United States Court of Appeals
FOR THE DISTRICT OF COLUMBIA CIRCUIT
Argued May 3, 2012 Decided July 20, 2012
No. 10-1252
NATIONAL ENVIRONMENTAL DEVELOPMENT ASSOCIATION'S
CLEAN AIR PROJECT,
PETITIONER
v.
ENVIRONMENTAL PROTECTION AGENCY,
RESPONDENT
LOUISIANA DEPARTMENT OF ENVIRONMENTAL QUALITY, ET
AL.,
INTERVENORS
Consolidated with 10-1254, 10-1255, 10-1256, 10-1258,
10-1259, 11-1073, 11-1080, 11-1081, 11-1090, 11-1092
On Petitions for Review of a Final Action of the
United States Environmental Protection Agency
Paul M. Seby, Special Assistant Attorney General, Office of
the Attorney General for the State of North Dakota, et al.,
argued the cause for petitioner State of North Dakota. Mark W.
DeLaquil argued the cause for petitioner ASARCO LLC. With
them on the briefs were Wayne Stenehjem, Attorney General,
Office of the Attorney General for the State of North Dakota,
2
Margaret I. Olson, Assistant Attorney General, Andrea Bear
Field, Lucinda Minton Langworthy, Robert N. Steinwurtzel,
Leslie Sue Ritts, Benjamin Joseph Alke, Robert K. Baldwin,
Marty J. Jackley, Attorney General, Office of the Attorney
General for the State of South Dakota, Roxanne Giedd, Deputy
Attorney General, Catherine Cortez Masto, Attorney General,
Office of the Attorney General for the State Nevada, Jasmine K.
Mehta, Deputy Attorney General, James D. “Buddy” Caldwell,
Attorney General, Office of the Attorney General for the State
of Louisiana, Megan K. Terrell, Assistant Attorney General,
Herman Robinson, Donald J. Trahan, Deidra L. Johnson,
Gregory W. Abbott, Attorney General, Office of the Attorney
General for the State of Texas, and Nancy E. Olinger, Assistant
Attorney General. Nhu Q. Nguyen, Senior Deputy Attorney
General, Office of the Attorney General for the State of Nevada,
entered an appearance.
Madison B.C. Miller was on the brief for amicus curiae
Oklahoma Department of Environmental Quality in support of
state petitioners, non-state petitioners and supporting
intervenors.
Norman L. Rave Jr. , Attorney, U.S. Department of Justice,
argued the cause for respondent. With him on the brief were
Steven Silverman, Attorney, United States Environmental
Protection Agency, and Michael Thrift, Attorney.
Seth L. Johnson and David S. Baron were on the brief for
respondent-intervenors American Lung Association, et al.
Before: SENTELLE, Chief Judge, KAVANAUGH, Circuit
Judge, and GINSBURG, Senior Circuit Judge.
Opinion for the Court filed by Chief Judge SENTELLE.
3
SENTELLE, Chief Judge: Several states and state regulatory
agencies, together with corporations and industrial associations,
petition for review of the Environmental Protection Agency’s
rule entitled “Primary National Ambient Air Quality Standard
for Sulfur Dioxide,” and of the subsequent denial of petitions for
reconsideration of the standard. Petitioners contend, first, that
EPA failed to follow notice-and-comment rulemaking
procedures, and second, that the agency arbitrarily set the
maximum sulfur dioxide (SO2) concentration at a level lower
than statutorily authorized. For the reasons discussed more fully
below, we conclude that the challenge to the rulemaking
procedure is not within our jurisdiction and must be dismissed.
We further conclude that EPA did not act arbitrarily in setting
the level of SO2 emissions and therefore deny that portion of the
petitions for review.
I. Background
A. The Clean Air Act
The Clean Air Act (CAA) in §§ 108 and 109 requires EPA
to establish, review, and revise air quality criteria and standards,
allowing an “adequate margin of safety.” 42 U.S.C. §§ 7408,
7409. The 1970 amendments to the Act required the
Administrator to publish a list of air pollutants it intended to
regulate under the Act, including all those pollutants the
Administrator found reasonably could be anticipated to
endanger public health. 42 U.S.C. § 7408(a)(1). For each listed
pollutant, the Administrator had to issue air quality criteria that
“accurately reflect[ed] the latest scientific knowledge useful in
indicating the kind and extent of all identifiable effects on public
health or welfare which may be expected from the presence of
such pollutant in the ambient air, in varying quantities,”
including the effects of a pollutant when it combines with other
factors such as atmospheric conditions or other pollutants. 42
4
U.S.C. § 7408(a)(2).
The CAA required the Administrator to promulgate a
primary and secondary National Ambient Air Quality Standard
(NAAQS) for each listed pollutant by 1971 and to review and
revise those standards as appropriate every five years. 42 U.S.C.
§ 7409(a), (d)(1). The Act requires that the primary standards
“be ambient air quality standards the attainment and
maintenance of which in the judgment of the Administrator,
based on such criteria and allowing an adequate margin of
safety, are requisite to protect the public health.” 42 U.S.C.
§ 7409(b)(1).
The Act vests each State with “the primary responsibility
for assuring air quality within the entire geographic area
comprising such State . . . .” 42 U.S.C. § 7407(a). After EPA
promulgates a new final standard, the Act gives States a chance
to recommend whether areas within their boundaries should be
designated as “nonattainment,” “attainment,” or
“unclassifiable,” and the Agency makes the final designation.
42 U.S.C. § 7407(d). States then must submit State
Implementation Plans (SIPs), which, after receiving EPA
approval, impose federally enforceable controls on air pollution
sources so States can attain and maintain the NAAQS. 42
U.S.C. §§ 7410, 7502, 7514-7514a.
B. Regulatory Background: Sulfur Dioxide NAAQS
Sulfur dioxide, a “highly reactive colorless gas,” derives
mostly from fossil fuel combustion. It smells like rotting eggs
and, at elevated concentrations in the air, can cause acid rain. Its
presence in the ambient air can cause adverse health effects,
particularly in asthmatics. See Am. Lung Ass’n v. EPA, 134
F.3d 388, 389 (D.C. Cir. 1998).
5
On April 30, 1971, EPA promulgated the first primary
NAAQS for SO2 concentrations in the ambient air. 36 Fed. Reg.
8186 (Apr. 30, 1971). The standard set a 24-hour concentration
limit of 140 parts per billion (ppb) SO2, and an annual average
limit of 30 ppb. Id. at 8187. Over the next three decades, EPA
reviewed the standard, but did not revise it. See 74 Fed. Reg.
64810, 64813 (Dec. 8, 2009) (providing history of the SO2
NAAQS).
In 1988, EPA declined to revise the NAAQS, but requested
comment on a proposal to add a new 1-hour primary standard of
400 ppb to protect against five- to ten-minute bursts of SO2
concentrations. 53 Fed. Reg. 14926 (Apr. 26, 1988). In
response to those comments and other developments, in 1994,
EPA offered several more options for comment, including the
addition of a five-minute standard of 600 ppb. 59 Fed. Reg.
58958 (Nov. 15, 1994). After concluding its review of these
proposals and comments in 1996, EPA announced it would not
revise the NAAQS. In its review, it found that under the current
standards at that time, thousands of asthmatics could be exposed
to enough short-term bursts of SO2 that their lung function could
be impaired. 61 Fed. Reg. 25566, 25572 (May 22, 1996). EPA
concluded, however, that such effects “do not pose a broad
public health problem when viewed from a national perspective”
and did not warrant revisions to the SO2 NAAQS. Id. at 25572,
25575.
The American Lung Association and the Environmental
Defense Fund challenged before this Court the Administrator’s
decision not to implement a five-minute standard. We found
that EPA had failed to explain adequately how it reached its
decision not to revise the NAAQS, given that the Administrator
had found that short-term exposures to bursts of SO2 could
significantly affect the lung function of thousands of asthmatics.
Am. Lung Ass’n, 134 F.3d at 392-93 (D.C. Cir. 1998).
6
Accordingly, we remanded the decision to EPA.
In response, EPA initiated the review of the SO2 NAAQS
that eventually led to this proceeding. See 71 Fed. Reg. 28023
(May 15, 2006). Based on that review, EPA proposed a rule to
revise the primary SO2 standard. EPA proposed, inter alia, to
revoke the current 24-hour and annual standards and to establish
a standard to target short-term bursts of SO 2
exposure—specifically, a 99th percentile 1-hour daily maximum
standard level set somewhere between 100 ppb and 50 ppb. 74
Fed. Reg. 64810, 64845-86 (Dec. 8, 2009). EPA also proposed
to amend ambient air monitoring, reporting, and network design
requirements. The proposal focused on increasing and updating
the monitoring network to support the proposed 1-hour standard.
74 Fed. Reg. at 64846-47.
C. The Final Rule
After receiving comments on its rule proposal, EPA issued
a final rule addressing the primary SO2 standard. 75 Fed. Reg.
35520 (June 22, 2010). Petitioners challenge two parts of the
final rulemaking, which we describe here—the level at which
EPA set the standard and a portion of its statements regarding
the implementation plan for the standard.
1. Level
EPA mandated that States must meet a new 1-hour SO2
standard using a 99th percentile form, set at 75 ppb maximum
SO2 concentration. 75 Fed. Reg. at 35548. The goal of the new
standard is to prevent asthmatics from being exposed to short-
term, five- to ten-minute bursts of SO2, which EPA found could
cause lung function decrements in asthmatics. Id.
7
EPA explained that it conducted substantial amounts of new
research to determine the appropriate level for the 1-hour SO2
NAAQS. 75 Fed. Reg. at 35524. In 2008, EPA staff prepared
an Integrated Science Assessment (ISA), which summarized the
latest scientific knowledge regarding effects of exposure to SO2.
In 2009, EPA staff prepared a Risk and Exposure Assessment
(REA) to quantify the public health effects of exposure to SO2
in the ambient air. See 75 Fed. Reg. at 35523-24 (discussing
development of ISA and REA). The ISA and REA focused on
two types of studies—controlled human exposure clinical
studies and epidemiologic studies. See 75 Fed. Reg. at 35525.
The controlled human exposure studies examined the effects of
varying levels of SO2 on unmedicated asthmatics performing
exercises. The studies did not test subjects with severe asthma
because of ethical concerns. 75 Fed. Reg. at 35533. The
epidemiologic studies considered whether a statistical
association exists between levels of SO2 in the ambient air and
the occurrence of events such as hospital admissions and
emergency room visits for respiratory ailments. The ISA and
REA also reviewed animal studies. 75 Fed. Reg. at 35525.
The epidemiologic studies showed that in geographic areas
meeting the previous 24-hour and annual concentration limits,
there were positive associations between ambient air
concentrations of SO2 and respiratory symptoms in children,
emergency department visits, and hospitalizations for respiratory
conditions. Id.
Clinical studies showed that mild and moderate asthmatics
exposed to SO2 concentrations as low as 200 to 300 ppb for five
to ten minutes experienced moderate or greater decrements in
lung function. See 75 Fed. Reg. at 35525. As SO2 exposure
increased, both the severity of the decrements and the number of
asthmatics affected increased. Id. At 400 ppb and greater, the
effects often were statistically significant at the group mean
8
level and were accompanied by respiratory symptoms. Id. In
the REA, EPA determined that a 1-hour NAAQS set at 50 to 100
ppb could limit exposures of exercising asthmatic children to
five-minute peak SO2 levels greater than or equal to 400 ppb. A
1-hour 150 ppb standard could limit their exposure to five-
minute 400 ppb concentrations, but would provide “appreciably
less” protection to five-minute exposures of 200 ppb
concentrations. 75 Fed. Reg. at 35528-29.
The EPA Administrator determined that the studies showed
that the NAAQS should protect asthmatics from 200 ppb short-
term bursts of SO2. She concluded that a 1-hour standard level
set at 75 ppb would accomplish this goal and provide an
adequate margin of safety. 75 Fed. Reg. at 35548.
2. Implementation
In the preamble to the final rule, EPA also explained that,
based on comments it received, “we are revising our general
anticipated approach toward implementation of the new 1-hour
NAAQS.” 75 Fed. Reg. at 35550. Instead of assessing
attainment of the standard primarily by monitoring the ambient
air, as it had stated it would in the proposed rule, EPA suggested
it would use a “hybrid analytic approach” that would combine
monitoring with computer modeling to determine compliance.
75 Fed. Reg. at 35551.
Several states and environmental regulatory bodies of states
charged with implementing the SO2 standards, along with
several companies and coalitions that represent industries that
emit SO2 as a byproduct of their industrial activities, petition for
review of the EPA’s rulemaking. Petitioners allege two errors
in the EPA’s proceedings. All Petitioners contend that EPA
violated the notice-and-comment rulemaking provisions of the
Administrative Procedure Act (APA), 5 U.S.C. § 553, by
9
mandating a hybrid modeling-monitoring implementation
approach rather than a monitoring-only approach. Only the non-
state petitioners bring the second challenge, contending that the
Agency’s decision to set the new 1-hour SO2 standard at 75 ppb
was arbitrary and capricious.
II. Analysis
A. Computer Modeling Provisions
Petitioners argue that EPA failed to follow the notice-and-
comment rulemaking provisions of the APA because it did not
include in its proposed rule any suggestion that EPA was
considering changing its method of determining attainment of
the SO2 standard from an air-monitoring approach to a hybrid
approach using computer modeling in combination with air
monitoring. The APA requires agencies to publish proposed
rules in the Federal Register and afford interested parties
opportunity to comment on the proposals. 5 U.S.C. § 553; see
PPG Indus., Inc. v. Costle, 659 F.2d 1239, 1241 (D.C. Cir.
1981) (requiring EPA to comply with APA notice-and-comment
rulemaking procedures when changing a NAAQS
implementation program). Petitioners point out that statements
in the preamble to the final rule suggesting a computer-modeling
and air-monitoring approach were not part of the original
proposed rule submitted for notice and comment, and that
therefore they had no opportunity to comment thereon.
We do not have jurisdiction, however, to consider whether
EPA has violated APA rulemaking procedures because the
challenged statements do not constitute final agency action. The
CAA provides that this Court has jurisdiction to review “any
national primary or secondary ambient air quality standard . . .
or final action taken . . . by the Administrator.” 42 U.S.C.
§ 7607(b) (emphasis added); see Indep. Equip. Dealers Ass’n v.
10
EPA, 372 F.3d 420, 426 (D.C. Cir. 2004). An agency action is
“final” if it meets two conditions: the action must “mark the
consummation of the agency’s decisionmaking process,” and the
action must be “one by which rights or obligations have been
determined, or from which legal consequences will flow.”
Bennett v. Spear, 520 U.S. 154, 177-78 (1997) (internal
quotation marks and citation omitted). The challenged
statements in the preamble here do not meet these conditions.
Petitioners compare the preamble’s statements here to the
statements in the preamble to the final rule challenged in
Natural Resources Defense Council v. EPA, 571 F.3d 1245
(D.C. Cir. 2009). There, we recognized that the challenged
statements were “not conjectural” and that their “terms [were]
clear,” so therefore it was “fair to infer that the EPA intended
the statements to create binding legal consequences.” Id. at
1252 n.2. Here, the preamble’s statements about the hybrid
approach are much less clear, demonstrating that the statements
do not mark the consummation of agency action, and they do not
create obligations from which legal consequences will flow.
The preamble explains that given the comments EPA received
regarding the proposed monitoring approach, “we are revising
our general anticipated approach toward implementation of the
new 1-hour NAAQS.” 75 Fed. Reg. at 35550. It cautions that
the discussions here “explain our expected and intended future
action in implementing the new 1-hour NAAQS—in other
words, they constitute guidance, rather than final agency
action—and it is possible that our approaches may continue to
evolve as we, States, and other stakeholders proceed with actual
implementation.” Id. In discussing the specifics of its intended
hybrid approach, EPA states that, “[w]e believe that some type
of hybrid approach is more consistent with our historical
approach and longstanding guidance toward SO2 than what we
originally proposed.” Id. at 35551. EPA explains that it
“anticipates making the determination of when monitoring alone
11
is ‘appropriate’ for a specific area on a case-by-case basis,
informed by the area’s factual record, as part of the designations
process,” citing as an example of a situation in which
monitoring might be the preferred approach an area in which a
shipping port is the only significant stationary source of SO2. Id.
at 35552 n.22. EPA notes that it “intends to solicit public
comment prior to finalizing this guidance.” Id. at 35552. This
language all suggests an indefinite, anticipated plan.
We do not suggest that if the language had imposed definite
requirements upon states or regulated industries we would be
bound by the agency’s characterization. Certainly if that were
the case, we could consider rejecting the characterization and
consider Petitioners’ challenges to what might then be final
agency action for purposes of judicial review. See Barrick
Goldstrike Mines Inc. v. Browner, 215 F.3d 45, 48 (D.C. Cir.
2000) (agency labels regarding finality or lack thereof are not
determinative). But the preamble imposes no such
requirements. EPA explained that it expected to make initial
attainment designations in 2012 based on existing monitoring
capabilities, as well as “any refined modeling the State chooses
to conduct specifically for initial area designations.” 75 Fed.
Reg. at 35552. That language does not impose new legal
obligations to use modeling.
To be sure, because EPA now intends to use this hybrid
approach, it has scaled back its proposed plans to develop a
more extensive monitoring network. 75 Fed. Reg. at 35551
(“This projected change in approach would necessarily result in
a lesser emphasis on the less appropriate, more expensive, and
slower to establish monitoring tool than did the proposed rule.
Therefore, the minimum requirements for the SO2 monitoring
network in this final rule are of a smaller scale than proposed
. . . .”). Petitioners do not argue, however, that they have
suffered an injury by not being required to build a more
12
extensive monitoring network. Petitioners will be free to
challenge any final action EPA takes that imposes an obligation
Petitioners must meet. The challenged provisions here do not
meet that standard.
B. NAAQS Level
Some Petitioners also challenge the level at which EPA set
the maximum concentration for the new 1-hour SO2 standard,
arguing that 75 ppb is lower than “requisite” to protect public
health. We have jurisdiction to consider this challenge under
CAA Section 307(b)(1), which provides that this Court has
exclusive jurisdiction over petitions for review of national
ambient air quality standards promulgated by the EPA
Administrator. 42 U.S.C. § 7607(b)(1). Under the CAA, we
will set aside the Agency’s determination only if it is “arbitrary,
capricious, an abuse of discretion, or otherwise not in
accordance with law.” 42 U.S.C. § 7607(d)(9)(A); see also
APA Section 706 (5 U.S.C. § 706); Ethyl Corp. v. EPA, 51 F.3d
1053, 1064 (D.C. Cir. 1995) (arbitrary and capricious standard
under the CAA is interpreted in “essentially the same” way as
the same standard under the APA). We owe deference to the
Administrator’s determination regarding the reliability of
scientific evidence. Am. Trucking Ass’ns v. EPA, 283 F.3d 355,
374 (D.C. Cir. 2002). Although we must perform a “searching
and careful” inquiry into the facts, we do not look at the decision
as would a scientist, but “as a reviewing court exercising our
narrowly defined duty of holding agencies to certain minimal
standards of rationality.” Ethyl Corp. v. EPA, 541 F.2d 1, 36-37
(D.C. Cir. 1976) (en banc).
Under the Clean Air Act, the EPA Administrator must set
NAAQS at a level “requisite to protect the public health,”
“allowing an adequate margin of safety.” 42 U.S.C.
§ 7409(b)(1). This Court has recognized that Congress defined
13
public health broadly, requiring NAAQS to “protect not only
average healthy individuals, but also ‘sensitive citizens,’” such
as children or people afflicted with asthma, emphysema, or other
conditions causing sensitivity to air pollution. Am. Lung Ass’n,
134 F.3d at 389 (citing SEN. REP. No. 91-1196, at 10 (1970)).
The Supreme Court has held that NAAQS are set at the
“requisite” level if they are set at a level “not lower or higher
than is necessary” to protect public health. Whitman v. Am.
Trucking Ass’ns, 531 U.S. 457, 475-76 (2001); see also 75 Fed.
Reg. at 35521 (EPA must “establish standards that are neither
more nor less stringent than necessary.”).
1. Interpretation of Human Clinical Studies
Petitioners first argue that the Administrator’s decision to
adopt a 75 ppb standard was arbitrary and capricious because
EPA misinterpreted the controlled human exposure clinical
studies. Specifically, they criticize how EPA applied guidelines
published by the American Thoracic Society (ATS) regarding
what constitutes an adverse effect of air pollution.
The ATS guidelines recommend that reversible loss of lung
function in individuals in combination with respiratory
symptoms should be considered adverse. See 75 Fed. Reg. at
35531. The guidelines also recommend that an increased risk to
a population caused by a pollutant, even if the risk to a single
individual is not increased, should be considered an adverse
effect because individuals within that group would have
diminished reserve function and would be at an increased risk if
affected by another agent. 75 Fed. Reg. at 35526.
Petitioners argue that EPA concedes that the clinical studies
only have shown that five-minute exposures to SO2 produce
adverse effects at 400 ppb and above. Pet. Br. at 45-46 (citing 75
Fed. Reg. at 35526-27). Thus, Petitioners reason, EPA has not
14
shown that five-minute exposures to SO2 levels below 400 ppb
cause adverse effects in individuals. Regarding the population-
level standard, petitioners claim first that EPA failed to find a
causal relationship between five-minute exposures to SO2 at
levels below 400 ppb and the decrements in lung function and,
second, that EPA extrapolated individual data from the clinical
studies to represent the effect of SO2 on a population level. Pet.
Br. at 46-48.
EPA, however, was not bound to set the SO2 standard
according to the ATS guidelines. The guidelines merely
provided one reference point to help EPA and the public
understand what should be considered an adverse effect of SO2
on human health.
On the other hand, the EPA Administrator is bound by
statute to promulgate NAAQS that are “requisite to protect the
public health” “allowing an adequate margin of safety.” 42
U.S.C. § 7409(b)(1). It could not then exceed EPA’s authority
to choose a level below that which produced adverse effects in
the clinical studies in order to set a standard that allows an
adequate margin of safety. Further, the clinical studies did not
test severe asthmatics or very young children. EPA concluded
that it was reasonable to assume that those vulnerable
populations would suffer more serious health effects than mild
and moderate asthmatics. 75 Fed. Reg. at 35526 & n.5. We
cannot say it was unreasonable for EPA to consider these
vulnerable populations in setting the standard.
Further, in issuing the final rule, EPA considered more than
the ATS adversity standards. EPA explains it considered the
advice and recommendations it received from Clean Air
Scientific Advisory Committee, an independent scientific review
committee, and the conclusions drawn from previous NAAQS
reviews. Id. at 35526. EPA also considered epidemiologic
15
studies, which we discuss in greater detail below, to inform its
view of the population-level risk.
2. Choice of Epidemiologic Studies
Petitioners contend that the decision to adopt a 75 ppb
standard was arbitrary and capricious because EPA “cherry-
pick[ed]” studies that supported its preferred result, while
ignoring studies that would support a higher standard. Pet. Br.
at 42. To make this argument, Petitioners point to language in
the final rule which they claim affirms that the epidemiologic
studies upon which EPA relied to set the 75 ppb standard
“suffered from a ‘major methodological issue’ arising from ‘the
extent to which other air pollutants,’ particularly fine particulate
matter, ‘may confound or modify SO2-related effect estimates.’”
Pet. Br. at 44 (quoting 75 Fed. Reg. at 35531); see also Pet. Br.
at 50-51, 53. Petitioners suggest that EPA “assumed away” the
problem by relying on controlled human exposure studies to lend
“biological plausibility” to the Agency’s preferred results. Pet.
Br. at 44.
The quotations selected by Petitioners, however, only
support Petitioners’ arguments when taken out of their original
context. In the final rule, EPA explains that it has conducted
substantial amounts of new research to determine the appropriate
level for the 1-hour SO2 NAAQS. 75 Fed. Reg. at 35524. As
described above, EPA reviewed controlled human exposure
studies in which exercising asthmatics were exposed to five- to
ten-minute bursts of SO2, epidemiologic studies relying mostly
on 1-hour and 24-hour standards, and animal toxicologic studies
examining animal exposures to SO2 for short periods of time. 75
Fed. Reg. at 35525. Out of context, the language Petitioners
quote, as presented above, would suggest that when reviewing
the epidemiologic studies, EPA failed to consider how other
pollutants might affect the results of those studies. In fact, the
16
full quotation states as follows:
Although EPA has recognized that multiple factors can
contribute to the etiology of respiratory disease and that
more than one air pollutant could independently impact
respiratory health, we continue to judge, as discussed in
the ISA, that the available evidence supports the
conclusion that there is an independent effect of SO2 on
respiratory morbidity. In reaching this judgment, we
recognize that a major methodological issue affecting
SO2 epidemiologic studies concerns the evaluation of
the extent to which other air pollutants, particular [sic]
PM2.5, may confound or modify SO2-related effect
estimates. The use of multi-pollutant regression models
is a common approach for evaluating potential
confounding by co-pollutants in epidemiologic studies.
It is therefore important to note that when the ISA
evaluated U.S. and international epidemiologic studies
employing multi-pollutant models, SO2 effect estimates
generally remained positive and relatively unchanged
when co-pollutants, including PM, were included.
75 Fed. Reg. at 35531. Thus, in context, the language Petitioners
cite to support their argument that EPA failed to consider the
effects of other pollutants in the epidemiologic studies actually
demonstrates that EPA did take other pollutants into
consideration, and even when it did so, the results “remained
positive and relatively unchanged.” EPA concedes that there are
“uncertainties” associated with separating the various pollutants’
effects, but that “the limited available evidence indicates that the
effect of SO2 on respiratory health outcomes appears to be
generally robust and independent of the effects of gaseous co-
pollutants, including NO2 and O3, as well as particulate co-
pollutants, particularly PM2.5.” 75 Fed. Reg. at 35531.
17
Further, although Petitioners accuse EPA of “cherry-
picking” its friends, EPA offers a reasonable explanation for why
it relied most heavily on three particular epidemiologic studies.
EPA explains that it relied on those three studies, out of the more
than fifty peer reviewed studies available, precisely because
these three were conducted in the United States and included
multi-pollutant models to help address the “major
methodological issue” that Petitioners contend EPA ignored. See
75 Fed. Reg. at 35547 (recognizing that “there is special
sensitivity in this review in disentangling PM-related effects . . .
from SO2-related effects in interpreting the epidemiologic
studies”). EPA concluded that the epidemiologic evidence
supported a 75 ppb standard, 75 Fed. Reg. at 35548, and the
controlled human exposure studies supported the plausibility of
the associations reported in the epidemiologic studies, 75 Fed.
Reg. at 35544.
Based on its review of all of those studies, EPA found a
“causal relationship between respiratory morbidity and short-
term (5-minutes to 24-hours) exposure to SO2.” 75 Fed. Reg. at
35525. A “causal relationship” finding is the strongest finding
the ISA can make. Id. EPA concluded that the collected
evidence showed that five- to ten-minute exposures to SO2
concentrations at least as low as 200 ppb can result in adverse
health effects in five to thirty percent of the exercising
asthmatics tested in the controlled human exposure studies, 75
Fed. Reg. at 35526, and that a 75 ppb 1-hour limit would
“substantially limit asthmatics’ exposure” to such concentrations,
allowing a reasonable margin for safety, 75 Fed. Reg. at 35548.
Based on the record discussed above, we cannot conclude
that the choice EPA made to give especial weight to the three
studies conducted in the United States that accounted for the
effects of SO2 concentrations using multi-pollutant regression
models was arbitrary or capricious.
18
3. Statistical Significance
Petitioners also argue that EPA acted arbitrarily in setting
the 1-hour standard at 75 ppb because EPA did not rely on
studies that showed a statistically significant association between
exercising asthmatics’ lung function decrements and short-term
exposures to air concentrations of 200 ppb SO2. Pet. Br. at 43,
45-47, 48. We disagree. EPA found that five- to ten-minute
exposures to SO2 caused statistically significant lung function
decrements when asthmatics were exposed to 400 ppb SO2 or
greater. 75 Fed. Reg. at 35525. Although EPA recognized that
the results did not remain statistically significant when the
concentrations sank to 200 to 300 ppb, EPA reasoned that the
clinical study results “could reasonably indicate an SO2-induced
shift in these lung function measurements for [exercising
asthmatics].” 75 Fed. Reg. at 35526. Further, the studies did not
include severe asthmatics. EPA concluded that it was
“reasonable to presume” that people with more severe asthma
would suffer more serious health consequences from short-term
exposures to 200 ppb SO2. 75 Fed. Reg. at 35526.
We have held before that EPA has discretion to set a
NAAQS at a concentration level below a level that has been
demonstrated to have a statistically significant association with
negative health effects. See Am. Trucking Ass’ns, 283 F.3d at
371. Just so now. We cannot say that the studies necessitated a
75 ppb standard, but we also cannot say that such a standard is
unreasonable or unsupported by the record before us. See id. at
372.
4. Current Air Quality Standards
Finally, Petitioners argue that the new SO2 standard is
arbitrary and capricious because EPA ignored its own finding
that the new standard would create few new health benefits
19
compared to current air quality standards and other CAA
provisions that would prevent air quality from deteriorating to
the level of the existing NAAQS. Pet. Br. at 55 (citing 75 Fed.
Reg. at 35533-34). Petitioners explain that the CAA only gives
EPA authority to revise NAAQS “as appropriate” and reason that
it is inappropriate for EPA to revise the standards when current
air quality does not warrant a revision to protect public health.
Pet. Br. at 57-58 (citing 42 U.S.C. § 7409(d)(1)).
Nothing in the CAA requires EPA to give the current air
quality such a controlling role in setting NAAQS. And as
Petitioners themselves note, the CAA gives EPA significant
discretion to decide whether to revise NAAQS. Further, in the
final rule, EPA cites evidence that current levels of SO2 in the
ambient air, even when the air quality meets the current SO2
NAAQS, still cause respiratory effects in some areas. 75 Fed.
Reg. at 35530-31. In short, EPA had discretion to revise the
NAAQS and Petitioners’ argument is unavailing.
III. Conclusion
For the foregoing reasons, we dismiss the petitions in part
and deny in part.