EPA-910/9-86-147
December 1986
COMPARISON OF
AIR QUALITY MODEL ESTIMATES
WITH MEASURED S02 CONCENTRATIONS
NEAR MARCH POINT, WASHINGTON
Prepared by
Kirk D. Winges
EPA Contract No. 68-02-3886
Project Officer
Robert B. Wilson
U. S. Environmental Protection Agency, Region 10
1200 Sixth Ave.
Seattle, Washington 98101
TRC Environmental Consultants, Inc.
15924 22nd Ave. SE
Mill Creek, Washington 98012
(206) 485-2992
-------
DISCLAIMER
This report has been reviewed by Region 10, U. S.
Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily
reflect the views and policies of the U. S. Environmental
Protection Agency, nor does mention of trade names or
commercial products consitute endorsement or recommendation
for use.
-------
COMPARISON OF
AIR QUALITY MODEL ESTIMATES
WITH MEASURED S02 CONCENTRATIONS
NEAR MARCH POINT, WASHINGTON
TABLE OF CONTENTS
1 .0 INRODUCTION 1
1.1 Background 1
1 . 2 Purpose of the Study 2
1.3 Organization of the Current Document 2
2 .0 METHODOLOGY 4
2 . 1 The SHORTZ Model 4
2. 2 The ISCST Model 5
2 . 3 Emission Information 6
2 . 4 Meteorology 12
2 . 5 Receptors 13
2.6 Other Model Information 16
3 . 0 ANALYSIS RESULTS 18
4 . 0 SENSITIVITY ANALYSIS 69
4 . 1 Stack Tip Downwash 69
4.2 Wind Direction 69
4 . 3 Wind Speed 74
4.4 Atmospheric Stability 74
5 . 0 CONCLUSIONS 75
REFERENCES 78
APPENDIX A SAMPLE COMPUTER OUPUT FOR THE TEST CASES
-------
1.0 INTRODUCTION
TRC Environmental Consultants, Inc. was retained by the
Environmental Protection Agency to investigate the performance
of the SHORTZ and ISCST air quality models in predicting
sulfur dioxide concentrations in the vicinity of March Point,
Washington. The March Point area, located just to the east of
Anacortes, Washington is the site for two oil refineries and
limited other industrial development. TRC's role was to
utilize the air quality models to predict sulfur dioxide
concentrations for a study period running from May through
November, 1985. The predicted concentrations were then to be
compared with the measured concentrations to evaluate model
performance.
1.1 Background
The three industrial concerns of the March Point area in this
investigation, Texaco, Shell and Allied Chemical, in
cooperation with the Northwest Air Pollution Authority
(NWAPA), the Washington State Department of Ecology and the
U.S. Environmental Protection Agency (EPA), began a program to
monitor sulfur dioxide concentrations in the vicinity of the
March Point refineries several years ago. In the early
1980's, sulfur dioxide concentrations collected by Allied
Chemical in the March Point area showed violations of the
local five-minute and one-hour standards. In 1984, the EPA
conducted an evaluation of the sulfur dioxide- concentrations
in the March Point to determine 1) the source contributions to
the measured values, 2) the concentrations at locations other
than the air quality monitoring stations, and 3)
recommendations for siting of new air quality monitors in
areas of high concentration.
-------
The EPA study was based on the use of air quality modeling
techniques to predict ambient concentrations of sulfur
dioxide. The principal air quality model used in the
evaluation of the sulfur dioxide concentrations was the SHORTZ
Model, developed by the H. E. Cramer Company for the EPA. The
SHORTZ Model is the model recommended by the EPA for use with
sulfur dioxide emissions from buoyant sources located in urban
areas of complex terrain, defined as the presence of terrain
heights above the stack height. The terrain to the south of
the March Point area includes terrain heights above stack
height.
Based on EPA recommendations, the State Department of Ecology
and the NWAPA established three temporary monitors for sulfur
dioxide in the area to the south of the March Point industrial
area. A test period established as May, 1985 through
November, 1985 was used to collect data for model validation.
The data collected at the three monitors would be used to test
the model accuracy, and ultimately to select permanent agency-
operated sulfur dioxide monitoring station locations.
1.2 Purpose of the Current Study
The current study is the performance of air quality modeling
for comparison with the measured concentrations during the
test period of May, 1985 through November, 1985. Ultimately
the results of the study will be used to determine if the
SHORTZ Model or an alternative, the ISCST Model is an accurate
tool for the siting of air quality monitors in locations
similar to the March Point setting, and, if possible, to
select permanent monitor sites for the sulfur dioxide
monitoring in the March Point area.
-------
1.3 Organization of the Current Document
The current report documents all the proceedings of the TRC
investigation of the air quality model performance for the
March Point area. Section 2.0 describes the methodology of
the current study, including a brief description of the SHORTZ
and ISCST air quality models, a discussion of the
meteorological, emission, and other inputs used by the models,
and a discussion of the key decisions in running the models
(e.g. , the determination of stack-tip downwash using
computation of the Froude Number). Section 3.0 discusses the
results of the direct modeling of the cases selected by the
Department of Ecology. Section 4.0 discusses the sensitivity
analysis, which describes how the model results vary depending
on values selected for the input parameters. Finally, Section
5.0 discusses the conclusions of the study. Appendix A
presents sample computer printouts for the SHORTZ and ISCST
runs .
-------
2.0 METHODOLOGY
2.1 The SHORTZ Model
The SHORTZ air quality model was developed by the H. E. Cramer
specifically for simulating air quality impacts from multiple
source developments in complex terrain. For the purposes of
the current air quality modeling, complex terrain is defined
as the presence of terrain elevations in the area to be
modeled that are higher than the stack heights of the emission
sources. For the current study there are a total of 20
sources of emission, with emission heights above sea level
varying from 52 to 82 meters above sea level. Terrain
elevations of over 90 meters above sea level are located to
the south of the refineries within a distance of 2-3
kilometers. As a result, the area is judged to be complex
terrain. The Guideline on Air Quality Models, a document
published by the EPA, provides guidance on the appropriate air
quality models to use for certain applications, and the SHORTZ
Model is recommended for urbanized or industrialized areas of
complex terrain for sources such as the three industrial
facilities on March Point. The SHORTZ Model has been used in
numerous previous air quality studies in Western Washington--
most notably the evaluation of the ASARCO Tacoma copper
smelter.
The SHORTZ air quality model is well documented in the "User's
Instructions for the SHORTZ and LONGZ Computer Programs,
Volumes I and II", published by the EPA (EPA-903/9-82-004) .
No attempt will be made to describe the SHORTZ Model here.
The major model inputs can be grouped in four general classes:
o emission information,
-------
o meteorological data,
o receptor locations, and
o other information.
Each of these data requirements will be discussed in the
following sections.
2.2 The ISCST Model
The Industrial Source Complex Short-Term (ISCST) Model was
also developed by the H. E. Cramer Company for regulatory use.
The ISCST Model is well documented in the User's Guide for the
Industrial Source Complex Dispersion Model (EPA-450/4-86-005)
and will not be discussed in detail. It is very similar in
many regards to the SHORTZ Model, but differs in a few key
areas. Primarily, the differences concern the treatment or
ability to treat the effects of terrain on plume dispersion.
The SHORTZ Model was specifically designed to treat rough
terrain settings, defined as the presence of terrain heights
above the stack height in the area. The ISCST Model
specifically cannot treat rough terrain settings. In fact,
the ISCST Model does not allow the specification of terrain
heights above stack level. The ISCST Model can, however,
treat rolling terrain with heights below stack height.
The other principal difference between ISCST and SHORTZ
concerns the treatment for downwash. The ISCST Model is
designed to treat the complex effects of building wake
downwash on plume dispersion. The SHORTZ Model, although
having a treatment for stack-tip downwash, is not designed to
treat the effects of building wakes. For the current project,
building wakes are not considered to have a significant effect
on plume dispersion.
-------
2.3 Emission Information
Both models require that each source be identified with a
specific source identification number. The information which
must be provided for each source includes the emission rate in
grams per second, the source location, the stack height, the
elevation of the stack base, and a number of stack parameters
such as the emission temperature, the volume of the stack
gases emitted and the stack radius. Sources can also be
grouped and the results printed out in terms of a group's
contribution at each receptor to the total impact.
For the current project, there are three industrial facilities
being modeled: the Texaco oil refinery, the Shell oil refinery
and the Allied Chemical plant. Emissions at an oil refinery
are not constant, but rather vary from day to day depending on
the sulfur in the feed stock, the operating conditions, or the
shutting down of certain sources for maintenance. For
determining the air quality models' performance during the
test period, it was necessary to determine the emission
conditions for each source during the test period. The
Department of Ecology reviewed the air quality data for the
entire test period and selected certain periods for modeling.
They then obtained emission data from the industrial
facilities for those periods. The runs of the both models
were accomplished by adjusting the input parameters to reflect
actual conditions for each of the periods to be modeled.
A total of 20 different periods were selected by the
Department of Ecology and modeled in the current study. Ten
of these periods were one-hour episodes, while the remaining
10 were three-hour episodes. Table 2-1 depicts the input
values used for each of the major parameters for each stack of
concern in the current study.
-------
Table 2-1
Emission Rates Used In the Air Quality Modeling
Constant Parameters:
Source
UTM-X
UTM-Y
Stack Base Stack
Ht. (m) Elev. (m) Radius (m)
Allied Chem.:
101 532722 5369522
30.5
30
0.61
Texaco :
201
Shell:
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
532661
531961
531945
531923
531897
532029
532178
532170
531833
531845
532125
532125
531932
531924
531915
531875
531887
531906
531898
5368539
5371117
5371117
5371117
5371117
5371120
5371115
5371132
5371030
5371030
5371190
5371202
5370843
5370843
5370843
5370845
5370845
5370843
5370843
52
37
40
46
46
40
54
53
40
40
38
38
52
52
52
52
52
52
52
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
30
14
14
14
14
14
14
14
14
14
14
14
20
20
20
20
20
20
20
1
0
0
0
0
0
1
1
0
0
0
0
0
0
0
1
1
0
0
.41
.88
.88
.99
.72
. 84
.45
. 14
.69
.45
.87
. 87
.76
. 84
.84
.37
.37
.76
. 69
Variable Parameters:
Source
Emission
Rate (g/sec!
Stack
Temp. (°K)
May 22, 1985 (Cases 1, 11 and 12)
Allied Chem:
101 2.68 350
545
601
486
584
Texaco :
201
Shell:
301
302
303
175
8
11
10
.40
.95
.72
.46
Volume
Flow (m3/sec)
11 .94
77 . 10
11 .76
10. 19
13 . 54
-------
Table 2-1
(continued)
Emission Stack Volume
Source Rate (g/sec) Temp. (°K) Flow, (jm3/sec)
523 3.79
515 8.54
497 51.01
526 44.70
610 5.75
615 1.38
466 13.31
472 15.45
626 7.12
481 4.04
441 1.51
508 17.98
513 18.16
475 1.57
715 0.92
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
June 20, 1985
Allied Chem:
101
Texaco :
201
Shell:
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
3.78
9.07
78.37
64.64
5.29
1 .26
17.01
19.91
5.80
5.42
1.89
19.53
19.53
2 . 14
0.63
(Case 2)
3.76
195.60
5.80
6.43
14.24
1 .51
4.28
44 .86
37.84
2.52
0.63
18.90
13.61
3.91
2 .02
0.76
9.70
9.70
1 .13
1 . 13
350 11.94
545 77.10
593 12.81
513 11.51
614 21.93
548 3.39
509 8.46
493 46.39
516 40.52
600 5.43
605 1.44
478 17.96
472 12.88
669 10.18
506 4.14
460 1.42
505 18.39
511 18.61
481 1.95
810 4 . 23
-------
June 24
Allied
Source
101
Texaco :
201
Shell:
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
August
Allied
101
Texaco :
201
Shell:
301
302
303
304
305
306
307
308
309
310
311
Table 2-1
( continued)
, 1985 (Cases 3 and 13)
Chem:
Emission
Rate (g/secj
4.43
195.60
6.30
7.69
8.06
1 .76
4.41
50.90
41 .58
2.52
0.76
9.58
6.80
4.66
2 . 52
0.63
10.08
10.08
1.26
1.01
15, 1985 (Cases 4
Chem:
4.03
181 .80
3.91
4. 28
7.43
1 . 13
4.79
53 .05
43.09
0.38
0.00
13 . 86
11.59
Stack
) Temp. (°K)
350
545
594
522
606
539
505
498
523
575
605
475
466
681
524
451
499
500
485
690
and 14)
350
545
523
480
571
473
451
497
523
586
615
480
486
Volume
Flow (m3/sec;
11 .94
77. 10
11 .75
12.54
18.23
3. 18
7.96
52.40
44.82
4.68
1 .48
11 .39
8.49
10.04
4.31
1 .04
16.61
16.64
1 .95
3.34
11 .94
77.10
7 . 25
7.58
12.85
1 .40
5.67
50.88
43.67
0.50
1 .38
13.21
11.23
-------
Source
Emission
Rate jg/sec)
312
313
314
315
316
317
318
September 27,
Allied Chem :
101
Texaco:
201
Shell:
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
4.91
1.51
1.76
11 .09
11 .09
1.39
2.27
1985 (Cases 5
4.70
0.00
0.00
11 .34
14.36
1 .51
5.92
80.39
65.90
3 .40
1.13
16.00
16.00
7.81
2.539
1 .51
11.97
11 .97
1.89
3 . 15
October 5, 1985 (Case 7)
Allied Chem:
101
Texaco :
201
Shell:
301
302
3.22
0.00
0.00
8.57
Table 2-1
(continued)
Stack
Temp. (°K)
612
478
477
478
496
473
810
Volume
Flow (m3/sec)
6.64
1.80
1 .98
12.23
12.69
1 .58
4.79
15, 16 and 17
350
NA
NA
501
614
493
469
505
526
561
598
475
489
641
483
463
480
490
491
810
350
NA
NA
484
11 .94
NA
NA
11 . 19
17.75
1.41
5.53
55 .87
47 .57
3.73
1 .40
14 .92
14. 88
11 . 24
2 .02
1 .44
11 .18
11 .41
1 -94
8 .47
11 .94
NA
NA
10.65
-------
Table 2-1
( continued)
Source
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
November 10,
Allied Chem:
101
Texaco :
201
Shell:
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
Emission
Rate (g/sec)
14.74
1.01
5.04
92.36
76.73
2.77
0.76
15.88
19.66
5.29
1 .39
1 .39
7.94
7.94
1 .26
0.00
1985 (Cases 8,
0.13
179.30
0.00
7.69
14.74
2 .02
6.43
91 . 22
68.54
4 .03
0.88
33.52
33.64
4 .03
2.90
1 .51
11 .34
11 . 34
2 .02
1 . 89
Stack
Temp. (°K)
626
481
465
503
523
579
581
478
489
634
470
464
470
490
483
810
9, 10, 18, 19
350
545
NA
481
621
523
458
503
523
622
611
478
489
621
510
483
498
494
519
793
Volume
Flow (m3/sec)
21 . 16
1 . 22
5 .67
52.37
44.92
3.93
1 . 16
15.79
19.67
8.61
1 .86
1 .64
11 .03
11 .50
1 .56
8.47
and 20)
11 .94
77.10
NA
9.77
17.78
2 .50
8. 16
79.56
62 .65
6.77
1.79
20.99
21 .50
7 .09
4.55
2 .43
15.84
15 .72
3.61
6 . 41
-------
2.4 Meteorology
The air quality models require meteorological information in
the form of wind speed, wind direction, atmospheric stability,
mixing height and ambient temperature. One of the major
limitations of a Gaussian Plume Model, such as the SHORTZ
Model and the ISCST Model (and virtually every other model in
the Guideline on Air Quality Models) is that it assumes that
the atmosphere is in steady-state over all space and time for
the individual period being modeled (the base meteorological
data input rate). For the current study the base data rate is
hourly, and the models assume a single value for wind speed,
wind direction, stability, mixing height and temperature
applies for the entire area for one hour.
There were a number of sources of meteorological data for use
in the air quality modeling, and the first step in the
modeling procedure was the selection of the single value to
use for each of the cases selected by the Department of
Ecology. The sources of meteorological information included
the three industry monitors (Texaco, Shell and Allied) and
various airport weather stations, including Bellingham,
Whidbey Island Naval Air Station, Friday Harbor, and Paine
Field (Everett) . A composite table including all the
meteorological data was prepared, and in a meeting between the
EPA and TRC, values selected for use. Ultimately, the Texaco
Monitor was selected for the wind speed and wind direction,
while the Bellingham airport data was used for selection of
the atmospheric stability, mixing height and temperature
information.
The basis for the selection involved the consistency of the
Texaco and Bellingham data with the other stations, and the
proximity of location of these monitors to the sources and
12
-------
receptors. The Texaco Monitor agreed well with the Allied
monitor, while the Shell monitor differed substantially. Also
the Shell data were not available for some of the period of
interest. The Texaco monitor was also closest to the receptor
locations, since the Texaco refinery is south of the Shell
refinery, and the receptors (state-operated monitors) were to
the south of the Texaco refinery. The Texaco monitor did not
collect cloud-cover data (used for atmospheric stability and
mixing height) or temperature data. The Whidbey Island Naval
Air Station data was consistently in disagreement with the
other three airports. The Bellingham station was the closest
of the three remaining airports, and appeared to be the most
representative of the March Point area.
The meteorological data selected for use in the air quality
modeling are summarized in Table 2-2. It will be noted that
the stability information are presented by a letter class
designation. The letter classes were developed by Mr. Bruce
Turner to simulate different atmospheric mixing conditions and
are taken from the cloud cover and wind speed information in a
procedure recommended by the EPA (EPA, 1970). The procedure
involves the computation of the solar angle and the
determination of an insolation class number. The National
Climatic Center uses precisely the same methodology as used
here to generate stability class for development of
statistical wind roses.
2.5 Receptors
The air quality models require the specification of locations
at which to compute concentrations, called receptor locations.
The three primary locations used here are the sites of the
three temporary air quality monitors. They are referred to by
the names "Beebe", "Island Warehouse" and "Bullfinch". The
figures to be presented in Section 3.0 illustrate the
13 '
-------
Table 2-2
Meteorological Data Used In the Air Quality Modeling
Wind Atm. Mixing Amb.
Direction Stab. Height Temp
degrees Class (m) (°K)
Date
Case 1 :
5/22/85
Case 2 :
6/20/85
Case 3 :
6/24/85
Case 4 :
8/15/85
Case 5 :
9/27/85
Case 6:
9/27/86
Case 7 :
10/5/86
Case 8 :
11/10/85
Case 9 :
11/10/85
Case 10:
11/10/85
Case 11:
5/22/85
Case 12:
5/22/85
Case 13:
6/24/85
Hours
1300
1300
1400
1100
1200
0900
1100
0900
1000
1100
1200
1300
1400
1300
1400
1500
1200
1300
1400
Wind
Speed
Cm/sec)
1.79
2.68
1 .79
8.94
5.36
4.47
0.89
7.15
7.15
7.15
2.68
1 .79
2.68
1 .79
2 .68
2.68
2.68
2 . 68
1.79
360
360
360
340
360
360
270
30
20
20
360
360
340
360
340
320
360
360
360
B
B
B
D
D
C
B
B
B
B
B
C
A
B
1500 295
1500
1000
750
750
750
1500
750
750
750
1000
1500
1500
1500
1500
1500
1000
1500
1500
292
289
300
293
290
286
274
275
275
295
295
296
295
296
298
287
288
289
-------
Date
Hours
Wind
Speed
(m/sec)
Table 2-2
(continued)
Wind Atm.
Direction Stab.
degrees Class
Mixing
Height
Case 14
8/15/85
1100
1200
1300
8.94
4.47
8.94
340
360
350
D
C
D
750
750
750
300
301
302
Case 15:
9/27/85
1000
1100
1200
5.36
4.47
5.36
360
350
360
D
C
C
750
750
750
291
292
293
Case 16
9/27/85
0900
1000
1100
4.47
5.36
4.47
360
360
350
C
D
C
750
750
750
290
291
292
Case 17
9/27/85
1100
1200
1300
4.47
5.36
5.36
350
360
360
C
C
C
750
750
750
292
293
293
Case 18:
11/10/85
0800
0900
1000
7. 15
7.15
7.15
30
30
20
D
D
D
750
750
750
274
274
275
Case 19:
11/10/85
0900
1000
1100
7. 15
7. 15
7.15
30
20
20
D
D
D
750
750
750
274
275
275
Case 2 0:
11/10/85
1100
1200
1300
7.15
7. 15
7. 15
20
10
20
D
D
D
750
750
750
275
275
275
-------
locations of these monitors. The information provided to the
models concerning these monitors include the location in
Universal Transverse Mercator (UTM) coordinates, and the
elevation of the ground at the receptor location. For the
current project, UTM coordinates had been provided to TRC in
the data for the air quality monitoring stations. However,
these coordinates did not match the map-identified locations
for the sources. Consequently, to be consistent with the
display maps, the UTM coordinates were modified slightly to so
they would plot correctly in the figures of Chapter 3.0.
Receptor heights were also provided to TRC with the data for
the monitor sites.
In addition to the three air quality monitoring sites, a grid
of receptors was determined for the air quality modeling. A
total of 143 receptors, spaced at 250 meters apart on an 11 by
13 grid were established. • For each receptor, the UTM
coordinate and terrain elevation were determined.
2.6 Other Model Information
The final block of information provided to the models included
the values to use for a number of switches and miscellaneous
parameters. In general, default values were used for most of
the other parameters, such as potential temperature gradients,
entrainment coefficients, accelerations due to gravity,
rectilinear plume expansion distance, power law exponents for
the wind speed, and the turbulence intensities for each
stability class.
One particular area deserves comment. The User's Instructions
for the SHORTZ Model provide guidance concerning stack-tip
downwash, a process whereby the plume is caused to decrease in
height due to the aerodynamic influence of the stack in the
wind. It has been determined from experimental evidence that
16
-------
the tendency of stack-tip downwash to influence a plume is a
function of the Froude Number for the stack, a mathematical
construct which ratios the momentum force of a plume to its
buoyant force. For plumes with Froude numbers greater than
3.0, the momentum dominates, and the stack tip downwash is
applicable. For plumes with Froude Numbers less than 1.0, the
buoyant forces dominate and the stack-tip downwash does not
apply. For stacks with Froude Numbers in the range between
1.0 and 3.0, the applicability of the stack-tip downwash is
not certain. For the current study, the value of 3.0 was used
to determine if stack-tip downwash should be used. However,
the sensitivity analysis discussed in Section 4.0 addresses
the use of the alternate (1.0) Froude Number criterion.
17
-------
3.0 ANALYSIS RESULTS
The SHORTZ Model was run for the 20 cases selected by the
Department of Ecology. For each case, concentrations were
computed at a total of 146 receptors — the three monitor
locations and the 143 grided receptors. Results at the
monitor locations are summarized in Table 3-1, while Figures
1-20 illustrate the full picture for both the grided receptors
and the three monitor locations.
The ISCST Model was run for the same 20 cases selected by the
Department of Ecology. For each case, concentrations were
computed at a total of 146 receptors -- the three monitor
locations and the 143 grided receptors. Results at the
monitor locations are summarized in Table 3-1, while Figures
21-40 illustrate the full picture for both the grided
receptors and the three monitor locations.
In addition to the summaries shown in the table and figures,
Appendix A contains sample SHORTZ and ISCST computer
printouts.
Table 3-1 also includes the measured values at the three
monitors for the period of interest. By comparison of the
measured versus the predicted values in Table 3-1, the overall
performance of the models can be assessed. Both the SHORTZ
and the ISCST Model had concentrations in the same order of
magnitude as the measured values. Neither of the models
predict concentrations which correlate well with the measured
values. A linear regression was performed for each of the
three sites with the result indicating that the correlation
coefficient (r-squared) was 0.04 and 0.02 for the one-hour and
three-hour concentrations respectively when evaluated with the
SHORTZ Model. For the ISCST Model the correlation
18
-------
Table 3-1
Comparison of Model Results with Measured Concentrations
SC>2 Concentration (ppm]
Case
Island Warehouse
Meas. Shortz ISCST
Bullfinch
Meas. Shortz ISCST
Beebe
Meas. Shortz ISCST
One-hour
1
2.
3
4
5
6
7
8
9
10
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
Cases :
03
02
03
03
05
01
03
02
01
09
Three-hour
11
12
13
14
15
16
17
18
19
20
0.
0.
0.
0.
0.
o.
0.
0.
0.
0.
02
02
02
03
03
02
03
02
04
08
0
0
0
0
0
0
0
0
0
0
.03
.08
.02
.00
.01
.01
.00
.00
.00
.00
0
0
0
0
0
0
0
0
0
0
.03
.07
.03
.01
.01
.01
.00
.00
.00
.00
0.08
0.02
0.03
0.01
NA
0.04
0.04
0.01
0.01
0.01
0
0
0
0
0
0
0
0
0
0
.07
.04
.06
.00
.03
.05
.00
.00
.00
.00
0
0
0
0
0
0
0
0
0
0
.07
.02
.05
.00
.04
.07
.00
.00
.00
.00
0
0
0
0
0
0
0
0
0
NA
.08
. 10
.09
. 16
. 11
.08
.09
. 10
.06
0
0
0
0
0
0
0
0
0
0
.05
.06
.03
.00
.03
.01
.00
.00
.00
.00
0.05
0.08
0.04
0.00
0.03
0.02
0.00
0.00
0.00
0.00
Cases :
0
0
0
0
0
0
0
0
0
0
.02
.02
.04
.05
.03
.03
.03
.00
.00
.00
0
0
0
0
0
0
0
0
0
0
.03
.02
.04
.03
.03
.03
.03
.00
.00
.00
0.06
0.06
0.03
NA
NA
0.02
NA
0.01
0.01
0.01
0
0
0
0
0
0
0
0
0
0
.04
.02
.04
.01
.03
.03
.02
.00
.00
.03
0
0
0
0
0
0
0
0
0
0
.05
.02
.04
.02
.04
.04
.03
.00
.00
.01
0
0
0
0
0
0
0
0
NA
NA
.06
.06
.10
.09
.09
.08
.08
.04
0
0
0
0
0
0
0
0
0
0
.03
.02
.04
.02
.03
.03
.04
.00
.00
.00
0.05
0 .02
0.05
0.03
0.04
0.04
0.05
0 .00
0.00
0 .00
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Environmental Consultants, Inc.
YMAX = 5368125
YMIN = 5365375
* Monitor Location
0
0.5 1.0
Scale in Kilometers
1.5
Figure 1
CASE 1 - SHORTZ
1-Hour Concentrations
for May 22, 1985, 1300 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
SOg CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC EnTiroiunental Consultants, Inc.
YMAX = 5368125
YMIN - 5365.375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 2
CASE 2 - SHORTZ
1-Hour Concentrations
for June 20, 1985, 1300 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Environmental Consultants, Inc.
YMAX = 5368125
\lslanii Warehouse
"IT
YMIN = 5365375
1
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 3
CASE 3 - SHORTZ
1-Hour Concentrations
for June 24, 1985, 1400 PST
(contour interval = 0.01 ppm)
— _.
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TKC Environmental Consultants, Inc.
YMAX
5368125
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 4
CASE 4 - SHORTZ
1-Hour Concentrations
for Aug. 14, 1985, 1100 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC BnTiroiunental Consultants, Inc.
YMAX = 5368125
Island Warehouse
*
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 5
CASE 5 - SHORTZ
1-Hour Concentrations
for Sept. 27, 1985, 1200 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TEC Environmental Consultants, Inc.
YMAX = 5368125
Island Warehouse
•*•
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 6
CASE 6 - SHORTZ
i-Hour Concentrations
for Sept. 27, 1985, 0900 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
SOg CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Environmental Consultants, Inc.
YMAX = 5368125
Island Warehouse
*
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 7
CASE 7 - SHORTZ
1-Hour Concentrations
for Oct. 5, 1985, 1100
(contour interval = 0.0
PST
1 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC En-Tironmental Consultants, Inc.
YMAX
5368125
Island Warehouse
*
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 8
CASE 8 - SHORTZ
1-Hour Concentrations
for Nov. 10, 1985, 0900 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
SOg CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC BnTironmental Consultants, Inc.
YMAX = 5368125
Island Warehouse
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 9
CASE 9 - SHORTZ
1-Hour Concentrations
for Nov. 10, 1985, 1000 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TEC EnTironmental Consultants, Inc.
YMAX = 5368125
Island Warehouse
*
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 10
CASE 10 - SHORTZ
1-Hour Concentrations
for Nov. 10, 1985, 1100 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC KnYiromnentol Consultants, Inc.
YMAX = 5368125
YMIN = 5365375
* Monitor Location
0 0.5 i.O 1.5
Scale in Kilometers
Figure 11
CASE 11 - SHORTZ
3-Hour Concentrations
for May 22, 1985, 1200-1400 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Enyironmental Consultants, Inc.
rMAX = 5368125
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 12
CASE 12 - SHORTZ
3-Hour Concentrations
for May 22, 1985, 1300-1500 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
SOg CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC EnTironmental Consultants, Inc.
TMAX = 5368125
YMIN
5365375
* Monitor Location
0 0.5 i.O 1.5
Scale in Kilometers
Figure 13
CASE 13 - SHORTZ
3-Hour Concentrations
for June 24, 1985, 1200-1400 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Environmental Consultants, Inc.
YMAX = 5368125
YMIN = 5365375
I
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 14
CASE 14 - SHORTZ
3-Hour Concentrations
for Aug. 15, 1985, 1100-1300 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Environmental Consultants, Inc.
TMAX
5368125
Island Warehouse
YMIN
5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 15
CASE 15 - SHORTZ
3-Hour Concentrations
for Sept. 27, 1985, 1000-1200 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC BnTironmental Consultants, Inc.
YMAX
5368125
Island Wanehouse
*
O
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 16
CASE 16 - SHORTZ
3-Hour Concentrations
for Sept. 27, 1985, 0900-1100
(contour interval = 0.01 ppm)
PS1
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Eimrorunental Consultants, Inc.
YMAX = 5368125
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 17
CASE 17 - SHORTZ
3-Hour Concentrations
for Sept. 27, 1985, 1100-1300
(contour interval = 0.01 ppm)
PST
-------
MARCH POINT MODEL
EVALUATION PROJECT
SOg CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TEC Environmental Consultants, Inc.
YMAX = 5368125
YMIN = 5365375
1
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 18
CASE 18 - SHORTZ
3-Hour Concentrations
for Nov. 10, 1985, 0800-1000 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
SOg CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Environmental Consultants, Inc.
YMAX = 5368125
Island Warehouse
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 19
CASE 19 - SHORTZ
3-Hour Concentrations
for Nov. 10, 1985, 0900-1100 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Environmental Consultants, Inc.
YMAX = 5368125
YMIN = 5365375
I
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 20
CASE 20 - SHORTZ
3-Hour Concentrations
for Nov. 10, 1985, 1100-1300 PST
(contour interval = 0.01 ppm)
.
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TEC Environmental Consultants, Inc.
YMAX = 5368125
lstcm
-------
MARCH POINT MODEL
EVALUATION PROJECT
SOg CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Knrironmental Consultants, Inc.
YMAX = 5368125
YMIN = 5365375
* Monitor Location
0 0.5 i.O 1.5
Scale in Kilometers
Figure 22
CASE 2 - ISCST
1-Hour Concentrations
for June 20, 1985, 1300 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Bnviroiunental Consultants, Inc.
YMAX = 5368125
Island warehouse
YMIN
5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 23
CASE 3 - ISCST
1-Hour Concentrations
for June 24, 1985, 1400 PST
(contour interval = 0.01 ppm)
.
-------
MARCH POINT MODEL
EVALUATION PROJECT
CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Environmental Consultants, Inc.
YMAX = 5368125
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 24 1
CASE 4 - ISCST
1-Hour Concentrations
for Aug. 14, 1985, 1100 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Environmental Consultants, Inc.
YMAX = 5368125
Island Warehouse
YMIN = 5365375
* Monitor Location
0
0.5 1.0
Scale in Kilometers
1.5
Figure 25
CASE 5 - ISCST
1-Hour Concentrations
for Sept. 27, 1985, 1200 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC BnTironmental Consultants, Inc.
YMAX = 5368125
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 26
CASE 6 - ISCST
i-Hour Concentrations
for Sept. 27, 1985, 0900 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Bnrironmental Consultants, Inc.
YMAX = 5368125
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 27
CASE 7 - ISCST
1-Hour Concentrations
for Oct. 5, 1985, 1100 PST
(no concentrations)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC KnTiroiunental Consultants, Inc.
YMAX = 5368125
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
i i
Figure 28
CASE 8 - ISCST
1-Hour Concentrations
for Nov. 10, 1985, 0900 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Environmental Consultants, Inc.
YMAX = 5368125
Island Warehouse
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 29
CASE 9 - ISCST
1-Hour Concentrations
for Nov. 10, 1985, 1000
(contour interval = 0.01
PST
ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Environmental Consultants, Inc.
YMAX = 5368125
Island Warehouse
*
YMIN = 5365375
* Monitor Location
Figure 30
o
0.5 1.0
Scale in Kilometers
1.5
CASE 10 - ISCST
1-Hour Concentrations
for Nov. 10, 1985, 1100 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
SOg CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC EnYironmental Consultants, Inc.
YMAX = 5368125
Inland Warehouse
*
YMIN = 5365375
* Monitor Location
0
0.5 1.0
Scale in Kilometers
1.5
Figure 31
CASE 11 - ISCST
3-Hour Concentrations
for May 22, 1985, 1200-1400 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TKC Environmental Consultants, IDC.
YMAX - 5368125
Island Warehouse
*
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 32
CASE 12 - ISCST
3-Hour Concentrations
for May 22, 1985, 1300-
(contour interval = 0.01
1500 PST
ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Environmental Consultants, IDC.
YMAX = 5.368125
YMIN
5365375
I
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 33
CASE 13 - ISCST
3-Hour Concentrations
for June 24, 1985, 1200-1400 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
SOg CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Environmental Consultants, Inc.
YMAX
5368125
YMIN = 5365375
* Monitor Location
0 0.5 1.0 -1.5
Scale in Kilometers
Figure 34
CASE 14 - ISCST
3-HoTir Concentrations
for Aug. 15, 1985, 1100-1300 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Environmental Consultants, Inc.
YMAX = 5368125
YMIN
5365375
0
Monitor Location
0.5 1.0
Scale in Kilometers
1.5
Figure 35
CASE 15 - ISCST
3-Hour Concentrations
for Sept. 27, 1985, 1000-1200 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TKC Environmental Consultants, Inc.
YMAX = 5368125
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 36
CASE 16 - ISCST
3-Hour Concentrations
for Sept. 27, 1985, 0900-1100
(contour interval = 0.01 ppm)
PST
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Environmental Consultants, Inc.
YMAX = 5368125
YMIN = 5365375
* Monitor Location
0
0.5 1.0
Scale in Kilometers
1.5
Figure 37
CASE 17 - ISCST
3-Hour Concentrations
for Sept. 27, 1985, 1100-1300 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
S02 CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Enriroiunental Consultants, Inc.
YMAX = 5368125
Island Warehouse
*
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 38
CASE 18 - ISCST
3-Hour Concentrations
for Nov. 10, 1985, 0800-1000 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Bnrironmental Consultants, Inc.
YMAX = 5368125
Island Warehouse
•*•
YMIN = 5365375
0
Monitor Location
0.5 1.0
Scale in Kilometers
1.5
Figure 39
CASE 19 - ISCST
3-Hour Concentrations
for Nov. 10, 1985, 0900-1100 PST
(contour interval = 0.01 ppm)
-------
MARCH POINT MODEL
EVALUATION PROJECT
SOg CONCENTRATIONS (ppm)
Date Prepared:
Dec. 7, 1986
TRC Environmental Consultants, lac.
YMAX = 5366125
YMIN = 5365375
* Monitor Location
0 0.5 1.0 1.5
Scale in Kilometers
Figure 40
CASE 20 - ISCST
3-Hour Concentrations
for Nov. 10, 1985, 1100-1300 PST
(contour interval = 0.01 ppm)
_ i
-------
coefficients were even lower (0.0001 and 0.005 for the one-
hour and three hour cases respectively.
The data from Table 3-1 have been plotted in Figures 41 and 42
in the format of a scatter plot. The lack of correlation is
evident by the wide spread from the perfect agreement line
which would run on a diagonal from the lower left of the box
to the upper right in each of the plots.
Linear regressions are certainly not the only means of
evaluation of a model's performance. In fact the linear
regression is not often not used in air quality model
evaluation, because linear regression illustrates how well two
data sets are correlated, but not how accurate the model is at
predicting concentrations. The model might over-predict by a
factor of five and still give perfect correlation.
More importantly, many regulatory applications concern only
the ability of the air quality model to predict the peak or
worst-case concentration, not the entire distribution of
concentrations. Thus, often the model is evaluated simply in
terms of its ability to predict the highest concentrations
measured over the entire field of receptors. Cumulative
frequency plots are made of the model's performance, where
highest predicted is compared to highest measured value
without regard to whether the two values coincide in space and
time.
A new technique has recently been prepared to assess air
quality model performance. The technique, presented by Cox,
et. al. (1985) involves the computation of two parameters: a
fractional bias of the average values (FB) and a fractional
bias of the standard deviation (FO) . The FB and FO are then
plotted on a special graph and the closer the values come to
the center of the graph, the better the agreement of the model
60
-------
MARCH POINT MODEL EVALUATION
One—Hour Average Concentrations
?
Q_
0.
c
0
+J
0
"c
-------
MARCH POINT MODEL EVALUATION
Three-Hour Average Concentrations
?
Q.
a
c
0
£
0
+J
c
0
0
C
0
o
n
-------
predictions with observed concentrations.
Figure 43 illustrates such a graph for the current project for
the one-hour concentrations. Figure 44 illustrates the same
information for the three-hour concentrations. The
performance of the ISCST and SHORT air quality models is seen
in these figures. The box at the center of each figure is
said to represent the "factor of two" agreement that is often
referenced for air quality models. As the two figures show,
only one of the four points plotted is inside the factor of
two box, and even that value (ISCST, 3-hour concentrations) is
almost out of the box. In general, then the SHORTZ and ISCST
air quality models are not performing within the factor of two
performance level when predictions and observations are paired
in space and time.
The models both agree on the source apportionment. In
general, for the receptors close to the Texaco refinery, the
Texaco source does not contribute to the calculated
concentrations. The Allied source is a minor contributor
during all conditions, due to the low emission rate. The
large number and emissions of the Shell sources make them the
major contributors for the close receptors, although under
some conditions, the Texaco source was seen to contribute 20%
of the concentration. For the Bullfinch receptor, the
contribution of the Texaco source increases to 40% or more.
The reason is that the receptor is higher and the Texaco plume
no longer passes overhead as it does with the closer
receptors. Additionally, since the source/receptor distance
is large, there is more time for the plume to mix to the
ground in transit, and since the Texaco source is a major
source and closer to the Bullfinch receptor than the Shell or
Allied sources, it's percentage contribution increases.
One additional analysis was performed by comparing the maximum
63
-------
March Point Model Evaluation
Model Performance for 1—Hour Averages
1 —'
-------
March Point Model Evaluation
Model Performance for 3-Hour Averages
1 —
W
1-1
o
m
0
-1 —
-2-
- ISCST
- SHORTZ
-2
1 0
Bias of Average
Figure 44
-------
prediction and observed concentrations from each time period,
regardless of location (paired in time, but unpaired in
space). The results, depicted in Figures 45 and 46 indicate
agreement between measured and predicted is much better. The
implication of this final analysis is that the models are
capable of predicting the maximum concentrations, and even
capable of predicting when they may occur, but not capable of
predicting the location. Therefore the models may not provide
accurate siting information for the location of monitors in
the vicinity of sources.
The general conclusion is that both the SHORTZ and ISCST air
quality models perform poorly with the March Point data.
Figures 43 and 44 clearly show a tendency of both models
toward underprediction.
-66
-------
March Point Model Evaluation
Model Performance for 1-Hour Averages
CO
4-1
o
OT
CQ
H
-1
_l
-2-
A - ISCST
-------
March Point Model Evaluation
Model Performance for 3—Hour Averages
-I
1 —
I
CO
4-1
o
1/5
£
o
-l
-I
-2-
A - ISCST
* - SHQRTZ
-2
-1 0
Bias of Average
Figure 46 Comparison of Measured Versus Predicted for
Data Paired in Time, but not in Space.
-------
4.0 SENSITIVITY ANALYSIS
The following sections discuss the sensitivity of the
predicted model results to the values assumed for the inputs.
4.1 Stack Tip Downwash
As discussed in Section 2.0, the Froude Number is computed for
the SHORTZ Model to determine if stack tip downwash is to be
used for a particular source. The value to use as a criterion
for applying the downwash correction based on the Froude
Number is a point of some uncertainty. In the current
analysis, a value of 3.0 was used as a criterion. Stacks with
Froude Numbers greater than 3.0 were assumed to experience
stack tip downwash, while those with Froude Numbers less than
3.0 were not. A sensitivity analysis was conducted to
determine the effect on the results if the Froude Number
criterion had been 1.0 instead of 3.0. For two cases (Cases 1
and 9), the model predictions were repeated with the Froude
Number criterion changed, and the computed concentrations were
identical to those with the Froude Number criterion of 3.0.
Thus, the Froude Number criteria is determined to have no
influence on the modeled concentrations.
The Froude number computation is not a part of the ISCST Model
analysis.
4.2 Wind Direction
The model predictions at a given location are highly dependant
on wind direction. The effect results because the wind
directions are imprecisely known, and because any short-term
Gaussian Plume model will have a strong concentration gradient
69
-------
in the cross wind direction. Examination of Figure 1 for the
SHORTZ Model near the Beebe monitor shows this gradient
particularly well. To illustrate the effect of a change in
wind direction on the results, Figure 47 has been prepared
which shows Case 1, repeated with the wind direction modified
by 10 degrees either to the east or to the west using the
SHORTZ Model. A similar plot is show in Figure 48 for the
ISCST Model. Although the plots are reduced, and somewhat
difficult to read, the effect on concentration of changing the
wind direction is dramatic, and can easily be seen by
examining the position of the Beebe and Island Warehouse
monitors. For the Case 1 plot (center of Figure 47), the
Beebe monitor is located near the 0.05 isopleth. When the
wind shifts 10 degrees to the east (lower plot), the Beebe
monitor is moved to the center of the plume and concentrations
are increased to over 0.08 ppm.
The opposite occurs when the wind is shifted the other
direction. The shift of only 10 degrees results in a decrease
in concentration at the Beebe monitor to only 0.01 ppm. The
net effect of a 20 degree change in wind direction is a change
in the- concentration by a factor of 8. The same effect is
seen at the Island Warehouse receptor- The effect is present,
although less pronounced at the Bullfinch monitor. In
general, the sensitivity to wind direction changes decreases
with increasing distance from the source. Table 4-1
summarizes the sensitivity analysis for Case 1. As Table 4-1
shows, virtually the identical sensitivity to wind direction
is observed for the ISCST Model.
Since the wind direction is imprecisely known and could easily
vary by 10 degrees or more within an hour, the magnitude of
change in the concentrations greatly reduces the confidence in
the model predictions.
70
-------
Figure 47 Wind Direction Sensitivity for SHORTZ
Wind Direction Shifted 10 Degrees West
YUM - 034MTB
Wind Direction as in Case 1
YUMC - 9MM0
Wind Direction Shifted 10 Degrees East
-------
Figure 48 Wind Direction Sensitivity for ISCST
Wind Direction Shifted 10 Degrees West
Wind Direction as in Case 1
YWX - 9MMB
Wind Direction Shifted 10 Degrees East
-------
Table 4-1
Sensitivity Analysis for Wind Direction,
Wind Speed, and Stability
Model Predicited Concentration
of Sulfur Dioxide in ppm
Case Island Warehouse Bullfinch Beebe
SHORTZ
Case 1 Unchanged 0.03 0.07 0.05
Wind Direction:
10° East 0.00 0.07 0.01
10° West 0.08 0.03 0.08
Wind Speed:
1 m/sec increase 0.05 0.06 0.04
1 m/sec decrease 0.03 0.08 0.05
Stability:
1 Class less stable 0.06 0.05 0.05
1 Class more stable 0.01 0.07 0.04
ISCST
Case 1 Unchanged 0.03 0.07 0.05
Wind Direction:
10° East 0.00 0.05 0.01
10° West 0.08 0.03 0.08
Wind Speed:
1 m/sec increase 0.04 0.06 0.04
1 m/sec decrease 0.04 0.08 0.08
Stability:
1 Class less stable 0.08 0.03 0.07
1 Class more stable 0.02 0.09 0.06
-------
4.3 Wind Speed
The effect of wind speed on the model prediction of
concentrations is also significant. To illustrate the
influence of wind speed, Case 1 was modeled with the wind
speed increased by 1 m/sec and decreased by 1 m/sec. The
results are summarized for the three monitor locations in
Table 4-1. As the table indicates, the concentrations are
generally increased for a reduction in wind speed, while an
increase in wind speed usually results in a decrease in
concentrations. The results are not as sensitive to wind
speed as they are to wind direction. The sensitivity
decreases as wind speed increases, so for some of the other
cases, where wind speeds were higher, (e.g. Cases 4, 5, 6, 8,
9, 10, 14, 15, 18, 19 and 20) the sensitivity should not be as
great. A 1 m/sec variation in the wind speed is not an
unexpected level of uncertainty for such measurements.
4.4 Atmospheric Stability
The atmospheric stability influences the mixing in the
atmosphere and hence the dilution of the plume as it moves
downwind. As a result, the stability assumed in the modeling
has a significant influence on the model concentrations. To
illustrate the effect of stability, Case 1, which was
originally modeled as a class "B" stability has also been
modeled as a class "A" and a class "C" stability. The effect
is shown in Table 4-1.
As the table shows, the model results are very significantly
influenced by the assumed stability. Since stability was
determined for the current analysis from cloud cover
observations at Bellingham and on site wind speed observations
on the Texaco refinery, there are large uncertainties in the
stability class assignments for each of the cases.
74
-------
5.0 CONCLUSIONS
The current analysis has been performed to evaluate the
ability of the SHORTZ and ISCST air quality models to predict
sulfur dioxide concentrations in the vicinity of March Point,
Washington. Both models were used to predict concentrations
for a total of 20 test cases for an experimental period
running from May, 1985 through November, 1985. Both one-hour
and three-hour cases were considered, and the results compared
to measured concentrations at three monitoring sites located
just to the south 'of the industrialized area of March Point.
While measured concentrations were relatively low, the ten
highest one-hour and three-hour average concentrations were
selected for evaluation.
Neither model was judged to give good agreement between
measured and predicted concentrations paired in space and
time. A major reason for the poor performance is the
inaccuracy of the input information. A sensitivity analysis
illustrated both models' extreme sensitivity to values of
input parameters, particularly wind direction. The inability
to accurately specify the wind direction for an hourly average
could result in concentrations being off by close to an order
of magnitude. Another element of uncertainty is the knowledge
of the emission information. For many of the sources the
stack parameters were only imprecisely known, and better
information on the exact emission rates and emission
conditions would probably greatly improve model reliability.
When the predictions and observations were paired in space and
time, the models were biased toward underprediction. When the
predictions and observations were unpaired in space but paired
in time, the models performed more favorably. In fact the
75
-------
overall magnitude of the measured values was quite similar to
the model predictions, so that on a cumulative frequency
basis, both models may have done acceptably. However, the
models did not predict within the customary "factor-of-two"
performance usually given to air quality models when the data
are paired in space and time.
The inaccuracy of the input information, while a major source
of error, may not be the only problem. Complex terrain
settings, such as March Point are very difficult to model
accurately. In particular, the assumption of steady state in
space (the assumption that a single value of the wind
direction and speed applies for all space), is simply not
valid for rough terrain settings. It is true there are not
other options in the absence of additional data, and for
regulatory purposes, the Gaussian-plume (steady state) models
will continue to be used because they have wide agency
acceptance. Situations like the March Point analysis are the
inevitable consequence of the reliance on Gaussian dispersion.
To improve the March Point model performance, better on-site
data should be collected and reduced. In particular, detailed
knowledge of the wind direction both at the source and the
receptor would enable a more accurate air quality analysis.
The results of the current analysis do not favor one model
over the other; therefore, no recommendation can be given
concerning the most appropriate air quality model to use for
the March Point location, except for the areas where receptor
heights are greater than stack height, and the SHORTZ Model
must be used since the ISCST Model does not permit receptor
heights greater than stack height.
An important question which must be asked is the need for
76
-------
continued monitoring at March Point. The concentrations of
sulfur dioxide were not approaching any applicable air quality
standard at any of the monitors and it might be concluded that
the public is not at risk from exposure to sulfur dioxide.
However, oil refineries can change emission rates drastically
depending on the quality of the feed stock and the fuels
combusted at the site. Future monitoring may see higher
concentrations if conditions change at the refinery and it may
be important to continue to monitor in the March Point area.
The Beebe residence is the location where maximum
concentrations have been measured previously, and it should
continue to be the point of measurement. Future ambient
monitoring, if performed with accurate meteorological and
emissions sampling programs could yield a valuable air quality
data set for model validation and calibration in the March
Point area.
It should be noted here that the conclusions stated here
concern solely sulfur dioxide concentrations. No
consideration has been given to other chemical species which
might be emitted by any of the facilities in the March Point
area.
77
-------
REFERENCES
U.S. EPA, 1970. "Workbook for Atmospheric Diffusion
Estimates", EPA Document No. AP-26, Research Triangle Park,
NC.
U.S. EPA, 1982. "User's Instructions for the SHORTZ and LONGZ
Computer Programs, Volumes I and II", EPA Report Number EPA-
903/9-82-004.
U.S. EPA, 1986. "User's Guide for the Industrial Source
Complex Model", EPA Document Number EPA-450/4-86-005.
78
-------
Appendix A
Sample Computer Output for the Test Cases
-------
SHORTZ AVERSION 92326^
AN A!RVQUALITY DISPERSION MODEL IN
SECT!ON 2. NON-GUIDELINE MODELS,
TN "NAMAP 'VERSION 5^ DEC °2
SOURCC- i:T! c 2° ON 'JNAMAP MAGNETIC TAPF CROM NTIS
-------
ONE HOUR CASE 1 - NO DOWNIKASH FOR FR < 3, WS=4 MPH, !KD=360, STAB=B
TABLE 1
GENERAL INPUT DATA -
NUMBER OF INDUT SOURCES
NUMBER OF X GRID COORDINATES
NUMBER OF Y GRID COORDINATES
TOTAL NUMBER OF HOURS IN EACH DAY
NUMBER OF DAYS OR CASES
NUMBER OF CONCENTRATION REPORTS (SOURCE COMBINATIONS)
NUMBER OF DISCRETE CALCULATION POINTS
MET DATA INPUT CARD RATE (1=HOURLY,2^2 HOURLY,ETC)
IS CONCENTRATION CALCULATED AT BASE RATE PRINTED
NO. OF HOURS IN FIRST AVERAGE CONCENTRATION PRINTED
NO. OF HOURS IN SECOND AVERAGE CONCENTRATION PRINTED
NO. OF HOURS IN THIRD AVERAGE CONCENTRATION PRINTED
ARE TERRAIN ELEVATION HEIGHTS USED
IS WIND SPEED TERRAIN FOLLOWING
ARE CONCENTRATIONS AVERAGED OVER DAYS OR CASES
IS THE FORMA7 COR SOURCE DATA READ
IS COORDINATE SYSTEM CARTESIAN (=0) OR POLAR (=1)
ARE DISCRETE RECEPTORS CARTESIAN (=9) OR °OLAR (=1)
ARE SOURCE COORDINATES CARTESIAN (=0) OR POLAR (=1)
SIGEPU SIGAPU FOR ALL SOURCES OPTION
RURAL/URBAN MODE OPTION (RURAL=0),(URBANE)
MODEL UNITS CONVERSION FACTOR
ACCELERATION OF GRAVITY
HEIGHT OF MEASUREMENT OF WIND SPEED, ETC
ENTRAPMENT COEFFICIENT FOR UNSTABLE ATMOSPHERE
ENTRAPMENT COEFFICIENT FOR STABLE ATMOSPHERE
DISTANCE OVER WHICH RECTILINEAR PLUME EXPANSION OCCURS
DECAY COEFFICIENT COR PHYSICAL OR CHEMICAL DEPLETION
ANGULAR DISPL OF GRID SYSTEM CROM TRUE NORTH
ELEVATION OF BASE OF WEATHER STATION
X ORIGIN OF POLAR COORDINATES
Y ORTG'N OF POLAR COORDINATES
*-* COORDINATE SYSTEM X AXIS (METERS) *-*
:3
-------
ONE "HOUR CASE 1 - NO DOWNWASH FOR CR < 3, WS=4 MPH, WO=35Q, STASIS
TAS!c ^
- GENERAL INP'JT CATA
*-* GRID SYSTEM TERRAIN HEIGHTS (METERS) *-*
- GRID SYSTEM X AXIS (METERS) -
DATE
, CASE 1, °AGE
AXIS (METERS
\
5363000.000
5357750.900
5357509.
5357250.
5357000.
5365750,
5355500.
5356250.
5365009.
qocjin^n
5355500.
,009
onn
. J y ^
909
,000
.000
.000
.000
onn
.999
5
15
2".
35
51
54
51
54
C7
57
.0000000
.1000000
.2000000
.4000290
.5099990
.8000000
.9000000
.0000000
.0000000
1 innnnn
1 Qnnnnn
6,
Q
K
5
15.
36.
33
13
.0000000
.0000000
.1000000
.1090009
2900900
.1000000
.2000000
.5000000
.5000000
^nnnnnn
nrinnnnn
fi
5
12
12
6
6
3
u u . j i; j
.0009000
.0000000
.1000000
.1990000
290090n
.2000000
.1000000
.1000000
.9999009
.9099000
9099ni9
jo ; j
6.
30.
36.
51.
57.
30.
U.'JUU 3JUUU.UUU JJiiJ'J . UUU 3Ji3UU . 'JUU JO^ ; ,)'J . J'. w j.vJvJv. . *.
HEIGHT -
0000000 B 1Pnnnnn 1R ^nnnnnn 3POr,noo «o 7onr.0or ci nnn^
5000000 18 3090090 24 4009009 32 5999999 "' 9909999 51 00009C
0990009 3.0000000 6.1OC0900 15.2000999 42.7909999 33.5929"'
"000000 9n"iOOn 3 nOOn900 9900099 "* ^n"0""^ '3 '^•^'"•"
0000009 .0000000 .0000000 .0990992 .909099? .292?:''
GRID SYSTEM X AXIS (METERS)
533259.999
AXIS (METERS
5358000
OCT7C1
5357500
5367250
5367090
5356750
5355590
5356250
.990
inn
.090
.900
.000
.909
.900
.009
j
1[;
1 c
18
18
15
12
5
5355000.000
cocci^n
cscccnn
/
(*ETEPS)
nnn
inn
(UP
V
TERS
.2000000
onnnniin
.3000000
.3000000
.2299000
.2000000
.1000000
.0000000
.9009000
9090009
.9099900
533500.000
12
12
12.
12
5
5
.2000000
.2000000
.2000000
.2000009
.1090900
.1000000
.0000000
.0099900
.0000000
.0000000
.9000000
UCTQ(JT
\
533750.000
12
12
12
5
5
18
27
35
36
.2900000
.2000000
.2000000
.1000000
.1909000
.3009000
.4000000
5090900
.5090099
48.8000909
54
*-*
X
.9999000
DISCRETE
534000.000
5.
6.
HEIGHT -
1000000
1000000
12.2000000
18.
33.
48.
48.
48.
51
54.
51.
POINT
V
(METERS) (METERS
3000000
5000000
8000000
8000000
3000009
8009000
9000000
0000000
TERRAIN HEIGHTS (METERS) *-*
HEIGHT X v HEIGHT
) (METERS) (METERS )
532752.9 5367619.0
17.7000000
532332.0 5366404.0 19.8009009
532509 9 536'78''5 9
-------
ONE HOUR CASE 1
OOWNWASH CQR FR < 3, WS=4 MPH, !KO=360, STASIS
SATE
TABLE
SOURCE INPUT DATA
SOURCE INVENTORY
C T SOURCE T SOURCE
A A NUMBER Y STRENGTH
R P P(GRAMS/SEC)
D c c
X
X
X
X
x
V
X
X
X
x
X
X
X
X
x
X
Y
X
Y
X
101 0
201 0
301 0
302 0
303 0
304 0
305 0
305 0
307 0
308 0
309 0
310 0
311 0
•319 n
313 n
311 0
o-ir n
316 0
917 n
318 0
2.680
175.400
8.950
11.720
10.460
3.780
9.070
78.370
54.540
5.290
1.260
17 om
19.910
5.300
5.420
1.390
10 530
19.530
2 140
.530
X
COORDINATE
(METERS)
532722.00
532661.00
531961.00
531945.00
531923.00
531897.00
532029.00
532178.00
532170.00
531833 CO
531845.00
53/>125 "0
532125.00
531932.00
531924.00
531915.00
531875.00
531887.00
531995.00
531898.00
Y HEIGHT IF TYPE=0
COOROINATE ABOVE TEM° (DEG K)
(METERS ) GROUND IF TYPE=!OR2
(METERS) LENGTH SHORT
SIDE (MTRS)
5369522.00
5368539.00
5371117.00
5371117.00
5371117.00
5371117.00
5371120.00
5371115.00
5371132.00
5371030.00
5371030.00
£3711 on on
5371202.00
5370843.00
5370843.00
5370843.00
5370845.00
5370845.00
5370843.00
5370843.00
30.50
52.00
37.00
40.00
46.00
48.00
40.00
54.00
53.00
40 00
40.00
38.00
38.00
52.00
52.00
52.00
52.00
52.00
52.00
34.00
350.000
545.000
501.000
486.000
584.000
523.000
515.000
497.000
526.000
510.000
515.000
456.000
472.000
526.000
481.000
441.000
508.000
513.000
475.000
715.000
IF TYPE^O
RT M**3/SEC
IF TYPE=10R2
LENGTH LONG
SIDE (MTRS)
1 1
77
1 1
10
13
•j
8.
51
44
5
i
13.
15.
7,
4,
i
17,
18.
1
.940
.100
.750
.190
.540
.790
.540
010
.700
.750
.380
.310
.450
.120
,040
,510
.980
,150
570
920
ANGLE
TA
LONG
SIDE
(DEG)
n
.0
n
0
.0
n
.0
,0
.0
.0
.0
.0
.0
.0
n
n
.0
n
n
.0
STACK
INNER
RADIUS
(METERS)
c i n
1.410
nnn
nnn
.000
nnn
.000
.000
1 1/JQ
.000
.000
.000
.870
.000
.000
000
.000
.000
.000
.000
!TI CyATTQkj OSpTTplJI ^TC ^JCTOTC
f,- SCTTLTNG FPFP'1!
crjr^ '/a r.rTTv ~l
PACC CVICTCPC /czr\ fl^mo'
(METERS) (FRACj
30,00
•>0 nn
14 On
1 .1 OP
'• f, 00
• (i nn
14.00
11.00
if nn
'4 00
14.00
14.00
1i 0^
20.00
20.00
2n 00
20.00
20.00
20.00
20.00
-------
ONE HOUR CASE 1 - NO DOWNWASH FOR CR •'. 3, WS=4 MPH, WD=3SC, STAS=8 OATE , CASE "1, °AGE
TASLE 4
- METEOROLOGICAL !NPUT DATA -
'[
UOUR WIND WIND LAYER AMBIENT VERT GRAO STAB WNO SPO ST9 DEV EL ST0 OEV AZ STD OEV 5L 5TD DEV »Z LATEF:A1
OIRECTION SPEED DEPTH TEMP OF POT TMP ILITY POWER LAW ANGLE, SOR A\'2LE, SOR ,'r-LE, 30R AVT.E, SCR DI—'JGIC"
(DEGREES) (MTR/SEC) (METERS) (DEG K) (DEG K/M) CAT. EXPONENT TYPE 0 TYPE 0 TY?E 'C?2 Tvoi ^^2 :CEcc:c:Ev';
THETA UBAR H^1 TA DPDZ ISTBLE P SIGEPU(RAD) SIGAPU(RAO) SIGEPL(RAD) SIGAPL(?AO; V_I:"J.A '
13CC 36C.GOOQ 1.7900 1500.000 295.000 .0000 B .1000 .1080000 1544000 .108000? .154IOCC .90CC
-------
ONE HOUR CASE 1 - NO DOWNWASH FOR FR < 3, WS=4 MPH, WD=360, STAB=8
TABLE 5
DATE
, CASE 1, PAGE
HOUR GROUND LEVEL CONCENTRATION PARTS PER MILLION
1! p Tfl
'
531000.000
Y AXIS (METERS )
FROM ALL SOURCES
GRID SYSTEM X AXIS (DETERS) -
(THE MAXIMUM CONCENTRATION IS .9940959 AT X= 532750.0, Y=5366250.Q)
531250.000 531500.000 531750.000 532000,000 532250.000 532500.000
CONCENTRATION -
532750.000 533000
I
5358000.000
5367750.000
535'750C.OOO
5367250.000
5367000.000
5356750.000
5366500.000
5366250.000
5365750.000
5355500.000
4)(rc ,MCTCRq
5358000.000
5357750.000
5367500.000
5367250.000
5367000.000
5355750.000
5356500.900
5355250.000
5366000.000
5365750.000
5365500,000
.0013136
.0020039
.9027958
.0035466
.0044844
.0053074
.0050200
.0056649
9077127
.0076743
.0080425
533250.000
.0005822
.0010093
not KJ nt
. J J . J -r -j -
.0021575
.0030434
.0045553
.0065769
.0091656
.0118777
.0143759
.0154874
.0080815
.0097181
.01 12334
.9124089
.0134629
.0140345
.0146337
.9151571
.0152617
.0152749
.0149171
(THE MAXIMUM
533500.000
.0000445
.0001046
nnnonpi
. U V J (, W 0 1
,0003525
.0005650
.0008344
.0012066
.0017706
.0025521
.0035602
.0046955
.0297979
.0305209
.0309430
.0305463
.0302431
.0294127
.0283043
.0272852
.0261784
.0251009
.0241302
GRID
CONCENTRATION
.0666656
0621 150
.0582270
.0555364
.0516195
.0486459
.0449189
0410121
.0375913
.0355154
.0338020
SYSTEM X AXIS
.0935208
.0839890
.0756527
.0695585
.0537!!15
.0579663
.0535175
.0505050
.0475557
.0457247
.0442922
(METERS)
IS .0940959 AT X= 532750
0841547 .0490503
0767540 .0432321
0695236 .0534218
0558253 .0652544
0631375 .0353948
0638013 .0880958
0648311 .0910704
0621194 .OS267'7C
0506381 .0756734
nKOfii/ic nT'-'Oct;
jJU^j-iw . w : *T j v w
n v-KicRicn n>
• W , <™wwUl'Cwu 1 nil
. j V.-T u \}-r i . J^
ii7nono^ o <) r
i o rt .-1 ^ *? o n H
• "Ji--w -' •-•
n o 1 7 Q p 7 oK
nq/inosq n,u
, J « -h u J ^ J . a -r •-
noo2g/c n^
n 7 n 1 1, ,} Q «•
.0559093 .0,,
533750.000 534000.000
- CONCENTRATION
.0000023
.0000077
.0000204
.0000450
.0000872
.0001534
.0002465
.0003762
.0005565
.0008231
.0011842
.0000001
.0000004
.0000014
,0000042
.0000106
.0000227
.0000428
.0000735
.0001139
.0001806
.0002593
-------
ONE HOUR CASE 1 - MO DODINWASH FOR FR < 3, 4IS=4 MPH, WD=350, STAS=S DATE , CASE 1, PAGE
TAOI C R /r>ri)\|-M
, HDi-i. - v~^.>. . ,
\ HOUR GROUND LEVEL CONCENTRATION PARTS PER MILLION CROM ALL SOURCES
0 HOUR(S) 0 T0 0
- DISCRETE °0!NT RECEPTORS -
(THE MAXIMUM CONCENTRATION IS, .0715010 AT X= 532332.0, Y-5355i04.0]
9 X Y CONCENTRATION X Y CONCENTRATION X Y CONCENTRATOR
(METERS) (METERS ) (METERS) (METERS ) (METERS) (METERS )
532762.0 5367619.0 .0261222 532332.0 5386404.0 .0715010 532509.0 5367825.0 .0471571
-------
ISCSTU (VERSION 86170)
AN AIR QUALITY DISPERSION MODEL IN
SECTION 2. NON-GUIDELINE MODELS.
IN UNAMAP (VERSION 5) JUNE 86.
SOURCE: UNAMAP FILE ON EPA'S UNIVAC 1110, RTP, NC.
1311021
20 13
J3100E+06
J3655E+07
.22000E+03
.OOOOOE+00
.22000E+03
.10000E+02
.21000E+03
.11000E+03
.20000E+03
.20000E+03
.18000E+03
.20000E+03
.17000E+03
.20000E+03
.12000E+03
.10000E+03
.90000E+02
JOOOOE+02
JOOOOE+02
JOOOOE+02
.20000E+02
JOOOOE+02
.OOOOOE+00
JOOOOE+02
J3276E+06
J3233E+06
J3251E+06
.OOOOOE+00
.38180E+03
00000000
11 3 0
' .25000E+03
.25000E+03
.oooaoE+oo
.OOOOOE+00
.11000E+03
.OOOOOE+00
.11000E+03
.OOOOOE+00
.12000E+03
.OOOOOE+00
JOOOOE+02
.20000E+02
.20000E+02
.40000E+02
JOOOOE+02
JOOOOE+02
.20000E+02
JOOOOE+02
.20000E+02
JOOOOE+02
.OOOOOE+00
JOOOOE+02
.OOOOOE+00
JOOOOE+02
J3676E+07
J3664E+07
J3678E+07
.OOOOOE+00
.OOOOOE+00
1 0 0 2 0 1 1 (
0 1
.OOOOOE+00
.OOOOOE+00
.OOOOOE+00
.OOOOOE+00
.10000E+02
.OOOOOE+00
.20000E+02
.OOOOOE+00
.20000E+02
.OOOOOE+00
.40000E+02
.20000E+02
.40000E+02
.20000E+02
.20000E+02
.40000E+02
.20000E+02
.40000E+02
.OOOOOE+00
.40000E+02
.OOOOOE+00
.40000E+02
J8000E+02
J5000E+02
J8000E+02
.OOOOOE+00
parts
3221121
1
.OOOOOE+00
.18000E+03
.OOOOOE+00
.16000E+03
.OOOOOE+00
.12000E+03
.10000E+03
.12000E+03
.19000E+03
.90000E+02
.20000E+03
JOOOOE+02
.12000E+03
.20000E+02
.10000E+03
.20000E+02
.20000E+02
.40000E+02
.OOOOOE+00
.40000E+02
.OOOOOE+00
.40000E+02
.OOOOOE+00
per million
1100000
.OOOOOE+00
.20000E+03
.OOOOOE+00
.18000E+03
.10000E+02
.17000E+03
JOOOOE+02
.16000E+03
JOOOOE+02
.16000E+03
JOOOOE+02
.16000E+03
.10000E+03
.11000E+03
.70000E+02
JOOOOE+02
.20000E+02
.40000E+02
.20000E+02
.20000E+02
.20000E+02
.20000E+02
.OOOOOE+00
0 0
0000
.OOOOOE+00
JOOOOE+02
.20000E+02
JOOOOE+02
JOOOOE+02
JOOOOE+02
.70000E+02
JOOOOE+02
.40000E+02
JOOOOE+02
JOOOOE+02
.OOOOOE+00
CONVERTED TO IBM PC
.OOOOOE+00
.OOOOOE+00
JOOOOE+02
.11000E+03
JOOOOE+02
JOOOOE+02
JOOOOE+02
JOOOOE+02
JOOOOE+02
JOOOOE+02
JOOOOE+02
.OOOOOE+00
JOOOOE+02
J4000E+03
.20000E+03
J4000E+03
JOOOOE+02
JOOOOE+02
JOOOOE+02
JOOOOE+02
JOOOOE+02
JOOOOE+02
-------
*** ONE HOUR CASE 1 - NO DOWNWASH FOR FR < 3 May 22, 1985
***
CALCULATE (CONCENTRATIONS,OEPOSITION=2)
RECEPTOR GRID SYSTEM (RECTANGULAR=1 OR 3, POLAR=2 OR 4)
DISCRETE RECEPTOR SYSTEM (RECTANGULAR=1,POLAR=2)
TERRAIN ELEVATIONS ARE READ (YES=1,NO=0)
CALCULATIONS ARE WRITTEN TO TAPE (YES=1,NO=0)
LIST ALL INPUT DATA (NO=O.YES=1,MET DATA ALSO=2)
COMPUTE AVERAGE CONCENTRATION (OR TOTAL DEPOSITION)
WITH THE FOLLOWING TIME PERIODS:
HOURLY (YES=1,NO=0)
2-HOUR (YES=1,NO=0)
3-HOUR (YES=1,NO=0)
4-HOUR (YES=1,NO=0)
6-HOUR (YES=1,NO=0)
8-HOUR (YES=1,MO=0)
12-HOUR (YES=1,NO=0)
24-HOUR (YES=1,NO=0)
PRINT 'N'-DAY TABLE(S) (YES=1,NO=0)
PRINT THE FOLLOWING TYPES OF TABLES WHOSE TIME PERIODS ARE
SPECIFIED BY ISW(7) THROUGH ISW(14):
DAILY TABLES (YES=1,NO=0)
HIGHEST & SECOND HIGHEST TABLES (YES=1,NO=0)
MAXIMUM 50 TABLES (YES=1,NO=0)
METEOROLOGICAL DATA INPUT METHOD (PRE-PROCESSED=1,CARO=2)
RURAL-URBAN OPTION (RU.=0,UR. MODE 1=1,UR. MODE 2=2,UR. MODE 3=3)
WIND PROFILE EXPONENT VALUES (DEFAULTS=1,USER ENTERS=2,3)
VERTICAL POT. TEMP. GRADIENT VALUES (DEFAULTS=1,USER ENTERS=2,3)
SCALE EMISSION RATES FOR ALL SOURCES "(NO=0,YES>0)
PROGRAM CALCULATES FINAL PLUME RISE ONLY (YES=1,NO=2)
PROGRAM ADJUSTS ALL STACK HEIGHTS FOR DOWNWASH (YES=2,NO=1)
PROGRAM USES BUOYANCY INDUCED DISPERSION (YES=1,NO=2)
CONCENTRATIONS DURING CALM PERIODS SET - 0 (YES=1,NO=2)
REG. DEFAULT OPTION CHOSEN (YES=1,NO=2)
TYPE OF POLLUTANT TO BE MODELLED (1=S02,2=OTHER)
DEBUG OPTION CHOSEN (1=YES,2=NO)
NUMBER OF INPUT SOURCES
NUMBER OF SOURCE GROUPS (=0,ALL SOURCES)
TIME PERIOD INTERVAL TO BE PRINTED (=0,ALL INTERVALS)
NUMBER OF X (RANGE) GRID VALUES
NUMBER OF Y (THETA) GRID VALUES
NUMBER OF DISCRETE RECEPTORS
NUMBER OF HOURS PER DAY IN METEOROLOGICAL DATA
NUMBER OF DAYS OF METEOROLOGICAL DATA
SOURCE EMISSION RATE UNITS CONVERSION FACTOR
HEIGHT ABOVE GROUND AT WHICH WIND SPEED WAS MEASURED
LOGICAL UNIT NUMBER OF METEOROLOGICAL DATA
ALLOCATED DATA STORAGE
REQUIRED DATA STORAGE FOR THIS PROBLEM RUN
ISW(2) -
ISW(3)
ISW(4) -
ISW(5)
ISW(6) =
ISW(7) =
ISW(8)
ISW(9)
ISW(10)
ISW(12) =
ISW(13) -
ISW(14) =
ISW(15) -
ISW(16) =
ISW(17) =
ISW(18) =
ISW(19) =
ISW(20)
ISW(21) -
ISW(22) -
ISW(23)
ISW(24)
ISW(25)
ISW(26)
ISW(27)
ISW(28)
ISW(29)
ISW(30)
NSOURC = 20
NGROUP 0
IPERD = 0
NXPNTS 13
NYPNTS = 11
NXWYPT 3
NHOURS 1
NDAYS 1
TK=.38180E+03
ZR - 10.00 METERS
IMET = 5
LIMIT - 43500 WORDS
MIMIT - 4914 WORDS
-------
*** ONE HOUR CASE 1 - NO DOWNWASH FOR FR < 3 May 22, 1985 ***
*** UPPER BOUND OF FIRST THROUGH FIFTH WIND SPEED CATEGORIES ***
(METERS/SEC)
1.54, 3.09, 5.14, 8.23, 10.80,
*** X-COORDINATES OF RECTANGULAR GRID SYSTEM ***
(METERS)
531000.0, 531250.0, 531500.0, 531750.0, 532000.0, 532250.0, 532500.0, 532750.0, 533000.0, 533250.0,
533500.0, 533750.0, 534000.0,
*** Y-COORDINATES OF RECTANGULAR GRID SYSTEM ***
(METERS)
5365500.0, 5365750.0, 5366000.0, 5366250.0, 5366500.0, 5366750.0, 5367000.0, 5367250.0, 5367500.0, 5367750.0,
5368000.0,
*** X,Y COORDINATES OF DISCRETE RECEPTORS ***
(METERS)
( 532762.0,5367619.0), ( 532332.0,5366404.0), ( 532509.0,5367825.0), (
-------
*** ONE HOUR CASE 1 - NO DOWNWASH FOR FR < 3 May 22, 1985
***
* ELEVATION HEIGHTS IN METERS *
* FOR THE RECEPTOR GRID *
Y-AXIS /
(METERS) /
5368000.0 /
5367750.0 /
5367500.0 /
5367250.0 /
5367000.0 /
5366750.0 /
5366500.0 /
5366250.0 /
5366000.0 /
5365750.0 /
5365500.0 /
531000.0
.00000
6.09601
15.24003
27.43205
36.57607
51.81610
54.86411
60.96012
64.00813
67.05613
57.05613
531250.0
.00000
.00000
5.09601
6.09601
15.24003
6.09601
15.24003
36.57607
33.52806
33.52806
.00000
531500.0
.00000
.00000
6.09601
5.09601
12.19202
12.19202
6.09601
6.09601
3.04801
.00000
.00000
X-AXIS (METERS)
531750.0 532000.0
.00000
.00000
6.09601
30.48006
36.57607
60.96012
57.91211
30.48006
.00000
.00000
.00000
6.09601
6.09601
6.09601
21.33604
30.48006
24.38405
18.28804
18.28804
3.04801
.00000
.00000
532250.0
18.28804
18.28804
12.19202
24.38405
21.33604
18.28804
18.28804
24.38405
6.09601
3.04801
.00000
532500.0
18.28804
18.28804
18.28804
18.28804
18.28804
18.28804
18.28804
33.52806
15.24003
.00000
.00000
532750.0
18.28804
18.28804
18.28804
18.28804
18.28804
24.38405
42.67208
60.96012
42.67208
3.04801
.00000
533000.0
18.28804
18.28804
18.28804
18.28804
30.48005
60.96012
60.96012
60.96012
33.52805
3.04801
.00000
-------
*** ONE HOUR CASE
DOWNIKASH FOR FR < 3 May 22, 1985
***
* ELEVATION HEIGHTS IN METERS *
* FOR THE RECEPTOR GRID *
Y-AXIS /
(METERS) /
5368000.0 /
5367750.0 /
5367500.0 /
5367250.0 /
5367000.0 /
5366750.0 /
5366500.0 /
5366250.0 /
5366000.0 /
5365750.0 /
5365500.0 /
533250.0
15.24003
15.24003
18.28804
18.28804
15.24003
12.19202
6.09601
.00000
.00000
.00000
.00000
533500.0
12.19202
12.19202
12.19202
12.19202
6.09601
6.09601
.00000
.00000
.00000
.00000
.00000
533750.0
12.19202
12.19202
12.19202
6.09601
6.09601
18.28804
27.43205
36.57607
36.57607
48.76810
54.86411
X-AXIS (METERS)
534000.0
6.09601
6.09601
12.19202
18.28804
33.52806
48.76810
48.76810
48.76810
51.81610
54.86411
60.96012
-------
*** ONE HOUR CASE 1 - NO DOWNWASH FOR FR < 3 May 22, 1985 ***
* ELEVATION HEIGHTS IN METERS *
* FOR THE DISCRETE RECEPTOR POINTS *
X - - Y - H6T. - X - Y - HGT. - X - Y - HGT.
532752.0 5367619.0 17.67.843 532332.0 5366404.0 19.81204 532509.0 5367825.0 17.67843
-------
*** ONE HOUR CASE 1 - NO DOWNWASH FOR FR < 3 May 22, 1985
***
*** SOURCE DATA ***
SOURCE
NUMBER
101
201
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
T W
Y A NUMBER
P K PART.
E E CATS.
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
EMISSION RATE
TYPE=0,1
(GRAMS/SEC)
TYPE=2 BASE
(GRAMS/SEC) X Y ELEV.
*PER METER**2 (METERS) (METERS) (METERS)
.26800E+01
.17540E+03
.89500E+01
.11720E+02
.10460E+02
.37800E+01
.90700E+01
.78370E+02
.64640E+02
.52900E+01
.12600E+01
.17010E+02
.19910E+02
.58000E+01
.54200E+01
.18900E+01
.19530E+02
.19530E+02
.21400E+01
.63000E+00
532722.0
532661.0
531961.0
531945.0
531923.0
531897.0
532029.0
532178.0
532170.0
531833.0
531845.0
532125.0
532125.0
531932.0
531924.0
531915.0
531875.0
531887.0
531906.0
531898.0
5369522.0
5368539.0
5371117.0
5371117.0
5371117.0
5371117.0
5371120.0
5371115.0
5371132.0
5371030.0
5371030.0
5371190.0
5371202.0
5370843.0
5370843.0
5370843.0
5370845.0
5370845.0
5370843.0
5370843.0
30.0
30.0
14.0
14.0
14.0
14.0
14.0
14.0
14.0
14.0
14.0
14.0
14.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
HEIGHT
(METERS)
30.50
52.00
37.00
40.00
46.00
46.00
40.00
54.00
53.00
40.00
40.00
38.00
38.00
52.00
52.00
52.00
52.00
52.00
52.00
34.00
TEMP.
TYPE=0
(DEG.K)j
VERT. DIM
TYPE=1
(METERS)
350.00
545.00
601.00
486.00
584.00
523.00
515.00
497.00
526.00
610.00
615.00
466.00
472.00
626.00
481.00
441.00
508.00
513.00
475.00
715.00
EXIT VEL.
TYPE=0
(M/SEC); SLOG. BLDG. 8LDG.
HORZ.DIM DIAMETER HEIGHT LENGTH WIDTH
TYPE=1,2 TYPE=0 TYPE=0 TYPE=0 TYPE=0
(METERS) (METERS) (METERS) (METERS) (METERS)
10.21
12.43
4.89
4.23
4.40
2.33
3.85
7.78
11.05
3.90
2.17
5.66
5.57
3.92
1.82
.58
3.05
3.08
.87
.52
1.22
2.81
1.75
1.75
1.98
1.44
1.68
2.89
2.27
1.37
.90
1.73
1.73
1.52
1.68
1.68
2.74
2.74
1.52
1.37
.00 .00 .00
.00 .00 .00
.00 .00 .00
.00 .00 .00
.00 .00 .00
.00 .00 .00
.00 .00 .00
.00 .00 .00
.00 .00 .00
.00 .00 .00
.00 .00 .00
.00 .00 .00
.00 .00 .00
.00 .00 .00
.00 .00 .00
.00 .00 .00
.00 .00 .00
.00 .00 .00
.00 .00 .00
.00 .00 .00
-------
MET. DA1K
DAY 1
«** ONE HOUR CASE 1 - NO OOWNWASH FOR FR < 3 May 22, 1985 ***
* METEOROLOGICAL DATA FOR DAY 1 *
POT. TEMP.
FLOW WIND MIXING GRADIENT WIND DECAY
VECTOR SPEED HEIGHT TEMP. (DEG. K STABILITY PROFILE COEFFICIENT
HOUR (DEGREES) (MRS) (METERS) (OEG. K) PER METER) CATEGORY EXPONENT (PER SEC)
180.0 1.79 1500.0 295.0 .0000 2 .0700 .OOOOOOE+00
-------
*** ONE HOUR CASE 1 - NO DOWNWASH FOR FR < 3 May 22, 1985
***
DAILY
1-H
SGRO
* DAILY 1-HOUR AVERAGE CONCENTRATION parts per million *
* ENDING WITH HOUR 1 FOR DAY 1 *
* FROM ALL SOURCES *
* FOR THE RECEPTOR GRID *
* MAXIMUM VALUE EQUALS .08727 AND OCCURRED AT ( 532000.0, 5368000.0) *
Y-AXIS /
(METERS) /
5368000.0 /
5367750.0 /
5367500.0 /
5367250.0 /
5367000.0 /
5366750.0 /
5366500.0 /
5366250.0 /
5366000.0 /
5365750.0 /
5365500.0 /
531000.0
.00455
.00579
.00696
.00800
.00885
.00957
.01006
.01042
.01065
.01078
.01083
531250.0
.01597
.01728
.01825
.01874
.01904
.01887
.01874
.01856
.01808
.01759
.01689
531500.0
.03946
.03825
.03694
.03518
.03351
.03166
.02978
.02815
.02665
.02536
.02431
531750.0
.06906
.06272
.05739
.05331
.04876
.04525
.04174
.03844
.03565
.03404
.03275
X-AXIS (METERS)
532000.0
.08727
.07734
.06891
.06270
.05743
.05303
.04994
.04802
.04576
.04429
.04297
532250.0
.08087
.07279
.06588
.06318
.06165
.06137
.06158
.06193
.05874
.05678
.05441
532500.0
.05650
.05482
.05688
.06218
.06808
.07221
.07378
.07618
.07018
.06493
.06161
532750.0
.03145
.03325
.03960
.04877
.05759
.06558
.07318
.07630
.06948
.06038
.05717
53300
.01
Oil
•
o
.0?
.02
o
.Ol
.04
.01
o|
.04
-------
DAILY:
1-HR/PO
SGROUPS
*** ONE HOUR CASE 1 - NO DOWNWASH FOR FR < 3 May 22, 1985
***
* DAILY 1-HOUR AVERAGE CONCENTRATION parts per million
* ENDING WITH HOUR 1 FOR DAY 1 *
* FROM ALL SOURCES *
* FOR THE RECEPTOR GRID *
MAXIMUM VALUE EQUALS
.08727 AND OCCURRED AT ( 532000.0, 5368000.0) *
Y-AXIS /
(METERS) /
5368000.0 /
5367750.0 /
5367500.0 /
5367250.0 /
5367000.0 /
5366750.0 /
5366500.0 /
5366250.0 /
5366000.0 /
5365750.0 /
5365500.0 /
533250.0
.00262
.00358
.00461
.00591
.00788
.01060
.01356
.01640
.01910
.02123
.02275
_533500.0
.00045
.00077
.00116
.00162
.00217
.00291
.00391
.00524
.00677
.00834
.00984
533750.0
.00006
.00013
.00024
.00039
.00058
.00083
.00116
.00161
.00219
.00295
.00380
X-AXIS (METERS)
534000.0
.00001
.00002
.00004
.00008
.00014
.00022
.00033
.00047
.00065
.00090
.00122
-------
JAIL',
11
S6R(
*** ONE HOUR CASE 1 - NO OOWNWASH FOR FR < 3 May 22, 1985 ***
* DAILY 1-HOUR AVERAGE CONCENTRATION parts per million *
* ENDING WITH HOUR 1 FOR DAY 1 *
* FROM ALL SOURCES *
* FOR THE DISCRETE RECEPTOR POINTS *
- X - - Y CON. - X - - Y - CON. - X - Y - CON.
532762.0 5367619.0 .03478 532332.0 5366404.0 .06653 532509.0 5367825.0 .05446
-------
REPORT DOCUMENTATION
PAGE
1. REPORT NO.
EPA-910/9-86-147
3. Recipient's Accession Na.
4. Title and Subtitle
COMPARISON OF AIR QUALITY MODEL
ESTIMATES WITH MEASURED S02 CONCENTRATIONS
NEAR MARCH POINT. WASHINGTON
5. Report Oat*
December 1986
7. Authors)
Kirk D. Winges
8. Performing Organization Rept. No.
3710-Q81
9. Performing Organisation Name and Addresa
TRC Environmental Consultants, Inc
15924 22nd Avenue SE
Mill Creek, Washington 98012
10. Praject/Taak/Work Unit No.
11. Contract(C) or Grant(G) No.
(0 68-02-3886
(G)
12. Sponsoring Organization Name and Addresa
U. S. Environmental Protection Agency
Region 10
1200 Sixth Avenue
Seattle, Washington 98101
13. Type of Report & Period Covered
Final
14.
IS. Supplementary Notes
It. Abstract (limit 200 words)
This report documents an air quality modeling study of sulfur
dioxide concentrations near March Point, Washington. Previous
Modeling conducted by the EPA was used to site air quality monitors
in the vicinity of March Point. The current study evaluated the
measured data from these air quality monitors with predictions using
the SHORTZ and ISCST air quality models. A series of 20 different
test periods were used in the model evaluations. Neither model
preformed well in a comparison of measured and predicted values when
the data are paired in space and time. However, model prediction
improved for both models when comparison was performed with the data
paired in time, but not in space. The main conclusions were that
air monitoring is not necessary, given the low level of impacts, and
that neither model offers significant advantages unless terrain
heights are higher than the stack heights, in which case the SHORTZ
Model is preferred.
17. Document Analysis a. Descriptors
Air Pollution, Meteorology, Turbulent Diffusion
b. Identifiers/Open-ended Terms
Dispersion Modeling, ISC, SHORTZ.
c. COSATI Held/Group
«. Availability Statement
Release unlimited
19. Security Class (This Report)
Unclassified
20. Security Class (This Page)
Unclassified
21. No. of Pages
97
22. Price
(SeeANSI-Z39.18)
See Instructions on Reverse
OPTIONAL FORM 272 (4-77)
(Formerly NTIS-3S)
Department of Commerce
------- |