06
0.17
0.16
0.75
0.03
0.43
0.25
0.22
Ratio of
Calculated
and Observed
Concentrations
0.23
1.41
0.26
0.43
0.12
0.22
0.68
1.73
0.52
0.47
0.21
0.25
0.05
0.18
0.10
1.09
1.32
0.97
0.67
0.73
0.08
0.94
1.93
0.33
1.00
1.27
1.28
0.14
1.15
3.92
0.20
0.60
0.17
0.35
0.61
1.45
4.69
0.12
1.43
0.71
0.73
4-28
-------
TABLE 4-6 (Continued)
Case
No.
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
Year/
Julian
Day
76/094
76/094
76/094
76/095
76/096
76/096
76/106
76/107
76/109
76/113
76/113
76/113
76/122
76/124
76/126
76/164
76/189
Hour
1800
2000
2300
0200
1600
1700
0200
1200
1000
0600
0700
0800
2200
1200
0200
2400
1800
Monitor
D03H
D03H
D03H
D03H
D13H
D93H
D13H
D93H
D23H
D63H
D83H
D23H
D63H
D83H
D63H
D83H
D63H
D83H
D13H
D13H
D23H
D63H
D13H
D93H
Observed
Concentration
(ppm)
0.26
0.27
0.59
0.27
0.30
0.16
0.60
0.16
0.64
0.10
0.26
0.35
0.15
0.36
0.21
0.31
0.25
0.31
0.25
0.27
0.29
0.28
0.30
0.10
Calculated
Concentration
(ppm)
0.27
0.10
0.14
0.15
0.05
0.44
0.85
0.18
0.24
0.21
0.60
0.23
0.14
0.60
0.22
0.41
0.26
0.51
0.71
0.62
0.11
0.12
0.00
0.18
Mean Ratio (MR)
Ratio of
Calculated
and Observed
Concentrations
1.04
0.37
0.24
0.56
0.17
2.75
1.42
1.13
0.38
2.10
2.31
0.66
0.93
1.67
1.05
1.32
1.04
1.65
2.84
2.30
0.38
0.43
0.00
1.80
0.86
4-29
-------
W
cu
o
G
CO
0
O
*4-4
M
CU
PH
rH
0)
t3
O
S
14H
O
CO
CU
1-1
3
CO
CO
s
C M
•H O
XI 4J CN
4J O
•rt CO VW
£2 fH O
^S CO
w
CO
- - - - - —
pi
S
C
cO
0
3 CO
XI CO
o u
CO G CO
3 3
TJ O -H
C P T3
tO CO
O OS
CU 00
1-1 ON
•H rH
a
to
H
CO
XI
CO 4-1 X-N
to a)|m
5 r—
G -ON
o cu ^
P 0
(0
a. M
•H O
H _
^4 CO f-~ >
O 60 co
tO M> r--
4J vH ON
CO VJ ^H
pq % ^
60 CO
G 4->
•H cu a
Tj ^ CU
rH CO >4-l
•H 3 MH
3 W
(0
, — |
cu
'O
0 CU
!C T)
e-> si
tn
M
*
*
^-^
x^-
O cy* in oo
__ _
1 CN | l-l
a) cu| o
cu » C
O t-i tO
G CU XI
tfl 0 4J
g to
O e_> cu
l»H 4-J
H cu to
cu x: cu
ft 4-> >-J
60
O O co
^
co cu cu
0) CO XI
3 e
M
en Q)
to x:
CU 4-1
3
CU
TJ
G 3
cu cu o
S
•X
•K
4-30
-------
(iudd)
OS
FIGURE 4-1.
Comparison of the observed cumulative 1-hour S09 concentration
frequency distribution for the DOW data set with the corresponding
distributions calculated by the ISCST program in the Rural Mode and
Urban Mode 2 using the Cramer, et^ al. (1975) stack-tip downwash cor-
rection for all stacks.
4-31
-------
NOIlVdlN3DNOO ZOS
FIGURE 4-2. Comparison of the observed cumulative 1-hour SO concentration
frequency distribution for the DOW data set with the corresponding
distributions calculated by the ISCST program in the Rural Mode and
Urban Mode 2 using the Cramer, e* al. (1975) stack-tip downwash
correction for stacks with Froude numbers above 3.0.
4-32
-------
sracks with Froudo numbers (Fr) greater than or equal lo 3.0, respect-
ively. Similarly, Figure 4-3 compares the observed concentration fre-
quency distribution with the distributions calculated by the ISC Model
using the Scire and Schulman "downwash radius" correction. Inspection
of Figures 4-1 through 4-3 shows that the Cramer, _et_ ad. stack-tip downwash
correction in Urban Mode 2 for stacks with Froude numbers above 3.0
and the Scire and Schulman "downwash radius" correction in both the
Rural Mode and Urban Mode 2 closely match the upper nercentiles of
the observed concentration distribution. However, as shown by Figure
3-9 in Section 3.4.3, the unmodified ISC Model also closely matches the
upper perccntiles of the observed concentration distribution for model
option Combination 6. Thus, we cannot determine on the basis of the
available information whether or not the Cramer, et_ al_. stack-tip
downwash correction or the Scire and Schulman "downwash radius" cor-
rection definitely improves the performance of the ISC model for the
DOW data set.
We point out that comparisons of calculated and observed
1-hour concentration cumulative frequency distributions such as shown in
Figures 4-1 through 4-3 generally should be restricted to hours with the
same meteorological conditions and approximately the same emissions par-
ameters. The DOW data set satisfies the first criterion because the
meteorological conditions during all of the hours selected for model
testing were neutral stability in combination with moderate or strong
winds. However, the DOW data set does not satisfy the second criterion
because of the highly variable SO^ emission rates and exit velocites for
the various stacks at the two DOW power houses (see Tables 3-31 and 3-32
in Section 3.4.1). Additionally, because of limitations In sample size
for a given monitor, Figure 3-9 in Section 3.4.3 and Figures 4-1 through
4-3 are based on the calculated and observed concentrations for all mon-
itors .
4-33
-------
CM
(ujdd) NOIlVdiN30NOO 2OS
KIGURE 4-3. Comparison of the observed 1-hour SCL concentration frequency
distribution for the DOW data set with the corresponding distri-
butions calculated by the ISCST program in the Rural Mode and Urban
Mode 2 using the Scire and Schulman (1980) "downwash radius" cor-
rection.
4-34
-------
SECTION 5
CONCLUSIONS
The key features that distinguish the ISC Model from other
generally available dispersion models are the gravitational settling/
dry deposition option and the building wake effects option. We conclude
from the tests of these key features described in Sections 2 and 3 that:
1. The gravitational settling/dry deposition option adds
capabilities lacking in most current models, and the
accuracy of this option for particulates with appreciable
gravitational settling velocities appears to correspond
to the approximate factor of 2 accuracy generally attrib-
uted to the results of short-term dispersion model
calculations in the absence of complicating factors (AMS,
1978) .
2. For plumes subject to building wake effects, the building
wake effects option significantly improves the perform-
ance of the ISC Model over that of the corresponding
models (CRSTER and MPTER) which do not consider building
wake effects when used to calculate concentrations near
the source.
However, the results of the model calculations for the Millstone data
set (Johnson, et^ al_., 1975) indicate that it may be desirable to modify
the model to consider a reduction in plume height due to initial entrainment
of insignificantly buoyant emissions into the cavity zone for stacks located
on or adjacent to squat buildings with stack height to building height
ratios less than 1.2.
If the mean ratio (MR) of calculated to observed concentrations,
the root mean square error (RMSH) and the percentage error bands are
5-1
-------
used as measures of model performance for the DOW data set, the ISC
Model appears to have a systematic tendency to underestimate the con-
centrations produced near the source by buoyant stack emissions subject
to building wake effects. As discussed in Section 4, we added the
Cramer, et al. (1975) stack-tip downwash correction and the Scire and
Schulman (1980) "downwash radius" correction to special versions of the
ISCST program and repeated the concentration calculations for the DOW
data set. Both corrections improved the model's performance, as indi-
cated by the MR, RMSE and percentage error bands. However, the MR, RMSE
and percentage error bands may not be the most appropriate measures of
model performance for the DOW data set because uncertainties about the locations
of the plume centerlines arising from the uncertainties about the mean
transport wind directions cannot be removed from the DOW data set as was
done in the diffusion experiments where the locations of the plume
centerlines could be determined from the dense monitoring networks. It
is not possible, from a comparison of the calculated and observed concen-
tration frequency distributions, to determine whether or not either of
the two corrections significantly improves the model's performance for
the DOW data set.
5-2
-------
REFERENCES
Abrams, I. B., 1978: Private communication (22 December 1978 letter to
J. F. Bowers, H. E. Cramer Company, Inc.).
Abrams, I. B., 1979: Private communication (26 February 1979 letter to
J. F. Bowers, H. E. Cramer Company, Inc.).
American Meteorological Society, 1978: Accuracy of dispersion models: A
position paper of the 1977 AMS Committee on Atmospheric Turbu-
lence and Diffusion. Bulletin American Meteorological Society,
5i, 1025-1026.
Atomic Energy Commission, 1972: On-site meteorological programs. U. S.
AEG Regulatory Guide 1.23.
Bjorklund, J. R. and J. F. Bowers, 1979: User's instructions for the
SHORTZ and LONGZ computer programs. Technical Report TR-79-
131-01, H. E. Cramer Company, Inc., Salt Lake City, Utah.
Briggs, G. A., 1969: Plume Rise. Available as TID-25075 from Clearing-
house for Federal Scientific and Technical Information, Spring-
field Va., 80.
Briggs, G. A., 1971: Some recent analyses of plume rise observations. In
Proceedings of the Second International Clean Air Congress, Aca-
demic Press, New York.
Briggs, G. A., 1973: Diffusion estimates for small emissions. ATDL Contri-
bution File No. (Draft) 79, Air Resources Atmospheric Turbulence
and Diffusion Laboratories, Oak Ridge, Tennessee.
Briggs, G. A., 1975: Plume rise predictions. Lectures on Air Pollution
and Environmental Impact Analyses. American Meteorological Soci-
ety, Boston, Massachusetts.
Bringfelt, B., 1968: Plume rise measurements at industrial chimneys.
Atmospheric Environment, _3(6), 575-598.
Brown, J. M., 1979a: Private communication (Telephone conversation on
17 February 1979 with J. F. Bowers, H. E. Cramer Company, Inc.).
Brown, J. M., 1979b: Private communication (Telephone conversation on
7 March 1979 with J. F. Bowers, H. E. Cramer Company, Inc.) .
Boyle, D. G. , ct a].., 1975: DC-7B aircraft spray system for large-area
insect control. PPG Document No. DPG-DR-C980A, U. S. Army
Dugway Proving Ground, Dugway, Utah 84022.
6-1
-------
REFERENCES (Continued)
Busse, A. D. and J. R. Zimmerman, 1973: User's guide for the Climatologi-
cal Dispersion Model. EPA Report No. EPA-RA-73-024, U. S. Envi-
ronmental Protection Agency, Research Triangle Park, North
Carolina.
Cramer, H. E., H. V. Geary and J. F. Bowers, 1975: Diffusion-model calcu-
lations of long-term and short-term ground-level S0~ concentra-
tions in Allegheny County, Pennsylvania. H. E. Cramer Company
Technical Report TR-75-102-01, prepared for the U. S. Environ-
mental Protection Agency, Region III, Philadelphia, Pennsylvania.
EPA Report 903/9-75-018. NTIS Accession No. PB-245262/AS.
Draxler, R. R., 1980: An improved Gaussian model for long-term average
air concentration estimates. Atmospheric Environment, 14(5),
597-601.
Dumbauld, R. K., J. R. Bjorklund and J. F. Bowers, 1973: NASA/MSFC multi-
layer diffusion models and computer program for operational pre-
diction of toxic fuel hazards. NASA Contractor Report NASA
CR-129006. H. E. Cramer Company Technical Report prepared for
National Aeronautics and Space Administration, George C. Marshall
Space Flight Center, Alabama 35812.
Ellis, H. M. and P. C. Liu, 1980: Discussion: An air quality model perform-
ance assessment package. Atmospheric Environment, J_A_, 1113.
Environmental Protection Agency, 1969: Air Quality Display Model. Pre-
pared by TRW Systems Group, Washington, D. C., available as
PB 189-194 from the National Technical Information Service,
Springfield, Virginia.
Environmental Protection Agency, 1977: User's manual for Single Source
(CRSTER) Model. EPA Report No. EPA-450/2-77-013, U. S. Environ-
mental Protection Agency, Research Triangle Park, 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. Environ-
mental Protection Agency, Research Triangle Park, North Carolina.
Environmental Protection Agency, 1979: Industrial Source Complex (ISC)
DLspersion Modol user's guide. EPA Reports EPA-450/4-79-Q30
(Volume T) and EPA-450/4-79-031 (Volume II), U. S. Environmental
Protection Agency, Research Triangle Park, North Carolina.
Fay, J. A., M. Escudier and D. P. Hoult, 1970: A correlation of field
observations of plume rise. Journal of the Air Pollution Control
Association, 20(6), 391-397.
6-2
-------
REFERENCES (Continued)
Halitsky, J. H. and K. Woodard, 1974: Atmospheric diffusion experiments
at a nuclear power plant site under light wind inversion condi-
tions. Preprint Volume for the Symposium on Atmospheric Diffu-
sion and Air Pollution, American Meteorological Society, Boston,
Massachusetts.
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, Massachusetts.
Islitzer, N. F., 1965: Aerodynamic effects of large reactor complexes upon
atmospheric turbulence and diffusion. USAEC Report No . IDO-12041,
U. S. Weather Bureau, Idaho Falls, Idaho.
Johnson, W. B., &t^ al_. , 1975: Gas tracer study of roof-vent effluent dif-
fusion at Millstone Nuclear Power Station. Atomic Industrial
Forum, Inc. Report No. AIF/NESP-0076, prepared by Stanford
Research Institute, Menlo Park, California.
Lee, R. F. , M. T. Mills and R. W. Stern, 1975: Validation of a single
source dispersion model. Paper presented at the 6th NATO/CCMS
International Technical Meeting on Air Pollution Modeling, Frank-
furt/Main, Germany, 24-26 September 1975.
McDonald, J. E., 1960: An aid to computation of terminal fall velocities
of spheres. J. Met. , 17, 463.
Munn, R. E. and A. F. W. Cole, 1967: Some strong-wind downwash diffusion
measurements at Douglas Point, Ontario, Canada. Atmospheric
Environment , 1^, 601-604.
Pasquill, F. , 1961: The estimation of the dispersion of windborne material.
Meteorology Magazine, 90, 33-49.
Pierce, T. K. and D. B. Turner, 1980: User's guide, for MPTER. EPA Report
No . EPA- 6 00 / 8-8 0- 01 6 , U. S. Environmental Protection Agency,
Triangle Park, North Carolina.
Sclro, J. S. and L. L. Schulman, 1980: Modeling Plume rise from low-level
buoyant line and point sources. Preprint Volume for the Second
Joint Conference on Applications of Air Pollution Meteorology,
American Meteorological Society, Boston, Massachusetts.
6-3
-------
REFERENCES (Continued)
Start, G. E., et_ al_., 1977: Rancho Seco building wake effects on atmo-
spheric diffusion. NOAA Technical Memorandum ERL ARL-64, Air
Resources Laboratories, Idaho Palls, Idaho.
Stewart, R. E., 1968: Atmospheric diffusion of particulate matter released
from an elevated source. Journal of Applied Meteorology, J7, 425-
432.
Turner D. B., 1964: A'>diffusion model for an urban area. Journal of
Applied Meteorology, _3(1) , 83-91.
Turner, D. B., J. R. Zimmerman and A. D. Busse, 1972: An evaluation of
some climatological dispersion models. Paper presented at the
3rd NATO/CCMS International Technical Meeting on Air Pollution
Modeling, Paris, France, October 1972.
Turner, D. B., 1979: Atmospheric dispersion modeling: A critical review.
Journal of the Air Pollution Control Association, 29X5), 502-519.
Walker, E. R., 1965: A particulate diffusion experiment. Journal of Ap-
plied Meteorology, 4, 614-621.
6-4
-------
APPENDIX A
CALCULATED AND OBSERVED DEPOSITION PROFILES
FOR THE WALKER (1965) AND STEWART (1968)
EXPERIMENTS
This appendix contains the plots of calculated and observed depo-
sition profiles for the Walker (1965) and Stewart (1965) deposition experi-
ments that are discussed in Section 2. Figures A-l through A-18 show the
observed and calculated normalized crosswind integrated deposition (CWJD)
profiles for the Walker (1965) trials. Two calculated profiles are given
for each trial. The mean layer wind speed a. was used in the deposition
calculations for the first profile and the mean wind speed at the height
of emission u{h} was used in the deposition calculations for the second
profile. The surface reflection coefficients y used to calculate the
deposition profiles shown in Figure A-l through A-12 were taken from
Figure 2-8 of the Industrial Source Complex (ISC) Model User's Guide (EPA,
1979). For the calculated profiles in Figures A-13 through A-18, it was
assumed that there was no reflection at the ground surface (y - 0).
Figures A-19 through A-28 show the calculated non-dimensional crosswind
integrated deposition profiles for the Stewart (1968) trials and the
observed maximum non-dimensional crosswind integrated deposition values.
A-l
-------
o
\
Q
>—i
£
ISC(QL)
0.2
200 300
DISTANCE (meters)
T
FIGURE A-l. Observed and calculated normalized crosswind Integrated deposi-
tion versus distance for Walker Trial A.
A-2
-------
E
o>
o>
^
O
Q
O
1 TRIAL B
OBSERVED
ISC (0L)
11-1 ISC (u{h})
'•t
0.2
200 300
DISTANCE (meters)
FIGURE A-2. Observed and calculated normalized crosswind integrated deposi-
tion versus distance for Walker Trial B.
A-3
-------
30
20
E
o>
o>
E
O
x.
Q
K—I
O
0.2
TRIAL C
OBSERVED
ISC (QL)
ISC lulh)) —
200
DISTANCE (meters)
300
FIGURE A-3. Observed and calculated normalized crosswind integrated
deposition versus distance for Walker Trial C.
A-4
-------
30
20
10
8
E
o>
\
o>
e
o
^
0
»—«
£
o
I
0.8
0.6
0.4
0.2
-.
ZT" ™1V_.
-t-t
i I..-
rffl
— I
R AL D
OBSERVED-
ISC (0L) —
ISC (ii{h})
—
T}-
- -n-
=flBi
+- -t -t -;
f—/I
trl:
w
i:i
mr:
t4r
100
200 300
DISTANCE (meters)
FIGURE A-4. Observed and calculated normalized crosswind integrated
deposition versus distance for Walker Trial D.
A-5
-------
E
o>
o>
J
O
>x
O
0.2
200 300
DISTANCE (meters)
FIGURE A-5. Observed and calculated normalized crosswind integrated
deposition versus distance for Walker Trial E.
A-6
-------
E
o>
o
\
Q
i—«
O
TRIAL F
OBSERVED
ISC (0L) —
ISC (u{h})
100
200 300
DISTANCE (meters)
FIGURE A-6, Observed and calculated normalized crosswind integrated
deposition versus distance for Walker Trial F.
A-7
-------
0>
E
o
v^
o
t—1
o
0.2
200 300
DISTANCE (meters)
FIGURE A-7. Observed and calculated normalized crosswind integrated
deposition versus distance for Walker Trial G.
A-8
-------
E
o>
o
^
Q
*-*
O
i MSC (QL)
ISC {u{h})
200 300
DISTANCE (meters)
FIGURE A-8. Observed and calculated normalized crosswind integrated
deposition versus distance for Walker Trial H.
A-9
-------
E
en
O
v,
Q
i—i
O
TRIAL I
OBSERVED
200
DISTANCE (meters)
300
FIGURE A-9. Observed and calculated normalized crosswind integrated
deposition versus distance for Walker Trial I.
A-10
-------
E
o>
V.
en
E
i-
o
o
0.2
100
200 300
DISTANCE (meters)
FIGURE A-10.
Observed and calculated normalized crosswind integrated
deposition versus distance for Walker Trial J.
A-ll
-------
30
20
E
o>
o
V.
o
I—I
£
o
0.2
OBSERVED-
s^]
ISC (QL)
ISC (o{h})
200 300
DISTANCE (meters)
FIGURE A-ll.
Observed and calculated normalized crosswind integrated
deposition versus distance for Walker Trial K.
A-12
-------
E
CT>
\
O>
E
i-
o
o
TRIAL L
OBSERVED
ISC (QL)
200 300
DISTANCE (meters)
FIGURE A-12.
Observed and calculated normalized crosswind Integrated
deposition versus distance for Walker Trial L.
A-13
-------
1
o>
o>
o
Q
1—4
O
TRIAL 6 (y = 0)
OBSERVED
ISC (QL)
ISC (u{h})
200
DISTANCE (meters)
300
FIGURE A-13.
Observed and calculated normalized crosswind integrated
deposition versus distance for Walker Trial G with no
reflection at the surface.
A-14
-------
E
I
O
Q
>—•
O
200
DISTANCE (meters)
300
FIGURE A-14.
Observed and calculated normalized crosswind integrated
deposition versus distance for Walker Trial H with no
reflection at the surface.
A-15
-------
E
I
o>
o
\
a
•—«
o
TRIAL I (
OBSERVED
ISC (QL)
ISC (u{h})
200 300
DISTANCE (meters)
FIGURE A-15.
Observed and calculated normalized crosswind integrated
deposition versus distance for Walker Trial I with no
reflection at the surface.
A-16
-------
E
I
o>
o
^
o
1—4
o
TRIAL J ly = 0)
OBSERVED
ISC (QL)
200
DISTANCE (meters)
300
FIGURE A-16.
Observed and calculated normalized crosswind integrated
deposition versus distance for Walker Trial J with no
reflection at the surface.
A-17
-------
E
i
o>
J
I-
O
TRIAL K (y = 0)
OBSERVED
ISC (DL)
ISC (u{h})-
200
DISTANCE (meters)
300
FIGURE A-17.
Observed and calculated normalized crosswind integrated
deposition versus distance for Walker Trial K with no
reflection at the surface.
A-18
-------
E
I
o>
\
o>
O
\
Q
•—i
O
TRIAL L (y = 0)
OBSERVED
ISC (QL)
ISC (o{h})
200
DISTANCE (meters)
300
FIGURE A-18.
Observed and calculated normalized crosswind integrated
deposition versus distance for Walker Trial L with no
reflection at the surface.
A-19
-------
10°
h
o
'
r i:
4 i-
TRIAL BI/200
OBSERVED MAXIMUM
ISC MODEL
SUBJECTIVE EXTRAPOLATION-
ir_ ' I
100
200
DISTANCE (meters)
300
FIGURE A-19.
Observed maximum and calculated profile of non-dimensional
cross-wind integrated deposition for Stewart Trial Bl/200.
A- 20
-------
o
o
TRIAL B2/200
OBSERVED MAXIMUM
ISC MODEL
^SUBJECTIVE EXTRAPOLATION
100
200
DISTANCE (meters)
300
FIGURE A-2Q. Observed maximum and calculated profile of non-dimensional
cross-wind integrated deposition for Stewart Trial B2/200.
A-21
-------
o
OBSERVED MAXIMUM
200 300
DISTANCE (meters)
FIGURE A-21. Observed maximum and calculated profile of non-dimensional
cross-wind integrated deposition for Stewart Trial B3/100.
A-22
400
-------
o
o
TRIAL 84/100
OBSERVED MAXIMUM
ISC MODEL
SUBJECTIVE EXTRAPOLATION
100 150
DISTANCE (meters)
FIGURE A-22. Observed maximum and calculated profile of non-dimensional
cross-wind integrated deposition for Stewart Trial B4/100.
200
A-23
-------
o
O
TRIAL B5/IOO
OBSERVED MAXIMUM
ISC MODEL
FIGURE A-23.
200 300
DISTANCE (meters)
Observed maximum and calculated profile of non-dimensional
cross-wind integrated deposition for Stewart Trial B5/100.
A-24
400
-------
10°
8
o
3=
I
TRIAL B5/200
OBSERVED MAXIMUM
rHJISC MODEL
^SUBJECTIVE EXTRAPOLATION
100 150
DISTANCE (meters)
200
FIGURE A-24.
Observed maximum and calculated profile of non-dimensional
cross-wind integrated deposition for Stewart Trial B5/200.
A-25
-------
o
o
-[^OBSERVED MAXIMUM
ISC MODEL
200
300
DISTANCE (meters)
400
FIGURE A-25.
Observed maximum and calculated profile of non-dimensional
cross-wind integrated deposition for Stewart Trial SI/50.
A-26
-------
Q
>—i
O
O
s-
. o-
TRIAL SI/100
OBSERVED MAXIMUM
ISC MODEL
-------
o
TRIAL S2/IOO
[OBSERVED MAXIMUM
ISC MODEL
-------
Q
o
TRIAL SI/100
OBSERVED MAXIMUM
ISC MODEL
200
300
400
DISTANCE (meters)
FIGURE A-26. Observed maximum and calculated profile of non-dimensional
cross-wind integrated depositi9n for Stewart Trial SI/I00.
A-27
-------
ol
TRIAL S2/IOO
OBSERVED MAXIMUM
ISC MODEL
300
FIGURE A-27.
400
500 600 700
DISTANCE (meters)
800
900
Observed maximum and calculated profile of non—dimensional
cross-wind integrated deposition for Stewart Trial S2/100.
A-28
-------
o
TRIAL S2/200
OBSERVED MAXIMUM
ISC MODEL
'200
FIGURE A-28.
300
400
500
DISTANCE (meters)
Observed maximum and calculated profile of non-dimensional
cross-wind integrated deposition for Stewart Trial S2/200.
A-29
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1 REPORT NO.
EPA-450/4-81-002
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
An Evaluation Study for the Industrial Source
Complex (ISC) Dispersion Model
5. REPORT DATE
January 1981
6. PERFORMING ORGANIZATION CODE
7 AUTHOfl(S)
J. F. Bowers and A. J. Anderson
8. PERFORMING ORGANIZATION REPORT NO.
9 PERFORMING ORGANIZATION NAME AND ADDRESS
H. E. Cramer Company, Inc.
University of Utah Research Park
P. 0. Box 8049
Salt Lake City, Utah 84108
10. PROGRAM~ELEMENT NO.
11 CONTRACT/GRANT NO.
68-02-3323
12. SPONSORING AGENCY NAME AND ADDRESS
U. S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Source Receptor Analysis Branch (MD-14)
Research Triannlo P*rk, NC 27711
13. TYPE OF RhPORT AND PERIOD COVERED
Einal
14. SPONSORING AGENCY CODE
EPA-450
15. SUPPLEMENTARY NO'TES
16. ABSTRACT
The Industrial Source Complex (ISC) Model contains several new features not
contained in other guideline atmospheric dispersion models. The important features
are the gravitational setting/dry deposition option and the building wake effects
option. Performance of the ISC Model is compared with that of CRSTER for a single
source and MPTER for multiple sources. Results indicate that these model options
significantly improve model performance. Additional suggestions for model improve-
ment are discussed. K
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Air Pollution
Atmospheric Models
Meteorology
Turbulent Diffusion
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Industrial Sources
Deposition
Downwash
Dispersion
fl. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (This Report)
Unclassified.
21, NO. OF PAGES
20.
22. PRICE
EPA Form 2220-1 (9-73)
------- |