-------
turbulence measurements to assign the Pasquill stability category to each
hour. Consequently, we recommended to EPA Region III that the model
performance evaluation be restricted to the second year of the Westvaco
monitoring program.
A major deficiency of the Westvaco data set is the absence of
valid onsite mixing depths. Minisondes were released from Luke Hill at
approximately 0700 and 1330 EST on 21 of the 29 days during February 1980,
the third month of the first quarter of the Westvaco monitoring program.
We used the minisonde vertical temperature profiles to estimate mixing
depths for comparison with the concurrent Luke Hill acoustic sounder mixing
depths and with mixing depths estimated from Greater Pittsburgh Airport
0700 and 1900 EST rawinsonde releases. The minisonde and Airport soundings
generally revealed the same major synoptic-scale features (for example, a
well-defined subsidence inversion). Also, the lower portions of the mini-
sonde temperature profiles closely resembled the temperature profiles con-
structed from the tower measurements, although the two temperature profiles
occasionally showed systematic differences in temperature that are probably
attributable to minisonde calibration errors. However, we found no corre-
lation between the onsite tower and minisonde mixing depths and the acoustic
sounder mixing depths. We therefore concluded that the acoustic sounder
mixing depths are invalid, a conclusion also reached by Hanna, et_ j_l_. (1982a),
In our opinion, the most likely explanation for the invalid acoustic sounder
mixing depths is calibration problems. For example, an independent quality
assurance audit by Radian Corporation dated 29 January 1982 found that the
transmitter's start pulse and operating pulse widths failed to meet manufac-
turer's specifications.
In the fall of 1977, the H. E. Cramer Company applied the SHORTZ
model to the Westvaco Luke Mill using hypothetical meteorological inputs
and concluded that, with the current SO- emission limitation, emissions
from the Main Stack might cause the short-term NAAQS to be slightly exceeded
in the vicinity of Monitors 8 and 9. Although the SHORTZ model predicted
31
-------
higher concentrations on elevated terrain closer to the Main Stack, the
reliability of these calculated concentrations was not known because they
occurred at distances less than the distance at which the buoyant plume
typically could be expected to stabilize. When standard "block averages"
are used to analyze the hourly S0« concentration measurements from the
entire 2-year Westvaco monitoring program, Table 3-10 shows 13 observed
3-hour average concentrations above the 3-hour NAAQS and Table 3-11 shows
11 observed 24-hour average concentrations above the 24-hour NAAQS. All of
these "block average" 3-hour and 24-hour average concentrations occurred on
elevated terrain at distances less than the distances to Monitors 8 and 9.
If non-overlapping running mean 3-hour and 24-hour average concentrations
are considered for the entire 2-year monitoring program, Tables 3-12 and
3-13 shows that 3-hour and 24-hour average concentrations slightly above
the 3-hour and 24-hour NAAQS were measured at Monitors 8 and 9.
3.2 RESULTS OF THE MODEL PERFORMANCE EVALUATION
3.2.1 Results of the Valley Model Performance Evaluation
We used the Valley model, as described in Section 2.2 and Appendix
F, to calculate 24-hour average SO- concentrations for the nine monitoring
sites considered in the dispersion model performance evaluation (see Appendix
A for a discussion of the selection of these sites). The emissions data
used in the Valley model calculations were for the calendar days during
each year of the 2-year Westvaco monitor program with the highest and second-
highest observed 24-hour average concentrations at the various monitoring
sites. The meteorological conditions assumed in the Valley model calcula-
tions were the conditions recommended by Burt and Slater (1977) for screening
analyses (F stability and a mean wind speed of 2.5 meters per second).
Table 3-14 compares the 24-hour average concentrations calculated for each
monitoring site with the observed highest and second-highest 24-hour average
concentrations. (The observed 24-hour average concentrations in Table 3-14
32
-------
TABLE 3-10
BLOCK AVERAGE 3-HOUR AVERAGE SO CONCENTRATIONS ABOVE THE
3-HOUR NATIONAL AMBIENT AIR QUALITY STANDARD
Date
Hours
(EST)
Monitor
Concentration
(yg/m )
(a) First Year of the Westvaco Monitoring Program
27 Mar 1980
31 Jul 1980
13 Nov 1980
0400-0600
0700-0900
0700-0900
1000-1200
5
5
4
6
1,472
1,415
1,402
1,813
(b) Second Year of the Westvaco Monitoring Program
13 Jan 1981
8 Apr 1981
22 Oct 1981
13 Nov 1981
19 Nov 1981
0400-0600
0400-0600
0700-0900
0700-0900
0100-0300
0400-0600
0700-0900
6
6
6
1
7
5
7
1
1,389
1,386
1,617
1,425
1,729
1,640
1,955
1,376
33
-------
TABLE 3-11
BLOCK AVERAGE 24-HOUR AVERAGE SO CONCENTRATIONS ABOVE THE
24-HOUR NATIONAL AMBIENT AIR QUALITY STANDARD
Date
Monitor
3
Concentration (yg/ra )
(a) First Year of the Westvaco Monitoring Program
27 Mar 1980
21 Nov 1980
5
6
435
448
(b) Second Year of the Westvaco Monitoring Program
5 Dec 1980
29 Dec 1980
6 Jan 1981
13 Jan 1981
8 Apr 1981
13 Nov 1981
6
1
3
3
4
6
6
5
383
388
427
377
406
409
401
417
34
-------
TABLE 3-12
NON-OVERLAPPING RUNNING MEAN 3-HOUR AVERAGE SO CONCENTRATIONS
ABOVE THE 3-HOUR NATIONAL AMBIENT AIR QUALITY STANDARD
Date
Hours
(EST)
Monitor
Concentration
(ug/m )
(a) First Year of the Westvaco Monitoring Program
27 Mar 1980
31 Jul 1980
3-4 Sep 1980
21 Nov 1980
13 Nov 1980
14 Nov 1980
0400-0600
0600-0800
0700-0900
0700-0900
2300-0100
0500-0700
1000-1200
0300-0500
5
8
9
5
4
5
1
6
6
1,472
1,622
1,310
1,417
1,402
1,336
1,386
1,813
1,386
(b) Second Year of the Westvaco Monitoring Program
13 Jan 1981
8 Apr 1981
22 Oct 1981
13 Nov 1981
19 Nov 1981
0200-0400
0400-0600
0700-0900
0700-0900
0100-0300
0400-0600
0700-0900
0900-1100
1
3
4
6
6
6
1
7
5
7
8
1
6
1,800
1,504
1,517
1,339
1,389
1,617
1,425
1,729
1,640
1,955
1,368
1,376
1,527
35
-------
TABLE 3-13
NON-OVERLAPPING RUNNING MEAN 24-HOUR AVERAGE SO CONCENTRATIONS
ABOVE THE 24-HOUR NATIONAL AMBIENT AIR QUALITY STANDARD
Date
Hours
(EST)
Monitor
Concentration
(ug/m )
(a) First Year of Westvaco Monitoring Program
26-27 Mar 1980
13-14 Nov 1980
20-21 Nov 1980
1000-0900
0800-0700
2200-2100
5
6
1
6
472
537
380
500
(b) Second Year of Westvaco Monitoring Program
5- 6 Dec 1980
13-14 Dec 1980
28-29 Dec 1980
6 Jan 1981
12-13 Jan 1981
16-17 Jan 1981
7- 8 Apr 1981
30 Apr-1 May 1981
8- 9 Nov 1981
12-13 Nov 1981
0900-0800
2100-2000
1900-1800
0100-2400
1500-1400
2400-2300
2300-2200
1800-1700
1900-1800
1300-1200
1200-1100*
2000-1900
6
6
1
3
3
4
6
1
6
6
9
1
7
4
5
6
8
445
375
419
427
367
419
417
474
372
403
390
380
692
369
424
411
385
* Monitor 7 concentration measurements missing after 1100 EST on 13
November 1981.
36
-------
TABLE 3-14
COMPARISON OF THE HIGHEST AND SECOND-HIGHEST OBSERVED 24-HOUR AVERAGE S0r
CONCENTRATIONS WITH THE CONCENTRATIONS CALCULATED BY THE VALLEY MODEL "
Monitor
Date
Concentration (ppm)
Observed
Calculated
Ratio of Calculated and
Observed Concentrations
(a) First-Year (December 1979 through November 1980)
1
3
4
5
6
7
8
9
10
31 Jul 80
21 Nov 80
30 Sep 80
14 Jun 80
31 Jul 80
25 Jun 80
27 Mar 80
23 Nov 80
21 Nov 80
24 Jun 80
23 Nov 80
20 Nov 80
27 Mar 80
23 Nov 80
27 Mar 80
23 Nov 80
28 Jan 80
2 Jan 80
0.111
0.103
0.079
0.073
0.104
0.087
0.166
0.086
0.171
0.124
0.127
0.081
0.101
0.097
0.103
0.084
0.077
0.052
0.762
1.280
1.171
1.153
0.695
0.824
0.676
0.682
1.379
1.121
0.782
0.947
0.485
0.491
0.436
0.442
0.000
0.000
6.86
12.43
14.82
15.79
6.68
9.47
4.07
7.93
8.06
9.04
6.16
11.69
4.80
5.06
4.23
5.26
0.00
0.00
(b) Second-Year (December 1980 through November 1981)
1
3
4
5
6
7
8
9
10
13 Jan 81
29 Dec 80
6 Jan 81
13 Jan 81
13 Jan 81
6 Jan 81
13 Nov 81
29 Dec 80
13 Jan 81
8 Apr 81
29 Dec 80
22 Apr 81
13 Nov 81
9 Jan 81
1 May 81
13 Nov 81
9 Jan 81
3 Jan 81
0.180
0.148
0.163
0.144
0.155
0.119
0.159
0.112
0.156
0.153
0.110
0.101
0.130
0.091
0.112
0.111
0.091
0.048
1.306
1.172
1.183
1.247
1.193
1.146
0.516
0.744
1.407
1.338
0.853
0.772
0.369
0.577
0.446
0.331
0.000
0.000
7.26
7.92
7.26
8.66
7.70
9.63
3.25
6.64
9.02
8.75
7.75
7.64
2.84
6.34
3.98
2.98
0.00
0.00
37
-------
have not been adjusted for the effects of "background," which we define in
this report as ambient SO. concentrations attributable to emissions from
sources other than the Westvaco Main Stack.) The ratio of calculated to
observed concentrations ranges from 2.98 to 15.79 for all monitors except
Monitor 10. Under the assumed "worst-case" meteorological conditions, the
Westvaco plume does not mix far enough downward in the Valley model calcula-
tions to cause a non-zero concentration at Monitor 10. In a regulatory
application of the Valley model, all elevated terrain at and beyond the
boundaries of the Westvaco property would be considered in the model analysis.
Monitors 8 and 9 are on elevated terrain near the southern boundary of the
Westvaco property, and the bias toward overestimation at these monitors
tends to support the continued use of the Valley model as a safe-sided
screening model.
3.2.2 Results of the Complex I and II Model Performance Evaluation
As discussed in Section 2 and Appendix F, the Valley and Complex
I models are based on very similar assumptions. If the wind is contained
within a 22.5-degree sector for 6 hours of a 24-hour period and the meteor-
ological conditions during these 6 hours consist of F stability and an
average wind speed of 2.5 meters per second, the 24-hour average concen-
trations calculated by the two models at receptors in the downwind 22.5-
degree sector are equivalent. Because of the conservativeness of the
Valley model for the Westvaco data set, it follows that the Complex I model
should also be a safe-sided screening model for the Westvaco data set. The
Complex II model predicts higher hourly concentrations than the Complex I
model because the crosswind concentration distribution is assumed to be
Gaussian (as described by Pasquill-Gifford lateral dispersion coefficients)
rather than uniform within a 22.5-degree sector. Consequently, the Complex
II model should also be a safe-sided screening model for the Westvaco data
set. Of the five complex terrain dispersion models described in Section 2,
only the generalized SHORTZ model and the site-specific LUMM model were
38
-------
considered to be likely candidates as refined (non-screening) models for
the Westvaco data set. It was therefore the joint decision of the H. E.
Cramer Company and the EPA Project Officer that the detailed model perfor-
mance evaluation should be restricted to the SHORTZ and LUMM models.
3.2.3 Results of the SHORTZ and LUMM Model Performance Evaluation
On 21 October 1982, Westvaco Corporation, the State of Maryland,
EPA Region III and EPA OAQPS agreed to a protocol for the selection of the
SHORTZ model or the LUMM model as the complex terrain dispersion model to
be used to establish an S09 emission limitation for the Main Stack at the
Westvaco Luke Mill. A copy of this model performance evaluation is con-
tained in Appendix A. As discussed in the 21 October 1982 protocol, the
model evaluation was restricted to the second year of the Westvaco moni-
toring program because of the large fraction of missing turbulence measure-
ments (key meteorological inputs to both the SHORTZ and LUMM models) during
the first year of the monitoring program. Westvaco, the State of Maryland
and EPA also agreed on 21 October 1982 that the performance evaluation for
the SHORTZ and LUMM models would be conducted by ERT and reviewed by the
H. E. Cramer Company. The results of the model performance evaluation have
been summarized by Hanna, et al (1982b) and verified by Bowers and Hargraves
(1982). The remainder of this subsection is primarily based on the report
by Bowers and Hargraves (1982).
Observed (Minus Background) S0? Concentrations
VJe used the observed hourly SO concentrations provided to EPA
Region III by ERT for the second year of the Westvaco monitoring program to
determine, for each monitor of concern for the model performance evaluation
(Monitors 1, 3, 4, 5, 6, 7, 8, 9 and 10), the 25 highest 1-hour, 3-hour and
24-hour average and the annual average observed (minus background) S0_
39
-------
concentrations. The procedures used to adjust the hourly concentrations
for background (defined as ambient S09 concentrations attributable to
emissions from sources other than the Westvaco Main Stack) and to account
for hours with missing concentration measurements are outlined in the 21
October 1982 model evaluation protocol contained in Appendix A. The results
of our analysis of the observed hourly SCL concentrations are presented
in Appendix B.
Table 3-15 compares the highest and second-highest observed (minus
background) 3-hour and 24-hour average SCL concentrations calculated at
each monitor of concern by ERT and the H. E. Cramer Company. To facilitate
comparison of the two sets of observed (minus background) S0_ concentra-
tions, we divided the concentrations reported by Hanna, et^ al_. (1982b) in
units of micrograms per cubic meter by 2,620 to obtain concentrations in
units of parts per million (ppm), the units used to archive the observed
hourly SO concentrations. The two sets of 3-hour average concentrations
in Table 3-15 agree identically and the 24-hour average concentrations
agree to within about 1 percent. However, not all of the observed (minus
background) concentrations calculated by ERT and the H. E. Cramer Company
show such close agreement. For example, Table 3-16 shows differences of up
to 8 percent in some of the 25 highest 24-hour average concentrations. We
manually verified our observed (minus background) 24-hour average concentra-
tions in Table 3-16 using the observed hourly SO^ concentrations provided
to EPA Region III by ERT. We subsequently learned that the differences in
the observed (minus background) concentrations independently calculated by
ERT and the H. E. Cramer Company are attributable to slightly different
interpretations of the background estimation procedures outlined in the
21 October 1982 model evaluation protocol.
In our determination of the observed (minus background) hourly
S0~ concentrations for the Westvaco model evaluation study, we defined
the background during each hour as the lowest observed hourly SO
40
-------
Table 3-15
COMPARISON OF THE HIGHEST AND SECOND-HIGHEST OBSERVED (MINUS BACKGROUND)
3-HOUR AND 24-HOUR AVERAGE S02 CONCENTRATIONS CALCULATED BY
ERT AND THE H. E. CRAMER COMPANY
Monitor
3 -Hour
Concentrations
(ppm)
ERT
HEC
24-Hour
Concentrations
(ppm)
ERT
HEC
Ratio of ERT and
HEC Concentrations
3-Hour
Concentrations
24-Hour
Concentrations
(a) Highest Concentrations
1
3
4
5
6
7
8
9
10
0.5412
0.4337
0.4560
0.6235
0.6142
0.7432
0.5195
0.4728
0.1592
0.5412
0.4337
0.4560
0.6235
0.6142
0.7432
0.5195
0.4728
0.1592
0.1666
0.1585
0.1411
0.1540
0.1487
0.0991
0.1247
0.1088
0.0433
0.1666
0.1585
0.1419
0.1546
0.1501
0.0994
0.1255
0.1099
0.0433
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.994
0.996
0.991
0.997
0.994
0.990
1.000
(b) Second-Highest Concentrations
1
3
4
5
6
7
8
9
10
0.5158
0.3842
0.3733
0.4518
0.5268
0.6575
0.3325
0.4008
0.1120
0.5158
0.3842
0.3733
0.4518
0.5268
0.6575
0.3325
0.4008
0.1120
0.1374
0.1309
0.1142
0.1011
0.1432
0.0946
0.0808
0.1061
0.0373
0.1375
0.1309
0.1147
0.1015
0.1432
0.0947
0.0820
0.1070
0.0373
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.999
1.000
0.996
0.996
1.000
0.999
0.985
0.992
1.000
41
-------
TABLE 3-16
COMPARISON OF SELECTED OBSERVED (MINUS BACKGROUND) 24-HOUR AVERAGE
CONCENTRATIONS CALCULATED BY ERT AND THE H. E. CRAMER COMPANY
AT MONITOR 10
Date
5 Jul 81
6 Oct 81
24-Hour Concentrations (ppm)
ERT
0.0190
0.0160
HEC
0.0206
0.0173
Ratio of
ERT and HEC
Concentrations
0.922
0.925
42
-------
concentration if this concentration was above the monitor threshold concen-
tration of 0.005 parts per million (13 micrograms per cubic meter). If the
lowest observed concentration was equal to 0.005 parts per million (ppm),
we arbitrarily defined the background concentration as 0.0025 ppm (6.5
micrograms per cubic meter). We then subtracted the background concentra-
tion estimated for each hour from all of the observed concentrations.
Thus, our minimum observed (minus background) hourly concentration was
0.0025 ppm. According to Memo No. PSC-1537 (30 November 1982, "Determi-
nation of Observed Concentrations Used in Model Performance Evaluation at
the Luke Mill"), ERT followed essentially the same approach. However, if
the lowest observed hourly concentration was 0.005 ppm during an hour, ERT
defined the observed (minus background) concentrations as zero at all moni-
tors with observed concentrations of 0.005 ppm and the observed (minus
background) concentrations as the observed concentrations minus 0.0025 ppm
at all monitors with concentrations above 0.005 ppm. The differences in
the observed (minus background) 24-hour average SO concentrations indepen-
dently calculated by ERT and the H. E. Cramer Company are explained by
ERT's use of several zero observed (minus background) hourly concentrations.
Table 3-17 lists the observed (minus background) annual average
SO concentrations calculated for the second year of the Westvaco monitoring
program by ERT and the H. E. Cramer Company. There is an exact agreement
between the two observed (minus background) annual average SO concentra-
tions calculated for Monitor 3. However, the remainder of the observed
(minus background) annual average concentrations calculated by ERT are
lower than the corresponding concentrations calculated by the H. E. Cramer
Company, with a maximum difference of 25 percent at Monitor 10. We believe
that the differences in the procedures used by ERT and the H. E. Cramer
Company to adjust the observed hourly concentrations for background probably
explain the differences in the two sets of observed (minus background)
annual average concentrations.
43
-------
TABLE 3-17
COMPARISON OF THE OBSERVED (MINUS BACKGROUND) ANNUAL AVERAGE S0r
CONCENTRATIONS CALCULATED BY ERT AND THE "
H. E. CRAMER COMPANY
Monitor
1
3
4
5
6
7
8
9
10
Annual Concentrations (ppm)
ERT
0.021
0.018
0.013
0.012
0.033
0.012
0.011
0.010
0.006
HEC
0.022
0.018
0.014
0.014
0.034
0.014
0.012
0.011
0.008
Ratio of ERT
and HEC
Concentrations
0.955
1.000
0.929
0.857
0.971
0.857
0.917
0.909
0.750
44
-------
The model evaluation protocol in Appendix A specified that a
3-hour period must include 3 valid concentration measurements to be consi-
dered in the set of the 25 highest 3-hour average concentrations for a
monitor. Our examination of the tables of 25 highest observed (minus back-
ground) 3-hour average concentrations listed in Appendix A of the Hanna, et
al. (1982b) report indicates that ERT did not follow this procedure. If a
3-hour period contained 2 hours of valid concentration measurements, ERT
defined the "3-hour average" concentration for the period by the 2-hour
average concentration for the two hours with valid concentration measure-
ments. The affected monitors are:
a Monitor 5 during the third 3-hour period on 20 May 1981
e Monitors 7, 8 and 9 during the second 3-hour period on
18 October 1982
The differences in the 25 highest observed (minus background) 3-hour average
SO concentrations calculated by ERT and the H. E. Cramer Company for
Monitors 5, 7, 8 and 9 are not sufficient to affect the outcome of the
model performance evaluation.
The model evaluation protocol contained in Appendix A states that
standard "block average" observed (minus background) and calculated 3-hour
and 24-hour average concentrations shall be compared. A 3-hour period must
contain 3 hours of valid concentration measurements to be considered in the
comparison, while a 24-hour period must contain 18 hours of valid data.
These criteria can have significant effects on the results of the model
performance evaluation. For example, the highest observed (minus background)
24-hour average SO concentrations calculated for Monitor 7 under these
criteria is 0.0994 ppm on 29 December 1980. The highest and second-highest
observed (minus background) 3-hour average S0_ concentrations at Monitor
7 both occurred on 13 November 1981, a day not considered in the 24-hour
block averages because of missing observations after 1100 EST on 13 November.
If the concentration at Monitor 7 is defined as zero during the period 1200
45
-------
through 2400 EST on 13 November, the resulting observed (minus background)
24-hour average concentration is 0.1971 ppm, which is 1.98 times the observed
(minus background) concentration on 29 December 1980. For the period 1200
EST on 12 November through 1100 EST on 13 November, the observed (minus
background) 24-hour average concentration at Monitor 7 is 0.2599 ppm, which
is 2.61 times the observed (minus background) concentration on 29 December
1980.
SHORTZ Model Calculations
To assist ERT in the LUMM and SHORTZ dispersion model evaluation,
the H. E. Cramer Company provided ERT with a computer tape containing: (1)
the SHORTZ hourly meteorological and source inputs developed for the second
year of the Westvaco monitoring program following the procedures specified
in the model evaluation protocol (see Appendix A), and (2) the computer
code for the SHORTZ model as described in the model evaluation protocol.
Additionally, the H. E. Cramer Company provided ERT with specific guidance
on how SHORTZ should be executed (see Table 9 of the protocol). Appendix C
of this report contains tables which list the 25 highest 1-hour, 3-hour and
24-hour average and the annual average S09 concentrations calculated at
each monitor of concern for the model evaluation by the H. E. Cramer Company
using the SHORTZ model. The cumulative frequency distributions of these
sets of 25 highest calculated short-term SO. concentrations are compared
with the cumulative frequency distributions of the corresponding observed
(minus background) short-term concentrations in Appendix D.
Table 3-18 compares the highest and second-highest 3-hour and
24-hour average SO concentrations calculated at each monitor of concern
by ERT and the H. E. Cramer Company using the SHORTZ model. Because the
two sets of calculated short-term concentrations agree to within less than
plus or minus 1 percent, we conclude that the differences in the two sets
of calculated concentrations are attributable to differences in the accuracy
of the computer systems used by ERT and the H. E. Cramer Company. On the
46
-------
Table 3-18
COMPARISON OF THE HIGHEST AND SECOND-HIGHEST 3-HOUR AND 24-HOUR
AVERAGE S02 CONCENTRATIONS CALCULATED BY ERT AND THE
H. E. CRAMER COMPANY USING THE SHORTZ MODEL
Monitor
3-Hour
Concentrations
(ppm)
ERT
HEC
24-Hour
Concentrations
(ppm)
ERT
HEC
Ratio of ERT and
HEC Concentrations
3 -Hour
Concentrations
24-Hour
Concentrations
(a) Highest Concentrations
1
3
4
5
6
7
8
9
10
3.1680
4.1082
2.4836
1.1894
3.1451
1.2502
0.6525
0.7586
0.1586
3.1736
4.1034
2.4823
1.1889
3.1442
1.2483
0.6529
0.7543
0.1591
0.7143
0.8524
0.4259
0.2866
0.8228
0.2334
0.0920
0.1026
0.0457
0.7163
0.8503
0.4258
0.2866
0.8225
0.2330
0.0922
0.1029
0.0458
0.998
1.001
1.001
1.000
1.000
1.002
0.999
1.006
0.997
0.997
1.002
1.000
1.000
1.000
1.002
0.998
0.997
0.998
(b) Second-Highest Concentrations
1
3
4
5
6
7
8
9
10
3.1459
2.5391
2.1491
1.0550
2.9054
0.9574
0.5834
0.5566
0.1335
3.1540
2.5276
2.1496
1.0557
2.9067
0.9606
0.5838
0.5581
0.1338
0.6689
0.6674
0.4051
0.2095
0.7709
0.1612
0.0820
0.0979
0.0413
0.6694
0.6660
0.4050
0.2096
0.7722
0.1610
0.0821
0.0974
0.0414
0.997
1.005
1.000
0.999
1.000
0.997
0.999
0.997
0.998
0.999
1.002
1.000
1.000
0.998
1.001
0.999
1.005
0.998
47
-------
other hand, Table 3-19 shows differences in the annual average SCL concen-
trations calculated by ERT and the H. E. Cramer Company that are more diffi-
cult to explain by differences in machine accuracy. However, we point out
that the differences in the two sets of calculated annual average concentra-
tions in Table 3-19 are not sufficient to affect the outcome of the model
performance evaluation.
As noted above, ERT included in the determination of the sets of
25 highest observed (minus background) 3-hour average SO concentrations
several 2-hour average concentrations for 3-hour periods with only 2 valid
hourly concentration measurements. This same procedure, which deviates
from the model evaluation protocol contained in Appendix A, was also followed
in processing the results of the hourly concentrations calculated by ERT
using the SHORTZ model. The SHORTZ hourly meteorological inputs contain a
total of 18 hours flagged as missing because of calm or variable winds at
all levels of Tower No. 1 and Tower No. 2 (see the second footnote at the
bottom of Table 2 in the model evaluation protocol). The inclusion by ERT
of 3-hour periods containing a single hour with missing calculated concentra-
tions affects the following monitors:
Monitor 6 during the eighth 3-hour period on 5 May 1981
e Monitors 8 and 9 during the first 3-hour period on
30 September 1981
The differences in the 25 highest 3-hour average concentrations calculated
by ERT and the H. E. Cramer Company for Monitors 6, 8 and 9 using the SHORTZ
model are not sufficient to affect the outcome of the model performance
evaluation.
Performance Statistics and Scores
As explained in the model evaluation protocol contained in Appendix
A, the following parameters were used to compute the scores for the LUMM
48
-------
TABLE 3-19
COMPARISON OF THE ANNUAL AVERAGE S02 CONCENTRATIONS CALCULATED BY
ERT AND THE H. E. CRAMER COMPANY USING THE SHORTZ MODEL
Monitor
1
3
4
5
6
7
8
9
10
Annual Concentrations (ppm)
ERT
0.0874
0.0462
0.0271
0.0115
0.1603
0.0099
0.0034
0.0050
0.0046
HEC
0.0905
0.0422
0.0279
0.0116
0.1573
0.0092
0.0037
0.0048
0.0045
Ratio of ERT
and HEC
Concentrations
0.966
1.095
0.971
0.991
1.019
1.076
0.919
1.042
1.022
49
-------
and SHORTZ models: (1) the ratio of the absolute value of the bias
(residual) for the model being evaluated to the absolute value of the mini-
mum bias (residual) for either of the two models, (2) the minimum of the
ratio of calculated and observed concentrations and the ratio of observed
and calculated concentrations, and (3) the minimum of the ratio of the
variances of the calculated and observed concentrations and the ratio of
the variances of the observed and calculated concentrations. We used the
results of our analysis of the observed hourly SO concentrations (see
Appendix B) and our analysis of the hourly SO concentrations calculated
by SHORTZ (see Appendix C) to compute biases (residuals) and variances for
comparison with the corresponding values computed by ERT. The agreement
between the two sets of performance statistics was consistent with the
agreement between the two independent analyses of the observed concentra-
tions and the two sets of SHORTZ model predictions discussed above. We
also used the observed (minus background) concentrations and the concentra-
tions calculated by the LUMM model that are given in Appendix A of the
Hanna, et al. (1982b) report to compute and verify the biases (residuals)
and variances listed for the LUMM model in Table 17 of the Hanna, et al.
(1982b) report. Finally, we checked the protocol scoring results shown in
Table 17. If the concentrations, biases (residuals) and variances given in
Table 17 are accepted as accurate, we disagree with ERT's scoring only for
the comparison of the 25 highest observed (minus background) 24-hour average
concentrations at Monitor 10 paired in space with the 25 highest 24-hour
average concentrations calculated by the LUMM model. From Equation (9) of
the model evaluation protocol and for the concentrations and biases in
Table 17, the score for LUMM is
{Score Model.} = --rr x MIN{C . /0,0/C. } x {Possible Points}
x |f x (20) = 3.4 (3-1)
= 3 to the nearest integer
The score in Table 17 is given as 4 rather than 3.
50
-------
Table 3-20 summarizes the qualitative performance of the LUMM and
SHORTZ models by monitor. The LUMM model has the highest score at six
monitors (Monitors 1, 3, 4, 5, 6 and 7) and the SHORTZ model has the highest
score at three monitors (Monitors 8, 9 and 10). Table 3-20 does not indi-
cate any clear trend in model performance according to monitor elevation
above the top of the Westvaco Main Stack. For example, the SHORTZ model
has the highest score at both the monitor with the highest elevation and
the monitor with the lowest elevation. On the other hand, Table 3-20 appears
to indicate a trend in model performance according to distance from the
Main Stack. The LUMM model has the highest score at each monitor less than
or equal to 1.1 kilometers from the Main Stack and the SHORTZ model has the
highest score at each monitor greater than or equal to 1.5 kilometers from
the Main Stack.
The scores given in Table 17 of the Hanna, et al. (1982b) report
for the LUMM and SHORTZ models are 363 and 168, respectively. Under the
assumption that all of the concentrations, biases (residuals) and variances
in Table 17 are correct, we believe that the score for the LUMM model should
actually be 362 for the reason given above. Also, as discussed above,
there are some differences in ERT's and the H. E. Cramer Company's analyses
of the observed hourly S0? concentrations and of the hourly S09 concentra-
tions calculated by the SHORTZ model. However, these differences are rela-
tively small and do not affect the total score for either model by more
than a few points because the score for each comparison is rounded to the
nearest integer. We conclude that we are in agreement with ERT that the
LUMM model has the highest score.
Summary and Conclusion of the Model Performance Evaluation
Westvaco Corporation, the State of Maryland and EPA agreed on 21
October 1982 to a protocol for the evaluation of the performance of the
LUMM and SHORTZ complex terrain dispersion models using the Westvaco data
set. According to this protocol, the model to be used to establish an
51
-------
TABLE 3-20
SUMMARY OF MODEL PERFORMANCE BY MONITORING SITE
Monitor
1
3
4
5
6
7
8
9
10
Distance from Main
Stack (km)
0.8
0.7
0.9
1.1
0.8
1.0
1.5
1.5
3.4
Elevation Above Main
Stack Top (m)
126
86
125
161
113
159
165
195
26
Model with
Highest Score
LUMM
LUMM
LUMM
LUMM
LUMM
LUMM
SHORTZ
SHORT Z
SHORTZ
52
-------
SO emission limitation for the Main Stack at the Westvaco Luke, Maryland
Mill is the model with the highest score. Westvaco1s consultant (ERT)
conducted the model performance evaluation and concluded that the scores
for the LUMM and SHORTZ models are 363 and 168, respectively. Under con-
tract to EPA, the H. E. Cramer Company independently verified all of the
results of the ERT model performance evaluation except the predictions of
the LUMM model. We identified one minor deviation by ERT from the 21 October
1982 protocol and what we believe to be a single-point scoring error.
Also, there are several minor differences in the analyses of the observed
and calculated (SHORTZ) hourly SO concentrations independently performed
by ERT and the H. E. Cramer Company. However, the deviation by ERT from
the protocol and the differences in the independent analyses of observed
and calculated (SHORTZ) hourly concentrations do not change the score for
either model by more than a few points. Under the terms of the protocol,
we therefore conclude that the LUMM model is the model to be used to esta-
blish an SO emission limitation for the Westvaco Main Stack.
3.3 RESULTS OF THE SO EMISSION LIMITATION CALCULATIONS
The site specific LUMM model was identified as the complex terrain
dispersion model to be used to establish an SO emission limitation for
the Main Stack at the Westvaco Luke Mill under the terms of the 21 October
1982 model evaluation protocol contained in Appendix A. Under the assumption
that the only constraint on the SO emission limitation for the Westvaco
Main Stack is that emissions do not endanger the NAAQS for S0?, ERT (December
1982) used the LUMM model to estimate an emission limitation of 75.1 tons
per day (3-hour average concentration for 50 percent buoyancy flux). Under
this assumption and assuming that the LUMM model calculations were performed
correctly, we agree with the ERT (December 1982) computation of an emission
limitation.
In the absence of the data from the 2-year Westvaco monitoring
program and the model performance evaluation, the SO emission limitation
53
-------
for the Westvaco Main Stack would probably be determined by EPA on the
basis of predictions made using the Complex I model with any available
onsite meteorological data. To gain insight to the differences in emission
limitations arising from the two different modeling approaches, we used the
Complex I model with the source and meteorological inputs developed for the
second year of the Westvaco monitoring program (see Table 1 in Appendix A)
to estimate an emission limitation. As discussed in Section 2, we modified
the Complex I model's computer code to read hourly values of the stack exit
temperature, the stack exit velocity and the vertical potential temperature
gradient and to key the selection of the stable or adiabatic plume rise
equation on the vertical potential temperature gradient.
The Complex I model requires an estimate of the Pasquill stability
category during each hour in order to assign the appropriate wind-profile
exponent and Pasquill-Gifford vertical dispersion coefficient. The October
1980 draft EPA report "Proposed Revisions to the Guideline on Air Quality
Models" suggests that the hourly vertical turbulent intensity (standard
deviation of the wind elevation angle in radians) measured 10 meters above
ground-level can be used to assign the appropriate stability category.
Table 3-21, which is based on Table C-l of the October 1980 draft EPA report,
shows the suggested ranges of vertical turbulent intensities corresponding
to the various stability categories for a surface roughness length z of
15 centimeters. The draft EPA report suggests that these ranges be adjusted
0 2
to the roughness length of the site using a z ' law. According to
the wind-tunnel and field studies of the Westvaco Mill described by Weil
(1979) and Weil, et^ a^. (1981), the roughness length in the vicinity of the
Westvaco Mill is about 16 meters. Table 3-22 gives the adjusted ranges of
vertical turbulent intensities for a 16-meter roughness length. Because
the A, B and C stability categories almost never occur in the Westvaco data
set with the stability classification scheme shown in Table 3-22, we used
the scheme shown in Table 3-21 after consultation with the EPA Project
Officer. Our first and second choices of hourly vertical turbulent inten-
sities for use with Table 3-21 were the values measured at the 10meter
levels of Tower No. 2 and Tower No. 1, respectively. If both of the 10-meter
54
-------
TABLE 3-21
STABILITY CLASSIFICATION SCHEME SUGGESTED BY EPA FOR A SURFACE
ROUGHNESS LENGTH OF 15 CENTIMETERS
Pasquill Stability
Category
10m Vertical Turbulent Intensity
(rad)
A
B
C
D
E
F
>0.2094
0.1746 to 0.2094
0.1362 to 0.1745
0.0874 to 0.1361
0.0419 to 0.0873
<0.0419
TABLE 3-22
PROPOSED EPA STABILITY CLASSIFICATION SCHEME FOR A SURFACE
ROUGHNESS LENGTH OF 16 METERS
Pasquill Stability
Category
A
B
C
D
E
F
10m Vertical Turbulent Intensity
(rad)
>0.5329
0.4441 - 0.5329
0.3464 - 0.4440
0.2221 - 0.3463
0. 1066 - 0.2220
<0.1066
55
-------
observations were missing, we selected from the available measurements the
vertical turbulent intensity from the next lowest level of Tower No. 2 or
Tower No. 1.
We used the actual hourly values of the stack exit velocity and
stack exit temperature in the Complex I model calculations to estimate an
SO. emission limitation for the Westvaco Main Stack. However, we assumed
a constant SO emission rate of 1,000 grams per second for convenience.
We used discrete receptors to form a polar receptor array centered on the
Main Stack with a 10-degree angular spacing between radials. Excluding the
Westvaco property shown in Figure 7-1 of the Hanna, et_ a_l. (1982a) report,
we placed receptors at 500-meter intervals along each radial to the highest
terrain feature between 2.5 and 5.0 kilometers from the stack. The receptors
on each radial extended a minimum of 2 kilometers beyond the Westvaco pro-
perty boundary. Additional receptors were placed on the prominent high
terrain features that were not adequately covered by the polar receptor
array. Because the Complex I model's computer code accepts a maximum of
only 180 receptors, it was necessary to perform two computer runs to include
all of the 262 receptors.
Table 3-23 gives the maximum annual and highest of the second-
highest 3-hour and 24-hour average S0? concentrations calculated by the
Complex I model for emissions from the Westvaco Main Stack assuming an
emission rate of 1,000 grams per second. All of the calculated concentra-
tions in Table 3-23 are located at a point 4.5 kilometers west-northwest of
the Main Stack that is about 110 meters above the stack-top elevation. The
meteorological conditions during the first 3 hours of 1981 Julian Day 45
consisted of light east-southeast winds (wind speeds less than 1.0 meter
per second, defined as 1.0 meter per second for use in the model calculations)
in combination with 2 hours of F stability and a single hour of D stability.
The meteorological conditions on 1981 Julian Day 287 consisted of light-to-
moderate winds (wind speeds from 1.0 to 4.5 meters per second) within the
sector 085 to 115 degrees in combination with 16 hours of E stability, 6
hours of D stability and a single hour each of A stability and F stability.
56
-------
TABLE 3-23
MAXIMUM ANNUAL AND HIGHEST OF THE SECOND-HIGHEST 3-HOUR AND 24-HOUR
AVERAGE SO CONCENTRATIONS CALCULATED BY THE COMPLEX I MODEL FOR
EMISSIONS FROM THE WESTVACO MAIN STACK ASSUMING AN EMISSION
RATE OF 1,000 GRAMS PER SECOND
Averaging
Time
3 Hour
24 Hour
Annual
Year/Julian
Day (Period)
1981/45(1)
1981/287
Location*
Distance
(m)
4,500
4,500
4,500
Azimuth
Bearing (deg)
290
290
290
Elevation
(m MSL)
588
588
588
Concentration
(jag/m3)
2,543
1,123
92.7
*Locations are with respect to the Westvaco Main Stack.
57
-------
Both the 3-hour and 24-hour periods were characterized by large hourly
vertical wind-direction shears and/or hourly lateral turbulent intensities.
Consequently, the "worst-case" short-term periods for the Complex I model
are relatively high dilution periods for the SHORTZ and LUMM models.
We computed an S09 emission limitation for the Westvaco Main
Stack under the following assumptions: (1) the NAAQS for SCL are the
only constraints, and (2) the background SCL concentration is 13 micrograms
per cubic meter for all concentration averaging times. Under these assump-
tions, the allowable S0« emission rate is given by
(g/sec) = 1000CNAAQS-13)
J\
where x is the calculated concentration from Table 3-23 and 1,000 is the
SO. emission rate used in the Complex I model calculations. The resulting
allowable emission rate is given in Table 3-24 for each NAAQS. As shown by
the table, the 24-hour NAAQS restricts the allowable SO emissions to
about 29.9 tons per day in the Complex I model calculations, a factor of
2.5 lower than the emission limitation of 75.1 tons per day determined by
ERT (December 1982) using the LUMM model.
58
-------
TABLE 3-24
MAXIMUM ALLOWABLE SO EMISSION RATES FOR THE WESTVACO MAIN STACK
BASED ON THE RESULTS OF THE COMPLEX I CALCULATIONS
Averaging
Time
3 Hours
24 Hours
Annual
*
Allowable S0_ Emission Rate
(g/sec)
506.1
313.4
722.8
(ton/day)
48.2
29.9
68.8
The allowable SO emission rates assume: (1) the background SO. concen-
tration is 13 micrograms per cubic meter for all concentration averaging
times, and (2) compliance with the National Ambient Air Quality Standards
(NAAOS) is the only constraint on Westvaco
SO emissions.
59
-------
(This Page Intentionally Blank).
60
-------
SECTION 4
CONCLUSIONS
In our opinion, the Westvaco data set is the most detailed and
best documented data set developed to date for the purpose of evaluating
and validating complex terrain dispersion models. However, we believe that
the Westvaco Luke Mill modeling problem is sufficiently unique that any
conclusions about the accuracy of dispersion models evaluated using the
Westvaco data set should be used with caution unless they are supported by
previous (and/or future) experience in testing the models in complex terrain.
Also, the Westvaco data set contains ambiguities that are unlikely ever to
be resolved. For example, we question whether the sequence of events and
the physical processes leading to the occurrence of the highest short-term
SO concentrations at the monitors on elevated terrain south of the Westvaco
Main Stack (Monitors 7, 8 and 9 in Figure 1-1) can ever be determined with
any acceptable degree of confidence using the archived hourly average data.
With these caveats, the following paragraphs discuss our conclusions about:
(1) complex terrain dispersion model performance, (2) meteorological measure-
ments to develop dispersion model inputs in areas of complex terrain, and
(3) the effects of monitoring network design on the results of dispersion
model performance evaluations.
Dispersion Model Performance
The complex terrain dispersion models considered in the Westvaco
model evaluation study can be divided into two general categories:
(1) screening models (the Valley, Complex I and Complex II models), and
(2) refined models (the SHORTZ and LUMM models). The objective of the
screening models is to provide safe-sided estimates of maximum short-term
concentrations when little or no onsite meteorological data are available.
Because this objective generally was satisfied by the three screening
models, we believe that they can be defined as state-of-the-art complex
61
-------
terrain screening models. The objective of the refined models is to use
onsite meteorological measurements to provide accurate and unbiased esti-
mates of the highest short-term concentrations. This objective was satis-
fied by the LUMM model at all monitors located between 0.7 and 1.1 kilometers
from the Westvaco Main Stack and by the SHORTZ model at all monitors located
between 1.5 and 3.4 kilometers from the Main Stack. The performance of the
LUMM and SHORTZ models for the Westvaco data set appears to be a function
only of distance from the Main Stack; the SHORTZ model has the best quanti-
tative performance at the monitors with both the lowest and highest eleva-
tions above the stack-top elevation. Assuming that the typical distance to
plume stabilization is on the order of ten stack heights (Briggs, 1969),
one possible interpretation of the results of the performance evaluation is
that the LUMM model is the state-of-the-art refined model at distances less
than the distance to plume stabilization and the SHORTZ model is the state-of-
the-art refined model at longer downwind distances. We conclude from these
results that both the LUMM and SHORTZ models are state-of-the-art refined
models for the Westvaco data set. However, under the terms of the 21 October
1982 model evaluation protocol, we also conclude that the LUMM model should
be used to establish an S0_ emission limitation for the Westvaco Main
Stack.
The LUMM model was specifically developed for application to the
Westvaco data set, and the final version of the model represents the combin-
ation of model constants and meteorological inputs that yields the best fit
to the air quality measurements. Consequently, the LUMM model cannot be
assumed to be a generalized state-of-the-art complex terrain dispersion
model. However, many of the concepts upon which the LUMM model is based
may be suitable for generalized applications, and we believe that it would
be desirable to conduct additional performance evaluations of these concepts.
The SHORTZ model, on the other hand, has been applied as a generalized
complex terrain dispersion model over a period of almost 8 years. Although
previous tests of the SHORTZ model (see Appendix H of Bjorklund and Bowers,
1982) have not been as rigorous as the Westvaco model performance evaluation,
the results of these tests have consistently shown a close agreement between
62
-------
calculated and observed short-term SO concentrations for sources located
in complex terrain. The Westvaco data set offered the first opportunity to
date to utilize the SHORTZ model's capability of directly relating turbu-
lence measurements to plume expansion. After modification of the SHORTZ
model to account for the large vertical wind-direction shears (sometimes as
large as 180 degrees) frequently encountered by the plume from the Westvaco
Main Stack within a transport distance of less than 1 kilometer, the perfor-
mance of the SHORTZ model at the monitors beyond the typical distance to
plume stabilization was consistent with previous SHORTZ model performance
evaluations and thus tends to support the continued use of the SHORTZ model
as a generalized complex terrain dispersion model. However, the SHORTZ
model's bias in the Westvaco model evaluation study toward overestimation
of concentrations on elevated terrain at distances less than the distance
to plume stabilization indicates that the model's predictions should be
used with caution and common sense in these situations.
Meteorological Measurements in Complex Terrain
The results of the Westvaco data analysis and model performance
evaluation studies clearly demonstrate the critical importance of represen-
tative onsite meteorological measurements to develop meteorological inputs
for refined complex terrain dispersion models. As discussed by Hanna, et
al. (1982a), the localized circulations affecting the initial transport and
dispersion of the plume from the Westvaco Main Stack are essentially decoupled
from the synoptic scale circulation and show virtually no correlation with
the wind data from the nearest airport (Greater Pittsburgh Airport). The
results of the Westvaco study also show the difficulty of specifying in
advance what meteorological measurements are representative for modeling
purposes. For example, the purpose of the Beryl Meteorological Tower (see
Figure 1-1) was to provide insight to the valley winds experience by the
Westvaco plume as it exited the Main Stack. The 100-meter Beryl Tower was
so sheltered by the topography during the 2-year monitoring program
63
-------
that over a third of the hourly vector average winds at the 100-meter level
were reported as variable or calm. Thus, the Beryl Tower wind data were of
no practical value in defining source-receptor relationships during the
light wind cases with high observed SCL concentrations at the monitors on
the ridgeline southeast of the Main Stack. In retrospect, a Doppler acoustic
sounder at the site of the Beryl Tower probably would have provided far
more meaningful information on stack height winds and turbulent intensities.
The combination of the multilevel 100-meter Tower No. 1 and the
multilevel 30-meter Tower No. 2 with tower base elevations ranging from 10
meters below the stack-top elevation (Tower No. 2) to 126 meters above the
stack-top elevation (Tower No. 1) enabled the construction of vertical
wind, turbulence and temperature profiles. Although the general representa-
tiveness of these profiles is not known because of the different locations
of the two towers, these profiles were of considerable practical value and
were extensively used in the development of meteorological inputs for the
SHORTZ and LUMM models. We therefore recommend that future field measurement
programs designed to collect data to evaluate complex terrain dispersion
models include routine measurements of the vertical profiles of winds,
turbulence and temperature. Remote sensing (Doppler acoustic sounders) is
probably the most practical way in which to obtain the wind and turbulence
profiles.
Monitoring Network Design
The results of the Westvaco model performance evaluation study
illustrate how the density and locations of the air quality monitors can
significantly affect conclusions about model performance. For example, if
only Monitors 8, 9 and 10 had been in place during the 2-year Westvaco
monitoring program, the results of the performance evaluation would lead to
the unambiguous conclusion that the SHORTZ model provides very accurate
estimates of the 25 highest 1-hour, 3-hour and 24-hour average SCL concen-
trations paired in space only, while the LUMM model systematically under-
estimates these concentrations. On the other hand, if only Monitors
64
-------
1, 3, 4, 5, 6 and 7 had been in place, the results of the performance eval-
uation would lead to the unambiguous conclusion that the LUMM model provides
very accurate estimates of the 25 highest 1-hour, 3-hour and 24-hour average
SO concentrations paired in space only, while the SHORTZ model systemati-
cally overestimates these concentrations. It follows that conclusions
about model performance that are based on comparisons of observed and calcu-
lated concentrations at a single monitor or at a limited number of monitors
may not be transferable to other locations in the vicinity of the same
source. Additionally, any model that must be calibrated or "tuned" to
match the observations at a limited number of monitoring sites cannot be
used with confidence at other receptor locations.
65
-------
(This Page Intentionally Blank)
66
-------
REFERENCES
Bjorklund, J. R. and J. F. Bowers, 1982: User's instructions for the SHORTZ
and LONGZ computer programs. EPA Reports EPA-903/9-82-004a and
004b (NTIS Accession Numbers PB83-146092 and 146100), U. S. Environ-
mental Protection Agency, Region III, Philadelphia, PA.
Briggs, G. A., 1969: Plume Rise. Available as TID-25075 from Clearinghouse
for Federal Scientific and Technical Information, Springfield,
VA, 80.
Briggs, G. A., 1971: Some recent analyses of plume rise observations. In
Proceedings of the Second International Clean Air Congress.
Academic Press, NY.
Briggs, G. A., 1972: Chimney plumes in neutral and stable surroundings.
Atmospheric Environment, ^>(7), 507-510.
Briggs, G. A., 1973: Diffusion estimates for small emissions. ATDL Contri-
bution File No. (Draft) 79, Air Resources Atmospheric Turbulence
and Diffusion Laboratory, Oak Ridge, TN.
Briggs, G. A., 1975: Plume rise predictions. Lectures on Air Pollution
and Environmental Impact Analyses. American Meteorological Society,
Boston, MA.
Burt, E. W., 1977: Valley model user's guide. EPA Report No. EPA-450/2-
77-018, U. S. Environmental Protection Agency, Research Triangle
Park, NC.
Burt, E. W. and H. H. Slater, 1977: Evaluation of the Valley model. Preprint
Volume for the Joint Conference on Applications of Air Pollution
Meteorology, American Meteorological Society, Boston, MA, 192-195.
Cramer, H. E., et^ _a_l, 1972: Development of dosage models and concepts.
Final~Report under Contract DAAD09-67-0020(R) with the U. S.
Army, Deseret Test Center Report DTC-TR-72-609, Fort Douglas, UT.
Cramer, H. E., H. V. Geary and J. F. Bowers, 1975: Diffusion-model calcu-
lations of long-term and short-term ground-level SO. concentrations
in Allegheny County, Pennsylvania. EPA Report No. EPA-903/9-75-018
(NTIS Accession No. PB-245262/AS), U. S. Environmental Protection
Agency, Region III, Philadelphia, PA.
Cramer, H. E., 1976: Improved techniques for modeling the dispersion of
tall stack plumes. Proceedings of the Seventh International
Technical Meeting on Air Pollution Modeling and Its Application,
NATO Committee on the Challenges to Modern Society, 731-780.
67
-------
Cramer Company, H. E., 1981: Westvaco Luke, Maryland monitoring program:
Data analysis and dispersion model evaluation (first two quarters).
H. E. Cramer Technical Report TR-81-202-01 prepared for the U. S.
Environmental Protection Agency under subcontract to Research
Triangle Institute, Research Triangle Park, NC.
Environmental Research & Technology, Inc., 1981: Addendum to the report
Diffusion Model Development and Evaluation and Emission Limitations
at the Westvaco Luke Mill, Environmental Research & Technology,
Inc., Concord, MA.
Hanna, S. R., 1982: Private communication (statements made at a 10 September
1982 meeting between representatives of the U. S. Environmental
Protection Agency, the State of Maryland and Westvaco Corporation).
Hanna, S. , et_ aL_. , 1982a: Diffusion model development and evaluation and
emission limitations at the Westvaco Luke Mill. Document PA439,
Environmental Research & Technology, Inc., Concord, MA.
Hanna, S. R., et^ jl. , 1982b: An evaluation of the LUMM and SHORTZ dispersion
models using the Westvaco data set. Document No. PA-439, Environ-
mental Research & Technology, Inc., Concord, MA.
Pasquill, F., 1976: Atmospheric dispersion-parameters in Gaussian plume
modeling. Part II, possible requirements for changes in the
Turner Workbook Values. EPA Report No. EPA-600/4-76-3606, U. S.
Environmental Protection Agency, Research Triangle Park, NC.
Pierce, T. E. and D. B. Turner, 1980: User's guide for MPTER. EPA Report
No. EPA-600/8-80-016, U. S. Environmental Protection Agency,
Research Triangle Park, NC.
Turner, D. B., 1964: A diffusion model for an urban area.
Applied Meteorology, 3(1), 83-91.
Journal of
Turner, D. B., 1970: Workbook of atmospheric dispersion estimates, Publi-
cation No. 999-AP-26, National Air Pollution Control Administration,
Cincinnati, OH.
Weil, J. C. , 1979: Modeling of buoyant plume dispersion in complex terrain.
Martin Marietta Corp. Report No. PPRP-35, Martin Marietta Corpor-
ation, Baltimore, MD.
Weil, J. C., J. E. Cermak and R. L. Petersen, 1981: Plume dispersion about
the windward side of a hill at short range: Wind tunnel vs field
measurements. Preprint Volume for the Fifth Symposium on Turbulence,
Diffusion and Air Pollution, American Meteorological Society,
Boston, MA, 159-160.
68
-------
APPENDIX A
MODEL EVALUATION PROTOCOL
This appendix contains the protocol for the evaluation of the
SHORTZ and LUMM dispersion models using the Westvaco data set that was
agreed upon on 21 October 1982 by Westvaco Corporation, the State of
Maryland, the U. S. Environmental Protection Agency (EPA) Region III and
the EPA Office of Air Quality Planning and Standards (OAQPS).
A-l
-------
(This Page Intentionally Blank)
A-2
-------
21 October 1982
PROTOCOL FOR THE EVALUATION OF THE SHORTZ AND
LUMM DISPERSION MODELS USING THE WESTVACO DATA SET*
1. BACKGROUND AND PURPOSE
The Westvaco data set consists of detailed records of hourly
emissions, meteorological and SO air quality data collected in the vicinity
of the Westvaco Corporation Paper Mill at Luke, Maryland during the 2-year
period December 1979 through November 1981. The purpose of the Westvaco
monitoring program was to acquire the data needed to select the most appro-
priate complex terrain dispersion model for use in establishing an S0_
emission limitation for the Westvaco Main Stack. Under Contract No. 68-02-
3577 with the U. S. Environmental Protection Agency (EPA), the H. E. Cramer
Company, Inc. is to use the Westvaco data set to assist in the selection of
the complex terrain dispersion model "which provides the best air quality
predictions, and which will be the basis for a permanent SO. emission
limitation." EPA Region III, the EPA Office of Air Quality Planning and
Standards (OAQPS), the State of Maryland and Westvaco Corporation have
agreed that the two complex terrain dispersion models most likely to be
applicable at the Luke Mill are the SHORTZ model and the Luke Mill Model
(LUMM). SHORTZ was developed and documented by the H. E. Cramer Company
under previous EPA contracts (Cramer, et^ aul. , 1975; Bjorklund and Bowers,
1982) and LUMM was developed for Westvaco Corporation by Hanna, et al.
(1982).
The purpose of this Technical Note is to summarize a protocol,
agreed upon in advance by EPA Region III, EPA OAQPS, the State of Maryland
and Westvaco Corporation, for the evaluation of the two candidate dispersion
models as required by the August 1981 EPA report "Interim Procedures for
Technical Note prepared by the H. E. Cramer Company, Inc., Salt Lake
City, Utah under Contract No. 68-02-3577 with the U. S. Environmental
Protection Agency.
A-3
-------
Evaluating Air Quality Models." The August 1981 report suggests that the
evaluation of a dispersion model begin with an examination of the specific
modeling problem to determine if the current Guideline on Air Quality Models
(EPA, 1978) recommends a reference model for the intended type of applica-
tion. Neither SHORTZ nor LUMM is cited in the current Guideline on Air
Quality Models. SHORTZ is arbitrarily defined as the reference model in
this model evaluation protocol because of previous applications of the
model to the Luke Mill by EPA Region III and the H. E. Cramer Company. The
interim procedures recommend that a technical comparison of the proposed
model (LUMM) and the reference model (SHORTZ) first be made to determine if
the proposed model is qualitatively better than, comparable to, or worse
than the reference model. This technical comparison is then followed by a
model performance evaluation which focuses on the model performance attri-
butes of concern for the intended application. If the results of the model
performance evaluation are inconclusive, the results of the technical evalu-
ation are used to select the most appropriate model. If the results of
both the model performance evaluation and the technical evaluation are
inconclusive, the Interim Procedures specify the use of the reference model.
Because of time and level-of-effort constraints, this protocol addresses
only the model performance evaluation.
2. GENERAL APPROACH
Eleven continuous S0_ monitors were operated in the vicinity of
the Westvaco Mill during the 2-year monitoring program. The monitors with
the highest observed concentrations were located on elevated terrain in the
90-degree sector southeast of the Westvaco Main Stack (Monitors 1, 3, 4, 5,
6, 7, 8 and 9). Because the distances from the Main Stack to these monitors
range from 740 to 1,500 meters, the Westvaco data set principally reflects
the concentrations at these relatively short distances. Monitor 10 (Stony
Run), which is on elevated terrain 3,400 meters northeast of the Main Stack,
is of particular importance because it is the only monitor at the typical
A-4
-------
distance from the Main Stack to the Westvaco property boundaries. The two
other monitors, Monitor 2 and Monitor 11 (Bloomington), are not of major
interest for the purpose of dispersion model evaluation for two reasons.
First, these monitors generally have the lowest observed concentrations.
Second, the distances from the Main Stack to these monitors are within the
distance range for the monitors in the sector southeast of the Main Stack.
The model performance evaluation is therefore restricted to Monitors 1, 3,
4, 5, 6, 7, 8, 9 and 10.
The S00 concentrations of primary concern for regulatory purposes
z.
are the maximum annual and the highest of the second-highest 3-hour and
24-hour average concentrations because these concentrations are required to
assess compliance with the current National Ambient Air Quality Standards
(NAAQS) for SO . If there are inadequacies in the available data base, the
maximum 3-hour and 24-hour average S0_ concentrations may also be of concern
for regulatory purposes (EPA,1978). Consequently, maximum observed and
calculated annual, 3-hour and 24-hour average concentrations and second-
highest observed and calculated 3-hour and 24-hour average concentrations
will be compared. For consistency with current EPA policy on the enforce-
ment of the NAAQS, the observed and calculated 3-hour and 24-hour average
concentrations will be for the standard clock hours and calendar days
("block averages"). The effects of "background" (ambient SO concentrations
attributable to sources other than emissions from the Westvaco Main Stack)
will be removed from the observed concentrations before performing the
comparisons of observed and calculated concentrations. The background
concentration during each hour will be defined as the minimum observed
concentration if this concentration is above the monitor threshold concentra-
tion of 0.005 parts per million (13 micrograms per cubic meter). Because
concentrations below 0.005 parts per million are recorded as 0.005 parts
per million in the VJestvaco data set, the background will be defined as
half the monitor threshold concentration (0.0025 parts per million or 6.5
micrograms per cubic meter) if the minimum observed concentration is
recorded as 0.005 parts per million.
A-5
-------
The possible pairings of observed and calculated concentrations
for the purpose of model evaluation include (Fox, 1981): (1) maximum or
total fields of observed and calculated concentrations paired in space and
time, (2) maximum observed and calculated concentrations paired in time
only, (3) maximum observed and calculated concentrations paired in space
only, and (4) maximum observed and calculated concentrations unpaired in
either space or time. Because of limitations and uncertainties in model
source and meteorological inputs, it is not feasible to model the highest
short-term concentrations paired in space and time. Consequently, only
annual average concentrations paired in space and time will be compared.
(The air quality data will be used to determine the location of the maximum
annual average concentrations paired in space and time.) Model evaluations
of maximum concentrations paired in time only are based on the premise that
the model can predict the magnitude of the maximum concentration during any
time period with greater accuracy than it can predict the location of the
maximum concentration. For example, an uncertainty in the transport wind
direction of only a few degrees can lead to large errors in the hourly
concentrations calculated at fixed monitor locations in spite of the fact
that the model may accurately predict the maximum concentrations at the
downwind distances of the monitors (see Figures E-2 and E-3 of Cramer, et
al. 1976). Assuming a "perfect model," the highest short-term concentra-
tions calculated over a long period of record at fixed monitor locations
should be in good agreement with the highest concentrations observed during
the same period if the uncertainties in the model's source and meteoro-
logical inputs are random rather than systematic. For this reason, this
protocol includes a comparison at each monitor of interest of: (1) the
maximum and second-highest 3-hour and 24-hour average observed and calcu-
lated concentrations unpaired in time, and (2) the 25 highest 1-hour, 3-hour
and 24-hour average observed and calculated concentrations unpaired in
time. In addition to the comparisons of the highest short-term concentra-
tions paired in space only, the 25-highest 1-hour, 3-hour and 24-hour average
observed and calculated concentrations unpaired in space or time will be
compared because these concentrations are of practical importance in
regulatory decisions.
A-6
-------
3. SHORTZ MODEL MODIFICATIONS FOR USE IN THE WESTVACO STUDY
SHORTZ differs from LUMM in that SHORTZ is a generalized model
designed for application to single or multiple sources in regions of complex
terrain, whereas LUMM is a single-source model specifically designed for
application to the Westvaco Luke Mill. It has been the experience of the
H. E. Cramer Company that the "universal function" implicit in the SHORTZ
equation for the lateral dispersion coefficient a adequately accounts for
the effects of vertical wind-direction shear on lateral plume expansion in
most situations (Bjorklund and Bowers, 1982, p. 2-33). However, based on
an examination of the hourly wind-direction and SO concentration measure-
ments from the first two quarters of the Westvaco monitoring program, the
H. E. Cramer Company (January 1981) reported to EPA Region III that the
plume from the Westvaco Main Stack is subject to very large vertical wind-
direction shears as it rises through the highly channeled valley flow and
enters the flow above the elevated terrain. Because of these large wind-
direction shears, our January 1981 report suggested that it would be appro-
priate to modify SHORTZ for application to the Luke Mill by inclusion of
the Cramer, et_ al^. (1972) technique for accounting for the effects of ver-
tical wind-direction shear on crosswind plume expansion. Following this
approach, the total lateral dispersion coefficient a T is given by
2 /A8'xx 2
a +
y
(1)
where a is the unmodified SHORTZ lateral dispersion coefficient, x is the
downwind distance and A6' is the wind-direction shear in radians for the
layer containing the plume. We also concluded in our January 1981 report
that the difference in wind direction between the upper levels of Tower No.
1 and Tower No. 2 (the Luke Hill Tower) probably provides the best
available objective indicator of A6*. We therefore developed a modified
version of SHORTZ for use in the Westvaco model evaluation effort that
incorporates Equation (1). Parenthetically, Hanna, e£ al. (1982)
independently arrived at similar conclusions about how best to account for
the effects of vertical wind-direction shear in LUMM.
A-7
-------
4. MODEL INPUTS
Hourly Meteorological Inputs
In the opinion of Westvaco Corporation, the hourly meteorological
inputs previously developed by Hanna, _et_ aJL. (1982) for use with LUMM are
an integral part of LUMM. Because EPA has accepted Westvaco's contention,
the LUMM hourly meteorological inputs for use in the model evaluation will
be as defined in Table A-l of the Hanna, et_ al. (1982) report except that
the set of artificial north winds will be removed. The SHORTZ hourly meteor-
ological inputs discussed below were selected to minimize any modifications
of the onsite measurements or substitutions for missing data in order to
preserve the scientific objectivity and validity of the model evaluation.
The primary SHORTZ hourly meteorological inputs for use in the
model evaluation are listed in Table 1. Previous work indicates that the
wind-direction measurements most representative of the transport wind direc-
tions vary with the meteorological conditions. If, for simplicity, wind
directions from only one tower and level are selected for use in the model
calculations, we believe that the wind directions from the 100-meter level
of Tower No. 1 are most representative of the transport wind directions for
all meteorological conditions. Although the 10-meter level of Tower No. 2
is closest to the elevation of the top of the Main Stack, this level is
likely sheltered by local roughness elements and terrain during hours with
winds toward the monitoring network. Consequently, we have selected the
30-meter level of Tower No. 2 to obtain the SHORTZ reference level wind
speeds. As noted in Section 3, we will use the wind-direction difference
between the upper levels of Towers No. 1 and No. 2 to estimate vertical
wind-direction shear for use in the modified SHORTZ calculations. The
wind-profile exponents will be estimated from the differences in wind speed
between the upper levels of the two towers, while the vertical potential
temperature gradients will be based on the differences in temperature
between the top of Tower No. 1 and the 10-meter level of Tower No. 2. The
ambient air temperatures used in the plume rise calculations will be from
-------
TABLE 1
PRIMARY SHORTZ HOURLY METEOROLOGICAL INPUTS
Input Parameter
Primary Source
Transport Wind Direction
Reference Level Wind Speed
Vertical Wind-Direction Shear
Wind-Profile Exponent
Vertical Potential Temperature
Gradient
Ambient Air Temperature
Lateral and Vertical Turbulent
Intensities
Mixing Depths
100m Level of Tower No. 1
30m Level of Tower No. 2
(Luke Hill Tower)
Direction Difference between
Upper Levels of Towers
No. 1 and No. 2
Based on Speed Difference between
Upper Levels of Tower No. 1
and No. 2
Based on Temperature Difference
between 10m Level of Tower No. 2
and 100m Level of Tower No. 1
10m Level of Tower No. 2
30m Level of Tower No. 2
A constant value of 1000m
A-9
-------
the 10-meter level of Tower No. 2. Because Tower No. 2 is upwind of the
Main Stack during hours with winds toward the main monitoring network, we
believe that the turbulent intensities from the 30-meter level of this
tower are most likely to be representative of the turbulence affecting the
plume. Based on a comparison of onsite minisonde measurements, tower temper-
ature measurements and acoustic sounder measurements, we previously con-
cluded that the acoustic sounder mixing depths are invalida conclusion
also reached by Hanna, _et^ aA. (1982). It is our opinion from an examination
of the Westvaco data set and preliminary SHORTZ calculations that the plume
from the Main Stack is almost always contained within the surface mixing
layer and that the restriction on vertical mixing usually has no effect on
the ground-level concentrations at the short downwind distances of the air
quality monitors. Consequently, in the absence of satisfactory measurements
of mixing depths, we will assume a constant mixing depth of 1,000 meters in
the SHORTZ calculations.
We searched the Westvaco data set to see how many hours during
each year of the 2-year monitoring program would require no data substitu-
tions if we used the primary hourly meteorological inputs in Table 1.
Complete primary hourly inputs are available for 3,608 hours during the
first year and for 5,282 hours during the second year. If calendar days
are considered, complete hourly inputs are available for 20 days during the
first year and for 140 days during the second year. Because of the
extremely large number of hours of missing meteorological data during the
first year, only the second year will be considered in the model evaluation
using data substitutions as shown in Table 2. Although we have serious
reservations about the use of data substitutions because they raise serious
questions about the validity of any conclusions that might be reached about
model performance, we believe the data substitutions listed in Table 2
comprise the most objective procedure for developing a complete set of
hourly meteorological inputs for the second year of data.
A-10
-------
TABLE 2
DATA SUBSTITUTIONS TO BE USED IN DEVELOPING SHORTZ HOURLY
METEOROLOGICAL INPUTS
Input
Parameter
Rank of Parameter
Source
Parameter Source
Transport Wind
Direction1
1
2
3
4
5
100m Level of Tower No. 1
50m Level of Tower No. 1
10m Level of Tower No. 1
30m Level of Tower No. 2
10m Level of Tower No. 2
Reference Level Wind
Speed2
1
2
3
4
5
30m Level of Tower No. 2
10m Level of Tower No. 1
50m Level of Tower No. 1
100m Level of Tower No. 1
10m Level of Tower No. 2
Vertical Wind-Direction
Shear3
Direction Difference between
100m Level of Tower No. 1
and 30m Level of Tower No. 2
Direction Difference between
50m Level of Tower No. 1 and
30m Level of Tower No. 2
Direction Difference between
10m Level of Tower No. 1 and
30m Level of Tower No. 2
Direction Difference between
100m Level of Tower No. 1 and
10m Level of Tower No. 2
Direction Difference between
50m Level of Tower No. 1 and
10m Level of Tower No. 2
Direction Difference between
10m Level of Tower No. 1 and
10m Level of Tower No. 2
Direction Difference between
100m and 10m Levels of Tower
No. 1
Direction Difference between
50m and 10m Levels of Tower
No. 1
Wind-Profile Exponent1*
Based on Speed Difference between
100m Level of Tower No. 1 and 30m
Level of Tower No. 2
A-11
-------
TABLE 2 (Continued)
Input
Parameter
lank of Parameter
Source
Parameter Source
Wind-Profile Exponent1*
(Continued)
7
8
Based on Speed Difference between
50m Level of Tower No. 1 and 30m
Level of Tower No. 2
Based on Speed Difference between
10m Level of Tower No, 1 and 30m
Level of Tower No. 2
Based on Speed Difference between
100m Level of Tower No. 1 and 10m
Level of Tower No. 2
Based on Speed Difference between
50m Level of Tower No. 1 and 10m
Level of Tower No. 2
Based on Speed Difference between
10m Level of Tower No. 1 and 10m
Level of Tower No. 2
Based on Speed Difference between
100m and 10m Levels of Tower No. 1
Based on Speed Difference between
50 and 10m Levels of Tower No. 1
Vertical Potential
Temperature Gradient5
Based on Temperature Difference
between 100m Level of Tower No. 1
and 10m Level of Tower No. 2
Based on Temperature Difference
between 10m Level of Tower No. 1
and 10m Level of Tower No. 2
Based on Temperature Difference
between 100m and 10m Levels of
Tower No. 1
Based on Temperature Difference
between 30m and 10m Levels of
Tower No. 2
Ambient Air Temperature
10m Level of Tower No. 2
10m Level of Tower No. 1
10m Level of Beryl Tower
Lateral and Vertical
Turbulent Intensities6
1
2
3
4
5
30m Level of Tower No. 2
10m Level of Tower No. 1
50m Level of Tower No. 1
100m Level of Tower No. 1
10m Level of Tower No. 2
A-12
-------
TABLE 2 (Continued)
If no non-variable wind direction is found, the hour will be flagged by
setting the wind direction equal to 090 degrees and the mixing depth
equal to 1 meter.
Wind speeds above 0, but less than 1 meter per second, will be set equal
to 1 meter per second. If all of the wind speeds are calm, the hour will
be flagged by setting the wind direction equal to 090 degrees and the
mixing depth equal to 1 meter.
If none of the data substitutions is possible, the wind-direction shear
will be set equal to zero.
The wind-profile exponent will be set equal to zero if the calculated
exponent is negative or if none of the data substitutions is possible.
The wind-profile exponent will not be allowed to exceed unity.
If none of the data substitutions is possible, the vertical potential
temperature gradient will be set equal to the moist adiabatic value of
0.003 degrees Kelvin per meter.
If no turbulence measurements are available, the lateral and/or vertical
turbulent intensities entered will be climatological values for the
combination of season, wind-speed and time-of-day categories.
A-13
-------
The wind-profile exponents for use in the SHORTZ calculations
will be based on the model's assumption that the wind speed at height z
above mean sea level is given by
u{z} =
z > z
u
R
_
R
R
(2)
where u is the wind speed at height z above the surface at a point with
K K
elevation z above mean sea level. In the SHORTZ calculations, z will be
a a
defined as the elevation at the base of Tower No. 2 (468 meters MSL) and z
K
will be defined as the Tower No. 2 wind-speed measurement height of 30
meters above ground level. The first line of Equation (2) may be rewritten
as
u /u..
(3)
with u , u , z and z1 as defined in Table 3 for the eight possible combin-
ations of wind-speed inputs. As shown by Table 3, z is the height above
ground level at which u1 is measured and for the first six choices of
wind-speed inputs z is the height above the base of Tower No. 2 at
which u_ is measured. Wind-speed measurements from Tower No. 1 alone are
used for the last two choices of wind-speed inputs, and z for these choices
is the height above the base of Tower No. 1 at which u0 is measured.
A-14
-------
w
hJ
PQ
w
H
u
rJ
U
O S3
H H
« O
W EC
CO CO
£3
EC
o
23 W
<; co
txl Pi
o
CO [n
EH
EC CO
O H
M 23
W W
O
W PM
> X
M W
w o
Pi
O PM
23 I
<; o
CO M
O S
w
p-l
CO
CO
c
C
H
4-1
CO
£3
cr
W
d
H
N
C
cfl
CM
N
W
4-1
H
Ol
cc
1 rJ
IM
O
01
o
^
o
CO
£N
1 3
14-1
0
Ol
O
t-4
3
O
CO
4-1
O
^
C!
CO
ei
, v
e
^-^
,1
N
'g"
* '
CN
N
CO
H
C
0
H
4J
CO
3
£
C
H
CO
,J
C!
O
H
4-1
CO
3
c-
W
C
H
01
O
(-1
3
O
CO
TD
0)
0)
a,
CO
1
c
1-1
o
CO
CN
CO
CM
CM
0
23
!_,
OJ
O
H
U-l
O
rH
0)
^
Ol
rJ
e
o
CO
tH
O
^
0)
o
U-4
O
rH
Ol
£>
0)
e
o
o
1 1
o
CO
CM
CO
!-H
CM
*
O
23
J_,
Ol
Jj
O
H
14-1
0
rH
Ol
£>
a>
>-3
e
o
CO
1 1
o
z
Ol
^
o
H
U-4
0
rH
01
J>
01
rJ
e
o
in
CM
0
CO
CN
0)
i-3
O
CO
tH
0
^
^
o
14-i
O
, j
0)
J>
Ol
rJ
B
o
i H
CO
O
i i
CM
CO
CM
CM
O
^J
01
15
0
H
14-1
O
rH
Ol
p>
01
o
t 4
1 1
0
23
01
O
H
U-l
o
rH
01
£>
OJ
rJ
e
o
0
r-H
*J-
o
1 1
CN
00
i I
CM
O
23
J_,
0)
o
H
M-f
0
rH
01
£>
Ol
o
r t
O
2
o>
J^
o
H
U-l
o
rH
01
J>
0)
e
o
m
in
o
i i
CM
^J-
i I
CM
O
i-J
Ol
o
H
U-l
o
rH
Ol
(1)
rJ
0
1 1
1 1
*
o
z
0)
o
H
U-l
0
rH
01
£>
o>
rJ
£
O
r 1
^O
o
1 1
0
o
i i
r 1
!§
M
[5
O
H
U-l
o
rH
Ol
CD
rJ
1
14
1 1
O
s
0)
&
o
H
U-l
o
rH
Ol
£>
01
£
O
o
1 1
r-^.
O
. i
0
m
i i
0
23
^
0)
o
H
U-4
O
rH
0)
o>
0
"
I 1
O
85
01
o
H
U-l
o
rH
0)
£>
01
rJ
g
O
OO
A-15
-------
The vertical potential temperature gradients will be calculated
from the onsite tower temperature measurements using the general expression
(T9(°K) - T (°K))
+ 0.01 (4)
Table 4 identifies T?, T and (z_-z ) for the four choices of temperature
measurements shown in Table 2.
As noted by the sixth footnote at the bottom of Table 2, we will
use climatological values of the vertical and/or lateral turbulent inten-
sities for the hours with no onsite turbulence measurements. We analyzed
the wind-speed and turbulence data from the 30-meter level of Tower No. 2
during the second year of the Westvaco monitoring program to determine
median turbulent intensities for each combination of season (winter, spring,
summer and fall), wind-speed category and time-of-day category. The seasons
were defined in the conventional sense for dispersion modeling. For example,
winter was comprised of December, January and February. Time of day was
based on sunrise and sunset and was defined as follows:
Morning - Sunrise plus 1 hour to sunrise plus 5 hours
Afternoon - Sunrise plus 5 hours to sunset minus 1 hour
Evening - Sunset minus 1 hour to sunset plus 2 hours
e Night - Sunset plus 2 hours to sunrise plus 1 hour
The resulting median lateral and vertical turbulent intensities are listed
in Tables 5 and 6, respectively.
We point out that, if the wind during an hour is calm or variable
at all levels of Tower No. 1 and Tower No. 2, we will define the calculated
A-16
-------
W
CJ
H
W
W
H M
< H
.J Pi
CJ O
i-3 CO
CJ PS
H
O M
H S
O W
W CO
co CJ
CO O
> CO
Pi E-i
w a
H W
M Q
H 2
PS O
O
M W
W Pi
SS C3
H
Q "^
< W
Pn
CO S
[V] [-}
Pi H
EI h*3
W H
P-< a
S w
W H
EH O
PH
Pi
M
H
a
w
PP
. 1
N
1
CXI
N
t-H /-x
CO g
£> ^-^
(-1
0)
4J
d
( I
,e
6C
H
H-t
C -^
H <1"
v-'
H C
O
14-1 -H
O 4-1
CO
Ol 3
o cr
Vi W
0
CO
d /-N
H -3"
x-^
CSI
H d
o
O 4-1
CO
01 3
o cr
}-i W
o
CO
0)
CJ
o
CO
4-1 01
O >-i
3
C CO
CO J-i
Pi 0)
e
HI
H
CXI
CXI
CXI
CN
0
a
}_l
0)
o
H
4-4
o
i-H
O)
O!
hJ
e
o
r 1
1 1
0
a
^1
01
o
H
4-1
O
t-l
O)
01
I-J
e
o
o
1 1
CX|
ro
i i
O
a
}_4
0)
o
H
M-4
o
i i
0)
5)
hJ
e
0
r 4
^-1
0
M
Ol
O
H
4-1
O
i 1
QJ
0)
i-J
e
o
, 1
0
OS
. 1
0
a
M
01
O
H
4-1
O
t-H
O)
01
i-J
E3
0
i i
i i
0
a
J_l
01
o
H
4-1
O
i 1
Ol
Ol
i-J
E
o
o
1 1
o
cxi
CXI
O
a
t-i
01
o
H
M-l
o
.H
0)
01
I-J
e
o
r{
0
a
M
0)
§
H
4-1
0
rH
0)
O)
hJ
e
o
CO
i cxi m
-------
TABLE 5
MEDIAN HOURLY LATERAL TURBULENT INTENSITIES AT THE 30-METER LEVEL
OF TOWER NO. 2 DURING THE SECOND YEAR OF THE
WESTVACO MONITORING PROGRAM
Time
of
Day
Wind Speed (m/sec)
0-1.5
1.6-3.1
3.2-5.1
5.2-8.2
(a) Winter
Night
Morning
Afternoon
Evening
0.45
0.40
0.50
0.55
0.15
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
8.3-10.8
0.25
0.25
0.25
0.25
10.8
0.25
0.25
0.25
0.25
(b) Spring
Night
Morning
Afternoon
Evening
0.45
0.60
0.70
0.55
0.25
0.25
0.45
0.35
0.15
0.25
0.35
0.25
0.25
0.25
0.30
0.25
0.25
0.25
0.25
0.25
A
*
0.25
*
(c) Summer
Night
Morning
Afternoon
Evening
0.45
0.55
0.70
0.60
0.15
0.25
0.35
0.25
0.15
0.25
0.35
0.15
0.20
0.25
0.25
0.25
0.25
0.25
0.25
*
*
*
*
*
(d) Fall
Night
Morning
Afternoon
Evening
0.40
0.45
0.65
0.55
0.15
0.25
0.35
0.25
0.15
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
*
0.25
0.25
*
No observations.
A-18
-------
TABLE 6
MEDIAN HOURLY VERTICAL TURBULENT INTENSITIES AT THE 30-METER LEVEL
OF TOWER NO. 2 DURING THE SECOND YEAR OF THE
WESTVACO MONITORING PROGRAM
Time
of
Day
Wind Speed (m/sec)
0-1.5
1.6-3.1
3.2-5.1
5.2-8.2
(a) Winter
Night
Morning
Afternoon
Evening
0.05
0.15
0.25
0.10
0.05
0.15
0.15
0.15
0.15
0.15
0.15
0.15
(b) Spring
Night
Morning
Afternoon
Evening
0.05
0.30
0.35
0.15
0.05
0.15
0.15
0.15
0.10
0.15
0.15
0.15
0.15
0.15
0.15
0.15
8.3-10.8
10.8
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
*
A
0.15
*
(c) Summer
Night
Morning
Afternoon
Evening
0.05
0.20
0.40
0.15
0.05
0.15
0.15
0.05
0.05
0.15
0.15
0.05
0.15
0.15
0.15
0.15
0.15
0.15
0.15
A
A
A
A
A
(d) Fall
Night
Morning
Afternoon
Evening
0.05
0.15
0.25
0.05
0.05
0.15
0.15
0.05
0.05
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.30
A
0.15
0.15
A
No observations.
A-19
-------
ground-level concentrations for the hour as missing. Also, a calculated
3-hour concentration must contain 3 hours of non-missing calculated hourly
concentrations to be considered, while a 24-hour concentration must contain
18 hours of non-missing calculated hourly concentrations to be considered.
For example, a 24-hour period with 3 calm (missing) hours will have a "24-
hour average concentration" defined by the 21-hour average concentration
for the 21 hours of non-calm winds. We will use the same procedure for
hours with missing concentration measurements to obtain the highest observed
3-hour and 24-hour average concentrations. Calm winds during hours with
valid concentration measurements will have no effect on our determination
of the highest observed 1-hour, 3-hour and 24-hour average concentrations.
Source Inputs
Table 7 gives the stack height, stack radius, Universal Transverse
Mercator (UTM) X and Y coordinates and stack base elevation for the Westvaco
Main Stack. The other critical SHORTZ hourly source input parameters are
the S0_ emission rate, the actual volumetric emission rate and the stack
exit temperature. The Westvaco data set includes hourly measurements of
the stack exit temperature, the S0_ emission rate in tons per hour and the
in-stack SO concentration in parts per million. We will use these para-
meters to compute the actual volumetric emission rate (stack gas flow rate)
from the expression
322.9 Q (ton/hr) T(°K)
-
g
V0»3/sec) - - S-- - (5)
where 'SO is the S0« emission rate, T is the stack exit temperature and
*L j^ S
SO is the in-stack SO concentration. If an hourly S0_ emission rate,
volumetric emission rate or stack exit temperature is missing, we will use
the last reported value.
A-20
-------
TABLE 7
WESTVACO STACK PARAMETERS
Parameter
Parameter Value
Stack Height (m)
Stack Radius (m)
Stack Coordinates
UTM X (m)
UTM Y (m)
Stack Base Elevation (m MSL)
189.7
1.68
667,091
4,370,759
288
A-21
-------
Receptor and Other Inputs
Table 8 gives the UTM X and Y coordinates and elevations of the
nine air quality monitors to be considered in the model performance evalu-
ation. Although not shown in Table 7, the hourly concentration measurements
from Monitor 2 (Luke Hill) and Monitor 11 (Bloomington) in addition to the
hourly concentration measurements from the monitoring sites identified in
Table 7 will be used to estimate background concentrations (see Section 3).
SHORTZ is a highly generalized dispersion model with a large
number of input and output options. Although detailed SHORTZ user's
instructions are given by Bjorklund and Bowers (1982), we believe that is
is important to identify the values of some of the critical program control
parameters and model constants for use in the Westvaco model evaluation.
As shown by Table 9, SHORTZ will in general be executed in its default
mode. Because the SO emission rate will be entered in grams per second,
the units conversion factor TK will be set equal to 381.68 to obtain concen-
trations in parts per million. (The default value of TK is 10 for concen-
trations in micrograms per cubic meter.) The elevation above mean sea
level of Tower No. 2 of 468 meters is defined in Table 9 as the weather
station elevation. The wind measurement height ZR is 30 meters, the height
above ground level of the upper level of Tower No. 2.
5. MEASURES OF MODEL PERFORMANCE
To the best of our knowledge, the only study to date which fully
addresses the proposed AMS Measure of Dispersion Model Performance is the
study by Londergan, et al. (1982), who note that (p. 64):
One conclusion is apparent from even a cursory inspection of the
... tables. The volume of model performance statistics which was
generated in this study is excessive. The amount of effort
required to analyze fully the information contained in these
tables is prohibitive. After a limited review, it is also appar-
ent that many of the statistics are relatively uninformative,
A-22
-------
TABLE 8
UNIVERSAL TRANSVERSE MERCATOR (UTM) X AND Y COORDINATES
AND ELEVATIONS ABOVE MEAN SEA LEVEL (MSL) OF THE
AIR QUALITY MONITORING SITES
Site
1
3
4
5
6
7
8
9
10 (Stony Run)
Coordinates
UTM X (m)
667,800
667,638
667,639
667,576
667,860
667,320
667,090
667,412
669,766
UTM Y (m)
4,370,360
4,370,259
4,370,060
4,369,729
4,370,604
4,369,780
4,369,277
4,369,278
4,372,851
Ground Elevation
(m MSL)
604
564
603
639
591
637
643
673
504
A-23
-------
TABLE 9
SHORTZ PROGRAM CONTROL PARAMETERS AND MODEL CONSTANTS FOR USE
IN THE WESTVACO MODEL EVALUATION STUDY
SHORTZ Input
Parameter
Meaning of Parameter
Parameter Value for the
Westvaco Model Evaluation
Study
ISW (7)
ISW (9)
TK
ZR
HA
GAMMA 1
GAMMA 2
XRY
ALPHA
IDECAY
ROTATE
Define as "1" if Terrain Eleva-
tions Are Used
Define as "1" for Wind Speed a
Function of Height AGL Rather
Than Height MSL
Units Conversion Factor
Wind Speed Measurement Height
| Above HA (m)
Elevation above MSL of Weather
Station
i
j Adiabatic Plume Rise Entrainment
i Coefficient
Stable Plume Rise Entrainment
Coefficient
Distance Over Which Rectilinear
Lateral Expansion Occurs (m)
Lateral Diffusion Coeffient
Exponential Decay Coefficient
(sec )
Angular Displacement of Receptor
Grid from True North
1
0 (Default Value)
381.68 (Concentration
in ppm)
30
467.57
0.60 (Default Value)
0.66 (Default Value)
50 (Default Value)
0.9 (Default Value)
0 (Default Value)
0 (Default Value)
A-24
-------
repetitious, and redundant. It is not very productive to demon-
strate, eight times over, many of the general performance charac-
teristics... In an effort to follow the AMS workshop recommen-
dations as closely as possible, TRC and EPA elected to implement
the full list of performance measures for all the data sets and
subsets specified. A thorough review of this final report is
warranted, with the goal of setting priorities and evaluating the
usefulness of various measures, in order to provide greater
flexibility and better focus for future model evaluation exercises.
In view of these comments and our experience to date with the AMS Measures
of Performance (Bowers, 1982; Bowers, et_ _al_. , 1982), this protocol considers
only a meaningful subset of the AMS Measures of Performance.
Of the numerous AMS Measures of Performance, this protocol uses
two parametric measures. The first measure of performance is the bias
(average difference between observed and calculated concentrations), which
is defined as
~7 1
Axi = xoi - xci
where x . is the i observed concentration, x is the i calculated
concentration and N is the number of paired observed and calculated concen-
trations. The second measure of performance is a measure of the "noise" in
the results of the model calculations and is provided by the variance of
the differences
N
-i- V
N-l Z__t
A-25
(8)
-------
The August 1981 draft EPA report "Interim Procedures for Evalua-
ting Air Quality Models" recommends that a model performance evaluation
plan be developed in advance of any model testing. This plan assigns to
the various measures of model performance specific numerical values (points)
which are dependent on the objectives of the model calculations. For exam-
ple, a first-order objective might be to determine which model best predicts
the highest concentrations that are required for regulatory decision making
and a second-order objective might be to determine which model best predicts
total concentration fields. This model evaluation protocol has a single
objective, the determination of the best model to be used to establish an
SO. emission limitation for the Westvaco Mill. As noted in Section 2, the
Westvaco data set is heavily weighted by concentrations on Westvaco property
at a downwind distance from the Main Stack of about 1 kilometer (eight out
of nine monitors). Monitor 10 (Stony Run) is the only monitor at the typ-
ical downwind distance of the Westvaco property boundaries. Because it is
our understanding that the S09 concentrations calculated at and beyond the
Westvaco property boundaries are of primary concern for regulatory purposes,
comparisons of concentrations paired in space at Monitor 10 are assigned
possible points in this protocol that are four times the corresponding
possible points assigned to each of the eight other monitors.
Table 10 gives the pairings of observed and calculated concentra-
tions, measures of performance and the points to be assigned to the various
comparisons of observed and calculated concentrations. The score for each
pairing of maximum and second-highest observed and calculated concentrations
is based on the absolute value of the residual JAxj and is given by
{Score Model.} = l|X|min ^ MIN{C_/0 ,0/C.} x {Possible Points} (9)
I A. -I I
A-26
-------
w
KJ
PQ
cu en
r 1 4-1
H -H
CO O
en CM
o
PH
raging
Time
cu
cu
4-1 O
o c
cfl
cu B
t-l M
3 0
CO U-l
CO (-1
cu cu
S P-i
cu
C -H
I i E i
n-l
cu
H CU
CO O
P-i co
CL
en
CO
C
o
H
4J
S-i
1 t
c
c
o
c
0
u
CM CM
0 0
4-1 4J
O O -H -H O O
CN CM C (3 CN CM
O O
t^H **-1
^t f£*
*" **>^
CN CN
CO CO CO CO i 1 r-l
V4 £-4 V-i M CO CO
33 33 3 3
o o o o a e
p^ |"^ p"] p^ p^ (^
<; <
co -i
033 033
33 0 0 33 0 0
33 33 33 33
i i i t
co *J" CO >^
CM CM
cu
u
en is
co co
H -H
PQ S-i
CO
*2>
o
z
o
*:$
4-1
CO
cu
n
60
H
33
in
CM
A-27
-------
a
cu
3
d
d
o
O
iH
w
0) CO
,__ | It
H~l 4-J
42 d
H «H
CD O
cn p..
0
PL,
M
d
H 01
M E
CO -H
S-i H
01
>
5
CU
U-l U
o d
CO
a) B
S-i S-i
3 0
CO U-l
CO S-i
0) CU
S PM
cu
d -H
rH H
a
cu
S-i
H CU
to o
PH CO
CX
CO
CO
d
o
H
CO
S-J
4-1
d
cu
o
d
o
o
co ro J- J- J-
CM CM CM cM CM CM
(^ ^ M >-i t-i )-i
000 000
H -H -H -H -H -H
add d d d
o o o o o o
s s s s s s
LTl u-i LT> CN CNI CN
M CO CO (-1 CO CO
3 ^ M 3 j-, )_,
O 3 3 O 3 3
WOO P2 O O
PC K 32 pa
i i I-H
m -a- ro
O
"Z
co
cu
><
tiplied by: (1) the ratio of the absolute value of the
absolute value of the residual for the model being
rH CU
3 42
B 4J
CO O
4-1 4-1
d
H rH
O CU
Cu 13
o
cu B
rH
43 l-i
H CU
CD 42
CO 4-1
O -H
Cu CU
CU S-I
42 O
4-J M <
J>s,
43 X
d
0)
> rH
H Cfl
bfi 3
T3
cn -H
H CD
cu
CU S-i
S-i
o B
0 3
CD B
H
cu d
42 -H
H E
of calculated and observed concentrations (C/0) and
ations (0/C) for the model being evaluated. The score
£-1
O 4-1
H d
4-1 CU
cfl o
s-i d
o
cu o
42
4J T3
Cl)
M-l 4-1
O CO
rH
E 3
3 a
B rH
H cfl
d o
H
G *"O
d
CU CO
42
4-1 TD
CU
£>
s~*. S-i
CN CU
^ - cn
43
T3 O
d
co m
o
*
13 O
CU -H
4-1 4-1
CO cfl
3 1-4
rH
CO CU
> 42
CU 4-J
d on Westvaco property at approximately the same distance
ints for the more distant and off-property Monitor 10 are
.
j_j
cu
60
CU
4-1
d
H
4-J
CD
CU
S-I
CO
CU
d
CU
42
4-1
O
4->
-a
cu
rrj
d
3
o
S-i
CD
H
CU
4-1
Cfl
O
o
rH
CU
}-l
CO
cn
S-i
o
4-1
H
C
O
e
01
d
H
c
cu
42
4->
IH
O
4-1
42
60
H
cu
a)
CO
3
cfl
a
OJ
PQ
O
Cu
CU
rH
43
H
CD
CD
0
Cu
CU
42
4-1
rt
^
CJ
cfl
4-J
C/J.
d
H
CO
S
0
O
cfl
£>
4-1
CO
CU
£2
CU
42
4-1
E
o
j-t
IJ-J
cn
S-i
0
4-1
-H
d
0
e
j-i
cu
42
4-1
O
CU
42
4-1
J-I
0
UH
d
£5
O
42
cn
cu
CO
O
42
4-1
imum and second-highest concentrations except that absolute
values of residuals and average concentrations are sub-
X
cfl
e
cu
42
4-1
}_l
0
cn
cfl
cu
E
CO
cn
cu
42
4-1
cn
H
S
cu
4-1
CD
J>^
CO
01
4-1
3
rH
O
CO
43
CO
M
O
UH
TD
CU
4J
3
4J
H
4-1
CO
43
3
CO
CU
S-i
CO
CO
CU
CO
cfl
H
6043
cn
CU
E
H
4J
S-i
3
o
M-4
d
H
£_i
0
CJ
CO
cu
42
E-.
^4-1
0
cn
cu
3
rH
CO
>
trations.
G
0)
o
d
0
CJ
4-1
CO
CU
42
60
H
42
1
T3-
d
o
CJ
CU
CD
S-l
O
B
3
B
H
X
cfl
B
£_4
O
4-1
Tj
CU
4-J
3
4-J
H
4J
CO
tiplied by the minimum of the ratio of the variances for
d the ratio of the variances for the calculated and
rH
3
E
CO
4-1
d
H
o
Cu
cu
rH
43
H
cn
CD
O
G
cfl
CD
d
O
H
4-1
CO
S-i
4-1
d
Q)
O
d
o
CJ
Cu TJ
CU
42
4-1
£>>
43
d
cu
[>
H
60
CO
H
0)
}-l
O
o
CO
CU
42
H
cu
4-1
cfl
rH
3
O
rH
Cfl
CJ
*"O
c
cci
13
01
£>
S-4
CU
CO
0
0)
42
4-1
.
to
d
o
H
4-1
cO
S-l
4-1
C
CU
CJ
d
o
a
T3
cu
e
CU
co
43
O
A-28
-------
where
AX min
MIN{C./0,0/C.}
the absolute value of the minimum residual for either model
th
model
the absolute value of the residual for the i
the minimum for the i model of the ratio of calculated
and observed concentrations (C/0) and the ratio of
observed and calculated concentrations (0/C)
The score calculated from Equation (9) is rounded to the nearest integer.
It follows from Equation (9) that the only model with the potential to be
awarded all of the possible points is the model with the minimum residual.
The first of the scores for each pairing of the 25 highest observed and
calculated concentrations is also given by Equation (9) with biases substi-
tuted for residuals and average concentrations substituted for maximum or
second-highest concentrations. The second of the scores for each pairing
of the 25 highest observed and calculated short-term concentrations is
given by
{Score Model.} = MIN{o" .2/a 2,
i ci o '
2 2
a /a . } x {Possible Points}
o ci
(10)
where
a . = the variance of the calculated concentrations
ci ,. , .tn , .
for the i model
a = the variance of the observed concentrations
o
The score calculated from Equation (10) is also rounded to the nearest
integer.
A-29
-------
Table 11 shows the computation of the maximum possible points
from Table 10 and Table 12 gives the allocation of these points by model
performance attribute. Of the 602 possible points overall, 260 points
(about 43 percent) are assigned to the concentration comparisons of concern
for regulatory decision makers (the maximum annual and short-term average
concentrations and the second-highest short-term average concentrations).
Of the 372 possible points overall for the maximum, second-highest and 25
highest short-term concentrations paired in space only, 124 points are
assigned to Monitor 10 because it is the only monitor at the typical dis-
tance of concern for setting an SO emission limitation. We have assigned
75 percent of the 342 available points for the 25 highest concentrations
paired in space only and unpaired in space or time to absence of bias
because we consider absence of bias in dispersion model predictions to be
of critical importance. More points are allocated to the two measures of
performance for the 25 highest short-term concentrations paired in space
only than to the performance measures for the 25 highest short-term concen-
trations unpaired in space or time because we believe that the comparisons
of the highest concentrations paired in space are scientifically more signi-
ficant than the comparisons of the the highest unpaired concentrations.
The model to be used in the dispersion model calculations to set an SO
emission limitation for the Westvaco Mill will be the model with the highest
score.
A-30
-------
W
CD
d
H
O
P-i
CU
i-H
H
CD
CO
O
P-i
iw
O
d
0
rH
4J
cfl
U
^
Q^
B
o
O
14-J 0)
O CJ
d
O) cO
r-l g
3 r-i
CO O
CO <4-l
CU M
S cu
0)
e
H
H
C
M
13
CU
r-l CU
M a
cO to
P-I P]
C/3
CO
d
o
4-1
CO
}_i
1 t
C
CU
CJ
d
o
u
CD
4J
CD d
4-1 -H
d 0
H P,
O
p , \^
t-H
CN
CO II
II 0
r 1
CD
CO r-l r-l
4-1 O O CD CO
d 4-J 4J 4J 4J
H -H -H d d
o d d -H -H
p. o o o o
B S P. CX
o
4 ^>-t
E
3
B
H
X
CO
^
CO
4-1
co '.< d
4J T-)
d o
H p,
O
Cu ^"
CN
oo
4-> 4J
H -H -H -H -H -H
o o d d d d
P, p< O O O O
B S E S
m in
r-~ r-H OO rJ 00 P
0 0
II II X 4-1 X <4-l
CD CO CO CD CD CO
0) 01 0) CU 0) 0)
B B B B B B
H -H -H -H -H -H
60 60 60 60 60 60
d d d d d d
H -H -H -H -r-l -r-t
60 60 60 60 60 60
cO CO CO CO CO cO
cu o) cu cu cu cu
cO CO CO cO (0 CO
CO CO CO CO CO CO
X X X X X X
CO CD CD CD CO CD
d d d d d d
H T-l iH -H -H -H
o o o o o o
p, Cu Cu P, p, Cu
in in m o CN oo
CM CM
cu a)
o a
CD d CD d
cO CO cO cO
H vH -r-l -H
pg J_j pQ )_!
cfl CO
> >
O O
^jr- ^T-
O CO
la cu
r*-1
4-1
CD
0)
r-*
60
H
X
in
CM
CM
O
vO
CO
4-1
d
H
0
P-i
cu
O
rH
CD
CO
0
pu.
3
B
H
X
CO
S
A-31
-------
TABLE
ALLOCATION OF POSSIBLE POINTS ACCORDING TO
MODEL PERFORMANCE ATTRIBUTE
Model Performance Attribute
Ability to Predict Maximum 3-Hour and 24-Hour
Average Concentrations Unpaired in Space or
Time
Ability to Predict Maximum 3-Hour and 24-Hour
Average Concentrations Paired in Space Only
Ability to Predict Maximum Annual Average
Concentrations Paired in Space and Time
or in Time Only
Ability to Predict Second-Highest 3-Hour and
24-Hour Average Concentrations Unpaired in
Space or Time
Ability to Predict Second-Highest 3-Hour and
24-Hour Average Concentrations Paired in
Space Only
Absence of Bias in Predicting the 25 Highest
Short-Term Concentrations Unpaired in Space
or Time
Absence of Bias in Predicting the 25 Highest
Short-Term Concentrations Paired in Space
Only
Variances of 25 Highest Observed and Calcu-
lated Short-Term Concentrations Unpaired in
Space or Time Do Not Differ
Variances of 25 Highest Observed and Calcu-
lated Short-Term Concentrations Paired in
Space Only Do Not Differ
Total
Possible
Points
40
48
40
60
72
75
180
15
72
602
Percent of Maximum
Possible Points
7
8
7
10
12
12
30
2
12
100
A-32
-------
REFERENCES
Bjorklund, J. R. and J. F. Bowers, 1982: User's instructions for the SHORTZ
and LONGZ computer programs. EPA Reports EPA-903/9-82-004a and
004b (in publication), U. S. Environmental Protection Agency,
Region III, Philadelphia, PA.
Bowers, J. F., 1982: Scientific review of the ten rural dispersion models
under consideration by the U. S. Environmental Protection Agency
for possible inclusion in the next Guideline on -Air Quality Models,
Paper prepared for the AMS Steering Committee for the cooperative
agreement between the American Meteorological Society and the U.
S. Environmental Protection Agency.
Bowers, J. F. , A. J. Anderson and W. R. Hargraves, 1982: Tests of the
Industrial Source Complex (ISC) Dispersion Model at the Armco
Middletown, Ohio Steel Mill. EPA Report No. EPA-450/4-82-006
prepared for U. S. Environmental Protection Agency, Research
Triangle Park, NC.
Cramer, H. E., et^ ad., 1972: Development of dosage models and concepts.
Final~~Report under Contract DAAD09-67-C-0020(R) with the U. S.
Army, Deseret Test Center Report DTC-TR-72-609, Fort Douglas,
UT.
Cramer, H. E., H. V. Geary and J. F. Bowers, 1975: Diffusion-model
calculations of long-term and short-term ground-level SO
concentrations in Allegheny County, Pennsylvania. EPA Report
903/9-75-018 (NTIS Accession No. PB 245262/AS), U. S.
Environmental Protection Agency, Region III, Philadelphia, PA.
Cramer, H. E., J. F. Bowers and H. V. Geary, 1976: Assessment of the air
quality impact of SO- emissions from the ASARCO-Tacoma smelter.
EPA Report No. EPA 910/9-76-028, U. S. Environmental Protection
Agency, Region X, Seattle, WA.
Cramer Company, H. E., 1981: Westvaco Luke, Maryland monitoring program:
Data analysis and dispersion model evaluation (first two quarters),
H. E. Cramer Company, Inc. Technical Report TR-81-202-01 prepared
for the U. S. Environmental Protection Agency under subcontract
to Research Triangle Institute, Research Triangle Park, NC.
Environmental Protection Agency, 1978: Guideline on air quality models.
EPA Report No. EPA-450/2-78-027, OAQPS No. 1.2-080, U. S.
Environmental Protection Agency, Research Triangle Park, NC.
Fox, D. G., 1981: Judging air quality model performance: A summary of the
AMS Workshop on Dispersion Model Performance. Bulletin American
Meteorological Society, 62(5), 599-609.
A-33
-------
REFERENCES (Continued)
Hanna, S. , _et_ _a!L. , 1982: Diffusion model development and evaluation and
emission limitations at the Westvaco Luke Mill. Document PA439,
Environmental Research & Technology, Inc., Concord, MA.
Londergan, R. J., et^ juL. , 1982: Evaluation of rural air quality simulation
models. TRC Project 1713-R80 prepared for U. S. Environmental
Protection Agency, Research Triangle Park, NC.
A-34
-------
APPENDIX>B
ANALYSIS OF OBSERVED HOURLY SO CONCENTRATIONS
This appendix presents the results of the H. E. Cramer Company's
analysis of the observed hourly S0? concentrations reported by Environmental
P,esearch & Technology, Inc. (ERT) to the U. S. Environmental Protection
Agency (EPA) Region III for the second year of the Westvaco monitoring
program. The 25 highest short-term (1-hour, 3-hour and 24-hour average)
observed (minus background) concentrations at Monitors 1, 3, 4, 5, 6, 7, 8,
9 and 10 are listed in Tables B-l through B-9. See Appendix A for a discus-
sion of the procedures used to calculate hourly background concentrations
(concentrations attributable to emissions from sources other than the West-
vaco Main Stack). The observed (with background) and observed (minus back-
ground) annual average S0_ concentrations at the nine monitors of concern
for the model evaluation are given in Table B-10. Based on Tables B-l
through B-9, the 25 highest 1-hour, 3-hour and 24-hour average observed
(minus background) S09 concentrations at all of the nine monitors of
concern are respectively given in Tables B-ll, B-12 and B-13.
Br-1
-------
TABLE B-l
TWENTY-FIVE HIGHEST OBSERVED (MINUS BACKGROUND) SHORT-TERM S02 CONCENTRATIONS
AT MONITOR 1 DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
1-Hour Average
Concentrations
Date
(Hour)
22 Oct 81 (07)
13 Jan 81 (04)
13 Jan 81 (02)
19 Nov 81 (08)
13 Jan 81 (03)
22 Oct 81 (08)
21 Oct 81 (09)
03 Apr 81 (08)
08 Apr 81 (06)
06 Jul 81 (20)
08 Feb 81 (01)
09 Nov 81 (06)
19 Nov 81 (09)
16 Feb 81 (01)
26 Nov 81 (24)
08 Apr 81 (08)
25 Jan 81 (08)
13 Jan 81 (05)
19 Feb 81 (09)
21 Aug 81 (08)
04 Nov 81 (11)
09 Nov 81 (08)
05 Oct 81 (03)
27 Nov 81 (01)
04 Dec 81 (04)
Concentration
(ppm)
0.7285
0.7160
0.6835
0.6780
0.6130
0.6085
0.5535
0.5155
0.5095
0.5065
0.4955
0.4935
0.4760
0.4535
0.4485
0.4455
0.4410
0.4355
0.4355
0.4345
0.4240
0.4135
0.4115
0.4090
0.4090
3-Hour Average
Concentrations
Date
(Period)
22 Oct 81 (3)
19 Nov 81 (3)
13 Jan 81 (1)
13 Jan 81 (2)
06 Jul 81 (7)
09 Nov 81 (2)
12 Nov 81 (8)
08 Feb 81 (1)
09 Nov 81 (3)
05 Dec 80 (8)
08 Apr 81 (3)
25 Jan 81 (3)
09 Nov 81 (1)
22 Apr 81 (3)
03 Apr 81 (3)
08 Apr 81 (2)
19 Feb 81 (4)
21 Aug 81 (3)
13 Jan 81 (3)
21 Oct 81 (3)
03 Apr 81 (2)
29 Dec 80 (1)
07 May 81 (2)
29 Dec 80 (4)
27 Nov 81 (1)
Concentration
(ppm)
0.5412
0.5158
0.4620
0.4612
0.3768
0.3528
0.3125
0.3082
0.3065
0.2992
0.2982
0.2890
0.2850
0.2675
0.2672
0.2655
0.2460
0.2385
0.2380
0.2298
0.2135
0.2117
0.2088
0.2070
0.2045
24-Hour Average
Concentrations
Date
13 Jan 81
29 Dec 80
09 Nov 81
22 Oct 81
05 Dec 80
19 Nov 81
04 Dec 80
06 Jul 81
03 Apr 81
21 Aug 81
08 Feb 81
07 May 81
12 Nov 81
25 Jan 81
19 Feb 81
26 Feb 81
26 Nov 81
20 Oct 81
16 Feb 81
21 Oct 81
14 Feb 81
26 Sep 81
12 Jun 81
18 Aug 81
16 Dec 80
Concentration
(ppm)
0.1666
0.1375
0.1330
0.1113
0.0974
0.0863
0.0849
0.0847
0.0779
0.0681
0.0665
0.0661
0.0605
0.0589
0.0565
0.0558
0.0537
0.0524
0.0516
0.0505
0.0494
0.0483
0.0481
0.0469
0.0467
B-2
-------
TABLE B-2
TWENTY-FIVE HIGHEST OBSERVED (MINUS BACKGROUND) SHORT-TERM S02 CONCENTRATIONS
AT MONITOR 3 DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
01
02
03
04
05
06
07
08
09
10
11
12
13
14 '
15
16
17
18
19
20
21
2.2
23
24
25
1-Hour Average
Concentrations
Date
(Hour)
13 Jan 81 (02)
13 Jan 81 (03)
22 Oct 81 (08)
19 Nov 81 (09)
13 Nov 81 (21)
06 Jan 81 (04)
06 Jan 81 (03)
20 Oct 81 (24)
03 Apr 81 (08)
22 Oct 81 (07)
26 Aug 81 (09)
13 Jan 81 (04)
01 Apr 81 (11)
08 Feb 81 (01)
19 Nov 81 (08)
09 Nov 81 (02)
14 Feb 81 (21)
14 Nov 81 (12)
11 Apr 81 (04)
12 Jun 81 (08)
18 Aug 81 (09)
26 Nov 81 (24)
04 Nov 81 (11)
28 Jun 81 (09)
20 Oct 81 (23)
Concentration
(ppm)
0.6475
0.6110
0.5735
0.5580
0.5375
0.5300
0.4855
0.4735
0.4705
0.4665
0.4555
0.4150
0.3895
0.3855
0.3810
0.3725
0.3660
0.3655
0.3595
0.3575
0.3535
0.3505
0.3450
0.3415
0.3345
3-Hour Average
Concentrations
Date
(Period)
13 Jan 81 (1)
22 Oct 81 (3)
19 Nov 81 (3)
13 Jan 81 (2)
06 Jan 81 (2)
20 Oct 81 (8)
06 Jan 81 (1)
13 Nov 81 (7)
06 Jan 81 (8)
09 Nov 81 (2)
01 Apr 81 (4)
09 Nov 81 (1)
08 Apr 81 (3)
11 Apr 81 (2)
08 Feb 81 (1)
14 Nov 81 (4)
22 Apr 81 (3)
06 Jan 81 (5)
12 Nov 81 (8)
21 Aug 81 (3)
14 Nov 81 (2)
13 Nov 81 (2)
27 Nov 81 (1)
03 Apr 81 (3)
26 Aug 81 (3)
Concentration
(ppm)
0.4337
0.3842
0.3478
0.3085
0.2890
0.2775
0.2702
0.2575
0.2490
0.2302
0.2297
0.2288
0.2122
0.2105
0.2068
0.2038
0.1945
0.1858
0.1858
0.1848
0.1818
0.1815
0.1778
0.1758
0.1722
24-Hour Average
Concentrations
Date
06 Jan 81
13 Jan 81
13 Nov 81
29 Dec 80
14 Nov 81
09 Nov 81
22 Oct 81
19 Nov 81
21 Aug 81
20 Oct 81
14 Feb 81
07 May 81
11 Apr 81
24 Mar 81
03 Apr 81
26 Nov 81
22 Apr 81
15 Oct 81
12 Jun 81
19 Feb 81
18 Aug 81
23 Dec 80
08 Feb 81
21 Oct 81
01 Apr 81
Concentration
(ppm)
0.1585
0.1309
0.1083
0.0968
0.0906
0.0884
0.0787
0.0681
0.0623
0.0539
0.0533
0.0498
0.0495
0.0480
0.0471
0.0468
0.0421
0.0407
0.0403
0.0401
0.0397
0.0392
0.0378
0.0375
0.0375
B-3
-------
TABLE B-3
TWENTY-FIVE HIGHEST OBSERVED (MINUS BACKGROUND) SHORT-TERM S02 CONCENTRATIONS
AT MONITOR 4 DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
01
02
03
04
05
06
07
08
09
10
11
12
13
14 '
15
16
17
18
19
20
21
22
23
24
25
1-Hour Average
Concentrations
Date
(Hour)
13 Jan 81 (03)
19 Feb 81 (10)
06 Jan 81 (03)
22 Apr 81 (08)
13 Jan 81 (02)
13 Jan 81 (05)
14 Feb 81 (21)
19 Feb 81 (11)
13 Nov 81 (04)
06 Jan 81 (23)
12 Jun 81 (08)
14 Nov 81 (12)
22 Oct 81 (08)
22 Oct 81 (07)
12 Nov 81 (23)
21 Aug 81 (04)
19 Feb 81 (09)
13 Jar. 81 (04)
13 Nov 81 (01)
19 Jun 81 (09)
21 Aug 81 (02)
15 Oct 81 (02)
21 Aug 81 (08)
06 Jan 81 (02)
08 Feb 81 (01)
Concentration
(ppm)
0.8740
0.6255
0.5215
0.5080
0.4795
0.4775
0.4510
0.4415
0.4285
0.4125
0.4035
0.3905
0.3895
0.3845
0.3615
0.3385
0.3375
0.3340
0.3235
0.3075
0.3075
0.3055
0.3045
0.3015
0.3015
3-Hour Average
Concentrations
Date
(Period)
13 Jan 81 (1)
19 Feb 81 (4)
13 Jan 81 (2)
22 Oct 81 (3)
06 Jan 81 (1)
22 Apr 81 (3)
12 Nov 81 (8)
06 Jan 81 (8)
13 Nov 81 (2)
21 Aug 81 (3)
13 Nov 81 (1)
14 Nov 81 (4)
08 Feb 81 (1)
14 Feb 81 (7)
04 Dec 80 (8)
21 Aug 81 (1)
13 Jan 81 (3)
12 Jun 81 (3)
16 Feb 81 (1)
08 May 81 (1)
15 Oct 81 (1)
20 Oct 81 (8)
21 Aug 81 (2)
13 Nov 81 (3)
01 Apr 81 (4)
Concentration
(ppm)
0.4560
0.3733
0.3528
0.3165
0.3005
0.2775
0.2722
0.2687
0.2675
0.2348
0.2318
0.2275
0.2158
0.2138
0.2032
0.2015
0.1850
0.1845
0.1768
0.1708
0.1688
0.1665
0.1658
0.1658
0.1587
24-Hour Average
Concentrations
Date
13 Jan 81
06 Jan 81
13 Nov 81
14 Nov 81
21 Aug 81
19 Feb 81
22 Apr 81
04 Dec 80
14 Feb 81
26 Nov 81
22 Oct 81
24 Mar 81
07 May 81
12 Nov 81
16 Feb 81
18 Aug 81
08 Feb 81
05 Dec 80
18 Mar 81
27 Dec 80
27 Feb 81
26 Sep 81
25 Jan 81
04 May 81
15 Oct 81
Concentration
(ppm)
0.1419
0.1147
0.1097
0.0805
0.0787
0.0703
0.0574
0.0571
0.0568
0.0511
0.0508
0.0498
0.0482
0.0472
0.0457
0.0423
0.0418
0.0418
0.0355
0.0350
0.0347
0.0347
0.0344
0.0343
0.0339
B-4
-------
TABLE B-4
TWENTY-FIVE HIGHEST OBSERVED (MINUS BACKGROUND) SHORT-TERM S02 CONCENTRATIONS
AT MONITOR 5 DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
2.2
23
24
25
1-Hour Average
Concentrations
Date
(Hour)
21 Aug 81 (03)
27 Nov 81 (01)
13 Nov 81 (02)
13 Nov 81 (05)
13 Nov 81 (06)
13 Jan 81 (03)
25 Jan 81 (24)
21 Sep 81 (02)
29 Dec 80 (01)
14 Feb 81 (22)
22 Apr 81 (07)
21 Sep 81 (03)
16 Feb 81 (01)
01 May 81 (04)
04 Dec 80 (20)
17 Jul 81 (21)
16 Feb 81 (02)
21 Aug 81 (02)
22 Apr 81 (08)
26 Sep 81 (06)
13 Nov 81 (04)
20 May 81 (08)
08 Apr 81 (05)
01 May 81 (03)
26 Nov 81 (01)
Concentration
(ppm)
0.8935
0.7750
0.7625
0.7585
0.7145
0.6550
0.6515
0.6365
0.5970
0.5770
0.5265
0.4945
0.4875
0.4735
0.4430
0.4415
0.4335
0.4305
0.4130
0.4035
0.3975
0.3835
0.3725
0.3685
0.3680
3-Hour Average
Concentrations
Date
(Period)
13 Nov 81 (2)
21 Aug 81 (1)
16 Feb 81 (1)
21 Sep 81 (1)
22 Apr 81 (3)
14 Feb 81 (8)
13 Nov 81 (1)
04 Dec 80 (7)
27 Nov 81 (1)
29 Dec 80 (1)
25 Jan 81 (8)
01 May 81 (1)
26 Sep 81 (2)
01 May 81 (2)
22 Dec 80 (2)
13 Jan 81 (1)
16 Dec 80 (3)
06 Feb 81 (1)
24 Mar 81 (2)
29 Dec 80 (6)
26 Nov 81 (1)
13 Jan 81 (4)
05 Aug 81 (8)
12 Jun 81 (3)
29 Dec 80 (3)
Concentration
(ppm)
0.6235
0.4518
0.4062
0.4038
0.3902
0.3597
0.2975
0.2743
0.2665
0.2577
0.2538
0.2382
0.2375
0.2252
0.2233
0.2227
0.2190
0.2095
0.1975
0.1863
0.1773
0.1717
0.1662
0.1582
0.1578
24-Hour Average
Concentrations
Date
13 Nov 81
29 Dec 80
16 Feb 81
13 Jan 81
21 Aug ,.81
01 May 81
22 Apr 81
14 Feb 81
21 Sep 81
17 Aug 81
04 Dec 80
24 Mar 81
26 Nov 81
22 Dec 80
26 Sep 81
17 Jul 81
25 Jan 81
14 Nov 81
16 Dec 80
20 May 81
19 Feb 81
06 Jan 81
09 Jan 81
08 Apr 81
27 Nov 81
Concentration
(ppm)
0.1546
0.1015
0.0828
0.0816
0.0813
0.0812
0.0760
0.0742
0.0684
0.0647
0.0608
0.0607
0.0522
0.0510
0.0499
0.0494
0.0471
0.0448
0.0425
0.0400
0.0391
0.0379
0.0374
0.0364
0.0356
B-5
-------
TABLE B-5
TWENTY-FIVE HIGHEST OBSERVED (MINUS BACKGROUND) SHORT-TERM S02 CONCENTRATIONS
AT MONITOR 6 DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
1-Hour Average
Concentrations
Date
(Hour)
13 Jan 81 (04)
19 Nov 81 (09)
08 Apr 81 (07)
08 Apr 81 (08)
08 Apr 81 (06)
16 Feb 81 (01)
08 Apr 81 (05)
04 Nov 81 (11)
13 Nov 81 (21)
05 Dec 80 (17)
29 Mar 81 (04)
05 Nov 81 (11)
27 Nov 81 (01)
15 Feb 81 (09)
19 Nov 81 (10)
19 Nov 81 (11)
03 Apr 81 (05)
26 Nov 81 (24)
29 Mar 81 (05)
28 Feb 81 (10)
05 Dec 80 (24)
13 Jan 81 (05)
14 Feb 81 (21)
13 Nov 81 (20)
24 May 81 (02)
Concentration
(ppm)
0.8660
0.8250
0.8165
0.7825
0.7585
0.7125
0.6915
0.6670
0.6325
0.5910
0.5545
0.4865
0.4730
0.4635
0.4510
0.4445
0.4415
0.4395
0.4365
0.4285
0.4245
0.4215
0.4210
0.4125
0.3970
3-Hour Average
Concentrations
Date
(Period)
08 Apr 81 (3)
08 Apr 81 (2)
13 Jan 81 (2)
29 Mar 81 (2)
19 Nov 81 (3)
13 Nov 81 (7)
04 Nov 81 (4)
03 Apr 81 (2)
19 Nov 81 (4)
03 Apr 81 (3)
27 Nov 81 (1)
05 Dec 80 (6)
05 Dec 80 (7)
05 Dec 80 (8)
16 Feb 81 (1)
17 Jan 81 (7)
15 Feb 81 (3)
13 Jan 81 (1)
28 Feb 81 (4)
14 Nov 81 (4)
09 Dec 80 (7)
13 Jan 81 (3)
04 Dec 80 (4)
05 Nov 81 (4)
02 Dec 80 (2)
Concentration
(ppm)
0.6142
0.5268
0.5135
0.4435
0.4228
0.3648
0.3382
0.3332
0.3048
0.2978
0.2942
0.2938
0.2752
0.2715
0.2672
0.2542
0.2528
0.2473
0.2455
0.2412
0.2375
0.2203
0.2132
0.2127
0.2122
24-Hour Average
Concentrations
Date
08 Apr 81
13 Jan 81
05 Dec 80
13 Nov 81
17 Jan .,81
02 Jan 81
14 Nov 81
29 Dec 80
03 Apr 81
14 Dec 80
19 Nov 81
07 Oct 81
28 Nov 81
16 Feb 81
27 Nov 81
13 Dec 80
10 Dec 80
28 Feb 81
29 Nov 81
17 Nov 81
04 Dec 80
12 Nov 81
18 Nov 81
16 May 81
20 Mar 81
Concentration
(ppm)
0.1501
0. 1432
0. 1299
0. L274
0.1257
0.1232
0.1219
0.1195
0.1191
0.1144
0.1082
0.1035
0.0983
0.0945
0.0897
0.0883
0.0862
0.0853
0.0831
0.0817
0.0799
0.0786
0.0774
0.0752
0.0746
B-6
-------
TABLE B-6
TWENTY-FIVE HIGHEST OBSERVED (MINUS BACKGROUND) SHORT-TERM S02 CONCENTRATIONS
AT MONITOR 7 DURING THE SECOND YEAR OF THE WESTVAGO MONITORING PROGRAM
Rank
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
1-Hour Average
Concentrations
Date
(Hour)
13 Nov 81 (05)
13 Nov 81 (02)
13 Nov 81 (06)
13 Nov 81 (01)
29 Dec 80 (01)
08 Feb 81 (01)
13 Nov 81 (04)
16 Feb 81 (01)
25 Jan 81 (24)
20 Aug 81 (24)
09 Jan 81 (04)
07 Feb 81 (24)
18 Aug 81 (23)
18 Oct 81 (05)
01 May 81 (04)
13 Mar 81 (03)
20 May 81 (08)
22 Apr 81 (08)
26 Sep 81 (06)
22 Apr 81 (07)
14 May 81 (08)
17 Jul 81 (21)
12 Nov 81 (23)
18 Aug 81 (22)
13 Jan 81 (03)
Concentration
(ppm)
0.8525
0.8475
0.7525
0.6685
0.6380
0.6355
0.6245
0.6035
0.5935
0.5705
0.5070
0.5040
0.4915
0.4875
0.4855
0.4565
0.4415
0.4390
0.4385
0.4325
0.4305
0.4285
0.4145
0.4135
0.4110
3-Hour Average
Concentrations
Date
(Period)
13 Nov 81 (2)
13 Nov 81 (1)
16 Feb 81 (1)
22 Apr 81 (3)
18 Aug 81 (8)
08 Feb 81 (1)
12 Nov 81 (8)
06 Jan 81 (8)
20 Aug 81 (8)
29 Dec 80 (1)
01 May 81 (2)
26 Sep 81 (2)
22 Apr 81 (2)
14 Feb 81 (8)
16 Dec 80 (3)
25 Jan 81 (8)
19 Aug 81 (2)
22 Mar 81 (3)
14 May 81 (3)
09 Jan 81 (2)
07 Feb 81 (8)
21 Mar 81 (8)
08 May 81 (1)
04 Dec 80 (7)
04 Dec 80 (8)
Concentration
(ppm)
0.7432
0.6575
0.3748
0.3712
0.3578
0.3362
0.3258
0.2690
0.2605
0.2600
0.2598
0.2588
0.2525
0.2340
0.2320
0.2248
0.2165
0.2058
0.2017
0.1987
0.1947
0.1892
0.1868
0.1867
0.1852
24-Hour Average
Concentrations
Date
29 Dec 80
22 Apr 81
12 Nov 81
26 Sep 81
01 May 81
16 Feb 81
14 Feb 81
18 Aug 81
04 Dec 80
20 Aug 81
13 Jan 81
30 Sep 81
08 Feb 81
09 Jan 81
17 Jul 81
06 Jan 81
07 May 81
19 Aug 81
24 Mar 81
28 Dec 80
20 May 81
18 Oct 81
22 Mar 81
17 Aug 81
16 Dec 80
Concentration
(ppm)
0.0994
0.0947
0.0707
0.0688
0.0686
0.0666
0.0626
0.0617
0.0589
0.0571
0.0556
0.0526
0.0505
0.0501
0.0477
0.0475
0.0461
0.0458
0.0444
0.0442
0.0438
0.0420
0.0419
0.0414
0.0406
B-7
-------
TABLE B-7
TWENTY-FIVE HIGHEST OBSERVED (MINUS BACKGROUND) SHORT-TERM S02 CONCENTRATIONS
AT MONITOR 8 DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
1-Hour Average
Concentrations
Date
(Hour)
13 Nov 81 (02^
13 Nov 81 (05)
27 Nov 81 (01)
20 Aug 81 (24)
21 Sep 81 (02)
04 Dec 80 (21)
01 May 81 (04)
13 Nov 81 (04)
14 Nov 81 (13)
08 May 81 (01)
28 Dec 80 (19)
29 Dec 80 (09)
01 May 81 (01)
22 Apr 81 (08)
01 May 81 (05)
30 Sep 81 (09)
14 Feb 81 (21)
13 Nov 81 (06)
04 Dec 80 (20)
21 Sep 81 (03)
13 Jan 81 (12)
09 Jan 81 (04)
03 Nov 81 (08)
18 Oct 81 (05)
29 Dec 80 (10)
Concentration
(ppm)
0.8435
0.8215
0.8070
0.6695
0.5185
0.4910
0.4565
0.4195
0.4135
0.3945
0.3705
0.3640
0.3525
0.3410
0.3225
0.3220
0.3210
0.3175
0.3150
0.3105
0.3070
0.3060
0.3045
0.3015
0.2970
3-Hour Average
Concentrations
Date
(Period)
13 Nov 81 (2)
13 Nov 81 (1)
01 May 81 (2)
21 Sep 81 (1)
04 Dec 80 (7)
01 May 81 (1)
27 Nov 81 (1)
20 Aug 81 (8)
22 Apr 81 (3)
21 Mar 81 (8)
22 Apr 81 (2)
29 Dec 80 (3)
28 Dec 80 (7)
14 Feb 81 (8)
14 Nov 81 (5)
30 Sep 81 (3)
06 Feb 81 (1)
19 Aug 81 (2)
24 Nov 81 (1)
13 Jan 81 (4)
29 Dec 80 (4)
04 Dec 80 (8)
20 Aug 81 (1)
08 May 81 (1)
09 Jan 81 (2)
Concentration
(ppm)
0.5195
0.3325
0.3012
0.2772
0.2743
0.2722
0.2702
0.2682
0.2625
0.1938
0.1905
0.1882
0.1848
0.1713
0.1695
0.1677
0.1605
0.1525
0.1398
0.1397
0.1377
0.1372
0.1362
0.1332
0.1297
24-Hour Average
Concentrations
Date
13 Nov 81
01 May 81
29 Dec 80
22 Apr 81
26 Nov ,,81
04 Dec 80
30 Sep 81
20 Aug 81
28 Dec 80
21 Sep 81
14 Feb 81
13 Jan 81
20 May 81
09 Jan 81
27 Nov 81
18 Oct 81
14 Nov 81
23 Nov 81
06 Jan 81
22 Dec 80
19 Aug 81
24 Mar 81
12 Nov 81
05 Dec 80
24 Nov 81
Concentration
(ppm)
0.1255
0.0820
0.0752
0.0656
0.0640
0.0639
0.0591
0.0540
0.0525
0.0509
0.0504
0.0463
0.0462
0.0435
0.0431
0.0400
0.0396
0.0346
0.0345
0.0338
0.0314
0.0310
0.0304
0.0303
0.0298
B
-------
TABLE B-8
TWENTY-FIVE HIGHEST OBSERVED (MINUS BACKGROUND) SHORT-TERM S02 CONCENTRATIONS
AT MONITOR 9 DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
01
02
03
04
05
06
07
08
09
10
11
12
13
14 '
15
16
17
18
19
20
21
22
23
24
25
1-Hour Average
Concentrations
Date
(Hour)
21 Aug 81 (03)
13 Nov 81 (05)
16 Feb 81 (02)
01 May 81 (01)
01 May 81 (04)
29 Dec 80 (09)
27 Nov 81 (01)
21 Sep 81 (02)
22 Apr 81 (07)
14 Feb 81 (22)
13 Nov 81 (06)
08 Feb 81 (01)
13 Nov 81 (02)
18 Oct 81 (05)
21 Sep 81 (03)
01 May 81 (05)
07 May 81 (24)
14 Feb 81 (23)
13 Jan 81 (12)
22 Apr 81 (06)
01 May 81 (03)
22 Apr 81 (08)
17 Jul 81 (21)
04 Dec 80 (21)
18 Aug 81 (23)
Concentration
(ppra)
0.7095
0.6225
0.6185
0.6095
0.6035
0.5900
0.5580
0.5485
0.5325
0.4930
0.4835
0.4685
0.4275
0.4235
0.4085
0.3965
0.3965
0.3810
0.3680
0.3645
0.3515
0.3430
0.3295
0.3190
0.3155
3-Hour Average
Concentrations
Date
(Period)
13 Nov 81 (2)
01 May 81 (1)
01 May 81 (2)
22 Apr 81 (3)
14 Feb 81 (8)
21 Sep 81 (1)
16 Feb 81 (1)
21 Aug 81 (1)
08 Feb 81 (1)
29 Dec 80 (3)
19 Aug 81 (2)
04 Dec 80 (7)
18 Aug 81 (8)
27 Nov 81 (1)
24 Mar 81 (2)
22 Dec 80 (2)
14 Nov 81 (5)
30 Sep 81 (3)
13 Nov 81 (1)
13 Jan 81 (4)
21 Mar 81 (8)
07 May 81 (8)
03 Nov 81 (3)
26 Sep 81 (2)
23 Nov 81 (8)
Concentration
(ppm)
0.4728
0.4008
0.3832
0.3612
0.3557
0.3198
0.2685
0.2575
0.2188
0.2185
0.2162
0.2000
0.1902
0.1885
0.1532
0.1490
0.1482
0.1453
0.1442
0.1427
0.1375
0.1375
0.1363
0.1325
0.1278
24-Hour Average
Concentrations
Date
01 May 81
13 Nov 81
14 Feb 81
22 Apr 81
26 Nov 81
29 Dec 80
16 Feb 81
30 Sep 81
19 Aug 81
21 Sep 81
09 Jan 81
24 Mar 81
04 Dec 80
07 May 81
22 Dec 80
17 Aug 81
14 Nov 81
08 Feb 81
21 Aug 81
20 May 81
17 Jul 81
27 Nov 81
23 Nov 81
18 Oct 81
13 Jan 81
Concentration
(ppm)
0.1099
0.1070
0.0699
0.0679
0.0607
0.0588
0.0551
0.0532
0.0530
0.0511
0.0460
0.0450
0.0439
0.0428
0.0422
0.0417
0.0415
0.0408
0.0405
0.0395
0.0368
0.0351
0.0313
0.0297
0.0284
B-9
-------
TABLE B-9
TWENTY-FIVE HIGHEST OBSERVED (MINUS BACKGROUND) SHORT-TERM S02 CONCENTRATIONS
AT MONITOR 10 DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
01
02
03
04
05
06
07
08
09
10
11
12
13
14 '
15
16
17
18
19
20
21
22
23
24
25
1-Hour Average
Concentrations
Date
(Hour)
06 Feb 81 (09)
06 Jan 81 (10)
15 Aug 81 (10)
08 Jun 81 (23)
06 Jan 81 (11)
18 Jun 81 (08)
23 Apr 81 (10)
06 Oct 81 (06)
09 Jan 81 (10)
09 Jan 81 (11)
15 Aug 81 (09)
09 Jan 81 (09)
18 Jun 81 (07)
08 Apr 81 (09)
20 Jul 81 (10)
21 Jun 81 (08)
11 Mar 81 (05)
30 Sep 81 (13)
25 Sep 81 (10)
05 Jul 81 (11)
30 May 81 (09)
13 Jan 81 (10)
03 Jan 81 (11)
30 May 81 (10)
23 Dec 80 (05)
Concentration
(ppm)
0.2105
0.2055
0.2055
0.1785
0.1765
0.1685
0.1615
0.1595
0.1560
0.1370
0.1305
0.1290
0.1225
0.1205
0.1155
0.1125
0.1095
0.1090
0.1075
0.1055
0.1025
0.1010
0.1000
0.0975
0.0965
3-Hour Average
Concentrations
Date
(Period)
06 Jan 81 (4)
09 Jan 81 (4)
18 Jun 81 (3)
15 Aug 81 (4)
06 Feb 81 (3)
03 Jan 81 (4)
05 Jul 81 (4)
09 Jan 81 (3)
23 Dec 80 (2)
03 Jan 81 (2)
08 Jun 81 (8)
13 Jan 81 (3)
22 Dec 80 (3)
23 Dec 80 (3)
13 Jan 81 (2)
23 Apr 81 (4)
12 Dec 80 (4)
11 Mar 81 (2)
21 Jun 81 (3)
06 Oct 81 (2)
08 Jan 81 (8)
12 Dec 80 (6)
07 Jan 81 (8)
29 Sep 81 (4)
10 Aug 81 (4)
Concentration
(ppm)
0.1592
0.1120
0.1078
0.1028
0.0992
0.0863
0.0855
0.0850
0.0785
0.0770
0.0752
0.0707
0.0692
0.0685
0.0652
0.0652
0.0647
0.0628
0.0618
0.0618
0.0608
0.0585
0.0583
0.0560
0.0547
24-Hour Average
Concentrations
Date
09 Jan 81
03 Jan 81
06 Jan 81
12 Dec 80
23 Dec 80
13 Jan 81
22 Dec 80
29 Dec 80
15 Aug 81
05 Jan 81
08 Jan 81
05 Jul 81
06 Feb 81
24 Nov 81
26 Nov 81
16 Dec 80
30 Sep 81
13 Feb 81
05 Feb 81
03 Nov 81
06 Oct 81
25 Dec 80
01 Jan 81
18 Jun 81
14 Jun 81
Concentration
(ppm)
0.0433
0.0373
0.0354
0.0316
0.0315
0.0294
0.0291
0.0273
0.0271
0.0263
0.0219
0.0206
0.0204
0.0200
0.0200
0.0198
0.0188
0.0184
0.0182
0.0180
0.0173
0.0173
0.0172
0.0170
0.0168
B-10
-------
TABLE B-10
OBSERVED ANNUAL AVERAGE S02 CONCENTRATIONS DURING THE SECOND YEAR
OF THE WESTVACO MONITORING PROGRAM
Site
1
3
4
5
6
7
8
9
10
No. of Hours
of Valid Data
7,807
6,628
7,828
7,977
7,123
7,939
8,209
8,291
8,507
Annual Average Concentration (pptn)
With Background
0.026
0.023
0.019
0.019
0.039
0.019
0.017
0.016
0.012
Without Background
0.022
0.018
0.014
0.014
0.034
0.014
0.012
0.011
0.008
B-ll
-------
TABLE B-ll
TWENTY-FIVE HIGHEST OBSERVED (MINUS BACKGROUND) 1-HOUR S02 CONCENTRATIONS
AT ALL MONITORS DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
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
Monitor
5
4
6
7
7
8
6
8
6
8
6
5
5
5
6
7
1
1
5
6
9
6
1
1
8
Date
(Hour)
21 Aug 81 (03)
13 Jan 81 (03)
13 Jan 81 (04)
13 Nov 81 (05)
13 Nov 81 (02)
13 Nov 81 (02)
19 Nov 81 (09)
13 Nov 81 (05)
8 Apr 81 (07)
27 Nov 81 (01)
8 Apr 81 (08)
27 Nov 81 (01)
13 Nov 81 (02)
13 Nov 81 (05)
8 Apr 81 (06)
13 Nov 81 (06)
22 Oct 81 (07)
13 Jan 81 (04)
13 Nov 81 (06)
16 Feb 81 (01)
21 Aug 81 (03)
8 Apr 81 (05)
13 Jan 81 (02)
19 Nov 81 (08)
20 Aug 81 (24)
Concentration
(ppm)
0.8935
0.8740
0.8660
0.8525
0.8475
0.8435
0.8250
0.8215
0.8165
0.8070
0.7825
0.7750
0.7625
0.7585
0.7585
0.7525
0.7285
0.7160
0.7145
0.7125
0.7095
0.6915
0.6835
0.6780
0.6695
B-12
-------
TABLE B-12
TWENTY-FIVE HIGHEST OBSERVED (MINUS BACKGROUND) 3-HOUR S02 CONCENTRATIONS
AT ALL MONITORS DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
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
Monitor
7
7
5
6
1
6
8
1
6
9
1
1
4
5
6
3
6
5
5
9
5
3
9
1
7
Date
(Period)
13 Nov 81 (2)
13 Nov 81 (1)
13 Nov 81 (2)
8 Apr 81 (3)
22 Oct 81 (3)
8 Apr 81 (2)
13 Nov 81 (2)
19 Nov 81 (3)
13 Jan 81 (2)
13 Nov 81 (2)
13 Jan 81 (1)
13 Jan 81 (2)
13 Jan 81 (1)
21 Aug 81 (1)
29 Mar 81 (2)
13 Jan 81 (1)
19 Nov 81 (3)
16 Feb 81 (1)
21 Sep 81 (1)
1 May 81 (1)
22 Apr 81 (3)
22 Oct 81 (3)
1 May 81 (2)
6 Jul 81 (7)
16 Feb 81 (1)
Concentration
(ppm)
0.7432
0.6575
0.6235
0.6142
0.5412
0.5268
0.5195
0.5158
0.5135
0.4728
0.4620
0.4612
0.4560
0.4518
0.4435
0.4337
0.4228
0.4062
0.4038
0.4008
0.3902
0.3842
0.3832
0.3768
0.3748
B-13
-------
TABLE B-13
TWENTY-FIVE HIGHEST OBSERVED (MINUS BACKGROUND) 24-HOUR S02 CONCENTRATIONS
AT ALL MONITORS DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
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
Monitor
1
3
5
6
6
4
1
1
3
6
6
6
8
6
6
6
6
4
6
1
9
4
3
6
9
Date
13 Jan 81
6 Jan 81
13 Nov 81
8 Apr 81
13 Jan 81
13 Jan 81
29 Dec 80
9 Nov 81
13 Jan 81
5 Dec 80
13 Nov 81
17 Jan 81
13 Nov 81
2 Jan 81
14 Nov 81
29 Dec 80
3 Apr 81
6 Jan 81
14 Dec 80
22 Oct 81
1 May 81
13 Nov 81
13 Nov 81
19 Nov 81
13 Nov 82
Concentration
(ppm)
0.1666
0.1585
0.1546
0.1501
0.1432
0.1419
0.1375
0.1330
0.1309
0.1299
0.1274
0.1257
0.1255
0.1232
0.1219
0.1195
0.1191
0.1147
0.1144
0.1113
0.1099
0.1097
0.1083
0.1082
0.1070
B-14
-------
APPENDIX C
RESULTS OF THE SHORTZ MODEL CONCENTRATION CALCULATIONS
This appendix presents the results of the SHORTZ model concentra-
tion calculations performed by the H. E. Cramer Company for the second year
of the Westvaco Monitoring program. The 25 highest short-term (1-hour,
3-hour and 24-hour average) SO concentrations calculated by the SHORTZ
model at Monitors 1, 3, 4, 5, 6, 7, 8, 9 and 10 are listed in Tables C-l
through C-9. The annual average concentrations calculated at these nine
monitors of concern for the model evaluation are shown in Table C-10.
Based on Tables C-l through C-9, the 25 highest 1-hour, 3-hour and 24-hour
average SO concentrations calculated by SHORTZ at all of the nine monitors
of concern are respectively given in Tables C-ll, C-12 and C-13.
. I
C-l
-------
TABLE C-l
TWENTY-FIVE HIGHEST SHORT-TERM S02 CONCENTRATIONS CALCULATED BY
SHORTZ AT MONITOR 1 DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
1-Hour Average
Concentrations
Date
(Hour)
29 Dec 80 (Itn
26 Nov 81 (23)
18 May 81 (02)
04 Nov 81 (21)
17 Feb 81 (21)
12 Feb 81 (22)
04 Dec 80 (21)
02 May 81 (23)
03 May 81 (24)
04 May 81 (23)
22 May 81 (24)
04 Nov 81 (20)
30 Nov 81 (05)
22 Apr 81 (06)
08 Feb 81 (08)
18 May 81 (03)
23 May 81 (05)
29 Dec 80 (21)
23 May 81 (02)
05 May 81 (01)
06 May 81 (02)
08 Feb 81 (02)
10 Apr 81 (03)
25 Mar 81 (21)
08 Aug 81 (01)
Concentration
(ppm)
6.8795
5.8871
5.7255
5.4999
4.7498
4.5901
4.5057
4.2689
4.2618
4.0195
3.9969
3.9622
3.9147
3.7116
3.7042
3.6565
3.6200
3.5932
3.5367
3.5339
3.4979
3.4590
3.3584
3.3150
3.2281
3-Hour Average
Concentrations
Date
(Period)
18 May 81 (1)
04 Nov 81 (7)
29 Dec 80 (4)
04 Dec 80 (7)
23 May 81 (1)
08 Feb 81 (1)
02 May 81 (8)
29 Dec 80 (7)
30 Nov 81 (2)
26 Nov 81 (8)
14 Nov 81 (8)
06 May 81 (2)
06 May 81 (1)
22 May 81 (8)
17 Feb 81 (7)
08 Feb 81 (3)
23 May 81 (2)
07 Jun 81 (2)
12 Feb 81 (8)
03 May 81 (8)
12 Nov 81 (2)
15 Jul 81 (7)
27 Jul 81 (7)
25 Mar 81 (7)
08 Jan 81 (7)
Concentration
(ppm)
3.1736
3.1540
2.5622
2.4438
2.3912
2.3584
2.3532
2.2477
1.9947
1.9624
1.9163
1.7978
1.7698
1.7091
1.6019
1.5866
1.5711
1.5549
1.5300
1.5082
1.4801
1.4701
1.4660
1.4243
1.4229
24-Hour Average
Concentrations
Date
29 Dec 80
08 Feb 81
06 May 81
23 May 81
18 May ,.81
07 Jun 81
03 Aug 81
02 May 81
03 May 81
04 Nov 81
12 Nov 81
25 Mar 81
15 Jul 81
22 May 81
12 Feb 81
30 Nov 81
17 Aug 81
11 Jun 81
27 Jul 81
04 Dec 80
30 Dec 80
08 Jan 81
14 Nov 81
26 Apr 81
29 Jul 81
Concentration
(ppm)
0.7163
0.6694
0.6067
0.5918
0.5175
0.5140
0.4894
0.4723
0.4594
0.4376
0.4281
0.4162
0.4121
0.3963
0.3939
0.3844
0.3655
0.3542
0.3412
0.3319
0.3117
0.3012
0.2928
0.2811
0.2761
C-2
-------
TABLE C-2
TWENTY-FIVE HIGHEST SHORT-TERM S02 CONCENTRATIONS CALCULATED BY
SHORTZ AT MONITOR 3 DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
1-Hour Average
Concentrations
Date
(Hour)
03 May 81 (23)
23 May 81 (04)
23 May 81 (03)
23 May 81 (05)
18 Jun 81 (01)
27 Aug 81 (21)
17 May 81 (24)
03 May 81 (01)
08 Jan 81 (22)
01 Jun 81 (02)
25 Aug 81 (02)
24 May 81 (04)
17 Aug 81 (21)
10 Apr 81 (02)
01 Jun 81 (01)
17 Jun 81 (24)
17 Aug 81 (04)
30 Jul 81 (01)
03 May 81 (02)
31 Jul 81 (04)
15 Jul 81 (22)
14 Feb 81 (21)
02 May 81 (24)
22 Jul 81 (24)
09 Jan 81 (09)
Concentration
(ppm)
6.5644
6.5305
5.8840
5.7796
4.0196
3.5355
3.5058
3.3168
3.2510
3.2469
2.9568
2.8205
2.7610
2.7427
2.6599
2.6310
2.6016
2.5970
2.5478
2.3634
2.3499
2.3079
2.1999
2.1345
2.1078
3-Hour Average
Concentrations
Date
(Period)
23 May 81 (2)
23 May 81 (1)
03 May 81 (8)
03 May 81 (1)
25 Aug 81 (1)
01 Jun 81 (1)
27 Aug 81 (7)
10 Jul 81 (8)
17 May 81 (8)
18 Jun 81 (1)
17 Jun 81 (8)
17 Aug 81 (7)
17 Aug 81 (2)
04 Dec 80 (7)
20 May 81 (1)
10 Apr 81 (1)
08 Jan 81 (8)
22 Jul 81 (8)
02 May 81 (8)
27 Jun 81 (1)
24 May 81 (2)
08 Feb 81 (1)
30 Jul 81 (1)
10 Jul 81 (7)
06 May 81 (1)
Concentration
(ppm)
4.1034
2.5276
2.3627
2.1184
1.9888
1.9689
1.9673
1.7685
1.6580
1.3399
1.3068
1.3021
1.2940
1.2490
1.1868
1.1131
1.0837
0.9610
0.9484
0.9439
0.9402
0.8748
0.8657
0.8558
0.8147
24-Hour Average
Concentrations
Date
23 May 81
03 May 81
17 Aug 81
10 Jul 81
15 Jul 81
25 Aug 81
17 May 81
01 Jun 81
27 Aug 81
17 Jun 81
02 May 81
22 Jul 81
26 Jun 81
06 May 81
14 Nov 81
24 Mar 81
20 May 81
30 Jul 81
18 Jun 81
18 Jul 81
04 Dec 80
23 Jun 81
31 May 81
12 Nov 81
29 Jul 81
Concentration
(ppm)
0.8503
0.6660
0.4989
0.3647
0.3524
0.3380
0.2666
0.2535
0.2495
0.2387
0.2337
0.2157
0.2040
0.1987
0.1985
0.1831
0.1732
0.1726
0.1675
0.1660
0.1659
0.1603
0.1594
0.1555
0.1552
C-3
-------
TABLE C-3
TWENTY-FIVE HIGHEST SHORT-TERM S02 CONCENTRATIONS CALCULATED BY
SHORTZ AT MONITOR 4 DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
1-Hour Average
Concentrations
Date
(Hour)
04 Dec 80 (20)
17 Jun 81 (22)
01 Aug 81 (04)
17 Jun 81 (24)
14 Feb 81 (21)
23 May 81 (04)
01 May 81 (04)
18 Jul 81 (21)
07 May 81 (03)
09 Jan 81 (09)
17 May 81 (23)
25 Aug 81 (04)
03 May 81 (22)
17 Aug 81 (05)
22 Jul 81 (24)
18 Jul 81 (20)
03 May 81 (02)
25 Aug 81 (02)
18 May 81 (01)
17 May 81 (24)
08 Feb 81 (01)
15 Jul 81 (22)
29 Jul 81 (24)
25 Aug 81 (03)
24 Mar 81 (05)
Concentration
(ppm)
5.1199
4.0664
3.3959
3.3805
2.8275
2.6215
2.5917
2.5317
2.5029
2.3561
2.2419
2.2166
2.2070
2.1527
2.1487
2.1291
2.0764
2.0632
2.0279
1.9811
1.9676
1.9023
1.8892
1.8249
1.8030
3-Hour Average
Concentrations
Date
(Period)
17 Jun 81 (8)
04 Dec 80 (7)
07 May 81 (1)
25 Aug 81 (1)
17 May 81 (8)
18 Jul 81 (7)
25 Aug 81 (2)
17 Aug 81 (2)
10 Jul 81 (8)
01 Aug 81 (2)
20 May 81 (1)
22 Jul 81 (8)
23 May 81 (2)
03 May 81 (1)
29 Jul 81 (8)
14 Feb 81 (7)
24 Mar 81 (2)
17 Aug 81 (8)
08 Feb 81 (1)
01 May 81 (2)
26 Jun 81 (8)
09 Jan 81 (3)
10 Jul 81 (7)
03 May 81 (8)
10 Apr 81 (1)
Concentration
(ppm)
2.4823
2.1496
1.7547
1.7084
1.6818
1.6502
1.2715
1.2188
1.2042
1.1345
1.0483
1.0234
0.9957
0.9899
0.9567
0.9425
0.9285
0.9049
0.8712
0.8639
0.8385
0.7854
0.7813
0.7657
0.7470
24-Hour Average
Concentrations
Date
17 Aug 81
17 Jun 81
25 Aug 81
07 May 81
24 Mar 81
04 Dec 80
17 May 81
10 Jul 81
03 May 81
18 Jul 81
15 Jul 81
14 Nov 81
22 Jul 81
01 May 81
29 Jul 81
26 Jun 81
18 May 81
27 Apr 81
23 May 81
01 Aug 81
08 Feb 81
25 Mar 81
17 Jul 81
20 May 81
18 Mar 81
Concentration
(ppm)
0.4258
0.4050
0.3889
0.3459
0.3152
0.2752
0.2738
0.2612
0.2348
0.2273
0.2231
0.1917
0.1836
0.1825
0.1733
0.1642
0.1579
0.1531
0.1426
0.1418
0.1373
0.1371
0.1345
0.1325
0.1317
C-4
-------
TABLE C-4
TWENTY-FIVE HIGHEST SHORT-TERM S02 CONCENTRATIONS CALCULATED BY
SHORTZ AT MONITOR 5 DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
01
02
03
04
05
06
07
08
09
10
11
12
13
14 '
15
16
17
18
19
20
21
22
23
24
25
1-Hour Average
Concentrations
Date
(Hour)
06 Aug 81 (04)
29 Dec 80 (09)
07 May 81 (04)
08 Feb 81 (01)
03 Aug 81 (07)
17 Jul 81 (21)
27 Apr 81 (07)
18 Jul 81 (21)
07 May 81 (03)
03 May 81 (22)
17 Feb 81 (20)
25 Mar 81 (23)
07 Aug 81 (24)
04 May 81 (22)
17 Aug 81 (22)
18 Jul 81 (20)
17 May 81 (22)
07 May 81 (06)
18 May 81 (04)
23 Jul 81 (01)
14 Nov 81 (21)
20 May 81 (02)
17 Jun 81 (21)
15 Sep 81 (22)
17 May 81 (23)
Concentration
(ppm)
1.7645
1.7342
1.5941
1.5281
1.4992
1.4428
1.3583
1.2958
1.2863
1.2841
1.2523
1.2131
1.2114
1.2016
1.1844
1.1523
1.1334
1.1144
1.0966
1.0622
1.0425
1.0061
0.9653
0.9150
0.8910
3-Hour Average
Concentrations
Date
(Period)
07 May 81 (2)
18 Jul 81 (7)
07 May 81 (1)
17 Jul 81 (7)
17 May 81 (8)
06 Aug 81 (2)
07 Aug 81 (8)
14 Nov 81 (7)
17 Aug 81 (8)
29 Dec 80 (3)
03 Aug 81 (3)
25 Aug 81 (2)
24 Mar 81 (2)
08 Feb 81 (1)
17 Jun 81 (7)
27 Apr 81 (3)
24 Mar 81 (3)
18 May 81 (2)
03 May 81 (8)
17 Feb 81 (7)
17 Aug 81 (2)
26 Jun 81 (8)
20 May 81 (1)
25 Mar 81 (8)
17 May 81 (7)
Concentration
(ppm)
1.1889
1.0557
0.7984
0.7245
0.6900
0.6874
0.6482
0.6186
0.5876
0.5781
0.5299
0.5261
0.5223
0.5112
0.4619
0..4528
0.4494
0.4350
0.4280
0.4174
0.4064
0.4063
0.4054
0.4044
0.4022
24-Hour Average
Concentrations
Date
07 May 81
24 Mar 81
17 Aug 81
17 Jul 81
14 Nov 81
17 May 81
18 Jul 81
27 Apr 81
18 May 81
25 Aug 81
15 Sep 81
07 Aug 81
06 Aug 81
03 Aug 81
17 Jun 81
25 Mar 81
01 May 81
29 Dec 80
03 May 81
08 Feb 81
16 Aug 81
06 May 81
21 Mar 81
26 Jun 81
15 Jul 81
Concentration
(ppm)
0.2866
0.2096
0.2024
0.1602
0.1460
0.1464
0.1374
0.1290
0.1190
0.1160
0.0965
0.0918
0.0867
0.0840
0.0817
0.0779
0.0764
0.0733
0.0732
0.0728
0.0695
0.0677
0.0653
0.0624
0.0579
C-5
-------
TABLE C-5
TWENTY-FIVE HIGHEST SHORT-TERM S02 CONCENTRATIONS CALCULATED BY
SIIORTZ AT MONITOR 6 DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
01
02
03
04
05
06
07
08
09
10
11
12
13
14 '
15
16
17
18
19
20
21
22
23
24
25
1-Hour Average
Concentrations
Date
(Hour)
20 May 81 (23)
13 Nov 81 (06)
23 May 81 (06)
09 Nov 81 (09)
11 Jul 81 (23)
14 Aug 81 (21)
29 Dec 80 (24)
07 Apr 81 (03)
01 Nov 81 (09)
21 Aug 81 (07)
31 Jul 81 (01)
23 May 81 (24)
22 May 81 (23)
07 Jun 81 (21)
06 Jul 81 (21)
18 Jul 81 (23)
31 Jul 81 (02)
29 Jun 81 (20)
27 Jul 81 (21)
27 Jul 81 (22)
08 Feb 81 (04)
23 May 81 (23)
01 Aug 81 (06)
21 May 81 (02)
26 Sep 81 (05)
Concentration
(ppm)
9.2365
8.3246
4.9185
4.8016
4.7761
4.5817
4.4951
4.4921
4.4405
4.2338
4.0898
4.0532
3.9886
3.9400
3.8995
3.8884
3.7332
3.7054
3.5055
3.4120
3.4070
3.3534
3.3010
3.1703
3.1527
3-Hour Average
Concentrations
Date
(Period)
20 May 81 (8)
29 Dec 80 (8)
23 May 81 (8)
13 Nov 81 (2)
31 Jul 81 (1)
07 Apr 81 (1)
14 Aug 81 (7)
06 Jul 81 (8)
21 May 81 (1)
16 Nov 81 (8)
04 May 81 (2)
03 Nov 81 (2)
22 May 81 (1)
21 May 81 (8)
17 Jul 81 (1)
03 Aug 81 (8)
26 Sep 81 (2)
11 Jul 81 (8)
15 Jul 81 (8)
22 May 81 (2)
09 Sep 81 (8)
23 Jul 81 (2)
23 May 81 (2)
17 Nov 81 (1)
28 Oct 81 (7)
Concentration
(ppm)
3.1442
2.9067
2.8897
2.7749
2.6077
2.4424
2.3394
2.3120
2.1010
2.0966
2.0463
2.0037
1.9546
1.9279
1.9278
1.9037
1.8629
1.8244
1.7932
1.7584
1.7330
1.7005
1.6398
1.6301
1.6263
24-Hour Average
Concentrations
Date
22 May 81
29 Dec 80
21 May 81
23 May 81
06 Jul 81
04 May 81
20 May 81
03 Nov 81
08 Feb 81
16 Nov 81
03 Aug 81
09 Sep 81
27 Jul 81
30 Jul 81
09 Nov 81
17 Jul 81
29 Sep 81
07 Dec 80
15 Nov 81
17 Nov 81
28 Oct 81
06 Dec 80
18 Sep 81
31 Jul 81
15 Jul 81
Concentration
(ppm)
0.8225
0.7722
0.7210
0.6912
0.6296
0.5938
0.5794
0.5755
0.5553
0.5393
0.5378
0.5081
0.5051
0.4963
0.4913
0.4845
0.4688
0.4673
0.4526
0.4391
0.4311
0.4192
0.4081
0.4071
0.4051
C-6
-------
TABLE C-6
TWENTY-FIVE HIGHEST SHORT-TERM S02 CONCENTRATIONS CALCULATED BY
SHORTZ AT MONITOR 7 DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
1-Hour Average
Concentrations
Date
(Hour)
29 Dec 80 (09)
04 May 81 (22)
06 Aug 81 (04)
07 Aug 81 (23)
01 May 81 (01)
17 Feb 81 (20)
17 May 81 (22)
18 Jul 81 (19)
07 May 81 (05)
07 Aug 81 (24)
28 Dec 80 (18)
08 Feb 81 (01)
22 Jul 81 (23)
07 May 81 (06)
17 Jul 81 (20)
07 May 81 (04)
14 Nov 81 (20)
16 Dec 80 (06)
17 Jun 81 (21)
20 May 81 (02)
07 May 81 (07)
17 Aug 81 (06)
17 Jul 81 (06)
17 Aug 81 (02)
21 Mar 81 (22)
Concentration
(ppm)
2.1438
1.8379
1.5619
1.5539
1.4955
1.4708
1.4034
1.3720
1.3359
1.3279
1.3180
1.2708
1.2640
1.2345
1.2234
1.1743
1.1126
1.0949
1.0201
0.9925
0.9542
0.9494
0.9407
0.7899
0.7866
3-Hour Average
Concentrations
Date
(Period)
07 May 81 (2)
07 Aug 81 (8)
29 Dec 80 (3)
21 Mar 81 (8)
06 Aug 81 (2)
17 Jul 81 (7)
04 May 81 (8)
18 Jul 81 (7)
14 Nov 81 (7)
17 May 81 (8)
01 May 81 (1)
17 Feb 81 (7)
18 May 81 (2)
18 Aug 81 (2)
28 Dec 80 (6)
17 Jun 81 (7)
08 Feb 81 (1)
22 Jul 81 (8)
05 Aug 81 (8)
28 May 81 (8)
17 Aug 81 (1)
24 Mar 81 (2)
16 Dec 80 (2)
07 May 81 (3)
24 Mar 81 (3)
Concentration
(ppm)
1.2483
0.9606
0.7146
0.7070
0.6663
0.6644
0.6126
0.5898
0.5600
0.5086
0.4985
0.4903
0.4792
0.4400
0.4398
0.4234
0.4236
0.4232
0.4085
0.3958
0.3912
0.3824
0.3650
0.3566
0.3527
24-Hour Average
Concentrations
Date
07 May 81
24 Mar 81
17 Jul 81
17 Aug 81
14 Nov 81
07 Aug 81
18 May 81
18 Aug 81
21 Mar 81
17 May 81
01 May 81
29 Dec 80
06 Aug 81
27 Apr 81
04 May 81
28 Dec 80
18 Jul 81
17 Feb 81
06 May 81
16 Dec 80
23 Mar 81
22 Jul 81
08 Feb 81
17 Jun 81
15 Sep 81
Concentration
(ppm)
0.2330
0.1610
0.1583
0.1458
0.1455
0.1271
0.1140
0.1093
0.1004
0.0993
0.0968
0.0894
0.0833
0.0811
0.0784
0.0774
0.0745
0.0613
0.0589
0.0559
0.0557
0.0556
0.0546
0.0532
0.0518
C-7
-------
TABLE C-7
TWENTY-FIVE HIGHEST SHORT-TERM S02 CONCENTRATIONS CALCULATED BY
SHORTZ AT MONITOR 8 DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
1-Hour Average
Concentrations
Date
(Hour)
07 Aug 81 (23^
29 Dec 80 (09)
28 Dec 80 (18)
16 Dec 80 (06)
30 Sep 81 (03)
07 Aug 81 (24)
17 Feb 81 (20)
21 Mar 81 (23)
21 Mar 81 (24)
03 May 81 (21)
18 May 81 (21)
04 May 81 (22)
01 May 81 (01)
18 Jul 81 (19)
21 Mar 81 (22)
18 May 81 (05)
22 Jul 81 (23)
17 Aug 81 (06)
17 Jul 81 (20)
05 Aug 81 (24)
01 May 81 (09)
14 Nov 81 (20)
18 May 81 (06)
17 May 81 (22)
16 Dec 80 (07)
Concentration
(ppra)
1.2892
1.0728
1.0373
0.9581
0.8018
0.6694
0.6579
0.6329
0.6311
0.6221
0.6065
0.5800
0.5326
0.5323
0.4875
0.4868
0.4586
0.3796
0.3743
0.3664
0.3650
0.3549
0.3541
0.3482
0.3450
3-Hour Average
Concentrations
Date
(Period)
07 Aug 81 (8)
21 Mar 81 (8)
18 May 81 (7)
29 Dec 80 (3)
28 Dec 80 (6)
16 Dec 80 (2)
18 May 81 (2)
18 Aug 81 (2)
03 May 81 (7)
28 May 81 (8)
17 Feb 81 (7)
14 Nov 81 (7)
18 Aug 81 (1)
04 May 81 (8)
06 Aug 81 (2)
07 May 81 (2)
18 Jul 81 (7)
01 May 81 (1)
05 Aug 81 (8)
22 Jul 81 (8)
13 Nov 81 (8)
17 Jul 81 (7)
27 Apr 81 (2)
17 Aug 81 (2)
17 Jul 81 (3)
Concentration
(ppm)
0.6529
0.5838
0.3804
0.3576
0.3546
0.3194
0.2897
0.2658
0.2298
0.2286
0.2193
0.2002
0.1948
0.1933
0.1919
0.1826
0.1778
0.1775
0.1548
0.1529
0.1373
0.1315
0.1295
0.1265
0.1227
24-Hour Average
Concentrations
Date
18 May 81
07 Aug 81
21 Mar 81
28 Dec 80
18 Aug ,,81
14 Nov 81
16 Dec 80
29 Dec 80
01 May 81
17 Aug 81
17 Jul 81
07 May 81
30 Sep 81
24 Mar 81
03 May 81
28 May 81
17 Feb 81
27 Apr 81
04 May 81
06 Aug 81
15 Sep 81
18 Jul 81
17 May 81
09 Nov 81
05 Aug 81
Concentration
(ppn)
0.0922
0.0821
0.0731
0.0731
0.0625
0.0620
0.0543
0.0471
0.0431
0.0402
0.0378
0.0365
0.0365
0.0347
0.0311
0,0289
0.0274
0.0260
0.0242
0.0240
0.0230
0.0222
0.0205
0.0204
0.0194
c-t
-------
TABLE C-8
TWENTY-FIVE HIGHEST SHORT-TERM S02 CONCENTRATIONS CALCULATED BY
SHORTZ AT MONITOR 9 DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
01
02
03
04
05
06
07
08
09
10
11
12
13
14 '
15
16
17
18
19
20
21
22
23
24
25
1-Hour Average
Concentrations
Date
(Hour)
29 Dec 80 (09)
07 Aug 81 (24)
07 Aug 81 (23)
17 Feb 81 (20)
04 May 81 (22)
06 Aug 81 (04)
01 May 81 (01)
28 Dec 80 (18)
16 Dec 80 (06)
17 May 81 (22)
18 Jul 81 (19)
08 Feb 81 (01)
22 Jul 81 (23)
07 May 81 (05)
03 Aug 81 (07)
17 Jul 81 (20)
14 Nov 81 (20)
07 May 81 (06)
07 May 81 (04)
17 Jun 81 (21)
27 Apr 81 (06)
20 May 81 (02)
07 May 81 (07)
18 Aug 81 (06)
15 Sep 81 (18)
Concentration
(ppm)
1.3754
1.2377
1.0251
1.0248
0.9263
0.8528
0.7691
0.7157
0.7002
0.6890
0.6598
0.6313
0.6087
0.6065
0.5712
0.5641
0.5514
0.5495
0.5181
0.4568
0.4421
0.4382
0.4247
0.4245
0.4234
3-Hour Average
Concentrations
Date
(Period)
07 Aug 81 (8)
07 May 81 (2)
29 Dec 80 (3)
18 May 81 (7)
06 Aug 81 (2)
21 Mar 81 (8)
17 Feb 81 (7)
04 May 81 (8)
17 Jul 81 (7)
14 Nov 81 (7)
15 Sep 81 (6)
18 Jul 81 (7)
01 May 81 (1)
17 May 81 (8)
28 Dec 80 (6)
18 May 81 (2)
16 Dec 80 (2)
08 Feb 81 (1)
28 May 81 (8)
22 Jul 81 (8)
18 Aug 81 (2)
27 Apr 81 (2)
03 Aug 81 (3)
05 Aug 81 (8)
17 Jun 81 (7)
Concentration
(ppm)
0.7543
0.5581
0.4585
0.3718
0.3633
0.3602
0.3416
0.3088
0.2908
0.2764
0.2722
0.2662
0.2564
0.2433
0.2388
0.2362
0.2334
0.2104
0.2086
0.2033
0.2015
0.1988
0.1946
0.1872
0.1861
24-Hour Average
Concentrations
Date
07 May 81
07 Aug 81
18 May 81
24 Mar 81
17 Jul 81
14 Nov 81
17 Aug 81
29 Dec 80
01 May 81
27 Apr 81
15 Sep 81
21 Mar 81
18 Aug 81
17 May 81
06 Aug 81
28 Dec 80
17 Feb 81
04 May 81
16 Dec 80
18 Jul 81
03 Aug 81
06 May 81
08 Feb 81
22 Jul 81
28 May 81
Concentration
(ppm)
0.1029
0.0974
0.0868
0.0765
0.0687
0.0682
0.0672
0.0573
0.0529
0.0519
0.0506
0.0499
0.0496
0.0461
0.0454
0.0431
0.0427
0.0392
0.0350
0.0335
0.0298
0.0282
0.0269
0.0264
0.0261
C-9
-------
TABLE C-9
TWENTY-FIVE HIGHEST SHORT-TERM S02 CONCENTRATIONS CALCULATED BY
SHORTZ AT MONITOR 10 DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
1-Hour Average
Concentrations
Date
(Hour)
13 Jan 81 (08)
08 Apr 81 (01)
23 May 81 (08)
07 Apr 81 (23)
23 Feb 81 (20)
07 Apr 81 (22)
06 Feb 81 (05)
08 Apr 81 (05)
07 Apr 81 (24)
08 Apr 81 (04)
06 Jul 81 (03)
18 Feb 81 (04)
09 Nov 81 (08)
08 Apr 81 (02)
28 Feb 81 (05)
23 Nov 81 (19)
21 Sep 81 (23)
18 Feb 81 (06)
13 Jan 81 (04)
07 Feb 81 (07)
18 Oct 81 (05)
17 Apr 81 (01)
23 Mar 81 (07)
13 Jan 81 (06)
16 Apr 81 (23)
Concentration
(ppm)
0.1992
0.1718
0.1663
0.1648
0.1640
0.1627
0.1564
0.1534
0.1498
0.1460
0.1437
0.1435
0.1421
0.1394
0.1378
0.1346
0.1335
0.1327
0.1289
0.1259
0.1229
0.1221
0.1218
0.1210
0.1194
3-Hour Average
Concentrations
Date
(Period)
07 Apr 81 (8)
08 Apr 81 (2)
23 May 81 (3)
08 Apr 81 (1)
13 Jan 81 (2)
18 Feb 81 (2)
21 Sep 81 (8)
18 Feb 81 (1)
23 Feb 81 (7)
13 Jan 81 (3)
06 Feb 81 (2)
09 Feb 81 (8)
26 May 81 (2)
16 Apr 81 (8)
04 Mar 81 (2)
16 Apr 81 (1)
15 Aug 81 (2)
06 Jul 81 (1)
06 Oct 81 (2)
17 Feb 81 (8)
13 Jan 81 (1)
09 Nov 81 (3)
25 May 81 (1)
06 Feb 81 (1)
26 Sep 81 (8)
Concentration
(ppm)
0.1591
0.1338
0.1268
0.1257
0.1226
0.1210
0.1171
0.1051
0.1045
0.0965
0.0883
0.0854
0.0847
0.0822
0.0821
0.0820
0.0793
0.0793
0.0787
0.0777
0.0777
0.0776
0.0768
0.0750
0.0730
24-Hour Average
Concentrations
Date
18 Feb 81
08 Apr 81
13 Jan 81
16 Apr 81
02 Dec 80
05 Nov 81
07 Apr 81
17 Feb 81
26 Mar 81
15 Aug 81
27 Aug 81
20 Oct 81
04 Mar 81
08 Jun 81
06 Feb 81
03 Apr 81
06 Jan 81
23 Feb 81
25 May 81
26 May 81
14 Sep 81
04 Oct 81
09 Nov 81
19 Jul 81
21 Sep 81
Concentration
(ppm)
0.0458
0.0414
0.0390
0.0325
0.0301
0.0267
0.0263
0.0257
0.0250
0.0250
0.0243
0.0242
0.0234
0.0226
0.0220
0.0219
0.0214
0.0203
0.0199
0.0198
0.0195
0.0194
0.0192
0.0185
0.0182
C-10
-------
TABLE C-10
ANNUAL AVERAGE S02 CONCENTRATIONS CALCULATED BY SHORTZ FOR THE
SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Site
1
3
4
5
6
7
8
9
10
No. of Hours
of Valid Data
8,742
8,742
8,742
8,742
8,742
8,742
8,742
8,742
8,742
Annual Average Concentration
(ppm)
0.0905
0.0422
0.0279
0.0116
0.1573
0.0092
0.0037
0.0048
0.0045
C-ll
-------
TABLE C-ll
TWENTY-FIVE HIGHEST 1-HOUR S02 CONCENTRATIONS CALCULATED BY SHORTZ
AT ALL MONITORS DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
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
Monitor
6
6
1
3
3
1
3
3
1
1
4
6
6
6
1
1
6
1
6
6
6
1
1
6
6
Date
(Hour)
20 May 81 (23)
13 Nov 81 (06)
29 Dec 80 (10)
3 May 81 (23)
23 May 81 (04)
26 Nov 81 (23)
23 May 81 (03)
23 May 81 (05)
18 May 81 (02)
4 Nov 81 (21)
4 Dec 80 (20)
23 May 81 (06)
9 Nov 81 (09)
11 Jul 81 (23)
17 Feb 81 (21)
12 Feb 81 (22)
14 Aug 81 (21)
4 Dec 80 (21)
29 Dec 80 (24)
7 Apr 81 (03)
1 Nov 81 (09)
2 May 81 (23)
3 May 81 (24)
21 Aug 81 (07)
31 Jul 81 (01)
Concentration
(ppm)
9.2365
8.3246
6.8795
6.5644
6.5305
5.8871
5.8840
5.7796
5.7255
5.4999
5.1199
4.9185
4.8016
4.7761
4.7498
4.5901
4.5817
4.5057
4.4951
4.4921
4.4405
4.2689
4.2618
4.2338
4.0898
C-12
-------
TABLE C-12
TWENTY-FIVE HIGHEST 3-HOUR S02 CONCENTRATIONS CALCULATED BY SHORTZ
AT ALL MONITORS DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
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
Monitor
3
1
1
6
6
6
6
6
1
3
4
1
6
1
1
3
1
6
6
1
4
3
6
6
6
Date
(Hour)
23 May 81 (2)
18 May 81 (1)
4 Nov 81 (7)
20 May 81 (8)
29 Dec 80 (8)
23 May 81 (8)
13 Nov 81 (2)
31 Jul 81 (1)
29 Dec 80 (4)
23 May 81 (1)
17 Jun 81 (8)
4 Dec 80 (7)
7 Apr 81 (1)
23 May 81 (1)
8 Feb 81 (1)
3 May 81 (8)
2 May 81 (8)
14 Aug 81 (7)
6 Jul 81 (8)
29 Dec 80 (7)
4 Dec 80 (7)
3 May 81 (1)
21 May 81 (1)
16 Nov 81 (8)
4 May 81 (2)
Concentration
(ppm)
4.1034
3.1736
3.1540
3.1442
2.9067
2.8897
2.7749
2.6077
2.5622
2.5276
2.4823
2.4438
2.4424
2.3912
2.3584
2.3627
2.3532
2.3394
2.3120
2.2477
2.1496
2.1184
2.1010
2.0966
2.0463
C-13
-------
TABLE C-13
TWENTY-FIVE HIGHEST 24-HOUR S02 CONCENTRATIONS CALCULATED BY SHORTZ
AT ALL MONITORS DURING THE SECOND YEAR OF THE WESTVACO MONITORING PROGRAM
Rank
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
Monitor
3
6
6
6
1
6
1
3
6
1
6
1
6
6
6
6
6
1
1
6
6
3
6
6
1
Date
(Hour)
23 May 81
22 May 81
29 Dec 80
21 May 81
29 Dec 80
23 May 81
8 Feb 81
3 May 81
6 Jul 81
6 May 81
4 May 81
23 May 81
20 May 81
3 Nov 81
8 Feb 81
16 Nov 81
3 Aug 81
18 May 81
7 Jun 81
9 Sep 81
27 Jul 81
17 Aug 81
30 Jul 81
9 Nov 81
3 Aug 81
Concentration
(ppm)
0.8503
0.8225
0.7722
0.7210
0.7163
0.6912
0.6694
0.6660
0.6296
0.6067
0.5938
0.5918
0.5794
0.5755
0.5553
0.5393
0.5378
0.5175
0.5140
0.5081
0.5051
0.4989
0.4963
0.4913
0.4894
C-14
-------
APPENDIX D
CUMMULATIVE FREQUENCY DISTRIBUTIONS OF THE 25 HIGHEST OBSERVED (MINUS
BACKGROUND) AND CALCULATED (SHORTZ) SHORT-TERM SO CONCENTRATIONS
This appendix compares the cumulative frequency distributions of
the 25 highest short-term (1-hour, 3-hour and 24-hour average) observed
(minus background) S09 concentrations from Appendix B with the corresponding
cumulative frequency distributions of the 25 highest calculated (SHORTZ)
short-term concentrations from Appendix C. Table D-l gives the figure
number for each combination of monitor and concentration averaging time.
D-l
-------
TABLE D-l
IDENTIFICATION OF FIGURE NUMBERS BY MONITOR AND
CONCENTRATION AVERAGING TIME
Figure No.
D-l
D-2
D-3
D-4
D-5
D-6
D-7
D-8
D-9
D-10
D-ll
D-12
D-13
D-14
D-15
D-16
D-17
D-18
D-19
D-20
D-21
D-22
D-23
D-24
D-25
D-26
D-27
D-28
D-29
D-30
Monitor
1
1
1
3
3
3
4
4
4
5
5
5
6
6
6
7
7
7
8
8
8
9
9
9
10
10
10
All
All
All
Averaging Time
1 Hour
3 Hours
24 Hours
1 Hour
3 Hours
24 Hours
1 Hour
3 Hours
24 Hours
1 Hour
3 Hours
24 Hours
1 Hour
3 Hours
24 Hours
1 Hour
3 Hours
24 Hours
1 Hour
3 Hours
24 Hours
1 Hour
3 Hours
24 Hours
1 Hour
3 Hours
24 Hours
1 Hour
3 Hours
24 Hours
D-2
-------
1 1 1] 1 1
1
I
1
1
1
1
1
1
1
t
1
1
1
1
1
1
~ 1
1
I
1
I
1
1
1
1
I
~_ \
-
- \
\
\
\
\
"
~
~
-
-
~
_
: co
: o
- i i
i i
: cc
- Q£
I
:H
: uj
y
: o
(_^
: g
cc
tt:
UJ
r cc
_
'- O Z3
H- 0
1-1 x:
2 i
O --i
31
1 1 1 1 L
i
O
CD
CD
t 1 i j 1 1 j 1 I i 1 1 I
t-
cc
o
CO
-
-
I
1 ~
1
1 ~~
I _
\
\
\ 1
v i _
\
\ \ -
I 1 ""
J
^
1 1 -
1 1 ^
1 \
\ \ -
\ \
J 1 -
1 1 ~
1 ~
1 1 -
1 -
1 ~
r~)
jl] ~| "~
Q H \
LU (t \
^^* -yj ~
f^ 7^ __
uj 3
CO _J
CQ CC ~
0 H :
,1
!i :
j
l
ii i ] i i : i i i i i
^
CD
CD
CD
CD
CD
CD
LD
CD
s^
^^
O) ^~^
* r^
UJ
C3
UJ
oo a:
U-
UJ
r-- >
i i
^
CO ^
' 13
2:
ID ID
O
^*
*
m
CM
.
U3
O
T-H
« h
0 ^
o 'o
4-1
CO
CU
,-C
&0
H
^
U-|
CNJ
01
4-1
4-1
O
C!
O
H
4-1
J~
H
S-l
4-1
CO
H
'O
r^
O
C
CU
cr
01
M
4-i
Ol
>
H
4-1
cfl
1 1
3
e
3
CJ
0)
f.
4-J
4-1
O
d
o
CO
-H
CO
CU
e
o
CJ
t H
Q
0)
3
6C
H
(4
M
0
4-1
H
C
O
^-*
4-1
CO
03
C
o
H
4-1
CO
J-j
4J
C
cu
o
c
o
0
c
O
CO
cu
60
CO
}_4
0)
£>
CO
^
3
O
JC
1
1 i
^^
ID
C
3
O
60
^
O
CO
&
CO
3
C
H
e
a
ai
(-1
Ol
CO
0
1
c^
i i
£5
E
3
O
0)
^ r^
4-1
iJ
H
*
cfl
60
0
a
bC
C
()^
JS-J
o
4J
H
c^
o
e
o
CJ
CO
.2
CO
CU
&
Ol
4-)
4-1
O
^
CO
Ol
^
T3
C
O
o
cu
CO
cu
^
4J
CO
c
o
H
4-1
CO
J-i
4-1
a
CJ
CJ
C!
0
CJ
CU
cO°
CU
>
cfl
^
3
C
(-*
1
11
4-1
Cfl
01
f:
60
J3
U~l
CN
CU
J2
4-1
O
CJ
O
1-1
4-1
3
H
1-j
4J
H
*XJ
O
£
0)
3
cr
cu
S-J
60 4-1
C
H
}_t
O
01
>
H
4-1
N)
3
o
cc
>,
o
-o
cu
4-1
r 1
3
a
i i
cO
D-3
-------
(lit 1
~
"
-
-
""
-
-
-
-
_
_
:
-
-
: co
: o
~- f-
cc
: a;
-_ j
: uj
: o
o
I o
- UJ
cb
(X
ft:
UJ
«-i z>
r cc
Hi Qi
O Z3
1 0
n :C
Z 1
O CO
till 1
-1
o
1
Q
UJ
a:
UJ
CO
en
o
i
j i i j i i i i i i i i
~
rtJ
h-
o
X
CO
-
"
V \ "
\ \ -
\
V \ -
\ \
\ :
i 1
i \
t \
i \
i \ ~
\ | i
V i
V. 1
\
1 \ -
\
\ \ ~
\ \ ^
1 I
V V J
\ 1 ;
1 1
i \ *~
' N :
i V
i i -
i I
a ' \ I
^ \ -
cc
^ 1
u '
_I '
CC
CJ
1 1 1 1 1 1 1 1 1 I 1
CD *-
O
CD
CD T3
CD QJ
CD >*>?
0) 4.) 12 >,
CO 0) 43
42 60 3
o c a"a
H 0) 0)
CD t) ^i u jj
m co 3 M-I co
en a) TJ ^
42 a> 3
60-H > U
H -H iH
43 J-i 4J c8
O CO o
L/"l 4J i I
CN -H 3 CO
C E C2
O> O 3 O
m 42 S 0 -H
OT ^ jj
CD 4-1 0) CO
O 4-1 4J
CO (2
C C 42 01
O O 4-> O
H -H -H (2
4-1 4-1 S O
3 CO CJ
J3 4- ^ S
TS -H 4J ca cu
. l-i (2 M 60
4-1 0) 60 CO
. CO O O ^
*" -H C (-1 O!
rn O T3 0 & >
. ^ o ca
UJ ^ M
^ cj cNt2 (-J
= C2 O -H 3
t* (U cy) M 0
UJ 3 0 42
m Qi cr Ol 4-1 1
. ij_ a) so -r-t m
>-j ca c
... 14-H (-1 O 4J
UJ CD S CO
r* -* -i > CO 0 42
( -H O tO
(-»- 4J S-i ca T-I
tn i CO 3 > 42
- ' tH O 4-1
^3 3 42 CO u-i
z: e i w CM
"? F^ 0^ ^ 0,
tj <-N 0) 42
O) T3 42 4-1
-* 42 C 4-1
4J 3 M-J
O 4H O
M-I S-l O
(0 0 60 C2
. 4x H O
CO CO -H
O CO Ol 4-1
CO 43 ^ 3
CSl -H 43
. t-n CO T3 -H
CO 3 C S-4
Cu 12 O 4-1
E -H U co
o e 01 .H
CJ x ' CO T3
^_^
CNI
1
LO «
° S
3
60
rH
^ .
CD
H
O
CO
D-4
-------
CO
^
o
cn
QZ
\~
z.
UJ
u
:s
o
CJ
\
v
i
d>
o; ct
O ZD
H- O
UJ Q Iv
Ct> UJ CE \
£ £ 3 >
UJ UJ CJ '
> to _j \
CE DQ CC \
O CJ '
cc
O
X
O
z:
CSJ
J L
J L
CD
O)
CD
CD
CD
CD
CD
CD
CD
UD
CD
CD
CJ
2.
UJ
o
UJ
co a:
u.
UJ
CJ
CO
CM
in
o
0>
1
i
0) CU 3
03 60 t-l Cl
43 c y-< ii
O -H CS
>-l
co TD -H cn
CO
c
H
4-1
X O £
4-1 3 VJ
LO -H CJ 4-1
c^ ^ C
o cu o)
cu S j= cj
JC 4-J C
4-14-1 O
co jr o
CM 4-1
C CO
c
c o
O -rt
o
co
ct)
g !-
H 4J CO O)
4-1 CO i-l >
3 M cc ca
X 4-1 O
H C M (J
S-4 01 p. 3
4-1 O O
CO C 60
1
c
o c
no cj
J-l CXI
>-. CNiO
O O 4-1 4-J
C CO v-t CO
o) co)
3 0) O 4=
cr eo g t;
Ol CC
»-l
4-1 CU CJ
> cS tn
01 ro > CN
> J-l
CO Ol
O -±
3
o
3 i a) M-i
6 > >-i
C cj
O cS
cn
cn co
d -H
O 13 >-,
H CJ ja
SJ CO Ol >,
CO 3 Cfi CJ TJ
p. c: c 01
6 -H Q) CU 4-1
O E
3
CT i
3
CO
o
D-5
-------
I 1 1 I 1
-
-
_
_
~
I
v
-
\
1
I
1
"»_
^^.^
-
_
-
: tn
o
- |-~4
i i
cc
: a:
~ 2
: uj
- 2
o
: o
- UJ C
3 u
cn i
££ Q
UJ U
m > u
cc a
: c
I o rs
f- o
1 | T*
Z 1
O i-l
-
I 1 1 1 !
H
O
i 1 1 j 1 1 1 i 1 1 i i
*
-
g .
O
X
CO
.
"
\ "
\ *
\ ""
1
I :
I
\
\
^ i
\ 1
i \
i \
\ \ '-
\ \ .
i i -
i I :
i 1
\ V
V N
\ \
\ 1 :
i \
i \ ~
i
1
X I
a \ I
UJ '
D F- '
J CC
' d »
J 3 \ ] :
3 CC '
3 O I
I
it i 1 I i I i i I I i
CD *-
0 '
0)
CD
f*T\
CD
O)
CD
CD
CD
*
CD
CD
in
CD
*
o>
CD
P^
m
CO
in
^.
m
CM
r-4
in
o
« .
O
^~
o
UJ
-J
a
UJ
a:
U-
UJ
^x
1 1
I
I
cc
33
3E
O
T3
d)
j^ 1
cu cb i
cn S-i 3
J3 M) <4-l CJ
o d t-i
H CU CO
4-1 i-i > a
cn 3 -H
(b T3 4-1 0}
rd CO d
00 01 rH O
H 3 -H
,e »< S -u
0 3 cfl
U1 4-1 O S-l
CNI -H 4-1
d cu d
cu o .d cu
jz 21 *-> u
4-1 d
4-1 ,d o
IW Cfl 4-1 CJ
O -H
CO 3 CU
d d oo
C O g CO
H iH CO (-1
4-1 4-1 >-l CU
3 CO 00 >
XI >-i O CO
H 4J M
t-J d CX S-i
4-1 CU 3
cn o 00 o
H d d .d
0 0 -H 1
U !-l ^H
>> 0
CJ CN14-I 4-1
d C -H cn
CU CO d CD
3 ox;
cr cu g to
CU 00 -H
^ 03 o x:
*4~i V4 CJ
cu co m
cu > > CN
> C8 4J
H cn cu
4-1 t-i cu x;
CO 3 ^3 4-i
i-H 0
3 XI CU 14-1
g 1 XI O
3 i 4-1
o d
x-x 14-1 O
Ol T3 O -H
X! d 4J
O CO Xi H
m >-i cu -H PH
O 00 ^ (-1 O
^ 4_J £C3
d cj T3 cn c/i
o ct d -H
cn ^n o T3 P^
H CJ XI
t-i cn cu >>
co 3 cn o -X3
p. d d cu
g -H CU CU 4-1
O 6 X! D CO
o ^ 4-1 cr r-t
B
1
o
cu
H
3
00
i-t
D-6
-------
I
-
-
.
_
_
~
"
_
~
-
_
_
-
-
-
:
-
~~
^_
-
i
^
"
-
~~
~-
-
_
_
-
~
~
~~
-
~
-
i _i
1
o
III J
\
I
I
\
to
o
» 1
1
cr
a:
i
UJ
o
o
o
UJ a
5 UJ
cc :>
a; ct:
UJ UJ
CO > CO
CC CQ
O
Hi ry
O ^D
1 O
Z 1
o en
^_
iii i
i
\
i
i
\
\
i
a
UJ
i
cc
_j
o
_I
CC
O
1
1
1
1
1
CD
CD -0
i I 1 i i I i i i i i
-
-
KJ
i_
* cc
\ £}
s
\
V
\
i
i
i ~
.
j \
* \
\ \
\ \ *"
\ \
\ \ 2.
\\ ~"
\
\ \ -
\ \
\ \
i \
i \ "
*v \ "
v \
\ \
** N» \ I
"\ ] '
I 1
I \ ~.
\ \ ~-
\ \ "
\ 1 ~
\
\ \ ;
s \ I
1
1 I
\ 1
V _
1
1
1 _
1
I
1
1
1
1 i
-
-
1 1 | I ! | 1 1 1 1 1
CD £
CD g
0)
en
4=
o
CD *->
3 1
60
H
in
CNl
eu
CD "5
'+->
O
d
o
H
4->
£3
tn -3
CD £
4J
en
>~ -H
CD ^ ^
" ^ cT
5 c
C3 a)
UJ 3
«£ Z
UJ ^
f^ > 0)
. n >
i_ -H
-5
' rJ s
zr &
^3 3
OJ
-^ £
. 4-1
ro *o
d
o
en
CM -|j
rt
S-
0
o
^
*~^
.
m
1
in (=>
° S
^3
U3
H
P-!
» 1
tQ T-< .
o 'o
M
d
H
J-l
3
-a
CO
^i
o
4J
H
d
2
^
4J
cfl
en
C
o
H
4-1
co
5-J
4-1
d
CU
o
d
o
u
i
H
4->
CO
r-l
^3
^
3
CJ
CU
^
4-1
H
S
6
co
t-t
00
0
&
oo
d
H
1
^J
O
i t
CO
CJ
d
o
H
4-1
CO
}_<
d
cu
o
d
o
CJ
CU
00
CO
^1
0)
>
CO
(_,
3
O
f~".
1
en
o
CMI-I 4J
O
CA1
cu
00
CO
CU
>
CO
)-l
3
O
1
!Z
'O
d
3
O
>-i
00
a
co
42
en
3
d
H
e
H
d
o
E
o
CJ
CO
>
4-J
en
01
0)
4=
4-1
U-l
O
^-1
co
0)
>,
T3
secon
01
4^
"*"*
01
4=
oo
H
4T
in
CNI
0)
4=
J-i
H |
0
d
o
H
4J
3
42
H
V-i
4-1
en
H
T3
o
d
01
3
cr
C*s3
H
Pi
O
33
42
-a
0)
44
CO
'
(WddJ
D-7
-------
i 11 i r
I
I
I
1
i
l
i
l
1
l
t
l
l
l
l
l
i i i i i i i r
CD
CD
CD
CD
CD
CD
CD
o
X
cc
ce:
UJ
CJ Q
UJ
UJ O f-
cc
ui or
UJ UJ
>- to
CC CD
I
V
I
I
I
I
\
V
I
I
a:
csl
O O
CO
CD
10
CD
UJ
:D
C3
UJ
oo
UJ
o
on
csl
LD
a
l i i i
i i i_
T3
CU
S-4
CU
CD 60
O -H
4J 3
CO 'O
CU
jz in
60
H J-i
f. o
4-1
m -H
CN C
o
cu s
I I
CU 3
S-i O
U-i r-l
en
0) O
.5 en
4J a
cfl O
E to
3 M
O 4-1
c
Q)
J= U
4-1 C
o
CO
c
o
H
±j
CO
J-i
4J
C
cu
CJ
c
o
H
^
^
CO
^
60
O
Jj
P ,
60
C
CU
oo
CO
J-i
CU
>
CO
j-i
3
O
r£-
1
-a u -H -, csio
a O 4-1 4J
c to -H co
cu c cu
3 OJ C ^3
cr 60 e eo
0) CO -H
J-l ^ O ^
<4-l 0) O
p- co in
cu co > csi
H M co cu
4-> 3 CJ f.
nj o S 4-1
31 cu *4-i
E , J-l O
.id 4J
C O
O CO
CO ,£) O
H CJ
J-l CO CU
CO
CO CO
C -H
3
o, a
B v-1
o B
CJ ^
I
Q
3
to
H
(i.
C CU
0) CU
f. 3
o
D-8
-------
UJ UJ
-=f > CO
cr CQ cc
O O
or
13
Ol
>
01
CO
43
o
4-1
CO
O>
43
60
H
^r4.
U~l
1
01
^t
6CM-I
d
H
S-i
3
T3
o-
1-4
O
4-1
01
>
H
4J
03
i 1
3
e
3
O
1
3
O
.-H
03
o
CO
d
O
H
4-1
cd
l-^
d 01 d
01 o 43 oi
43 S 4-> CJ
4-1 d
4-1 43 O
O
O)
00
03
U-l 03
O
CO
d d
o o
H -H
4-1 4-1 1-1 0)
3 0) 60 >
("1 1-4 O Cd
H 4-i S-J
!-J d CX 1-i
4JO) 3
CO O 60 O
d d
n o -H i
o >-> i
>> o
O rsu-i 4-i
d C -H co
a) c/> d cu
3 O 43
cr CD e 60
H
O 43
CJ
01 03 m
CU
l-i 03
03 4-1
CO 0)
l-l 01 43
o) 3
rH
3
3
CJ
CU "4-1
43 O
d
-I O
01 T3 O -H
4-1 3 t-l
O 03
14-1 !-l Ol
O to t*
d O T3
O 03 d
H Pi
l-i O
4-1 33
cn c/3
H
CO 42 O t3 >-,
r-l O ,0
1-1 CO CU >x
0) 3 CO CJ T3
p., d d 01
6 v-i oj 01 4-1
O E 43
3 ra
CT r-H
I
Q
oo
H
(Wdd)
D-9
-------
CO
12
O
or
a:
UJ
O
UJ
c±>
cn
UJ
-t >
cc
CC Qi
O H)
(- O
ii I
a
uj
UJ
co
CQ
o
a
UJ
f-
cc
O
_j
cn
o
I '
i
i
i
i
i
i
i
i
i
i
i
i
cn
en
cn
cn
cn
cn
cn
cn
cn
UD
cn
05 z
UJ
C3
UJ
co or
u.
UJ
r^
co
LO
"fr
CM
H-
a:
ID
O
o
-« .
'o
T3
0)
^ I
Ol 01 I
en >-i 3
,n 60 u-i o
o c r-i
ri Ol CO
4-1 )-J > O
W 3 -H
dJ T3 4.)
cn
Cfl C
il O
H 3 -H
JS }-i E 4->
O 3 CO
Ut 4-> U ^
CN
Ol
0) O
4-1
d
01
u
d
o
o
<4-l CO
O -H
CB 3 01
d d 60
o o E co
H -H Cfl i-l
4J 4-1 >-l Ol
3 CO 60 >
O CO
^
C CX >-i
4J Oi 3
CO O 60 O
H c d f.
T3 O -H I
u !-j ro
t^ o
CN1U
H 4-1
cn
cu
o
d o
0) CO
3
cr 01 E M
O) 60 -H
^ CO O
4-4 >-l CJ
Ol
Ol
Cfl
cn 0)
0)
LO
> > CN
3 4J
3 cn
O
x-x
0) T3
J2 C
4-1 3
O
U-l Jj
O 60
d
o
H
H Pi
M C
CO
d o T3 cn
O co d -H
cn ,n o 13
H o
VH cn 01 >~
co 3 co o
ex d d
E -H Ol Ol 4-1
o e A P
u ^ 4-1 cr
oo
I
a
3
6U
H
D-10
-------
T3
QJ
t I .
CD 0) 3
CD 60 >-( O
O -H
4-1 3
CB
OJ O
H CO
4-1 C
£ -H CJ 4-J
a c
o a) a)
a,
x: o
X 4-1 c
4-14J O
co x: o
M-t 4-1
O C/3 -H QJ
CO CO
O -H g ^
H 4J CO QJ
4-1 CO ^ >
3 M 00 cfl
X 4-1 O
H C (-J
QJ
O
3
O
en c 00 x:
H O C I
13 CJ -H , CslO
O O 4-1 4-J
C to -H W
a) Co)
Q) O X
' 60
3
cr1
QJ CO
}-i M O
M-( 0) CJ
> Cfl
0) cfl
> 4->
H VJ CO QJ
4J 3 CU X
cfl O 2 4-1
CN
I
14-1
O
3
3 CN
O C
s~* iw O
QJ ID O T-I
X C
4-1 3 M
O CO
O 60
3 Nl
n H
QJ -H Pi
en
C O t3
O cfl C -H
w x o "O
H o
M a) a)
CO 3
E -H
o
o
QJ
cj T3
c a)
B x: 3 co
I
c
3
GO
NOI1BH1N33N03
D-ll
-------
o
II
h-
cc
LU
CJ
O
cj a
LU
LU a h-
t±> LU cr
cr > _j
LU LU CJ
> CO _[
cr m cr
o o
O 13
I- O
I
CD
CD
CD
CD
CD
CD
CD
rw
o
X
CO
J 1 L
I I I L
CD
CD
ID
CD
CD
CD
LU
ZD
a
LU
Cd
U-
LU
CO
CJ
CO
CM
LO
O
O
o
o
-I .
'o
CU
t
CU
en
o
o
4-1
CO
cu
l~!
60
H
x:
u~^
CXI
CU
_(";
4-1
IH
O
C
O
H
4_J
3
H
H
4-1
CO
H
13
^"1
O
c
QJ
3
cr
cu
}_i
iw
cu
>
H
4-1
CO
i 1
3
B
3
O
cu
(~?
4-1
IW
O
C
0
en
H
^
CO
d
E
o
60
C
H
J-J
3
13
i/~t
^i
0
4J
H
c
O
^
4J
cfl
en
C
c
H
4-1
CO
!-
4->
C
HI
c
c
o
o
o
o
C/3
re
i-j
3
O
n
I
f-H
s~~^
*i3
c
3
0
^j
60
O
CO
o
CO
3
, c:
H
e
i
cu
)-i
I4H
CU
>
H
4-1
CO
,-H
3
B
3
O
cu
_r*1
4J
x:
4-1
H
^5
B
cfl
i-t
60
O
»-i
ex
60
C
H
}_i
O
44J
H
C
o
£
o
o
>
4-1
en
cu
£2
cu
^r"!
4-1
I4_i
O
}_l
CO
CU
p>-
13
C!
O
O
CU
en
cu
x:
1
3
O
r- 1
CO
CJ
en
C
o
H
4-J
CO
V^
4-1
C
cu
o
c:
o
o
cu
60
CO
Q>
^>
CO
i-j
3
O
r*
1
i i
4J
en
cu
Xi
6C
H
x:
CM
CU
c*,
4-1
M-t
O
C
o
H
4-1
3
o
H
. M
4 '
en
H
'O
r-*-!
CJ
c
cu
3
Cxi
f (
K
O
en
£*>
^
13
CU
4-1
CO
I
o
CU
t-l
60
H
(Wdd)
D-12
-------
O
tI
H-
cr
UJ
z
o
a
UJ
LD
uj a
cb uj a:
a: > _i
r^
UJ LU CJ
> cn _i
cc 03 cc
o cj
a; a:
o ID
l o
:z: i
O CO
CD
CD
O)
CD
CD
CD
_1_1 I 1_
1 I I I I L
0)
CO
UD
CD
CD
U
UJ
ZD
O
UJ
oo a:
u_
UJ
to
LD
OO
er
CJ
LD
O
O
o
T-t
o
M I
CU 01 I
CO S-i 3
o d
4-1
CO
01
x:
60
H
<~".
m
CNI
01
r^
4-1
14-J
O
C
O
H
3
XI
H
!-J
4-1
CO
H
13
j>^
O
C
0)
3
c*
Ol
(-J
IM
0)
>
H
4-1
CO
1t
3
e
3
CJ
01
r"!
4J
M-l
0
c:
o
CD
H
)_i
CO
P-
e
o
H
!-l
3
TD
m
^
o
4-1
H
£4
o
^
4-1
cO
en
d
o
H
CO
^
4J
C
01
O
d
o
o
CD
>
H
4-1
CO
r- 1
3
e
3
u
01
f]
4-1
f?
4J
H
^
e
CO
60
O
J-j
CX
60
C
H
>-i
O
CN1U
o
C/}
0)
CO
03
VJ
Ol
>
CO
)_i
3
O
r"!
1
ro
X X
11)
C
3
O
}-4
60
^
O
CO
Xi
en
3
C
-H
E
H
C
O
e
o
u
CO
>
4-1
CO
Ol
^2
0)
f!
4-1
I4_t
O
i-l
CO
0)
^
T3
C
O
O
cu
CO
Ol
X!
CO
U
CO
c
c
H
4_1
cfl
^
4-1
C
01
o
c
0
o
01
60
>->
>
cO
S-j
3
O
XI
1
f-1
4->
en
eu
Xi
60
H
XI
m
CN
0)
x:
4-1
M-4
O
d
o
H
4J
3
H
(-1
4-1
CO
H
TD
£>-,
O
c
O)
3
Es!
H
P^
O
X
C/]
>-.
r^
T3
Ol
4J
to
3
tc
H
'o
D-13
-------
^
Ol
CO
r*i
O
CO
01
_f?
60
H
n
m
CN
01
r?
4-1
O
C!
O
H
4-1
3
42
H
l-l
4J
CO
H
T3
O
C
O)
3
cr
01
i-i
<4-l
Ol
>
H
4J
CO
3
e
3
0
01
4-J
M-J
o
c
o
en
H
1-1
CO
d
6
o
60
C
H
U
r§
in
l~i
O
4J
H
£
O
S
4-1
CO
en
C
O
H
4-1
cd
i-i
4-1
C^
0)
CJ
Cl
0
a
C"
o
CO
cu
60
CO
M
01
>
CO
V-4
3
O
"f
-d-
CN
, ^
-o
C
3
o
l-<
oo
0
(0
43
en
3
. C
H
e
1
cu
^4
^4-1
cu
.^
4-1
CO
^-1
3
e
3
U
cu
4=
4-1
4=
H
3
6
cO
S-i
60
O
Jj
a.
1
3
O
t-l
cO
O
CO
c
0
H
4-1
CO
I-i
4->
C
01
o
c
0
a
QJ
60
cO
^-i
Ol
>
CO
!^
3
O
6042
c:
H
1-1
*JO
4-J
H
Cl
O
e
O
CJ
cfl
>
4-1
CO
0)
12
01
n
4-1
M-4
O
VJ
cfl
Ol
>-
a
c
o
CJ
01
CO
01
42
1
vj"
CN
4-1
en
cu
f!
6C
H
42
m
CN
01
_(~1
4-1
m
o
c
o
H
4-1
3
42
H
i 1-t
4J
CO
H
T3
K*"l
U
C
01
3
IS3
H
O^
O
EC
en
^
43
T3
CU
4J
Cfl
I
o
0)
1-1
60
o
D-14
-------
>l 1 1 I 1
1
1
I
1
1
1
1
1
1
1
1
J
1
1
1
1
1
1
- 1
1
1
1
1
t
\
~ I
_ 1
V
v.
s
^* .
^
1
~ I
1
1
\
_ \
1
1
1
1 1
1
\
: to
: o
i i
i h-
cr
cc a
: c
a; a;
o n
I O
>-i a:
"" -^£ 1
O *H
^
1 1 1 1 1
1
O
1 1 1 | 1
1
1
J
1
1
\
\
\
S
rH
[JT|
J dj:
** ~-I
j O
3 -4
a cc.
3 CJ
I
1
I
I
iii |
0
o
i 1 1 1 1 i 1
r*l i
*N
|».
cc
O
X
C/)
"*
_
""
\
\
I
\ :
\
\ "
\ -
\
\
\
\ :
N ~~
\
i -
\ :
J
_
:
i
\ ~
1
\
_
i
1 ~"
_
t ! f I t I 1
*-H
I
CD
CD
f~fi
O)
CD
CD
CD
CD
CD
CD
LiO
CD
CD
03
C*"*
CO
in
m
-t
CM
LD
o
,1
o
.
a
^
CJ
j<^
UJ
ZD
a
UJ
a;
u_
UJ
^^
14
1
CC
^3
z:
rD
CJ
CU
t, 1
CU CU 1
CO t-l 3
Q £j£ 14-, ^J
0 C rH
-H CU CO
4-1 S-l > U
CO 3 -H
a) t> 4J co
42 cfl d
00 ^D i 1 O
H 3 i-(
O 3 CO
m 4-i o i-i
CNI -H 4-1
d a) d
CU O 42 CU
J_J £^
4-J 42 O
4-1 CO 4J CJ
0 -H
CO > Q)
C d 60
O O E Cfl
H -H CC t-i
4J 4-1 ^ CU
3 CC 00 >
42 S-J O cfl
rt 4-J t-i
>-i d o. >-i
4-1 01 3
CO O M O
H d d 42
X) O -r-t 1
CJ h t l
>i O
CJ CM4J 4-1
d O -rt CO
3 ' 0 42
cr cu E be
0) 00 -H
i-j CO O 42
m i-i cj
cu cc m
CU > > CM
p> CC 4-1
H CO CU
4-1 J-J CU 42
cfl 3 S 4-J
iH O
3 42 CU 14-1
B 1 4= 0
3 *~~* 4-1
o d
/ . 4-| O
CU T3 O -H
42 C 4J
4-i 3 M 3 tsJ
O cfl 42 H
M-t t-J CU -rt PS
O 60 ^ t-i C
r^ 4-1 X
d O T3 CO LO
O CO d -H
CO 43 O TD >,
H CJ 42
i-i cn a) >^
CO 3 CO O T3
P, C C CU
E -H CU
-------
I I I 1 T
CO
-z.
O
cc
a:
UJ
CJ
z
o
o
ci>
cc
a;
UJ
cc
ct:
o
i
i i i i i
Qi
o
a i
UJ cc
a; r)
UJ o
CO _!
DO CC
O CJ
I
I
I
I
CD
CO
CD
cn
en
CD
CD
I I I I
CD
CD
UD
CD
CD
CJ
UJ
o
UJ
co a:
u_
UJ
ZD
z:
o ZD
o
cn
CM
to
o
cu
i-J
V
CO
XI 60
o d
co 3
0) 13
X
60 vD
O
m 4-i
CM -H
d
cu o
»^ ^H
4-1
4-1
m co
o
CO
d d
o o
H -H
4-1 4-1
3 cfl
X1 ^
H 4-J
!-i d
4-1 0)
CO O
H d
"O O
o
d O
<1) CO
3
cr a)
O
H
4J CO
co d
iH O
3 -H
e -w
3 cfl
u ^
4-1
ai d
X QJ
4-1 O
d
x: o
4-1 CJ
H
5 0)
60
' cfl
cd (-1
t-i 0)
60 >
O cfl
O^ ^_j
3
60 C
d xi
H I
H ro
O
U-l 4-1
i-l CO
d cu
O X
e 60
H
O X!
O
co m
> CM
4J
co a>
0) X
!3 4-1
C1J <4-l
o d
x-s M-i O
cu -a o 1-1
4J 3 S-i 3
O cfl 42
t4-4 V-< CD
(-1 O
4-J ffi
CO CO
d O T3
o co d -H
cn x o T3 >
H O X
I-i to 01 >,
cfl 3 CO O 13
a. d d cu
E -H QJ 0) 4-J
o E
CJ
I
Q
60
H
3
cr .-I
'o
(Wdd) NOUUM1N33NQ3
D-16
-------
I
~"
"~
-
""
~
-
""
^
-
-
*
_
_
-
~
-
-
_.
-
_
-
'-
~
-
:
-
-
i
>
o
\\ 1 1 I 1 t 1 J 1 1 1 i 1 1 1 1
ru
I
cc
o
w
-J
1
_
_
I \
\
11
\
I \
V \ *"
i I i
"
'
\
\
V 1
il
ii
'x 1 -
V 1
1 \
oo \ :
TZ. \
o \ :
i i \
M \ -
§ \ :
"* 1 2.
~^^ \ \
UJ \ \
0 I 1 :
i \ \ "
o * a \
' UJ \ I
CC > _J
oi a* ID
UJ ul o
CO > (T. _J
cc a: cc
c: o
a: ex ^ :
0 13 1
1 O 1
i-t 3T ' 1
1 i 1
^ 1 ' 1
o -* i
z: CM i
i i i i i i i i i i i i i i
i C\
ID i
CD
CD
CD
CD
CD
CD
CD
CD
*
\J3
CD
p
CD
00
r-
CO
U3
"^*
CO
CM
m
LO
O
^
.
o
^
CJ
'^t
UJ
C3
UJ
Q£
U-
UJ
>
i i
h-
cr
Z3
>
~n
O
T3
"
(U
CO
o
O
1 1
CD
0)
60
H
4U
LTl
ai
4-1
"4-1
O
c
0
H
4-1
^
^_£}
H
4-1
W
H
T3
J^
O
d
Q)
3
cr
}_i
U-l
CD
*>
V*
H
1 i
03
^
§
CJ
0)
4->
4-1
O
C
O
cn
H
^
CO
o
a
^~~\
i
a
CU
^
0£
H
l-w
M
d
H
^
3
13
V4O
^1
o
4-1
H
C!
O
i^
ICLt
4-1
Cfl
cn
d
o
H
4J
cfl
J_l
4-1
d
CU
U
d
o
o
c
O
CO
0)
CO
Vi
Ol
J>
CO
S-i
^j
O
-f
CN
^^
T3
§
O
5-1
60
O
cfl
cn
;3
^
1
0)
^_l
U-l
ai
j>
H
4-1
CO
r-H
3
E
3
O
0)
4-1
4J
H
^
E
cfl
^t
60
O
}-(
a
60
c
H
M
-JO
4-1
H
d
o
o
CJ
Cfl
J_J
Cfl
0)
£~£
0)
4-1
m
0
»-i
cfl
01
T3
d
O
CJ
Cl)
en
0)
4-1
1
CJ
i 1
Cfl
CJ
en
d
o
-H
4-1
CO
V«i
4J
d
0)
CJ
d
0
CJ
0)
60
cfl
5-i
CD
£>
cO
^|
3
O
4=
1
^3-
.
43
a
0)
4J
CO
rH
NOUBH1N33N03
D-17
-------
I I 1 i t
_
-
_
_
-
__
-
-
_
_
..
_
-
-
: co
: o
i
2. f
CC
: a:
- i
r^
"Z,
UJ
: o
o
I o
- UJ
C±>
cc
a;
UJ
r- >
r cc
I C£ C£
o :D
i-o
:z i
O i
zr
-
iiii i
I
a
UJ
a:
UJ
CO
m
o
i i ) t i 1 1 t i 1 E
i '
\
\
i
\
i
V
1
'
I
\
\
1 i
n~*
! ^
i i
i *"
-/ »
t
j
. ~
' -
1
i
V \
\ \
. \
. \ ]]
1 \ ""
\ ~
( \
1 \
. \ ~
1 1 2
\
\ \
\ \
\ \ j;
\ \
1 \
1 V
1 \ -
11 *"
V
\ 1
1 "
1 I :
\ \
i \
\
1
\ 2
:
UJ \
1 , _
5 1 1 -
_j \
cc \ -
0 1
1
i :
! >
i
i
i i i i i i t i i i i
cn _
CO ^
en >
cn M i
QJ CU 1
cn !-i 3
fi C30 4-1 O
0 C2 r-l
H 01 CO
CD W ^ > °
gj cn 3 -H
cn cu *o *-> cn
cn 42 co c
60 r-- rH o
H 3 -H
42 ^-i S 4-1
O 3 CO
u-l 4J O V-l
CN -H 4-1
£2 CU £2
CU O 42 CU
42 S 4J CJ
cn w c
OJ 4-i 42 O
' "4-1 CO 4-1 O
0 -H
cn ^ cu
C C 60
o o 6 co
H -H CO i-t
4J 4-1 J-l CU
3 CO 60 >
_ 43 *< O CO
JJ -H 4J !-i
cn v-i c: a, )-i
4-1 CU 3
cn o 60 o
>. -H C C 42
(j T3 O -H 1
cn .22, cj M i
* ^^ Q
UJ U CNtU 4->
ro c o -H cn
C3 cu co c cu
III 3 0 42
£ cr cu B 60
Ou Lt- CU 60 -i-l
U_ VJ CO O 42
U-l }-i CJ
UJ CU CO u-i
-^ ^ OJ > > CN1
r_i > co w
I"1 -i-i cn cu
r" 4-1 >J CU 42
CC cfl 3 [2 4-1
CO I r-i 0
=1 3 42 CU U-i
=2 E 1 42 0
* 3 i-i 4-1
\n ^ o c
c_) ^~, m o
CU T3 O -H
. 42 t2 4J .
^f 4-1 3 i-i 3 M
O CO 43 H
^ 0 60 >-, M O
CO 44 4-> 33
c2 o -n cn co
O CO £2 -T-I
cn 42 o 'O >>
^i 'H U 43
Csl M w a» p>>
co 3 cn o TD
a. c c cu
6 -H CU CU 4J
O E 42 3 CO
U ^ 4-1 CT ^H
^-4
.
^_,
g c
cu
3
60
H
-^
H 0 *-< .
o
o
D-18
-------
I i i i 1 1 i i i 1 1 1 I 1 1 1 i t
i
i
i
i
i
i ~
t
i
i
P
1 tr
o
j .x.
. (/) _
l
I
I
^
"" l 1
I
- 1
1 I
\ \
s, X.
\ ^v
\ \.
\ >v
I \ ^**s I
- 1 "
1 ""
1 -
, 1
I V \
1 \
\ \
_ \ \
- \ ^N :
__. . \ _
^ 1
"* \
: o ', :
i i \ \
if \ \ i
en ^ i
: OL i :
1 . \ \
r? 1 \
: UJ ' \ :
: o i \ :
TO \ \ ~.
u a \ \
UJ 1
- UJ a i * \ -
cb UJ cr \ \ -
a: > _i \ \ -
ex. oz ^ \
UJ UJ O \
c-^ > co _i ;
- a: m cp ', -
I o; a: ! I
- 2 g ' -
H, ^ i -
~^ i I ,
o en
z:
1 1 1 1 1 1 | 1 1 1 1 1 1 | 1 ! 1
< O ,-
CO O '
0) v
CO W !-i 3
f> txm-t O
en 3 -H
cu -a 4-1 en
co x! co c:
CO 60 r-~ i I o
CO -H 3 -H
.^ M B 4J
O 3 CO
u~| 4-1 O S-l
oa -i-i 4J
C CD C
cu o x: cu
rC S 4-1 0
4-1 C
4-1 f. O
Q) H-l CO 4-1 O
en o -H
u; w s cu
C C 60
o o e co
H -H tfl >-l
4-1 4-1 S-J CU
3 CO 60 >
-Q (-1 O cfl
H 4-1 >-l
^ c a, >-i
LD 4-i 0) 3
CO CO O 60 O
"H C C- S^
a o -H i
>_ O ^J 01
f s £^ O
CO i^ O CNMJ 4J
-^ c! O -H tn
LU a) co c cu
ID 3 04=
a cr cu E 60
m£ 2 ^°2
CO Q- u-i !-i O
U_ cu co ui
QJ > > 04
m > co 4-1
_ CT -H CO QJ
^ -r. 4-1 >-i cu x
l~t cfl 3 3 4J
1 i 1 O
CC 3 4= a) 4-i
CD _j e i ,c o
=i 3 ro 4J
r: o c
-^ j^-« s l[ | Q
U3 ZD CU T3 O -H
t_) 42 C 4-1
4-1 3 S-i 3 Nl
_. O CO 42 H
^r M-I )-i QJ -H K
O 60 >> )-< O
^ 4-1 ffi
d O T3 CO CO
CO O co C -H
. en 43 O 13 >
H O 43
S-l CO CU >>
CO 3 to U 13
CN ex C C OJ
6 -H QJ CU 4J
0 6 4= 3 CO
cj - 4-1 cr i-i
^H
1
o
10 cu
O M
3
60
H
fii
^
^
CO
(Mdd)
D-19
-------
to
o
cc
ct:
UJ
o
o
CJ
s
a
UJ
c-> :>
a:
tz. oz ^
UJ UJ U
to _i
CC 03 CC
O O
a: ft.
o n
h- o
'"^ -I.
2: i
o ^j-
z: CM
CD
CD
]CD
'CD
CD
CD
CD
O
to
1 1 1 ! 1 1 1 |
CD
CD
LD
CD
00
UJ
Z3
a
UJ
UJ
* ZD
zr
LO m
o
CO
CM
LO
0
o
'o
T-t
NOI1BH1N33N03
D-20
CU
en
.0
o
4-1
en
01
f?
6C
H
,j~!
U~l
CN1
CU
_[~]
4-1
t-W
o
C!
O
H
4-1
3
O
H
S-i
4-1
en
H
13
>>
O
Ci
01
3
cr
CU
^
y-i
CU
>
H
4-1
CO
3
e
3
O
CU
ri
4-J
4-1
60
Ci
H
)J
3
T3
i^-
!-i
O
4J
H
fi
O
^
4-1
CO
en
C
O
4-J
cO
t-i
4-1
C
CU
u
C
O
o
CU
>-j
4-1
01
>
H
4-1
CO
<-{
3
6
3
CJ
CU
n
4-1
X!
1)
H
s
e
CO
^_l
M
O
)-i
cx
60
C
H
S_j
3
O
i 1
CO
O
W
d
O
H
4J
CO
^-i
4-1
C
CU
o
C
0
0
Ol
6C
CO
)-l
0)
^>
CO
M
3
0
1
co
Vj
3
O
1
.^j-
OJ
^ N
'O
C
3
O
>-j
4-1
,-j
d
0
B
0
CJ
ca
>
4J
CO
Ol
3
CU
f~l
4-1
4-4
O
(-1
03
Ol
4-1
CO
01
JZ
6C
H
(~l
LO
CN
Ol
f~^
4-1
M-l
o
C
0
H
4-1
3
rO
H
»
C-0
H
(£
O 60
S-i O
O T3 CD c/3
O cfl
CO ,£] O
H CJ
j-i cn 01
cO .
a, d
C -H
>,
en o
C!
01
E -H 01 Ol 4-1
o B j2 3 ca
O "^ 4-1 CT1 rH
00
rH
Q
0)
3
t£
-------
CO
o
cr
a;
LU
o
o
o a
LU
LU a i
o LU cr
en > _i
LU LU O
OO > CO _I
cr CD en
o o
tx az
o :D
i o
CJ3
03
103
O3
03
O3
O3
ff
a:
o
x
to
i i i
J I II 1 L
03
O3
LD
O3
O5
00
CO
LO
CO
CM
O
LU
ZD
LU
U_
LU
cn
0
LO
O
O
O
O
'o
'"a"
)_,
Ol
CO
,0
O
4-1
CO
01
rf"1
oo
d
H
^-1
3
'O
1
Ol
^1
M-l
01
£>
H
4-1
CO
00 CO r-H
H
r"?
t/^
CN
Ol
_r^
4-1
M-J
O
d
o
H
4-1
3
H
t-i
4-1
CO
H
ID
K*I
O
d
Ol
3
cr
01
j_i
14-1
o>
£>
H
4-1
CO
t-H
3
e
3
O
0)
^ f.
4-1
UH
O
d
o
CO
H
VJ
ca
P,
e
o
u
^_l
o
4-1
H
d
o
S
4-1
CO
CO
d
0
H
4-J
cfl
^_i
3
O
,d
I
p-H
, <
t3
d
3
c
t-l
oo
sj
o
CO
_(~1
CO
3
d
H
g
N '
3
e
3
CJ
CD
f!
4-1
r^
4-1
H
^
e
CO
^
00
o
^-1
a.
00
d
H
j-i
0
NlU
H
d
0
e
0
CJ
CO
£>
4J
CO
Ol
Is
QJ
^C
4-J
^4-4
0
^_l
CO
01
p>^
T3
d
o
CJ
o
co
Ol
(~!
4-1
1
3
O
I 1
ca
CJ
CO
d
o
H
4-1
CO
j-j
4.)
d
0)
o
d
0
o
0)
oc
ca
^_i
0)
^>
ca
^_j
3
O
,(~!
1
4-1
CO
Ol
,d
60
H
_ r~|
in
CNl
01
^;
4-1
M-l
O
d
o
H
4-1
3
t(~\
H
J-l
4-1
CO
H
-a
£>*,
o
d
Ol
3
cr
NJ
H
Pi
O
EE
C/3
^>>,
f^i
T3
01
4-1
CO
ri
I
O
3
ft.
D-21
-------
1
I
I
I
I
QL
O
X
oo
CO
o
II
I
a:
i
2:
LU
o
o a
ua
UJ a t
ci> LU en
a: :> _i
a: or ZD
LU UJ CJ
> co _j
cc ca a:
o o
l- o
O 00
CD
en
CD
en
CD
en
CD
CD
CD
Jl
CD
CD
UJ
r3
a
LU
00
UJ
>
I I
CO 5
2:
in =>
o
LD
a
T3
CD
>
0)
CO
o
O
4J
CO
60'
CJ
H
t-t
3
o>
^-t
4-1
t
OJ
>
H
01 T3 4J
c~|
cC
60 CO i 1
H
n
U~l
CM
0)
f.
4-1
4-1
O
c
o
H
4-1
3
o
H
J-i
4-1
CO
H
T3
J>^
O
c
0)
3
cr
0)
^i
4-1
0)
>
H
4-1
C3
i |
3
e
3
o
0)
4=
4-1
4-1
O
C
O
CO
H
±4
cd
&
B
o
o
^j
0
4J
H
£^
O
^
4_J
ta
CO
c
O
H
4J
a!
J-i
4J
C
0)
o
£i
O
o
r
O
C/)
0)
60
cd
}_4
a)
>
cB
^
3
O
n
1
CO
X N
-a
c
3
O
3
B
o
0)
J3
4_l
rt~\
4-J
-H
^
6
^_|
60
0
^4
a,
60
c
,-j
M
o
sJU
H
a
o
B
o
o
co
>
4-1
CO
01
3
0)
n
4-1
4-1
O
^1
cd
0)
\
3
O
1
cd
o
CO
c
o
H
4-1
CO
^4
U
C
0)
u
c
o
u
a)
60
cfl
t-i
0)
£>
cd
v~i
3
0
rf
1
CO
4_)
CO
ai
f~!
60
H
f?
m
CM
a)
f~i
4-1
4-J
O
c
o
H
4-1
3
H
bO >, l-i
^
u
CO
o
CO
3
*H
B
^^
a
f^
o
o
0)
CO
Q)
{~?
' 4->
4J
CO
H
T3
P*~>
O
c
cu
t
cr
tv]
H
Pd
O
JT]
CO
^>
,o
TD
OJ
4-J
cd
i !
o
CM
0)
t-1
o
H
O
D-22
-------
to
z
a
cc
a:
UJ
z
o
o o
UJ
uj a h-
c±> uJ cr
cr > _i
UJ UJ CJ
co >- co _i
cr en cr
o o
a: ce
o
in
i
«*
i i i
p
cr
o
:c
V)
CD
CD
CD
CD
05
CD
CD
CD
CD
U3
CD
CD
00
o
UJ
73
C3
UJ
u_
UJ
=3
X
UD ZD
CJ
-=!
00
ICsl
LD
O
a)
en
I I
CU 3
l-i O
o -H
en 13
01
f, 00
60
H S-J
.£ O
0) o
>
4-> d
o
g CO
-H o
03
en -H CD
& 60
6 2
H 4-1 CO CU
4J CO f-l >
3 S-J 60 cfl
43 4-i O
H d i-l J-l
J-l 0) OH 3
4-1 O O
en d 60j=
H o d i
S-i CM
>i CNO
CJ O 4J 4-1
C
a)
-H
c
a) o 42
cr 60 6 60
CD CO -H
>-i t-i O JS
i*-( 01 O
> co m
CU CO > CN
f> 4-1
H !-( W 0)
4-1 3
-------
tilt 1
_
.
-
_
-
_
.
-
_
"
-
_
_
_
-
: cn
: o
i i
cn
- ce:
- \ ,
r^
~ >
*£,
UJ
: o
0
: o
UJ C
cb u
cn i
a: a
UJ U
cn r>
en a
I c
a: a;
CD o
h- 0
_ ^- i
CD »-
z:
i
o
t i i i j i i a
i
i
t
*
i
i
i
i
*
\
i
i
i
i
i
i
i
i
i
i
i
\
\
\
\
i
1
\
\
\
\
\
\
\
\
\
y
\
a
UJ
3 J
j cn
X* ~}
J CJ
3 _J
3 cn
3 O
1
1
1
1
1
1
II 1 till
CD
o
ill 1
_
Ml
£
o
I
cn
"
-
«,
_
_
_
1
I ^
\ _
I
\ 2
V
\ j
v \
\ V -
\ ^ j;
' I
\ \
I \ _
1 \
i\
1 \
\ \
\ I ;
i \ 1
i \
i \
t \
1
1
I "
I 1 1 1 1
v~
I
cn
cn
^T^
Of
O)
cn
cn
cn
cn
cn
U3
cn
^-
°* 2:
" UJ
Z3
C3
UJ
co Qi
U-
UJ
1> >
t 1
1 .
J .
co^
" ~>
2T
U3 ^
CJ
.
*^t*
CO
CM
j
*
U)
o
< .
0
-a
0)
0) 0) 1
w h 3
J3 &D U-i O
O C r-l
H CU cfl
4J M > O
cn 3 -H
01 T3 4J cn
& co c:
60 C^ tH O
H 3 -H
ja >-i e 4J
O 3 cfl
ID 4-1 O l-i
CM -H 4-1
c oi c
o> o ,c o)
^2 S 4J CJ
4-1 C!
4-> X O
m cfl 4J cj
O -H
cn 3 CD
C C! 60
0 0 6 cfl
H -H cfl S-i
4J 4J f-l CU
3 cfl 60 >
,jh S-J O cfl
H 4-1 S-i
S-i Ci tX U
u a 3
CO CJ 60 O
H C C X
T3 O -H 1
O S-i i i
>> o
O CNHJ 4-1
d O -H cfl
0) CO C 0)
3 O rC
cr 01 e eo
OJ 60 -t-l
S-l CT) 0 ^
14-1 S-I O
o) cfl in
OI > > CNI
> CO 4-1
H CO 0)
*j s-j ai js
cfl 3 S 4J
i-H O
3 ^ 0) M-l
e i ^ o
^3 i 1 4_)
0 G
^^N *4-1 O
01 T3 O -H
j: c 4-1
4-1 3 (-1 3 N5
O cfl .a H
14-1 s-i o) -H erf
O 60 P>^ S-i O
v; 4-» rn
C o T3 cn co
o ca d -H
CO ^Q O T3 ^
H O ,£3
S-l W 0) >-,
cfl 3 co a TD
o, a co
E -H Ol 0) 4.)
O E rC 3 Cfl
CJ ' 4-1 CT rH
CNI
CNI
Q
1)
S-i
3
H
NOI1BH1N33N03
D-24
-------
o
X
o>
O)
7 en
en
CD
en
05
CO
o
tI
H-
CE
UJ
o
z
o
o
a:
a;
UJ
O ID
\- O
1-1 3:
z. \
o en
a
LU
a h-
UJ CC
:> _j
a; ZD
UJ O
CO _!
03 o:
o o
en
en
LO
CD
III
rs
C3
UJ
oo ft:
UJ
cr
_j
UD
CJ
cn
CM
UD
O
M 1
(D CU 1
en >-< 3
42 60 4-1 U
O d r-i
H Q> cfl
4J J-l > O
en 3 -H
cu T3 4-J en
42 cfl d
60 ON i-H O
OJ
Jj
4_)
*-M
0
d
o
H
4J
3
43
H
^
4-1
en
H
*O
p>^
CJ
d
cu
^j
0*
0)
J-l
14-1
CD
>
H
4J
CO
i i
3
6
3
CJ
cu
^r?
4_J
<4-l
O
d
o
en
H
^
03
a.
6
o
d
o
^r]
4_)
Cfl
en
d
o
H
4_J
cd
^
4_l
d
CU
o
d
o
CJ
0
O
oo
0)
60
CO
j_j
0)
>
cfl
£_l
3
O
r^
1
CO
^~s
'TU
d
3
O
^_i
60
O
cfl
o
en
^3
d
H
e
OJ
n
4_J
t~|
4-1
H
^
e
cfl
}_i
60
O
^_j
G.
60
d
H
t-i
O
sM-l
H
d
o
e
o
o
CO
>
4-1
en
cu
3
0)
r~!
4-1
U-4
O
V4
cfl
CU
!^>
T3
c^
O
CJ
eu
en
cu
42
d
CU
O
d
o
o
cu
60
cfl
^
cu
£>
cfl
^_J
^
o
42
1
o-i
4-1
eo
cu
42
60
»!-t
42
LO
CN
Q)
42
4-1
<4^
C
d
o
H
4J
3
o
H
t-1
^J
en
H
*a
^
o
d
01
3
Cs]
H
Pi
0
^"T-<
K*^
O
13
CU
4-1
rd
O
Csl
I
Q
3
CO
O
-H ,
a
D-25
-------
i\iiiir
i r
tn
z
o
a:
ct:
UJ
O
O
o a
UJ
uj a f
o ui cc
cc > _i
ex. a: ID
UJ UJ O
cn aa cn
O CJ
O ID
f- O
*"^ _L-
O5
O)
TCD
cn
CD
cn
OJ
P
cc
o
X
CO
CD
in
CD
>-
CJ
3
UJ
C3
UJ
u_
UJ
ID
ZL
n ^3
*
m
cn
CM
LO
O
M
01
CO
^rs
0
4-1
CO
01
f^
60
H
rr
u-i
CM
01
f"!
4-1
14-4
0
c
o
H
4-1
3
fl
H
^
4J
0}
H
TJ
O
C
Ol
3
c-
Ol
i-l
M-i
0)
H
4-J
CO
[
3
e
3
U
01
_<""
4-1
<4-l
O
C
o
cn
H
v^
cfl
O.
e
o
U
CM
1
01
3
60
C
1-1
}~l
3
tJ
o°\
Jj
o
4J
H
C
0
z.
4J
cO
CO
c;
o
H
4J
CO
v^
4-1
C
^J
o
c
o
a
o
c/>
0)
ec
n3
j-(
QJ
^r
CO
jj
^2
O
1
CM
xx
a
d
3
o
^l
co
l*st
a
CO
r^
U5
3
, C
r-t
E
1
01
J_4
U-i
0)
^,
H
4-1
CO
i 1
3
E
3
O
<3J
4J
f
4-1
H
^
E
CO
j-j
60
O
^i
p f
1
3
CJ
I1
CO
O
cn
c
o
H
4-1
G3
^4
4-1
C
01
O
c
O
CJ
01
60
CO
i-i
a
[>
CO
S-t
3
O
60 ,C
C
I
«-\
VJt-J
H
d
0
E
o
CJ
cO
>
cn
0)
S
0)
4J
4-1
O
i-l
CO
01
>
r^J
C
o
CJ
at
cn
OJ
TP;
4-1
1
^3
cr
OJ
C-H
p^|
O
a:
CO
K*~>
r\
TD
Ol
4_J
CO
11
cc
H
to
D-2A
-------
1 ! i 1 ; i i i
1
1
1
1
1
1
i
i
-
_
-
~*
-
..
-
_
: co
: o
- t-H
- i ^
if a:
tK
1
-^
: UJ
: o
I*£
o
: o o
UJ
uj a H-
cb ui cr
cr > _i
a: oi ID
UJ Ui O
o :» co _i
-- cr ca cr
I oo
a; a;
o ra i
^ 0 1
n 3: i
-z. \ \
O ^ 1
z: i
! ! 1 1 1 I 1 1
>
O
j i i ; ; ; ; i !
~
~
rw
H
flC - -
I 0
i X
I CO
I
1
I
1
t -
1
I
I
I
1
-
-
\ ~~
\ __
1 ~
\
\l
\\
V :
u
1\ i
u
v\ =
\ -
1 \
t ^ ^
' \
1 \ -
1 \
1 \
1
\ ^
\ \ -
\ \ -
\ \ -
I
1
1 i
1 1 1 1 1 1 1
T-. CS
I C3 '
0) ^
O) >
C7> a) a) i
. CO 60 M 3
42 C 4-< O
O -H t t
s-i a 03
4J 3 > 0
CO T3 -H
OJ a> 4-1 cn
C75 4-- O 03 G
rn 60 -! r-t o
. -H 3 -H
4= l-i E 4-i
0 3 03
LO 4-1 CJ ^1
CM -H 4J
C 01 C
a> o 42 aj
4^ S 4J 0
4J G
O) w "5 °
en 3 01
G C 60
0 0 S 03
H -H 03 S-i
4-J 4-J J-< O)
3 03 60 >
4D S-J O 03
H 4-) S-i
O> 4J (U 3
. CO O 60 O
r^ C C 42
.. T3 O -H 1
t\ O S-i ^
o) y >^ °
. 1Z O CM4-J 4J
[|l G O -H cn
^ QJ CO d Ol
O3 O 42
... cr a) e 60
UJ 0) 60 -H
CO Of >-i o) O 42
. n_ in s-i o
-H cn 01
»-* 4J s-i ai 4:
| 03 3 13 4J
cr >-i o
CO 3 42 OJ <4-l
. =i g 1 42 0
-J 3 . < 4->
z: o G
LO "^1 -^"^ ^^ C
. {_j 0) T3 O -H
4J 3 J-i 3 N]
xj" O CE ^3 f-*
M-I j-i 01 -H pi
O 60 >> S-i O
,M 4-1 02
(n G O T3 CO CO
, O 0) G -H
10 42 O T3 >-
H 0 43
M CO 0) >,
CJ OS 3 CO CJ 13
, p. d da)
E -H a) oj 4-1
O E 42 3 03
U '4-1 CT1 rH
L/1
CN;
1
in
O aj
* p
6C'
H
r-4
.
O
NOI1BH1N33N03
D-27
-------
to
z
o
UJ
o
o
o a
UJ
UJ a t-
o uj cc
_i
a: a: ra
UJ UJ CJ
> to _j
CD
03 60
rf) pj
O -H
4-1 3
03 T3
0)
42 O
I
CU I
f-i 3
4-) O
rH
0) nj
> o
H
4-1 03
ca d
i-l O
3 -H
g 4J
3 cS
CNJ -H
d
V O
4-1
d
QJ
o
d
o
o
0)
60
cfl
4-1
MH 03
O
03
d c:
o o
H -H
4-1 4-1 1-1
3 a 60
43 V* O
H 4J p
4-1 2
03 U 60
H C d
T3 O -H
O >-l
K^> O
O CM4-)
d o -H
> CM
> a) 4J
H 03 y> a) 42
CO 3 3 4J
H O
3 4= 0)
g I 42
3 CO 4-1
O
XS U-l
ai -o o
42 C
4-> 3 M
O CO
4H ^ d)
O 60 >-
i-i
3
O
JS
I
co
03
0)
-C
60
H
f.
U-l
O
C!
O
H
d O T3
o co d
03 43 O
H CJ
>-< cn a)
CO 3 03
a. d
6 v-t Q)
O g 42
CM
I
o
OJ
t-l
3
62
l-i O
4-1 CC
CO CO
H
-o ^
U
d
OJ
3
cr I
-------
CO
o
ii
\-
cc
h-
UJ
CJ
o
o a
UJ
UJ a \-
cb LU cc
a: > i
r(< r>-r ~i
UJ LU CJ
CC CD CC
O CJ
O
O
z:
CD
in
i
CM
CD
cn
03
CD
CO
CD
CD
o:
o
CO
CD
05
U3
CD
03 ^
UJ
a
UJ
oo a:
UJ
>
in
o
in
o
S-i
CU
CO
42
O
4-1
CO
O>
f!
OC
H
4=
in
CM
cu
4=
4-1
14-1
O
C
C
H
4J
3
42
H
jj
4-1
CO
H
ID
CJ
C
01
3
0"
0)
S-i
4-1
01
H
4-J
cd
j
e
3
O
0)
4=
^j
"4-1
O
C
o
cn
H
j_i
03
p,
E
O
CJ
oc
c.
H
S-i
3
""O
o
(1
J-l
o
4-1
H
C
0
s
4-J
CO
cn
C
O
H
4-1
ct
^
4J
C
0)
u
c
o
CJ
C*
o
C/3
0)
OC
CO
S-I
cu
£>
cd
s^
o
i
-3-
CM
s~^
^
d
3
O
S-i
oc
A!
CJ
CO
o
cn
3
C
H
E
x^-
1
cu
S-i
M-i
01
^
H
4_)
CO
rH
3
E
3
O
CU
r"
4-1
f"
4-1
H
^
E
CO
S_i
oo
0
S-i
p 1
00
c
H
S-i
^JO
4-1
H
c
o
6
o
CJ
CB
>
CO
S
0)
^c
4-1
U-!
o
J^
CO
cu
K*^
T3
d
o
CJ
cu
cn
cu
r~|
4-1
1
3
CJ
i i
03
CJ
cn
c
o
H
4J
CO
S-j
4_)
c
01
CJ
c
o
CJ
0)
00
CO
J-4
O
£>
CO
J_i
D
O
*J
i
^j-
CM
4-1
CO
0)
f!
to
H
r^
u~*
CM
OJ
4-1
14H
C
c
0
H
4-1
3
H
i S-i
4-1
cn
r-l
T3
^>~,
CJ
d
0)
3
cr
N]
H
P^
O
£C
CO
^>-,
_o
T3
CU
4-J
CO
r~ 1
CM
I
O
CU
S-i
3
(X)
D-29
-------
jl 1 f 1 J 1 I I |
1
I
\
1
i
i
|
1
- 1
1
1
1
" 1
1
- 1
1
\
V
" \
V
1 i
\
I
I
\
\
\
I
I
I
I I
i
\
: so
~. c*
- j^ j
en '
: ^ \
'- m i
- HT \
I -^ 1
> * 1
cj ' a
I ' UJ
UJ b H-
cb fij cc
cc ? i
CO IV fr* ' 1
i>- |^i jTi r j
o i> to _j
7 h- cc op cr
i i O CJ
I "2. CX '
o :D
- = i
i j i
_i *-<
a:
i i i i i i i i i
H O
o o
1 1 1 1 1 1 t
rw
fe
o
X
(/J _
"
-
;
-
_
-
"
_
-
:
i
-
-
\ i
;
-
;
\ ]
,
1
1 1 1 | t I 1
^_
CD
CD
C>
CD
CD
CD
CD
CO
to
CD
*
^~
CD y
' UJ
C3
UJ
oo ttr
U
UJ
C** -s>"
' 1
1
CO ^~i
=3
LO Z3
CJ
(T)
CM
*
» 1
*
U3
O
i-i
'
O
o
tu
>
VJ
(LI
CO
fi
0
4-J
to
tu
j2
60
H
r"
UTl
CN1
CD
42
4-1
M-l
O
d
o
H
4-1
3
r>
H
l-i
4-1
CO
H
ID
>.
O
d
0)
3
cr
tu
i_i
14-1
0)
>
H
4-J
CO
t 1
3
B
3
O
(1)
4-1
^4-4
o
d
o
w
H
)-l
CO
p.
g
O
o
CO
CN
1
Q
H
4-1
CO
r i
3
^
3
U
tu
42
4->
^c
4-1
H
£
0
cd
^-i
60
O
1-4
p.
60
d
H
l-i
O
44-1
H
d
o
e
o
o
cfl
£>
4J
to
0)
13
0)
^r*!
4-1
U-l
O
J-4
Cfl
T)
d
o
o
a)
CO
1)
4=
4-J
1
3
O
( j
rt
o
to
d
0
H
4-1
CO
4-1
d
tu
o
0
CJ
(LI
oo
CO
1-t
tu
>
cfl
^_t
3
O
J3
1
14
4_J
to
0)
_(~^
60
H
_r^
u-j
CN
QJ
(~*
4-1
+-I
0
d
o
H
4-J
3
42
H
4-1
CO
H
-a
£>^
CJ
d
cu
3
tr1
tsj
H
&
O
ffl
CO
>.
o
T3
0)
4-)
CO
i 1
D-30
-------
ill! 1
-
"
_
-
_
-
-
_
_
: co
: o
f-i
i i
en
: a;
-_ t~
_ ^2
UJ
: cj
o
: cj
UJ
cb
cr
CO K.
Ct: UJ
o >
t cr
i i
I 2 CK
a rD
2T O
_! \
_i en
cr
III! |
1
O
n 1 i i i rti i i i i i
i '
i
i
i
\
i
^
i _
i
t
i
i
P
ry
\ o
T~
i *JL-
i <"
I
t
I
^
i
_
"
I ~
I
\
\
t
| 1
\ \
1 ^ :
v 1
i _
\ ;
] __
I 1
I \
I ^ ;
I
1 1 I
* 1 -
\ 1
1 \ -
1 1
\ _
I \
1 j I
a \ I
UJ \ \ H
a t- v \
UJ cr
> _i
a: HD
UJ CJ
CO _J
to cr
O CJ
\ '-
I
I
I
ii i 1 i i i i i i
o
o '
CD
CD
U3
CD
«
CD
CD
CD
CD
CD
*
LO
CD
*
^
CD "
^*
UJ
~-)
C3
UJ
oo a;
u_
UJ
r*" t^*
. i
1-
"d
3C
U3 =J
O
»
CO
Csl
m
in
o
^_,
.
0
'o
a)
>
V^
0)
CO
42
4J
CO
0)
60
H
LO
CN
0)
4-1
O
d
o
r-i
4-1
3
_ri
.,_!
4J
W
H
T3
^
U
d
a)
cr1
OJ
M
4-4
0)
>
H
4J
CO
I (
3
e
o
0)
42
4-1
4-1
O
d
0
CO
H
^
n3
fX
O
CNI
1
Q
CD
3
60
H
60
d
rl
^j
rQ
CO
O
4-1
H
O
i \
, |
n3
4J
cfl
CO
d
o
H
4-1
CO
U
4J
d
0)
o
d
o
CJ
c*
O
CO
0)
CO
cfl
m
Jo
J»l
3
O
42
rn
^--^
d
3
o
J_l
co
CJ
ca
43
CO
d
H
e
1
OJ
>H
4-4
CD
>
H
Cfl
iH
5
e
CJ
0)
4-1
f.
H
^
B
CO
^-i
60
O
P.
60
d
H
o
^«-l
H
d
e
o
CJ
CO
>
4->
CO
01
ts
a)
^
4-1
o
>-!
cO
cu
T3
d
o
o
CD
en
cu
42
4-1
1
3
9
CO
O
CO
d
o
H
4-1
CO
t-i
4-J
d
o
o
o
cu
60
CO
^
0)
>
CO
j_i
3
O
f~i
1
m
4-1
en
cu
60
H
U~l
CNI
0)
4-1
4-4
O
d
o
H
4-1
3
H
4-1
CO
H
o
^
quenc
IS]
H
J-V|
O
CO
>^
43
T3
0)
cfl
1 1
NOIiUyiN33N03
D-31
-------
i * i i
-
-
-
i
i
\
V
\
V
\
\
1
1
\
\
\
1
1
~_ V.
-
: co
; o
i f
i
i
cr
: a;
-_ \-
' LU
: cj
f~\
I CJ
- LU C
c±> u
cr ^
co az a
a; LU u
0 > U
i cr a
I "-i C
:z or
o ra
zz o
31
I |
I -t
cr CM
i i i
3
O
III I 1
-
-
w
*y^,
0
7/>
{//
-
~
\-
~
-
_
-
_
:
~
-
-
:
-
~
.
2
-
~
_
~
-
-
-
2
:
-
'- 3 « 1 :
J CJ ' \ I
3 -J
3 cr
3 cj i :
! 1
1 '
!
T~
1
CD 'o
CD 01
t-fn > h£
m Sell
UJ OJ -H 0) 3
' to t-i !-i CJ
43 3 4-1 i-l
O PO ^3
Q) O
4-1 UJ p>
QJ CO (-1 T-I W
/-n o) o 4-1 ti
%> 42 4J 03 0
03 OO -H i i -H
-H C 3 4J
4= 0 g 03
6 3 ^
m o 4-i
CM i 1 d
i-H 01 01
4= 4-1 d
4J 4-1 O
CD CD f. u
CD 4n 4-1
O CO -H 01
d £3 oo
do 03
O -H g !-i
H 4-> 03 O)
4-1 Cd S-l >
3 ^ 00 CO
43 4J O
in -H d M >-i
en s-i o p. 3
05 4-J U 0
w d oo 42
H 0 d 1
>~ T3 O -i-l ^> c^io
' m u O 4J 4J
i±r d co vH cc
ri CD d 01
C3 3 01 O 43
UJ cr to 6 oo
°?£ £ 2 o2
4-1 Ol CJ
. > 03 m
UJ o) ai > CN
C-^ ~> > 4J
. (_ i -r-i s-i en 01
. 4-i 3 (U 42
t R) 0 3 4-.
CD 5 ^"f 0^4,
:D e <- 42 o
3- 3 CN 4J
in 13 ° c
. fl ^-x 4-1 0
LJ CD Td O -H
42 d 4J
-**. 4-1 3 VJ 3 N
. O CO 43 H
4-J ^-1 Ol -H Pi
o op f^ M O
^ d o 13 co c/i
O 03 d -H
CO 43 O "O >,
H CJ 43
rvi M TO ai >,
LN to 3 M O T3
* cud d cu
E -H Ol 01 4-1
O e 42 3 03
u ^-' 4J cr t-H
^ ^
o
CO
I
LD 0
° a)
* V-i
3
OO
H
0
D-32
-------
APPENDIX E
CUMULATIVE FREQUENCY DISTRIBUTIONS OF THE 25 HIGHEST OBSERVED (MINUS
BACKGROUND) AMD CALCULATED (LUMM) SHORT-TERM S00 CONCENTRATIONS
This appendix compares the cumulative frequency distributions of
the 25 highest short-term (1-hour, 3-hour and 24-hour average) observed
(minus background) and calculated (LUMM) SO concentrations from Appendix
A of the report by Hanna, e_t^ _al_. (1982b). Table E-l gives the figure number
for each combination of monitor and concentration averaging time. As noted
in Section 3.2, the observed (minus background) 3-hour average SO concentra-
tions in this appendix are not necessarily the same as in Appendices B and
D because 3-hour periods with 2 hours of valid concentration measurements
were included by ERT in the determination of the 25 highest 3-hour average
concentrations.
E-l
-------
TABLE E-l
IDENTIFICATION OF FIGURE NUMBERS BY MONITOR AND
CONCENTRATION AVERAGING TIME
Figure No.
E-l
E-2
E-3
E-4
E-5
E-6
E-7
E-8
E-9
E-10
E-ll
E-12
E-13
E-14
E-15
E-16
E-17
E-18
E-19
E-20
E-21
E-22
E-23
E-24
E-25
E-26
E-2 7
E-28
E-29
E-30
Monitor
1
1
1
3
3
3
4
4
4
5
5
5
6
6
6
7
7
7
8
8
8
9
9
9
10
10
10
All
All
All
Averaging Time
1 Hour
3 Hours
24 Hours
1 Hour
3 Hours
24 Hours
1 Hour
3 Hours -
24 Hours
1 Hour
3 Hours
24 Hours
1 Hour
3 Hours
24 Hours
1 Hour
3 Hours
24 Hours
1 Hour
3 Hours
24 Hours
1 Hour
3 Hours
24 Hours
1 Hour
3 Hours
24 Hours
1 Hour
3 Hours
24 Hours
E-2
-------
CD
CD
CD
CD
CD
CD
CD
CO
CD
U?
CD
CD ^
UJ
O
UJ
oo az
UJ
a:
u
H
4-1
CO
rH
^J
6
3
0
o>
_r~!
4->
UH
O
d
O
CO
H
»-l
CO
(^
e
o
o
i i
i
w
0)
3
60
H
t I
V-i
o
4-J
H
C
JO
*
4-1
cfl
en
d
o
H
4J
CO
4-1
C
CU
CJ
C
O
CJ
c~
O
CO
o)
60
CO
0)
>
cfl
J-t
3
O
1
i i
^ N
T3
C
3
O
^
60
^bii
CJ
CO
o
en
3
ci
H
e
* '
T3
CU
>
M
0)
f">
0
1
CO
r 1
3
g
^3
CJ
J3
4-1
4-1
H
^
E
cfl
60
O
^_j
P-
to
C
H
o
4-1
H
C
O
E
0
CJ
CO
>
4-1
CO
CU
^
CU
(-*
4J
4-1
O
1-1
CO
CU
K*"»
T3
d
0
o
CU
CO
CU
r~j
4-1
60
C
H
J_t
3
T3
cn
C
o
H
4_1
co
S-i
4-J
C
o
c
o
H
-U
^rj
S
£3
)l
r*~i
r*
T.1
OJ
4-J
C3
r I
D
O
i 1
cd
o
o
o
(Mddl
E-3
-------
CO
z.
o
tI
h-
CE
UJ
o
o a
LU
ua a f-
CjJ LU CE
CE > _!
LU UJ U
*-i > CO _l
CC DQ CE
O C_>
a; a;
o rn
l- o
1-1 or
2: i
o cn
CD
CD
JJ
CD
CD
CD
CD
CD
ID
CD
CD
Hi
13
C3
LU
00
CO
UJ
CE
CO
tM
LD
O
O
-I
1(0
o
d >,
-a
cu
> cu
M X!
CU 4J
cn cu
X 60 3
O C
H CU CU
I i j_j ^_j [ i
cn 3 M-J co
CU T3 iI
.d cu 3
60 i > O
H i-l iH
,d }-1 4-1 CO
O CO O
CM -H 3 cn
d e d
0) O 3 O
j= a o -H
4J 4-1
4J CU Cfl
M-l CO 4= ^
O 4-14-1
cn d
d d ,d 01
O O 4J O
H -H i-l d
4-1
3
cfl
H 4-1
M d
4-1 CD
cn u
H d
*o o
a
>>
u c^
d o
cu c/j
3
cr cu
cu 60
!-4 cfl
^0
O
E
co at
s-i so
to co
o >-i
M cu
ex >
co
cu e
cu >
> cO
CO
>-(
3
O
3 42
E
I
id 5-i
H 3
S-J O
o f:
4-> I
H m
d
Q *-
cn
o)
O ,c
o 60
CO -H
CO u~i
-J
o- d o 4-1
E -H o cn
O 6 CU -H P3
c_>
CNI
I
w
3
60
ft.
cn 13 (-;
E-4
-------
i 1 1 » i 1
_
_
_
_
_
_
-
I
_
_
.
-
r
: to
: o
f-H
r cc
: a:
i h-
: UJ
: o
o
I o
- UJ Q
O UJ
o: >
QZ QZ
UJ UJ
T-t :> tO
tr m
I o
a; a;
O ' *
_ l o
IZ 1
o -<<
21 CM
1 1 1 1 1 1
t
o
t 1 1 jtitlll 1 t
\
\
\
\
\
-
\
1
t
II 2
1 ^^
ID
1 1
1
1
1
1 _
1
1
1 ""
1
i '_
ll
\M
\ -
\
1 \
\
\
X
1 .
1 _
V
I \ -
I \ "
1 \. "
\ ^ :
* \ "
\ j;
V
v ~
i «
\
\ ] Z
a i I :
§ \ ^
ID
CJ
tr -
CJ ~
i -
i
' i l i i i i i t i i
--I CS
I0 I
OJ
en
jj
OJ
CD
OJ
OJ
CO
ID
OJ
OJ
00
J-^
*
CO
*
in
tlL
m
m
CM
"~2
ID
O
-i
1°
O
Jp
o
UJ
13
a
UJ
a:
u_
UJ
;>
-1
? *
cc
T"
CJ
T3
\4
CU
CO
,0
0
4-1
CO
cu
60
H
r*.
tn
CM
CO
a
CU
4-J
U-l
o
1-4
CU
c
o
CJ
CU
CO
cu
4-1
1
3
CJ
i i
CO
CJ
CO
C
0
H
4J
cfl
^-i
4-1
C
cu
CJ
c
o
o
CU
60
cfl
J-J
cu
J>
cfl
3
o
1
-3-
CM
4J
CO
CU
,j3
fcC
H
r1.
in
CM
cu
^r"
w
O
c
c
H
4-J
3
H
4-1
co
H
quency
&
^
>,
-0
-------
01 01 I
CD S-l 3
42 ec n-i o
o c; r-n
H O) CO
O
cn 3
0) T3
>
H
bO CO
H
CO
d
o
H
O 3 CO
in 4-1 o s-i
CXI -H 4-1
d 01 c
oi o 42 oi
42 S 4J o
4J d
U-i rO 4-J O
O -H
co 3 o)
C d 60
O O S cfl
H -H CO (-1
4-1 4-1 (-1 0)
3 cfl bO >
H 4J M
J-i d O. H
4-1 CU 3
cn cj bo o
H d d
T3 O -H
U S-l <
>s O
CJ CNI4-I 4J
C O -H CO
o) w d 01
3 042
cr CD g ac
0) 60 -H
S-< cd O J3
4-1 ^J O
a) co m
01 >>
|
>
H
rt 4-1
CO 0)
01 4=
i-l O
3 £ 0) >4-i
6 I 4= O
3 iI 4J
o d
^ u-i O
O> T3 O -H
42 d 4-1
4-1 3 M 3
O cfl 43 .
M-j ^ CJ -H S
O bO >^ Vi S
d o T3 cn H_I
o co d -H
cn 43 o 13 >>
H CJ 43
VJ cn oi ^
co 3 cn o "O
cx d d 01
6 -i-i ai o) 4-1
o S 4= 3 co
u » 4-1 cr ii
3
bO
E-6
-------
13
O)
£
01
Cfl
\
Ol
1-1
40 60 <+-!
O
4-1
Cfl
^
O
d
4)
^3
cr
0)
V-i
4-1
0)
£>
H
4-1
CO
i 1
3
B
o
OJ
_n
4-J
14-1
O
d
o
Cfl
H
j-i
cO
p i
e
o
u
d
H
V-
3
*"O
CO
J^J
0
4-1
H
d
o
^
4-1
Cfl
Cfl
C2
O
H
4-1
CO
I-i
4-1
d
01
o
d
o
CJ
CJ
>
H
4J
Cfl
^H
3
e
j3
CJ
cu
t-£^
4_)
r^
J_)
H
£5
B
CO
^_i
bO
0
i-i
p ,
bO
d
H
l-l
O
1
3
0
i-H
CO
O
Cfl
d
o
H
4J
CO
^_i
4_J
d
0)
CJ
d
o
o
OJ
bO
cfl
0)
>
cO
^
3
O
42
1
CNI4J 4-1
o
c/3
Ol
60
CO
^4
CD
£>
cfl
}_l
3
O
_(~1
1
ro
^-^
13
d
3
O
^i
bO
^
O
cO
o
en
3
d
'e
N '
H
C
o
B
o
o
cfl
£>
4-1
Cfl
0)
3
0)
r!
4-1
<4-4
O
(-1
CO
CU
£x,
13
d
o
CJ
O)
en
CD
42
4-1
Cfl
o
.C
bO
H
42
in
CM
0)
42
4-1
U-J
O
C
o
H
4-1
3
40
I-I
4-1
Cfl
H
"O
p^
O
d
CD
;3
cr
*5"l
"
1 1
( ~]
>-
o
13
CD
4-1
CO
i I
O!
f-i
3
bC
(Udd) NOIlHyiN33N03
E-7
-------
(Wdd) NOIlBMlN33NOa
E-8
cu
3 M 60 CO
,JO 4-1 O
H d S-i i-i
!-J CU CX 3
4-1 CJ O
CO C 60 X
H O d I
13 u -H -^r
S-J CM
>^ CNIO
O O 4-1 4-1
d en -H CD
cu d cu
3 CU O f.
&
cu
S-i
M-l CU O
0) CO
E 60
H
O J2
un
CM
H i-l CO CU
4-1 3 CU ,C
CO O S 4-1
iI ,d
3 I CU i*-J
6 -a- j3 o
3 CM 4-1
o c
/-N U-l O
CU T3 O -H
x: d 4->
4J 3 VJ 3
O CO ,£>
4H t-i CU -H
O 60 ^ >-J
riS4 I '
d cj
O cfl
to Xl O 13
H O ,O
J-J CO CU >-.
CO 3 CO CJ 'O
fx d d cu
0 -H Ol CU 4-1
o e x: 3 to
CJ) ^' 4-1 cT iI
CO
d -H
w
CU
i-i
tc
H
-------
1 1
_
-
-
_
-
__
_
_
_
_
-
~
-
^
-
-
-
:
1 1 1 1 1 1 .J.I.I 1 1. ..1
~
-
~
-
1 1
1
0
\ 'V
1 \
I \
, \
\ \
i \
\
1 \
I \
\ \
' \
1 I
\ \
I
1
1
I
1
I
1
1
1
I
\
I
I
1
»
t
I
\
\
\
1
1
X
X
1
1
\
V
V,
1
CO ^
3 >
^T..
0
t-H
H-
tr
H-
LU
O
O
cj a
LU
LU a h-
c!) LU cr
a: > _i
at a: ^
LU LU CJ
-4- > CO _]
cr en cc
O CJ
02 ct:
O =>
h O
. . ^ "
0 -H
1 1 1 1 1 1 III
O
o
1 1 1 1 1 1
_
^
\ i :
\
\
\
\
I
I
\
\ ""
\
\
\ -
1 -
\ ~
\
\
1 '-
\
\ -
\
\ .
\ _
\ -
V ~
\ -
k \
N, \
\
\ 1 I
V. \
\ \
1 1 ^
t :
\ I j;
\ '-
\
\ -
\
\
\
\
'
1 _
1
1
1
1
1
1
» i
1
CD
CD
CD
CD
CD
CD
CD
CD
LD
CD
^_
CD ^
LU
~^
O
LU
oo a:
u_
UJ
* ^^
1
CE
ft* i
^
21
LD rD
(_)
*
"
CD
CM
*"^
LD
a
^
o
o
QJ
!-< 1
-l > O
en 3 -H
a) "a 4-1 co
J3 cd C
&0 ^d" i ' O
H 3 -H
0 3 cd
un 4^ cj )j
0-1 -H 4J
C3 0) C
CD O ,C 0)
^ S 4-1 a
t-* C
M X O
4-1 Cd 4J O
O -H
en S cu
C C 60
o o 6 cd
H -H cd V-i
3 cd 60 >
r£3 M O CO
H 4^ J-J
)-< C P4 (-1
4-J CD 3
CO O 60 O
H C C .C
T3 O -H 1
U S-J ^H
>, 0
O CMP 4J
C O -H CC
CU C/3 C CU
3 0 .C
cr QJ e to
CU 60 -H
^ cd c ^:
M-l ^i O
cu cd in
CU > > CSi
> ca 4-1
H CO 0)
4J M HI J^
cd 3 rs 4J
t-H O
3 js CU ^
e i j= o
3 ' ' 1 '
o c
x-v 14-1 O
a) T3 o -i-i
4J 3 >-J 3
O rd ^2
M-I j-i a) -H s
Comparison o
(minus backg
the second y
quency distr
lated by LUM
i
w
OJ
to
r-^
(Wdd)
E-9
-------
CO
o
UJ
o
o a
UJ
UJ a f
c±> iti cr
cr > _i
a: ct: zj
UJ UJ CJ
«* > co _i
cr CD cr
o o
ft: ct:
O ZJ
l- o
O CO
CD
CD
3J
CD
CD
CD
a_i i i i
_L
J L
CD
CD
ID
CD
CD
z.
UJ
C3
UJ
en a:
u_
UJ
CO
up
CO
CM
U3
o
o
o
-1 .
ID
NOUBH1N33N03
E-10
O JZ
c o
O CO
cn ,a
CX3
I
w
M
H
-------
to
o
a:
o;
UJ
o
o a
UJ Q t
rh || i rp
cn > _i
UJ UJ CJ
-4- :» (n _j
cr CD a:
O CJ
a: or
o rs
I- O
Csl
CD
cn
CD
O)
CD
to
CD
05 z
UJ
a
UJ
ao a:
U-
UJ
03 _J
ID 13
CJ
-t
CO
Csl
UD
O
o
o
csl .
ICD
t I I
01 O> 3
CO 60 >-i CJ
d
O -H
to
}-l CU O
CO T3 -H CO
01 4-1 d
42 <) co c
60 iI -H
H l-l 3 4-1
42 O E CB
4-1 3 i-l
U-1 1-1 CJ *J
CN d d
O CU 0)
X 4-i d
4-14-1 O
cd 42 o
O CD
4J
H
0)
60
d o co
o -H e s-i
H 4-> CO O)
3 !-i 60 co
43 4J O
H d ^ M
S-i 0) d, 3
4-1 0 O
CO d 60 42
H O d I
T3 O iH -3"
>, CNO
O O 4-1 4-1
d w -H co
d CD
a)
cr ec 6 60
O) CO -H
4-1 Ol CJ
>
m
QJ Cfl > 04
H M cn cu
4-1 3 0) 42
CO O 13 4-1
i^
3 CN
a
CU M-l
-C 0
d
o
(U T3 C -i-l
d d 4J
4-1 3 M 3
O CO t ^
O 60
cu
CO
d CJ T3
o co d
CO 43 O
H CJ
M cn cu r^,
CO
3
CL, C
O)
CO CJ
. . .. d _
E -H a) a> 4-1
O E 42 3
c^
I
w
a)
)-l
oc
H
PIH
NOUUM1N33N03
E-ll
-------
cu
w
0
4^
CD
01
_("1
60
C
H
£_l
3
T3
1
cu
<4-l
CO
>
H
4-1
CO
60 m IH
H
4=
m
CM
CU
f!
4-1
T-l
4-1
03
t 1
3
E
3
cj
Q)
f
4J
<4-(
O
c
o
en
H
S-i
03
D.
e
0
C_3
}_J
O
4-1
H
C
o
^
4-1
CO
crj
C!
O
1-1
4-1
CO
S-i
4_l
C
d,1
O
C
o
0
o
c
c/:
0)
60
03
J-i
CU
>
cfl
}_l
3
O
f~1
1
, 1
,-^,
T3
£
3
O
I-i
60
cj
cO
n
en
3
C
H
E
v-^
3
E
^
o
cu
r^
4J
ff
4-1
H
^
E
co
i-i
60
O
S-l
a.
60
a
i-i
S-i
o
~14->
H
C
o
E
o
a
CO
>
4-1
cn
cu
13
01
^f~!
4J
U-l
O
S-i
CO
CU
>^
T3
cj
O
0
a>
CO
CU
^
u
3
O
1 i
CO
O
en
C
0
H
4J
cC
J_l
4-1
d
cu
0
c
o
CJ
0)
6£
CO
I-i
0)
£>
CO
}_l
3
O
n
1
i i
4J
CO
cu
-^«
60
H
J=
L/~l
CN|
CU
c~;
4J
14-1
O
d
o
H
4J
3
o
H
I-i
4«J
en
H
T3
>.
o
d
cu
3
cr
y
§
hJ
P^
_O
'O
OJ
4-i
CT3
i 1
I
w
cu
S-i
3
60
H
(Wdd)
E-12
-------
Ol
>
(-1
Ol
en
n
O
4->
cn
0)
(-;
60
H
f~!
LO
CM
Ol
_f-*
4J
4-,
O
C!
O
H
4J
3
H
J_4
4-1
CO
H
T)
>>
O
C
0)
3
cr
01
}-i
y-i
01
H
4-J
03
i 1
3
e
3
O
0)
_("]
4-1
IH
O
c:
o
CD
H
^
CO
O-
e
o
60
C
H
h
3
T3
U"l
h
O
4J
H
c
0
S
±j
n)
u;
c;
o
H
4-1
CtJ
J-l
4J
C
Ol
U
c
0
c~
O
CO
0)
60
CO
h
01
>
to
^_j
3
O
f*
1
01
x^V
T3
c:
3
o
h
60
^
U
CO
o
w
3
c;
H
e
i
Ol
1-1
U-l
01
>
H
4-1
tO
1 1
3
e
3
o
0)
^
4->
_C"t
4-1
H
^
e
CU
^4
OJD
0
M
O.
60
C
H
0
~H-I
H
C!
O
e
o
o
cfl
>
w
a)
£3
01
^r*
4J
m
0
^
ca
Ol
p>^
T3
C
O
CJ
CU
U3
0)
42
1
^3
O
r (
CO
CJ
w
c
o
H
4-1
CO
}-i
4-1
C
01
o
c
o
CJ
11
60
CO
>-.
0)
^
CO
}_i
3
O
rr^
1
4-1
cn
0)
rf~!
60
r-l
,r^
LTl
CM
0>
,r^
4J
M-H
o
cs
O
H
4-1
j3)
l n
H S
V-t S
4-1 ^3
cn i-J
H
T3 >,
n
K^l
a -a
c 01
Ol 4-1
3 cO
w
6C
H
E-13
-------
till 1
_
-
..
-
-
"
-
_
.
-
I CO
: o
I-H
~ i ,
r cr
- ct:
1. "
- ^
- UJ
^
: °
3
~ cb
cc
a:
UJ
10 >
r cc
I a; a:
o ra
t- o
-« r:
~ -«^ 1
o >
a;
UJ
to
03
O
i
i i \ i i 1 1 i i 1 1 \
\
\
\
\
\
\
\
\
\
\
\
! \
1 \
1 \ «=
i \ 2 .
1 \ «5 -
I \ ^
1 \ 3
1 \
1 \
1 \
| \
1 \
1 \
\ \ ~
^ _
S *.
^s
\
1
-
.
T
Vv
\\
\ ^
i 1
\ | ~
i
\xs :
i N -
\ ^i ~
i x "
1 ^^ _
\ ^
\ * "
\ X -
a \ \ !
UJ \ -
i i ! =
_t 1
z \ --
\ -
\
\ :
i i 1 1 i i i i i i
^ cs
ID i
a> ^
" >
J5 M
CD a>
en
J2
o
4-1
«» w
J> 01
CD .c
CD tic
. -H
vf~^
iri
CN
U
rC
4J
s *
c
o
H
4-1
3
^D
ID £
3) *->
en
H
>- *o
- %
33 ^j ^,
' § S
r3 ^
. l-< J_,
r~ co
y I i I
CD i 3
. =i e
i 3
to :3
(_) a)
^. 4-1
* 4-i
O
to c
o
en
H
C-J co
I*
O
t
* CN
(-H
1
ID
O a)
3
H
^.
J°
O
60
C
i-i
!-j
3
t3
LTl
>-l
O
4J
H
C
0
S
4-1
CO
d
o
4-1
CO
$-1
4-1
c:
cu
o
c
o
o
o
C
0)
60
CO
J-i
01
>
CO
^_l
3
O
,£!
|
CN!
/"^.
X3
C
O
^-i
60
O
CO
en
3
C
H
E
1
0)
^
M-l
at
£>
H
4-1
CO
r- 1
3
6
3
CJ
J3
4-1
H
^
S
CO
la
60
O
M
ft
60
C
H
!-i
iO
4-1
H
c:
o
e
o
CJ
CO
>
4-1
en
cu
£S
O)
4J
M-J
O
S-i
CO
0)
>>
n
c:
o
a
01
en
0)
4-1
1
u
i
CO
CJ
en
C
0
H
4-1
cO
S-i
4-1
C
0)
CJ
c;
o
o
01
60
CO
0)
£>
CO
>-i
0
f
CNJ
4-1
CO
0)
60
H
m
CNl
0)
_c
4-1
M-l
O
c
o
H
4-1
3
o
H
S-l
en
H
TD
j>^
quenc
s
s
hJ
T3
0)
4-1
CO
i 1
E-14
-------
CD
CO
CD
CO
O
CO
CO
CO
CO
U)
CO
UJ
^3
C3
UJ
00 OS
U_
LU
to _,
ZD
2T
ID ra
-*
CM
O
CD
O
(Udell NOIlbyiN33N03
E-15
t>
cu
>
»-i
cu
cn
42
60 »*-! O
o d I-H
H 0) CO
J-l > O
3 -H
0) T3
42
cn
cn
cfl d
6C v£> i1 O
H
42 H
i w
U 42
14-1 CO 4-1
O -H
C S *
ooe
O 3 CO
LO 4-1 O J-J
CN -H 4-1
c o) d
0) O 42 OJ
42 S 4-> 0
C
o
U
0)
(SO
(0
H -H cfl t-i
4-1 4-1 (-1 QJ
3 CO 60 >
42 VJ O Cfl
H 4J S-J
1-4 C O. t-l
4J , O
O CNU-) 4J
C O -H 03
oj co d 0)
O 42
cr
(!)
t-i
<4-l
aj
>
H
4-1
cfl
iH
e
3
0
0)
42
4-1
LM
O
C
o
03
H
*-l
CO
p.
e
o
U
a)
60
CO
J-i
cfl
M
3
O
1
, 1
^ x
T)
d
3
O
M
oo
Ai
CJ
cfl
42
CO
3
d
H
^
e
o
o
cfl
>
4-1
cn
0)
S
42
4-1
M-l
0
>-l
CO
0)
^
T3
d
o
o
0)
in
0)
42
j_)
60
H
42
LTl
CN
0)
42
4-1
M t
O
d
o
H
4-i
3
43
H
^
4-1
cn
H
T3
>.
CJ
c:
0)
3
cr
y
§
p^
hJ
>^
43
T3
OJ
4J
CO
1 1
I
w
01
t-1
-------
NOUBMiNHDNOG
E-16
-------
111! t
~
"
{ ._,
_
-
"
~"
""
-
_
_
: to
: o
K-l
~ i
F cc
- Qi
1 *"~
_ "^x
: u-J
^
o
: o
UJ
CD
CC
- Q£
UJ
CD i>
~ cc
I or a;
0 Z3
i-o
2 i
o -
CU
cn
o
J_)
cn
01
&
bO
H
r^
m
CM
01
&
4-1
U-l
0
c
0
H
4-1
3
o
H
J_j
4J
CO
H
T3
0
C!
QJ
3
cr
cu
14-1
cu
H
4-1
CO
I 1
3
o
cu
4-1
M-l
O
c:
o
CO
H
S-i
to
o
o
LO
1 I
1
w
cu
P
bO
rl
60
C
H
}-i
3
"tj
vD
j_,
O
4-1
r-1
C
O
^
4J
cn
c;
o
H
4-1
fC
^-1
4-1
C
cu
a
c
o
u
c
O
CO
cu
bo
!-i
0)
J>
CO
J-l
3
O
1
CM
/-*,
C
3
C
S-i
M
o
03
minus
> '
cu
S-i
MH
0)
[>
H
4_1
cfl
i-H
3
e
3
o
CU
(~!
4-1
4-1
H
£5
e
S-J
M
O
S-i
p.
bO
C
H
JO
4-1
H
d
o
e
o
CJ
03
4-1
cn
CU
£2
CU
4J
U-l
0
V-4
ca
0)
>
-a
c:
o
o
CU
cn
CJ
4-1
3
CJ
i I
03
CJ
CO
d
O
T-l
4-1
CO
}_<
4-1
d
(1)
u
d
0
o
cu
bO
£_l
a)
cO
^4
3
0
1
^3-
CM
cn
cu
bo
H
in
CM
CU
f;
4J
M-l
o
C
c
H
4-1
3
H
S-i
cn
H
T3
CJ
d
cu
3
a*
5^
g
J
j^
n
13
CU
4-1
03
NOIlByiN33NOO
E-17
-------
CD
CD
05
CD
CD
CD
CD
CD
cn
j i L
ID
CD
CD
o
UJ
ra
o
UJ
ao
UJ
CD
ID Z3
O
CM
ID
O
O
-« .
'o
(Udd) NOUbMiNHONOO
E-18
I
w
01
^4
3
(JC
-------
o =3
I- o
o en
CO
o
II
I
cr
UJ
CJ
o
CJ O
UJ
UJ a f
c±> uJ cc
ec > _j
a: a: rD
UJ UJ CJ
> co _i
a: m cr
O CJ
CB
en
en
en
en
CD
en
en
en
en
CO
o
UJ
13
C3
UJ
U.
UJ
cc
* Z3
z:
UJ ZD
CJ
uo
o
Q
-« .
o
I
ai
O)
S-i
01
CO
.O 60 M-i O
O C
CO
cj
en
en
ID TD
.C
60 r~-
d
o
S-i E 4-1
O 3 cc
in 4J o s-i
CNl -H 4-1
d tu
CO
.. d
o o e
60
. CO
CO S-4
Ol
CO
H -H
4-1 4-1
3 cfl
,0 s-< o
H 4J (-1
S-i d o S-J
4-1 Ol 3
en o 60 o
H d d ,c
^3 O -H I
o s-i m
>-. o
CJ CNJ4J 4J
d O -H co
CD CO d Ol
3 O ,d
cr 01 6 60
Ol 60 -H
s-i cfl o ^:
M-i S-l O
oi co m
> CN
4-1
CO 0)
01 f:
cfl 3 ^3 4-1
r-l O
3 f: ai 14-1
g i ,d o
30-14-1
cj d
^ 14-1 O
OJ T3 O -H
^n d 4-1
4-131-13
O CO ,£)
y-i s-i ai -H ;
O 60 >-, S-i
0)
CO
4-1 S-l
O cfl
co ,n o
H a
>-l M 0)
c3 3
a d
e
o
U CO i-3
d -H
en
B -C 3
-^ 4-1 C"
r~-
ti
1^
01
3r
H
NOUHH1N33NOO
E-19
-------
tt
I IT
1 I
\ _J
; I
\
\
I
\ 1
\ \
. \ "
- \ \ -
- \ \^ .
- V \ I
\ \
\ I
\ 1
.
I u J
- V -
- \ -
r
IS
- \ ^V -
- Z \ ^ -
: o \ lx :
" K \ x :
To: \ x\ ~
- QZ \ \ -
-^ \ 1
1 UJ \ \ '
- -z \ -
o *
: o a \ :
UJ 1 *
- UJ a i- \ \
CD LU CE \ 1
a: >- _J »
QZ a: ^3 \
UJ UJ 0 \
t^- >- to _j \
CE CO CE i H
: o o \ d
a; a; '
0 => I -
t z ! ' 1
** ^? 11
0 "t 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
J* d) CU 3
O* en M !-< u
42 C M-l r-l
O -H CO
V4
O) o en -H cu
C £ bO
CO CD
0 -H E \->
H 4J CO CU
*J CO !-i >
3 l-i M cfl
42 4J O
H C *-i 5-1
irj vj - M Csl
-.,(_) >-. csio
°f ZZ 0 O 4J 4J
y ^ cm
;J 3 O O 43
C3 cr 60 S oc
UJ CU CtJ -H
m Q; 5-i !-i O 43
~ rr LW o o
"- > co m
cu d > csi
UJ > 4-1
r^ ;> -H 5-i en cu
. *-* 4J 3 01 43
i_ co O J3 4J
rt- "-1 42
,,, O- 3 1 cu M-I
tD _I E ^ 43 0
' ID 3 c-j 4J
Z- cj C
n =; ^-v U-) 0
"f r» <1) T3 0 -H
O ^3 C 4-1
4-i 3 M 3
-rf. O a) 42
. U-i 5-4 CU -H S
0 M >, 5-1 §
_, C 0 T3 03 rJ
01 o n c -H
* en 42 o T3 >>
H O 43
5-J CO CU >s
P. i cfl 3 cn o -a
^ a, C C cu
' 6 -H 0) CU 4J
0 B 43 3 «
C_3 s_x j_| 571 r-l
1 1
CO
-H
1
8
3
M
*r^
3 -r-i CM1"?
o 'o o
E-20
-------
O5
01
05
01
O)
O)
O)
CD
CD
L£>
CO
UJ
^3
C3
111
oo a:
UJ
c-- >
LO
o
'o
0)
>
S-i
QJ
CO
,0
o
4-1
CO
QJ
f!
60
d
H
S-i
3
'O
1
d)
S-i
<4-i
QJ
>
H
4J
CO
6000 i-H
H
r1.
in
CNl
(1)
f*
4-1
M-l
0
d
0
H
4-1
3
H
J_j
4J
CO
H
T3
>>
CJ
£
0)
3
cr
0)
}_i
U-i
l-i
O
4-1
H
d
0
^r]
4_l
CO
CO
d
c
H
4-1
CO
S-i
4J
d
a)
u
d
o
o
o
o
C/3
Q)
60
CO
)-l
01
3
g
^3
0
QJ
^
4-1
_f~;
4-1
H
J5
E
CO
S-i
60
o
S-i
P ,
60
d
H
S-i
o
M-l
H
d
o
B
o
o
CO
1
3
o
, 1
cO
0
CO
d
o
H
4-1
cO
^4
4J
d
0)
CJ
^
o
u
01
60
cfl
}J
0)
>
£-*
3
o
r~!
1
. 1
[ 1
CO
0)
60
H
f!
U~l
> CO 4-1
CO O1
QJ X
4-1 S-I
cfl 3 13 4J
t-H O
3 .d 0) U-i
e i ^ c
3 ^ 4J
o d
<~^ >4-i O
QJ 13 O T-I
J3 d 4-1
4-1 3 S-i 3
O CO JD
M-l S-i CD -H g
^ 4J S
d O T3 CO nJ
O CO d -r-l
CO ^3 O T3 >i
H U ,£>
S-l CO Q) >^
co 3 to o -a
ex d d O)
O 6 ,c! 3 CO
u ^^ 4-1 cr ^-1
I
w
3
60
H
NOUBH1K33NOO
E-21
-------
1 1 1 i II 1 11
\ I
\ I
\
\ \
\ \
\
\\
\
I
1
t
1
I \
\ '
\
\
\
\
\
\
\
\
\
1 \
\
-
~
_
-
-
: ^
: to
: o
i i
: cr
r ^
: uj
- z
; o
I UJ
UJ a i
cb UJ cc
cr i» _j
a: a: rs
UJ U-l U
co :> en _j
cr DQ cr
I o o
or a:
o ra
h- 0
1-1 X
z i
O CO
z:
i i i i i i i
3
O
1 j t 1 1 I 1 1 i t
-
5
^
3
i
t
\
i
i
i
\
\
\
\
i
i i
i
\
\
\
v
\
V
\
V ~
"1 Vxv ^
:
\ 1 ^
1 \ ^
\ '. =
V \ :
1 ' :
\ 1
I
\
1
I '-
1
1 1 1 1 1 I 1 I I 1
-I CN
i i
CD
O)
O)
CD
*
0)
CD
CD
CO
CO
10
CO
*
V-
en"
s
§
UJ
03 QZ
U-
UJ
l> >
I"H
1
T
' =>
ID 3
CO
CM
.
LD
O
»- 1
j .
O
NOIiBMiN30NOO
E-22
60
O 60 >-, 1-J g
d CJ TD CO >-J
o cd d -H
CO 43 O TD >
H CJ 43
t-l CO 01 >,
CO 3 CO CJ *X3
ex c d 01
O)
B
o
u
o
CM
I
w
60
-d 3
-------
i 1
_
_
__
_
_
-
_
-
.
_
-
-
_
-
^_
1
~
-
~_
-
-
__
_
:
_
"
, ,
J i 1 I I 1 \ | 1 1 1 1 I 1 i I
\ \
\ \
\ \
\ \
\ \
\ '
\ \
\ \
\ \
i v
\ i
\ v S
\ i 5*
\ i *^!r
\ » ZZD
\ \ -1
3 \
\ i
.
*
\ \
\ *
\ \
\ \
\ \ ~
\ v
1 \
\ \ *
\ \ -
\ \
1
1
\
1 \ 2
\ S,
\ %X -
I \ ~
I
\ ""
\ \ ~
co i \ :
"Z. \ \
P \ \ ^
cr \ \ -
K \ i
i \ %
* \ \ _
uj V *' "
u \ i :
o i \
u a 1^1
UJ \ I
UJ a t- \ i _
o uj cr ^
fi~ ^^ i \
o; o; ZD \ I
UJ UJ CJ \
CD >- CO _l n
cr CD cc ! -
O CJ I
tr or ^
0 13 I
>- 0 1
t-i 2: i '
211 1 -
o -_
14-1
o
>
H
4-1
CO
3
CJ
0)
4-1
O
d
o
CO
H
)-l
CO
Cu
O
u
.
1 1
CN
1
W
CU
t-t
3
60
fe
M
C
7-1
l-t
3
T3
CC
^J
o
4-1
H
C
O
T"|
4-1
CO
CO
d
o
H
4-1
cC
i-i
4-1
C
at
u
d
0
CJ
i
V-4
M-l
0)
>
H
4-1
CO
r-H
3
E
3
CJ
CU
4^
4-1
4T
4-1
H
^
B
CO
Vv
60
O
J-j
1
CJ
i 1
CO
U
cn
d
o
H
4-1
co
I-i
4-1
d
u
d
0
CJ
0)
6C
CO
0)
>
CO
j_j
D, 3
0
6043
d
H
1
v^f
csi
CNO
O 4-1 4-1
CO
0)
vera
CO
}_i
3
O
1
CN
T3
d
3
C
60
U
CO
43
inus
^
H
C
O
B
0
CJ
CO
>
4J
CO
CU
01
4=
4-1
O
J-i
ca
CU
'd
o
CJ
01
CU
42
4-1
CO
01
60
H
CN
0)
42
4-1
M-l
O
d
o
H
4-1
3
rH
t-l
4-1
CO
H
TD
CJ
C
0)
3
C71
«
§
l^-l
-1
^*l
43
OJ
4-1
C3
i 1
o
'o
f-l
(Udd) NOUBM1N30NOO
E-23
'o
-------
en
CD
CD
CO
en
CD
CD
CD
CD
10
CD
CD^
UJ
ID
C3
UJ
oo or
u_
UJ
=3
ID 13
CJ
(M
LO
O
-o
0)
a)
CO
r> 60
O C
H
4J M
co 3
Ol -a
60 ON
CNl
d
0) O
I
01 I
^ 3
4-1 CJ
0) Cfl
> o
H
4J M
CO d
i-H O
3 -H
6 4J
3 cfl
O M
M-f tfl
O
w
c c
o o
H -H
U 4-1
3 cfl
J3 )~i
H 4J
i-ici
4J CU
cc a
H c
U O
O
>^
CJ C
d O
O) C/5
3
cr CD
OJ M
Vi (0
M-( V<
d)
0) >
c
a)
u
d
o
CJ
& 01
60
E CO
CO H
S-i 01
60 >
O CO
(^t ^J
60 O
d ,G
H I
rl 03
d a)
o X
6 60
H
O J2
O
cfl in
> cxi
4-1
en a)
D j;
4-1 S-<
CO 3
H O
3 ,£:
B I
o
0) 13
,c d
J^J ^J
o
O 60
d O T3 CO ,-J
O CO d -H
en ^) o T3 >%
i-4 C) O
t-l ^
CO 3 CO CJ TD
O
4-1
C.
U-l O
O -H
cu d
e -H
o e
CN
OJ
I
w
OJ
(-.
Eb
c
01 0)
-d 3
4-1 cr i
O
"^
"o
E-24
-------
erve
en
42
o
4-1
en
01
r*.
60
H
42
in
CN
0)
42
4-1
4-1
0
C2
0
H
4-1
3
H
J_l
4J
cn
H
T3
K*~>
o
fi
01
2
cr
CD
S-i
4-4
CD
p>
H
4-1
CO
i-H
3
6
3
a
0)
f!
4-1
4-1
O
C
O
to
H
}-i
nj
P.
6
O
c_>
too
C
-H
S-,
3
T3
Ol
J-l
O
4-1
H
C2
0
^
4-1
CO
en
ci
o
H
4-1
Cfl
^-1
4J
C
01
CJ
c
o
0
r
O
C/)
0)
60
CO
}_i
01
^
cfl
^i
p
o
_ r4.
\
cn
^-^
TD
C
3
O
M
M
a
CO
fl
cn
3
c
H
B
^-^
1
a;
^_i
4-t
0)
>
4-J
CO
i i
3
B
3
U
0)
42
4-1
42
4-1
rH
^
B
cfl
^
60
0
^_t
a.
60
c
H
}-l
O
4U
H
(2
O
g
O
o
cfl
£>
4-1
cn
0)
[2;
0)
r^
4J
4-1
0
^4
CO
01
>>
13
C
O
o
0)
cn
a>
42
4J
1
3
O
i i
CO
0
en
d
o
H
4-1
CO
J_l
4J
C
0)
o
c
o
o
0)
60
^
Ol
£>
cfl
j_i
^
O
1
cn
4_)
en
0)
42
60
H
42
u-i
CN
cy
f~^
4J
4-J
O
C2
C
H
4-1
3
42
H
>-!
41
cn
H
T3
^
O
C
0)
3
er
5^
§
3
p^
42
T3
CU
4_)
cfl
t 1
ro
CN
I
W
E-25
-------
en
a:
O
o
zr
(O
z
o
cc
Of
UJ
o
O CD
1U
UJ a »-
CD uj a:
en :> _J
o: a: 3
UJ UJ O
:> co _i
a: OQ a:
o o
ZD
o
3Z
I
-t
csi
CD
CD
O)
CD
CO
CD
CO
I I 1 L
CO
CO
10
CO
COS
UJ
=3
C3
UJ
GO C£
UJ
(O
LO
Of)
csl
cc
10
o
lo
-«
(Mdd) NOUBH1N33N03
E-26
'o
o
CU
>
to
CU
en
43
0
4-1
en
CU
42
00
H
42
LTl
CM
01
42
4_t
U-l
0
c
o
H
4-1
3
^
H
!-i
4_t
en
H
T3
>s
u
c
CU
£3
0"
0)
1-1
<4-l
H
4_)
CO
I
r~i
3
£3
3
O
CU
42
^
M-l
O
d
o
en
H
to
tO
cx
£3
o
00
d
H
^
3
'O
CT>
^_l
O
4-1
H
C
0
^
4.J
CO
en
d
O
H
4-1
?3
!~i
4J
d
CU
CJ
i
CU
^_J
<4-l
0)
>
J_J
CO
1 i
;3
e
j3
U
OJ
41!
4-J
^f*]
jj
*H
^
&
cd
^
00
o
to
CX
1
3
O
I 1
CO
U
en
£j
O
H
4J
}_4
4J
d
CU
o
d
o
u
CU
oo
cfl
CU
£>
CO
to
3
O
d 0042
O
O
C"
O
CO
0)
00
tO
i-l
CU
cfl
(-4
3
O
1
^j-
CN|
/ s
T3
d
o
^
op
o
tO
43
en
3
d
H
e
d
H
S-l
-10
4-1
H
d
O
e
o
o
>
en
CU
3
CU
r*.
4-1
M-t
O
rt
CU
>^
o
d
o
o
CU
w
CU
42
1
^J-
CM
4J
en
CU
42
H
t~1
CM
CU
f~!
4-1
IH
O
d
o
H
4->
£
H
(-1
CO
H
13
^
a
d
CU
.
2*
§
KJ
^
43
T3
CU
4_)
to
CM
I
w
CU
J-
3
00
H
-------
i 1 1 1 i
_
.
__
.
.
-
"
i
i
: to
z.
- o
1 1
1 h-
: a:
i
i
: LU
I CJ
: o
: g
1 ^ CC
I Hi tti
o rD
h- 0
Z 1
o ^
z:
III! |
)
O
i 1 1 1 1 1 ! 1 1 1 |
_
^w
<2z
5
7-^
J
j
i
i
\ ! :
\ '
\ i
\ i
\ *
i i
i i
\ \
\ \
1 '
I * -
A ^^ "
\ * "
I '
\ '
\ 1 *"
\ \
\ \
1 1 =
1 ' -
\ 1 -
LU \ j I
a i- \ '
LU CE \
>- _l *
Q; ZD
LU 0
CO _I
aa en
O CJ I
*i CS
ID '
en
en
en
O)
en
en
en
en
i
LO
en
00
r-
*
LO
.
^^
tn
*
CM
LO
O
* i
1°
O
^»
o
LU
^^
CS
LU
or
u_
LU
h-
^f"
z:
^3
CJ
T3
0)
l-l
0)
CO
r^
O
4-1
CO
cu
.fi
6C
H
f-1
m
Csl
0)
4-J
U-l
O
C
o
H
4-1
3
r'.
H
4->
CO
H
X3
CJ
C
cu
3
cr
cu
<4-l
CU
H
4-1
CO
i 1
3
3
CU
r*
4-J
4-1
O
C
O
CO
H
VJ
CO
cx
o
un
CN!
OJ
3
00
H
60
C
H
>-!
3
T3
O
^_j
O
4-1
H
C
£
4J
CO
CO
c
0
H
4-1
CO
t-t
4-)
C
Oi
CJ
C
o
a
c
O
C/J
cu
60
CO
0)
>
CO
S-i
3
O
1
i i
/* s
TJ
C
3
O
60
CJ
CO
43
CO
3
C
Is
1
a)
}-i
u-i
CU
>
H
4-1
CO
i 1
3
e
3
O
CU
_(~!
4-1
r<~?
4-J
H
^
e
CO
^
60
0
(J
ft
60
C
H
M
O
sU-l
H
C
o
e
o
CJ
CO
>
4J
CO
CU
CU
^
4-J
M-l
O
S-i
CO
CU
T3
e
o
0
CU
co
4-1
1
3
O
I 1
CO
O
CO
c
0
H
4-1
CO
S-i
4-1
C
cu
c
o
o
CU
60
CO
S-i
CO
1^
3
0
1
« i
4-1
CO
CU
,c
60
H
J2
in
CNJ
a>
£,
4J
M-l
C
O
H
4-1
^3
lf~i
H
S-l
4-1
CO
X!
CJ
C
3
cr
5
.£*
KJ
J3
T3
CU
s
NOUBHIN33NOO
E-27
-------
fill I 1
-
-
-
_
-
~
-
~
-
~
-
-
r
_
~
"
~ CO
"Z.
: o
t 1
(__
r cc
: a:
i »-
- UJ
: o
~*2L
0
I o
- UJ a
5 iu
cc :>
r- oi a:
L uJ UJ
i O :> CO
^ cc m
I o
QZ t£
o ID
1- 0
»-i n:
z i
O CO
^
-
iiii i i
3
O
CO
CD
,,,,,,,,
1 »
\ I
\ t
\ i
I * "
\ \ '
\\
'.
\
»
1
I ^
! 2
1 3
\ \
1 !
\ i
\ 1
\ '
\ l
I I
i *
l 1
l
I '
\ ^
\ \ 2
\ \
\ *
\ v
T V
\ I
V \ 1
\ N
I ^ -
\ » '-
\ ' -
\ I
1 \
I * ""
i
1
11 ~*
1 i
\ \
i
a \
UJ
H- {
cc \
1 1
CJ
1 \
cc \
f i V
I
i
i i i i i i i i i i i
«^w
CD
CO
CO
CD
CD
CD
LO
CD
CO
00
b^
CO
ID
^
m
CSl
TH
LO
0
f-i
1-1 CM .
lo O
^
CJ
^-^-
UJ
C3
UJ
eg
u_
UJ
^^
1 1
1
cc
J
ZD
ZI
~^
u
n
01
S-i
0)
w
o
o
4-J
cn
0)
oo
H
f~;
u~l
CSl
0)
XI
4-1
4-1
O
d
0
H
4-J
3
n
H
(-1
4-J
cn
H
O
^*>
a
d
Ol
3
cr
01
j-i
01
^>
H
4-J
CO
tH
3
E
j3
O
0)
t~?
4-1
o
d
o
cn
H
CO
ex
B
o
.
vO
CSl
1
W
OJ
j_i
3
00
H
Pu
00
d
H
^-1
3
C
i i
j_j
O
4-1
H
d
Q
^
4-1
in
d
o
H
4-1
CO
J-4
4-1
d
CU
CJ
d
o
u
1
QJ
S-
4-1
0)
£>
H
4-1
CO
iH
3
6
3
CJ
0)
4=
4-1
.d
4-J
H
J5
e
CO
^_i
00
o
1-1
ex
oo
d
H
o
CSUJ
O -H
CO
o
00
a
CJ
CO
^_i
3
O
^r^
i
ro
T3
d
3
O
(-4
M
0
CO
,Q
cn
3
d
H
3
d
o
o
a
cfl
4J
cn
01
&
CU
43
4-1
4-1
O
J-j
CO
0)
T3
d
O
O
O)
cn
0)
4-1
,
3
o
i i
CO
CJ
cn
d
o
H
4-1
CO
^
4-J
d
0)
o
d
o
CJ
OJ
oo
cO
}_l
cfl
l-l
3
0
1
ro
4J
W
1)
00
H
43
CSJ
0)
rt
4-1
4-J
C
e
0
H
4-1
3
o
H
4J
CO
H
T3
CJ
d
01
3
cr
|
, i
>~,
01
4-1
CO
t 1
(Wdd) NOUBM1N30NOO
E-28
-------
o :>
CO
cc
Qi
UJ
o
z
o
o a
Hi
UJ a H-
C3 uj a:
cc :>- _i
Q; a; ZD
UJ U-l O
to
CC CD CE
O O
o m
h- o
*-i or
z I
o -<*
2T CM
CD
CO
CO
CO
CO
CO
CO
V -
\ -
I -
I - O
I .
CO
CO
CO
»
CO
03
CO
m
io
en
CSJ
o
:z
UJ
Z3
o
UJ
a:
u_
UJ
o
13
CU
>
i-i
0>
CO
n
0
CO
01
f?
oo
H
,J=
U~l
CM
H
^
4-1
CO
H
c
oo
c
H
M
3
T3
0
T H
}_4
O
4-1
H
d
0
*^
4-1
CO
CO
d
o
,_i
4-1
CO
p
4-1
C
cu
u
d
o
o
|
CU
)-l
14-t
a»
H
4_J
CO
r *
3
s
3
CJ
0)
,-C
4_J
^c
4-J
H
^
£
03
^g
OC
o
^-
c.
oo
d
H
S-i
|
^3
O
r
CO
U
CO
d
o
rt
4-1
CO
t-l
4-1
d
a
a
d
o
CJ
(U
00
CO
CJ
^>
CO
^
3
O
r"
1
*xT
CM
>. CMC
O
d
cu
3
CD
S-i
m
o
H
4J
CO
i l
3
e
3
U
01
f"
4-1
U-(
O
d
o
CO
H
(-1
rt
c.
e
o
u
o
C/j
o
oc
CO
S-i
OJ
>
CO
J-4
3
O
jn
i
^j-
CNI
x \
T3
C
3
O
Iw
00
^
CJ
CO
rn
CO
3
d
iH
E
^-^
4J
H
d
o
e
o
o
CO
f>
CO
a
^:
0)
^c
4-1
U-4
0
^4
cO
OJ
^,
T3
d
o
CJ
CU
w
01
n
4J
4-1
CO
CU
fi^
00
H
c~*
L-l
CM
CJ
,d
4-1
Lt-i
C
d
o
n
4-1
3
o
r-i
S-l
4-1
CO
H
T3
^
a
d
cu
3
cr
vj
^"|
1 i
_!
p>^
n
TD
0)
4-1
CO
i i
csj
I
W
CJ
t-
CM
(Udd)
E-29
-------
ill! I
~
-
*
-
""
_.
.
_
-
~"
.
"
*
-
-
»
Z
: co
: o
- HH
l l-
cn
: o;
"" ^r
UJ
: cj
D
: o
1 UJ C
cb u
cr ^
co a; o
or UJ u
o ^~ v
- i cc a
~ -z. tx
O 3
2:0
_! 1
-I *""*
a:
i t i i i
^
0
-, , i
»
1
»
\
\
^
\
1
1
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
1
1
1
1
t
1
\
V
\
1
1
1
1
1
1
1
\
\
\
vv
\
1
1
t
1
1
1
\
1
1
a \
UJ
3 t !
J a: '
-J \
' Z3 ^
J 0 »
3 -J
3 cr
3 O
III 1 1
o
o
till 1 1
~
*J
^~
^^
Z)
_J
_
_
""
-
q
_
^
i
CD
CD
CD
CD
CD
CD
CD
CD
CD
ID
CD
^»
O) ^
, *^r
UJ
"^
C3
UJ
co ac
u_
UJ
r*^ ^^
i i
h-
CO t
-~
z:
ID 3
CJ
^f-
.
CD
*
CNI
.
ID
O
*
r-l
<
O
T3
0)
t
01
03
42
O
4-1
03
01
,C
to
H
42
LO
CN1
0)
42
4-}
U-i
0
d
o
H
4-1
3
,0
H
M
4-1
to
H
T3
>,
0
d
Ol
3
cr
41
J-l
t-l
0)
>
H
4J
co
r-(
3
e
3
O
a)
,c
4->
"4-1
O
c
o
03
H
M
co
a.
e
o
CJ
00
£
H
M
3
T3
03
M
O
4-1
-H
(S
O
g
rH
t-H
S3
4J
CO
03
cs
o
H
4-1
cfl
I-i
4J
c:
0)
u
C!
0
O
«"
o
w
QJ
60
cd
M
O
>
cd
jj
3
o
42
i
i i
^ N
a
a
3
0
u
00
Ai
u
cd
43
cn
3
d
H
6
1
CJ
(-1
14-4
0)
>
H
4->
td
iH
3
e
£j
u
0)
^c
4-1
J3
4-1
H
s
e
cd
t-i
M)
0
S-4
a.
M
d
H
^
0
4W
H
C
0
e
0
u
c3
>
4J
03
CD
S
0)
4S
4_)
U-l
0
J-l
cd
01
>.
T3
d
o
o
01
03
0)
.C
4-1
1
3
U
i-H
cO
CJ
03
C
o
rl
4-1
cd
t-i
4J
d
01
o
a
o
u
01
M
cd
>-<
0)
>
cd
(-<
3
0
JS
1
14
4->
03
0)
^
00
H
f.
ITt
0-1
01
.£
4J
U-l
O
d
o
H
4-1
3
43
H
S-i
4J
CC
H
T3
>,
CJ
d
01
3
cr
s
s
p^
H4
^
43
-a
OJ
4-1
CO
t-H
I
w
NOI1BH1M33N03
E-30
-------
(lit 1 I
-
-
-
~
-
~
-
_
_
-
-
-
~
~*
: co
; o
i i
r CC
: 01
i i-
1 UJ
: o
- z:
- o
- UJ a
cb uJ
a: :=-
to a: a:
a; UJ UJ
o > co
1- CC CD
I «- o
z a:
o ID
y- a
" J 1
_4 cn
cr
-
i i i i i i
t
o
I i Ijlllili 1 1
I
^
1
*
1
\
\
\
\
\
\
\
\
Iv_ ~
^^
V 2E
V '
I '
1
*
1
t
\
\ -
\
\ -
I
\ I
v V
s A
j :
5 ~
V
\ j
ii
u :
u
u ^
\ ' ~
\ 1 "~
Y ^
II ~
i *
a '
uJ V
cc J
i
~i
CJ
I J
CC 5
cj J :
i
u
-
"
-
1 1 1 1 1 1 1 1 1 t 1
O -r-
0 '
CO
CO
CO
*
CO
CO
CO
CO
CO
«
ID
CO
CO
CD
r***
m
CD
ID
*^*
m
CM
*^
LD
O
^H
CD
o
^"
o
z:
UJ
a
UJ
u_
UJ
^>>
l__l
I
1
CC
*~^
TT"
^
U
T3
0)
S-i
cn
PI
O
4_J
cn
o>
60
H
r!
i/~l
CN)
cu
t-1.
1 1
4-1
O
c
o
H
4_l
^
o
H
4-1
cn
H
TT^
^
o
CU
3
cr
o>
4-1
01
J>
H
4-1
Cfl
i 1
3
3
O
0)
4-1
4-1
O
C
0
H
S-i
CO
fX
E
o
c_>
B
a>
CN
i
w
CU
S-i
3
H
fn
C
H
S-i
3
O
cn
O
4J
H
C
C
E
i t
r t
Cfl
4J
cfl
cn
c
o
H
4-1
Cfl
S-fl
4J
c:
01
CJ
c
o
CJ
o
o
CO
01
60
cfl
CU
cfl
S-i
3
O
1
ro
d
3
O
S-4
60
O
CO
o
cn
3
C
H
E
~~^
1
0)
S-i
4-1
01
£>
H
4-1
CO
3
E
3
CJ
CU
^
4-1
_£.
4-1
H
£»
E
CO
60
O
ex
60
c
H
S-i
o
41-1
H
C
0
E
0
o
CO
4J
CO
CU
0)
,r^
4-}
4-1
o
S-j
CO
4)
T3
C
O
CJ
CU
cn
0)
ft
4-1
1
3
CJ
rH
CO
O
cn
C
O
H
4-1
CO
S-i
4-1
C
cu
o
£
o
u
01
60
CO
S-i
0)
£>
Cfl
S-4
3
0
1
cn
4-1
CC
01
60
H
CN
CU
fj
4-1
4-1
o
c
o
H
4-1
3
H
S-4
4-1
cn
H
n
CJ
c
O)
3
cr
1
,-J
>
"O
CU
4J
CO
t 1
E-31
-------
1
^
_
_
-
m
-
"
~
-
-
-
-
:
:"
-
~
;
_
-
-
-
-
"
_
-
~
"
-
I
3
0
I
CO
a:
o
l-
i i
_i -
a: OH 3 '
UJ UJ CJ 1 _
> co _j
cr tn cr
O CJ I
f)^
ID
o
-*
I
^-
CM
i
-
l _
i i i i i
V« ?> CUj
O3 ^ c: \ \
, 01 -H 01 3
03 I-I »-( U
42 3 4-1 T 1
O T3 cC
01 O
4-1 CO >
O) 03 I-i -H W
O) o> o 4J d
O) 4=: 4-1 co o
. 60 -H f-H -H
H C 3 4-)
42 O E rt
S 3 !-<
U") U 4-*
CM rH d
i l dJ 01
O> 03 42 0
-c 4-1 d
OJ 4-1 4J 0
m ca 4= a
UJ 4-1 4J
" o w -H cu
d > OD
dO CO
o -H e i-i
H 4-1 CO
3 J-i M) cO
42 4-J 0
113 -H d M P
C35 >-> 0) P. 3
4-1 O O
en d so 42
>_ -M o d i
rT T3 o -H , CMO
LU O C 4-1 4-1
33 d c/1 -H w
C3 a) da;
[7j 3 0) O 42
_ jri o" cso E &fl
00 O- 01 cfl -i-i
U_ l-i I-i O 42
4-1 01 O
UJ > cd m
H -H |J W <3>
p- 4J 3 CD 42
cr co o rs 4J
CO _| M 42
=5 3 1 CO 4-1
=i 6 -=t 42 o
*- 3 CNl 4-1
10 =3 0 d
« CJ ^w o
0) TD O -H
42 d 4-1
"t 4J 3 H 3
O CO 43
4-1 I-i 0) -H S
tn ,M 4-i S
. d o -a en >-j
o ro d !-(
en 43 o T3 >^
«. -H 0 43
CSJ I-i ec 01 >,
cd 3 en O -a
P. c dcu
E -i-i co a) 4-1
o E 42 3 to
O s> ' 4-) CT* i 1
,
O
CO
ID [i
O
. 0)
l-i
3
H
«-%
«-« .
IQ
(WddJ
E-32
-------
APPENDIX F
DESCRIPTION OF THE DISPERSION MODELS EVALUATED
F.I
DESCRIPTION OF THE SHORTZ MODEL
The SHORTZ model is a highly generalized dispersion model, based
on the steady-state Gaussian plume concept, that is designed to calculate
the shortterm ground-level concentrations produced at a large number of
receptors by emissions from multiple stack, building and area sources. The
SHORTZ model was first used by Cramer, £t_ al. (1975) to calculate the short-
term air quality impacts of emissions from the major SO sources located
in and adjacent to Allegheny County, Pennsylvania. Subsequently, the SHORTZ
model has been used in numerous dispersion model analyses throughout the
United States, especially in areas of complex terrain (see Appendix H of
Bjorklund and Bowers, 1982). A detailed technical discussion of the SHORTZ
model and user's instructions for the SHORTZ computer code are contained in
the report by Bjorklund and Bowers (1982). The following discussion con-
siders only those features of the SHORTZ model applicable to the Westvaco
model evaluation study.
Plume Rise Equations
The effective stack height H of a buoyant plume is given by the
sum of the physical stack height h and the buoyant rise Ah. For an adiabatic
or unstable atmosphere, the final buoyant rise Ah is given by
Ah,
1
u {h} \2y
3F
1
1/3
(10h)
2/3
(F-l)
F-l
-------
where the expression in the brackets is from Briggs (1969; 1971; 1972) and
u{h} = the mean wind speed at the stack height h (m/sec)
y, = the adiabatic entrainment coefficient ~ 0.6 (Briggs, 1972)
1
F =
4 3
the initial buoyancy flux (m /sec )
T,
c
T"
V = the volumetric emission rate of the stack (m /sec)
= TT r w
r = inner radius of stack (m)
w = stack exit velocity (m/sec)
g = the acceleration due to gravity (m/sec )
T = the ambient air temperature (°K)
d.
T = the stack exit temperature (°K)
The factor f, which limits the plume rise as the mean wind speed at stack
height approaches or exceeds the stack exit velocity, is defined by
(F-2)
; u{h} < w/1.5
/3w - 3u{h}\ /, r -ri i
1 . w/1.5 < u{h} < w
\ w /
u{h} < w
(F-3)
F-2
-------
The Cramer, et al. (1975) stack-tip downwash correction factor f is
generally not applied to stacks with Froude numbers less than about 3.0.
The corresponding Briggs (1969; 1971; 1972) final rise formula for a stable
atmosphere (potential temperature gradient greater than zero) is
Ah = -<
s
6F
1/3
; TT u{h} S 1/2 < lOh
3F
1 - cos
u{h}
u{h}
1/3
-i /?
; IT u{h} S ' > lOh
f (F-4)
where
'2
S
= the stable entrainment coefficient ~ 0.66 (Briggs, 1972)
_g _3_e
T 9z
a
= vertical potential temperature gradient (°K/m)
The entrainment coefficients
Y0 are based on the suggestions of
Briggs (1972). It should be noted that Equation (F-4) does not permit the
calculated stable rise Ah to exceed the adiabatic rise Ah , as the
s N
atmosphere approaches a neutral stratification (96/8z approaches 0). A
procedure of this type is recommended by Briggs (1972).
Ground-Level Concentration Equations
The steady-state Gaussian plume equation used by the SHORTZ model
to calculate the ground-level concentration at downwind distance x and
crosswind distance y is
F-3
-------
KQ
TT u H a a
y
{Vertical Term} {Lateral Term}
where
K = scaling coefficient to convert input parameters to dimen-
sionally consistent units
Q = °source emission rate (mass per unit time)
u{H} = mean wind speed at the plume stabilization height H (m/sec)
(F-5)
a ,a
Y
= standard deviations of the lateral and vertical concentra-
tion distributions at downwind distance x (m)
The Vertical Term refers to the plume expansion in the vertical
or z direction and includes a multiple reflection term that limits plume
growth to the surface mixing layer.
{Vertical Term} =
exp
2 a
n=l
exp
, /2n H + H1
1 / m
exp
. /2n H -
If m
(F-6)
where H is the depth of the surface mixing layer. The exponential terms
in the infinite series in Equation (F-6) rapidly approach zero near the
source. At the downwind distance where the exponential terms exceed
exp (-10) for n equal 3, the plume has become approximately uniformly mixed
within the surface mixing layer. In order to shorten computer computation
time without loss of accuracy, Equation (F-6) is changed to
F-4
-------
a
{Vertical Term} = - (F-7)
m
beyond this point. Equation (F-7) changes the form of the vertical
concentration distribution from Gaussian to rectangular. If H exceeds
H , the Vertical Term is set equal to zero which results in zero values
m
for the ground-level concentrations.
In complex terrain, the SHORTZ Vertical Term is modified by the
use of effective plume stabilization heights and mixing depths under the
following assumptions:
The actual top of the surface mixing layer extends over
the calculation grid at a constant height above mean
sea level; the actual top of the surface mixing layer
should not be confused with the effective top of the
surface mixing layer, which is a mathematical device
used to preclude violations of the Second Law of
Thermodynamics when plumes pass over elevated terrain
The axis of a plume contained within the surface mixing
layer remains at the plume stabilization height above
mean sea level, and the plume may impact elevated
terrain within the surface mixing layer under stable,
neutral or unstable conditions
Plumes that stabilize above the top of the surface
mixing layer do not contribute to significant
ground-level concentrations at any receptor (this
assumption also applies to flat terrain), including
receptors that are above the top of the surface mixing
layer
F-5
-------
In order to determine whether the stabilized plume is contained
within the surface mixing layer, it is necessary to calculate the mixing
depth H*{z } at the source from the relationship
m s r
Hm{zs}
H + z - z
mas
(F-8)
where
H = the depth of the surface mixing layer measured at a point
with elevation z above mean sea level (m)
a
m
z = the height above mean sea level of the source (m)
s
Equation (F-8) is represented schematically in Figure F-l, which assumes
that z is the elevation of an airport. As shown by the figure, the
3.
actual top of the surface mixing layer is assumed to remain at a constant
elevation above mean sea level. If the height H of the stabilized plume
above the base of the stack is less than or equal to H*{z }, the plume
Til S
is defined to be contained within the surface mixing layer.
The height H of the stabilized plume above mean sea level is
given by the sum of the height H of the stabilized plume above the base of
the stack and the elevation z of the base of the stack. At any
elevation z above mean sea level, the effective height H'{z} of the plume
centerline above the terrain is then given by
H'{z} =
H - z ; H - z > 0
o ' o
0 ; H - z < 0
o
(F-9)
F-6
-------
C
H
01
C
H
CO
4-)
C
O
O
(U
0)
N
H
"tfl
4J
W
O
J3
4-1
JJ
ai
5
a)
0)
c
H
4-1
QJ
13
0)
en
n
r^ 0)
tn>^
N ca
re oo
c
J3 -H
U CO
(X -H
(!) g
60 O
C cd
H >n
X VJ
H 3
S W
W
F-7
-------
The effective mixing depth H'{z} above a point at elevation z
above mean sea level is defined by
H'{z} =
m
H
m
H + z - z
m a
; z > z
a
; z < z
' a
(F-10)
Figure F-2 illustrates the assumptions implicit in Equation (F-10). For
grid points at elevations below the measurement base elevation (assumed in
Figures F-l and F-2 to be an airport), the effective mixing depth H'{z}
is allowed to increase in a manner consistent with Figure F-l. However, in
order to prevent a physically unrealistic compression of plumes as they
pass over elevated terrain, the effective mixing depth is not permitted to
be less than the mixing depth measured at the airport. It should be
emphasized that the concentrations are set equal to zero for grid points
above the actual top of the mixing layer (see Figure F-l).
The Lateral Term refers to the crosswind expansion of the plume
and is given by the expression
{Lateral Term} = exp
y
(F-ll)
where y is the crosswind distance from the plume centerline to the point at
which concentration is calculated.
The SHORTZ model uses a wind-profile exponent law to adjust the
observed mean wind speed from the measurement height to the stack height h
for use in the plume rise calculations and to the plume stabilization
height H for use in the concentration calculations. In complex terrain,
the SHORTZ model assumes that the mean wind speed at any given height above
-------
cn
o
H
3
O
cfl
a
U
C
a)
o
C
o
o
CO
4J
e
r-i
o
ex
-a
H
j-i
M
0)
60
H
cn
cn
rt
-' e
sa
o.
CU
60
C
H
X
cu
>
o
0)
u-t
w
I
fn
W
P4
O
O
F-9
-------
mean sea level is constant. Thus, the mean wind speed u measured at a
R
height z above the surface at a point with elevation z above mean sea
ix 3.
level is adjusted to the stack height for the plume rise calculations by
the relationship
u{h} = -I
- /ho' Za\
! I ; h > z + z_,
R\ ZR I ° a R
; h < z + z,,
' o a R
(F-12)
where h is the height above mean sea level of the top of the stack and p
is the wind-profile exponent. Similarly, the wind speed u{H} used in the
concentration calculations is given by
u{H> =
H - z
R z
R
; H < z + z_,
' o a R
(F-13)
Dispersion Coefficients
The SHORTZ model uses Cramer (1976) dispersion coefficients,
which are of the form
a' f {x} x
A y
(F-14)
and
a' f {x} x
t z
(F-15)
F-10
-------
where a (a ) is the lateral (vertical) dispersion coefficient, o'
y z A.
(a') is the lateral (vertical) turbulent intensity or standard deviation
LJ
of the wind azimuth (elevation) angle in radians and f (f ) is the
lateral (vertical) "universal function." The generalized a and a
y z
equations used by the SHORTZ model include provisions for lateral and
vertical virtual distances to account for the effects of entrainment on the
initial dispersion of a buoyant plume.
Under the assumptions that a is proportional to x over a dis-
J a
tance x downwind from an ideal point source and to x at longer down-
wind distances, Cramer, et al. (1972) derive the a expression
y
ay{x}
a! x
A ry
x + x - x (1-a)
Y ry'
ax
L ry J
-lO.
(F-16)
ax
JLL
ry \ x a;
\ ry A
iE.
I/a
R
- X
R
xry(l-a)
S X
ry
> X
ry
(F-17)
where a is the lateral dispersion coefficient at downwind distance
yR
x . The SHORTZ model does not allow the lateral virtual distancex to be
less than zero and, in the default mode, defines x as 50 meters and a
ry
as 0.9.
As discussed in Appendix A, it has been the experience of the
H. E. Cramer Company that the lateral "universal function" implicit in
F-ll
-------
Equation (F-16) adequately accounts for the effects of vertical wind-
direction shear on lateral plume expansion in most situations (see
Bjorklund and Bowers, 1982, p. 2-33). However, based on an examination of
the hourly wind-direction and SO. concentration measurements from the
first two quarters of the Westvaco monitoring program, the H. E. Cramer
Company (January 1981) reported to EPA Region III that the plume from the
Westvaco Main Stack is subject to very large vertical wind-direction shears
as it rises through the highly channeled valley flow and enters the flow
above the elevated terrain. Because of these large wind-direction shears,
:>
our January 1981 report suggested that it would be appropriate to modify
SHORTZ for application to the Luke Mill by inclusion of the Cramer, et al.
(1972) technique for accounting for the effects of vertical wind-direction
shear on crosswind plume expansion. Following this approach, the total
lateral dispersion coefficient a is given by
V
r 2i1/2
2 Me'xf
ay A.3JJ
(F-18)
where a is the unmodified SHORTZ lateral dispersion coefficient, x is
the downwind distance and A6 is the wind-direction shear in radians for the
layer containing the plume. Equation (F-18) was used by the SHORTZ model
in the Westvaco model performance evaluation described in Section 3.2.
The SHORTZ model assumes that, if a multiple reflection term
(Equation (F-6)) is used to confine the plume within the surface mixing
layer, a may be assumed to be proportional to x at all downwind
z
distances. The resulting a expression is
z
°z{x} =
F-12
-------
°zR QzR
°E ^ ' °E
o - °ZR
u , ,
E
^XR
^ -v
(F-20)
where a is the vertical dispersion coefficient at downwind distance
xR -j
Briggs (1972) notes that photographs of buoyant plumes show the
radius to be approximately 0.5 times the plume rise. Under the assumption
of Gaussian lateral and vertical concentration distributions at the downwind
distance of final plume rise x , the SHORTZ model defines the lateral and
K
vertical dispersion coefficients at this distance as
0.5 Ah
2.15
(F-21)
The downwind distance to final rise x is given by
R
lOh
tr u{h} S"
lOh
30
If
>0
> 0 and TT u{h} S 1/2 > lOh
(F-22)
F-13
-------
F.2 DESCRIPTION OF THE VALLEY MODEL
The EPA Valley model (Burt, 1977) is primarily designed to cal-
culate maximum 24-hour average ground-level concentrations produced by
stack emissions in complex terrain. The Valley model is a screening model
and is intended for use with hypothetical rather than actual meteorological
inputs (Burt and Slater, 1977). The Valley model makes the "worst-case"
assumptions that a plume in an elevated stable layer is confined within a
22.5-degree sector for 6 hours during a 24-hour period and that this plume
directly impinges on any terrain at the height of the plume centerline.
Common practice with the Valley model is to assume that the mean wind speed
at plume height during the 6 hours of impingement is 2.5 meters per second
and that the vertical dispersion in the elevated stable layer is equivalent
to that predicted by the Pasquill-Gifford a curve for F stability (Turner,
z
1969). The following description of the Valley model is of the version
contained in the UNAMAP-4 series of models and used in the Westvaco model
evaluation.
Plume Rise Equations
The Valley model uses the Briggs (1971; 1975) plume rise equations.
At downwind distances less than the distance to final plume rise, plume
rise is given by the adiabatic rise equation
x2/3 (F-23)
u
If YI in Equation (F-l) is set equal to the SHORTZ model's default value
of 0.6 and the stack-tip downwash correction factor f is set equal to unity,
Equations (F-l) and (F-23) are identical at a downwind distance of ten
stack heights (lOh). Under stable conditions, the plume rise given by
Equation (F-23) is terminated in the Valley model when it equals the final
plume rise predicted by the Briggs (1975) stable plume rise equation
F-14
-------
1/3
Ah
2.6
(F-24)
u S
If y in Equation (F-4) is set equal to the SHORTZ model's default value
of 0.66 and the downwash correction f is set equal to unity, the top line
of Equation (F-4) is the same as Equation (F-24) except that the coefficient
2.6 in Equation (F-24) has a value of 2.4 in Equation (F-4). In other
words, the final plume rise calculated by the Valley model under stable
conditions exceeds the corresponding final plume rise calculated by the
SHORTZ model by a factor of 1.08 (2.6/2.4). The Valley model assumes that
the vertical potential temperature gradient with F stability is 0.035 degrees
Kelvin per meter.
Ground-Level Concentration Equations
The Valley model gives the 24-hour average ground-level concentra-
tion attributable to 6 hours of plume impingement on elevated terrain under
stable meteorological conditions as
KQ
2/27 u a , A9f x
zT
exp
C{z,HQ}
(F-25)
where A6' is the sector width (0.3927 radians) within which the plume is
assumed to be contained during the 6 stable hours. The effective plume
height H' under stable conditions is given by
H'{z} =
H - z ; H - z > 10m
o o
10m ; H - z < 10m
o
(F-26)
F-15
-------
Equations (F-9) and (F-26) are equivalent for all practical purposes except
that Equation (F-9) is only used by the SHORTZ model when the plume is
contained within the surface mixing layer. The correction C{z,H }, which
reduces the calculated concentration to zero on terrain 400 meters or more
above the plume centerline, is given by
C{Z,HO} =
(401 - D{z,H })
o
400
(F-27)
where
D{z,H
z-H ; 1 < z - H < 401m
o o
401 ; z - H > 401m
o
(F-28)
Vertical Dispersion Coefficients
The Valley model requires only the vertical dispersion coefficient
because emissions are assumed to be uniformly distributed in the horizontal
within a 22.5-degree sector. The total vertical dispersion coefficient
a is given by
1/2
(F-29)
where a is the Pasquill-Gifford vertical dispersion coefficient (Turner,
£
1969), represented in the Valley model by an equation of the form
a = a x + c
z
F-16
(F-30)
-------
The coefficients a, b and c in Equation (F-30) are functions of stability
and downwind distance. The second term on the right hand side of Equation
(F-29) is intended to account for the effects of entrainment ("bouyancy
induced dispersion") on the initial vertical growth of a buoyant plume.
The treatment of the effects of entrainment on initial plume growth in the
SHORTZ and Valley models differs in two ways. First, the SHORTZ model assumes
a Gaussian distribution of material at the distance of final rise, whereas
the Valley model assumes a uniform ("top hat") distribution of material.
Second, the SHORTZ model uses virtual distances to account for the enhanced
initial growth of the plume, whereas the Valley model adds variances.
F.3 DESCRIPTION OF THE COMPLEX I AND COMPLEX II MODELS
The undocumented EPA Complex I and Complex II models were de-
veloped from the computer code for the EPA MPTER model (Pierce and Turner,
1980). The Complex I and II models are designed to use actual rather than
hypothetical hourly meteorological inputs, and both models make stability-
dependent assumptions about how complex terrain affects plume height above
terrain. The primary difference between the two models is that the Complex
I model assumes that emissions during each hour are uniformly distributed
in the horizontal within a 22.5-degree sector, whereas the Complex II model
assumes a Gaussian hourly crosswind concentration distribution. The
following paragraphs describe the Complex I and II models as used in the
Westvaco model evaluation study and recommend improvements to the models.
Plume Rise Equations
The Complex I and II models use the Briggs (1971; 1975) plume
rise equations. If the Pasquill stability category is neutral (D) or un-
stable (A, B or C), both models assume an adiabatic thermal stratification
and use Equation (F-23) to calculate plume rise as a function of downwind
distance. Final plume rise is assumed to occur at the distance 3.5 x*
(Briggs, 1971) , where x* is given by
(F-17)
-------
<* =
14 F5/8 ; F < 55 m4/sec3
34 F2/5 ; F > 55 m4/sec3
(F-31)
If the Pasquill stability category is stable, the Complex I and II models
use Equation (F-23) to calculate distance-dependent plume rise with the
final rise occurring at the distance where the plume rises given by Equations
:>
(F-23) and (F-24) are equal. The default vertical potential temperature
gradients assumed by both models during hours of E and F stability are
0.020 and 0.035 degrees Kelvin per meter, respectively.
It should be emphasized that the official versions of the Complex
I and II models key the selection of adiabatic or stable plume rise equation
on the Pasquill stability category rather than on the vertical potential
temperature gradient. Our analysis of the meteorological conditions during
the periods with the highest observed SO concentrations at the monitoring
network on the elevated terrain in the sector southeast of the Westvaco
Main Stack (see Figure 1-1) indicated that a stable thermal stratification
usually is required for the plume centerline to be low enough for the plume
to affect the monitoring network. Because the Pasquill stability categories
indicated for the hours with high observed concentrations by objective
stability classification schemes (for example, Turner, 1964) frequently
were neutral or unstable, the Complex I and II models automatically assumed
an adiabatic stratification which caused the models to overestimate plume
rises and hence to underestimate the ground-level concentrations at the
monitoring network. We therefore modified both models to read hourly values
of the vertical potential temperature gradient and to key the selection of
the adiabatic or stable plume rise equation on the potential temperature
gradient, the same approach as used by the SHORTZ model. Additionally, we
added logic to ensure that the calculated stable plume rise did not exceed
the corresponding calculated adiabatic plume rise.
F-18
-------
Ground-Level Concentration Equations
The Complex I model assumes that the plume is uniformly distri-
buted in the horizontal within a 22.5-degree sector during each hour. The
ground-level concentration at downwind distance x within this sector is
given by
2KO
x(x} = {Vertical Term} (F-32)
/2TT u{h} a , A6' x
zl
If only the first term of the full Vertical Term given by Equation (F-6) is
considered under stable conditions, Equation (F-32) has the same form as
the corresponding equation for the Valley model (Equation (F-25)) except
that Equation (F-32) has not been divided by a factor of 4 to account for
the Valley model's assumptions that es-sentially the same hourly concentra-
tion occurs during 6 hours of a 24-hour period. The Complex II model gives
the hourly ground-level concentration at downwind distance x and crosswind
distance y as
X(x,y} = {Vertical Term} {Lateral Term} (F-33)
azT °yT
The Complex I and II models assume that the mixing height H is
infinite with E or F stability and is terrain following with A, B, C or D
stability. Both models define the effective plume height H1 as
H' = MAXJFH, H - (l-F)(Ho-z)j (F-34)
where the "plume path coefficient" F is assumed to be 0.5 for the unstable
(A, B and C) and neutral (D) Pasquill stability categories and to be zero
F-19
-------
for the stable (E and F) Pasquill stability categories. Thus, the Complex
I and II models make the same basic assumption about plume height above
terrain under stable conditions as the Valley model. For consistency with
the Valley model, the Complex I and II models do not allow H' to be less
than 10 meters.
The Complex I and II models assume that wind speed is a function
of height above local ground level (rather than above mean sea level as
assumed by the SHORTZ model). The wind speed UD measured at height
K
z is adjusted to the stack height h for use in both the plume rise and
K
concentration calculations by the expression
u{h} = u -M (F-35)
R/
where the wind-profile exponent p is assigned on the basis of the Pasquill
stability category.
The Complex I and II models accept sequential hourly emission
rates and calculate hourly stack exit velocities from a single input exit
velocity under the assumption that the exit velocity is directly proportional
to the emission rate. Because of variations in coal-sulfur content, this
approximation is not highly accurate for the Westvaco Main Stack. Also,
the Complex I and II models do not allow for hour-to-hour variations in the
stack exit temperature. We therefore modified the computer codes for the
two models to allow them to accept hourly values of the stack exit velocity
and exit temperature.
Dispersion Coefficients
The Complex I and Complex II models use Equation (p-29) to
calculate the total vertical dispersion coefficient, with the Pasquill-
F-20
-------
Gifford vertical dispersion coefficient a represented by an equation of
z
the form
= ax
(F-36)
rather than of the form of Equation (F-30). The coefficients a and g in
Equation (F-36) are functions of stability and downwind distance. The
Complex I model does not require lateral dispersion coefficients because of
the sector-averaging assumption. The Complex II model defines the total
lateral dispersion coefficient a as
1/2
r 2 2]12
= [a + (Ah/3.5rJ (F-37)
where a is the Pasquill-Gifford lateral dispersion coefficient,
represented by equations of the form
a = 465 x tan(TH) (F-38)
TH = 0.01745 (d - e £n(x)) (F-39)
The coefficients d and e in Equation (F-39) are functions of stability.
The second term on the right-hand side of Equation (F-37) is intended to
account for the effects of entrainment on lateral plume growth.
F-21
-------
F.4 DESCRIPTION OF THE LUMM MODEL
The Luke Mill Model (LUMM) is a single-source model that was
developed by Hanna, ^jt al. (1982a) for application to the Westvaco Main
Stack. The LUMM and SHORTZ models are qualitatively similar in that both
models use direct turbulence measurements to predict plume expansion, both
models attempt to account for the effects of vertical wind-direction shear
on lateral plume growth, and both models key the selection of the appro-
priate plume rise equation on the vertical potential temperature gradient.
Also, both the LUMM and SHORTZ models utilize onsite meteorological measure-
ments to the maximum extent possible. Hanna, ot_ al^. (1982a) evaluated the
performance of six versions of the LUMM model, two of which differed only
in meteorological inputs. The version of the LUMM model considered by
Hanna, _e_t a^. (1982a) to give the best overall performance (Model 4) is
briefly described below.
Plume Rise Equations
The LUMM model considers only two stabilities: (1) stable, and
(2) "neutral" (adiabatic or unstable). If the vertical potential temper-
ature gradient is positive, the final plume rise is given by Equation (F-24)
To ensure that the stable plume rise Ah does not exceed the corresponding
adiabatic plume rise Ah , the LUMM model also uses Equations (F-23) and
(F-31) to calculate Ah.T for hours with stable (positive) potential temper-
N
ature gradients. Neutral conditions are assumed if Ah is less than
Ah or if the potential temperature gradient is less than or equal to
s
zero. Otherwise, stable conditions are assumed.
Ground-Level Concentration Equations
During hours with neutral conditions, the LUMM model uses Equation
(F-33) to calculate the ground-level concentration at downwind distance x
and crosswind distance y except that only the first term of the generalized
Vertical Tern (see Equation (F-6) ) is used by the LUMM model. That is, the
F-22
-------
multiple reflection portion of the Vertical Term that confines the plume
within the surface mixing layer is not incorporated in the LUMM model. The
effective plume height H1 under neutral conditions is given by Equation
(F-34) with the "plume path coefficient " F set equal to 0.4 to fit the
observations at the Luke Mill (Hanna, _et_ al., 1982a, p. 5-7). Unlike the
SHORTZ, Complex I and Complex II models, the LUMM model does not use a
wind-profile exponent law to extrapolate the wind speed from the measure-
ment height to the stack or plume height. Instead, the LUMM model esti-
mates the appropriate wind speeds from the onsite tower wind measurements
o
(see Table A-l of Hanna, et_ _al. , 1982a).
During hours with stable conditions, the effective plume height
assumed by the LUMM model depends on the relationship between the plume
height above plant grade and the height of a critical streamline H above
plant grade. The critical streamline height is given by
H = Az (1 - Fr) (F-40)
c max v '
where the Froude number is defined as
Fr = ryr- (F-41)
Az S '
max
and Az is (Hanna, et al., 1982a, p. 5-8) the "maximum mountain height
max
above the stack base in the direction of interest within about 10 km of
the stack." If H is greater than H , the ground-level concentration is
calculated from Equation (F-33), with H' given by Equation (F-34). If H is
less than H and the receptor is below H , two concentrations are calculated.
c v c
The first concentration is given by Equation (F-33) with H' redefined as
F-23
-------
H' = MAX
(F-42)
The second concentration calculated under stable conditions when H is less
than H is
c
RKQ
yT
exp
(F-43)
The reflection coefficient R in Equation (F-43) is set equal to 1.2 in the
LUMM model because Hanna, et al. (1982a) consider this value to- be "appro-
priate for terrain slopes of about 20 to 30° encountered in the area sur-
rounding the Luke Mill." The minimum of the two concentrations calculated
when H is less than H is assumed by the LUMM model to be the correct
concentration. Additionally, if a receptor is above H and H is below
H , the concentration at the receptor is assumed to be zero (Hanna, 1982).
Dispersion Coefficients
The LUMM model's total lateral dispersion coefficient a ~ is of
the form
r 2 2 2 1
= a {turbulence} + a {buoyancy} + a {wind shear} (F-44)
L y y y J
F-24
-------
Similarly, the LUMM model's total vertical dispersion coefficient a is
of the form
°zT
= a {turbulence} + a {buoyancy} (F-45)
I Z Z _J
The turbulence contributions to a and a are given by
a {turbulence} = I f {x} x (F-46)
o {turbulence} = I f {x} x ('F-47')
y z z v '
where I and I are the lateral and vertical turbulent intensities,
y z
respectively. (The lateral turbulent intensity I in Equation (F-45)
corresponds to a' in Equation (F-14)» while the vertical turbulent inten-
£\
sity I in Equation (F-47) corresponds to a' in Equation (F-15).)
Z ill
Comparison of Equations (F-46) and (F-47) with Equations (F-14) and (F-15)
shows that the turbulence-induced components of the dispersion coefficients
used by the LUMM and SHORTZ models are based on the same concepts. However,
the two models differ in the assumed forms of the "universal functions"
f and f .
Y z
The "universal functions" used by the LUMM model were inferred
from the equations proposed by Briggs (1973) for rural dispersion coefficients.
The lateral "universal function" for both neutral and stable conditions is
defined in the LUMM model as
F-25
-------
f {x} = (1 + O.OOOlx)
-1/2
(F-48)
The vertical "universal function" varies with stability and wind speed in
the LUMM model and is given by
f {x} =
z
1.0
(1 + 0.0003x)
(1 + 0.0015x)
-1
; neutral and u < 8 m/sec
; stable and u < 8 m/sec
; u > 8 m/sec
(F-49)
The LUMM, Valley, Complex I and Complex II models all assume a
uniform ("top hat") concentration distribution for the plume at the downwind
distance of final plume rise. Additionally, all four'models include the
effects of entrainment by the buoyant plume ("buoyancy induced dispersion")
by adding variances. However, the buoyancy contribution assumed by the
LUMM model, which is given by
a {buoyancy} = a {buoyancy} = 0.4 Ah
y z
(F-50)
is a factor of 1.4 larger than assumed by the three other models.
The LUMM model assumes that the contribution of vertical
wind-direction shear to lateral plume expansion is given by
a {wind shear} = 0.34 A9' x
y
F-26
(F-51)
-------
This expression is the same as used by the SHORTZ model (see Equation
(F-18)) except that the LUMM model's shear coefficient of 0.34 is a factor
of 1.46 larger than the SHORTZ model's shear coefficient. The first
version of the LUMM model assumed a shear coefficient of 0.17 on the basis
of suggestions made by Pasquill (1976). Although there are theoretical
arguments in support of the doubling of the shear coefficient in the
subsequent versions of the LUMM model, this doubling was also required to
fit the LUMM model to the Westvaco data set (Hanna, 1982).
As noted above, the LUMM model does not include the multiple
reflection portion of the full Vertical Term (see Equation (F-6)) to confine
the plume within the surface mixing layer. To avoid underestimation of
concentrations at downwind distances where the effects of the restriction
on vertical mixing at the top of the surface mixing layer begin to affect
ground-level concentrations, the LUMM model does not allow the total
vertical dispersion coefficient a to exceed 340 meters. The first two
Z i.
versions of the LUMM model, which did not allow a to exceed 136 meters,
z
systematically underestimated the concentrations observed at Monitor 10
(the Stony Run monitor in Figure 1-1) because a larger o was required to
Z
mix the plume downward to this monitor (Hanna, 1982).
F-27
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-903/9-83-OQ2
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Westvaco Luke, Maryland Monitoring Program: Data
Analysis and Dispersion Model Validation (Final
Report)
5. REPORT DATE
June 1983
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
J. F. Bowers, H. E. Cramer, W. R. Hargraves and
A. J. Anderson
8. PERFORMING ORGANIZATION REPORT NO.
TR-83-153-01
9, PERFORMING ORGANIZATION NAME AND ADDRESS
H. E. Cramer Company, Inc.
P. 0. Box 8049
Salt Lake City, UT 84108
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
Contract No. 68-02-3577
Modification No. 2
12. SPONSORING AGENCY NAME AND ADDRESS
U. S. Environmental Protection Agency, Region III
6th and Walnut Streets
Philadelphia, Pennsylvania 19106
13. TYPE OF REPORT AND PERIOD COVERED
Final: July 1981-Feb 1983
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The Westvaco data set consists of detailed records of hourly emissions, mete-
orological and SCL air quality data collected in the vicinity of the Westvaco
Corporation Paper Mill at Luke, Maryland during the period December 1979 through
November 1981. The purpose of the Westvaco monitoring program was to acquire the
data needed to select the most appropriate complex terrain dispersion model for use
in establishing an S02 emission limitation for the Luke Mill. The major objectives
of the work described in this report were to: (1) analyze and evaluate the Westvaco
meteorological and air quality data in order to develop the most suitable data set
to evaluate complex terrain dispersion models; and (2) use the Westvaco data set to
evaluate the performance of EPA's Valley, Complex I and Complex II models, the H. E.
Cramer Company's SHORTZ model and Westvaco Corporation's Luke Mill Model (LUMM).
The results of the model performance evaluation support the use of the Valley, Com-
plex I and Complex II models as safe-sided screening models when little or no onsite
meteorological data are available. The SHORTZ and LUMM models provided accurate and
unbiased estimates of the 25 highest 1-hour, 3-hour and 24-hour average concentra-
tions at some of the monitors and systematically biased estimates at the other moni-
tors. Based on the terms of a model evaluation protocol, the LUMM model was selec-
ted for use in establishing an SO emission limitation for the Luke Mill. ,
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Air Pollution
Turbulent Diffusion
Meteorology
Mathematical Models
Computer Models
Dispersion Models
Complex Terrain
Valley Model
Complex I Model
Complex II Model
SHORTZ Model
3. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
20 SECURITY CLASS (Tins page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-------