ASSESSMENT OF THE AIR QUALITY IMPACT
OF SO_ EMISSIONS FROM THE
£i
ASARCO-TACOMA SMELTER
Final Report
Prepared By
H. E. Cramer, J. F. Bowers, and H. V. Geary
EPA 910/9-76-028
July 1976
Prepared For
U. S. ENVIRONMENTAL PROTECTION AGENCY
Region X
Surveillance and Analysis Division
1200 Sixth Avenue
Seattle, Washington 98101
H. E. CRAMER COMPANY, INC.
University of Utah Research Park
P. O. Box 8049
Salt Lake City, Utah 84108
-------
ASSESSMENT OF THE AIR QUALITY IMPACT
OF S02 EMISSIONS FROM THE
ASARCO-TACOMA SMELTER
Errata
P. XVI Sixth line change telemertered to telemetered
P. XVIII Heading of third column should read 1-hour standard
of 0.40ppm
-------
EPA 910/9-76-028
July 1976
ASSESSMENT OF THE AIR QUALITY IMPACT OF SO EMISSIONS
FROM THE ASARCO-TACOMA SMELTER
prepared by
H. E. Cramer, J. F. Bowers, and H. V. Geary
prepared for
U. S. ENVIRONMENTAL PROTECTION AGENCY
REGION X
Surveillance and Analysis Division
1200 Sixth Avenue
Seattle, Washington 98101
Contract No. 68-02-1387
Task Order No. 4
EPA Project Officer: D. A. Wilson
H. E. Cramer company, inc.
UNIVERSITY OF UTAH RESEARCH PARK
POST OFFICE BOX 8049
SALT LAKE CITY, UTAH 84108
-------
This air pollution report is issued by Region X, Environmental Protection
Agency, to assist state and local air pollution control agencies in carrying out their
program activities. Copies of this report may be obtained, for a nominal cost,
from the National Technical Information Service, 5285 Port Royal Road, Springfield,
Virginia 22161.
This report was furnished to the Environmental Protection Agency by the
H. E. Cramer Company, Inc. of Salt Lake City, Utah in fulfillment of EPA Contract
No. 68-02-1387, Task Order No. 4. This report has been reviewed by Region X,
EPA and approved for publication. Approval does not signify that the contents
necessarily reflect the views and policies of the Environmental Protection Agency,
nor does mention of trade names or commercial products constitute endorsement
or recommendation for use.
Region X Publication No. EPA 910/9-76-028
i £
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA 910/9-76-028
2.
3. RECIPIENT .CTESSIOWNO.
4. TITLE AND SUBTITLE
ASSESSMENT OF THE AIR QUALITY IMPACT OF SC>2
EMISSIONS FROM THE ASARCO-TACOMA SMELTER
5. REPORT DATE
July 1976
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
H. E. Cramer, J. F. Bowers and H. V. Geary
8. PERFORMING ORGANIZATION REPORT NO.
TR-76-105-01
9. PERFORMING ORGANIZATION NAME AND ADDRESS
H. E. Cramer Company, Inc.
P. O. Box 8049
Salt Lake City, Utah 84108
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
Contract No. 68-02-1387
Task Order No. 4
12. SPONSORING AGENCY NAME AND ADDRESS
U. S. Environmental Protection Agency, Region X
1200 Sixth Avenue
Seattle, Washington 98101
13. TYPE OF REPORT AND PERIOD COVERED
Final June 1975 - June 1976
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT rpj^ major purpOse of the work described in this report was to use diffusion modeling
techniques to calculate the impact on ambient air quality of SO2 emissions from the ASARCO
copper smelter in Tacoma, Washington for the existing smelter configuration (51-percent con-
stant emissions control) and for 20 alternative smelter configurations with varying degrees of
constant emissions control. The accuracy of the modeling techniques was established by the
close correspondence obtained between calculated and observed short-term ground-level SO£
concentrations for 20 selected historical cases, when high hourly SO2 concentrations were
measured in the area surrounding the smelter, as well as by the close agreement between cal-
culated and observed annual average concentrations for 1972. The results of the model calcu-
lations show that the maximum allowable constant SO2 emission rate consistent with maintain-
ing the Washington Department of Ecology (DOE) and Puget Sound Air Pollution Control Agency
(PSAPCA) air quality standards ranges from 2000 to 2500 pounds per hour, depending on the
source configuration. These same SO2 emission rates will also preclude violations of the
National SO2 air quality standards. A review of the ASARCO Supplementary Control System
(SCS) shows that it has been effective in eliminating almost all violations of the National short-
term SO2 air quality standards. However, as evidenced by air quality measurements to date,
we believe the ASARCO SCS is not a viable method for preventing violations of the PSAPCA
and DOE 5-minute and 1-hour SO2 standards.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Air Pollution
Copper
Diffusion Models
Exhaust Gases
Meteorology
Monitor
Smelters
Sulfur Dioxide
13. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS {ThisReport)
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
iii
-------
IV
-------
ACKNOWLEDGEMENTS
Throughout the program of work culminating in the preparation of this
report, the H. E. Cramer Company, Inc. has greatly benefited from the assistance,
cooperation and guidance provided by many individuals.
We are especially indebted to Mr. Dean Wilson, EPA Region X Meteorolo-
gist and our Project Officer. Mr. Wilson assisted us in obtaining the data required
for our study, closely monitored the progress of our study, and coordinated our
contacts with EPA Region X and the Puget Sound Air Pollution Control Agency
(PSAPCA). Others within EPA Region X who provided guidance during the course
of our study include: Mr. G. Hofer, Chief, Air Surveillance and Investigation
Section; Mr. C. Findley, Chief, Air Technical Section; and Mr. R. Malatesta,
Technical Advisor. Mr. L. Sims, Chemical Engineer, assisted in the development
of source and emissions data for the possible future configurations of the ASARCO
smelter. Mr. W. Russell, Meteorologist, situated a continuous SO monitor
2t
within an area of predicted high short-term SO concentrations and operated the
^
monitor for a four-month period.
We are also greatly indebted to Mr. A. Dammkoehler, Air Pollution
Control Officer of PSAPCA, and to the members of his professional staff for their
assistance and cooperation. Mr. A. Kellogg, Chief-Technical Services; Mr. M.
Svoboda, Senior Meteorologist and Data Analyst; and Mr. K. Knechtel, Meteorolo-
gist and Systems Analyst, provided us with copies of the various ASARCO reports
to PSAPCA and with PSAPCA meteorological and air quality data. The historical
cases modeled in this study were, to a large extent, selected on the basis of the
recommendations of these gentlemen. Also, we wish to thank Mr. J. Roberts,
Air Pollution Engineer, for his assistance in obtaining emissions data for the
ASARCO smelter and the other major SO sources in the Seattle-Tacoma area.
Lt
-------
Finally, we wish to acknowledge the cooperation and assistance of the
personnel of the ASARCO-Tacoma Copper Smelter. Specifically, Mr. R. Welch
provided detailed information about the operation of the smelter1 s Supplementary
Control System and Mr. C. Randt provided detailed information about plant
operations and emissions.
VI
-------
EXECUTIVE SUMMARY
The assessment of the air quality impact of SO emissions from the
£t
ASARCO copper smelter in Tacoma, Washington described in this report was
divided into four phases:
• Background Study
• Model Development
• Model Application
• Evaluation of the ASARCO Supplementary Control System (SCS)
The Background Study Phase consisted of a review of all available meteor-
ological and air quality data for the Seattle-Tacoma area and of all available SO
£t
emissions data for the ASARCO smelter and other major SO sources. In addition,
£
ASARCO SCS procedures were studied, previous diffusion studies of the air quality
impact of SO emissions from the ASARCO smelter were analyzed and various
2
diffusion models were studied to determine those most likely to be applicable to the
ASARCO smelter. On the basis of this review, three critical meteorological
regimes were identified as being conducive to the highest short-term ground-level
SO concentrations resulting from ASARCO stack emissions. The critical
*L
wind-speed condition (moderate-to-strong winds in combination with neutral stabi-
lity) was determined to be the meteorological regime most likely to lead to
maximum 3-hour and 24-hour ground-level SO concentrations. The limited-mixing
2
condition (light-to-mode rate winds with the ASARCO plume contained within a
relatively shallow surface mixing layer) was found to be the meteorological regime
most frequently associated with violations of the 5-minute and 1-hour PSAPCA
standards. The fumigation condition was found to result in the simultaneous
occurrence of high ground-level SO concentrations at many points in the Seattle-
£i
Tacoma area. One long-term diffusion model was selected to calculate average
vii
-------
annual concentrations and five short-term diffusion models were selected to
calculate concentrations during the three critical meteorological regimes.
In the Model Development Phase, the validity of the long-term diffusion
model was tested by using it with 1972 meteorological data and emissions data
for ASARCO and the other major SO sources in the Seattle-Tacoma area to calculate
£i
annual average concentrations at the various SO monitor sites. In these calcu-
£
lations, the effects of the ASARCO Supplementary Control System (SCS) on SO
z
emissions were simulated by assigning an average emission rate, based on an
analysis of the 1972 smelter curtailment reports, to each joint combination of
Pasquill stability category and wind-speed category. The calculated and observed
1972 annual average SO concentrations at all monitors in the Tacoma area
th
(except the ASARCO monitor at N26 and Pearl) differed by less than one-half
the monitor threshold and accuracy (i.e., by less than - 0.005 parts per million).
To test the validity of the short-term models and to determine the models best
suited for application in this study, emissions data and meteorological data were
obtained for 20 selected cases during the 3-year period from 1972 through 1974
when high SO concentrations of one hour or longer duration were observed at
£
SO monitors. The 20 cases, which were selected with the assistance of PSAPCA
u
personnel, included meteorological conditions representative of all three critical
meteorological regimes. The emissions and meteorological data were used with
the short-term models to calculate SO ground-level concentrations at air
Zi
quality monitor sites for each case. These calculated concentrations were then
compared with the corresponding observed SO concentrations. In the case of the
£t
model selected for the limited-mixing and critical wind-speed conditions, the
average ratio of the calculated and observed hourly concentrations was unity
(all observed hourly concentrations considered in the comparison were above the
PSAPCA 1-hour standard of 0.25 parts per million). With respect to the model
selected for the fumigation condition, the average ratio of the calculated and
Vlll
-------
observed concentrations for the period of fumigation was equal to 0.8. The
calculated maximum hourly fumigation concentrations were approximately one-half
the observed hourly maximums. We believe this is principally explained by the
assumption made in the model calculations—in the absence of requisite wind
observations—that the wind-direction distribution at plume height prior to the
occurrence of fumigation is circular. It is important to note that all of the long-
term and short-term diffusion-model calculations were made without recourse to
calibration constants or any other arbitrary adjustment techniques. The Model
Development Phase thus established the validity of the selected modeling techniques
for estimating the impact of ASARCO SC>2 emissions in the Seattle-Tacoma area.
In the Model Application Phase, the validated modeling techniques were
used to calculate maximum long-term and short-term ground-level SO£ concentra-
tions for 21 Control Alternatives comprising the existing smelter configuration
without SCS and 20 hypothetical smelter configurations with varying degrees of
constant emissions control. Table I gives, for each Control Alternative, the total
SO emission rate, the approximate percent constant SO emissions control as
2t £
well as a brief description of the emissions controls and source configuration.
From discussions with PSAPCA and EPA Region X, it appears that ASARCO has
actually been able to achieve only about 45 percent constant emissions control
for Control Alternative 1 (the existing smelter configuration without SCS) instead
of the 51 percent given in Table I because the operating efficiency of present
emissions control equipment is below design specifications. Tables n and III
respectively identify the Control Alternatives that will meet, without any SCS, the
National and the PSAPCA and Washington Department of Ecology SO2 air quality
standards. The Washington Department of Ecology (DOE) standards are essentially
the same as the PSAPCA standards. As shown in Table II, Control Alternatives 7,
8, 9, 10, 20 and 21 meet all the National SO2 air quality standards. Similarly, Table
III shows that Control Alternatives 7, 9, 20 and 21 also meet all PSAPCA and
IX
-------
TABLE I
DESCRIPTION OF THE ASARCO CONTROL ALTERNATIVES
Control
Alternative
I
2
3
4
5
6
7
8
Total SO2
Emission
Rate(lb/hr)
24, 802
24,802
24,802
14,490
14,250
14,250
1,900
1,900
Percent
Constant
Emissions
Control
51
51
51
71
72
72
96
96
Description
Existing source configuration without SCS.
Existing emissions controls with separate
stacks for the existing acid plant and
liquid SO2 plant.
Existing emissions controls using new
main stack and modifications outlined in
ASARCO variance application.
Existing emissions controls on reverbs
and converters, new acid plant on roasters
after enrichment by SOg injection, existing
main stack and modifications outlined in
ASARCO variance application.
Existing emissions controls on reverbs
and converters , scrubber on roasters,
existing main stack and modifications
outlined in ASARCO variance application.
Same as Control Alternative 5 with a new
main stack.
Electric arc furnace with acid plant on
roasters and part of converter stream after
enrichment by SO2 injection, liquid SO2
plant on remainder of converter stream,
existing main stack and modifications out-
lined in ASARCO variance applications
Same as Control Alternative 7 with a new
main stack.
-------
TABLE I (continued)
DESCRIPTION OF THE ASARCO CONTROL ALTERNATIVES
Control
Alternative
9
10
11
12
13
14
15
16
Total SO2
Emission
Rate(lb/hr)
2,240
2,240
24,750
24,750
22,840
22,840
19,750
19,750
Percent
Constant
Emissions
Control
96
96
51
51
55
55
61
61
Description .
Existing configuration with scrubber on
roaster and reverb streams, existing
main stack and modifications outlined in
ASARCO variance application.
Same as Control Alternative 9 with a
new main stack.
Existing emissions controls with modifi-
cations outlined in ASARCO variance
application ,
Same as Control Alternative 11 with a
new main stack.
Existing emissions controls with 20 per-
cent of roaster stream through existing
acid plant after enrichment by SOn
injection, existing main stack and
modifications outlined in ASARCO
variance application.
Same as Control Alternative 13 with a
new main stack.
Liquid SO2 plant and existing acid plant
control converter stream, new acid plant
controls 50 percent of roaster stream
after enrichment by SO2 injection, exis-
ting main stack and modifications outlined
in ASARCO variance application
Same as Control Alternative 15 with a
new main stack.
XI
-------
TABLE I (continued)
DESCRIPTION OF THE ASARCO CONTROL ALTERNATIVES
Control
Alternative
Total SO2
Emission
Rate(lb/hr)
Percent
Constant
Emissions
Control
Description
17
14,160
72
Liquid SO2 plant on part of converter
stream, new acid plant on remainder of
converter stream and 100 percent of
roaster stream after enrichment by SOg
injection, existing main stack and modifi-
cations outlined in ASARCO variance
application.
18
12,730
75
Existing emissions controls on converter
stream, new acid plant on reverb stream
after enrichment by SO2 injection, exis-
ting main stack.
19
12,400
76
Liquid SOg plant on part of converter
stream, new acid plant on remainder of
converter stream and reverb stream
after enrichment by SOg injection, exis-
ting main stack.
20
2,470
95
Existing emissions controls on converter
stream, roaster and reverb stream to
new acid plant following enrichment by
SO2 injection, existing main stack.
21
2,230
96
Existing emissions controls on converter
stream, electric arc furnace stream
combined with roaster stream to a new
acid plant following enrichment by SOg
injection, existing main stack.
xn
-------
TABLE H
CONTROL ALTERNATIVES THAT WILL MEET THE
NATIONAL AMBIENT AIR QUALITY
STANDARDS FOR SO^
Control
Alternative
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
National Air Quality Standards for SO2
Annual Standard
(0. 03 ppm)
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
3 -Hour Standard
(0. 50 ppm)
X
X
X
X
X
X
X
X
X
X
X
24 -Hour Standard
(0. 14 ppm)
X
X
X
X
X
X
xiii
-------
TABLE III
CONTROL ALTERNATIVES THAT WILL MEET THE PSAPCA
AMBIENT AIR QUALITY STANDARDS FOR SO0
Control
Alternative
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
PSAPCA Air Quality Standards for SO2
Annual
Standard
(0. 02 ppm)
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Monthly
Standard
(0. 04 ppm)
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
24-Hour
Standard
(0. 10 ppm)
X
X
X
X
X
X
1-Hour
Standard
(0. 40 ppm)
X
X
X
X
X
X
1-Hour
Standard
0. 25 ppm)
X
X*
X
X*
X
X
5 -Minute
Standard
(1. 0 ppm)
X
X
X
X
X
X
* Marginal compliance, infrequent violations possible.
xiv
-------
Washington DOE standards. Control Alternatives 8 and 10, as indicated in Table III,
are marginal with respect to compliance with the PSAPCA 1-hour standard of
0.25 parts per million. Although the model calculations clearly indicate that any
violoations of this standard would be infrequent, we were unable to eliminate the
possibility that the standard might be exceeded at any point more than twice in any
7-day period. As shown in Table I, the Control Alternatives that meet all National,
PSAPCA and Washington DOE standards all have SO emission rates ranging
^
from about 2000 to 2500 pounds per hour, depending on the source configuration.
These SO emission rates correspond approximately to 96 and 95 percent constant
£i
emissions control based on the emission rate of 24,802 pounds of SO per hour
£t
(51-percent control) for Control Alternative 1.
The ASARCO SCS curtails SO emissions and/or uses stack heaters to
&
increase plume rise whenever high ground-level SO concentrations that threaten
^
the short-term standards are either anticipated or observed to occur at the ASARCO
and/or PSAPCA monitors. According to our calculations, the maximum decrease
in ground-level SO concentrations that can normally be achieved through the use
&
of the stack heaters alone varies from about 14 to 19 percent, depending on
whether one or both stack heaters are used. A severe constraint in the use of the
ASARCO SCS to achieve the stringent 5-minute and 1-hour PSAPCA SO standards
£t
arises from an inherent minimum time delay, which varies from about 20 minutes
to 1 hour or longer, between the time a curtailment decision is made and the time
the decision is implemented and the reduction in emissions can affect ambient
air quality at SO monitors in the Tacoma area. It follows that it is generally not
possible for the ASARCO SCS to prevent violations of the PSAPCA 5-minute and
1-hour standards on the basis of telemetered air quality observations from the
ASARCO and PSAPCA monitoring networks. Instead, occurrences of high
5-minute and 1-hour concentrations at monitor locations must be anticipated and
curtailment decisions must be made by the ASARCO SCS at least 20 minutes to 1 hour
xv
-------
in advance of such occurrences. As discussed below, although the ASARCO
meteorologists have become highly skilled in anticipating the occurrence of high
SO concentrations, many violations of the PSAPCA 5-minute and hourly standards
&
continue to occur. The time-delay constraint in the operation of the ASARCO SCS
is generally not limiting in the case of the 3-hour and 24-hour National standards
and thus telementered air quality monitor observations can normally be used
effectively as a basis for curtailments required to meet these standards.
The effectiveness of the ASARCO SCS in preventing violations of the
National and PSAPCA SO short-term standards is shown in Tables IV and V which
Ll
list the total number of violations per year observed at the PSAPCA monitor
j_i^
located at N26 and Pearl. This monitor has been in operation continuously since
1968 and is the monitor most frequently affected by ASARCO emissions. As
shown in Table IV, no violations of the National standards were observed at the
monitor after 1971 (the SCS was started in 1970) until the first part of 1976 when
one violation of the 3-hour standard occurred. Also, although there were no
violations of the 24-hour National standard, there was one observation above the
standard and one observation equal to standard during the first part of 1976. The
success of the ASARCO SCS in preventing almost all violations of the National
short-term standards is in part due to the curtailment actions and stack-heater
operations undertaken in attempting to meet the more stringent PSAPCA short-term
standards.
Table V shows that the number of violations of the PSAPCA short-term
standards decreased dramatically from 1970 through 1973. Since 1973, however,
no appreciable decrease in the number of violations has occurred in spite of the
addition in 1974 of a liquid SO plant which increased the constant emissions
Lt
control from 17 to 51 percent. We believe that Table V shows that the ASARCO
SCS reached a limit of effectiveness during 1973 and 1974. We also believe that,
xvi
-------
TABLE IV
TOTAL NUMBER OF VIOLATIONS OF THE NATIONAL AIR QUALITY
STANDARDS FOR SO AT N26th AND PEARL
£1
Year
1976**
1975
1974
1973
1972
1971
1970
1969
1968
Number of Violations*
3 -Hour Secondary Standard
1 (2)
0 (0)
0 (1)
0 (0)
0 (0)
1 (2)
5 (6)
5 (6)
0 (1)
24-Hour Primary Standard
0 (1)***
0 (0)
0 (0)
0 (0)
0 (0)
1 (2)
1 (2)
4 (5)
1 (2)
* Total number of observations above the standard are enclosed by parentheses
** As of 6 February 1976
*** One observed concentration exactly equal to the 24-hour standard also occurred
during 1976
xvii
-------
TABLE V
TOTAL NUMBER OF VIOLATIONS OF THE PSAPCA AIR
QUALITY STANDARDS FOR SO AT N26th AND PEARL
Lt
Year
1976*
1975
1974
1973
1972
1971
1970
1969
1968
Number of Violations
5 -Minute
Standard of
1. 0 ppm
10
8
10
16
27
39
169
199
124
1-Hour
Standard of
0. 10 ppm
10
6
6
4
11
20
62
84
56
1-Hour
Standard of
0. 25 ppm
15
35
21
27
34
52
140
219
125
24-Hour
Standard of
0. 10 ppm
2
0
1
0
0
3
4
15
10
* As of 6 February 1976
xviii
-------
because of the severe delay-time constraint and the consequent dependence of
curtailment decisions on meteorological forecasts of plume behavior, it is
extremely unlikely that any further significant reductions in the number of
violations of the PSAPCA 5-minute and 1-hour standards can be achieved by SCS
techniques with the present source configuration and 51-percent constant emissions
control. In our opinion, with the present source configuration, violations of the
PSAPCA 5-minute and 1-hour standards can only be prevented by reducing the
maximum allowable stack emissions to about 2500 pounds of SO per hour. This
Lt
is approximately equal to the constant hourly emission rate for the Control
Alternatives in Tables n and in that meet all the National and PSAPCA SO
standards.
In summary, the principal conclusions of this study are:
• Without the use of SCS, the present smelter source
configuration (51-percent constant emissions control)
would result in violations of all National, PSAPCA
and Washington DOE short-term (5-minute, 1-hour,
3-hour and 24-hour) SO air quality standards
^j
• Depending on the source configuration, the maximum
allowable SO emission rate consistent with maintaining
£
all PSAPCA and Washington DOE air quality standards for
SO ranges from 2000 to 2500 pounds per hour; these
^
same emission rates will also ensure compliance with
all National air quality standards for SO
&
• The ASARCO SCS is capable of preventing most
violations of the National 3-hour and 24-hour air
quality standards, but SCS is not a viable means
of preventing violations of the more stringent
PSAPCA 5-minute and 1-hour standards
xix
-------
Recommendations made in this report include:
• Investigation of the impact of low-level fugitive SO
£j
emissions and possible stack downwash, accomplished
in part by the installation and operation of additional
SO monitors in the Tacoma area, located closer to
£t
the smelter than the existing monitors,
• Inclusion by ASARCO in the smelter operation reports
submitted to PSAPCA of average hourly values of the
stack exit temperature, actual volumetric flow rate
and SO emission rate in pounds per hour
&
• Change of the format of the ASARCO curtailment
reports to PSAPCA so that curtailment requests
are expressed as a maximum allowable SO emission
£
rate in pounds per hour rather than as a percent
curtailment
• Investigation of the impact of SO emissions from the
^
Kaiser aluminum plant in Tacoma which, according to
our diffusion-model calculations, may be threatening
the National and PSAPCA SO air quality standards
Li
XX
-------
TABLE OF CONTENTS
Section Title Page No.
ACKNOWLEDGMENT v
EXECUTIVE SUMMARY vii
1 INTRODUCTION 11
1.1 Background 1
1.2 Purpose and Approach 4
1.3 Report Organization 5
2 SO2 EMISSIONS FOR THE ASARCO-TACOMA SMELTER 9
2.1 Stack Parameters and SO2 Emission Rates 10
for 17- and 51-Percent Control
2.2 Stack Parameters and SO2 Emissions Data 14
for Possible Future Plant Configurations
(Control Alternatives)
3 METEOROLOGICAL DATA 29
3.1 Meteorology of the Seattle-Tacoma Area 29
3.2 CriticalMeteorologicalRegim.es 39
3.3 General Meteorological Inputs 42
4 HISTORICAL CALCULATIONS 49
4.1 Selection of Diffusion Models 49
4.2 1972 Annual Average Concentration Calculations 52
4.3 Selected Short-Term Calculations 60
4.3.1 Calculation Procedures 60
4.3.2 Results of the Short-Term 67
Historical Calculations
xxi
-------
TABLE OF CONTENTS (continued)
Section Title
5 EVALUATION OF THE CONTROL ALTERNATIVES
5.1 Maximum Annual Average Concentrations
5.2 Maximum Monthly Average Concentrations
5. 3 Maximum 3-Hour and 24-Hour Concentrations
5.4 Maximum Hourly and 5-Minute Concentrations
Page No.
73
73
75
78
84
EVALUATION OF THE EFFECTIVENESS OF THE
ASARCO SCS
6.1 Background: The SCS Concept
6.2 Description of the ASARCO SCS
6. 3 Critique of the ASARCO SCS
6. 4 Effectiveness of the ASARCO SCS
91
91
92
97
102
CONCLUSIONS AND RECOMMENDATIONS
7.1 Conclusions
7.2 Recommendations
107
107
109
REFERENCES
111
Appendix
A
MATHEMATICAL MODELS USED TO CALCULATE A-l
GROUND-LEVEL CONCENTRATIONS
A. 1 Introduction A-l
A. 2 Plume Rise Formulas A-5
A. 3 Short-Term Concentration Model A-7
A. 4 Long-Term Concentration Model A-14
A. 5 Application of the Short-Term and Long-Term A-19
Concentration Models in Complex Terrain
A. 6 Procedures Used to Calculate Occurrence A-24
Frequencies of Hourly Concentrations
xxii
-------
TABLE OF CONTENTS (continued)
Appendix Title Page No.
B ANNUAL FREQUENCY DISTRIBUTION OF WIND SPEED B-l
AND WIND DIRECTION AT N26*h AND PEARL
EMISSIONS DATA FOR THE OTHER MAJOR SO2 C-l
SOURCES IN THE SEATTLE-TACOMA AREA
D EXIT-VELOCITY RESTRICTIONS ON PLUME RISE D-l
AND THE ASARCO PLUME
E DISCUSSION OF THE SHORT-TERM HISTORICAL CASES E-l
E. 1 Introduction E-l
E.2 Discussion of the Individual Cases E-4
xxiii
-------
-------
SECTION 1
INTRODUCTION
1.1 BACKGROUND
The American Smelting and Refining Company (ASARCO) copper smelter
in Tacoma, Washington is one of the largest copper smelters in the United States,
currently accounting for about eight percent of the nation's copper. The smelter,
which was acquired by ASARCO in 1905, began as a lead smelter in 1890 with
copper smelting added in 1902. In 1911, lead smelting was discontinued in favor
of copper smelting. Copper has been the principal product of the Tacoma plant
since 1911.
Copper smelting at the Tacoma plant has always resulted in substantial
emissions to the atmosphere of sulfur dioxide (SOo). During the first 60 years of
operation, a tall stack was the principal method used to reduce the impact on ambient
air quality of these emissions. In 1950 an acid plant was installed which recovered
17 percent of the total SOg generated by smelting the copper ore concentrate. The
addition of a liquid SO2 plant in 1974 increased the percentage of SO2 recovery to a
maximum of 51 percent. According to EPA Region X and PSAPCA personnel, the
smelter recovers on the average only about 45 percent of the generated SO-
because the emissions control equipment operates below design efficiency.
Figure 1-1 is a topographic map of the Tacoma area. The filled circle
near the tip of the Tacoma peninsula shows the location of the ASARCO smelter's
main stacko As shown by the figure, terrain elevations in the area surrounding
the ASARCO smelter range from sea level to about 150 meters (500 feet) above sea
level. Water surrounds the smelter on three sides, and the land surface varies
from relatively smooth tidal flats and beaches to rugged hills. The meteorology
and complex topography of the Tacoma area frequently combine to produce poor
-------
FIGURE 1-1.
Topographic map of the Tacoma area. The filled circle shows
the location of the main ASARCO stack. Terrain elevations are
in feet above mean sea level, and the contour interval is 100 feet
(30.5 meters).
2
-------
atmospheric dilution conditions (i. e., conditions that are conducive to high ground
level SO concentrations as a result of the smelter1 s emissions).
£
For many years, the ASARCO Tacoma smelter voluntarily curtailed plant
operations when meteorological conditions were conducive to ground-level SO
&
concentrations that endangered vegetation or caused public complaints. Prior to
1969, curtailment decisions at the Tacoma smelter were made by the plant manager.
Thus, this form of intermittent curtailment came to be known as "Sea-Captaining. "
In late 1969, the Tacoma smelter employed a professional meteorologist to develop
and operate an Intermittent Control System (ICS), now called a Supplementary
Control System (SCS), in order to improve the ambient air quality in the vicinity
of the smelter.
Serious attempts to apply emissions and air quality standards to the
Tacoma smelter began with the formation of the Puget Sound Air Pollution Control
Agency (PSAPCA) on 1 July 1967. On 13 March 1968, PSAPCA adopted Regula-
tion I which included emissions and air quality standards for SO . However, civil
£
penalties for violations of Regulation I were not assessed until July 1969. On
29 December 1969, ASARCO applied for a variance from the ambient air quality
standards for SO . This variance was subsequently denied by PSAPCA, but ASARCO
£i
petitioned for judicial review. On 8 July 1970, PSAPCA revised Regulation I to
include the more stringent Washington Department of Ecology (DOE) air quality
standards and to limit sulfur emissions to 10 percent of the total sulfur generated
during the smelting process. After public hearings, the PSAPCA Board granted
ASARCO a three-year variance from the new emissions standard on 13 January 1971.
This variance was appealed by ASARCO to the Pollution Control Hearing Board
(PCHB). The PCHB remanded the problem to the PSAPCA Board which, after
a public hearing, granted a variance that would terminate on 31 January 1976 if
ASARCO did not commit to control 90 percent of the input sulfur. On 9 December 1974,
-------
ASARCO notified PSAPCA that the smelter did not intend to comply with the
PSAPCA 10-percent SO2 emissions standard. In February 1976 the PSAPCA
Board granted the smelter a new five year variance from the 10-percent SC>2
emissions standard. As part of the agreement, ASARCO promised to commit
nearly $4. 7 million toward additional emissions controls for pollutants other
than SC^.
The U. S. Environmental Protection Agency (EPA) has also been
concerned about the air quality impact of the Tacoma smelter's emissions,
including SO , particulates and arsenic. In June 1975, EPA contracted with the
£t
H. E. Cramer Company, Inc. of Salt Lake City, Utah to study the impact on
ambient air quality of SO emissions from the smelter and to evaluate the effective-
£t
ness of the smelter' s Supplementary Control System in preventing violations of
SO air quality standards.
£l
1. 2 PURPOSE AND APPROACH
The primary purpose of the study described in this report was to use
diffusion modeling to determine the degree of constant SO emissions control
&
required for the ASARCO-Tacoma smelter to assure attainment and maintenance
of the National, Washington DOE and PSAPCA air quality standards listed in
Table 1-1. An additional purpose was to evaluate the reliability of the ASARCO
Supplementary Control System (SCS).
The study consisted of four phases. During Phase I, a background study
was conducted to review meteorological and air quality data, ASARCO SCS
procedures, previous efforts to model smelter SO emissions, and emissions
^
data for ASARCO and the other major SO sources in the Seattle-Tacoma area.
LA
During Phase II, one long-term and five short-term diffusion models were tested
to determine the models most applicable to the unique topographical and
-------
meteorological conditions in the vicinity of the ASARCO smelter. Phase III
consisted of the application of the models selected during Phase H to determine
the degree of constant emissions control required to meet the various air quality
standards listed in Table 1-1. Finally, the reliability of the ASARCO SCS
program was assessed during Phase IV in order to determine whether the
smelter' s SCS is a viable method of maintaining the applicable air quality standards.
1.3 REPORT ORGANIZATION
In addition to the Introduction, this report consists of six major sections
and four appendices. Section 2 contains descriptions of the past and present stack
and emissions data for the ASARCO smelter as well as stack and emissions data
for possible future smelter configurations. Section 3 contains a discussion of the
meteorology of the Seattle-Tacoma area, descriptions of the meteorological regimes
associated with the highest ground-level SO2 concentrations and listings of the
meteorological input parameters used in the diffusion-model calculations. Section 4
presents the results of calculations of annual average SO2 concentrations in Tacoma
for the year 1972 and the results of calculations for 20 Short-Term Historical Cases
when high SO2 concentrations were observed for one or more hours at air quality
monitors; comparison of calculated and observed concentrations is also included in
Section 4. In Section 5, the modeling techniques used in Section 4 are employed to
assess the impact on ambient air quality of both current smelter SO2 emissions
(without the use of SCS) and smelter emissions for possible future plant configurations.
An evaluation of the effectiveness of the smelter's SCS is presented in Section 6. The
conclusions and recommendations of the study are given in Section 7.
The diffusion models used in this study are described in detail in Appendix
A. Appendix B provides tabulations of the annual frequency of occurrence of wind-
speed and wind-direction categories in Tacoma, classified according to the Pasquill
stability categories. Emissions data for the major SO2 sources in the TacQma area,
-------
TABLE 1-1
SO AIR QUALITY STANDARDS APPLICABLE
TO THE ASARCO SMELTER
Time Period
5 minutes
Ihour
1 hour
3 hours
24 hours
30 days
Annual
National Staj
Primary
—
—
—
—
0.14a
—
b
0.03
tidards (ppm)
Secondary
—
—
—
a
0.50
—
Washington DOE
Standards (ppm)
a
0.40
c
0.25
o.ioa
b
0.02
PSAPCA
Standards (ppm)
d
1.00
b
0.40
c
0.25
o.iob
b
0.04
b
0.02
a—Not to be exceeded more than once per year
b—Never to be exceeded
c—Not to be exceeded more than two times in any seven consecutive days
d—Not to be exceeded more than once in any eight consecutive hours
-------
other than the ASARCO smelter, are given in Appendix C. Appendix D contains a
discussion of exit-velocity restrictions on buoyant plume rise for stacks in general
and for the main ASARCO stack in particular. Appendix E contains detailed
discussions of the 20 Short-Term Historical Cases.
-------
-------
SECTION 2
SO2 EMISSIONS DATA FOR THE ASARCO TACOMA SMELTER
The basic SO2 emissions data used in this study are for normal smelter
operations at full production with the following percent constant emissions control:
• 17 percent—applies to smelter operations prior
to March 1974
• 51 percent—applies from March 1974 to present
• >51 percent—applies to possible future smelter
configurations used in Control Alternative Calculations
to determine the degree of constant emissions control
required for compliance with air quality standards
SO2 emissions data for 17-percent constant emissions control were supplied by
ASARCO and are estimates derived from material balance calculations. These
estimates were obtained by forming ratios of the total weight of sulfur contained
in the slag, copper, copper by-products, and sulfuric acid produced
by the smelter to the total weight of sulfur contained in the copper ore prior to
processing. For a sulfur balance, the difference between the total weight of sulfur
in the ore and the weight of the sulfur contained in the slag and smelter by-products
must equal the weight of the sulfur contained in the SO2 discharged to the atmos-
phere. Calculations of the amount of sulfur discharged to the atmosphere through
the 172-meter (565-foot) main stack based on average flow rates and in-stack SO2
concentration measurements agree, within a few percent, with the amount required
to achieve a sulfur balance. It follows that average SO2 low-level fugitive emis-
sions are quite small compared to the average SO2 emissions discharged from the
main stack. This conclusion is supported by comparisons of calculated annual
average SOo concentrations for 1972 with observed air quality at SO2 monitoring
-------
sites located in the Tacoma area. The best agreement between the calculated and
observed concentrations is obtained when the average annual low-level fugitive
emissions are effectively set to zero. During normal smelter operations, low-
level fugitive emissions occur intermittently during loading, unloading and transfer
of the concentrate as well as during the rolling of the converters. Also, for the
existing 51-percent control smelter configuration, other fugitive emissions occur
whenever the liquid SC>2 plant is down. Stations of the existing SC>2 monitoring
network are too far away from the smelter to detect the effects of low-level fugitive
emissions. It is possible, however, that fugitive emissions may account for the
high SC>2 concentrations measured in Tacoma by the EPA mobile 862 monitor (see
Appendix D) which was located approximately 1. 2 kilometers from the main stack.
2.1 STACK PARAMETERS AND SO2 EMISSION RATES
FOR 17- and 51-PERCENT CONTROL
Basic stack parameter values and SOo emissions data for smelter operations
at full production are listed in Table 2-1 for both 17- and 51-percent constant emis-
sions control. As mentioned above, the stack and emissions data for 17-percent
control were obtained from ASARCO and the latter are principally based on sulfur
balance calculations. Although the emissions data are the best estimates available
and appear to be accurate long-term averages, it is not possible to fix the magnitude
of short-term (hourly, daily and weekly) variations that may have occurred. In
addition to the SO2 emissions from the main stack, small amounts of SO2 were
emitted from the acid plant and anode furnace stacks for the 17-percent control
configuration. These emissions were on the order of 0.1 percent of the maximum
uncurtailed emission rate for the main stack and thus had a minimal impact at the
distances of the SO2 monitors. Consequently, in our short-term and long-term
historical calculations for the ASARCO smelter, we did not include the acid plant
stack (17-percent control) or the anode furnace stack (17- and 51-percent control).
10
-------
TABLE 2 -1
ASARCO MAIN STACK PARAMETERS AND SO2 EMISSIONS DATA
FOR 17- AND 51-PERCENT CONSTANT EMISSIONS
CONTROL AT FULL PRODUCTION
Parameter Values
Constant Emissions Control (Percent)
17 Percent
51 Percent
Stack Height (ft)
Elevation of Stack
Base Above Mean
Sea Level (ft)
Inner Stack Diameter (ft)
UTM Coordinates (m)
X
Y
Volumetric Emission Rate
(acfm)
No Stack Heaters
One Stack Heater
Two Stack Heaters
Stack Exit Temperature (°F)
No Stack Heaters
One Stack Heater
Two Stack Heaters
Maximum Uncurtailed SOg
Emission Rate (Ib/hr)
565
150
24
537,370
5,238,080
524, 769
587,482
636,237
180
240
280
52,000
565
150
24
537,370
5,238,080
553,000
553,000
553,000
163
225
260
30,000
11
-------
In the spring of 1974, ASARCO added a liquid SC>2 plant that increased the
level of constant emissions control to 51 percent. SC>2 emissions from the pre-
existing acid plant as well as the new liquid SC>2 plant were ducted to the main
stack, thus eliminating the acid plant stack as a source. To assist in establishing
representative values for the main stack parameters and SC>2 emission rates for
the 51-percent control configuration, ASARCO provided detailed records of plant
operations, in-stack SOr, measurements, stack temperatures and flow rates for
a 30-day period in the fall of 1975. These data were used to develop the parameters
values and SC>2 emission rate for 51-percent control shown in Table 2-1. Addi-
tionally, a detailed analysis was made of the hourly data for twelve of the 30 days
during which the smelter operations covered a broad range of production levels.
This analysis showed that the actual volumetric emission rate from the main stack
was unaffected in any systematic way by the number of stack heaters in operation
and was therefore effectively constant. A second result of this detailed analysis of
hourly data is presented in Figure 2-1 which shows a scatter diagram and a fitted
linear least-squares regression curve relating measured in-stack SOr, concentrations
to SO2 emission rates. The upper and lower curves in the figure are calculated
95-percent confidence limits. The relationship between the SO2 emission rate Q
in tons per hour and the measured in-stack SO2 concentration p in parts per million
(ppm) given by the least-squares regression curve is
Q (tons/hr) = 0. 0028 p (ppm) - 0. 1165 (2-1)
The calculated linear correlation coefficient is 0. 922 and the number of paired data
points used in the calculation is 283. According to the confidence limits shown in
Figure 2-1, there is a 95-percent probability that the actual SO2 emission rate is
within ± 20 percent of the rate calculated from Equation (2-1) using the measured
in-stack SO0 concentration. This equation was used to estimate the hourly SO2
£i
emission rates for three historical cases (18, 19, and 20) thatoccurred after the
installation of the liquid SO2 plant (see Section 4.3).
12
-------
cs>
c
o
LJ
g
CO
CO
2
LJ
CJ
O
CO
FIGURE 2-1.
1000
2000 3000 4000 5000
IN STACK S02 CONCENTRATION (ppm)
6000
7000
Scatter diagram of in-stack SC^ concentration measurements versus SO2 emission
rates. Line drawn through the data points is the least-squares regression line while
the upper and lower lines show the 97. 5 and 2. 5 percent confidence limits.
-------
2.2 STACK PARAMETERS AND SO2 EMISSIONS DATA FOR
POSSIBLE FUTURE PLANT CONFIGURATIONS (CONTROL
ALTERNATIVES)
According to EPA Region X, a study of the control of SO from the
4
ASARCO smelter indicates that two levels of constant emissions control beyond
the current level of 51-percent control are technically feasible. By controlling
either the roaster or reverberatory furnace streams, EPA Region X believes
the sulfur capture can be increased to about 70 percent of the input sulfur. Sulfur
removal on the order of 90 percent is possible if both streams are controlled. EPA
Region X requested that we evaluate the air quality impact of 21 possible future
plant configurations of the ASARCO smelter. These configurations, which are
termed "Control Alternatives", start with the existing smelter configuration and
cover a broad spectrum of methods available for the control of SO2 emissions from
the smelter. At one end of the spectrum, it is assumed that part of the roaster
stream is controlled through an upgraded acid plant. At the other end of the
spectrum, it is assumed that the smelter converts to electric smelting and uses a
new, larger acid plant. It is assumed that degrees of control intermediate between
these two extremes are achieved by upgrading SO£ concentrations in various
streams by means of SOg injection and by using the existing acid plant and/or a
new acid plant for the final control devices.
Tables 2-2 through 2-13 list the stack and emissions data for the 21
Control Alternatives. The percent constant SO2 emissions control as well as the
types of emissions controls and details of the source configuration are listed at
the top of each table. Source numbers given in the tables reference the source
identification code used in the computer printout sheets containing the detailed
diffusion-model calculations which are on file at EPA Region X and at PSAPCA in
Seattle, Washington. Table 2-14 gives the assumed locations, elevations and
source numbers of the hypothetical new stacks. The hypothetical new main stack
is situated 70 meters (230 feet) west of the existing main stack. The other
14
-------
hypothetical stacks are located within the plant production area.
In our calculations for the existing plant configuration (Control Alternative
1), we used the average volumetric and uncurtailed SO£ emission rates for 51-
percent control given in Table 2-2; the SC>2 emission rate given in Table 2-1 is the
maximum uncurtailed rate. Also, we used the stack exit temperature for no stack
heaters in operation because the stack heaters are considered to be a part of the
smelter's Supplementary Control System (SCS). Because the Froude number for
the existing ASARCO stack is less than 3, the minimum value for which exit-
velocity (downwash) restrictions on plume rise are believed to apply, no exit-
velocity restrictions on plume rise were assumed in the calculations for this
stack (see Appendix D).
It is important to note that the ASARCO smelter has apparently been unable
to achieve the 51-percent constant emissions control called for in Control Alternative
1 (the existing plant configuration). According to EPA Region X and PSAPCA, the
average SO recovery is less than 45 percent because the existing emissions
control equipment often operates below design efficiency.
15
-------
TABLE 2-2
STACK AND EMISSIONS DATA FOR CONTROL ALTERNATIVE 1
51 Percent Control Using Existing Emissions Controls and
Existing Source Configuration without SCS.
Existing Main Stack (Source #1001)
Height (ft)
Exit Diameter (ft)
Average Flow Rate (acfm)
Minimum Flow Rate (acfm)
Maximum Flow Rate (acfm)
Exit Temperature (°F)
No Heaters
One Heater
Two Heaters
SO Emission Rate (Ib/hr)
LA
565
24
549,000
415,400
692,900
163
225
260
24,802
16
-------
TABLE 2-3
STACK AND EMISSIONS DATA FOR CONTROL ALTERNATIVE 2
51 Percent Control Using Existing Emissions Controls, Existing Main Stack for
Roasters and Reverbs with New Liquid SO2 Plant Stack and New Acid Plant Stack.
Acid Plant Stack (Source #1002)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO Emission Rate (Ib/hr)
2i
Exit Temperature (°F)
Liquid SQ2 Plant Stack (Source #1003)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO2 Emission Rate (Ib/hr)
Exit Temperature (°F)
Existing Main Stack (Source #1004)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO Emission Rate (Ib/hr)
^
Exit Temperature (°F)
100
2.5
26,508
483
170
100
3.0
51, 945
1,250
170
565
24.0
452,118
23,070
150
17
-------
TABLE 2-4
STACK AND EMISSIONS DATA FOR CONTROL ALTERNATIVES 3 AND 4
Control Alternative 3--51 Percent Control Using New Main Stack, Existing
Emissions Controls and Modifications Outlined in ASARCO Variance Application.
New Main Stack (Source #1005)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO Emission Rate (Ib/hr)
^
Exit Temperature (°F)
500
12
516,546
24,802
170
Control Alternative 4—71 Percent Control Using Existing Main Stack, Existing
Emissions Controls on Reverbs and Converters, New Acid Plant on Roasters after
Enrichment by SO£ Injection and Using Modifications Outlined in ASARCO
Variance Application.
Existing Main Stack (Source #1006)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO Emission Rate (Ib/hr)
Li
Exit Temperature (°F)
565
24
400,288
14,490
170
18
-------
TABLE 2-5
STACK AND EMISSIONS DATA FOR CONTROL ALTERNATIVES 5 AND 6
Control Alternative 5—72 Percent Control Using Existing Main Stack, Existing
Emissions Controls on Re verbs and Converters, Scrubber on Roasters and
Using Modifications Outlined in ASARCO Variance Application.
Existing Main Stack (Source #1007)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO2 Emission Rate (Ib/hr)
Exit Temperature (°F)
565
24
388,443
14,250
150
Control Alternative 6—72 Percent Control Using New Main Stack,
Existing Emissions Controls on Reverbs and Converters, Scrubber on Roasters
and Using Modifications Outlined in ASARCO Variance Application.
New Main Stack (Source #1008)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO2 Emission Rate (Ib/hr)
Exit Temperature (°F)
500
12
388,443
14, 250
150
19
-------
TABLE 2-6
STACK AND EMISSIONS DATA FOR CONTROL ALTERNATIVES 7 AND 8
Control Alternative 7—96 Percent Control Using Existing Main Stack, Electric
Arc Furnace, Acid Plant on Roasters and Part of Converter Stream after
Enrichment by SO2 Injection, Liquid SO2 Plant on Remainder of Converter
Stream and Modifications Outlined in ASARCO Variance Application.
Existing Main Stack (Source #1009)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO2 Emission Rate (Ib/hr)
Exit Temperature (°F)
565
24
344,592
1,900
150
Control Alternative 8—96 Percent Control Using New Main Stack, Electric
Arc Furnace, Acid Plant on Roasters and Part of Converter Stream after
Enrichment by SOo Injection, Liquid SO£ Plant on Remainder of Converter
Stream and Modifications Outlined in ASARCO Variance Application.
New Main Stack (Source #1010)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO2 Emission Rate (Ib/hr)
Exit Temperature (°F)
500
8
344,592
1,900
150
20
-------
TABLE 2-7
STACK AND EMISSIONS DATA FOR CONTROL ALTERNATIVES 9 AND 10
Control Alternative 9—96 Percent Control Using Existing Main Stack, Existing
Source Configuration, Scrubber on Roaster and Reverb Streams and with
Modifications Outlined in ASARCO Variance Application.
Existing Main Stack (Source #1011)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO Emission Rate (Ib/hr)
^
Exit Temperature ( F)
565
24
385, 842
2,240
150
Control Alternative 10—96 Percent Control Using New Main Stack with Existing
Source Configuration, Scrubber on Roaster and Reverb Streams and with
Modifications Outlined in ASARCO Variance Application.
New Main Stack (Source #1012)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO Emission Rate (Ib/hr)
^
Exit Temperature (°F)
500
10.5
385, 842
2,240
150
21
-------
TABLE 2-8
STACK AND EMISSIONS DATA FOR CONTROL ALTERNATIVES 11 AND 12
Control Alternative 11—51 Percent Control Using Existing Main Stack, Existing
Source Configuration and Existing Emissions Controls with Modifications Out-
lined in ASARCO Variance Application.
Existing Main Stack (Source #1013)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO Emission Rate (Ib/hr)
2t
Exit Temperature (°F)
565
24
402,962
24,750
170
Control Alternative 12—51 Percent Control Using New Main Stack, Existing
Source Configuration and Existing Emissions Controls with Modifications Out-
lined in ASARCO Variance Application.
New Main Stack (Source #1014)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO Emission Rate (Ib/hr)
^
Exit Temperature (°F)
500
10
402,962
24,750
170
22
-------
TABLE 2 -9
STACK AND EMISSIONS DATA FOR CONTROL ALTERNATIVES 13 AND 14
Control Alternative 13—55 Percent Control Using Existing Main Stack, Existing
Emissions Controls with 20 Percent of Roaster Stream through Existing Acid
Plant after Enrichment by SO£ Injection and with Modifications Outlined in
ASARCO Variance Application.
Existing Main Stack (Source #1015)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO2 Emission Rate (Ib/hr)
Exit Temperature (°F)
565
24
401,607
22,840
170
Control Alternative 14—55 Percent Control Using New Main Stack, Existing
Emissions Controls with 20 Percent of Roaster Stream through Existing Acid
Plant after Enrichment by SO2 Injection and with Modifications Outlined in
ASARCO Variance Application.
New Main Stack (Source #1016)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO2 Emission Rate (Ib/hr)
Exit Temperature (°F)
500
10
401,607
22,840
170
23
-------
TABLE 2-10
STACK AND EMISSIONS DATA FOR CONTROL ALTERNATIVES 15 AND 16
Control Alternative 15—61 Percent Control Using Existing Main Stack with Liquid
SO2 Plant on Part of Converter Stream, New Acid Plant on 50 Percent of Roaster
Stream after Enrichment by SO2 Injection and with Modifications Outlined in
ASARCO Variance
Existing Main Stack (Source #1017)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO Emission Rate (Ib/hr)
Exit Temperature (°F)
565
24
400,846
19,750
170
Control Alternative 16—61 Percent Control Using New Main Stack with Liquid
SO2 Plant on Part of Converter Stream, New Acid Plant on 50 Percent of
Roaster Stream after Enrichment by SO2 Injection and with Modifications Out-
lined in ASARCO Variance Application.
New Main Stack (Source #1018)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO Emission Rate (Ib/hr)
£*
Exit Temperature (°F)
500
10
400,846
19,750
170
24
-------
TABLE 2-11
STACK AND EMISSIONS DATA FOR CONTROL ALTERNATIVES 17 AND 18
Control Alternative 17—72 Percent Control Using Existing Main Stack with
Liquid SO2 Plant on Part of Converter Stream, New Acid Plant oh Remainder of
Converter Stream and on 100 Percent of Roaster Stream after Enrichment by SO2
Injection and Using Modifications Outlined in ASARCO Variance Application.
Existing Main Stack (Source #1019)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO2 Emission Rate (Ib/hr)
Exit Temperature (°F)
565
24
399,503
14,160
170
Control Alternative 18—75 Percent Control Using Existing Main Stack with
Existing Emissions Controls on Converter Stream and New Acid Plant on
Reverb Stream after Enrichment by SO2 Injection.
New Main Stack (Source #1020)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO Emission Rate (Ib/hr)
Ll
Exit Temperature (°F)
565
24
400,205
12,730
170
25
-------
TABLE 2-12
STACK AND EMISSIONS DATA FOR CONTROL ALTERNATIVES 19 AND 20
Control Alternative 19—76 Percent Control Using Existing Main Stack with
Liquid SO£ Plant on Part of Converter Stream and New Acid Plant on Remainder
of Converter Stream and on Reverb Stream after Enrichment by SO2
Injection.
Existing Main Stack (Source f 1021)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO Emission Rate (Ib/hr)
£j
Exit Temperature (°F)
565
24
399,277
12,400
170
Control Alternative 20—95 Percent Control Using Existing Main Stack with
Existing Emissions Controls on Converter Stream and with New Acid Plant
on Roaster and Reverb Streams after Enrichment by SO2 Injection.
New Main Stack (Source #1022)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO2 Emission Rate (Ib/hr)
Exit Temperature (°F)
565
24
385,693
2,470
150
26
-------
TABLE 2-13
STACK AND EMISSIONS DATA FOR CONTROL ALTERNATIVE 21
Control Alternative 21—96 Percent Control Using Existing Main Stack, Existing
Emissions Controls on Converter Stream, Electric Arc Furnace with New
Acid Plant on Combined Furnace and Roaster Streams after Enrichment by
SO2 Injection.
Existing Main Stack (Source #1023)
Height (ft)
Exit Diameter (ft)
Flow Rate (acfm)
SO2 Emission Rate (Ib/hr)
Exit Temperature (°F)
565
24
345,283
2,230
150
TABLE 2-14
ASSUMED LOCATIONS AND ELEVATIONS OF
HYPOTHETICAL NEW ASARCO STACKS
Source Name
New Main Stack
Acid Plant Stack
SO2 Plant Stack
Source No.
1005,1008
1010,1012
1014,1016
and 1018
1002
1003
Location
UTM X (m)
537,300
537,400
537,500
UTM Y (m)
5,238,080
5,238,400
5,238,400
Elevation
(ft above MSL)
121
30
33
27
-------
28
-------
SECTION 3
METEOROLOGICAL DATA
3.1 METEOROLOGY OF THE SEATTLE-TACOMA AREA
The meteorology of the Seattle-Tacoma area reflects the complex topo-
graphy and the large number of land-water interfaces. Seattle and Tacoma are
situated on the eastern shore of Puget Sound, which has an approximate north-
south orientation. To the west of Puget Sound, the Olympic Mountain Range
effectively shields the entire area from west winds. Terrain elevations in the
immediate Seattle-Tacoma area range from sea level to 150 meters above sea
level. Additionally, the land surface varies from relatively smooth tidal flats
and beaches to rugged hills. These variations in terrain height, type and orienta-
tion significantly affect the wind flow. Variations in wind speed, wind direction
and atmospheric stability indicated by an analysis of meteorological measure-
ments at four locations in the Seattle-Tacoma area are described below.
Wind-Direction Distributions
Figure 3-1 shows the 1972 annual frequency distributions of wind direc-
tion at the N26 and Pearl station in Tacoma (solid line) and at McChord Air
Force Base (dashed line). The directions in Figure 3-1 have been reversed 180
degrees to show the directions toward which the wind is blowing and thus the effects
of wind direction on the areal distribution of pollutants. The N26 and Pearl
station is approximately 3 kilometers south of the stack of the ASARCO smelter,
and McChord AFB is about 18 kilometers south of the stack. The two wind-
direction distributions are similar and indicate a general north-south circulation
in the Tacoma area. As shown by Figure 3-1, the wind-direction distributions
at the two sites would be nearly identical if the McChord wind directions were
rotated 22. 5 degrees clockwise. We believe this difference is due to differences
in the orientation of Puget Sound.
29
-------
N26th 8 PEARL
NW
McCORD AFB
NNW
N
NNE
WNW
WSW
ENE
ESE
SSW
SSE
SE
FIGURE 3-1. 1972 annual frequency distributions of wind direction at N26*h
and Pearl (solid line) and Me Chord Air Force Base (dashed
line). The directions are directions toward which the wind
is blowing and the percentage frequency scale is shown in
the lower half of the figure.
30
-------
Figure 3-2 shows the 1972 annual frequency distributions of wind direction
at the Seattle-Tacoma Airport (solid line) and at Boeing Field (dashed line). The
directions in Figure 3-2 are also directions toward which the wind is blowing.
The Seattle-Tacoma Airport is located about 23 kilometers northeast of the ASARCO
stack while Boeing Field is located about 30 kilometers north-northeast of the
stack. The wind-direction distribution at the Seattle-Tacoma Airport is similar
to the distributions at Me Chord AFB and the N26 and Pearl station. The wind-
direction distribution for Boeing Field appears invalid because of the low frequency
of occurrence of north and south winds as compared to the high frequency of occur-
rence of north-northwest, north-northeast, south-southeast and south-southwest
winds. However, the possibility of errors in the Boeing Field wind-direction data
does not affect the results of this study.
Wind-Speed Distributions
Table 3-1 lists the 1972 annual frequency distributions of wind speed at
th
N26 and Pearl, McChord AFB, the Seattle-Tacoma Airport and Boeing Field.
j_U
As shown by the table, the wind-speed distributions at McChord AFB and N26
and Pearl are essentially the same; also, the wind-speed distributions at the
Seattle-Tacoma Airport and Boeing Field are nearly identical. However, the
wind-speed distributions for N26 and Pearl and McChord AFB differ significantly
from the wind-speed distributions for the Seattle-Tacoma Airport and Boeing Field.
Specifically, very light surface winds (wind speeds less than about 1. 5 meters per
second) occur about four times more frequently in the Tacoma area than in the
Seattle area.
31
-------
— SEATAC
— BOEING FIELD
NNW
NNE
WNW
ENE
WSW
ESE
FIGURE 3-2. 1972 annual frequency distributions of wind direction at the
Seattle-Tacoma Airport (solid line) and Boeing Field (dashed
line). The directions are the directions toward which the
wind is blowing and the percentage frequency scale is shown
in the lower half of the figure.
32
-------
TABLE 3-1
1972 ANNUAL WIND-SPEED DISTRIBUTIONS AT N26
AND PEARL, MC CHORD AIR FORCE BASE,
SEATTLE-TACOMA AIRPORT (SEATAG)
AND BOEING FIELD
th
Wind Speed
(m/sec)
0 -1.5
1.6- 3.0
3.1- 5.1
5.2- 8.2
8.3 - 10.8
>10. 8
Annual Percent Frequency of Occurrence
N26th and Pearl
35.63
29.94
21.05
11.64
1.54
0.20
McChord AFB
33.89
33.80
17.65
13.13
1.17
0.36
SEATAC
6.75
41.96
31.96
16.97
2.18
0.18
Boeing Field
9.99
40.68
31.98
16.42
0.85
0.08
33
-------
Stability Distributions
Turner (1964) has developed a procedure to determine the Pasquill stability
category on the basis of airport surface weather observations. The Turner defini-
tions of the Pasquill stability categories are based on solar radiation (insolation),
as estimated from cloud cover and time of day, and wind speed. Tables 3-2 and
3-3 list the parameters that define the various stability categories. The thermal
stratifications represented by the six stability categories are:
• A - Very unstable
• B - Unstable
• C - Slightly unstable
• D - Neutral
• E - Stable
• F - Very stable
Table 3-4 lists the 1972 annual frequency distributions of the Pasquill
stability categories at N26 and Pearl, Me Chord AFB, Seattle-Tacoma Airport and
Boeing Field. Because no cloud cover observations were available for N26 and
Pearl, cloud cover observations at McChord AFB were used in conjunction with
concurrent wind measurements at N26 and Pearl to determine the stability
categories at N26 and Pearl. The stability distributions at McChord AFB and
N26 and Pearl are essentially the same and the stability distributions at the
Seattle-Tacoma Airport and Boeing Field are very similar. However, unstable
and very stable conditions are more frequent at the Tacoma area sites (McChord
AFB and N26 and Pearl) than at the Seattle area sites (Seattle-Tacoma Airport
and Boeing Field). The higher frequency of occurrence of light winds in the
34
-------
TABLE 3-2
PASQUILL STABILITY CATEGORY AS A FUNCTION
OF INSOLATION AND WIND SPEED
Wind
Speed
(knots)
0,1
2,3
4,5
6
7
8,9
10
11
>12
Insolation Index
4
A
A
A
B
B
B
C
C
C
3
A
B
B
B
B
C
C
C
D
2
B
B
C
C
C
C
D
D
D
1
C
C
D
D
D
D
D
D
D
0
D
D
D
D
D
D
D
D
D
-1
F
F
E
E
D
D
D
D
D
-2
F
F
F
F
E
E
E
D
D
TABLE 3-3
INSOLATION CATEGORIES
Insolation
Insolation Category Number
Strong
Moderate
Slight
Weak
Overcast < 7000 feet (day or night)
Cloud Cover > 4/10 (night)
Cloud Cover < 4/10 (night)
4
3
2
1
0
-1
-2
35
-------
Tacoma area (see Table 3-1) is consistent with these differences in stabilities.
As indicated by Tables 3-2 and 3-3, a higher frequency of light winds will result
in a higher frequency of unstable and very stable conditions. It should be noted
that all four sites in the Seattle-Tacoma area show that the neutral D stability
category occurs more than half of the time.
The 1972 annual frequency distribution of wind-speed and wind-direction
categories at N26 and Pearl, classified according to the Pasquill stability cate-
gories, is presented in Appendix B. We believe that, of the available wind data,
the N26 and Pearl data are the most representative of the winds that affect the
transport and dispersion of emissions from the ASARCO smelter.
Wind Persistence
In order to assess the frequency of occurrence of maximum 3-hour and
24-hour ground-level concentrations, it is necessary to know the combined per-
sistence of wind speed and wind direction for time periods up to 24 hours. Table
3-5 lists the total number of occurrences at N26 and Pearl during the period
January through December 1972 of the persistence of wind directions and wind
speeds greater than 5.1 meters per second for time periods from 1 to 24 hours.
As shown by the table, the most persistent wind directions in the Tacoma area
are north-northeast, east-northeast, south-southwest and west-southwest. The
maximum persistence during 1972 of moderate to strong winds within a wind-
direction sector was 20 hours for north-northeast winds.
Mixing Depths
We have analyzed tabulations of the seasonal frequency of occurrence of
surface mixing depths measured at the Seattle-Tacoma Airport during the period
36
-------
TABLE 3-4
1972 ANNUAL DISTRIBUTIONS OF THE PASQUILL STABILITY
CATEGORIES AT N26th AND PEARL, MC CHORD
AFB, THE SEATTLE-TACOMA AIRPORT
AND BOEING FIELD
Pasquill Stability
Category
A
B
C
D
E
F
Annual Percent Frequency of Occurrence
N26th and Pearl
0.42
6.91
9.89
53.94
7.93
20.91
McChord AFB
0.80
9.81
10.96
55.66
5.67
17.10
SEA TAG
0.18
4.87
9.96
61.46
13.15
10.39
Boeing Field
0.10
4.34
9.81
60.46
11.54
13.74
37
-------
TABLE 3-5
TOTAL NUMBER OF OCCURRENCES DURING 1972 OF THE PERSISTENCE OF WIND DIRECTION
AND WIND SPEEDS GREATER THAN 5.1 METERS PER SECOND AT N26th AND PEARL
Number of
Hours of
Persistence
s 1
a 2
^ 3
a 4
a 5
2= 6
a 7
a 8
a 9
a 10
a 11
a 12
& 13
a 14
a 15
a 16
£ 17
£ 18
2 19
2; 20
2:21
= 22
a 23
^ 24
N
22
6
3
2
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
NNE
165
66
31
21
14
9
6
5
4
3
2
2
2
2
I
I
I
I
1
I
0
0
0
0
NE
74
25
11
7
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
ENE
16
8
4
3
2
3
3
1
1
1
1
1
1
1
0
0
0
0
0
0
0
0
0
0
E
7
2
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Wind Direction (Sector)
ESE
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
SE
2
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
SSE
4
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
s
77
27
11
7
3
3
2
2
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
ssw
301
109
53
31
19
11
7
4
3
2
1
1
1
1
0
0
0
0
0
0
0
0
0
0
sw
266
88
38
22
12
5
3
2
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
wsw
179
62
28
18
11
7
4
3
3
3
2
2
1
1
0
0
0
0
0
0
0
0
0
0
w
14
5
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
WNW
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
NW
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
NNW
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
-------
1959 through 1961 (Environmental Data Service, 1967) in order to determine
seasonal mixing depths for each time-of-day and wind-speed category. Table 3-6
lists the median early morning and afternoon mixing depths for each season.
Although the early morning mixing depths calculated by the Environmental Data
Service are assumed to apply at 0800 PST, we believe that they are representative
of nighttime conditions in the Seattle-Tacoma area because of the effects of heat
sources and surface roughness. Table 3-6 shows that the diurnal variation of the
mixing depth is at a minimum in the winter and at a maximum in the spring and
summer. For wind speeds less than 5.1 meters per second, mixing depths during
the winter are quite shallow throughout the day.
3. 2 CRITICAL METEOROLOGICAL REGIMES
Based on our review of meteorological, emissions and air quality data,
we believe that there are three meteorological regimes in the Tacoma area that
are conducive to high ground-level SO concentrations produced by stack emis-
&
sions from the ASARCO smelter. These critical regimes are:
• The critical wind-speed condition
• Fumigation
• The limited-mixing condition
The Critical Wind-Speed Condition
The critical wind-speed condition (neutral stability in combination with
moderate to strong winds) is the most persistent of the three meteorological regimes
and is thus the regime most likely to produce the highest ground-level SO concen-
^
trations for averaging times of 3 and 24 hours. Because high ground-level concen-
trations produced by the critical wind-speed condition are generally confined
39
-------
TABLE 3-6
MEDIAN MIXING DEPTHS MEASURED AT THE
SEATTLE-TACOMA AIRPORT DURING
THE PERIOD 1959
THROUGH 1961
Wind Speed
(m/sec)
0 - 1.5
1.6- 3.0
3.1- 5.1
5.2- 8.2
8.3 - 10.8
>10. 8
Median Mixing Depth (m)
Winter
Night
125
125
375
625
625
625
Afternoon
375
375
375
625
625
625
Spring
Night
125
375
675
875
1250
1250
Afternoon
1250
1250
1250
1250
1250
1250
Summer
Night
125
375
375
625
1250
1250
Afternoon
1250
1250
1250
1250
1250
1250
FaU
Night
125
125
375
875
875
875
Afternoon
625
875
875
875
875
875
40
-------
within a narrow angular sector, it is difficult to detect maximum ground-level
concentrations that occur under this regime with a conventional air quality monitoring
network because the requisite density of fixed monitoring stations is logistically
infeasible.
Fumigation
When a stack plume is contained in a stable layer above the top of the
surface mixing layer, the plume remains aloft and does not mix to the ground.
Concentrations within the plume are high because of the very limited turbulent
mixing. When the top of the surface mixing layer reaches the height of the stack
plume, the pollutants are suddenly brought to the ground. This process, called
fumigation, is transient and typically results in high ground level concentrations
that rarely persist for longer than 1 or 2 hours. Turner (1969) assumes a narrow
ribbon-like plume in the stable layer prior to fumigation. However, we found no
evidence of the classical ribbon-like plume in the fumigation cases we studied.
Instead of being confined in a narrow plume, the emissions at plume stabilization
height appear to be spread over a wide angular sector that may be as large as
360 degrees prior to fumigation. Briggs (1965) describes a similar behavior for
plumes from TVA power plants. We believe that daytime heating is generally the
cause of the fumigation in the Tacoma area. An exception is one case of a nighttime
fumigation (3 October 1972, see Section 4.3), possibly associated with vertical
wind shear. We have not found any evidence of sea-breeze fumigation as described
by Lyons and Cole (1973), probably because the land-water temperature difference
in the Tacoma area is generally much less than the land-water difference for the
cases studied by Lyons and Cole. Also, the over-water fetch of air reaching the
Tacoma area is perhaps shorter than the over-water fetch associated with the
usual sea-breeze fumigations.
41
-------
The Limited-Mixing Condition
The most common critical meteorological regime, the limited-mixing
condition, occurs with light winds whenever the ASARCO plume is contained within
the surface mixing layer. It should be noted that this definition differs from the
TVA definition of the limited-mixing condition (Carpenter, et al., 1971). We have
used the term "limited mixing" because it most accurately characterizes the
meteorological structure in the Tacoma area during these periods.
3. 3 GENERAL METEOROLOGICAL INPUTS
The following procedures were used to specify the general meteorological
inputs required by the short-term and long-term diffusion models described in
Appendix A.
Vertical Turbulent Intensities
Our vertical expansion (a ) equation, which includes the effects of the
z
initial vertical plume or building dimension, relates the vertical turbulent intensity
directly to plume growth (see Equation (A-13) of Appendix A). Table 3-7 lists the
values of the standard deviation of the wind elevation angle cr' corresponding to
E
the Pasquill stability categories. These values are based in part on the measure-
ments of Luna and Church (1971) and are consistent with the cr1 values implicit in
E
the vertical expansion curves presented by Pasquill (1961). In our calculations,
we have combined the E and F stability categories because we believe that surface
roughness elements and heat sources in the vicinity of the ASARCO smelter are
incompatible with the small diffusion coefficients and minimal turbulent mixing
associated with the Pasquill F stability category.
42
-------
Lateral Turbulent Intensities
Our lateral expansion (cr ) equation, which also includes the effects of the
J
initial lateral plume or building dimension, relates the lateral turbulent intensity
directly to plume growth (see Equation (A-ll) of Appendix A). In our calculations,
we assumed that the standard deviation of the wind azimuth angle cr' is equivalent
A
to cr' for a 10-minute averaging period. This assumption is supported by bivane
Ej
measurements made by Luna and Church (1971) and others at heights on the order
1/5
of 100 meters above the surface. The t law suggested by Osipov (1972) and
others was then used to obtain hourly a' values. That is, a ' for a given stability
1/5
category was multiplied by 1. 43 (6 ) to obtain the corresponding hourly cr' value.
J\
Table 3-7 also lists the cr' values used in our calculations.
A
Wind-Profile Exponents
In the diffusion models, the variation with height of the wind speed in the
surface mixing layer is assumed to follow a wind-profile exponent law of the form
\P
r ) (3-1)
R/
where
u{z) = wind speed at height z above the surface
u {z } - wind speed at reference height z above the surface
rl R
p = the wind-profile exponent
The wind-profile exponent law is used to adjust the mean wind speed from the
reference height (in this study, the measurement height at N26 and Pearl) to the
stack height for the plume-rise calculations and to the plume stabilization height
for the concentration calculations. Values of the wind-profile exponent p assigned
to the various combinations of wind speed and stability for the short-term and long-
term calculations are listed in Table 3-8. These exponent values are based on
the results obtained by DeMarrais (1959) and Cramer, et al. (1972).
43
-------
TABLE 3-7
VERTICAL AND LATERAL TURBULENT INTENSITIES
Pasquill Stability Category
A
B
C
D
E
V
(rad)
0. 1745
0.1080
0.0735
0. 0465
0.0350
"A
(rad)
0. 2495
0. 1544
0.1051
0. 0665
0.0501
TABLE 3-8
WIND PROFILE EXPONENTS
Pasquill
Stability
Category
A
B
C
D
E
Wind-Speed Category (meters per second)
0.0 - 1.5
0.10
0.10
0.20
0.25
0.30
1.6 - 3.1
0.10
0.10
0.15
0.20
0.25
3.2 - 5.1
0.10
0.10
0.15
0.25
5.2 - 8.2
0.10
0.10
— —
8.3 - 10.8
0.10
0.10
—
>10. 8
0.10
0.10
44
-------
Mixing Depths
Table 3-9 lists the mixing depths assigned to the various combinations of
wind-speed and stability categories. The mixing depths assigned to the unstable
A, B and C stability categories are the average of the seasonal median afternoon
mixing depths (see Table 3-6). Similarly, the mixing depths assigned to the
combined E and F stability categories are the average of the median early morning
mixing depths. The median early morning and afternoon mixing depths were
averaged and assigned to the D stability category. The mixing depths in Table 3-9
were used in the long-term calculations. The mixing depths used in the short-term
calculations were estimated from the concurrent upper-air sounding at the Portage
Bay EMSU station closest to the time period of interest.
Ambient Air Temperatures
The Briggs (1971) plume-rise equations given in Section A. 2 of Appendix A
require the ambient air temperature as an input. The ambient air temperatures
used in the short-term calculations were the temperatures measured at Me Chord
Air Force Base. The ambient air temperatures used in the long-term calculations
are listed in Table 3-10. These ambient temperatures were determined by averaging
the monthly average temperatures for Seattle, Washington which are published in
the Climatic Atlas of the United States (1968). The normal daily maximum tempera-
ture was assigned to the A, B and C stability categories; the average daily tempera-
ture was assigned to the D stability category; and the normal daily minimum tempera-
ture was assigned to the combined E and F stability categories.
Vertical Potential Temperature Gradients
The Briggs (1971) plume-rise equations given in Section A. 2 of Appendix A
also require the vertical potential temperature gradient as an input. Table 3-10
also lists the potential temperature gradients assigned to the various combinations
45
-------
TABLE 3-9
MIXING DEPTHS IN METERS
Pasquill
Stability
Category
A
B
C
D
E
Wind -Speed Category (meters per second)
0.0 - 1.5
875
875
875
500
125
1.6 - 3.1
940
940
940
595
250
3.2 - 5.1
940
940
690
440
5.2 - 8.2
1000
875
8.3 - 10.8
1000
1000
> 10.8
1000
1000
TABLE 3-10
AMBIENT AIR TEMPERATURES AND VERTICAL
POTENTIAL TEMPERATURE GRADIENTS
Pasquill
Stability
Category
A
B
C
D
E
Ambient Air
Temperature
(OK)
288
288
288
284
279
Potential Temperature Gradient (°K/m)
Wind-Speed Category (meters per
0.0-1.5
0.00
0.00
0.00
0.00
0.03
1.6-3.1
0.00
0.00
0.00
0.00
0.02
3.2-5.1
0.00
0.00
0.01
5.2-8.2
0.00
0.00
second)
8.3-10.8
0.00
0.00
>10. 8
0.00
0.00
46
-------
of wind-speed and stability categories. The potential temperature gradients in
Table 3-10 are based on the Turner (1964) and Pasquill (1961) definitions of the
Pasquill stability categories, the measurements of Luna and Church (1971), and
our own previous experience. The potential temperature gradients in Table 3-10
were used in the long-term calculations. The potential temperature gradients used
in the short-term calculations were estimated from the concurrent upper-air sounding
at the Portage Bay EMSU station.
47
-------
48
-------
SECTION 4
HISTORICAL CALCULATIONS
4.1 SELECTION OF DIFFUSION MODELS
During our background study of the ASARCO smelter, we reviewed a
number of long-term and short-term atmospheric diffusion models that might be
applicable to the topographical and meteorological factors affecting the dispersion
of smelter SOo emissions. On the basis of our survey, we selected one long-term
and five short-term models that were tested during the Model Development Phase
of our study. The long-term model was the H. E. Cramer Company generalized
long-term diffusion model (Cramer, et al., 1975). As noted in Section A. 4 of
Appendix A, this model is very similar to the EPA's Climatological Dispersion
Model (COM). The results of our calculations using the long-term model are
discussed in detail in Section 4.2. The short-term models that we tested were:
• The H. E. Cramer Company short-term model (Cramer,
et_al., 1975)
• The Lyons and Cole (1973) sea-breeze fumigation model
• The Turner (1969) fumigation model
• The H. E. Cramer Company air-stagnation model
(Cramer and Bowers, 1974)
• The Briggs (1965) air-stagnation model
With the exception of the H. E. Cramer Company short-term and long-term models,
it was necessary to modify all of the above models to include the effects of variations
in terrain height over the calculation grid. The results of trial calculations using
the five short-term models are briefly discussed in the following paragraphs.
49
-------
The H. E. Cramer Company short-term model (see Section A. 3 of Appendix
A) is a generalized model that accepts site-specific meteorological input parameters
and includes terrain effects. We have used this model in a number of air quality
impact studies (for example, Cramer, et al., 1975), and the correspondence between
calculated and observed concentrations has generally been excellent. As discussed
in detail in Section 4. 3, the model worked well in historical calculations for two of
the three critical meteorological regimes: the critical wind-speed condition and
the limited-mixing condition (see Section 3.2).
In our analysis of the air quality and meteorological data, we found two
hours (6 September and 7 October 1974) when sea-breeze fumigation may have
resulted in high ground-level SO2 concentrations in the Tacoma area. Sea-breeze
fumigation is a process that can occur when the land is significantly warmer than an
adjacent large body of water and a sea breeze transports a stable air mass that has
formed above the water over the land. A new unstable boundary layer with a semi-
parabolic shape then begins to form at the land-water interface. The plume from a
stack located near the shoreline that stabilizes above the new (unstable) boundary
layer will travel inland with minimal growth until it intersects the thermally-
unstable boundary layer, where the plume is brought quickly to the ground. For the
two hours when sea-breeze fumigation may have occurred, our short-term model
matched the observed concentrations as well as, or better than, the Lyons and Cole
(1973) sea-breeze fumigation model. As noted in Section 3. 2, we believe that the
land-water temperature difference in the Tacoma area is generally too small for
sea-breeze fumigation as described by Lyons and Cole to occur. Also, the over-
water fetch in the Tacoma area may not be long enough to permit the formation
of a deep stable layer.
We have found no evidence in the air quality data for the fumigation of a
narrow, ribbon-like plume as described by Turner (1969). Instead of being confined
in a narrow plume, the emissions at plume stabilization height appear to be spread
50
-------
over a wide angular sector that may be as large as 360 degrees prior to the fumiga-
tion. When fumigation begins, high ground-level SCv> concentrations occur almost
simultaneously over the entire area. Briggs (1965) describes a similar behavior
for plumes from TVA power plants. However, the Briggs (1965) model overpre-
dicted the observed concentrations by as much as a factor of 3 to 5. The Cramer
and Bowers (1974) air-stagnation fumigation model tended to underestimate the
maximum hourly concentrations, but generally agreed well with the average concen-
trations for the period of fumigation. The Cramer and Bowers model is a modifi-
cation of the long-term diffusion model described in Section A. 4 of Appendix A.
The model makes the following assumptions:
• Plume rise is determined by the stable meteorological
conditions prior to fumigation
• The wind-direction distribution in the stable layer is
circular (i.e., pollutants in the stable layer are spread
about the stack in a circular pattern)
• When the mixing depth reaches the plume stabilization
height, the pollutants are rapidly mixed to the ground
(the rate of vertical mixing is determined by the vertical
turbulent intensity in the unstable surface layer)
Based on the trial calculations using the five short-term models described
above, we selected two models for use in the short-term historical calculations
which are summarized in Section 4. 3 and described in detail in Appendix E. The
H. E. Cramer Company generalized short-term model was used for all short-term
historical cases except the fumigation cases; the H. E. Cramer Company air-
stagnation fumigation model was used for the fumigation cases.
51
-------
It should be noted that the short-term diffusion models that we selected
for use in this study are advanced forms of the Gaussian plume model and thus are
very similar to the models used by EPA and others. However, the principal
advantage of the selected models is that they use meteorological inputs that can be
directly related to observed meteorological conditions and also include the most
recent developments in the treatment of buoyant plume rise as well as variations
in terrain height.
4.2 1972 ANNUAL AVERAGE CONCENTRATION CALCULATIONS
Variations in stack exit temperatures, flow rates and SOg emissions
resulting from the operation of the ASARCO SCS complicated the selection of ASARCO
source input parameters used in modeling the 1972 annual average concentrations.
Analysis of the 1972 curtailment reports submitted by ASARCO to PSAPCA showed
that the two stack heaters were generally turned on during the daytime and at such
other times when high SO2 ground-level concentrations were either observed or
expected to occur at the monitors. Similarly, the 1972 curtailment reports indicated
that smelter SO2 emissions were routinely curtailed 50 to 90 percent or more when
both stack heaters were turned on. We interpret this to mean that, whenever the
plume from the main stack was observed or expected to be in the surface mixing
layer (which is the required condition for high ground-level concentrations at the
monitors impacted by the ASARCO plume), operation of the SCS resulted in both
stack heaters being turned on and emissions curtailments of 50 to 90 percent.
To take into account the effects of the ASARCO SCS on SO2 emissions
during 1972, a hypothetical SCS operating regime was developed for the 1972 annual
concentration calculations. Any such approximation of the SCS operating regime
must be consistent with the curtailment reports and the total SO2 emissions during
1972. In our hypothetical SCS operating regime, the following assumptions were
made:
52
-------
• The exit temperature and volumetric emission rate
given in Table 2-1 for the main ASARCO stack with
both heaters in operation applied whenever the ASARCO
plume was contained within the surface mixing layer
• On the average, the smelter SCS reduced SO« emissions
on the basis of wind speed and the Pasquill stability
category as shown in Table 4-1
• As shown in Table 4-2, the SO2 emission rate for
each joint combination of wind-speed and stability
categories was obtained from the product of the maximum
uncurtailed emission rate for 17-percent permanent
control (26 tons per hour) and the percent emissions
reduction given in Table 4-1
If the hourly emission rate given in Table 4-2 for each wind-speed/
stability combination is multiplied by the number of hours during 1972 of
that particualr wind-speed/stability combination (see Appendix B) and the
annual emissions thus obtained for each combination are summed, the calculated
total SO2 emissions from the main ASARCO stack during 1972 are 155, 562 tons.
This is very close to the total annual emissions for 1972 of 156,000 tons reported
by ASARCO. Thus, our hypothetical SCS operating regime is consistent with both
the curtailment reports and the total emissions reported by ASARCO for 1972.
However, we recognize that this hypothetical SCS operating regime for 1972 is
only one of a number of possible regimes.
The general meteorological inputs discussed in Section 3.3, the annual
wind distributions at N26tn and Pearl given in Appendix B, the ASARCO emissions
data discussed above and the emissions data for the other sources in the Seattle-
Tacoma area listed in Appendix C were used with the long-term concentration
53
-------
TABLE 4-1
PERCENT REDUCTION IN SO2 EMISSIONS ASSUMED IN THE
1972 ANNUAL CONCENTRATION CALCULATIONS
Pasquill
Stability
Category
A
B
C
D
E
F
Wind Speed (m/sec)
0.0-1.5
90
90
80
0
0
0
1.6-3.0
90
90
80
0
0
0
3.1-5.1
—
90
80
60
50
—
5.2-8.2
—
—
80
60
—
—
8.3 - 10.8
—
—
80
60
—
—
S10.8
—
—
—
80
—
—
TABLE 4-2
SO2 EMISSION RATES IN TONS PER HOUR ASSUMED IN THE
1972 ANNUAL CONCENTRATION CALCULATIONS
Pasquill
Stability
Category
A
B
C
D
E
F
Wind Speed (m/sec)
0.0-1.5
2.6
2.6
5.2
26.0
26.0
26.0
1.6-3.0
2.6
2.6
5.2
26.0
26.0
26.0
3.1-5.1
2.6
5.2
10.4
13.0
5.2-8.2
5.2
10.4
8. 3 - 10. 8
5.2
10.4
2=10.8
10.4
54
-------
model described in Section A. 4 of Appendix A to calculate 1972 annual average
ground-level SO2 concentrations. Kaiser Aluminum was not included in the calcu-
lations because SO2 emissions were discharged from a 152-meter (500-foot) stack
during 1972 and we do not know the exit parameters for this stack. Concentrations
were calculated for a total of 1012 grid points on a 20-kilometer grid. The grid
spacing within 6 kilometers of the ASARCO smelter was 0. 5 kilometers; a 1-
kilometer grid spacing was used beyond 6 kilometers. Concentrations were also
calculated for the locations of the PSAPCA and ASARCO air quality monitors. The
procedures described in Section A. 5 of Appendix A were used to account for varia-
tions in terrain height over the calculation grid. Also, the SO2 decay constant $
(see Equations (A-9) and (A-23) in Appendix A) was set equal to zero because the
rate of decay, although not well established, is believed to be insignificant for the
distance and time intervals considered in this study (see Eliassen and Saltbones,
1975).
Figure 4-1 shows the calculated isopleths of annual average ground-level
SO2 concentration produced by the combined sources. The location of the main
ASARCO stack is indicated by the star in the figure. The calculated maximum
annual average SO2 concentration of 0. 0145 parts per million (ppm) is located
about 6. 5 kilometers southeast of the main ASARCO stack. The St. Regis Paper
Mill is principally responsible for this calculated concentration; the contribution
of ASARCO emissions is negligible at this point. The calculated concentration
isopleths in Figure 4-1 indicate that the 1972 annual average SO2 concentrations
in the Tacoma area were below the National and PSAPCA ambient air quality
standards of 0. 03 and 0. 02 ppm, respectively.
Figure 4-2 shows the calculated isopleths of annual average ground-level
SO2 concentration due to the ASARCO emissions. The isopleth pattern closely
resembles the reversed wind-direction distribution at N26th and Pearl shown in
Figure 3-1. The calculated maximum annual average concentration produced by
55
-------
FIGURE 4-1.
Calculated isopleths of 1972 annual average ground-level SOg
concentration in parts per million produced by emissions from
all sources. Locations of the ASARCO and PSAPCA SC-2 monitors
are indicated by the + symbols.
56
-------
530
FIGURE 4-2
Calculated isopleths of 1972 annual average ground-level SO
concentration in parts per million produced by ASARCO emissions.
The star shows the location of the main ASARCO stack.
57
-------
ASARCO emissions is 0.0083 ppm. This concentration occurs in Tacoma, about
6. 3 kilometers south-south west of the ASARCO stack. The other SO2 sources
contribute about 0. 0003 ppm at this point, leading to an annual average SO0
^j
concentration produced by all sources of 0.0086 ppm. A calculated secondary
maximum annual average concentration produced by ASARCO emissions of
0.0056 ppm occurs on Maury Island about 6. 9 kilometers north-northeast of the
ASARCO stack. Other major SO2 sources contribute about 0. 0022 ppm at this
point to yield a calculated annual average SO2 concentration from all sources
combined of 0. 0078 ppm.
Table 4-3 compares the calculated and observed SO2 concentrations for
the PSAPCA and ASARCO SO2 monitors in the Tacoma area. The locations of
the monitors are shown by the + symbols in Figures 4-1 and 4-2; the monitor
symbols used in the figures are enclosed by parentheses following the monitor
names in Table 4-3. In comparing the calculated and observed concentrations, it
is important to note that the threshold and accuracy of the two types of SO2 monitors
used by PSAPCA and ASARCO are both equal to 0. 01 ppm. With the exception of
the ASARCO monitor at N26th and Pearl, all of the observed annual average con-
centrations are below the accuracy and threshold of the various SO2 monitors. If
we assume that all of the observed annual concentrations have an error range equal
to plus or minus one-half the monitor threshold and accuracy of 0.01 ppm (i. e.,
0. 005 ppm) and if we further assume that the annual concentration measured by
PSAPCA at N26th and Pearl is more representative than the corresponding concen-
tration measured by the ASARCO monitor, Table 4-3 shows that all of the calculated
concentrations are within the range of error inherent in the measurements. We
have no explanation for the difference in the annual average concentrations measured
by the ASARCO and PSAPCA monitors at N26th and Pearl. The PSAPCA Meeker
and the ASARCO Meeker-Brown monitors, which are also located in close proximity,
measured identical annual average concentrations during 1972.
58
-------
TABLE 4-3
COMPARISON OF CALCULATED AND OBSERVED 1972 ANNUAL
AVERAGE GROUND-LEVEL SO CONCENTRATIONS
£l
Monitor
Vashon Island (K56)
N26th and Pearl
PSAPCA(P2)
ASARCO(Al)*
Fife Sr. H. S. (P3)
Meeker
PSAPCA (P6)
ASARCO (A5)*
Adams St (P7)
Reservoir (A2)*
Highlands (A3)*
University Place (A4)*
Calculated Concentration (ppm)
ASARCO
0. 0026
0. 0023
0. 0000
0.0013
0. 0000
0.0010
0. 0042
0. 0033
Other Sources
0.0016
0. 0014
0.0012
0. 0045
0.0017
0. 0014
0. 0015
0.0018
All Sources
0. 0042
0. 0036
0.0012
0.0057
0.0018
0. 0024
0.0057
0.0051
Observed
Concentration
(ppm)
0.005
0.007
0.022
0.003
0.004
0.004
0.004
0.003
0.005
0.002
* ASARCO SO2 monitor
59
-------
It is emphasized that in our 1972 annual average SC>2 concentration
calculations, we used a simple hypothetical approximation to represent the effects
of the complicated ASARCO SCS operating regime. Our hypothetical SCS operating
regime is consistent with the total SO£ emissions during 1972, but it is only one of
a number of possible approximations. We therefore did not expect that our calcu-
lated annual concentrations would exactly match the observed concentrations. How-
ever, if the uncertainties in the observed concentrations caused by instrument
threshold and response are considered, the calculated concentrations are in very
good agreement with the air quality data.
4. 3 SELECTED SHORT-TERM CALCULATIONS
4. 3.1 Calculation Procedures
During the Model Development Phase of our study, we analyzed meteoro-
logical and air quality data provided by PSAPCA and selected 20 dates when high
SOg concentrations were measured for one or more hours at one or more monitors
in the Tacoma area. These 20 Short-Term Historical Cases include examples of
the three critical meteorological regimes: the critical wind-speed condition, the
limited-mixing condition and fumigation. A discussion of these regimes is given
in Section 3. 2.
The short-term diffusion model described in Section A. 3 of Appendix A,
including the terrain-adjustment procedures discussed in Section A. 5, was used
in the calculations for the critical wind-speed and limited-mixing conditions. In
applying this model, we varied the wind direction by as much as 10 to 20 degrees
to obtain the best correspondence at all SO£ monitors between calculated and
observed concentrations. This procedure was necessary because the hourly mean
wind direction at plume stabilization height used in the model calculations must be
accurate within 1 or 2 degrees if the plume is to be accurately positioned with
60
-------
respect to fixed points. The mean hourly wind directions obtained from the 15-meter
level at the PSAPCA N26^n and Pearl station can easily differ from the mean wind
direction at the stabilization height of the ASARCO plume by 10 to 20 degrees or
more. In the Tacoma cases where high concentrations were concurrently observed
at two or more monitors, it was often possible to estimate the wind direction at
plume stabilization height within about 1 degree on the basis of a comparison of the
calculated and observed concentrations. A more detailed discussion of the sensi-
tivity of short-term ground-level concentrations calculated at fixed points to small
errors in the wind direction is given in Appendix E in the discussion of Case 4
(5 July 1972).
The long-term diffusion model described in Section A. 4 of Appendix A,
including the terrain-adjustment procedures discussed in Section A. 5, was used
in the fumigation calculations by applying the following procedures:
• The mean wind speed, wind-profile exponent, ambient
air temperature and vertical potential temperature
gradient used in each fumigation calculation were set
equal to values representative of the stable conditions
prior to the onset of fumigation
• The wind-direction distribution was assumed to be
circular (i.e., f. . was equal to A0'/27r)
!» j» K, g,
• The vertical turbulent intensity
-------
The meteorological inputs used in the calculations for the first 19 Short-
Term Historical Cases are listed in Table 4-4. The meteorological inputs for
Case 20 (25 December 1974) are included with the discussion of that case (see
Appendix E). The wind speeds and the wind directions enclosed by parentheses
in Table 4-4 are the hourly values measured at the PSAPCA N26th and Pearl
station; the wind directions without parentheses are the wind directions used in
the calculations. In each case, the mixing depth and potential temperature gradient
were estimated from the most recent Portage Bay EMSU sounding. The McChord
Air Force Base cloud-cover observations were combined with the N26^ and Pearl
wind speeds in assigning a Pasquill Stability category to each hour following the
Turner (1964) criteria (see Tables 3-2 and 3-3). The lateral and vertical turbulent
intensities corresponding to the various stability categories are given in Table 3-7.
The ambient air temperatures in Table 4-4 are the temperatures measured at
McChord Air Force Base. The wind-profile exponents were assigned on the basis
of the wind speed and Pasquill stability category (see Table 3-8).
Table 4-5 lists the hourly values of the SO2 emission rates, the volumetric
emission rates and the stack exit temperatures for the main ASARCO stack used
in the calculations for the first 19 Short-Term Historical Cases. The source inputs
for Case 20 are provided with the discussion of that case in Appendix E. For each
case listed in Table 4-5, the volumetric emission rate and stack exit temperature
were determined on the basis of the number of stack heaters stated in the ASARCO
curtailment report to be in operation (see Section 2.1). Where possible, the SOo
emission rates were estimated from in-stack SOg concentration measurements. In
the absence of in-stack SO2 data (see Table 4-5), the emission rates were approxi-
mated by the simple equation
r/inn- 1>"| (4_1}
62
-------
TABLE
METEOROLOGICAL INPUT PARAMETERS FOR THE FIRST 19
E-IORT-TEEM HISTOBICAL CASES
01
w
Case
No.
1
2
3
4
5
6
7
8
9
10
11
Date
29 March 1972
4 May 1972
4 June 1972
5 Julv 1972
30 October 1972
11 October 1972
24 October 1972
26 May 1973
4 June 1973
19 June 1973
25 June 1973
Hour
(PST)
1200
1300
1400
2000
1400
0300
0000
0100
0200
0300
0400
0900
1000
2200
2300
0700
1100
2100
1100
1200
1300
Meteorological
Regime
Fumigation
Critical Wind Speed
Limited Mixing
Limited Mixing
Fumigation
(Wind Shear)
Critical Wind Speed
Limited Mixing
Limited Mixing
Limited Mixing
Limited Mixing
Limited Mixing
Wind*
Direction
irteg)
Vrbl
014(009)
014(020)
010(007)
Vrbl
018(032)
012(032)
012(042)
196(250)
013(019)
007(026)
012(Vrbl)
013(329)
013(022)
013(013)
018(038)
021(047)
Wind
Speed
(m/secl
0.940
5. 092
2.058
1.595
0. 323
6.996
6.431
1.235
3.087
1.801
2.315
1.183
1.338
3.192
2.264
2.881
2.727
Ambient
Air
Temp.
<°K>
273
288
295
287
233
276
276
278
281
281
280
283
290
292
297
298
299
Potential
Temperature
Gradient
(°K/m)
0.008
0.007
0.008
0.016
0.004
0.005
0.005
0.003
0.003
0.003
0.010
0.003
0.003
0.006
0.003
0.003
0.003
Mixing
Depth
(m)
528
1000
600
620
600
530
530
500
1500
1000
500
1300
1300
1800
1000
1000
1000
Wind
Profile
Exponent
0.30
0.15
0.10
0.25
0.30
0.10
0.10
0.25
0.20
0.25
0.25
0.25
0.10
0.20
0.10
0.10
0.10
Pasquill
Stability
Category
B
D
C
E
B**
D
D
D
D
E
E
D
A
E
B
B
B
* Wind directions enclosed by parentheses are directions measured at N26th and Pearl; directions not enclosed by parentheses are directions used
in calculations.
** Although this stability category did not exist, we estimate that the vertical turbulent intensity in the surface mixing layer was equivalent to the
value associated with this category.
-------
TABLE 4-4 (conf d)
METEOROLOGICAL INPUT PARAMETERS FOR THE FIRST 19
SHORT-TERM HISTORICAL CASES
Case
No.
12
13
14
15
16
17
18
19
Date
30 July 1973
n July 1973
17 October 1973
28 January 1974
9 February 1974
26 February 1974
G September 1974
7 October 1974
Hour
(PST)
0900
1000
0500
0600
0700
0900
1000
1300
0300
0400
1200
1000
1 1 no
i J.UU
1200
0400
1600
0200
1200
1300
Meteorological
Regime
Limited Mixing
Limited Mixing
Limited Mixing
Critical Wind Speed
Fumigation
Limited Mixing
Limited Mixing
Limited Mixing
Wind*
Direction
(deg)
224(232)
223(241)
259(255)
260(251)
266(266)
223(246)
228(228)
228(242)
218(210)
213(215)
218(215)
Vrbl
218(196)
018(040)
010(020)
019(020)
018(014)
Wind
Speed
(m/sec)
1.69S
1.698
2.006
2.058
1.749
1.235
1.3S9
2.624
6.739
5.247
7.151
0. 514
2.727
1.183
2.161
3.447
3.653
Ambient
Air
Temp.
(°K)
288
290
285
. 285
285
287
287
292
231
281
282
276
278
297
278
290
292
Potential
Temperature
Gradient
<°K 'm)
0.007
0.007
0.008
0.008
O.OOs
0.008
0.008
0.004
0.002
0.002
0.002
0.003
0.003
0.005
0.017
0.003
0.003
Mixing
Depth
(m)
450
450
800
800
800
800
800
1150
600
600
600
736
1000
1950
980
980
980
Wind
Profile
Exponent
0.20
0.20
0.20
0.20
0.20
0.25
0.25
0.15
0.10
0.10
0.10
0.25
0.20
0.10
0.25
0.10
0.10
Pasquill
Stability
Category
D
D
D
D
D
D
D
C
D
D
D
B
D
B
E
B
C
* Wind directions enclosed by parentheses are directions measured at N26th and Pearl; directions not enclosed by parentheses are directions used
in calculations.
-------
TABLE 4-5
EMISSIONS DATA FOR THE MAIN ASARCO STACK FOR THE
FIRST 19 SHORT-TERM HISTORICAL CASES
O5
cn
Case
No.
1
2
3
4
5
6
7
8
9
10
11
Date
29 March 1972
4 May 1972
4 June 1972
5 July 1972
3 October 1972
11 October 1972
24 October 1972
26 May 1973
4 June 1973
19 June 1973
25 July 1973
Hour
(PST)
1200
1300
1400
2000
1400
0300
0000
0100
0200
0300
0400
0900
1000
2200
2300
0700
1100
2100
1100
1200
1300
SO Emission
Rate (Ib/hr)
52,000*
31,720*
4,420*
21,493*
52,000*
44, 850*
11,960*
9,880*
40, 032
34,778
15, 888
25, 145
4,379
17,139
6,505
5,880
6,255
Volumetric Emission
Rate (m /sec)
300.23
300.23
300.23
300.23
277.22
247.63
300.23
300.23
300.23
300. 23
300.23
300.23
300.23
300.23
300.23
300.23
300.23
Stack Exit
Temperature (°K)
411.0
411.0
411.0
411.0
388.8
355.4
411.0
411.0
411.0
411.0
411.0
411.0
411.0
411., 0
411.0
411.0
411.0.
*Emission rate estimated from Equation (4-1) and the curtailment report
-------
TABLE 4-5 (cont* d)
EMISSIONS DATA FOR THE MAIN ASARCO STACK FOR THE
FIRST 19 SHORT-TERM HISTORICAL CASES
05
Oi
Case
No.
12
13
14
15
16
17
18
19
Date
30 July 1973
31 July 1973
17 October 1973
28 January 1974
9 February 1974
26 February 1974
i
6 September 1974
7 October 1974
Hour
(PST)
0900
1000
0500
0600
0700
0900
1000
1300
0300
0400
1200
1000
1100
1200
0400
1600
0200
1200
1300
SO Emission
Rate (Ib/hr)
32,776
29,023
31, 775
27,772
18,515
27,772
18,515
32,276
30,976
31,650
21,517
24,248
38,115
4,527
33,507
6,067
8,167
Volumetric Emission
Rate (m /sec)
300.23
300.23
300,23
300.23
300.23
300.23
300.23
300.23
277.22
300.23
300.23
277.22
277.22
260.95
260. 95
260.95
260.95
Stack Exit
Temperature (°K)
411.0
411.0
411.0
411.0
411.0
411.0
411.0
411.0
388.8
411oO
411.0
388.8
388.8
399.9
380.4
399.9
399.9
-------
where Q is the SC>2 emission rate, Q0 is the maximum uncurtailed emission rate
(see Section 2.1) and C is the percent curtailment indicated by the ASARCO cur-
tailment report. In some cases it was necessary to consider the delay time in the
downwind transport of stack emissions so that the emissions parameters in Table
4-5 do not always correspond to the hours listed in the table. These cases are
specifically noted in the discussion of the individual cases presented in Appendix E.
The concentration calculations for the Short-Term Historical Cases in-
cluded the contributions of all sources listed in Appendix C except Kaiser Aluminum.
As discussed in Section 4.2, Kaiser Aluminum was excluded from the historical
calculations because of uncertainties about the emissions parameters. Also, as
explained in Section 4.2, the SC>2 decay constant 0 was set equal to zero in all of
the snort-term calculations.
4. 3.2 Results of the Short-Term Historical Calculations
The results of the concentration calculations for the 20 Short-Term
Historical Cases are summarized below and are discussed in detail in Appendix
E. Table 4-6 shows the calculated and observed hourly SO2 concentrations
for the critical wind-speed and limited-mixing cases when the observed hourly con-
centrations exceeded 0.25 ppm. In each case, ASARCO emissions account for
at least 99 percent of the calculated concentration. The numbers enclosed by
parentheses in the table refer to concentration measurements by the ASARCO SO2
monitor at N26tl:i and Pearl. The hours listed in Table 4-6 include only those hours
when the ASARCO plume was simultaneously detected by two or more SO2 monitors
so that the wind direction could be accurately estimated. Thus, the Maury Island
and Vashon Island cases of high SO£ concentrations are not included in Table 4-6
but are discussed separately. As shown by the mean ratios of calculated and
observed concentrations at the bottom of Table 4-6, the calculated concentrations
67
-------
TABLE 4-6
COMPARISON OF CALCULATED AND OBSERVED SO2 CONCENTRATIONS
FOR CRITICAL WIND-SPEED AND LIMITED-MIXING CASES WHEN THE
OBSERVED HOURLY CONCENTRATION EXCEEDED 0. 25 PPM
Case No.
2
3
4
6
7
8
9
10
12
13
Monitor
N26th and Pearl
Reservoir
Highlands
N26th and Pearl
N26th and Pearl
Reservoir
N26th and Pearl
N26th and Pearl
N26th and Pearl
N26th and Pearl
Reservoir
N26th and Pearl
N26th and Pearl
N26th and Pearl
McMicken Heights
Meeker
Meeker-Brown
Meeker
Meeker-Brown
Meeker
Meeker -Brown
Observed
Concentration*
(ppm)
(1.23)
0.27
0.73
0.40 (0.46)
0. 56 (0. 68)
0.62
0.30 (0.22)
0.26 (0.26)
0. 60 (0. 46)
0.32 (0.21)
0.37
0. 37 (0. 67)
0.31 (0.21)
0.25 (0.25)
0.50
0.28
0.31
0.38
0.42
0.59
0.49
Calculated
Concentration
(ppm)
0.88
0.04
0.44
0.25
0.61
0.64
0.30
0.23
0.26
0.44
0.39
0.64
0.11
0.34
0.52
0.28
0.35
0.42
0.50
0.56
0.53
Calc. Cone.**
Observed Cone,
(0. 72)
0.15
0.60
0.63 (0.54)
1.09 (0.90)
1.03
1.00 (1.36)
0.88 (0.88)
0.43 (0.57)
1.38 (2.10)
1.05
1.73 (0.96)
0.35 (0.52)
1.36 (1.36)
1.04
1.00
1.13
1.11
1.19
0.95
1.08
68
-------
TABLE 4-6 (Continued)
Case No.
14
19
Monitor
McMicken Heights
Tukwila
N26th and Pearl
Reservoir
N26th and Pearl
Observed
Concentration
(ppm)
0.30
0.41
0. 42 (0. 35)
0.50
0.27 (0.17)
Calculated
Concentration
(ppm)
0.09
0.10
0.87
1.00
0.10
Mean Ratio
Calc. Cone.**
Observed Cone.
0.33
0.24
2.07 (2.49)
2.00
0.37 (0.59)
0.97 (1.00)
*Numbers enclosed by parentheses are concentrations measured by the
ASARCO SO2 monitor at N26th and Pearl.
**Numbers enclosed by parentheses are the ratios of calculated and observed
concentrations for the ASARCO monitor at N26 and Pearl.
69
-------
are, on the average, equal to the observed concentrations. However, some of the
calculated and observed concentrations for the individual cases differ by a factor
of two or more. Differences of this order between observed and calculated concen-
trations for a specific hour at individual monitoring sites are not unexpected because
of uncertainties in meteorological and emissions input data. Even with the advantage
gained in specifying the wind direction by having two or more monitors in the plume,
it is still not possible to position the plume with sufficient accuracy to preclude
uncertainties of 25 or 30 percent. However, we believe that uncertainties in the
ASARCO hourly SC>2 emission rates are the principal cause of the large differences
between the calculated and observed concentrations for the individual cases.
Although the Maury Island and Vashon Island cases are not included in
Table 4-6, the calculated concentrations at these sites are entirely consistent
with the observed concentrations. That is, in each case the concentration calcu-
lated by assuming that the ASARCO plume passed directly over the SC>2 monitor
is greater than or equal to the observed concentration. For example, the observed
concentration on Maury Island for the hour ending at 0400 PST on 26 February
1974 (Case 17) was 0.33 ppm. The corresponding calculated centerline concen-
tration is 0.47 ppm, and a 2- to 3-degree shift in the wind direction used in the
model calculations brings the calculated and observed concentrations into exact
agreement. We point out that, when the wind blows from the ASARCO stack toward
either the Maury Island or the Vashon Island SO2 monitor, ASARCO is the only
source that contributes to the concentration calculated for the monitor.
Table 4-7 shows the calculated and observed SO2 concentrations for the
fumigation cases when the observed hourly concentrations exceeded 0.25 ppm. The
calculated concentrations are compared with both the maximum hourly concentration
during the period of fumigation and the average concentration during the period of
fumigation. As shown by the mean ratios of calculated and observed concentrations
70
-------
TABLE 4-7
COMPARISON OF CALCULATED AND OBSERVED SO2 CONCENTRATIONS
FOR FUMIGATION CASES WHEN THE OBSERVED HOURLY
CONCENTRATION EXCEEDED 0.25 PPM
Case
No.
1
2
16
Monitor
N26th and
Pearl
Adams St.
Reservoir
N26th and
Pearl
N26th and
Pearl
Observed Concentration*
toDm)
Maximum
1-Hour
0. 19 (0. 36)
0.36
0.38
0.35 (0.16)
0.49 (0.52)
Avg. For
Period Of
Fumigation
0.13 (0.21)
0.17
0.17
0.19 (0.10)
0.40 (0.40)
Calc.
Cone.
(ppm)
0.16
0.14
0.18
0.14
0.08
Mean Ratios
Calculated Cone. **
Observed Cone.
Maximum
1-Hour
0.84 (0.44)
0.39
0.47
0.40 (0.88)
0. 16 (0. 15)
0.45 (0.47)
Avg. For
Period Of
Fumigation
1.23 (0.76)
0.82
1.06
0.74 (1.40)
0.20 (0.20)
0.81 (0.85)
^Numbers enclosed by parentheses are the concentrations measured by
the ASARCO monitor at N26th and Pearl.
**Numbers enclosed by parentheses are the ratios of calculated and observed
concentrations for the ASARCO monitor at N26^ and Pearl.
71
-------
at the bottom of the table, the calculated fumigation concentrations are, on the
average, about half of the maximum hourly concentrations and about 15 to 20
percent lower than the average concentrations. This result is not surprising in
view of the assumption of a circular wind distribution in the stable layer prior to
fumigation which is necessitated by the lack of wind data at plume height as well as
plume observations. Although fumigation is more difficult to model than the critical
wind-speed or limited-mixing conditions, our study indicates that the highest
ground-level SO2 concentrations in the Tacoma area are not associated with
fumigation. As noted previously, this conclusion is supported by the existing
air quality data.
72
-------
SECTIONS
EVALUATION OF THE CONTROL ALTERNATIVES
5.1 MAXIMUM ANNUAL AVERAGE CONCENTRATIONS
The general meteorological inputs discussed in Section 3.3, the 1972
annual wind distributions for N26 and Pearl given in Appendix B, the ASARCO
emissions data for the 21 Control Alternatives given in Section 2.2 and the
emissions data for the other major SO sources in the Seattle-Tacoma area
£
listed in Appendix C were used with the long-term concentration model described
in Section A. 4 of Appendix A to calculate annual average ground-level SO
£t
concentrations for each Control Alternative (See Section 2.2). Concentrations
were calculated for a total of 1012 grid points on a 20-kilometer by 25-kilometer
grid. The grid spacing within 6 kilometers of the ASARCO smelter was 0. 5 kilo-
meters; a 1-kilometer grid spacing was used beyond 6 kilometers. Concentrations
were also calculated for the locations of the PSAPCA and ASARCO air quality
monitors. The procedures described in Section A. 5 of Appendix A were used
to account for variations in terrain height over the calculation grid.
Table 5-1 lists, for each Control Alternative, the magnitude and location
of the calculated maximum annual average ground-level SO concentration pro-
^
duced by ASARCO emissions. Also, the table gives both the calculated contribution,
at the point of the maximum ASARCO concentration, of the other major SO
£t
sources in the Seattle-Tacoma area and the total concentration calculated for the
combined sources. As indicated by the table, Control Alternatives 2, 3, 12 and
14 fail to meet both the National annual air quality standard for SO of 0.03 ppm
2
and the PSAPCA annual standard of 0. 02 ppm. Control Alternatives 6 and 16
will meet the National annual standard, but not the PSAPCA annual standard.
The calculated maximum annual concentrations for the remaining Control
73
-------
TABLE 5-1
MAGNITUDES AND LOCATIONS OF CALCULATED MAXIMUM ANNUAL
AVERAGE GROUND-LEVEL SO2 CONCENTRATIONS
PRODUCED BY ASARCO EMISSIONS
Control
Alternative
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Location of
Maximum Concentration
UTM X
(m)
534,500
537,000
536,000
534,500
534,500
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
UTM Y
(m)
5,232,500
5,237,500
5,235,000
5,232,500
5,232,500
5,235,000
5,235,000
5,235,000
5,235,000
5,235,000
5,235,000
5,235,000
5,235,000
5,235,000
5,235,000
5,235,000
5,235,000
5,235,000
5,235,000
5,235,000
5,235,000
Calculated Concentrations
(ppm)
ASARCO
0. 01573
0. 14983
0. 03080
0. 00954
0.00961
0. 02077
0.00113
0. 00289
0. 00126
0.00327
0. 01253
0. 03374
0.001158
0.03118
0. 01002
0.02698
0. 00720
0. 00647
0. 00631
0. 00139
0.00132
Other
Sources
0.00039
0. 00165
0.00163
0.00039
0. 00039
0.00163
0.00163
0.00163
0. 0016~3
0.00163
0. 00163
0.00163
0.00163
0.00163
0. 00163
0.00163
0. 00163
0.00163
0. 00163
0. 00163
0.00163
Combined
Sources
0,01612
0. 15148
0. 03243
0. 00993
0.01000
0. 02240
0. 00276
0. 00452
0. 00289
0. 00490
0.01416
0.03537
0. 01321
0.03281
0.01165
0. 02861
0.00883
0. 00810
0. 00794
0. 00302
0.00295
74
-------
Alternatives are below both annual standards, even if the SO2 background produced
by the other sources is included.
Figure 5-1 shows the calculated isopleths of annual average ground-level
SO concentration produced by ASARCO emissions for Control Alternative 1 (the
existing plant configuration without SCS). The isopleth pattern closely resembles
the reversed wind-direction distribution shown in Figure 3-1 for N26 and Pearl.
The highest calculated concentrations are located south-southwest and northeast
of the smelter. According to the calculations, ASARCO emissions have little
impact on the long-term ambient air quality in areas northwest and southeast of
the smelter. The annual isopleth patterns for the other Control Alternatives also
closely resemble the reversed wind-direction distribution for N26 and Pearl.
Figure 5-2 shows, for the combined sources and ASARCO Control Alterna-
tive 1, the calculated isopleths of annual average ground-level SO concentration.
As can be seen in the figure, calculated SO concentrations above the National and
£t
PSAPCA annual air quality standards occur about 11 kilometers southeast of the
ASARCO stack in an area essentially unaffected by ASARCO emissions. These
high calculated concentrations are almost entirely due to low-level SO emissions
from the Kaiser Aluminum Plant. Assuming that the stack and emissions parameters
used in the calculations are correct, Kaiser Aluminum has a greater impact on
long-term ambient air quality than the ASARCO smelter.
5. 2 MAXIMUM MONTHLY AVERAGE CONCENTRATIONS
In addition to the PSAPCA annual ambient air quality standard of 0.02 ppm,
there is a PSAPCA monthly standard of 0. 04 ppm. During 1972, the maximum
monthly average ground-level SO concentration measured by the PSAPCA monitor
th
at N26 and Pearl was 0. 016 ppm for the month of October. This is 2.29 times
75
-------
FIGURE 5-1.
Calculated isopleths of annual average ground-level SO
concentration in parts per million produced by ASARCO
emissions for Control Alternative 1.
76
-------
FIGURE 5-2.
Calculated isopleths of annual average ground-level SO2
concentration in parts per million produced by the combined
sources for Control Alternative 1.
77
-------
the annual average concentration of 0. 007 ppm measured by the PSAPCA monitor.
We interpret this result to mean that the worst-case month will have a maximum
monthly average ground-level SO concentration equal to or greater than 2.3 times
£
the corresponding maximum annual average concentration. On this basis, it
follows that Control Alternatives 2, 3, 6, 12, 14 and 16 will not meet the PSAPCA
monthly standard. Thus, the Control Alternatives that fail to meet the PSAPCA
annual standard also fail to meet the PSAPCA monthly standard.
5. 3 MAXIMUM 3-HOUR AND 24-HOUR CONCENTRATIONS
As pointed out in Section 3.2, we believe that the critical wind-speed
condition (moderate-to-strong winds in combination with neutral stability) is the
principal meteorological regime associated with the maximum 3-hour and 24-hour
ground-level SO concentrations produced by ASARCO stack emissions. The wind-
^
persistence statistics presented in Table 3-5 indicate that wind speeds above 5.1
meters per second at N26 and Pearl are most likely to persist from the north-
northeast. We therefore analyzed the hourly wind data from N26 and Pearl for
the year 1972 in order to isolate a 24-hour period with persistent north-northeast
winds. The hourly data for this selected "worst-case" 24-hour period were used
in the short-term concentration calculations for the 21 Control Alternatives.
Table 5-2 lists the hourly meteorological inputs for the "worst-case"
24-hour period (1400 PST on 24 January to 1300 PST on 25 January 1972). The
wind speeds and wind directions are the hourly values measured at N26tn and Pearl.
The N26 and Pearl wind speeds were used in conjunction with concurrent cloud-
cover observations at Me Chord Air Force Base to assign the Pasquill stability
category to each hour following the Turner (1964) procedures. The lateral and
vertical turbulent intensities that correspond to the Pasquill stability categories
are given in Table 3-7. The ambient air temperatures in Table 5-2 are the
78
-------
TABLE 5-2
METEOROLOGICAL INPUTS FOR THE WORST-CASE 24-HOUR PERIOD
Hour
00
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Wind
Direction
(deg)
020
020
020
020
020
020
020
020
020
020
020
020
020
020
020
020
030
020
030
020
020
030
030
040
Wind
Speed
(m/sec)
4.116
5.144
5.144
6.173
5.659
6.173
7.717
8.231
8.231
10. 803
10. 803
11.318
10. 803
10. 803
9.774
10. 289
10. 803
9.774
8.745
7.202
7.202
7.717
7.202
6.688
Ambient
Air
Temperature
(°K)
274
274
274
274
274
273
273
273
272
271
270
270
270
270
270
270
269
269
269
269
269
269
269
269
Potential
Temperature
Gradient
(°K/m)
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0,003
Mixing
Depth
(m)
690
690
690
875
875
875
875
875
875
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
875
875
875
875
875
Wind
Profile
Exponent
P
0.15
0.15
0.15
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
Pasquill
Stability
Category
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
CO
-------
temperatures measured at Me Chord Air Force Base. For each hour, the wind-speed
and stability categories were used to select the wind-profile exponent from Table 3-8.
Because no Portage Bay EMSU soundings were made during this period, we set the
potential temperature gradient equal to the moist-adiabatic value which appears to
be characteristically associated with the occurrence of high short-term SO
^
concentrations (see Section 4.3). The hourly mixing depths were assigned on the
basis of wind speed and stability (see Table 3-9).
The meteorological inputs in Table 5-2, the ASARCO emissions data for
the 21 Control Alternatives (see Section 2. 2) and the emissions data for the other
major SO sources in the Seattle-Tacoma area listed in Appendix C were used with
^
the short-term concentration model described in Section A. 3 of Appendix A to
calculate 1-, 3- and 24-hour average ground-level SO concentrations for each
£*
Control Alternative. Concentrations were calculated for a total of 1012 grid points
on a 20-kilometer by 25-kilometer grid. The grid spacing within 6 kilometers of
the ASARCO smelter was 0. 5 kilometers; a 1-kilometer grid spacing was used
beyond 6 kilometers. Concentrations were also calculated for the locations of
the PSAPCA and ASARCO SO monitors. The procedures described in Section A. 5
^J
of Appendix A were used to account for variations in terrain height over the
calculation grid.
Table 5-3 lists the locations and the magnitudes of the calculated maximum
3-hour ground-level SO concentrations produced by ASARCO emissions for the
^
various Control Alternatives. All of the calculated maximum 3-hour concentrations
occur during the first 3 hours of the "worst-case" 24-hour period (see Table 5-2).
Because the average rather than the maximum uncurtailed SO emission rate was
Lt
used in the calculations for Control Alternative 1 (the existing configuration without
SCS) and because ASARCO has had difficulty in maintaining 51-percent permanent
control (see Section 2.2), the calculated maximum 3-hour concentration given in
Table 5-3 for this case may be too low.
80
-------
TABLE 5-3
MAGNITUDES AND LOCATIONS OF CALCULATED MAXIMUM 3-HOUR
GROUND-LEVEL SO2 CONCENTRATIONS PRODUCED BY
BY ASARCO EMISSIONS
Control
Alternative
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Location of
Maximum Concentration
UTM X
(m)
536,000
537,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
536,000
UTM Y
(m)
5,234,500
5,237,000
5,234,500
5,234,500
5,234,500
5,234,500
5,234,500
5,234,500
5,234,500
5,234,500
5,234,500
5,234,500
5,234,500
5,234,500
5,234,500
5,234,500
5, 234, 500
5,234,500
5,234,500
5,234,500
5,234,500
Calculated Concentrations
(ppm)
ASARCO
0. 63307
0. 89401
0. 88399
0.41222
0.43077
0. 59254
0. 05995
0.08214
0.06788
0. 09336
0.70281
0. 97214
0. 64945
0. 89822
0.56201
0. 77724
0. 40348
0.36248
0.35341
0.07486
0. 07032
Other
Sources
0. 00214
0. 00229
0. 00214
0. 00214
0. 00214
0.00214
0.00214
0. 00214
0.00214
0. 00214
0.00214
0. 00214
0. 00214
0. 00214
0. 00214
0. 00214
0. 00214
0. 00214
0. 00214
0.00214
0.00214
Combined
Sources
0. 63521
0. 89630
0.88613
0.41436
0.43291
0. 59468
0. 06209
0. 08428
0. 07002
0. 09550
0. 70495
0. 97428
0.65159
0. 90036
0.56415
0. 77938
0.40562
0. 36462
0.35555
0. 07700
0. 07246
81
-------
As shown by Table 5-3, Control Alternatives 1, 2, 3, 6, 11, 12, 13, 14,
15 and 16 have calculated maximum 3-hour ground-level SO concentrations that
&
exceed the National 3-hour ambient air quality standard of 0.50 ppm. For
these Control Alternatives, the calculated second highest 3-hour concentrations
during the worst case 24-hour period also exceed the 3-hour standard. It should be
noted that, under the selected "worst-case" meteorological conditions, the other
major SO sources in the Seattle-Tacoma area do not significantly contribute to
^
the calculated concentrations at the locations of the maximum concentrations
produced by ASARCO emissions.
Table 5-4 lists, for each Control Alternative, the location and magnitude
of the calculated maximum 24-hour average ground-level SO concentration produced
Lt
by ASARCO emissions. As noted above, the calculated 24-hour concentration for
Control Alternative 1 may be too low. The results of the calculations indicate that
Control Alternatives 1, 2, 3, 4, 5, 6, 11, 12, 13, 14, 15, 16, 17, 18 and 19 will
not meet the National 24-hour ambient air quality standard of 0.14 ppm or the
PSAPCA 24-hour ambient air quality standard of 0.10 ppm.
The meteorological inputs for the "worst-case" 24-hour period include
a 20-hour persistence of north-northeast winds above 5.1 meters per second.
According to Table 3-5, this condition can be expected to occur about once per
year. Table 3-5 also indicates that north-northeast winds above 5.1 meters per
second can be expected to persist for 14 or more hours about twice per year. It
follows that the second highest 24-hour average SO concentrations will be
&
approximately 70 percent (14/20) of the "worst-case" 24-hour concentrations
listed in Table 5-4. Thus, the second highest 24-hour average SO concentrations
Lt
for Control Alternatives 1, 2, 3, 4, 5, 6, 11, 12, 13, 14, 15, 16, 17, 18 and 19
should also exceed both the National and the PSAPCA 24-hour standards.
82
-------
TABLE 5-4
MAGNITUDES AND LOCATIONS OF CALCULATED MAXIMUM 24-HOUR
GROUND-LEVEL SO2 CONCENTRATIONS PRODUCED
BY ASARCO EMISSIONS
Control
Alternative
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Location of
Maximum Concentration
UTM X
(m)
536,000
536,500
536,000
536,000
536,500
536,000
536,500
536,000
536,500
536,000
536,500
536,000
536,500
536,000
536,500
536,000
536,500
536,500
536,500
536,500
535,500
UTM Y
(m)
5,234,500
5,235,500
5,234,500
5,234,500
5,235,500
5,234,500
5,235,500
5,234,500
5,235,500
5,234,500
5,235,500
5,234,500
5,235,500
5,234,500
5,235,500
5,234,500
5,235,500
5,235,500
5,235,500
5,235,500
5,234,500
Calculated Concentrations
(ppm)
ASARCO
0.42514
0.55083
0.59124
0. 26844
0. 28792
0.37668
0. 04062
0. 05152
0. 04541
0. 05925
0.46187
0. 62958
0.42698
0. 58148
0.36958
0. 50304
0. 26543
0. 23841
0.23251
0.05008
0. 04763
Other
Sources
0. 00088
0.00091
0. 00088
0. 00088
0.00091
0. 00088
0.00091
0. 00088
0. 00091
0. 00088
0. 00091
0. 00088
0.00091
0. 00088
0. 00091
0. 00088
0.00091
0. 00091
0. 00091
0. 00091
0.00091
Combined
Sources
0.42602
0. 55174
0.59212
0. 26932
0. 28883
0.37756
0. 04153
0. 05240
0. 04632
0.06013
0.46278
0. 63046
0.42789
0.58236
0. 37049
0.50392
0.26634
0.23932
0. 23342
0. 05099
0. 04854
83
-------
We point out that the area of high calculated short-term ground-level SO
concentrations for the selected worst-case 24-hour period is west of the existing
SO monitoring network. Thus, the maximum hourly ground-level SO concentrations
2 ^
produced by actual ASARCO emissions during this period probably went undetected.
The PSAPCA monitor at N26 and Pearl, which is on the eastern edge of this area,
measured moderate hourly SO concentrations (0. 05 to 0.28 ppm) during 11 hours
£.1
of this period.
As noted in Section 5.1, our calculations suggest that low-level SO
emissions from Kaiser Aluminum have a greater impact on long-term ambient air
quality than the ASARCO stack emissions. However, the maximum 24-hour SO
£j
concentration calculated for Kaiser Aluminum during the worst-case 24-hour period
is only 0.04 ppm. This follows because the meteorological condition that maximizes
the ground-level concentrations produced by buoyant stack emissions (the critical
wind-speed condition) minimizes the ground-level concentrations produced by low-
level emissions. Light winds in combination with a stable thermal stratification
characteristic of air stagnation comprise the meteorological condition that typically
produces the highest ground-level concentrations from low-level emissions.
5.4 MAXIMUM HOURLY AND 5-MINUTE CONCENTRATIONS
It follows from the preceding discussions that only Control Alternatives
7, 8, 9, 10, 20 and 21 will meet both the National and PSAPCA SO ambient air
£i
quality standards for averaging times greater than one hour. We have assumed
that the Control Alternatives that do not meet all PSAPCA and National ambient
air quality standards for averaging times greater than one hour also do not meet
the 5-minute and hourly PSAPCA standards. Thus, only the six Control Alternatives
cited above are considered in this section.
84
-------
It should be noted that the maximum hourly ground-level SO concentrations
Lt
calculated for the Control Alternatives using the meteorological inputs in Table 5-2
may not necessarily represent the actual maximum hourly concentrations for the
following reasons:
• The density of the 0.5-kilometer grid spacing may not
be adequate to isolate locations of the maximum hourly
ground-level concentrations
• Meteorological conditions that result in the highest
hourly ground-level concentrations may differ from
the persistent moderate-to-strong wind conditions
associated with maximum 3-hour and 24-hour
concentrations
For the six Control Alternatives with the potential to meet all PSAPCA
air quality standards (Control Alternative 7, 8, 9, 10, 20 and 21), we used the
general meteorological inputs given in Section 3.3 and Equation (A-41) in Appendix A
to calculate, for each combination of wind-speed and stability categories, maximum
hourly ground-level SO concentrations. In these calculations, we assumed
2t
the average terrain elevation within Tacoma to be 100 meters (330 feet) above mean
sea level (see Figure 1-1). The annual frequencies of occurrence of the various
i"V\
combinations of wind speed and stability at N26 and Pearl (see Appendix B) were
used to estimate the total number of hours per year of hourly SO concentrations
£
above the PSAPCA hourly standards of 0.40 and 0.25 ppm.
Table 5-5 presents, for the six Control Alternatives, the calculated maximum
hourly ground-level SO concentrations and Table 5-6 presents the estimated number
£i
of hours per year of hourly concentrations above the PSAPCA hourly standards. It
should be noted that two pairs of Control Alternatives (7 and 8; 9 and 10) are identical
except that the existing 172-meter (565-foot) main stack is used for Control
85
-------
TABLE 5-5
CALCULATED MAXIMUM HOURLY GROUND-LEVEL
SO CONCENTRATIONS
Control Alternative
7
8
9
10
20
21
Maximum Hourly SO Concentration
(ppm)
0. 14239
0. 31465
0. 16223
0. 39443
0. 17891
0.17052
86
-------
TABLE 5-6
ESTIMATED NUMBER OF HOURS PER YEAR THAT THE
MAXIMUM GROUND-LEVEL SO2 CONCENTRATION
EXCEEDS THE PSAPCA HOURLY STANDARDS
Control Alternative
7
8
9
10
20
21
Hours Per Year That The Hourly Concentration Exceeds
0. 40 ppm*
0
0
0
0
0
0
0. 25 ppm**
0
948
0
948
0
0
*PSAPCA ambient air quality standard, never to be exceeded.
**PSAPCA ambient air quality standard, not to be exceeded at any given point
more than twice in any 7-day period.
87
-------
Alternatives 7 and 9 while a new 152-meter (500-foot) main stack is used for
Control Alternatives 8 and 10. Also, the base of the new main stack is assumed to
be located 9 meters (30 feet) below the base of the existing stack. Because the
new main stack is shorter and the base is at a lower elevation than the existing stack,
the maximum concentrations calculated for the new main stack are larger than
the corresponding concentrations calculated for the existing main stack.
The results of the calculations presented in Tables 5-5 and 5-6 may be
summarized as follows:
• Control Alternatives 7, 9, 20 and 21 clearly meet all
PSAPCA hourly standards
• The calculated maximum hourly concentrations for
Control Alternatives 8 and 10 are relatively low and
concentrations above 0. 25 ppm are estimated to occur
so infrequently that we believe it is likely that these
Control Alternatives will be in marginal compliance with
the PSAPCA hourly standard of 0. 25 ppm (not to be exceeded
at any given point more than twice in any 7-day period)
The conclusion given above that Control Alternatives 8 and 10 are likely to
be in marginal compliance with the PSAPCA hourly standard of 0. 25 ppm requires
additional explanation. PSAPCA Regulation I permits two hourly SO concentrations
^
above 0. 25 ppm (but less than 0.40 ppm) at any given point during a 7-day period.
Assuming that these excursions are evenly spaced in time, a single point can exper-
ience up to 104 excursions per year without violating the PSAPCA hourly standard
of 0.25 ppm. Table 5-6 lists, for each Control Alternative, the total number of
hours per year that the calculated maximum hourly ground-level SO concentration
A
exceeds 0. 25 ppm. Trial calculations for Control Alternative 1 using the procedures
described in Section A. 6 of Appendix A indicate that no single point will experience
88
-------
more than about 4 percent of this total. It follows that, at the point most frequently
affected by ASARCO emissions, Control Alternatives 8 and 10 will cause excursions
above the 0.25 ppm standard about 40 hours per year. In each case, it is possible
that three of these excursions could occur within a 7-day period. Thus, infrequent
violations of the 0. 25 ppm standard are possible for Control Alternatives 8 and 10.
The only other short-term PSAPCA SO ambient air quality standard
£t
that must be considered is the PSAPCA 5-minute standard of 1.0 ppm, not to be
-1/5
exceeded more than once in any 8-hour period. Applying the t law (Osipov,
1972 and others) to the Control Alternatives listed in Table 5-5, none of the
calculated maximum 5-minute concentrations exceeds 1. 0 ppm. We conclude
that the Control Alternatives that will meet the PSAPCA hourly standards will also
meet the PSAPCA 5-minute standard.
89
-------
90
-------
SECTION 6
EVALUATION OF THE EFFECTIVENESS OF THE ASARCO SCS
6.1 BACKGROUND: THE SCS CONCEPT
A Supplementary Control System (SCS) is a system that curtails stack
pollutant emissions during periods when meteorological conditions conducive to
ground-level concentrations in excess of applicable air quality standards exist or
are anticipated to exist.
The U. S. Environmental Protection Agency (1976) has recently published
an SCS guidelines document describing the basic elements, functions, development
and operation of Supplementary Control Systems. As explained in the EPA
guidelines document, an SCS operates in two different modes:
• An open-loop mode for normal smelter or plant operations in
which a predictive model that quantitatively relates stack
SO emissions, stack parameters and meteorological
&
parameters to maximum short-term ground-level SO
Lt
concentrations is routinely used to determine the emissions
curtailment required to prevent violations of the applicable
air quality standards
• A closed-loop mode used to curtail emissions in emergency
situations when air quality measurements from monitors
located in the vicinity of a smelter or power plant show that
short-term air quality standards are threatened
There are many technical as well as logistical and economic problems
associated with the development and operation of an SCS. These problems stem
in part from the uniqueness of the meteorological, topographical and source
91
-------
parameters associated with each industrial smelter or fossil fuel power plant. The
meteorological parameters include wind circulation patterns and atmospheric
stability conditions controlling the transport and dispersion of stack discharges.
Topographical parameters include prominent terrain features that channel or
block the normal airflow and/or induce vortices that affect the dispersion of
stack plumes, as well as terrain elevations that are significantly higher than the
elevation of the stack base. Source parameters include, in addition to the usual
stack parameters (stack height, exit velocity, exit temperature, effluent composi-
tion and density, and pollutant discharge rate), plant layout of buildings and
other structures that may produce wake effects, as well as process details that
determine the rate of emissions curtailment. Other SCS technical problems
include the development of a predictive model that accurately relates stack and
fugitive emissions to ambient air quality, the design and operation of automated
in-stack measurement systems, air quality monitoring networks, meteorological
measurement networks, as well as automated data-acquisition, data-processing,
data-storage and data-retrieval systems. Additional problems arise from the
fact that the detailed meteorological data, air quality data and emissions/stack
parameter data required for designing an SCS are generally not available.
Consequently, it is usually necessary to revise and update the original SCS
design as the result of the experience gained during a year or more of actual
operation.
6. 2 DESCRIPTION OF THE ASARCO SCS
As stated in the ASARCO Supplementary Control System Operating Manual
(Welch, 1976), the objective of the Tacoma Plant SCS is to maintain an air quality
that meets the local, state and National ambient air quality standards by keeping
a proper balance between the SO emission rate and the meteorological elements
&
of wind and stability that determine the degree of plume dispersion and thus the
92
-------
sulfur dioxide concentration. Development of the ASARCO SCS started in 1970
well before any detailed guidelines such as those recently published by EPA were
available. The rationale for the ASARCO SCS stems principally from the
stringent PSAPCA short-term SO standards, particularly the 5-minute and
£t
1-hour standards, which we believe cannot normally be met with either the
17-percent or 51-percent constant emissions controls. The basic operational
concept of the ASARCO SCS involves the use of meteorological forecasts to
anticipate the times when the main stack plume will be in contact with the ground,
as the result of turbulent mixing in the air layer between the plume stabilization
height and the earth' s surface, and the curtailment of stack emissions and/or the
use of stack heaters sufficiently in advance of these times to prevent the occurrence
of concentrations that approach or exceed the PSAPCA short-term air quality
standards. The required emissions curtailments are quite large. According to
the emission rate guidelines in the ASARCO SCS Operating Manual, they range
from about 50 to 90 percent or more of the uncurtailed emission rate for 51-percent
constant emissions control, depending on meteorological conditions. Because of
the time required to effect the requisite emissions curtailments and the time
before the reduced stack emissions can affect the air quality at the ASARCO or
PSAPCA monitors, there is an inherent time delay in the ASARCO SCS of 20
minutes to 1 hour or more (see Section 6. 3). It follows that the occurrence of
high 5-minute and 1-hour concentrations at the monitors must be anticipated in
advance and appropriate curtailment actions must be taken in advance if violations
of the corresponding air quality standards are to be prevented. These facts
explain the emphasis placed on meteorological forecasting in the routine operation
of the ASARCO SCS and clearly demonstrate that the ASARCO SCS cannot be
operated as a closed-loop system (i. e., by basing curtailment actions on air
quality observations at the monitors) to achieve compliance with the 5-minute
and 1-hour PSAPCA air quality standards. Although there are some exceptions,
the ASARCO SCS appears to be generally capable of preventing violations of the
93
-------
of the National 3-hour and 24-hour SO ambient air quality standards by a closed-
£t
loop mode of operation, especially because of the curtailment actions taken to
attempt to prevent violations of the PSAPCA 5-minute and 1-hour standards.
The ASARCO SCS is staffed by six professional meteorologists. Curtail-
ment decisions are made by the meteorologist on duty at the Emissions Control
Center located in the smelter administration building. According to the SCS
Operating Manual, the normal daily SCS operating routine is divided into three
phases:
(1) Initial Phase, corresponding to the early part of the day
when the ASARCO plume is typically above the surface
mixing layer and not in contact with the ground. In this
phase, a meteorological forecast is made of the:
• Time at which the top of the surface mixing layer
is expected to reach plume level
• Intensity of the initial mixing
• Wind speeds at the surface and at plume level
during the initial mixing period
(2) Intermediate Phase, corresponding to the middle part
of the day when the ASARCO plume is in the surface
mixing layer and is in contact with the ground. In this
phase, predictions are made of the:
• Degree of mixing
• Wind flow at the surface and at plume height
• Stability within the mixing layer
94
-------
(3) Final Phase, corresponding to the latter part of the day
when the surface mixing layer contracts and the ASARCO
plume ceases to mix to the ground. In this phase, curtailment
is reduced and the emission rate increased as the meteoro-
logical indicators show increasing stability and the plume moves
out of the surface mixing layer.
The ASARCO SCS Operating Manual provides, for the first two phases, emission
rate guidelines (expressed as fractions of the uncurtailed rate ranging from about
0.10 to 0. 50) to assist the meteorologist in making curtailment decisions.
According to the SCS Manual, specific emission rates cannot be designated in the
Manual because of the almost infinite number of possible meteorological conditions.
The actual emission rate can thus only be determined by the experienced judgment
of the meteorologist who must base his decision upon his interpretation of the
meteorological conditions. During the Final Phase, the emission rate is increased
as meteorological indicators show increasing stability and all curtailment action
is normally stopped when the tower temperature difference* shows an inversion
greater than 3 degrees Fahrenheit and the insolation is less than 0. 05 gram calories
per square centimeter per minute. The ASARCO SCS Operating Manual also
specifies guidelines for dealing with upset conditions, the occurrence of unexpected
high concentrations at any ambient monitoring stations that are projected to approach
or exceed an air quality standard, air stagnation episodes and the operation of the
stack heaters.
The following meteorological, air quality and smelter operation data are
received at the Emission Control Center by telemetry or teletype:
*The tower in question is not identified in the ASARCO SCS Operating Manual; we
believe it is the 150-foot microwave tower at N26^n and Pearl.
95
-------
Air Quality Data
• SC>2 concentrations from five ASARCO monitors are
telemetered to a multipoint recorder; the monitors are
ASARCO Autometers and operate on a 30-minute cycle
• SC>2 concentrations are received by teletype from
14 PSAPCA SOo monitors; normally, 5-minute average
SC>2 concentrations at each monitor are received every
15 minutes and hourly average and 24-hour running
average SC>2 concentrations at each monitor are received
each hour
• 30-Minute average coefficients of haze (CoHs) at each
PSAPCA monitor site are received by teletype every
half hour and hourly average and 24-hour running average
CoHs are received each hour
Meteorological Data
• Wind speed and wind direction data are telemetered from
all ASARCO SO monitor sites except University Place,
th
from the top of a 150-foot tower at N26 and Pearl,
j.r_
and from a tower located at N48 and Baltimore
• Temperature data are telemetered from sensors located
at the top of the 150-foot tower at N26th and Pearl, the
nearby ASARCO monitor site, the 510-foot level of the
main stack and at the Emissions Control Center; also,
temperature profile data from an acoustic sounder
located outside the Emissions Control Center are
continuously recorded
96
-------
• Wind speed and wind direction data (5-minute and
hourly averages) are received by teletype from all
fifteen PSAPCA wind stations
• The Emission Control Center also has a teletype
receiver for the National Weather Service Circuit
A transmissions, a teletype receiver-transmitter
for a local National Weather Service network and
a facsimile recorder for the NAFAX circuit
Smelter Operation Data
The following in-plant telemetered information is recorded in the
Emissions Control Center
• Converter operations
• Reverberatery furnace feed rate
• Stack heater operations
• Stack gas SC>2 concentrations
• Acid plant and liquid SO2 plant gas flows
6. 3 CRITIQUE OF THE ASARCO SCS
Comparison of the ASARCO SCS with the SCS concept described in the
recently-published EPA guidelines document (U. S. Environmental Protection
Agency, 1976) shows the principal difference to be in the open-loop mode of
operation. The ASARCO SCS does not use a predictive model directly relating
97
-------
the stack SO2 emission rate, stack parameters (stack height, exit velocity or flow
rate and exit temperature) and meteorological parameters to ground-level concen-
trations. Although some attempts have apparently been made to develop a pre-
dictive model for the ASARCO Tacoma smelter, these were not successful. In
place of a predictive model of the type specified in the EPA guidelines, ASARCO
has substituted the meteorological forecasting procedures described above coupled
with the judgment and experience of the meteorologist on duty at the SCS Emissions
Control Center to determine the proper emission rate and stack heater operation
required to prevent violations of the short-term air quality standards. The current
operational procedures of the ASARCO SCS appear to be almost entirely based on
the experience gained from operating the system over the past six years without
recourse to diffusion modeling of the impact of the smelter SO2 emissions.
The combination of this strong empirical base with the high degree of
subjectivity inherent in the meteorological prediction procedures and curtailment
decisions makes an objective analysis of the ASARCO SCS procedures very difficult.
For example, we have had considerable difficulty in reconciling SO0 emission rates
^
for the historical cases used in the selection and testing of short-term diffusion
models (see Section 4.3 and Appendix E) indicated by the curtailment reports
supplied to PSAPCA by ASARCO with similar rates estimated from the in-stack
SO2 monitor readings. Figure 6-1 shows a graph of average hourly SO2 emission
rates for 25 July 1973 estimated from curtailment reports and from in-stack SO2
concentration data using Equation (4-1). Reasonably close agreement, which may
be fortuitous, occurs only for about 6 of the 24 hours. The largest differences
occur during the early morning and nighttime hours when the percent curtailment
is small or zero and during the middle of the day when the percent curtailment
is 90 percent or larger. We interpret Figure 6-1 to mean that there is no accurate
method for determining the actual SO emission rate from the ASARCO curtailment
Li
reports and that the emission rate for 100 percent curtailment, prior to the liquid
SO plant, was approximately 4, 000 pounds of SO per hour.
98
-------
I04
LJ
t£
2
g
CO
5
LJ
CM
O
CO
LJ
O
cr
LJ
< I03
o
i
I02
25 JULY 1973
L-
L_J
= Q ESTIMATED FROM CURTAILMENT REPORTS
= Q ESTIMATED FROM S02 CONCENTRATIONS
I
I
I
I
10 12 14 16
HOUR (PDT)
18 20 22
24
FIGURE 6-1. Comparison of hourly average SO2 emission rates estimated from
curtailment reports and from in-stack SO_ concentration data
for 25 July 1973.
99
-------
Other features of the ASARCO SCS that merit discussion are the density
of the existing SO air quality monitoring network, the inherent time delay in
implementing curtailment decisions and the effectiveness of the stack heaters
in reducing ground-level SO concentrations. Although air quality data are
£t
telemetered to the ASARCO Emissions Control Center from five ASARCO monitors
and fourteen PSAPCA monitors, only three of the PSAPCA monitor sites (N26
and Pearl, Meeker and Maury Island) are normally affected by the ASARCO SO
stack emissions. ASARCO and PSAPCA monitors are collocated at the N26 and
Pearl and Meeker sites. Thus, telemetered air quality data are available for
only six monitor locations routinely affected by the ASARCO plume. There are
sound reasons for believing that an increase in the number of SO monitors,
^
especially if additional monitors were located 1 to 2.5 kilometers south-southwest
and southwest of the main stack, would result in observations of high short-term
ground-level concentrations at these locations produced by low-level fugitive
emissions and also by stack emissions during periods of moderate-to-strong winds.
Although it is admittedly speculative, we believe an increase in the density of
the existing SO2 monitoring network along the lines outlined above would result in:
• Appropriate revisions in the current ASARCO SCS
operating procedures
• A significant increase in the number of violations
of the PSAPCA 5-minute and 1-hour air quality
standards
The time normally required to reduce smelter stack emissions in response
to a specific curtailment request appears to vary from about 10 to 30 minutes.
The additional time required for the reduced stack emissions to begin to change
ground-level SO concentrations at the monitors depends on the wind speed, wind
^
d irection and stability conditions. This additional time delay typically varies
from a minimum of 10 minutes to 1 hour or more. We conclude that the ASARCO
100
-------
SCS thus has an inherent characteristic time delay of 20 minutes to 1 hour or more,
which includes both the time required to implement a curtailment decision and
the time required for the reduced emissions to start to affect the air quality at
the SC> monitor sites located closest to the ASARCO smelter.
O
Operation of one or both of the stack heaters is an important feature of
the ASARCO SCS. We have calculated the increase in plume rise resulting from
the operation of the stack heaters by using the current stack and emissions data
for 51-percent constant emissions control in Table 2-1 in the plume rise equations
presented in Section A. 2 of Appendix A. These calculations show that the use of
one stack heater increases plume rise by about 14 percent and the use of two stack
heaters increases plume rise by about 19 percent. To a first approximation,
these percentage increases in plume rise will result in almost equivalent percentage
reductions in the maximum short-term ground-level SO concentrations. Specifi-
&
cally, the maximum short-term ground-level concentration is approximately
inversely proportional to the square of the sum of the stack height and the buoyant
plume rise. For mean wind speeds of 3 or 4 meters per second, the buoyant
plume rise without stack heaters is approximately equivalent to the actual stack
height of 172 meters (565 feet). With both stack heaters in operation, the buoyant
plume rise is increased 19 percent and is thus equal to 205 meters (672 feet). By
adding the stack height to the buoyant plume rise, squaring each sum and taking
the ratio of the squared sums, one obtains a reduction in the maximum ground-
level concentration of about 20 percent due to the operation of both stack heaters.
Thus, the effect of the operation of the stack heaters on ground-level concentra-
tions is relatively small except when the enhancement in plume rise is sufficient
to cause the plume to stabilize above the top of the surface mixing layer.
101
-------
6.4 EFFECTIVENESS OF THE ASARCO SCS
Since the ASARCO Tacoma SCS began operation in 1970, there has been
a significant improvement in the ambient air quality at the locations of the
PSAPCA and ASARCO SO monitors. This improvement is illustrated by
^
Tables 6-1 and 6-2 which show, for the period 1968 to present, the number of
violations at the PSAPCA SO monitor at N26 and Pearl of the National and
PSAPCA ambient air quality standards, respectively. The N26 and Pearl
monitor is the PSAPCA monitor most frequently affected by ASARCO emissions
and it has been in operation longer than any other PSAPCA monitor. Table 6-1
shows that no violations of the 3-hour and 24-hour National standards occurred
•f"l"j
at N26 and Pearl after 1971 until the first part of 1976 when the 3-hour
standard was violated. Since 1971, there have been a total of three observations
(one in 1974 and two in the first part of 1976) above the 3-hour standard and one
observation (in the first part of 1976) above the 24-hour standard. Table 6-2
shows that the number of violations of the PSAPCA short-term standards decreased
significantly through 1973. ASARCO (Welch, 1975) attributes the majority of the
1974 violations to start-up and shake-down problems with the liquid SO plant.
£1
However, the entries in Table 6-2 indicate the number of violations has remained
fairly constant since 1973 and there is a good possibility that the number of
violations during 1976 will exceed those for 1975 and 1974.
We believe that Tables 6-1 and 6-2 demonstrate that the ASARCO SCS
reached the limit of effectiveness during 1973 and 1974 in terms of its capability
to prevent violations of both the National and PSAPCA ambient air quality standards.
With respect to the National standards, the effectiveness of the ASARCO SCS is
very high although, as shown by the experience in early 1976, it is incapable of
preventing all violations of the 3-hour and possibly the 24-hour standards. Table
6-2 shows that all of the PSAPCA short-term standards are still being violated,
especially the 5-minute and 1-hour standards.
102
-------
TABLE 6-1
TOTAL NUMBER OF VIOLATIONS OF THE NATIONAL AIR QUALITY
STANDARDS FOR SO AT N26th AND PEARL
Year
1976**
1975
1974
1973
1972
1971
1970
1969
1968
Number of Violations*
3 -Hour Secondary Standard
1 (2)
0 (0)
0 (1)
0 (0)
0 (0)
1 (2)
5 (6)
5 (6)
0 (1)
24-Hour Primary Standard
0 (1)
0 (0)
0 (0)
0 (0)
0 (0)
1 (2)
1 (2)
4 (5)
1 (2)
*Total number of observations above the standard are enclosed by parentheses
**As of 6 February 1976
103
-------
TABLE 6-2
TOTAL NUMBER OF VIOLATIONS OF THE PSAPCA AIR
QUALITY STANDARDS FOR SO2 AT N26* AND PEARL
Year
1976*
1975
1974
1973
1972
1971
1970
1969
1968
Number of Violations
5 -Minute
Standard of
1. 0 ppm
10
8
10
16
27
39
169
199
124
1-Hour
Standard of
0.40 ppm
10
6
6
4
11
20
62
84
56
1-Hour
Standard of
0. 25 ppm
15
35
21
27
34
52
140
219
125
24 -Hour
Standard of
0. 10 ppm
2
0
1
0
0
3
4
15
10
*As of 6 February 1976
104
-------
As explained in Sections 6. 2 and 6.3, because of the delay time of
20 minutes to 1 hour or more inherent in the ASARCO SCS and because of the
practical impossibility of anticipating all of the times that the plume will mix to
the ground, it is most unlikely that any further improvement in ambient air
quality can be achieved (with the present source configuration) without a substantial
reduction in the maximum allowable emission rate. It is interesting in this
regard to note that the addition of the liquid SO plant in 1974 and the increase
Li
in the constant emissions control from 17 to 51 percent appears to have had no
effect on the number of violations at the N26 and Pearl monitor of either the
National or PSAPCA air quality standards. We believe a completely effective
SCS would require that the smelter operate most of the time at a curtailed rate
which would not differ appreciably from the constant emission rate in Section 5.3
of 2000 to 2500 pounds of SO per hour given for the Control Alternatives that met
£t
the PSAPCA 1-hour standard of 0. 25 ppm. Consequently, we doubt that an SCS
is a viable method of maintaining the 5-minute and 1-hour PSAPCA standards
for the present ASARCO smelter configuration or for any of the alternative
smelter configurations (Control Alternatives) considered in this report that
endanger these standards.
105
-------
106
-------
SECTION 7
CONCLUSIONS AND RECOMMENDATIONS
7.1 CONCLUSIONS
The principal conclusions of this study of the impact on ambient air quality
of SO2 emissions from the ASARCO-Tacoma smelter are:
1. Without the use of Supplementary Control System (SCS) techniques,
SO2 emissions from the present smelter source configuration with 51-percent
constant emissions control would result in violations of all National, PSAPCA and
Washington DOE short-term (5-minute, 1-hour, 3-hour and 24-hour) SO£ air
quality standards; the National, PSAPCA and Washington DOE annual standards
as well as the 30-day PSAPCA standard would not be violated.
2. The percent constant emissions control required, without the use of
SCS techniques, to prevent violations of the short-term air quality standards by
SO2 emissions from the ASARCO-Tacoma smelter depends on the source config-
uration. For the 20 hypothetical source configurations (Control Alternatives)
evaluated in this study, the maximum allowable SO2 emission rate consistent with
meeting all PSAPCA and Washington DOE short-term standards varies from about
2,000 to 2, 500 pounds of SO2 per hour. Based on an assumed current average
uncurtailed emission rate of 24, 800 pounds of SO2 per hour for the present 51-
percent constant emissions control, the above rates of 2, 000 to 2, 500 pounds per
hour respectively represent 96- and 95-percent constant emissions control. These
maximum allowable emission rates will also ensure compliance with all National
SO2 air quality standards.
3. Our evaluation of the effectiveness of the ASARCO SCS in meeting the
National short-term SO2 air quality standards, which is supported by air quality
observations from the PSAPCA monitors, shows that it is capable of preventing
107
-------
most violations of the National 3-hour and 24-hour standards. In part, the past
success of the ASARCO SCS in meeting the National short-term standards is
attributable to the stack heater operations and emissions curtailment activities
arising from attempts to meet the more stringent PSAPCA short-term standards.
4. Our evaluation of the effectiveness of the ASARCO SCS in meeting the
PSAPCA and Washington DOE short-term air quality standards, which is also
supported by air quality observations from the PSAPCA monitors, shows that it is
not a viable means of preventing violations of these standards, especially the 5-
minute and 1-hour standards. The fundamental difficulty in relying on SCS
techniques to meet these standards is the minimum delay of 20 minutes to 1 hour
or longer inherent in the ASARCO SCS from the time a curtailment decision is made
until the time that the implemented curtailment decision can significantly affect the
air quality at the closest SO2 monitors. This means that curtailment decisions
cannot be made on the basis of telemetered data from the air quality monitors but
must instead always be made, using meteorological forecasting techniques, at
least 20 minutes to 1 hour in advance of any occurrence of high ground-level SO
£t
concentrations. Given the large percentage curtailment required (40 to 90 percent
of the average uncurtailed emission rate for full production with 51-percent constant
emissions control) and the large uncertainties inherent in meteorological predictions,
we believe it clearly impossible for ASARCO to achieve the nearly-perfect fore-
casting capability required in using any SCS techniques to prevent violations of the
5-minute and 1-hour PSAPCA standards.
5. For the reasons outlined above, we believe that the only way in which
the PSAPCA 5-minute and 1-hour SO2 standards can be met with the present source
configuration is to limit the maximum stack emission rate to about 2,500 pounds of
SO2 per hour, which is approximately equivalent to the maximum allowable constant
emission rate for the Control Alternatives that meet all the National, PSAPCA and
Washington DOE air quality standards for SO2-
108
-------
7.2 RECOMMENDATIONS
During our study of the air quality impact of SO emissions from ASARCO
Z
and other major SO sources in the Seattle-Tacoma area, we identified three
£
problem areas that we believe require further investigation and/or action. We
therefore recommend that:
1. The air quality impact of low-level fugitive SO emissions from
^
the ASARCO smelter as well as the impact of possible downwash effects from the
main ASARCO stack should be further investigated by placing additional SO2
monitors closer to the smelter than the existing monitors. The existing PSAPCA
and ASARCO SO monitors are located too far from the smelter to detect these
£t
effects. The EPA SO monitor that was located approximately 1.2 kilometers
£i
south-southwest of the main stack from 18 September 1975 through 27 January 1976
(see Appendix D) measured high hourly SO concentrations. However, from the
Lt
available emissions and meteorological data, we are not able to isolate the
contributions of low-level fugitive emissions and stack downwash effects to these
observations.
2. The smelter operation reports submitted by ASARCO to PSAPCA
should include hourly average values of stack exit temperature, actual stack
volumetric flow rate, in-stack SO concentration and the stack SO9 emission
2 ^
rate in pounds per hour. If possible, the curtailment requests should be reported
in terms of maximum allowable SO emission rates in pounds of SO per hour or
in terms that can easily be converted to emission rates in pounds of SO per hour
^
by regulatory agency personnel.
3. The air quality impact of low-level SO emissions from the
z
Kaiser Aluminum Plant in Tacoma should be investigated by diffusion modeling
and/or air quality monitoring to determine whether emissions from this source
are endangering the National and PSAPCA air quality standards. Diffusion-model
109
-------
calculations made during this study indicate that emissions from Kaiser Aluminum
may be endangering these standards. A detailed investigation of the air quality
impact of SO emissions from Kaiser Aluminum was beyond the scope of the
^
present study.
110
-------
REFERENCES
American Society of Mechanical Engineers, 1973: Recommended guide for the pre-
diction of the dispersion of airborne effluents, Second Edition. ASME
Air Pollution Control Division, New York, N. Y.
Briggs, G. A., 1965: A plume rise model compared with observations. Journal of
the Air Pollution Control Association, 15, 433.
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 Pro-
ceedings of the Second International Clean Air Congress, Academic Press,
New York.
Briggs, G. A., 1972: Chimney plumes in neutral and stable surroundings. Atm.
Env., 6(7), 507-510.
Briggs, G. A., 1973: Diffusion estimates for small emissions. ATDL Contribution
File No. (Draft) 79, Air Resources Atmospheric Turbulence and Diffusion
Laboratories, Oak Ridge, Tennessee.
Bringfelt, B., 1968: Plume rise measurements at industrial chimneys. Atmos-
pheric Environment, 2(6), 575-598.
Calder, K. L., 1971: A climatological model for multiple source urban air pollution.
Proc. 2nd Meeting of the Expert Panel on Air Pollution Modeling, NATO
Committee on the Challenges of Modern Society, Paris, France, July 1971,
33.
Carlson, P. E. and J. A. Scuderi, 1974: Meteorological study for Tooele Army
Depot. DPG-FR-C965A-I, U. S. Army Dugway Proving Ground, Dugway,
Utah.
Cramer, H. E., etal., 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, Utah.
Cramer, H. E. and J. F. Bowers, 1974: Calculated impact on ambient air quality
of current and projected SO2 and fluoride emissions from the Agricultural
Products Corporation Plant at Conda, Idaho. H. E. Cramer Company,
Inc. Technical Report TR-74-103-01 prepared for U. S. Environmental
Protection Agency, Research Triangle Park, N. C.
Ill
-------
Cramer, H. E., H. V. Geary and J. F. Bowers, 1975: Diffusion-model calculations
of long-term and short-term ground-level SC>2 concentrations in Allegheny
County, Pennsylvania. EPA 903/9-75-018, U. S. Environmental Protection
Agency, Region III, Philadelphia, Pennsylvania.
Carpenter, S. B., T. L. Montgomery, J. M. Leavitt, W. C. Colbaugh and F. W.
Thomas, 1971: Principal plume dispersion models: TVA power plants.
Journal of the Air Pollution Control Association. 21/8), 491-495.
DeMarrais, G. A., 1959: Wind speed profiles at Brookhaven National Laboratory.
J. Met., 16, 181-190.
Eliassen, A. and J. Saltbones, 1975: Decay and transformation rates of SO2, as
estimated from emissions data, trajectories and measured air concen-
trations. Atmospheric Environment, 9_, 425-429.
Environmental Data Service, 1967: Tabulation I, frequency of occurrence, average
wind speed through mixing depth for Seattle, Washington. Job No. 6234,
National Climatic Center, Federal Building, Asheville, North Carolina.
Environmental Data Service, 1968: Climatic Atlas of the United States. U. S.
Department of Commerce, Washington, D. C.
Environmental Protection Agency, 1969: Air Quality Display Model. Prepared by
TRW Systems Group, Washington, D. C., available as PB 189-194 from
the National Technical Information Service, Springfield, Virginia.
Environmental Protection Agency, 1976: Guidelines for evaluating supplementary
control systems. Prepared by Monitoring and Data Analysis Division,
Research Triangle Park, North Carolina, available as EPA-450/2-76-003
from the Air Pollution Technical Information Center, EPA, Research
Triangle Park, North Carolina.
Fay, J. A., M. Escudier and D. P. Hoult, 1970: A correlation of field observations
of plume rise. Journal of the Air Pollution Control Association, 20 (6),
391-397.
Luna, R. E. and H. W. Church, 1971: A comparison of turbulence intensity and
stability ratio measurements to Pasquill turbulence types. Paper pre-
sented at a Conference on Air Pollution Meteorology, Raleigh, N. C.,
April 5-9, 1971.
Lyons, W. H. and H. S. Cole, 1973: Fumigation and plume trapping on the shores
of Lake Michigan during stable onshore flow. Journal of Applied Meteo-
ology, 12_, 494.
112
-------
Osipov, Y. S., 1972: Diffusion from a point source of finite time of action. In
AICE Survey of USSR Air Pollution Literature - Volume PCX, distributed
by National Technical Information Service, Springfield, Virginia.
Pasquill, F., 1961: The estimation of the dispersion of windborne material.
Met. Mag., 90, 33-49.
Pasquill, F., 1962: Atmospheric Diffusion. D. Van Nostrand Co., Ltd.,
London, 297.
Pasquill, F., 1974: Atmospheric Diffusion (Second Edition). Ellis Horwood Limited,
Sussex, England, 420.
Turner, D. B., 1964: A diffusion model for an urban area. J. Appl. Meteor.,
3 (1), 83-91.
Turner, D. B., 1969: Workbook of Atmospheric Dispersion Estimates.
PHS Publication No. 999-AP-26. U. S. Department of Health, Education
and Welfare, National Air Pollution Control Administration, Cincinnati, Ohio.
Weisenberg, I. J. and J. C. Serne, 1976: Design and operating parameters for
emission control studies: ASARCO, Tacoma copper smelter. EPA Report
No. EPA-600/2-76-036k prepared for U. S. Environmental Protection
Agency, Office of Research and Development, Washington, D. C.
Welch, R. E., 1975: Intermittent control of sulfur dioxide emissions from a copper
smelter. Paper presented at the Annual Meeting of the Pacific Northwest
International Section of the APCA, Vancouver, British Columbia,
November 19-21, 1975.
Welch, R. E., 1976: Supplementary control system operating manual. ASARCQ
Technical Note submitted to Puget Sound Air Pollution Control Agency,
Seattle, Washington.
113
-------
114
-------
APPENDIX A
MATHEMATICAL MODELS USED TO CALCULATE
GROUND-LEVEL CONCENTRATIONS
A. 1 INTRODUCTION
The computerized diffusion models described in this appendix fall into
two general categories: (1) Short-term models for calculating time-averaged
ground-level concentrations for averaging times of 1, 3, 8 and 24 hours; (2) Long-
term models for calculating seasonal and annual ground-level concentrations.
Both the short-term and long-term concentration models are modified versions
of the Gaussian plume model for continuous sources described by Pasquill (1962).
In the short-term model, the plume is assumed to have Gaussian vertical and
lateral concentration distributions. The long-term model is a sector model similar
in form to the Environmental Protection Agency's Climatological Dispersion Model
(Calder, 1971) in which the vertical concentration distribution is assumed to be
Gaussian and the lateral concentration distribution within a sector is rectangular
(a smoothing function is used to eliminate sharp discontinuities at the sector
boundaries). The cr vertical expansion curves and the cr lateral expansion
curves are determined by using turbulent intensities in simple power-law expres-
sions that include the effects of initial source dimensions. In both the short-term
and long-term models, buoyant plume rise is calculated by means of the Briggs
(1971) plume-rise formulas. An exponent law is used to adjust the surface wind
speed to the source height for plume-rise calculations and to the plume stabiliza-
tion height for concentration calculations. Both the short-term and the long-term
models contain provisions to account for the effects of complex terrain.
Table A-l lists the hourly meteorological inputs required by the short-
term concentration model. Lateral and vertical turbulent intensities a! and a'
A E
A-l
-------
TABLE A-l
HOURLY METEOROLOGICAL INPUTS REQUIRED BY THE
SHORT-TERM CONCENTRATION MODEL
Parameter
Definition
u
R
TA
o"
E
H
m
dz
Mean wind speed at height z
R
Mean wind direction at height z
R
Wind-profile exponent
Wind azimuth-angle standard deviation in radians
Wind elevation-angle standard deviation in radians
Ambient air temperature ( K)
Depth of surface mixing layer
Vertical potential temperature gradient
A-2
-------
may be directly specified or may be assigned on the basis of the Pasquill stability
category. The Pasquill stability category is determined from surface weather
observations using the Turner (1964) wind-speed and solar-index values. Mixing
depths may be obtained from rawinsonde or pibal measurements, or they may be
assigned on the basis of tabulations of the frequency of occurrence of wind speed
and mixing depth (available from the National Climatic Center for synoptic rawin-
sonde stations). Potential temperature gradients may be measured or assigned
on the basis of climatology.
Table A-2 lists the meteorological inputs required by the long-term con-
centration model. Joint-frequency distributions of wind-speed and wind-direction
categories according to the Pasquill stability categories may be obtained from the
National Climatic Center. Alternately, surface wind observations may be analyzed
to generate wind-frequency distributions by time-of-day categories (night, morning,
afternoon and evening). Vertical turbulent intensities may be determined from a
climatology of actual measurements or may be assigned on the basis of the Pasquill
stability categories. Median mixing depths may be determined from the seasonal
tabulations of the frequency of occurrence of wind speed and mixing depth. Vertical
potential temperature gradients may be assigned to stability or time-of-day cate-
gories on the basis of climatology.
A. 2 PLUME-RISE FORMULAS
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 buoyant rise Ah is given by
/ \. t Ir\
(lOh)2/3 f (A-l)
u{h}
A-3
-------
TABLE A-2
METEOROLOGICAL INPUTS REQUIRED BY THE
LONG-TERM CONCENTRATION MODEL
Parameter
Definition
fi i k t (Table)
!> J j K» ''
R
p (Table)
K., 1
(Table)
a;M
90
dz
H .
(Table)
(Table)
v '
(Table)
Frequency distribution of wind-speed and wind-
direction categories by stability or time-of-day
categories for the a, season
Height at which wind-frequency distributions
were obtained
Wind-profile exponents for each stability or
time-of-day category and i wind-speed
category
Standard deviation of the wind-elevation angle in
radians for the i wind-speed category and k
stability or time-of-day category
Ambient air temperature for the k stability
or time-of-day category and I* season
Vertical potential temperature gradient for the
{th wind-speed category and k"-" stability or
time-of-day category
Median surface mixing depth for the i wind-
speed category, k^ stability or time-of-day
category and $, season
Mean wind speeds at height z
R
A-4
-------
where the expression in the brackets is from Briggs (1971; 1972) and
u{h} = the mean wind speed at the stack height h
y = the adiabatic entrainment coefficient ~ 0. 6
F = the initial buoyancy flux
(A-2)
V = the volumetric emission rate of the stack
2
= TT r w
r = inner radius of stack
w = stack exit velocity
g = the acceleration due to gravity
T = the ambient air temperature ( K)
£t
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
e —
I ; u{h) < w/1.5
3w - 3u|h}\
w
0
I ; w/1.5 < u{h) < w
; u{h} 2: w
(A-3)
The empirical correction factor f is generally not applied to stacks with Froude
numbers less than about unity. The corresponding Briggs (1971) rise formula
for a stable atmosphere (potential temperature gradient greater than zero) is
A-5
-------
6F
u(h} y 2 S
1/3
3F / /10S1/2hV)
r , 2 I1 "UUD\ u(h} //
_u(h} y2 S \ N ^> X/J
-1 /2
; TT u{h} S < lOh
1/3
-1 /2
; TT u{h} S > lOh
Vf (A-4)
where
y = the stable entrainment coefficient ~ 0. 66
s = -S_ £i
Ta 9Z
80
8z
= vertical potential temperature gradient
The entrainment coefficients y and y are based on the suggestions of Briggs
1 ij
(1972). It should be noted that Equation (A-4) does not permit the calculated stable
rise Ah to exceed the adiabatic rise Ah as the atmosphere approaches a neutral
stratification (dd/dz approaches o) . A procedure of this type is recommended
by Briggs (1972).
A.3
SHORT-TERM CONCENTRATION MODEL
A. 3.1 Elevated Sources
The atmospheric dispersion model used to calculate hourly
average ground-level concentrations downwind from an elevated continuous source
is given by
x{x,y} =
KQ
TT u{ H} a cr
y z
(Vertical Term} (Lateral Term} (Decay Term} (A-5)
A-6
-------
where
K = scaling coefficient to convert input parameters to dimensionally
consistent units
Q = source emission rate
U{H} = mean wind speed at the plume stabilization height H
a CT = standard deviations of the lateral and vertical concentration
V 7
distributions at downwind distance x
The Vertical Term refers to the plume expansion in the vertical
or z direction and includes a multiple reflection term that limits cloud growth to
the surface mixing layer.
{Vertical Term) = < exp
exp
n=l
/2n H - H'
11 m
2 \ or
exp
2n H +H1
m
a
(A-6)
where H is the depth of the surface mixing layer. The exponential terms in the
m
infinite series in Equation (A-6) rapidly approach zero near the source. At the
downwind distance where the exponential terms are non-zero for n equal 3, the
plume has become approximately uniformly mixed within the surface mixing layer.
In order to shorten computer computation time, Equation (A-6) is changed to the
form
/i7cr7
(Vertical Term} =
2H
(A-7)
m
beyond this point. Equation (A-7) changes the form of the vertical concentration
distribution from Gaussian to rectangular. If H exceeds H , the vertical term is
m
set equal to zero which results in a zero value for the ground-level concentration.
A-7
-------
The Lateral Term refers to the crosswind expansion of the
plume and is given by the expression
i- 2
(Lateral Term} = exp - - f ^- J
I \ v /
(A-8)
where y is the crosswind distance from the plume centerline to the point at which
concentration is calculated.
The Decay Term, which accounts for the possibility of pollutant
removal by physical or chemical processes, is of the form
(Decay Term} = exp [- $ X/U(H}] (A-9)
where
the washout coefficient A (sec ) for precipitation scavenging
0 692
—; , where T , is the pollutant half life for physical or
1/2 chemical removal
0 for no depletion (ip is automatically set to zero by the computer
program unless otherwise specified)
In the model calculations, the observed mean wind speed u is
R
adjusted from the measurement height z to the source height h for plume-rise
calculations and to the stabilization height H for the concentration calculations by
a wind-profile exponent law
u(z} = U-(ZR} ( ^
\ R
A-8
-------
The exponent p, which is assigned on the basis of atmospheric stability, ranges
from about 0. 1 for very unstable conditions to about 0.4 for very stable conditions.
According to the derivation in the report by Cramer, et al_.
(1972), the standard deviation of the lateral concentration distribution a is given
by the expression
cr {x} = cr' x
y A ry
x + x - x (1-cv)
V ryv
ax
ry
a
(A-ll)
x =•<
y
ry \ x a'
3 \ ry A
a
R
"
T — x
A '
. JH
(A-12)
where
a' = the standard deviation of the wind-azimuth angle in radians
x = distance over which rectilinear plume expansion occurs
downwind from an ideal point source (~ 50 meters)
cr = the standard deviation of the lateral concentration distribution
at downwind distance x^
a - the lateral diffusion coefficient (~0.9)
The virtual distance x is not permitted to be less than zero. The lateral turbulent
y
intensity CT ' may be specified directly or may be assigned on the basis of the
Pi.
Pasquill stability category.
Following the derivation of Cramer, et aL (1972) and setting
the vertical diffusion coefficient /3 equal to unity, the standard deviation of the
vertical concentration distribution a is given by the expression
A-9
-------
cr {x} = a' (x + x
z1 J E \ z
(A-13)
where
x
z
"
zR
(A-14)
a
zR
- standard deviation of the wind-elevation angle in radians
= the standard deviation of the vertical concentration distribution
at downwind distance x
The vertical turbulent intensity cr' may also be obtained from direct measure-
hi
ments or may be assigned according to the Pasquill stability categories. When cr'
E
values corresponding to the Pasquill stability categories are entered in Equation
(A-13), the resulting curves will differ from the corresponding Pasquill-Gifford
curves in that Equation (A-13) assumes rectilinear expansion at all downwind dis-
tances. Thus, a values obtained from Equation (A-13) will be smaller than the
z
values obtained from the Pasquill-Gifford A and B curves and larger than the
values obtained from the D, E and F curves at long downwind distances. However,
the multiple reflection term in Equation (A-6) which confines the plume to the
surface mixing layer accounts for the behavior of the D, E and F curves (decrease
in the expansion rate with distance) in a manner that may be related to the
meteorology of the area.
Following the recommendations of Briggs (1972), the lateral
and vertical standard deviations of a stabilized buoyant plume are defined by
<7 = cr
yR zR
0.5 Ah
2.15
(A-15)
A-10
-------
The downwind distance to stabilization x is given by
XR
J
lOh ; - ^ 0
TT u{h} S~1/2 ; |? > 0 and TT u{h} S~1/2 < lOh
Q /J _ —1 /2
lOh ; -r2 > 0 and TT u h S > lOh
az
(A-16)
A. 3.2 Application of the Short-Term Model to Low-Level Emissions
The short-term diffusion model in Section A. 3.1 may be used to
calculate ground-level concentrations resulting from low-level emissions such as
losses through building vents. These emissions are rapidly distributed by the
cavity circulation of the building wake and quickly assume the dimensions of the
building. Ground-level concentrations are calculated by setting the release height
h and the buoyancy parameter F equal to zero. The standard deviation of the
lateral concentration distribution at the source a is defined by the building
crosswind dimension y divided by 4.3. The standard deviation of the vertical
o
concentration distribution at the source cr is obtained by dividing the building
zo
height by 2.15. The initial dimensions cr and cr are assumed to be applicable
& J yo zo
at the downwind edge of the building. It should be noted that separate turbulent
intensities cr' and cr' may be defined for the low-level sources to account for the
A IL
effects of surface roughness elements and heat sources.
A. 3. 3 Short-Term Concentration Model for Area Sources
The atmospheric dispersion model used to calculate ground-
level concentrations at downwind distance x from the downwind edge of an area
source is given by the expression
A-ll
-------
(Vertical Term}
(A-17)
where
(Lateral Term} (Decay Term}
Q = area source strength in units of mass per unit time
y = crosswind source dimension
o
' x
E o
In
"E(X+Xo) + h
; x < 3 x
a' (x+x /2)+h ; x > 3x
EV o/ ' o
(A-18)
x - alongwind dimension of the area source
o
h = the characteristic height of the area source
The Vertical Term for an area source is given by
(Vertical Term} =
,exp
n=l
2H
m
; exp
n /6H '
1J m
~ 2 V o-
\ z .
2-1
= 0
exp
i/.
2
6H
2-1
m
C7
A-12
-------
The Lateral Term is given by the expression
{Lateral Term} = { Vertical Term}
(A-22)
x' = distance downwind from the upwind edge of the area source
A. 4
LONG-TERM CONCENTRATION MODEL
A. 4.1 Elevated Sources
The atmospheric dispersion model for elevated point and volume
sources is similar in form to the Air Quality Display Model (Environmental Pro-
tection Agency, 1969) and the Climatological Dispersion Model (Calder, 1971). In
the model, the area surrounding a continuous source of pollutants is divided into
A-13
-------
sectors of equal angular width corresponding to the class intervals of the
seasonal and annual frequency distributions of wind direction. The emission
rate during a season or year is partitioned according to the relative wind-
direction frequencies. Ground-level concentration fields for each source are
translated to a common reference coordinate grid system and summed to obtain
the total due to all emissions. For a single source, the mean seasonal concentra-
tion at a point (r, 9) is given by
2KQ
--*
. I
a
(A-23)
V.
- exp
where
exp
;i, k,f, •
exp
2nH
- H
(A-24)
= frequency of occurrence of the i wind-speed category,
' j wind-direction category and k stability or time-of-
day category for the ^ season
A#' - the sector width in radians
= a smoothing function
A-14
-------
s{0) =
A0' - 10! - 0'
.1 J
A0'
0
;
;
0!-0-
0! - 0'
3
== A0'
> A01
(A-25)
3' =
the angle measured in radians from north to the centerline
of the th wind-direction sector
0' = the angle measured in radians from north to the point (r, 0)
As with the short-term model, the Vertical Term given by
Equation (A-24) is changed to the form
V.
2H
(A-26)
when the exponential terms in Equation (A-24) become non-zero for n equal 3.
The remaining terms in Equations (A-23) and (A-24) are identical to those pre-
viously defined in Section A. 3.1 for the short-term model, except that the turbulent
intensities and potential temperature gradients may be separately assigned to each
wind-speed and/or stability (or time-of-day) category; the ambient air temperatures
may be separately assigned to each stability (or time-of-day) category for each
season; and the surface mixing depths may be separately assigned to each wind-
speed and/or stability (or time-of-day) category for each season.
As shown by Equation (A-25), the rectangular concentration distribution
within a given angular sector is modified by the function S{0) which smoothes dis-
continuities in the concentration at the boundaries of adjacent sectors. The center-
line concentration in each sector is unaffected by contributions from adjacent
sectors. At points off the sector centerline, the concentration is a weighted
A-15
-------
function of the concentration at the centerline of the sector in which the calculation
is being made and the concentration at the centerline of the nearest adjoining
sector.
The mean annual concentration at the point (r, 9) is calculated from the
seasonal concentrations using the expression
A. 4.2 Application of the Long-Term Model to Low-Level Emissions
Long-term ground-level concentrations produced by low-level
emissions are calculated from Equation (A-23) by setting the source height h and
the buoyancy parameter F equal to zero. The standard deviation of the vertical
concentration distribution at the downwind edge of the building u is defined as
z o
the building height divided by 2.15. Separate vertical turbulent intensities cr' may
E
be defined for the low-level sources to account for the effects of surface heat
sources and roughness elements. A virtual point source is used to account for the
initial lateral dimension of the source in a manner identical to that described below
for area sources.
A. 4. 3 Long-Term Concentration Model for Area Sources
The mean seasonal concentration at downwind distance r with
respect to the center of an area source is given by the expression
A-16
-------
2KQ
iTrR A0
itJi
u.{h} . ,
i z;i, k
V
i,M
(A-28)
exp -
' - ro)/u.{h}J
where
R = radial distance from the virtual point source to the receptor
1/2
- ((r' +x )2 +y2)
r1 =
r =
o
y =
x -
distance from source center to receptor, measured along the
plume axis
effective source radius
lateral distance from the cloud axis to the receptor
virtual distance
r cot
O
(A-29)
z;i,k
2o-,
E;i,k o
In
a' . (r' + r } +h
E;i,k\ ol
ff' . . fr'-r }+h
. E;i,kV o) J
-' . . r'+h
E;i,k
r < r'
o
6r
; r' > 6r
(A-30)
A-17
-------
V.
1+2
o
E
'2nH
exp
'•i k
J j if IV
•V/27T 0" . ,
z;i.k
2H . , .
2-1
6H
5 exp
; exp
z;i,k
- 0
MA-31)
and the remaining parameters are identical to those previously defined.
For points interior to the area source, the concentration for
seasonal models is given by the expression
/
X (r ^
2K Q
— ^
4 y
o Jo
E
f
i.i.k,^ .
_ r. -, . in
u.{h} o-' .
1L J E;i,k
r+h]
»' , fc + h
L t;i, k J
i, k,|
(A-32)
where
r" = the downwind distance, measured along the plume axis from the
upwind edge of the area source
A. 5 APPLICATION OF THE SHORT-TERM AND LONG-TERM
CONCENTRATION MODELS IN COMPLEX TERRAIN
The short-term and long-term concentration models described in Sections
A. 3 and A. 4 are strictly applicable only for flat terrain where the base of the stack
(or the building source) and the ground surface downwind from the source are at the
same elevation. However, both models may also be applied to complex terrain by
defining effective stabilization heights and mixing depths. The following assump-
tions are made in the model calculations for complex terrain:
A-18
-------
• The top of the surface mixing layer extends over the
calculation grid at a constant height above mean sea
level
• Ground-level concentrations at all grid points above
the top of the surface mixing layer are zero
• Plumes that stabilize above the top of the surface
mixing layer do not contribute to ground-level concen-
trations at any grid point (this assumption also applies
to flat terrain)
In order to determine whether the stabilized plume is contained within
:e mixing layer, it is n«
the source from the relationship
the surface mixing layer, it is necessary to calculate the mixing depth H *{z } at
in s
where
H *{z } = (E +z - z } (A-33)
ml sj \ m a s/
H = the depth of the surface mixing layer measured at a point
with elevation z above mean sea level
a
z = the height above mean sea level of the source
s
Equation (A-33) is represented schematically in Figure A-l. As shown by the
figure, the 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 is defined to be con-
m^ sj
tained within the surface mixing layer.
A-l 9
-------
Top of Mixing Layer
Mixing Depth
z
Mixing Depth Measured
>- at Airport Equals
Minimum Depth
<) (No calculations
made for grid
points with
terrain elevations
above top of /
mixing layer (msl)
at airport)
Assigned to
Source
FIGURE A-l.
Mixing depth Hm(zs} used to determine whether the stabilized plume is 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,
s
the effective height H'{z} of the plume centerline above the terrain is then given
by
H'{z} =
H - z
o
H -z > 0
o
H - z < 0
o
(A-34)
For building sources, H'{z) is always set equal to zero.
The effective mixing depth H' {z} above a point at elevation z above
mean sea level is defined by
H1 (z) =
ml J
H
m
; z a: z
H +z-z ; z -= z
ma a
(A-35)
Figure A-2 illustrates the assumptions implicit in Equation (A-35). For grid
points at elevations below the airport elevation, the effective mixing depth H ' {z}
is allowed to increase in a manner consistent with Figure A-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 noted that the concentration is
set equal to zero for grid points above the actual top of the mixing layer (see
Figure A-l).
The terrain adjustment procedures also assume that the mean wind
speed at any given height above sea level is constant. Thus, the wind speed u
measured at height z_ above the surface at a point with elevation z above mean
H a
A-21
-------
tss
to
Effective Top of Mixing Layer
Effective
Mixing Depth
H
Assigned to
Grid Point
Mixing Depth Measured
at Airport Equals
Minimum Depth
(No calculations
made for grid
points with
terrain elevations
above top of
mixing layer (msl)
at airport
FIGURE A-2.
Effective mixing depth H ' {z} assigned to grid points for the concentration calculations.
-------
sea level is adjusted to the stack height for the plume-rise calculations by the
relationship
u{h} =
R\ z
R
; h s z +z_
o a R
; h < z +z
o a R
(A-36)
where h is the height above mean sea level of the top of the stack. Similarly,
the wind speed U{H} used in the concentration calculations is given by
u{H) =•
; H < z +z
o a R
(A-37)
It should be noted that the terrain-adjustment procedures outlined above
provide a very simple representation of complex plume-terrain interactions that
are not yet well understood. Because the model assumptions are generally con-
servative, it is possible that concentrations calculated for elevated terrain,
especially elevated terrain near a source, exceed the concentrations that actually
occur. It should also be noted that the procedures described above differ from
previous "terrain-intersection" models in that terrain intersection is only per-
mitted for a plume contained within a mixing layer. That is, terrain intersection
is permitted for all stability categories, but only for a plume contained within the
surface mixing layer.
A-23
-------
A. 6 PROCEDURES USED TO CALCULATE OCCURRENCE FREQUENCIES OF
HOURLY CONCENTRATIONS
This section describes procedures to calculate the annual probability of
measuring hourly ground-level concentrations, produced by emissions from a
single stack, that are greater than or equal to a selected threshold concentration
XTTj. Meteorological inputs required for the calculations include the annual distri-
bution of wind-speed and wind-direction categories classified according to stability
or time-of-day categories. By knowing the fraction of the time that the threshold
concentration is exceeded, the calculated probabilities can be converted to con-
venient units such as hours per year. Occurrence frequencies calculated using
these procedures will be essentially the same as those calculated using a year of
sequential hourly meteorological data.
Whenever the maximum ground-level concentration x exceeds the
6 max
threshold concentration x , an approximately elliptical area will exist within
TH
which the ground-level concentration is greater than or equal to X_Tr- This con-
TH
cept is illustrated in Figure A-3. The x axis in the figure is the downwind axis,
the y axis is the crosswind axis and the vertical axis is the concentration axis.
The ratio y of the threshold concentration xmTT to the maximum ground-level
1 rl
concentration x determines the size of this ellipse.
max
The maximum ground-level concentration may be easily calculated by
making the following simplifications in the short-term concentration model
described in Section A. 3:
• The mixing depth H is assumed to be much greater
than the plume stabilization height H (i. e., H » H)
• Equation (A-ll) is replaced by the simplified expression
a = a' xa (A-38)
A-24
-------
101008
N>
Ul
FIGURE A-3. Illustration of the approximately elliptical area within which the ground-level
concentration is greater than or equal to the threshold concentration x
TH
-------
• Equation (A-13) is replaced by the simplified expression
a = a' x .. (A-39)
Z £j
Under the above assumptions, the maximum ground-level concentration obtained
from Equation (A-5) is given by
q+1
[.
•ma* ... , n(B+l) SXP I' <—>l
-------
The maximum distance x at which concentrations greater than or equal to v H
^ J. XT
can occur is given by
x9 = C {y} -T- (A-43)
2 2 a£
The distance x from the source to the center of the ellipse is then given by
ymax
x = — " (A-44)
ymax 2
Finally, the semi-minor axis y of the ellipse is given by
max
). 9
The coefficients C {7}, C {y} and C {y} are listed in Table A-3. It should be
\. Zt o
noted that these coefficients and Equations (A-42) through (A-45) assume that the
lateral diffusion coefficient a is equal to 0. 9.
Equations (A-41) through (A-45), which allow one to readily determine the
area within which the threshold concentration is equalled or exceeded, form the
basis for the following probability calculation procedures:
(1) For each wind-speed/stability combination, calculate
max;i, k*
(2) For each wind-speed/stability combination, calculate
•v = y /Y
ri,k TH/ max;i,k
A-27
-------
TABLE A-3
COEFFICIENTS FOR CALCULATING THE DIMENSIONS OF
ELLIPSES WITHIN WHICH THE CONCENTRATION
IS GREATER THAN OR EQUAL TO X
y
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
0.95
0.99
1.00
cjy}
0.323106
0.357991
0.386236
0. 412444
0.438581
0.466111
0.496744
0.533358
0.583162
0.620825
0.675747
0.725476
c2«
3.951545
2.690688
2.126670
1.783548
1.542262
1.356514
1.203265
1.068030
0.936680
0.863202
0.781617
0.725476
c3M
2.941260
2.003900
1.551049
1.255806
1.033335
0. 849163
0.684611
0.525411
0.351782
0.242594
0.106428
0.000000
A-28
-------
(3) All wind-speed/stability combinations for which j S: 1
i, k
need not be considered; assign zero frequencies to these
combinations and normalize the frequencies of occurrence
of the remaining wind-speed/stability combinations within
each wind-direction sector by dividing by the total frequency
of occurrence for all wind-direction sectors of the wind-speed/
stability combinations for which y < 1. These adjusted
i, k
frequency distributions will be used to determine the probability
of observing concentrations above the threshold x at each
IH
grid point (r, 0).
(4) For each of the remaining wind-speed/stability combinations,
calculate x, . . , x , x . . and y ...
l;i,k 2;i,k ymax;i,k max;i,k
(5) For each grid point (r, 9), calculate the probability f'{r, 0} .
i, k
of observing a concentration above the short-term threshold
concentration X™^ as a result of the i wind-speed cate-
th
gory and k stability category, where
f.',{r,0} =Y* P. . ,{r,0) f! . , (A-46)
i.k1- t-t i,j,kv i,j,k v
J
and f! . , is the adjusted (normalized) frequency of
!»J» K
occurrence of the i wind-speed category, j wind-
direction category and k stability category. The equations
defining P. . are given below.
!> j> K
(6) For each grid point (r, 9), calculate the total probability
F{r, d} of observing concentrations greater than or equal
to X™^ from the equation
1H
A-29
-------
(A~47)
The weighting factor P. . {r, 0} is determined by the ratio of the
angular width of the portion of the elliptical area contained within the wind-direction
sector at downwind distance r to the angular width A0' of the sector. That is,
P. is given by
; B a A01
pi 1 kr'
J-» J> K
76.
\ i,
=£ B < A0'
'-6. {r}) (A0'+6. {r})
Lk V i,k >>
- (A-48)
T5 —
27T -
0' - 0!
j
7T
I 0' - 0'.
'• J
(A-49)
6 . {r} = 2 arctan
y{r).
(A-50)
A-30
-------
y(r).
max;i,k
1 -
4 fr-x . . Y
ymax;i, k /
1/2
r» • i
2;i,k
; r
or
r =* x
2;i,k
(A-51)
It should be noted that P. . is not permitted to exceed unity and <5 is assumed
i, j, k i, K
to be less than A01.
If T is the fraction of the time during the year that X is greater than
max
or equal to x • the number of hours per year that the hourly concentration at the
TH
point (r, 9) is greater than or equal to X™u can be estimated from the expression
1H
{Hours/year} = F{r, 0} (T) (365) (24)
= 8760 F{r, 6} T
(A-52)
A-31
-------
03
to
-------
APPENDIX B
ANNUAL FREQUENCY DISTRIBUTION OF WIND SPEED
AND WIND DIRECTION AT N26th AND PEARL
Tables B-l through B-6 list the annual joint frequency of occurrence by
Pasquill stability category of wind-speed and wind-direction categories at N26
and Pearl for the year 1972. These distributions were developed from concurrent
hourly cloud cover observations at Me Chord Air Force Base and hourly wind
measurements at the Puget Sound Air Pollution Control Agency1 s meteorological
tower at N26 and Pearl by the STAR program of the National Climatic Center.
This computer program uses the Turner (1964) definitions of the Pasquill stability
categories (see Section 3.1). As explained in Section 3. 3, we combined the E and
F stability categories in our calculations because we believe that surface roughness
elements and heat sources in the Tacoma are incompatible with the minimal
turbulent mixing associated with the Pasquill F stability category. Also, the A
and B stability categories were combined in the annual concentration calculations
because the A stability category only occurs about 0.4 percent of the time in the
Tacoma area (see Table B-l).
B-l
-------
TABLE B-l
1972 ANNUAL FREQUENCY DISTRIBUTION OF WIND SPEED AND WIND DIRECTION
AT N26th AND PEARL FOR THE A STABILITY CATEGORY
Direction
(Sector)
N
NNE
NE
ENE
E
ESE
SE
SSE
S
ssw
sw
wsw
w
WNW
NW
NNW
Total
Wind Speed (meters per second)
00.0-01.5
.0000
.0000
.0616
.0369
.0000
.0000
.0000
.0000
.0000
.0123
.0000
.0123
.0369
.0246
.0246
.0493
.2586
01.6-03.0
.0000
.0493
.0246
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0246
.0246
.0256
.0123
.0000
.0000
.1601
03. 1-05. 1
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
. 0.000
05.2-08.2
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
08.3-10.8
.0000
.0000
.0000
.0000
.0000
.0000
.0000
. Oo'oo
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
> 10.8
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
Total
.0000
.0493
.0862
.0369
.0000
.0000
.0000
.0000
.0000
.0123
.0246
.0369
.0616
.0369
.0246
.0493
.4187
W
N)
-------
TABLE B-2
1972 ANNUAL FREQUENCY DISTRIBUTION OF WIND SPEED AND WIND DIRECTION
AT N26th AND PEARL FOR THE B STABILITY CATEGORY
Direction
(Sector)
N
NNE
NE
ENE
E
ESE
SE
SSE
S
SSW
SW
wsw
w
WNW
NW
NNW
Total
Wind Speed (meters per second)
00.0-01.5
.1486
.4197
.5219
.2364
.0642
.0128
.0000
.0391
.0519
.1027
.1455
.3657
.5130
.2846
.3221
.2322
3.4606
01.6-03.0
.1724
.5049
.4926
.1232
.0000
.0000
.0000
.0123
.0123
.0000
.0985
.4433
.2833
.0493
.0246
.0246
2.2414
03.1-05.1
.1601
.3325
.2956
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.1601
.2094
.0493
.0000
.0000
.0000
1.2069
05.2-08.2
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
08.3-10.8
.0000
.0000
.0000
.0000
.0000
.0000
.0000
. oo'oo
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
> 10.8
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
Total
.4812
1.2572
1.3101
.3596
.0642
.0128
.0000
. 0514
.0642
.1027
.4041
1.0184
.8455
.3339
.3467
.2568
6.9089
W
co
-------
TABLE B-3
1972 ANNUAL FREQUENCY DISTRIBUTION OF WIND SPEED AND WIND DIRECTION
AT N26thAND PEARL FOR THE C STABILITY CATEGORY
Direction
(Sector)
^l>J w l^ Ul/1. /
N
NNE
NE
ENE
E
ESE
SE
SSE
S
SSW
SW
wsw
w
WNW
NW
NNW
Total
Wind Speed (meters per second)
00.0-01.5
.3229
.2518
.4598
.1970
.0972
.0424
.0547
.1136
.1204
.1204
.2969
.2723
.2354
.1574
.1136
.1245
2.9803
01.6-03.0
.1970
.4187
.6897
.1724
.0123
.0123
.0000
.0369
.0985
.0985
.3325
.4803
.1478
.0616
.0369
.0123
2.8079
03.1-05.1
.3695
1.3054
.7882
.0246
.0123
.0000
.0000
.0123
.0369
.1108
.4433
.4064
.0493
.0123
.0000
.0123
3.5837
05.2-08.2
.0616
.3202
.0739
.0000
.0000
.0000
.0000
.0000
.0000
.0123
.0369
.0123
.0000
.0000
.0000
.0000
.5172
08.3-10.8
.0000
.0000
.0000
.0000
.0000
.0000
.0000
. o'o'oo
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
> 10.8
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
Total
.9510
2.2961
2.0115
.3941
.1218
.0547
.0547
.1628
.2559
.3421
1.1097
1.1713
.4324
.2313
.1505
. 1492
9.8892
W
-------
TABLE B-4
1972 ANNUAL FREQUENCY DISTRIBUTION OF WIND SPEED AND WIND DIRECTION
AT N26th AND PEARL FOR THE D STABILITY CATEGORY
Direction
(Sector)
N
NNE
NE
ENE
E
ESE
SE
SSE
S
ssw
sw
wsw
w
WNW
NW
NNW
Total
Wind Speed (meters per second)
00.0-01.5
.7113
.9294
1.1694
.6241
.3558
.5550
.4507
1.1473
1. 6430
1.3568
1.3308
1.2177
1. 1254
.5000
.3215
.2933
13.7315
01.6-03.0
.5172
1.3300
1.3300
.3079
.3079
.2217
.1847
.5049
1.4778
1.9335
2.6232
2. 0443
.8374
.1355
.0739
.0739
13.9039
03.1-05.1
.1970
1.3670
1.2192
.1601
.0739
.1478
.0862
.2340
.8251
3.2759
3.9286
1.6995
.2094
.0000
.0000
.0123
13.4360
05.2-08.2
.1108
1.3916
1.0345
.1355
.0739
.0123
.0246
.0739
.5049
2.9557
3.2882
1.4409
.0616
.0000
.0000
.0123
11.1207
08.3-10.8
.0000
.1601
.0369
.0739
.0000
.0000
.0000
. OO'OO
.0862
.3941
.3818
.4064
.0000
.0000
.0000
.0000
1.5394
> 10.8
.0000
.0123
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0493
.0493
.0985
.0000
.0000
.0000
.0000
.2094
Total
1. 5364
5. 1904
4.7901
1.3015
.8115
.9368
.7463
1.9601
4.5371
9.9651
11.6018
6.9074
2.2338
.6355
.3954
.3918
53.9409
W
-------
TABLE B-5
1972 ANNUAL FREQUENCY DISTRIBUTION OF WIND SPEED AND WIND DIRECTION
AT N26thAND PEARL FOR THE E STABILITY CATEGORY
Direction
N
NNE
NE
ENE
E
ESE
SE
SSE
S
SSW
SW
wsw
w
WNW
NW
NNW
Total
Wind Speed (meters per second)
00.0-01.5
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
01.6-03.0
.1108
.7882
.6897
.1232
.1232
.0739
.1232
.3079
.8744
.7266
.7143
.3571
.0369
.0369
.0123
.0123
5.1108
03. 1-05.1
.0493
.7759
.8744
.0862
.0369
.0369
.0123
.0985
. 1232
.2463
.1970
.2586
.0246
.0000
.0000
.0000
2. 8202
05.2-08.2
.0000
. 0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
08.3-10.8
.0000
.0000
.0000
.0000
.0000
.0000
.0000
. o'o'oo
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
> 10.8
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
Total
.1601
1. 5640
1. 5640
.2094
.1601
.1108
.1355
.4064
.9975
.9729
.9113
.6158
.0616
.0369
.0123
.0123
7.9310
td
i
-------
TABLE B-6
1972 ANNUAL FREQUENCY DISTRIBUTION OF WIND SPEED AND WIND DIRECTION
AT N26th AND PEARL FOR THE F STABILITY CATEGORY
Direction
(Sector)
N
NNE
NE
ENE
E
ESE
SE
SSE
S
ssw
sw
wsw
w
WNW
NW
NNW
Total
Wind Speed (meters per second)
00.0-01.5
1.1341
1. 5243
1.8145
.8769
.4616
.4469
.4622
1.4925
1.2592
1.2529
1.3990
.7687
.6606
.5511
.5112
.5814
15.1970
01.6-03.0
.3202
1.1700
1.2931
.3325
.1355
.0123
.0123
.2833
.3941
.4310
.5911
.5172
.1355
.0000
.0246
.0616
5.7143
03.1-05.1
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
05.2-08.2
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
08.3-10.8
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
> 10.8
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
Total
1.4543
2. 6943
3.1076
1.2094
.5970
.4593
.4746
1.7758
1.6533
1.6839
1.9901
1.2859
.7960
.5511
.5358
. 6430
20.9113
W
-------
oo
I
m
-------
APPENDIX C
EMISSIONS DATA FOR THE OTHER MAJOR SO
SOURCES IN THE SEATTLE-TACOMA AREA 2
Table C-l lists the locations and the stack and emissions data for the
major SO sources in the Seattle-Tacoma area other than the ASARCO smelter.
^2
The parameters in Table C-l, which were developed from information provided by
PSAPCA, were used in all diffusion-model calculations described in this report
unless otherwise noted.
In two cases (Tacoma City Light and Seattle City Light), a source complex
consisted of two or more stacks in close proximity and with identical stack and
emissions parameters. In order to reduce computer computational time without
loss of accuracy, the SO emissions for these stacks were combined and assumed to
&
originate from a single stack with a volumetric emission rate and exit temperature
equal to the volumetric emission rate and exit temperature of the individual stacks.
For these two cases, Table C-l also lists the number of stacks.
For the reasons discussed in Appendix D, exit-velocity restrictions on
plume rise were not included in the diffusion-model calculations for stacks with
Froude numbers less than 3. 0. The stack radius for a stack with Froude number
less than 3. 0 is shown as zero in Table C-l.
C-l
-------
TABLE C-l
LOCATIONS AND STACK AND EMISSIONS DATA FOR THE MAJOR
SO SOURCES IN THE SEATTLE-TACOMA AREA
Source
Name
Dupont de
Nemours & Co.
Hooker Chemi-
cals & Plastics
Pennwalt
St. Regis Kraft
Mill
Source
Number
1
2
3
4
5
6
7
8
9
10
11
12
S°2
Emissions
(tons/yr)
101.0
196.0
36.6
36.6
95.0
150.0
97.0
237.0
4.0
6.0
964.0
375.0
UTMX
Coordinate
(m)
525800
545240
547240
547240
543300
543300
543300
543300
543300
543300
543300
543300
UTM Y
Coordinate
(m)
5217250
5236190
5234890
5234890
5234800
5234800
5234800
5234800
5234800
5234800
5234800
5234800
Elevation of
Stack Base
(m above MSL)
70
2
2
2
2
2
2
2
2
2
2
2
Stack
Height
(m)
46
58
18
46
53
46
88
84
23
53
30
25
Stack
Radius
(m)
-0-
-0-
.65
-0-
-0-
.84
1.33
2.13
1.07
1.30
.99
.68
Volumetric
Emission
Rate
(m3/sec)
2.74
16.33
9.34
9.34
42.03
70.05
56.04
151.78
38.29
58.38
24.28
17.00
Stack
Exit
Temp.
(°K)
436
426
436
533
405
405
335
451
480
491
422
422
o
I
to
-------
TABLE C-l
LOCATIONS AND STACK AND EMISSIONS DATA FOR THE MAJOR
SO SOURCES IN THE SEATTLE-TACOMA AREA
Li
Source
Name
Quemetco
U.S. Oil &
Refining
U. S. Gypsum
Boise Cascade
Boeing-Renton
Kaiser-Aluminum
Western Seattle
Steam
Shell Oil
Texaco
U. S. N. -Sand
Point
Source
Number
13
14
15
16
17
19
20
21
22
23
24
25
S°2
Emissions
(tons/yr)
1,410.4
310.0
52.7
34.8
8.8
33.7
1,931.3
152.0
94.0
70.0
26.1
31.4
UTMX
Coordinate
(m)
548800
545800
547730
532250
532250
558600
547600
549900
549140
549140
548900
555500
UTM Y
Coordinate
(m)
5269500
5233610
5234600
5225500
5225500
5261500
5234000
5272500
5269800
5269800
5269700
5281400
Elevation of
Stack Base
(m above MSL)
3
4
6
58
58
9
4
20
3
3
3
15
Stack
Height
(m)
27
33
7
61
17
12
9
40
30
9
21
11
Stack
Radius
(m)
.46
-0-
.23
-0-
.61
.43
.97
-0-
-0-
-0-
-0-
Volumetric
Emission
Rate
(m3/sec)
9.33
22.38
4.25
26.15
9.34
8.50
30.0*
56.63
3.50
1.40
.42
8.64
Stack
Exit
Temp.
<°K)
366
626
394
505
519
422
30*
450
422
422
467
505
o
I
CO
* Indicates building source; building length and width are entered as stack exit temperature and volumetric
emission rate.
-------
TABLE C-l
LOCATIONS AND STACK AND EMISSIONS DATA FOR THE MAJOR
SO SOURCES IN THE SEATTLE-TACOMA AREA
Source
Name
Veterans Hospi-
tal - Tacoma
Me Chord AFB
Pier 91
McNeil Peniten-
tiary
Milwaukee RR
Switchyard
Tacoma City
Light Plant 2
Seattle City
Light Lake
Union Plant
Seattle Public
Safety Bldg.
Seattle Center
Source
Number
26
27
29
30
31
32
34
41
42
s°2
Emissions
(tons/yr)
73.2
159.0
143.0
97.9
29.5
566.0
(2 stacks)
410.9
(7 stacks)
50.4
40.6
UTMX
Coordinate
(m)
532300
539150
546400
525500
544500
546900
550900
550300
548900
UTM Y
Coordinate
(m)
5220000
5220020
5275500
5227500
5234500
5235900
5276400
5272300
5274200
Elevation of
Stack Base
(m above MSL)
75
84
2
55
2
6
43
31
35
Stack
Height
(m)
38
6
9
46
10
42
51
55
27
Stack
Radius
(m)
-0-
-0-
-0-
-0-
-0-
-0-
-0-
-0-
.51
Volumetric
Emission
Rate
(m3/sec)
4.67
23.33
18.36
2.26
1.20
35.87
6.61
2.36
9.80
Stack
Exit
Temp.
(°K)
422
491
505
422
450
505
477
366
477
o
,1
-------
TABLE C-l
LOCATIONS AND STACK AND EMISSIONS DATA FOR THE MAJOR
SO SOURCES IN THE SEATTLE-TACOMA AREA
2
Source
Name
University of
Washington
Veterans Hospi-
tal-Seattle
Seattle-Tacoma
Airport
Standard Oil
Source
Number
43
44
45
46
S°2
Emissions
(tons/yr)
96.7
37.6
129.2
52.0
UTMX
Coordinate
(m)
552000
552000
553010
549600
UTMY
Coordinate
(m)
5277500
5267800
5253890
5270200
Elevation of
Stack Base
[m above MSL)
27
98
117
9
Stack
Height
(m)
61
37
15
10
Stack
Radius
(m)
1.68
-0-
.30
-0-
Volumetric
Emission
Rate
m3/sec)
139.38
2.83
2.36
.94
Stack
Exit
Temp.
(°K)
394
422
450
533
o
-------
o
05
-------
APPENDIX D
EXIT-VELOCITY RESTRICTIONS ON PLUME RISE
AND THE ASARCO PLUME
In most plume rise equations, including those of Briggs (1969, 1970, 1971
and 1972), momentum effects are neglected when calculating buoyant plume rise.
There are sound theoretical and empirical reasons to believe that momentum effects
are negligible in comparison with buoyancy effects for a hot stack. However,
Briggs (1969), Fay, et_aL, (1970) and others have expressed the belief that their
plume rise equations may not be applicable when the ratio of the stack exit velocity
w to the mean wind speed u is less than 1. 5 to 1. 2 unless the Froude number Fr is
less than about unity. The Froude number Fr, which in essence is the ratio of
momentum to buoyancy, is defined by Briggs (1969, p. 6) as
Fr - fn, _ T v (D-l)
(
where
T = the stack exit temperature (°K)
s
T - the ambient air temperature (°K)
a
D = the stack diameter (m)
2
g = the acceleration due to gravity (m/sec )
Thus, a stack with a small Froude number will have a high heat content. For w/u
less than 1. 2, Briggs (1970) states that "downwash of the plume into the low pres-
sure region in the wake of the stack is likely to occur." The ASME Guide (1973,
p. 23-25) recommends that a stack should be designed so that w/u is greater than
or equal to 2 most of the time. As shown in Table D-l, most of the stacks
used to validate the Briggs plume rise equations have either high exit velocities or
D-l
-------
TABLE D-l
EXIT VELOCITIES AND FROUDE NUMBERS OF STACKS USED TO
VALIDATE THE BRIGGS (1969) PLUME RISE EQUATIONS
Source*
Harwell A**
Harwell B
Bosanquett**
Darmstadt**
Duisburg
Tallawarra**
Lakeview**
CEGB Plants
Barley
Earley
Castle Donington
Castle Donington
Northfleet**
Northfleet**
TVA Plants
Shawnee**
Colbert**
Johnsonville
Widows Creek
Gallatin
Gallatin
Paradise
Paradise
Exit Velocities (m/sec)
9.9
9.9
9.7
4.8
8.5
3.7
19.8
5.6
17.1
12.5
16.7
14.1
21.3
14.8
13.1
28.9
21.8
16.0
7.2
15.6
17.4
Froude Number
17.6
17.6
3.0
0.9
5.6
0.5
19.9
1.9
17.7
5.3
9.5
10.9
24.9
11.9
7.7
43.8
18.5
8.9
1.5
20.6
8.9
* See Table 5.1 of Briggs (1969) for data references.
**Not included in "selected data" by Briggs (1969) because factors such as low exit
velocity, terrain downwash, etc. may affect plume rise.
D-2
-------
low Froude numbers. Thus, any exit velocity effects on plume rise would not be
observed for these stacks except on rare occasion.
During 1974 and 1975, the H. E. Cramer Company developed a computer-
ized set of short-term and long-term urban diffusion models and used them to
calculate ground-level concentrations produced by the major stationary 862
sources in Lansing, Michigan and Allegheny County, Pennsylvania. During the
two studies, we found that visual observations of the behavior of the plumes from
a number of stacks did not correspond to the behavior predicted by the Briggs
equations when the wind speed approached the stack exit velocities. Observations
suggested that, rather than attaining the calculated plume rise, the plumes did
not achieve any buoyant rise when the wind speed was greater than or equal to
the stack exit velocities. In general, these stacks were characterized by Froude
numbers appreciably larger than unity.
On the basis of our observations in Lansing and Allegheny County, we
hypothesized that plume rise is precluded for most stacks when w/u is less than
or equal to unity. Following the suggestions of Briggs (1973), we assumed that
the exit-velocity effect starts to occur when w/u is equal to 1. 5. We therefore
multiplied the plume rise Ah calculated from the Briggs equations by a factor f
defined by
I ; u •= w/1.5
3w - 35
s
w
0
. 5 =s u
w
w ^ u
(D-2)
The expression in Equation (D-2) for u between 0. 67 w and w is a simple linear
interpolation between the two extremes that we hypothesized. Equation (D-2)
corresponds to Equation (A-3) in Appendix A.
D-3
-------
The use of Equation (D-2) in long-term and short-term diffusion model
calculations substantially improved the correspondence between the calculated and
observed concentrations in every case. One 24-hour historical case in Allegheny
County was of particular interest. For this case, we had reported visual observa-
tions that the plumes from a power plant passed directly over an air quality monitor
for much of the day. The plumes behaved in the manner predicted by our modified
plume rise formulas, and the ratios of calculated and observed maximum 3-hour and
24-hour average concentrations were 1.17 and 1.10, respectively. This case is
discussed in a recent report prepared for EPA (Cramer, et al., 1975).
Since completion of the Allegheny County and Lansing studies, we have
continued our study of exit-velocity restrictions on plume rise. We have obtained
aircraft measurements made by the EPA Environmental Monitoring and Support
Laboratory (EMSL) of plume rise for three coal-fired power plants during periods
when the wind speed at stack height was equal to or greater than the stack exit
velocity. Also, we have analyzed the plume-rise data of Bringfelt (1968), although
Bringfelt and Briggs (1970) point out that there are some deficiencies in these
data. We have found that, for stacks with Froude numbers greater than 3. 0, the
Briggs plume rise equations significantly overestimate plume rise as the wind
speed approaches or exceeds the stack exit velocity. If our semi-empirical
correction factor f is applied to the Bringfelt trials with the potential for exit-
velocity effects (w/u ^1.5 and Fr —3.0), the correspondence between calculated
and observed plume rise is essentially the same as obtained for the trials without
exit-velocity effects. We have found no evidence in the Bringfelt data or the EMSL
observations of exit-velocity restrictions on plume rise for stacks with Froude
number less than 1. 0. For stacks with Froude numbers between 1. 0 and 3. 0,
some stacks appear to exhibit exit-velocity effects and some do not. Because the
Froude number of the ASARCO stack varies from 1.5 to 2. 8 (depending on the
number of stack heaters), there is some uncertainty about the behavior of the
D-4
-------
ASARCO plume as the wind speed at stack height approaches the stack exit-velocity
(about 6 meters per second).
During our background study of the ASARCO smelter, we hypothesized
that, if exit-velocity restrictions on plume rise applied to the ASARCO stack,
high short-term ground-level concentrations would begin about 1 kilometer
southwest of the stack during periods of moderate-to-strong northeast winds. To
test our hypothesis, EPA Region X located a continuous SO monitor at a distance
u
of about 1. 2 kilometers and at an azimuth bearing of 215 degrees from the stack.
Table D-2 lists the hourly ground-level SO concentrations greater than or equal
Zi
to the PSAPCA hourly standard of 0. 25 ppm measured during the period
18 September 1975 through 27 January 1976. For days with several consecutive
hours with concentrations above 0. 25 ppm, the time and magnitude of the maximum
hourly concentration are shown in parentheses in the table.
Not all of the high SO concentrations listed in Table D-2 occurred during
Lt
periods of moderate-to strong northeast winds, the conditions that we hypothesized
as leading to the highest short-term concentrations. Some of the high concentra-
tions occurred with light northeast winds or during air stagnation. The most
clearly defined cases of observed high hourly concentrations at the EPA monitor
occurred on the night of 18-19 September 1975. The hourly ground-level concen-
trations calculated from meteorological data and ASARCO in-stack SO concentra-
^
tion measurements provided by PSAPCA, assuming exit-velocity restrictions on
plume rise, ranged from 2.5 to 10 times lower than the observed concentrations.
The calculated concentrations that were about 10 times lower than the observed
concentrations could be substantially increased by a slight increase in the wind speed
or a slight decrease in the stack exit-velocity (the correction factor f is extremely
sensitive to the ratio w/u for w/u. between 1. 5 and 1). For the hour ending at
D-5
-------
TABLE D-2
MAXIMUM HOURLY GROUND-LEVEL SO2 CONCENTRATIONS
MEASURED BY THE EPA SO2 MONITOR IN TACOMA
Date
18 September 1975
19 September 1975
24 September 1975
30 September 1975
1 October 1975
24 November 1975
19 December 1975
20 January 1976
Time (PST)
2100 - 2200
2200 - 2300
(2136 - 2236)
0200 - 0300
1048 - 1148
1324 - 1424
1700 - 1800
1800 - 1900
(1712 - 1812)
1724 - 1824
1448 - 1548
1100 - 1200
1200 - 1300
1300 - 1400
(1206 - 1306)
0900 - 1000
Concentration (ppm)
0.43
0.59
(0. 79)
0.30
0.27
0.29
0.71
0.31
0.75
0.44
0.28
0.35
0.45
0.27
0.47
0.43
D-6
-------
2300 PST on 18 September, the wind speed at stack height exceeded the exit
velocity (i. e., the calculated plume rise was zero) and the calculated concentra-
tion was 2.5 times lower than the observed concentration. The ASARCO in-stack
SO concentration data indicated a very low emission rate for this hour. If the
2
maximum uncurtailed emission rate is used, the calculated concentration exceeds
the observed concentration (the curtailment report does not indicate any curtailment
in smelter operations during this hour). We repeated the concentration calculations
without assuming exit-velocity restrictions and, in every case, the calculated
concentration at the EPA monitor was zero.
For the majority of the cases listed in Table D-2, we believe that the
observed wind directions clearly establish that the smelter was the source of the
SO measured by the EPA monitor. Assuming that the ASARCO in-stack SO
2 2
concentration data are valid on a short-term basis, the majority of the measured
SO concentrations at the EPA monitor could not have been caused by emissions
2i
from the main stack. It is therefore necessary to hypothesize continuous low-level
SO emissions on the order of 1, 000 pounds per hour to account for the observed
LI
concentrations. Low-level fugitive emissions occur during certain phases of
production (Weisenberg and Serne, 1976), but there is no record of either their
magnitude or time duration.
In summary, we were not able to establish whether exit-velocity restric-
tions on plume rise apply to the main ASARCO stack because of uncertainties in
the SO emissions from the main stack and from low-level fugitive sources. The
£t
PSAPCA and ASARCO SO monitors in the area are too far from the stack to be
2t
useful in establishing whether exit-velocity effects are occurring. Because there
remains a reasonable doubt that exit-velocity restrictions on plume rise should be
applied to the ASARCO stack, we did not include exit-velocity effects in any of our
calculations for the ASARCO stack. If exit-velocity effects had been included, the
calculated maximum short-term ground-level SO concentrations would have been
&
increased by almost a factor of 4.
D-7
-------
00
I
Q
-------
APPENDIX E
DISCUSSION OF THE SHORT-TERM HISTORICAL CASES
E. 1 INTRODUCTION
As noted in Section 4. 3, we analyzed meteorological and air quality data
provided by PSAPCA and selected 20 dates when high hourly SO2 concentrations
were measured at one or more SO2 monitors for one or more hours. We then
used concurrent meteorological and emissions data to calculate concentrations at
the monitor sites for comparison with the observed concentrations. The source
and meteorological inputs, the calculation procedures and the results of the cal-
culations are discussed in Section 4. 3. This Appendix provides a detailed dis-
cussion of the individual cases.
In Figure E-l, a map of the Tacoma area, the + symbols show the loca-
tions of the ASARCO and PSAPCA SO2 monitors. The ASARCO monitors are
labeled Al through A5; the PSAPCA monitors are labeled with the PSAPCA station
number. Table E-l gives the names of the ASARCO and PSAPCA SO2 monitors
identified in Figure E-l. It should be noted that the PSAPCA and ASARCO monitors
at N26 and Pearl are located in close proximity, as are the PSAPCA Meeker and
the ASARCO Meeker-Brown monitors.
Several of the Short-Term Historical Cases treat the occurrence of
high SO2 concentrations at the PSAPCA McMicken Heights and Tukwila monitors,
which are located outside of the area shown in Figure E-l. The McMicken Heights
monitor is 23. 7 kilometers northeast of the main ASARCO stack (indicated by the
star in Figure E-l) and the Tukwila monitor is 26. 0 kilometers northeast of the
stack.
E-l
-------
- - - 5230
Figure E-l. Map of the Tacoma area. The + symbols show the locations of
the ASARCO and PSAPCA SO2 monitors. The star shows the
location of the main ASARCO stack.
E-2
-------
TABLE E-l
NAMES OF ASARCO AND PSAPCA SO MONITORS
SHOWN IN FIGURE E-l
Monitor Label in
Figure E-l
Al
A2
A3
A4
A5
P2
P3
P6
P7
P8
K56
K62
Monitor Name
N26th and Pearl
Reservoir
Highlands
University Place
Meeker-Brown
th
N26 and Pearl
Fife Sr. H. S.
Meeker Jr. H. S.
Adams Street
Williard Elementary School
Vashon Island
Maury Island
E-3
-------
Because the purpose of the historical calculations was to demonstrate the
validity of the diffusion models and modeling procedures, we usually calculated
concentrations only at the locations of the PSAPCA and ASARCO SC>2 monitors.
The exceptions are Case 4 (5 July 1974) and the single 24-hour case (Case 20:
25 December 1974). Thus, we did not estimate the maximum ground-level SO
dt
concentrations for most of the historical cases. However, maximum ground-level
SO concentrations have been calculated for the existing configuration of the ASARCO
£A
smelter (without SCS) and for possible future smelter configurations using the same
modeling techniques. The results of these calculations are presented in Section 5.
E.2 DISCUSSION OF THE INDIVIDUAL CASES
E.2.1 Case 1: 29 March 1972 (Fumigation Case)
Surface winds during the early morning hours of 29 March 1972
were light and variable. With the ASARCO plume stabilizing above the top of the
surface mixing layer, the ASARCO smelter operated without emissions curtailment
until 0930 PST when the smelter's SCS requested 60 percent curtailment. Shortly
after 1100 PST, high ground-level SO£ concentrations began to occur in the Tacoma
area. The following SO2 concentrations in parts per million (ppm) were observed
at the ASARCO and PSAPCA SO2 monitors during the period 1100 to 1400 PST:
Hour (PST) N26th and Pearl Adams St. Reservoir
0.09
0.38
0.03
0.17
where the concentrations enclosed by parentheses are the concentrations measured
by the ASARCO monitor at N26th and Pearl.
1200
1300
1400
Average
0.16 (0.19)
0.19 (0.36)
0.03 (0.08)
0.13 (0.21)
0.00
0.36
0.14
0.17
E-4
-------
The simultaneous occurrence at midday of high SO concentrations
N26th and Pearl, Adams Street and the Reservoir suggests a widespread fumigation.
Although the ASARCO smelter's SCS requested 78 percent curtailment at 1115 PST
and 88 percent curtailment at 1200 PST, it was too late to prevent hourly SO2
concentrations above the PSAPCA hourly standard of 0.25 ppm at three separate
locations for the hour ending at 1300 PST.
The calculated concentrations in ppm for the period of fumigation are:
Monitor Calculated Concentration
N26th and Pearl 0.16
Adams St. 0. 14
Reservoir 0.18
ASARCO emissions account for about 99 percent of these calculated concentrations.
The concentration calculated for N26tn and Pearl is approximately equal to the
hourly concentrations measured by the PSAPCA monitor at N26 and Pearl for
the hours ending at 1200 and 1300 PST. However, the calculated concentration
is only about half the concentration measured by the ASARCO monitor at N26th and
Pearl for the hour ending at 1300 PST. Similarly, the concentrations calculated
for Adams Street and the Reservoir are less than half the measured concentrations
during the period 1200 to 1300 PST. However, the average concentrations for the
period of fumigation agree well with the calculated concentrations.
E.2.2 Case 2: 4 May 1972 (Critical Wind-Speed Case)
During this critical wind-speed case, high SO£ concentrations
were measured at N26^ and Pearl, the Reservoir and Highlands during the hour
ending at 2000 PST. The ASARCO SCS did not request emissions curtailment until
1930 PST, when 78 percent curtailment was requested. The observed and calculated
E-5
-------
SO0 concentrations in ppm for the hour ending at 2000 PST are:
Lt
Monitor Observed Concentration Calculated Concentration
N26th and Pearl (1.23) 0.88
Reservoir 0.27 0.04
Highlands 0.73 0.44
The parentheses for the observed concentration at N26 and Pearl indicate the
ASARCO monitor value$ no data are available for the PSAPCA monitor for this
hour. It should be noted that the contributions of emissions from the non-ASARCO
SO sources in the Seattle-Tacoma area to the calculated concentrations are
£i
negligible.
On the average, the calculated concentrations are about half the observed
concentrations. The ASARCO SO2 emission rate used in the calculations was
estimated from the ASARCO curtailment report using Equation (4-1), and we have
found that the reported curtailment often does not reflect the actual emissions
from the smelter. The uncertainly about the actual SO2 emission rate helps to
account for the discrepancies between calculated and observed concentrations, at
least at N26^ and Pearl and Highlands, Also, there are some uncertainties in the
measured concentrations (note that the 1972 annual average concentrations measured
by the ASARCO and PSAPCA SO2 monitors at N26th and Pearl differ by more than a
factor of 3; see Section 4.2).
E.2.3 Case 3: 4 June 1972 (Limited-Mixing Case)
Light westerly winds on the morning of 4 June 1972 shifted to
northeast at midday resulting in high SOg concentrations in the Tacoma area for
the hour ending at 1400 PST. During the period 1300 to 1330 PST, the ASARCO
SCS requested 88 percent emissions curtailment; the requested curtailment was
increased to 95 percent at 1330 PST0 The observed and calculated SOo concentrations
E-6
-------
in ppm for the hour ending at 1400 PST are:
Monitor Observed Concentration Calculated Concentration
N26th and Pearl 0.46(0.40) 0.25
Reservoir 0.14 0.10
Highlands 0.24 0.13
where the number enclosed by parentheses is the concentration measured by the
ASARCO monitor at N26tn and Pearl. Emissions from the main ASARCO stack
are solely responsible for the calculated concentrations.
On the average, the calculated concentrations are about 63 percent of the
observed concentrations. Based on a comparison of SCvj emission rates estimated
for other cases from Equation (4-1) and from in-stack SOg concentration data, it
appears that Equation (4-1) significantly underestimates the SCv emission rate when
the requested curtailment exceeds about 90 percent. Thus, it is not surprising that
the observed concentrations exceed the calculated concentrations for this case.
E.2.4 Case 4: 5 July 1972 (Limited-Mixing Case)
With the ASARCO smelter operating without curtailment, the
wind shifted from northwest to north-northeast during the early morning hours of
5 July 1972. Although the smelter's SCS requested 88 percent curtailment at
0220 PST, the hourly SO2 concentrations measured at N26tn and Pearl and the
Reservoir exceeded the PSAPCA hourly standard of 0.40 ppm. The PSAPCA
monitor at N26th and Pearl measured a concentration of 0. 56 ppm, while the
ASARCO monitor measured a concentration of 0.68 ppm. At the Reservoir, the
ASARCO monitor measured a concentration of 0. 62 ppm.
Figure E-2 shows the isopleths of ground-level SO2 concentration produced
by emissions from the main ASARCO stack calculated using the observed surface
E-7
-------
wind direction at N26th and Pearl of 007 degrees. As shown by Figure E-2, the
Reservoir monitor (labeled A2 in the figure) is on the plume axis, while the
and Pearl station (labeled P2 and Al) is on the edge of the plume with a calculated
concentration of about 0.1 ppm. Figure E-3 shows the same isopleth pattern for
a mean wind direction of 010 degrees. In this case, the calculated concentrations
at N26tn and Pearl and the Reservoir are 0. 61 and 0. 64 ppm, respectively. Thus,
a 3-degree shift in the surface wind direction yields calculated concentrations that
are within 10 percent of the observed concentrations. If the surface wind direction
f*Vi
had not been adjusted, the calculated concentration at N26 and Pearl would have
been about 15 percent of the observed concentration and the calculated concentration
at the Reservoir would have been about double the observed concentration.
Figures E-2 and E-3 clearly demonstrate the sensitivity of short-term
concentrations calculated at fixed receptor points to slight changes in the wind
direction used in the calculations. If the observed surface wind direction is used
in the calculations, a comparison of calculated and observed concentrations often
indicates an apparent failure of the short-term diffusion model. However, a
slight change in the surface wind direction usually brings the calculated and observed
concentrations into close agreement at all monitors. Short-term concentrations
calculated at fixed receptor points are more sensitive to slight uncertainties in
the wind direction than to slight uncertainties in any source or other meteorological
input parameter. Thus, we believe that the procedure that we have followed in the
short-term historical calculations of varying the wind direction by several degrees
to find the optimum wind direction is fully warranted for the cases when two or more
monitors simultaneously detected the ASARCO plume.
Figures E-2 and E-3 also indicate that the PSAPCA and ASARCO air
quality monitoring networks did not detect the calculated maximum hourly ground-
level SO concentration for the hour ending at 0300 PST on 5 July 1972. The
^
figures indicate that the hourly ground-level SO concentration exceeded 1. 0 ppm
Li
E-8
-------
5240
5235
5230
535
540
545
550
5225
FIGURE E-2.
Calculated isopleths of hourly ground-level SO2 concentration in parts per million
produced by ASARCO emissions for the hour ending at 0200 PST on 5 July 1972,
assuming a wind direction of 007 degrees.
-------
5240
H
i
530
- 1 5235
T A! C 0 MIA
5230
545
550
FIGURE E-3.
Calculated isopleths of hourly ground-level SC>2 concentration in parts per million
produced by ASARCO emissions for the hour ending at 0300 PST on 5 July 1972,
assuming a wind direction of 010 degrees.
-------
within a narrow elliptical area beginning about 2.4 kilometers from the base of
the main ASARCO stack and ending about 10 kilometers from the stack. The
calculated maximum hourly concentration of 1.98 ppm is located about four
kilometers from the ASARCO stack.
E.2.5 Case 5: 3 October 1972 (Fumigation Case)
During the night of 2 -3 October 1972, the ASARCO smelter
operated without curtailment and with a single stack heater. Shortly before mid-
night, high ground-level SO2 concentrations began to occur at N26**1 and Pearl,
the Reservoir and Adams Street. The following SO2 concentrations in ppm were
observed:
Hour (PST)
02/2400
03/0100
03/0200
Average
N26tn and Pearl
0. 03 (0. 04)
0. 35 (0. 16)
0.05 (0.03)
0.14 (0.08)
Reservoir
0.08
0.23
0
0.10
Adams St.
0.01
0.18
0.01
0.07
where the numbers enclosed by parentheses are the concentrations measured by
the ASARCO monitor at N26til and Pearl.
The simultaneous occurrence of high SO0 concentrations at the three
£t
monitors suggests a widespread fumigation. Because the high concentrations
occurred in the middle of the night, solar heating obviously could not have been
the cause of this fumigation. For some other reason, it appears that the mixing
depth increased to above the plume height and the plume, which had been elevated,
was suddenly mixed to the surface. Carlson and Scuderi (1974) report a series of
diffusion experiments that included at least one trial that they believe showed this
phenomenon. They hypothesize that the vertical wind shear increased until there
was a sudden transfer of momentum to the surface, resulting in a fumigation.
E-ll
-------
In our calculations for this case, we assumed that the vertical turbulent
intensity in the surface layer during the period of fumigation was equivalent to the
median value for the Pasquill B stability category. The calculated fumigation
concentrations in ppm are:
Monitor Calculated Concentration
N26th and Pearl 0.14
Reservoir 0.13
Adams St. 0.16
ASARCO emissions account for 98 to 99 percent of these calculated concentrations.
The calculated concentrations agree well with the average concentrations during
the period of fumigation. However, the calculated concentrations at Adams Street
and the Reservoir are about 30 percent lower than the maximum observed hourly
concentrations,, At N26 and Pearl, the calculated concentration is only 40 percent
of the maximum hourly concentration measured by the PSAPCA monitor and about
88 percent of the maximum hourly concentration measured by the ASARCO monitor.
E.2.6 Case 6: 11 October 1972 (Critical Wind-Speed Case)
During this critical wind-speed case, high SO2 concentrations
were detected whenever the wind caused the ASARCO plume to pass over the N26*h
and Pearl and the Highlands SO2 monitors. At 0215 PST, the smelter SCS turned
on both heaters and requested 55 percent emissions curtailment. At 0320 PST,
the SCS requested 88 percent curtailment. The following SO concentrations in
-------
where the numbers enclosed by parentheses are the concentrations measured by the
ASARCO monitor at N26tn and Pearl. The relatively low concentrations at the
Highlands monitor and the near-threshold concentrations at University Place suggest
that the ASARCO plume may have had a curvilinear trajectory during this period
(the Highlands and University Place SO2 monitors are on the same azimuth bearing
from the ASARCO stack).
The calculated concentrations in ppm for the hours ending at 0300 and 0400
PST are:
Hour (PST) N26th and Pearl Highlands
0300 0.30 0.97
0400 0.23 0.04
Only emissions from the main ASARCO stack contribute significantly to the calculated
concentrations. With the exception of the concentration calculated for Highlands for
the hour ending at 0300 PST, the calculated and observed concentrations are in good
agreement. A slight deviation of the ASARCO plume from a straight-line trajectory
could easily explain the discrepancy between the calculated and observed concen-
trations for this monitor and hour (see Figures E-2 and E-3).
E.2.7 Case 7: 24 October 1974 (Limited-Mixing Case)
As early morning ground fog began to clear and light and variable
surface winds increased and became northeasterly, high ground-level SO2 concen-
trations began to occur at N26th and Pearl at about 0800 PST on 24 October 1974.
The ASARCO SCS, which had previously requested 60 percent emissions curtailment,
requested 88 percent curtailment at 0815 PST. Nevertheless, the hourly SO2 con-
centrations measured by both monitors at N26^ and Pearl exceeded the PSAPCA
E-13
-------
hourly standard of 0.40 ppm. The observed and calculated concentrations in ppm
for the hour ending at 0900 PST are:
Monitor Observed Concentration Calculated Concentration
N26th and Pearl 0. 60 (0. 46) 0.26
Reservoir 0.10 0.08
Highlands 0.07 0.13
where the number enclosed by parentheses is the concentration measured by the
ASARCO monitor at N26^ and Pearl. As in the previous cases of northeast winds,
emissions from the main ASARCO stack account for over 99 percent of the calcu-
lated concentrations.
The two concentration measurements at N26 and Pearl suggest that the
SO2 emission rate for the main ASARCO stack was about 20,000 pounds per hour
rather than the rate of 10, 000 pounds per hour used in the calculations. An
emission rate of 20,000 pounds per hour would double the concentrations calculated
at the Reservoir and Highlands, but the differences between calculated and observed
concentrations would be less than the difference in the concentrations measured by
the two monitors at N26*n and Pearl. As noted in the following discussion of Case 8,
in-stack SO2 concentration data indicated an SOg emission rate of 16,000 pounds
per hour during a period when 100 percent curtailment was requested. The ASARCO
curtailment report and Equation (4-1) were used to estimate the emission rate of
10, 000 pounds per hour for this case. Thus, it is quite possible that the SO emis-
£i
sions during the period 0800 to 0900 PST on 24 October 1972 were about 20,000 pounds.
E.2.8 Case 8: 26 May 1973 (Limited-Mixing Case)
During the morning of 26 May 1973, surface wind directions at
the various PSAPCA meteorological towers in the Tacoma area differed significantly.
For example, the 1000 PST wind direction at N26tn and Pearl was 250 degrees, while
E-14
-------
the wind direction at Vashon Island was 130 degrees. The average of these wind
directions is 190 degrees, which is consistent with the south winds required to
transport ASARCO emissions to the island (the observed concentration at Vashon
Island for the hour ending at 1000 PST was 0. 36 ppm). The wind shifted to north-
northeast by early evening, resulting in high SO2 concentrations in the Tacoma
area. The emissions curtailment requested by the ASARCO SCS ranged from 0 to
100 percent on 26 May 1973. However, in-stack SO2 concentration data indicate
that the SO0 emission rate was never less than about 16, 000 pounds per hour for
£t
the hours included in this case.
Assuming that the wind direction at plume height was 196 degrees for the
hour ending at 1000 PST, the calculated SO2 concentration at the Vashon Island
monitor is 0.44 ppm. This calculated concentration is entirely due to ASARCO
emissions. Thus, ASARCO emissions could have produced the observed concen-
tration of 0. 36 ppm. However, we cannot accurately position the ASARCO plume
during this hour because there is only one SO2 monitor on Vashon Island. As
previously noted, we can often estimate the wind direction at plume height to
within 1 or 2 degrees when the plume is simultaneously detected by two or more
monitors. In these cases, the correspondence between calculated and observed
SOp concentrations at the various monitors usually provides strong evidence that
the observed concentrations were produced by emissions from the ASARCO smelter.
On the evening of 26 May 1973, the following SO2 concentrations in ppm
were measured in the Tacoma area:
Monitor 2200 PST 2300 PST
N26th and Pearl 0,32 (0.21) 0 (0.03)
Reservoir 0.07 0.37
Highlands 0.11 0.00
University Place 0.15 0.05
E-15
-------
where the numbers enclosed by parentheses are the concentrations measured by the
ASARCO monitor at N26t^1 and Pearl. The corresponding calculated concentrations
are:
Monitor 2200 PST 2300 PST
N26th and Pearl 0.44 0.02
Reservoir 0.03 0.39
Highlands 0.22 0.00
University Place 0.21 0.00
Comparison of the calculated and observed concentrations for the hour ending at
2200 PST suggests that the SO2 emission rate used in the calculations is about 40
percent too high. For this hour, the differences between the calculated and observed
concentrations at the Reservoir (0. 03 ppm) and at Highlands (0.11 ppm) are less than
or equal to the difference in the concentrations measured by the ASARCO and PSAPCA
SO2 monitors at N26th and Pearl. For the hour ending at 2300 PST, the calculated
and observed concentrations at the Reservoir are in excellent agreement; the cal-
culated and observed concentrations at N26*n and Pearl, Highlands and University
Place are all near the threshold of the SO2 monitors.
E.2.9 Case 9: 4 June 1973 (Limited-Mixing Case)
Low clouds and fog covered the Tacoma area on the morning of
4 June 1973. High SO2 concentrations began to occur at the N26*h and Pearl SO~
monitors at about 0600 PST and the ASARCO SCS requested 100 percent emissions
curtailment at 0625 PST. However, in-stack SO2 concentration data indicate that
over 25,000 pounds of SO2 were emitted during the period 0600 to 0700 PST. The
early morning fog and low clouds cleared by mid-morning, leading to unstable
meteorological conditions with light and variable winds. Consequently, the ASARCO
SCS requested emissions curtailment ranging from 88 to 100 percent throughout the
morning.
E-16
-------
For the hours ending at 0700 and 1100 PST on 4 June 1973, the observed
SO concentrations in ppm were:
&
Monitor 0700 PST 1100 PST
N26th and Pearl 0. 37 (0. 67) 0. 31 (0. 21)
Reservoir 0.00 0.23
Highlands 0.13 0.08
University Place 0. 00 0. 03
where the numbers enclosed by parentheses are the concentrations measured by
the ASARCO monitor at N26 and Pearl. The corresponding calculated concen-
trations are:
Monitor 0700 PST 1100 PST
N26th and Pearl 0. 64 0. 11
Reservoir 0.20 0.13
Highlands 0.26 0.05
University Place 0. 19 0. 03
For the hour ending at 0700 PST, the calculated concentration for N26th
and Pearl is in very good agreement with the concentration measured by the
ASARCO monitor. However, the calculated and observed concentrations at the
other monitors are not in good agreement. We have no explanation for the low
observed concentrations at the Reservoir, Highlands and University Place, but
the differences between the calculated and observed concentrations at these monitors
are less than the difference in the concentrations measured by the collocated moni-
tors at N26th and Pearl.
For the hour ending at 1100 PST, comparison of the calculated and observed
concentrations at N26"1 and Pearl, the Reservoir and Highlands suggests that the
SO emission rate of 4,400 pounds per hour estimated from the in-stack SO
E-17
-------
concentration data and used in the calculations is only about 57 percent of the actual
emission rate during this period. The calculated and observed concentrations at
University Place are so near the monitor threshold that a direct comparison of these
concentrations is likely to be misleading.
E.2.10 Case 10: 19 June 1973 (Limited-Mixing Case)
Broken to overcast middle clouds covered the Tacoma area until
late afternoon on 19 June 1973. The ASARCO SCS requested emissions curtailment
ranging from 60 to 88 percent between 0700 and 1740 PST, at which time curtailment
was ended. At about 2000 PST the wind shifted from northeast to north-northeast
and high SO2 concentrations began to occur at the SO2 monitors in Tacoma. The
ASARCO SCS ordered 100 percent curtailment at 2015 PST and, as concentrations
began to decrease, the requested curtailment was changed to 37 percent at 2045 PST.
In-stack SO2 concentration data indicate that the main ASARCO stack emitted about
17, 000 pounds of SO2 between 2000 and 2100 PST. For the hour ending at 2100 PST,
the observed and calculated SOg concentrations in ppm are:
Monitor Observed Concentration Calculated Concentration
N26th and Pearl 0.25 (0.25) 0.34
Reservoir 0.07 0.01
Highlands 0.07 0.09
University Place 0.06 0.08
where the number enclosed by parentheses is the concentration measured by the
ASARCO monitor at N26th and Pearl.
Excluding the Reservoir, a comparison of the calculated and observed
concentrations for each monitor indicates that the estimated SO~ emission rate
is about 33 percent higher than the actual emission rate. At the Reservoir, the
E-18
-------
calculated concentration is about 0. 06 ppm lower than the observed concentration,
which was only 0. 07 ppm. As previously noted, concurrent concentration measure-
ments by the collocated SO monitors at N26th and Pearl frequently differ by 0.10
ppm or more.
E.2.11 Case 11: 25 July 1973 (Limited-Mixing Case)
Early morning ground fog cleared by about 1000 PST on 25 July
1973, followed by moderately unstable meteorological conditions. With light north-
east to north-northeast winds, appreciable SO2 concentrations began to occur at the
SO2 monitors in Tacoma at about 1000 PST. The ASARCO SCS requested emissions
curtailment of 87 to 98. 5 percent during the morning and afternoon, and in-stack
SO2 concentration data indicate that SO2 emissions from the main stack were re-
duced to about 6,000 pounds per hour. The following SO2 concentrations in ppm
were observed in the Tacoma area at midday:
Monitor
N26til and Pearl
Reservoir
Highlands
University Place
1100 PST
0.11 (0.19)
0.14
0.02
0.01
1200 PST
0.13 (0.23)
0.00
0.08
0.07
1300 PST
0.09 (0.15)
0.00
0.04
0.05
where the numbers enclosed by parentheses are the concentrations measured by
the ASARCO monitor at N26^n and Pearl. The corresponding calculated concen-
trations in ppm are:
Monitor
N26th and Pearl
Reservoir
Highlands
University Place
1100 PST
0.20
0.16
0.09
0.05
1200 PST
0.12
0.05
0.08
0.04
1300 PST
0.08
0.02
0.08
0.04
E-19
-------
For the hour ending at 1100 PST, the calculated and observed concentrations
at N26 and Pearl and the Reservoir are in good agreement if the measurement by
the ASAECO monitor at N26th and Pearl is assumed to be more representative than
the measurement by the collocated PSAPCA monitor. The measured concentrations
at Highlands and University Place are so close to the monitor threshold that a com-
parison of calculated and observed concentrations is likely to be misleading. How-
ever, the differences between the calculated and observed concentrations at these
sites are less than the difference in the concentrations measured by the two monitors
at N26 and Pearl during this hour.
For the hour ending at 1200 PST, the calculated and observed concentrations
at Highlands and at N26 and Pearl are in good agreement if the measurement by the
PSAPCA monitor at N26^ and Pearl is assumed to be more representative than the
measurement by the ASARCO monitor. It should be noted that the observed concen-
trations at the Reservoir, Highlands and University Place are quite low, and the
differences between the calculated and observed concentrations at the Reservoir and
University Place are about half the difference between the two observed concentra-
tions at N26tn and Pearl.
For the hour ending at 1300 PST, the observed concentrations at the Reser-
voir, Highlands and University Place are so low that a comparison of calculated and
•f"Vi
observed concentrations is probably not warranted. At N26 and Pearl, the calcu-
lated concentration is about 10 percent lower than the concentration measured by the
PSAPCA monitor and about half the concentration measured by the ASARCO monitor.
E.2.12 Case 12: 30 July 1973 (Limited-Mixing Case)
On occasion, the ASARCO plume can affect the ambient air quality
at long distances from the smelter. For example, the PSAPCA SO<, monitors at
McMicken Heights and Tukwila are affected by ASARCO SO9 emissions during periods
E-20
-------
of sustained southwest winds. The McMicken Heights monitor is about 23. 7
kilometers northeast of the main ASARCO stack and the Tukwila monitor is about
26. 0 kilometers northeast of the stack. Because the delay time in the transport of
ASARCO emissions to these monitors is typically several hours, the ASARCO SCS
must anticipate the onset of conditions leading to high concentrations at these
monitors and must curtail emissions several hours in advance. If the SCS fails
to anticipate adverse meteorological conditions and does not curtail emissions
until after the high concentrations begin to occur, high concentrations tend to
persist for a time approximately equal to the travel time between the smelter and
the monitors.
On 30 July 1973, southwest winds during the early morning hours carried
the ASARCO plume northeast toward the McMicken Heights and Tukwila monitors.
The mean wind speed at plume height was about 2 meters per second so that the
dealy time in the transport of ASARCO emissions to the monitor was at least 3
hours. Taking this delay into account, the in-stack SOo concentration data indicate
that the effective SO2 emission rate was about 30,000 pounds per hour during the
period 0800 to 1000 PST. Although the emissions curtailment requested by the
ASARCO SCS was increased from 56 to 74 percent at 0800 PST, it was too late to
preclude high hourly SO0 concentrations at McMicken Heights and Tukwila. The
£t
following SOg concentrations in ppm were observed at the two monitors:
Monitor 0900 PST IQOQ PST
McMicken Heights 0.50 0.23
Tukwila 0.18 0.13
The corresponding calculated concentrations in ppm are:
Monitor 0900 PST 1000 PST
McMicken Heights 0.52 0.16
Tukwila 0.31 0.06
E-21
-------
These calculated concentrations are entirely due to ASARCO emissions.
The calculated concentrations for the hour ending at 0900 PST are in good
agreement with the observed concentrations. The agreement between calculated
and observed concentrations for the hour ending at 1000 PST is not nearly as good
as the agreement for the preceding hour. However, the differences between the
calculated and observed concentrations are less than the differences in concurrent
observed concentrations that frequently occur for the collocated SOg monitors at
N26th and Pearl.
E.2.13 Case 13: 31 July 1973 (Limited-Mixing Case)
West winds during the early morning hours of 31 July 1973
transported the ASARCO plume toward the PSAPCA Meeker and the ASARCO
Meeker-Brown SO0 monitors. At 0520 PST the ASARCO SCS requested 100 per-
z
cent emissions curtailment; the requested curtailment was reduced to 94 percent
at 0600 PST. The in-stack SO2 concentration data indicate that the SO2 emission
rate slowly decreased from about 25,000 pounds per hour to about 6, 000 pounds
per hour during the next 3 hours. Because of the delay time required to reduce
emissions and the approximate 1-hour travel time to the Meeker monitors, high
concentrations persisted until the wind shifted to the southwest shortly before 0700
PST. With the onset of southwest winds, high concentrations began to occur at
the McMicken Heights and Tukwila monitors sometime after 0800 PST.
The observed SO concentrations in ppm at the Meeker SO0 monitors
2 <2
during the early morning hours of 31 July 1973 were:
Hour (PST) Meeker Meeker-Brown
0500 0.28 0.31
0600 0.38 0.42
0700 0.59 0.49
E-22
-------
Although the Meeker and Meeker-Brown SO0 monitors are in such close proximity
Ll
that they are shown as being in the same location in Figure E-l, we used the 180-
meter separation to estimate the wind direction at plume height to within about 1
degree. Assuming a 1-hour travel time to the Meeker monitors, the calculated
concentrations in ppm are:
Hour (PST) Meeker Meeker-Brown
0500 0.28 0.35
0600 0.42 0.50
0700 0.56 0.53
All of the calculated concentrations are within 20 percent of the observed concen-
trations. On the average, the calculated concentrations are about 8 percent higher
than the observed concentrations. It should be noted that ASARCO is the only
source that contributes to the calculated concentrations.
The observed SO2 concentrations in ppm at the McMicken Heights and
Tukwila SC>2 monitors were:
Hour (PST) McMicken Heights Tukwila
0900 0.16 0.12
1000 0.14 0.19
Assuming a travel time to these monitors of about 3 hours, the corresponding
calculated concentrations in ppm are:
Hour (PST) McMicken Heights . Tukwila
0900 0.21 0.05
1000 0.13 0.20
E-23
-------
The correspondence between calculated and observed concentrations for the hour
ending at 0900 PST is only fair, although the maximum difference between cal-
culated and observed concentrations is only 0.07 ppm. For the hour ending at
1000 PST, the calculated and observed concentrations are in very good agreement.
As in the previous short-term calculations for McMicken Heights and Tukwila, the
calculated concentrations are entirely due to ASARCO emissions.
E.2.14 Case 14: 17 October 1973 (Limited-Mixing Case)
Light and variable surface winds during the morning of 17
October 1973 changed to light southwest winds by late morning and high SO2 con-
centrations began to occur at McMicken Heights and Tukwila at about 1200 PST.
At 1230 PST the ASARCO SCS requested that emissions curtailment be increased
from 37 to 55 percent. Nevertheless, the observed SO2 concentration at Tukwila
for the hour ending at 1300 PST exceeded the PSAPCA hourly standard of 0.40
ppm. The observed and calculated SO2 concentrations in ppm for the hour ending
at 1300 PST are:
Monitor Observed Concentration Calculated Concentration
McMicken Heights 0.30 0.09
Tukwila 0.41 0.10
The calculated concentrations assume that the travel time from the ASARCO
smelter to the two monitors was about 2 hours on 17 October 1973.
Of all of the historical cases for the limited-mixing condition, this case
has the poorest correspondence between calculated and observed concentrations.
Assuming the validity of the SO emission rate estimated from the in-stack SO2
concentration data, it appears that the meteorological inputs used in the calculations
are not representative of the actual conditions during the period 1200 to 1300 PST.
E-24
-------
The early morning Portage Bay EMSU sounding showed a shallow mixing layer of
about 125 meters and the 1300 PST sounding showed a deep mixing layer of about
1150 meters. The early afternoon mixing depth was used in the calculations.
However, it is possible that there was a significant increase in the mixing depth
between 1200 and 1300 PST. If the average mixing depth for the period was about
400 meters, the calculated and observed concentrations would be in much better
agreement.
E.2.15 Case 15: 28 January 1974 (Critical Wind-Speed Case)
During this critical wind-speed case, relatively high SO2
concentrations were measured at Maury Island during the early morning hours
and again at midday. As high SOo concentrations began to occur at about 0200
PST on 28 January 1974, the ASAECO SCS requested 55 percent emissions curtail-
ment. Although curtailment was reduced to 17 percent at 0230 PST, the SCS re-
quested 55 percent curtailment and turned on the second stack heater at 0330 PST.
High SO concentrations again began to occur at Maury Island at about 1100 PST.
^
Emissions curtailment requested by the ASARCO SCS during the period 1100 to
1200 PST ranged from 50 to 88 percent.
For the hours ending at 0300, 0400 and 1200 PST on 28 January 1974, the
observed and calculated SO0 concentrations in ppm at Maury Island are:
z
Hour (PST) Observed Concentration Calculated Concentration
0300 0.10 0.25
0400 0.20 0.32
1200 0.18 0.16
E-25
-------
The calculated concentrations assume that the ASARCO plume was trans-
ported directly over the Maury Island monitor. ASARCO emissions account for
100 percent of the calculated concentrations. Because the calculated centerline
concentrations are greater than or approximately equal to the measured concentra-
tions, ASARCO emissions could easily have produced the observed SO concentra-
Lt
tions at Maury Island on 28 January 1974.
E.2.16 Case 16: 9 February 1974 (Fumigation Case)
When early morning ground fog cleared in the Tacoma area
on the morning of 9 February 1974, high ground-level SO concentrations occurred
£t
over a widespread area. The ASARCO SCS requested emissions curtailments of
40 percent at 0535 PST, 88 percent at 0545 PST, 50 percent at 0720 PST and
82 percent at 0830 PST. As the high SO concentrations that began at about
^
0830 PST continued, the SCS requested 100 percent curtailment between 0900 and
1300 PST. The following SO concentrations in ppm were measured in Tacoma
^
and the surrounding area on the morning of 9 February 1974:
Monitor 1000 PST 1100 PST 1200 PST Avg
N26th and Pearl 0.42(0.40) 0.49(0.52) 0.29(0.27) 0.40(0.40)
Maury Island 0.01 0.09 0.18 0.09
Reservoir 0.07 0.21 0.14 0.14
Highlands 0.12 0.19 0.21 0.17
where the concentrations enclosed by parentheses are the concentrations measured by
tv»
the ASARCO monitor at N26 and Pearl. The calculated fumigation concentrations
in ppm are:
E-26
-------
Monitor Calculated Concentration
N26th and Pearl 0. 08
Maury Island 0. 03
Reservoir 0.09
Highlands 0.06
The contribution of ASARCO emissions to the calculated concentrations ranges
from 93 percent at Maury Island to 98 percent at the Reservoir and N26 and
Pearl.
The calculated concentrations are about 15 to 45 percent of the maximum
observed hourly concentrations and about 20 to 65 percent of the average concen-
trations measured during the period of fumigation. In the previous fumigation
cases (Cases 1 and 5), the calculated concentrations typically were about half of
the maximum hourly concentrations and approximately equal to the average concen-
trations for the period of fumigation. The poor correspondence between the calcu-
lated fumigation concentrations and the average observed concentrations in this
case suggests that the wind-direction distribution departed significantly from the
circular wind-direction distribution assumed by the model. Also, the SO
£t
emission rate for the ASARCO stack used in the calculation (the average rate
indicated by the in-stack SO data for the period 0500 to 0900 PST) may have been
^
too low.
E. 2.17 Case 17: 26 February 1974 (Limited-Mixing Case)
Broken to overcast low and middle clouds covered the Tacoma
area during the early morning hours of 26 February 1974. Light rain was also
reported at Me Chord Air Force Base during this period. The ASARCO smelter
operated with one stack heater during the period 0300 to 0400 PST; the smelter1 s
SCS did not request any emissions curtailment. The observed SO concentration
E-27
-------
at Maury Island for the hour ending at 0400 PST was 0.33 ppm. Assuming that
the centerline of the ASARCO plume was transported directly over the Maury
Island monitor, the calculated concentration is 0.47 ppm. A 2- to 3-degree shift
in the wind direction would bring the calculated and observed concentrations into
exact agreement. As previously noted, when the wind blows directly from the
ASARCO stack toward the Maury Island monitor, ASARCO is the only source that
contributes to the concentration calculated for the monitor.
E. 2.18 Case 18: 6 September 1974 (Limited-Mixing Case)
Northeast winds and unstable meteorological conditions
during the afternoon of 6 September 1974 caused the ASARCO SCS to request
emissions curtailment ranging from 97 to 100 percent. Nevertheless, relatively
high SO concentrations were measured in the Tacoma area during the period
&
1500 to 1600 PST. The observed and calculated SO concentrations in ppm are:
£t
Monitor Observed Concentration Calculated Concentrations
N26t and Pearl 0.23(0.21) 0.20
Reservoir 0.09 0.10
Highlands 0.08 0.14
where the number enclosed by parentheses is the concentration measured by the
ASARCO monitor at N26 and Pearl.
This is one of the two hours included in the 20 Short-Term Historical
Cases when we believe that sea-breeze fumigation as described by Lyons and
Cole (1973) may have been occurring. However, the correspondence between the
observed concentrations and the concentrations calculated using the short-term
diffusion model described in Section A. 3 of Appendix A is excellent.
E-28
-------
E. 2.19 Case 19: 7 October 1974 (Limited-Mixing Case)
High SO concentrations were measured in the Tacoma area
£t
during the early morning of 7 October 1974 and again during the early afternoon.
After high concentrations began to occur at about 0100 PST, the ASARCO SCS
requested 100 percent emissions curtailment at 0145 PST. The requested curtail-
ment was reduced to 60 percent at 0230 and ended at 0315 PST. During the late
morning and early afternoon of 7 October 1974, the ASARCO SCS requested
emissions curtailment ranging from 84 to 96 percent. The following concentrations
in ppm were measured in the Tacoma area on 7 October 1974:
Monitor 0200 PST 1200 PST 1300 PST
N26th and Pearl 0.42(0.35) 0.27(0.17) 0.14(0.11)
Reservoir 0.50 0.03 0.02
Highlands 0.02 0.13 0.13
University Place 0.01 0.12 0.08
where the numbers enclosed by parentheses are the concentrations measured by
the ASARCO monitor at N26 and Pearl. The corresponding calculated concentra-
tions in ppm are:
Monitor 0200 PST 1200 PST 1300 PST
th
N26 and Pearl 0.87 0.10 0.14
Reservoir 1.00 0.04 0.02
Highlands 0.01 0.07 0.16
University Place 0.00 0.04 0.09
A comparison of the calculated and observed concentrations for the hour
ending at 0200 PST suggests that the SO emission rate used in the calculations
£
is about double the actual emission rate. The emission rate used in the calcula-
tion was estimated from Equation (2-1) and, as noted in Section 2.1, the emission
E-29
-------
rate obtained from Equation (2-1) may differ from the actual emission rate by
about 20 percent. Also, the average in-stack SO concentration for this hour is one
^
of the highest that we have seen for the smelter operating at 51 percent permanent
control. Because the ratios of calculated and observed concentrations at N26 and
Pearl and the Reservoir are nearly identical, we believe that this case demonstrates
the uncertainty in the ASARCO smelter1 s hourly SO emission rate rather than a
£i
failure of the short-term diffusion model.
For the hour ending at 1200 PST, the calculated concentrations are only
about one-third to one-half the observed concentrations. This hour may also
illustrate an uncertainty in the SO emission rate. However, the differences
^
between the calculated and observed concentrations are less than the difference
in the concentrations observed by the two monitors at N26 and Pearl.
The hour ending at 1300 PST is the second hour when we believe that
sea-breeze fumigation as described by Lyons and Cole (1973) may have been
occurring. As in the other case of possible sea-breeze fumigation, the correspon-
dence between the observed concentrations and the concentrations calculated using
the short-term diffusion model described in Section A. 3 of Appendix A is excellent.
Thus, it is not necessary to invoke the concept of sea-breeze fumigation in
order to explain the observed concentrations.
E.2.20 Case 20: 25 December 1974 (Limited-Mixing Case)
We selected 25 December 1974 for study because there were
no SCS activities at the ASARCO smelter throughout the entire 24-hour period
(i.e., no stack heaters were used and no emissions curtailment was requested).
Table E-2 lists the meteorological input parameters used in the calculations for
25 December 1974. The wind speeds and wind directions in Table E-2 are the
values measured at N26 and Pearl, except that Maury Island wind speeds and
E-30
-------
TABLE E-2
METEOROLOGICAL INPUT PARAMETERS FOR 25 DECEMBER 1974
Hour
(PST)
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
Wind
Direction
(deg)
212
241
216
217
220
254
223
207
206
214
204
257
272
325
335
296
274
226*
234*
224
149*
232
149*
139*
Wind
Speed
(m/sec)
3.807
4.424
5.196
5.813
4.373
2.829
3.138
3.395
1.595
1.440
2.161
1.749
1.440
1.080
0.823
1.029
1.183
1.235*
1.440*
0.823
1.440*
0.772
1.595*
1.595*
Ambient
Air
Temperature
(°K)
279
278
277
277
277
277
276
274
275
277
279
280
280
279
279
278
278
278
278
278
277
277
276
276
Potential
Temperature
Gradient
(°K/m)
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
Mixing
Depth
(m)
875
875
875
875
875
875
875
875
875
875
875
875
875
875
875
875
875
875
875
875
875
875
875
875
Wind
Profile
Exponent
P
0.15
0.15
0.15
0.10
0.20
0.25
0.20
0.15
0.20
0.20
0.15
0.15
0.20
0.20
0.20
0.20
0.25
0.25
0.25
0.25
0.30
0.25
0.20
0.25
Pasquill
Stability
Category
D
D
D
D
E
E
E
D
D
C
C
C
C
C
C
C
D
D
D
D
E
D
D
D
I
co
Wind direction and wind speed measured at Maury Island; N26 and Pearl winds were variable.
-------
wind directions are listed for the hours (1800, 1900, 2100, 2300 and 2400 PST)
when the winds at N26 and Pearl were reported as variable. No attempt was
made to estimate the hourly wind directions at plume height; the wind directions
in Table E-2 are unadjusted surface wind directions. The N26 and Pearl
(or. Maury Island) wind speeds were used in combination with cloud-cover obser-
vations at Me Chord Air Force Base to assign the Pasquill Stability Category to
each hour following the Turner (1964) procedures. The mixing depths and potential
temperature gradients obtained from the Portage Bay EMSU soundings on the
mornings of 24 and 26 December 1974 were averaged and assumed to apply through-
out 25 December 1974. Because broken to overcast low and middle clouds covered
the Tacoma area during most of this period, it is reasonable to assume that there
was little diurnal variation of either the mixing depth or the potential temperature
gradient. Table E-3 lists the emissions data for the main ASARCO stack on
25 December 1974. Although no emissions curtailment was requested on 25
December 1974, Table E-3 indicates that there were significant hour-to-hour
variations in the SO emission rate (Equation (2-1) was used to calculate the
^
hourly SO emission rates given in the table).
^
Figure E-4 shows the calculated isopleths of 24-hour average ground-level
SO concentration in ppm for 25 December 1974. The concentration isopleths in
Zt
Figure E-4 include the calculated effects of all major SO sources in the Seattle-
£t
Tacoma area. Table E-4 lists the calculated maximum hourly and 24-hour average
ground-level SO concentrations for 25 December 1974. ASARCO emissions
£
account for 99. 5 to 100 percent of the calculated concentrations listed in Table E-4.
The calculated maximum 24-hour average concentration is located approximately
6. 8 kilometers northeast of the main ASARCO stack. This concentration is below
the National and PSAPCA 24-hour Air Quality Standards of 0.14 and 0.10 ppm,
respectively. However, the calculations indicate that the PSAPCA hourly standard
of 0.40 ppm (never to be exceeded) was exceeded during 21 hours on Christmas
Day 1974. Also, the calculated maximum 3-hour concentration of 0. 52 ppm slightly
E-32
-------
TABLE E-3
EMISSIONS DATA FOR THE MAIN ASARCO STACK
ON 25 DECEMBER 1974
Hour
(PST)
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
SO2 Emission
Rate
(Ib/hr)
28,607
22,447
21,607
25,975
30,175
23, 007
23,567
22,447
18,975
19,927
21,495
22,335
21,327
21,775
17,407
19,927
20,375
25,247
21,887
20,487
21,887
25,807
20,207
19,647
Volumetric Emission
Rate
(m3/sec)
260.95
260.95
260.95
260.95
260.95
260.95
260.95
260.95
260.95
260.95
260.95
260.95
260.95
260. 95
260.95
260.95
260.95
260.95
260. 95
260.95
260.95
260.95
260. 95
250.95
Stack Exit
Temperature
(°K)
346.0
346.0
346.0
346.0
346.0
346.0
346.0
346.0
346.0
346.0
346.0
346.0
346.0
346.0
346.0
346.0
346.0
346.0
346.0
346.0
346.0
346.0
346.0
346.0
E-33
-------
FIGURE E-4.
Calculated isopleths of 24-hour average ground-level SC>2 concen-
tration in parts per million produced by emissions from the combined
sources on 25 December 1974.
E-34
-------
TABLE E-4
LOCATIONS AND MAGNITUDES OF CALCULATED MAXIMUM HOURLY
AND 24-HOUR SO CONCENTRATIONS ON 25 DECEMBER 1974
Hour
(PST)
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
24 -Hour
Average
Maximum Calculated
Concentration (ppm)
0.69
0.36
0.31
0.35
0.44
0.77
0.44
0.41
0.58
0.61
0.52
0.66
0.68
1.24
1.20
0.75
0.60
0.85
0.68
0.90
0.59
1.17
0.61
0.73
0.09
Location
UTM X(m)
541,000
542,500
541,000
541,000
541,500
544, 000
542,500
540, 000
540,000
540, 000
539,000
541,500
541,500
539,500
539,000
541,500
543,500
542, 000
542, 000
542, 000
534,000
543, 000
534, 000
533, 000
542, 000
UTM Y(m)
5,244,000
5,241,000
5,243,500
5,243,000
5,243,000
5,240,000
5,243,500
5,243,500
5,243,500
5,242,000
5,242,000
5,239,000
5,238,000
5,235,000
5,234,500
5,236,000
5,237,500
5,242,500
5,241,500
5,243,000
5,243,500
5,242,500
5,243,500
5,243,000
5,243,000
E-35
-------
exceeds the 3-hour Secondary Air Quality Standard of 0. 50 ppm. This concentra-
tion, which is located 3.1 kilometers southeast of the ASARCO stack, is calculated
for the 3-hour period ending at 1500 PST.
Table E-5 compares the calculated and observed hourly and 24-hour
average SO concentrations at the Maury Island, Meeker and Meeker-Brown SO
4 2t
monitors. As shown by the table, there is poor agreement between the hour-by-
hour calculated and observed concentrations at the three monitors. As previously
noted, we believe that this discrepancy is principally due to a fundamental
lack of accuracy in the available wind-direction data which precludes the accurate
positioning of plumes with respect to specific grid points on an hourly basis. For
periods of 24 hours, however, the effects of inaccuracies in the hourly meteorologi-
cal data are considerably reduced by the averaging process. As indicated by
Table E-6, the calculated and observed 24-hour average SO concentrations at
z
the various ASARCO and PSAPCA SO monitors are in good agreement. This
&
result is consistent with our previous experience in urban diffusion modeling
(for example, Cramer, et_al., 1975) and the experience of others as summarized
by Pasquill (1974, p. 309).
E-36
-------
TABLE E-5
COMPARISON OF CALCULATED AND OBSERVED CONCENTRATIONS
ON 25 DECEMBER 1974
Hour
(PST)
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
24 -Hour
Average
Maury Island
Calculated
Concentration
(ppm)
0.02
0.00
0.13
0.19
0.22
0.00
0.01
0.00
0.00
0.17
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.04
0.00
0.00
0.00
0.00
0.03
Observed
Concentration
(ppm)
0.00
0.06
0.08
0.02
0.01
0.01
0.04
0.04
0.02
0.26
0.06
0.00
0.00
0.00
0.00
0.00
0.02
0.25
0.12
0.05
0.00
0.00
0.00
0.00
0.04
MeekerJMeeker-Brown)
Calculated
Concentration
(ppm)
0.00(0.00)
0. 00(0. 00)
0.00(0.00)
0.00(0.00)
0. 00(0. 00)
0. 00(0. 00)
0. 00(0. 00)
0.00(0.00)
0.00(0.00)
0.00(0.00)
0. 00(0. 00)
0.14(0.16)
0.10(0.09)
0.00(0.00)
0. 00(0. 00)
0.00(0.00)
0.00(0.00)
0.00(0.00)
0. 00(0. 00)
0. 00(0. 00)
0.02(0.01)
0.00(0.00)
0. 02(0. 01)
0. 00(0. 00)
0.01(0.01)
Observed
Concentration
(ppm)
0.00(0.00)
0. 00(0. 00)
0. 00(0. 00)
0.00(0.00)
0.00(0.00)
0.00(0.00)
0.01(0.00)
0.00(0.00)
0.00(0.00)
0.00(0.00)
0.01(0.00)
0.00(0.00)
0.00(0.00)
0.01(0.01)
0. 01(0. 01)
0.10(0.05)
0.03(0.03)
0.00(0.00)
0.00(0.00)
0.00(0.00)
0.00(0.00)
0.00(0.00)
0.00(0.01)
0.01(0.02)
0.01(0.01)
E-37
-------
TABLE E-6
COMPARISON OF CALCULATED AND OBSERVED 24-HOUR AVERAGE
SO CONCENTRATIONS ON 25 DECEMBER 1974
£t
Monitor
N26 and Pearl
PSAPCA
ASARCO
Reservoir
Highlands
University Place
Meeker Jr. H. S.
Meeker-Brown
Maury Island
Calculated Concentration
(ppm)
0.00
0.00
0.00
0.00
0.01
0.01
0.03
Observed Concentration
(ppm)
0.00
0.00
0.00
0.00
0.01
0.01
0.04
E-38
------- |