Report Detail

Administrative Law Judges recommend that the Texas Commission on Environmental Quality rule against renewal of ASARCO's permit 20345 due to pollution concerns

April 04, 2007 -- TEXAS COMMISSION ON ENVIRONMENTAL QUALITY
AN ORDER concerning Application of ASARCO, Incorporated to Renew
Air Quality Permit No. 20345, TCEQ Docket No. 2004-0049-AIR, SOAH
Docket No. 582-05-0593
On ______________, the Texas Commission on Environmental Quality (Commission or
TCEQ) considered the application of ASARCO, Incorporated, to Renew Air Quality Permit
No. 20345. The application was presented to the Commission with a Proposal for Decision by the
Honorable William G. Newchurch and Veronica S. Najera, Administrative Law Judges (ALJs) with
the State Office of Administrative Hearings (SOAH).
After considering the ALJs’ Proposal for Decision (PFD) and the evidence and arguments
presented, the Commission makes the following Findings of Fact (FOF) and Conclusions of Law
(COL):
I. FINDINGS OF FACT
Introduction
1. On March 28, 2002, ASARCO, Incorporated, (Applicant or ASARCO) applied to the
Commission to renew its Air Quality Permit No. 20345 (Permit, Current Permit, or Permit
20345).
2. The requested renewal would allow Applicant to resume its copper smelting operations,
which it ceased in 1999.
2
3. On April 28, 2004, during its open meeting and public comment period, the Commission
received a request for hearing on the renewal issue.
4. On May 14, 2004, the Commission, exercised its plenary authority to hold a hearing in the
public interest and issued an interim order referring two issues to SOAH:
a. Whether the operation of the El Paso Copper Smelter under the terms of the
proposed permit will cause or contribute to a condition of air pollution; and
b. Whether the Applicant’ s compliance history for the last five years of
operation of the El Paso Primary Copper Smelter warrant the renewal of
Air Quality Permit No. 20345.
5. The Commission also assigned the burden of proof on these issues to ASARCO.
Procedural History and Parties
6. On January 27, 2005, the ALJs held a preliminary hearing in this matter at the University
of Texas at El Paso.
7. Notice of the preliminary hearing was published in the El Paso Times, a newspaper
generally circulated in El Paso County, on December 26, 2005, and mailed by the
Commission’ s Chief Clerk to persons who had previously requested such notice.
8. At the preliminary hearing, parties were admitted and aligned as follows:
3
ADMITTED PARTIES REPRESENTATIVE
ASARCO Mr. Eric Groten and Mr. Patrick Lee
City of El Paso (El Paso) Mr. Erich M. Birch
Executive Director (ED) Mr. Daniel Long and Mr. Brian MacLeod
Office of Public Interest Counsel (PIC) Ms. Anne Rowland
Sierra Club, et al. (Sierra Club)
• Quality of Life El Paso
• El Paso County Medical Society
• Get the Lead Out Coalition
• Senator Eliot Shapleigh, individually
• UTEP Students Against ASARCO
• UTEP Students Government
Association
• El Paso High Neighborhood
Association
• Matthew F. Carroll, individually
• Debra Kelly, individually
• Juan Garza, individually
Mr. Richard W. Lowerre
and Ms. L. Layla Aflatooni
Sandoval, et al. (Sandoval or Anapra Group)
• Southside Low Income Housing
Development
• Linda Sandoval, individually
• Michelle Velasco, individually
• Olga Arguelles, individually
Mr. Taylor Moore
Sunset Heights ACORN, et al. (ACORN)
• Henry L. Pfafflin, individually
• Edward C. Patrykus, individually
• Rodolfo Urias, individually
• Blanca Vega de Urias, individually
• Dr. Fidel Urrutia, individually
• Arturo Moreno, individually
Mr. Michael R. Wyatt, Mr. Enrique
Valdivia, and Ms. Veronica Carbajal
9. The PIC is currently represented by Emily A. Collins. Ms. Rowland has left the PIC.
4
10. On March 7, 2005, Juan Garza filed a motion to withdraw as a party. His motion was
granted via Order No. 10.
11. On May 31, 2005, the El Paso Medical Society filed a motion to withdraw as a party. Its
motions was granted via Order No. 24.
12. On March 31, 2005, the El Paso High Neighborhood Association filed a motion to
withdraw as a party. Its motions was granted via Order No. 9.
13. On March 31, 2005, Matthew F. Carroll filed a motion to withdraw as a party. His
motion was granted via Order No. 9.
14. Subsequent to the preliminary hearing, the ALJs established a docket control order
designed to complete the proceeding within the maximum expected duration set by the
Commission. In its Interim Order, the Commission set October 27, 2005, as the date by
which the PFD would be due.
15. The following are the principal procedural events in the case:
5
DATE PROCEDURAL SCHEDULE
Jan. 27, 2005 Preliminary hearing at which parties were designated and aligned.
March 14,
2005
Deadline for each party to serve TRCP 194 disclosures. Discovery began.
March 21,
2005
ASARCO pre-filed its direct-case evidence in writing, including all
testimony and exhibits.
May 6, 2005 First prehearing conference.
May 6, 2005 All parties, other than ASARCO, El Paso, and the ED, pre-filed their direct
case evidence in writing, including all testimony and exhibits.
May 13, 2005 Second prehearing conference.
May 18, 2005 El Paso pre-filed its direct-case evidence in writing, including all testimony
and exhibits.
May 23, 2005 Deadline to submit written discovery requests.
June 13, 2005 ED pre-filed his direct-case evidence in writing, including all testimony and
exhibits.
June 27, 2005 Close of discovery/Final day to take depositions/Deadline to file objections
to and motions to strike pre-filed evidence/Deadline for ASARCO to file
list of rebuttal witnesses and brief summary of each’ s rebuttal
testimony/Deadline to file dispositive motions.
July 5, 2005 Deadline to file responses to objections to pre-filed evidence and to
dispositive motions.
July 8, 2005 Third Prehearing conference.
July 11- 22,
2005
Hearing on the merits.
August 19,
2005
Deadline to file closing briefs.
August 29,
2005
Deadline to file replies to closing briefs.
October 27,
2005
Deadline to issue Proposal for Decision.
6
General Background
16. ASARCO has operated a smelting and refining operations at its El Paso facility for over onehundred
years. The original plant was built in 1887, along the Rio Grande, to process lead
ores from the mines in Mexico and the Southwest.
17. In 1899, the smelter incorporated into the American Smelting and Refining Company, and
it so operated until 1975, when the company officially became ASARCO, Incorporated.
18. The ASARCO EL Paso Plant is situated at the juncture of two countries (the United States
and Mexico) and three states (Texas, New Mexico, and the Mexican state of Chihuahua).
The ASARCO plant is located immediately north and east of the Rio Grande. It lies in the
Rio Grande Canyon between the Franklin Mountains and the Cerros del Muleros in Mexico.
19. The ASARCO EL Paso Plant is bounded by Interstate 10 on the east, Executive Center
Boulevard to the north, the American Canal to the southwest, and Paisano Boulevard to the
west.
20. Before closing operation, ASARCO smelted copper in El Paso using a Continuous Top-Feed
Oxygen Process (ConTop).
21. Permit 20345, which this case concerns, was issued by the Texas Air Control Board (TACB)
in 1992 to permit the new ConTop reactors at the ASARCO El Paso Plant.
22. The ConTop reactors replaced ASARCO’ s previously grandfathered copper-smelting
facilities.
7
23. ASARCO also holds Permit No. 4151, which authorizes unloading operations, certain
conveyance systems, and other operations up to and including the bedding building at the El
Paso plant.
24. ConTop was implemented in March 1993, and has been the exclusive operating unit used for
the production of copper anodes since then.
25. Since Permit 20345’ s 1992 issuance, several permit amendments and alterations have been
approved by the ED without contested case hearings.
26. Applicant ceased its copper smelting operations in 1999 and remains in an extended
condition of inoperation.
27. Until production was discontinued in 1999, copper smelting was ASARCO’ s primary
activity at its El Paso plant, resulting in the production of copper anodes that are sent to
other ASARCO facilities.
28. Applicant also generated sulfuric acid as part of the off-gas treatment process from the
emissions from the copper smelting process.
Air Pollution
Authorized Emissions
29. Permit 20345 contains a maximum allowable emission rate table (MAERT) that authorizes
ASARCO to emit the following, which the permit specifically refers to as “ air
contaminants,” at various locations and in various amounts:
10 • Particulate matter equal to or less than 10 microns in diameter (PM );
8
10 • Particulate matter (PM), including PM , often called total suspended particulates (TSP);
X 2 • Oxides of nitrogen (NO ), which includes nitrogen dioxide (NO );
• Volatile organic compounds (VOCs);
2 • Sulfur dioxide (SO );
• Carbon monoxide (CO);
2 4 • Sulfuric acid (H SO ); and
• Lead.
30. Additionally, ASARCO will emit other compounds, which are included in its PM and VOC
emissions.
31. In November 1994, an uncontested amendment to Permit 20345 was granted by the
Commission to adjust heavy metal emission rates from the original representations to actual
rates that were measured during required stack sampling. According to that amendment, the
following compounds were authorized to be emitted at various locations and in various
amounts:
• Arsenic
• Chromium
• Chrome VI
• Copper-dust
• Copper-fume
• Lead
• Nickel
• Zinc
• Chromium
• Chrome VI.
32. Although not specifically named in Permit 20345, the permit has authorized and if renewed
would authorize ASARCO to emit manganese, barium, carbon and cadmium
33. Permit 20345 has never authorized and would not authorize ASARCO to emit hydrogen
sulfide, beryllium, dioxins, furans, or fluoride.
9
34. If renewed, Permit 20345 would also authorize ASARCO to emit trace quantities of other
compounds that would not cause or contribute to air pollution.
NAAQS
35. The U.S. Environmental Protection Agency (EPA) has established National Ambient Air
2 2 10 2.5 Quality Standards (NAAQS) for lead, NO , CO, sulfur oxides (including SO ), PM , PM ,
and ozone. 40 Code of Federal Regulations (C.F.R.) §§ 50.4, 50.5, 50.6, 50.7, 50.8, 50.9,
50.10, 50.11, and 50.12.
36. Each of the NAAQS is listed below:
10
NAAQS
[micrograms per cubic meter (:g/m ) or parts per million (ppm) as indicated] 3
Pollutant Averaging Time Primary
Standard
Secondary
Standard
Carbon Monoxide 8-hour 9 ppm
10,000 :g/m3
None
1-hour 35 ppm
40,000 :g/m3
None
Lead Quarterly Average 1.5 :g/m Same as Primary 3
Nitrogen Dioxide Annual (Arithmetic Mean) 0.053 ppm
100 :g/m3
Same as Primary
10 PM Annual (Arithmetic Mean) 50 :g/m Same as Primary 3
24-hour 150 :g/m None 3
2.5 PM Annual (Arithmetic Mean) 15.0 :g/m Same as Primary 3
24-hour 65 :g/m None 3
Ozone 8-hour 0.08 ppm Same as Primary
Sulfur Oxides Annual (Arithmetic Mean) 0.03 ppm
80 :g/m3
None
24-hour 0.14 ppm
365 :g/m3
None
3-hour None 0.5 ppm
1300 :g/m3
The Commission’s NGLC Rules
37. With certain exceptions, the Commission generally prohibits any person in Texas from
causing, suffering, allowing, or permitting emissions of the following substances from sources
on contiguous properties to exceed the following net ground level concentrations (NGLCs):
11
Net Ground Level Concentration Standards
(:g/m unless otherwise indicated) 3
Substance Concentration Averaging Time
sulfur dioxide 0.4 ppm
(755 :g/m ) 3
30-minutes
TSP 200 3 hours
400 1 hour
2 4 H SO 15 24 hours
50 more than once in a 24-hour
period
100 any time
30 TEX. ADMIN. CODE ANN. (TAC) §§ 111.155, 112.3(a), and 112.41(a) (2005).
2 SO Area Control Plan
38. An “area control plan” is a site-specific regulatory scheme for which the owner of an
2 SO -emitting source can apply for approval as an alternative to compliance with the generally
2 applicable SO NGLC standard. 30 TAC § 112.19.
39. Upon application by a regulated entity and recommendation of the ED, the Commission may
approve such a regulatory control plan. 30 TAC § 112.20.
40. The area around the ASARCO El Paso plant is covered by an area control plan that sets 0.5
ppm, or 1137 :g/m , over two consecutive half-hour averages as the compliance standard. 3
Effects Screening Levels
41. Since at least the mid-1970s, the Commission staff has developed effects screening levels
12
(ESLs) for ground level concentrations of emitted constituents.
42. The ESLs are prepared by the staff of the Commission’s Toxicology Section and identify the
levels at which the members of that section believe that a constituent may unquestionably be
emitted without causing adverse health or other effects.
43. The staff uses toxicological information from animal studies, exposure limits set for
occupational situations, epidemiological studies, and Material Safety Data Sheets to identify
concentrations of constituents at which no adverse health effect has been observed. When
specific information on a constituent is not available, the staff uses information that is
available on constituents with similar chemical structures and toxicological properties to fill
the gaps.
44. Having determined a concentration of a contaminant with no reported or estimated health
effect, the staff divides that number by multiple safety factors of ten to account for
differences between animals and humans (when the underlying data was based on a study of
animals), between people (to account for particularly sensitive individuals), and in exposure
time and for the contribution of multiple sources of the same pollutant in an area.
45. Thus, to account for the shorter-term exposure effects, the staff generally sets a 24-hour
average ESL that is 1 percent of the occupational exposure limit. To account for longer-term
exposure effects, they generally set an annual average ESL that is 1/1000 of the occupational
standard.
46. ESLs set by the above-described method are very conservative and protective of children,
the elderly, and people with pre-existing conditions and account for long term exposures.
47. The Commission and its predecessor agencies have a long history of finding the above ESL
methodology sound in prior cases. See Asarco Incorporated, TACB Docket No. 92-07
13
(Board Order) (May 8, 1992)(FOF 22). Also Southwestern Refining Company, Inc.,
TNRCC Docket No. 95-0431-AIR (An Order Renewing Air Quality Permit
No. R-3153)(Jul. 13,1995)(FOF 27 and 28); and In the Matter of the Application of TXI
Operations, L.P. for Permit No. HW-50316-001 (An Order Granting the Application of TXI
Operations, L.P. for Permit No. HW-50316-001)(Mar. 10, 1999)(FOF 423 et seq.)
48. Below are ESLs for particulate matter components that ASARCO could emit under Permit
20345 if it were renewed:
Contaminant ESL
(:g/m ) 3
Averaging Time
Arsenic 0.4 24-hour
0.1 Annual
Chromium 0.4 24-hour
0.1 Annual
Chrome VI 0.4 24-hour
0.1 Annual
Copper-dust 4 24-hour
1 Annual
Copper-fume 0.4 24-hour
0.1 Annual
Nickel 0.06 24-hour
0.015 Annual
Zinc 20 24-hour
5 Annual
Manganese 2 24-hour
0.2 Annual
Barium 5 24-hour
14
0.5 Annual
Cadmium 0.1 24-hour
0.01 Annual
Iron salts 4 24-hour
1 Annual
Respirable silica 0.4 24-hour
0.1 Annual
Carbon
49. Carbon emissions, in and of themselves, would not cause adverse health or other effects.
VOCs
X 50. VOCs and NO form ozone, hence VOCs are indirectly regulated by the NAAQS for ozone.
51. ASARCO would not emit a quantity of any specific VOC that would cause adverse health or
other effects.
Copper and Iron Salts
52. Because the ESL exceedances are relatively small and the margin of safety in setting an ESL
is so large, a 24-hour copper dust concentration of 5.2 :g/m , which is 1.3 times the ESL, and 3
a 24-hour iron salt concentration of 4.43 :g/m , which is 1.1 times the ESL, would not cause 3
adverse health or other effects.
15
Respirable Silica
53. The primary concern with silica is a chronic effect, silicosis, hence the short-term level could
be much higher.
54. The California EPA’s chronic exposure level for respirable silica is 3 :g/m . 3
55. A 2.90 :g/m 24-hour and a 0.43 :g/m annual-average concentration of respirable silica 3 3
would not cause adverse health or other effects.
Arsenic
56. A 1.32 :g/m 24-hour average concentration of arsenic, though higher than the ESL, would 3
not cause adverse health or other effects.
57. The unpolluted, cleanest air in Texas, with no significant industrial sources of contaminants,
is near the McDonald Observatory in west Texas.
58. Annual-average arsenic concentrations in the air near the McDonald Observatory range from
0.01 to 0.02 :g/m . 3
59. Above-average levels of cancer are not found in the area near the McDonald Observatory.
60. A 0.2 :g/m annual-average of arsenic would not cause adverse health or other effects. 3
Risk Assessment
61. There is no such thing as zero risk.
16
62. EPA uses a range of risk factors from one-in-10,000 to one-in-a-million, depending on the
circumstances, and most state environmental agencies use that same range.
63. EPA and the Commission are approaching risk-analysis issues along similar lines in all
environmental programs and steadily moving toward greater consistency.
64. No state requires a one-in-a-million or lower risk level.
65. One-in-a-million is often used by environmental agencies as a de minimus value, which
requires no further scrutiny. A greater risk typically requires a more site-specific evaluation.
66. TCEQ’s Texas Risk Reduction Program (TRRP) concerns corrective action when land has
been contaminated. However, such land cleanups can impact air quality. 30 TAC §§ 350.1
and 350.2(a).
67. The TRRP generally requires cleanups to reduce emissions of carcinogenic air contaminants
to a risk level of 1-in-100,000 for off-site receptors except when a very detailed analysis of
exposure pathways indicates that few people are likely to be exposed. 30 TAC
§ 350.72(a)(1), 350.74 and 350.75.
68. EPA guidance states that even levels of risk calculated to be slightly in excess of 1-in-10,000
can still be acceptable, based on site-specific and chemical-specific information.
69. The Commission and EPA are moving toward a consistent carcinogenic-risk target of not more
than one-in-100,000 unless evidence indicating that far fewer than 100,000 people would be
exposed to the risk, which might make a target as low as one-in-10,000 acceptable.
17
EPA’s Integrated Risk Information System
70. EPA’ s Integrated Risk Information System (IRIS) electronic database is available on the
Internet and lists toxicity values for different exposure paths, e.g. oral, air, etc.
71. The methodologies underlying IRIS have been thoroughly reviewed by peer experts both
within and outside EPA.
72. IRIS inhalation reference concentrations (IRIS values) are calculated based on EPA’ s
assumptions of the average volume of air that an average-sized exposed person would
breathe in a day, e.g. 20 m /day and 70 kg body weight. 3
73. IRIS values are based on the assumption that a person is exposed to the concentration of a
contaminant for an entire lifetime.
74. IRIS values are calculated based on both cancer and non-cancer health risks and are intended
to be used to evaluate long-term community exposure rather than shorter exposure.
75. The IRIS values show concentrations of a contaminant that EPA has calculated could cause
one additional cancer in 10,000 one in 100,000 and one in one million for people who
receive lifetime exposure to the contaminant through that path.
76. IRIS inhalation exposure levels are useful in identifying potential health hazards and selecting
a response, but they have many limitations.
77. IRIS values have a uncertain spanning, i.e., margin of error, of perhaps an order of
magnitude.
18
78. IRIS inhalation concentrations cannot be validly used to accurately predict the incidence of
human disease or the types of effects that chemical exposures have on humans.
Cadmium
79. EPA has assigned a risk factor to cadmium.
80. The highest reported level of cadmium not affecting exposed workers is 10 :g/m per year. 3
81. In 1986, the annual-average concentration of cadmium in the ambient air in El Paso was
0.018 :g/m , and in 1987 it was 0.014 :g/m . 3 3
82. IRIS data indicates that there is a 0.0018 risk of additional lifetime cancers per :g/m of 3
lifetime exposure to cadmium.
83. Using IRIS data, an exposure to a 0.018 :g/m concentration of cadmium for life could lead 3
to 3.2 extra cancers per 100,000 population.
84. Except when a very detailed analysis of exposure pathways indicates that few people are likely
to be exposed, annual-average cadmium concentrations should not be greater than 0.01 :g/m , 3
which is the ESL.
Monitoring Data
85. From 1993 through 1999, when the ASARCO El Paso plant operated under the Permit 20345,
El Paso owned and operated an ambient-air monitor (El Paso Monitor) that was 1.25 miles east
of the ASARCO site.
19
86. TCEQ also owned and operated a monitor (TCEQ Monitor) that was 1.5 miles from the
ASARCO facility.
2 87. ASARCO maintained a network of five to six continuously operating ambient SO monitors
(ASARCO Monitors) around the El Paso plant since the 1970s. The ASARCO Monitors are
two to three miles to the southeast, east, and northwest of the facility.
88. The available data from the TCEQ and El Paso Monitors showed no exceedance of the
NAAQS for lead when ASARCO previously operated under Permit 20345. That data showed
the following peak lead level:
Monitored Maximum Levels Compared to NAAQS
Contaminant Averaging Period NAAQS Highest Level Year of Highest
Lead Quarterly Average 1.5 :g/m 0.4 :g/m 1996 3 3
10 89. There is some data from the TCEQ and El Paso Monitors for PM , but it is fragmentary and
has large gaps.
90. From 1993 through 1999, when ASARCO previously operated under Permit 20345, the
2 maximum 3-hour, 24-hour, and annual-average concentrations of SO recorded by the El Paso
and TCEQ Monitors never came close to the NAAQS, steadily moved downward until
approximately 2000, and then stayed at very small fractions of each NAAQS thereafter.
2 2 91. During that same period of operation, the following were the highest levels of NO and SO
recorded by either the TCEQ or the El Paso Monitor during the periods before, during, and
after ASARCO operated ConTop:
20
Monitored Maximum Levels Compared to NAAQS
(1992-2004)
averaging period NAAQS Highest Level Year of Highest
Sulfur Dioxide Annual Arithmetic Mean 0.03 ppm 0.012 ppm 1992
24-hour 0.14 ppm 0.095 ppm 1990
3-hour 0.5 ppm
(1300
:g/m ) 3
0.4 ppm 1990
2 NO Annual 0.053 ppm 0.023 ppm 1994-1996
92. The TCEQ Monitor and El Paso Monitor were placed to monitor the overall quality of the
air in the region and not specifically to monitor the localized impact of ASARCO’ s or any
other entity’ s emissions.
93. The computer modeling for lead that ASARCO submitted to the TACB when Permit 20345
was originally issued predicted a max GLC at a point west of the ASARCO facility, which
was miles from the TCEQ, El Paso, or ASARCO Monitors.
2 94. In March 1995, to support an amendment to its permit, ASARCO modeled SO for a broad
grid of points surrounding the ASARCO facility and produced a set of maps showing the
2 predicted annual SO concentrations. Those maps also showed the max GLC points that the
modeling predicted for each averaging period.
2 95. The March 1995 modeling predicted an annual SO max GLC, before accounting for
2 background SO contribution, of 15.6 :g/m at a point virtually on ASARCO’ s eastern 3
property line. The maximum predicted 30-minute impact was 1,135 :g/m at a point 3
approximately 8,200 feet, or 1.5 miles, east of the ASARCO facility. The 3-hour average
max GLC was northeast of the ASARCO facility.
21
2 96. The El Paso Monitor is approximately one-quarter mile from the 30-minute SO peak
location predicted in the 1995 modeling, but far less close to the other peak locations. The
TCEQ Monitors are even farther away from those max GLC locations.
2 97. The annual-average SO concentrations predicted in the 1995 modeling generally fall from
the predicted 15.6 :g/m peak at ASARCO’ s property line to approximately half of that a 3
mile to the east, while declining and then rising to 12 :g/m over two miles to the north. 3
98. For purposes of determining maximum ground level concentrations, short distances matter.
Air Dispersion Modeling
99. Atmospheric dispersion modeling, also called air dispersion modeling, is a computerized
mathematical tool based on the principles of physics that simulates the dispersion of an
emission from the source to the location where it is received and provides an estimate of the
concentration at the receptor location.
100. Various factors are fed into a computer program, which than predicts concentrations of the
contaminant at various locations. Among those factors are the type of contaminant; the
temperature, point, and elevation of the emission; the speed and direction of the wind; the
turbulence of the atmosphere; and the elevation of the surrounding terrain.
101. Atmospheric dispersion modeling is the most suitable tool for predicting the ambient
concentration of a particular pollutant that will result from the emissions from a particular
source.
102. The TCEQ staff and EPA rely almost exclusively on modeling to determine whether a
particular source will cause or contribute to a condition of air pollution.
22
103. For at least the last 30 years, both agencies’ policies and procedures have directed applicants
to use models.
Overview of Models
104. Though not required, EPA and the TCEQ staff currently prefer that regulated entities generally
use the Industrial Source Complex Model, Version 3 (ISC3) for the following regulatory
applications:
• Industrial source complexes;
• Rural or urban areas;
• Flat or rolling terrain;
• Transport distances less than 50 kilometers;
• 1-hour to annual averaging times; and
• Continuous toxic air emission.
40 C.F.R. Part 51, Appendix A to Appendix W of Part 51—Summaries of Preferred Air Quality
Models.
105. ASARCO has never modeled the dispersion of its emissions under Permit 20345 using ISC3.
106. In 1992, the preferred dispersion models were the original version of the Industrial Source
Complex Model (ISC1) and COMPLEX I, which was another EPA approved model that took
into account the changes in the elevation of the terrain over which the emissions would be
dispersed.
107. After ASARCO originally applied for Permit 20345 in 1991, it prepared and submitted ISC1
and COMPLEX I modeling runs in early 1992 (1992 Modeling).
23
108. In 1992, one had to run both the ISC1 and the COMPLEX I models to determine maximum
ground level concentrations (max GLCs) at different terrain elevations. One had to run the two
different models separately, look at the results, and see which ones were higher.
109. The ISC3 model incorporates the features of the ISC1 and COMPLEX I models.
110. Superficially, there are differences in the interfaces with the ISC models. In ISC3, one can
now use menus to enter data for emission rate, stack height, stack diameter, etc., and the menus
transmit that data to the model input file. In 1991, one had to enter that data directly into a
model input file in FORTRAN code. Numbers for the inputs–diameter, velocity, etc.–were
entered left to right with no separation between them.
111. The only other air dispersion modeling that ASARCO has ever run and submitted to TCEQ
concerning Permit 20345 was prepared to support its 1994 application to amend the permit.
2 That application was primarily to change its authorized SO emission rates.
2 112. To support that 1994 application, ASARCO in 1995 modeled those then-proposed SO
emissions using BEEST-X, Version 1.3 (1995 modeling). That was a private vender’s
software that combined the algorithms from a later version of ISC–Industrial Source Complex
Short-Term Model (ISC2)–and COMPLEX I into one model.
113. ASARCO did not prepare an up-to-date dispersion model for this case. Instead, ASARCO
relied on its 1992 and 1995 modeling, neither of which modeled all of the concentrations of
each pollutant that the Permit, if renewed, would authorize ASARCO to emit.
Appropriateness of Using Older Models
114. The basic science of plume dispersion has not changed.
24
Scaling
115. GLCs for any emission rate could be calculated based on the modeling of a different
emission rate for the same pollutant by calculating the ratio between the two rates and
multiplying the modeled GLC by that same ratio. This process is referred to as scaling.
116. In scaling from older modeling to support this renewal application, ASARCO always
reverted back to the latest modeling that was conducted for the pollutant at issue. In the case
2 of SO , it was the 1995 modeling. In the case of the other pollutants, it was the 1992
modeling. To avoid distortions, no number that was previously scaled was scaled again.
117. Scaling is an acceptable technique when performed correctly and is not forbidden by the
EPA’ s modeling guidance.
118. Scaling is not an appropriate approach when the heights of the emission sources dramatically
differ.
Conservatism In Modeling
119. In its 1992 and 1995 modeling, ASARCO made certain conservative assumptions that tended
to over-predict the maximum ground level concentrations that the modeled emissions would
cause.
120. That modeling assumed that all modeled sources were operating at the same time, the events
were coincidental, and there were worst-case meteorological conditions. All of those
conditions rarely occur at the same time, hence that was a very conservative set of
assumptions.
25
121. Although emissions hug the ground and cannot go up-hill unless there is turbulence, the ISC1
model assumed that ground-level and low-level releases from ASARCO could travel uphill
and impact receptors higher than their points of emission.
122. When receptors were lower than the ASARCO plant elevation, they were nevertheless
modeled as if they were at the plant grade. That meant that the plume did not have to travel
downward and disperse to reach the receptors, hence the results assumed that the receptors
were experiencing an exaggerated impact of those emissions.
123. Although the EPA guidance for the ISC models states that one should not model receptors
higher than the stack height, those higher receptors were modeled, which overestimated the
impacts on them.
124. Not all of ASARCO’ s sources would operate 24 hours a day, yet the modeling assumed that
ASARCO’ s sources were operating 24 hours a day. That included nighttime, in stable
conditions, which are worst-case conditions for many of ASARCO sources.
125. The rural instead of the urban switch was selected for the 1992 and 1995 modeling. Using
the urban switch assumed lower turbulence and predicted higher GLCs than the if the urban
switch had been used.
Partial vs. Full Receptor Grid
126. In 1992 and 1995, ASARCO modeled five relatively small grids near likely sensitive
receptors. Even in the early 1990s computer computation speeds were so slow that modeling
a broad range of points was very time-consuming and rather difficult. Modeling only the
five sensitive-receptor areas, rather than a very broad area, reduced the quantity, length, and
cost of the computer runs.
26
127. The five small receptor areas were:
• La Calavera, a residential area immediately adjacent to the ASARCO facility;
• Executive Center, another residential area a bit farther away;
• Mesita Elementary School, the school in Texas that is nearest to the ASARCO facility;
• The nearest dorms at the University of Texas at El Paso; and
• A fifth location that is not indicated in the evidence.
PM Emissions under Permit 20345 Slone
128. ASARCO’ s 1992 modeling predicted the following maximum ground level concentrations
10 from PM and PM emissions:
1992 PM Modeling Results
(:g/m ) 3
Averaging Time Highest Concentration EPA Significance Level NAAQS or
NGL
10 PM 24-hour 3.09 5 150
annual 0.37 1 50
PM 1-hour 311 - - 400
3-hour 177 - - 200
129. That 1992 modeling was prepared based on the following assumptions:
1992 PM Modeling Assumptions
(:g/m ) 3
lbs/hour tpy
PM 95 371
10 PM 92.5 368
27
130. The permit at issue in this case has been amended since 1992. The PM that ASARCO would
be authorized to emit is lower than it was in the 1992 permit. The renewed permit would
10 authorize 98 percent of the hourly and 95 percent of the annual PM and PM emissions that
the 1992 permit did. Rounding off, the renewed permit would authorize the following:
PM Emissions Under the Renewed Permit
(:g/m ) 3
lbs/hour tpy
PM 93.3 353
10 PM 90.7 350
131. There are a lot of options when modeling wide-spread fugitive dust sources. Their precise
location does not matter that much and can be modeled as a single source. That allows PM
emissions to be scaled as a group.
132. Since the renewed permit would authorize 98 percent of the hourly and 95 percent of the
10 annual PM and PM emissions that the 1992 permit did, scaling from the 1992 modeling
would yield the following max GLCs under Permit 20345 alone, if it is renewed:
Renewed Permit Maximum PM Impact Scaled from 1992 PM Modeling Results
(:g/m ) 3
Averaging Time Highest Concentration EPA Significance Level NAAQS or
NGLC
10 PM 24-hour 3.0282 5 150
annual 0.3515 1 50
PM 1-hour 304.78 - - 400
3-hour 168.15 - - 200
2.5 133. Due to significant technical difficulties in directly estimating PM from industrial facilities
28
and estimating secondarily-formed fine particles through chemical reactions in the
10 atmosphere, current EPA policy is to allow a source to use its PM NAAQS demonstration
2.5 as a surrogate for making a PM NAAQS demonstration.
2.5 134. Based on the above, the maximum ground level concentration of PM due to ASARCO
emissions under Permit 20345 alone can be estimated as follows:
2.5 Renewed Permit Maximum PM Impact Based on and Scaled from 1992 PM Modeling
Results
(:g/m ) 3
Pollutant Averaging Time Primary
Standard
NAAQS
2.5 PM 24-hour 3.0282 65 :g/m3
Annual 0.3515 15.0 :g/m3
Specific PM Constituents
135. PM is a catchall category that includes many particular pollutants.
136. In the 1992 modeling, ASARCO modeled contiguous ASARCO sources and predicted max
GLCs for the following PM constituents, which are compared to the non-polluting levels,
generally the ESLs, found above:
29
1992 Modeling of PM Constituents Compared to Maximum Non-polluting Levels
(:g/m ) 3
Contaminant 1992 Model Max GLC Non-polluting Level Averaging Times
Arsenic 0.11 1.32 24-hour
0.02 0.2 Annual
Chromium

Related Reports

http://www.soah.state.tx.us/pfdsearch/pfds/582/05/582-05-0593-po.pdf

Fair Use Notice
This site contains copyrighted material the use of which has not always been specifically authorized by the copyright owner. We are making such material available in our efforts to advance understanding of environmental, political, human rights, economic, democracy, scientific, and social justice issues, etc. We believe this constitutes a "fair use" of any such copyrighted material as provided for in section 107 of the US Copyright Law. In accordance with Title 17 U.S.C. Section 107, the material on this site is distributed without profit to those who have expressed a prior interest in receiving the included information for research and educational purposes. For more information go to: http://www.law.cornell.edu/uscode/17/107.shtml. If you wish to use copyrighted material from this site for purposes of your own that go beyond "fair use", you must obtain permission from the copyright owner.

e-mail this to a friend | print this article