United States Court of Appeals
FOR THE DISTRICT OF COLUMBIA CIRCUIT
Argued April 10, 2008 Decided June 6, 2008
No. 07-1053
NATURAL RESOURCES DEFENSE COUNCIL AND LOUISIANA
ENVIRONMENTAL ACTION NETWORK,
PETITIONERS
v.
ENVIRONMENTAL PROTECTION AGENCY,
RESPONDENT
AMERICAN CHEMISTRY COUNCIL,
INTERVENOR
On Petition for Review of an Order
of the Environmental Protection Agency
John D. Walke argued the cause for petitioners. With him
on the briefs were Aaron S. Colangelo, Patrice Simms, Adam
Babich, and Jill Witkowski.
David S. Gualtieri, Attorney, U.S. Department of Justice,
argued the cause for respondent. With him on the brief were
John C. Cruden, Deputy Assistant Attorney General, and
Michael W. Thrift, Counsel. Kent E. Hanson, Attorney, U.S.
Department of Justice, entered an appearance.
2
Leslie A. Hulse, Charles H. Knauss, Robert S. Taylor, and
Robert V. Zener were on the brief for intervenor.
Before: GRIFFITH and KAVANAUGH, Circuit Judges, and
SILBERMAN, Senior Circuit Judge.
Opinion for the court filed by Senior Circuit Judge
SILBERMAN.
SILBERMAN, Senior Circuit Judge: Synthetic organic
chemicals have few direct consumer uses, but they often serve
as raw materials in the production of plastics, rubbers, fibers,
protective coatings, and detergents. Petitioners, the Natural
Resources Defense Council and the Louisiana Environmental
Action Network, challenge EPA’s residual risk rulemaking
under subsection 112(f) of the Clean Air Act for facilities that
use or produce synthetic organic chemicals (“the industry”).
Petitioners also challenge EPA’s technology review under
subsection 112(d)(6). In a rather unusual bit of rulemaking, the
agency determined by rule not to change its previous rule, which
gave rise to petitioners’ challenge. We deny the petition.
I.
Section 112 of the Clean Air Act regulates hazardous air
pollutants. When the Act was passed in 1970, hazardous air
pollutant was defined as a substance “which may reasonably be
anticipated to result in an increase in mortality or an increase in
serious irreversible, or incapacitating irreversible, illness.”
Sierra Club v. EPA, 353 F.3d 976, 979 (D.C. Cir. 2004) (citation
omitted). The Administrator of the EPA was required to prepare
a list of air pollutants that fell within this definition, then
promulgate standards to protect the public health from these
substances with an “ample margin of safety.” Id.
3
As we have explained, this arrangement proved
problematic. Id. In light of unrealistic time frames and
scientific uncertainly over which substances posed a threat to
public health, EPA only listed eight pollutants as hazardous
between 1970 and 1990. Id. In 1990, Congress sought to hasten
the process by adopting a new regulatory approach for
hazardous air pollutants. Rather than have EPA list one-by-one
those substances likely to be harmful, the amended version of
section 112 provides a list of 191 substances that Congress
deemed to be hazardous. 42 U.S.C. § 7412(b)(1). EPA could
subsequently add to or subtract from this list. Id. § 7412(b)(2)-
(3).
Instead of basing its regulations on health risks (the “ample
margin of safety”), EPA was required by the 1990 amendments
to adopt technology-based standards in the first instance. That
is to say, in the first round of regulation, the agency was obliged
to look to the best available control technology to control
emissions for each category of major sources that emits one or
more of the listed hazardous air pollutants.1 Id. § 7412(d)(2)-
(3). (The control technology used to meet this standard is
referred to as the “maximum achievable control technology” or
“MACT.”) Under this technology-based approach, standards for
new sources may not be less stringent than “the emission control
that is achieved in practice by the best controlled similar
source.” Id. § 7412(d)(3). And for existing sources, the
emission standards may not be less stringent than “the average
emission limitation achieved by the best performing 12 percent
1
The Act defines “major source” as “any stationary source or
group of stationary sources located within a contiguous area and under
common control that emits or has the potential to emit considering
controls, in the aggregate, 10 tons per year or more of any hazardous
air pollutant or 25 tons per year or more of any combination of
hazardous air pollutants.” 42 U.S.C. § 7412(a)(1).
4
of the existing sources.” Id. After setting the “floor” – i.e., the
minimum required reduction in emissions for a new or existing
source – EPA has discretion to require an even greater reduction
in emissions, taking into account costs, health effects,
environmental effects, and energy requirements.2 Id. §
7412(d)(2).
In the second stage of regulation, EPA was obliged to
review any residual health risks that had not been eliminated by
the initial technology-based standards. Id. § 7412(f). This
second stage is described as “risk-based” or “health-based”
because it requires EPA to set a standard based on a medical
assessment of a given pollutant’s health risks (as was true of the
pre-1990 statute), rather than the current state of control
technology. See generally Percival, Schroeder, Miller & Leape,
ENVIRONMENTAL REGULATION: LAW, SCIENCE & POLICY 126
(4th ed. 2003). Within six years of promulgating the technology-
based standards, EPA was required to prepare a report to
Congress analyzing any residual health risks. If Congress did
not act on the report, then EPA was to conduct residual risk
analysis under subsection 112(f)(2).
***
EPA initially promulgated technology-based emission
standards for the industry in 1994 (there are 238 facilities in the
2
Although EPA considered costs (as we discuss supra) in
setting its technology-based standards, we subsequently held that it
was not appropriate to consider costs in establishing the maximum
achievable control technology “floor” (although the agency is
permitted to consider cost in deciding whether to require further
“beyond the floor” reductions). See Nat’l Lime Ass’n v. EPA, 233
F.3d 625, 640 (D.C. Cir. 2000); see also NRDC v. EPA, 489 F.3d
1364, 1375-76 (D.C. Cir. 2007).
5
United States that produce or use synthetic organic chemicals).3
Those standards required the use of control technologies such as
recovery devices, thermal oxidizers, carbon absorbers, and
steam strippers. After submitting the required report to
Congress in 1999, the agency commenced residual risk
rulemaking, apparently because – as we discuss below – it read
the statute as requiring a rulemaking proceeding to consider
whether to revise the technology-based standards, since the
industry’s emissions pose lifetime excess cancer risks of greater
than one-in-one million.
In the notice of proposed rulemaking, EPA listed two
options for the residual risk rulemaking, one of which would
have imposed somewhat stricter standards. 71 Fed. Reg. 34,422,
34,438 (2006). But the other, which EPA adopted in the final
rule, 71 Fed Reg. 76,603 (2006), was a reaffirmation of the
existing rule. EPA determined that under the existing
technology-based standard, no individual would face an excess
lifetime cancer risk of greater than 100-in-one million, which
EPA regards as the “presumptively acceptable” level under its
precedents. In the same regulatory procedure, EPA sought to
satisfy another statutory requirement, subsection 112(d)(6),
which commands the Administrator to “review, and revise as
necessary” the technology-based standards in light of
technological developments at least every eight years. 42
U.S.C. § 7412(d)(6). It concluded there were no such
developments.
Petitioners challenge EPA’s actions on several grounds.
Their primary argument is one of statutory construction. They
contend that the statute obliged EPA, in the residual risk
rulemaking, to tighten the standards for the industry so that the
3
See National Emission Standards for Organic Hazardous
Air Pollutants, 59 Fed. Reg. 19,402 (1994).
6
lifetime excess cancer risk to exposed persons would be no
greater than one-in-one million. It is also argued that in
reviewing the technology-based standards, EPA violated
subsection 112(d)(6) by taking costs into account. Alternatively,
even if EPA complied with the statute, petitioners claim that the
rulemaking violated the APA, as arbitrary and capricious,
because it relied on faulty data and overlooked significant
sources of emission.
II.
A.
Petitioners contend that subsection 112(f)(2)(A) obliged
EPA to revise industry standards to reduce lifetime excess
cancer risk to one-in-one million. Petitioners rely primarily on
the last sentence of that subsection, whereas EPA looks to the
whole subsection. That provision states in full:
If Congress does not act on any recommendation submitted
under paragraph (1), the Administrator shall, within 8 years
after promulgation of standards for each category or
subcategory of sources pursuant to subsection (d) of this
section, promulgate standards for such category or
subcategory if promulgation of such standards is required
in order to provide an ample margin of safety to protect
public health in accordance with this section (as in effect
before November 15, 1990) or to prevent, taking into
consideration costs, energy, safety, and other relevant
factors, an adverse environmental effect. Emissions
standards promulgated under this subsection shall provide
an ample margin of safety to protect public health in
accordance with this section (as in effect before November
15, 1990), unless the Administrator determines that a more
7
stringent standard is necessary to prevent, taking into
consideration costs, energy, safety, and other relevant
factors, an adverse environmental effect. If standards
promulgated pursuant to subsection (d) of this section and
applicable to a category or subcategory of sources emitting
a pollutant (or pollutants) classified as a known, probable or
possible human carcinogen do not reduce lifetime excess
cancer risks to the individual most exposed to emissions
from a source in the category or subcategory to less than
one in one million, the Administrator shall promulgate
standards under this subsection for such source category.
42 U.S.C. § 7412(f)(2)(A).
It is undisputed that facilities that produce or use synthetic
organic chemicals emit carcinogens and are, therefore, within
the reach of the last sentence. It is also undisputed that, in light
of the fact that existing technology-based standards do not
reduce the risk to less than one-in-one million, EPA was obliged
to “promulgate standards” under subsection 112(f). Petitioners
contend that the third sentence obviously means that residual
risk standards must meet the threshold test – i.e., EPA must
reduce such risks to one-in-one million. That may well be a
possible interpretation, but the sentence contains a glaring
omission; it does not say what petitioners would like us to infer.
Rather, that sentence instructs the Administrator to “promulgate
standards,” but it says nothing about the substantive content of
those standards. If Congress had wished to set a “bright line”
standard, it would have been rather easy for the draftsmen to say
just that. The failure to do so could not have been accidental.
In light of the rest of the subsection’s language (and other
provisions), it seems to us that the subsection was drafted as a
8
deliberately ambiguous compromise.4
We reach that conclusion because the second sentence,
which sets forth the substantive standard to be applied, simply
calls for standards that “provide an ample margin of safety to
protect public health” (unless the Administrator wishes to go
further to avoid adverse environmental effects). No distinction
is drawn between carcinogens and non-carcinogens. The third
sentence, on which petitioners rely, not only lacks the language
that petitioners ask us to infer; it also specifically states that if
the one-in-one million trigger is met, the Administrator must
promulgate standards “under this subsection,” which, perforce,
takes us back to the second sentence.5
EPA’s construction of the subsection is bolstered by another
paragraph, 112(f)(2)(B), which states:
4
Congress rejected the Senate version of the bill – which
mandated a bright line standard for carcinogens – in favor of the
House version, which gave the Secretary more discretion under the
“ample margin of safety” standard. Compare A LEGISLATIVE
HISTORY OF THE CLEAN AIR ACT AMENDMENTS OF 1990, at 4445, with
id. at 2139-40.
5
As one commentator has noted, subsection 112(f)(2) “does
not require that the residual risk standard for a category be set at a
level that would force the highest risk source in that category to
achieve the one-in-one-million benchmark, but merely mandates an
additional round of regulation.” Bradford C. Mank, What Comes After
Technology: Using an “Exceptions Process” to Improve Residual Risk
Regulation of Hazardous Air Pollutants, 13 STAN. ENVTL. L.J. 263,
276 (1994). The author continues: “By not requiring a one-in-a-
million or any other residual risk standard in section 112(f), Congress
essentially left the difficult task of defining an ‘ample margin of
safety’ to the EPA’s discretion.” Id. at 277.
9
Nothing in subparagraph (A) or in any other provision of
this section shall be construed as affecting, or applying to
the Administrator’s interpretation of this section, as in
effect before November 15, 1990, and set forth in the
Federal Register of September 14, 1989 (54 Federal
Register 38044).
42 U.S.C. § 7412(f)(2)(B). The cited item in the Federal
Register is EPA’s emission standard for benzene, which is a
carcinogenic hazardous air pollutant. In the Benzene
rulemaking, EPA set forth its interpretation of “ample margin of
safety,” as that term was used in the 1970 version of the Clean
Air Act. It said that the “ample margin” was met if as many
people as possible faced excess lifetime cancer risks no greater
than one-in-one million, and that no person faced a risk greater
than 100-in-one million (one-in-ten thousand). 54 Fed. Reg. at
38,044-45. In other words, the Benzene standard established a
maximum excess risk of 100-in-one million, while adopting the
one-in-one million standard as an aspirational goal. This
standard, incorporated into the amended version of the Clean Air
Act, undermines petitioners’ assertion that EPA must reduce
residual risks to one-in-one million for all sources that emit
carcinogenic hazardous air pollutants.
Petitioners respond that subsection 112(f)(2)(B) is a savings
clause that only preserves EPA’s specific regulations regarding
benzene. But the text belies this contention. Subsection
112(f)(2)(B) makes clear that nothing in subparagraph (A) shall
be construed as “affecting, or applying to the Administrator’s
interpretation” of section 112, as set forth in the Benzene
standard. The word “interpretation” indicates that the savings
clause is not limited to EPA’s benzene-specific determinations,
but applies broadly to the agency’s construction of the Clean Air
Act in the Benzene standard. Petitioners also contend that
subsection (B) should only be read as applying to non-
10
carcinogens, but this is not persuasive. Subsection 112(f)(2)(B)
incorporates EPA’s “interpretation” of the Clean Air Act from
the Benzene standard, and the text of this provision draws no
distinction between carcinogens and non-carcinogens. Indeed,
benzene is itself a carcinogen, 54 Fed. Reg. at 38,048, so it
would make little sense for Congress to incorporate this standard
only for non-carcinogens.
The parenthetical clause in the second sentence of
subsection 112(f)(2)(A) lends further support to EPA’s position.
That sentence states “[e]missions standards promulgated under
this subsection shall provide an ample margin of safety to
protect public health in accordance with this section (as in effect
before November 15, 1990) . . . .” EPA interprets the
parenthetical as a “shorthand reference” to the Benzene standard,
given that subsection (B) uses almost identical language,
incorporating “the Administrator’s interpretation of this section,
as in effect before November 15, 1990, and set forth in the
Federal Register . . . .” The phrase “this section (as in effect
before November 15, 1990)” is certainly broad enough to
encompass EPA’s prior interpretations of “this section” as well
as the text itself. In fact, the operative provision of the pre-1990
version of section 112 uses the exact same “ample margin of
safety” language as subsection 112(f)(2)(A) currently uses. See
Sierra Club, 353 F.3d at 979-80. Thus, the parenthetical must
refer to something more than the bare text of “this section,” or
else it would be surplusage.
Petitioners insist that EPA’s interpretation renders the third
sentence effectively meaningless. To be sure, the third sentence,
as EPA interprets it, seems relatively anodyne; it lacks
substantive force. But, at least as EPA reads it, the word
“promulgate” means the agency is obliged to conduct a
rulemaking to consider residual risks for sources that emit
carcinogens. That extra procedural step is not a trivial
11
obligation. Congress often imposes procedural requirements
without dictating substantive outcomes. See, e.g., Strycker’s
Bay Neighborhood Council v. Karlen, 444 U.S. 223, 227-28
(1980) (discussing the National Environmental Policy Act). We
also disagree with petitioners’ argument that EPA did not
“promulgate standards” under subsection 112(f)(2) because it
simply readopted the initial standards. This position finds no
support in the text of the statute. Subsection 112(f)(2) only
mandates that residual risk standards “provide an ample margin
of safety to protect public health.” If EPA determines that the
existing technology-based standards already provide an “ample
margin of safety,” then the agency is free to readopt those
standards during a residual risk rulemaking.
Finally, petitioners argue that EPA unlawfully considered
cost while setting the “ample margin of safety” in the residual
risk standards. Petitioners are correct that the Supreme Court
has “refused to find implicit in ambiguous sections of the [Clean
Air Act] an authorization to consider costs that has elsewhere,
and so often, been expressly granted.” Whitman v. Am. Trucking
Ass’n, 531 U.S. 457, 467 (2001). In this case, however, we
believe the clear statement rule has been satisfied. As explained
above, subsection 112(f)(2)(B) expressly incorporates EPA’s
interpretation of the Clean Air Act from the Benzene standard,
complete with a citation to the Federal Register. In that
rulemaking, EPA set its standard for benzene “at a level that
provides ‘an ample margin of safety’ in consideration of all
health information . . . as well as other relevant factors including
costs and economic impacts, technological feasibility, and other
factors relevant to each particular decision.” 54 Fed. Reg. at
38,045 (emphasis added). EPA considered cost in Benzene, and
subsection 112(f)(2)(B) makes clear that nothing in the amended
version of the Clean Air Act shall “affect[]” the agency’s
interpretation of the statute from that rulemaking.
12
In sum, we conclude that EPA’s interpretation of subsection
112(f)(2), although not an inevitable one, certainly is, at least, a
reasonable construction of the statute. See Chevron U.S.A., Inc.
v. NRDC, 467 U.S. 837, 843 (1984).
B.
Petitioners’ second statutory argument is based on
subsection 112(d)(6), which states:
The Administrator shall review, and revise as necessary
(taking into account developments in practices, processes,
and control technologies), emissions standards promulgated
under this section no less often than every 8 years.
42 U.S.C. § 7412(d)(6). It is argued that EPA was obliged to
completely recalculate the maximum achievable control
technology – in other words, to start from scratch. We do not
think the words “review, and revise as necessary” can be
construed reasonably as imposing any such obligation. Even if
the statute did impose such an obligation, petitioners have not
identified any post-1994 technological innovations that EPA has
overlooked.
More troublesome, however, is petitioners’ assertion that
the agency improperly considered costs in considering whether
to revise the standards. EPA did, in fact, state in its notice of
proposed rulemaking that “leakless components” should not be
considered the “maximum achievable control technology”
because of the high cost of replacing existing components. 71
Fed. Reg. at 34,438. That could be thought in tension with our
cases holding that EPA may not consider costs in setting the
maximum achievable control technology “floors,” but only in
determining whether to require “beyond the floor” reductions in
13
emissions.6 See Nat’l Lime Ass’n v. EPA, 233 F.3d 625, 640
(D.C. Cir. 2000); see also NRDC v. EPA, 489 F.3d 1364, 1375-
76 (D.C. Cir. 2007). EPA may have done just that in setting the
initial floors. Yet the time period for challenging those
standards has long since passed, 42 U.S.C. § 7607(b)(1), which
raises the question whether EPA’s reaffirmation of its cost-based
reasoning in its technology review gives rise to a new
opportunity for petitioners to challenge this apparent defect.7
Fortunately, we do not have to decide this question because
in its final rule, EPA squarely found that there were no
“significant developments in practices, processes and control
technologies,” and petitioners do not challenge this conclusion.8
71 Fed. Reg. at 76,605. Since that is the core requirement of
subsection 112(d)(6) and EPA’s finding satisfies that
requirement, it is irrelevant whether EPA considered costs in
arriving at the initial MACT floor and reaffirming that standard
in the residual risk rulemaking. Petitioners argue that EPA’s
consideration of cost somehow “tainted” the entire technology
review. But under the deferential “arbitrary and capricious”
6
As we indicated, supra, costs may be considered in
appraising risk – i.e., the “ample margin of safety.” 42 U.S.C. §
7412(f)(2)(B); 54 Fed. Reg. at 38,045.
7
In its brief, EPA asserts that it may consider “cost,” but the
agency draws no distinction between MACT floors and beyond-the-
floor reductions. Nor does the agency address our decision in
National Lime. Petitioners, on the other hand, do not discuss the
significance of the fact that the initial MACT standards were
unchallenged.
8
Even if petitioners did dispute this point, it involves a
factual finding; we would thus grant significant deference to the
agency’s conclusion. See NRDC v. EPA, 194 F.3d 130, 136 (D.C. Cir.
1999) .
14
standard, we may not set aside an agency’s factual finding based
on amorphous allegations of “taint.”
III.
There remains petitioners’ claim that EPA’s analysis of the
residual health risks from facilities that use or produce synthetic
organic chemicals was arbitrary and capricious (unreasonable).
In conducting its risk assessment, EPA relied upon industry-
supplied data – collected by the American Chemistry Council –
that was submitted on a questionnaire approved by the agency.
Based on this data, EPA determined that no source presented a
lifetime cancer risk of greater than 100-in-one million, and that
only two sources presented a risk equal to that threshold.
Petitioners argue that EPA should have handled the data
collection itself, and that the industry-supplied data was
defective in several respects.
Under section 114 of the Clean Air Act, EPA “may require”
the owner of an emissions source to keep records, make reports,
install monitoring equipment, take emissions samples, and
“provide such other information as the Administrator may
reasonably require.” 42 U.S.C. § 7414(a). Petitioners contend
that EPA’s risk analysis was flawed because the agency did not
exercise its authority under that section; instead, EPA relied
upon data voluntarily supplied by the industry. But section 114
is not a mandatory provision – it only states that EPA “may”
require sources to supply data. This wording gives the
Administrator discretion to decide what types of data should be
used for a risk assessment. Indeed, EPA has explained that
relying on data from industry sources is a well-established
practice. In its 1999 report to Congress on how the agency
planned to address residual risks, EPA stated that “source and
emissions data can be derived from broad-scale emissions
15
inventories, specific data collection efforts with particular
industries, or information from regional, State, or local air toxics
agencies.” Residual Risk Report to Congress (Mar. 1999), at 35
(emphasis added). As EPA’s counsel explained at oral
argument, it is very costly and time-consuming – for both the
agency and the emissions sources – to issue information requests
under section 114. Tr. of Oral Arg. at 21-22. It was therefore
not unreasonable for the agency to decline to invoke its section
114 authority when more efficient data-collection methods were
available.
Nevertheless, petitioners assert that the industry-supplied
data was flawed. They contend that many of the questionnaires
were incomplete, and that the data will understate health risks
because high-emissions sources have an incentive to withhold
data from the agency. Although there were some gaps in the
data, EPA ultimately received responses from 44% of all
sources, including sources with both low and high emissions
levels. The agency also explained that when certain data points
were missing, EPA used “environmentally protective defaults”
in its models. (Petitioners do not dispute this point.) In other
words, EPA acknowledged that the data was not comprehensive,
but compensated for this uncertainty by erring on the side of
protecting public health. We think that is a reasonable position.
We also disagree with petitioners’ assertion that the
emissions data – which dated back to 1999 – was unreliable.
Although the risk assessment was completed in 2006, it
obviously began much earlier than that. Data from 1999 seems
a bit old, but EPA explained that a significant amount of time
was needed to collect the data, run the models, analyze the
results, and prepare the rulemaking. EPA persuasively argues
that by 1999, the technology-based standards had already been
in place for several years, so it was unlikely that there would be
a substantial increase in emissions between 1999 and 2006.
16
Petitioners, moreover – and this is a key omission – do not assert
that emissions actually increased over this period.
EPA compared the industry-supplied data to the National
Emissions Inventory, which is a database of air emissions
information supplied by state and local air agencies, tribes, and
industry sources.9 The agency determined that the two data sets
yield similar results in terms of maximum risk. Petitioners
challenge this conclusion, pointing to specific facilities in which
the Emissions Inventory data show higher risks than the
industry-supplied data. But EPA explained that the Emissions
Inventory data for each facility includes hazardous air pollutants
other than synthetic organic chemicals, which means that the
Emissions Inventory is likely to show higher risks than the
industry-supplied data. Intervenors further emphasize that in at
least one of the three “high risk” facilities in the Emissions
Inventory data, the emissions of synthetic organic chemicals
were less than half of the facility’s overall emissions. EPA
noted that the Emissions Inventory included conservative
assumptions that caused this data set to overstate the risks from
synthetic organic chemicals. And EPA never contended that the
two data sets perfectly overlapped; it only stated that “the
highest risks from using the [Emissions Inventory] data were of
the same order of magnitude as those using the industry data.”
71 Fed. Reg. at 76,607.
Be that as it may, EPA’s analysis of the Emissions
Inventory data was only used to check the agency’s risk analysis
based on the industry-supplied data. Even if the correlation
were not perfect, that would not necessarily show that EPA’s
use of industry-supplied data was unreasonable. Indeed, even
though the industry data does not include as many facilities as
the Emissions Inventory data, it is more detailed in other ways.
9
See http://www.epa.gov/ttn/chief/net/neiwhatis.html.
17
For example, the industry data provides far greater detail on
emission point locations and release parameters than the
Emissions Inventory.
In sum, petitioners’ arguments boil down to one simple
point: EPA could have used better data in conducting its risk
analysis. Whether or not this is true, it misstates the inquiry
under the arbitrary and capricious standard. As we have
explained:
EPA typically has wide latitude in determining the extent of
data-gathering necessary to solve a problem. We generally
defer to an agency’s decision to proceed on the basis of
imperfect scientific information, rather than to invest the
resources to conduct the perfect study.
Sierra Club v. EPA, 167 F.3d 658, 662 (D.C. Cir. 1999) (citation
omitted). In other words, the sole question before us is whether
EPA has acted reasonably, not whether it has acted flawlessly.
On the record before us, EPA explained why it chose to rely on
industry-supplied data, and it reasonably responded to
petitioners’ objections to its data analysis.
***
Petitioners finally argue that EPA failed to address several
different types of emissions from synthetic organic chemical
manufacturing facilities, such as emissions from cooling towers,
emissions of inorganic hazardous air pollutants, and emissions
from “clusters” of nearby facilities. In making these arguments,
petitioners often reiterate their contentions that EPA relied upon
faulty data and did not reduce risks to one-in-one million. We
rejected those arguments above, and we need not address them
again here. With respect to the other arguments, we have
considered them and we find them to be without merit. EPA
18
adequately responded to each of the alleged deficiencies in the
residual risk assessment.
IV.
For the aforementioned reasons, the petition for review is
denied.
So ordered.