New York v. Nuclear Regulatory Commission

United States Court of Appeals
FOR THE DISTRICT OF COLUMBIA CIRCUIT

Argued March 16, 2012 Decided June 8, 2012

No. 11-1045

STATE OF NEW YORK, ET AL.,
PETITIONERS

v.

NUCLEAR REGULATORY COMMISSION AND UNITED STATES OF
AMERICA ,
RESPONDENTS

STATE OF NEW JERSEY , ET AL.,
INTERVENORS

Consolidated with 11-1051, 11-1056, 11-1057

On Petitions for Review of Orders
of the Nuclear Regulatory Commission

Monica Wagner, Deputy Bureau Chief, Office of the
Attorney General for the State of New York, argued the cause
for petitioners States and Prairie Island Indian Community
Petitioners. With her on the briefs were Eric T. Schneiderman,
Attorney General, Office of the Attorney General for the State
of New York, John J. Sipos and Janice A. Dean, Assistant
Attorneys General, Barbara D. Underwood, Solicitor General,
Brian A. Sutherland, Assistant Solicitor General of Counsel,
 2

Jeffrey S. Chiesa, Attorney General, Office of the Attorney
General for the State of New Jersey, Kevin P. Auerbacher,
Assistant Attorney General, Ruth E. Musetto, Deputy Attorney
General, William H. Sorrell, Attorney General, Office of the
Attorney General for the State of Vermont, Thea Schwartz,
Assistant Attorney General, George Jepsen, Attorney General,
Office of the Attorney General for the State of Connecticut,
Robert Snook, Assistant Attorney General, and Joseph F.
Halloran.

Geoffrey H. Fettus argued the cause for petitioners the
Environmental Groups. With him on the briefs were Andres J.
Restrepo and Diane Curran.

Robert M. Rader, Senior Attorney, U.S. Nuclear Regulatory
Commission, argued the cause for respondents. With him on the
brief were John E. Arbab, Attorney, U.S. Department of Justice,
Stephen G. Burns, General Counsel, U.S. Nuclear Regulatory
Commission, and John F. Cordes Jr., Solicitor.

David A. Repka argued the cause for intervenors Nuclear
Energy Institute, et al., in support of respondents. With him on
the brief were Brad Fagg and Jerry Bonanno. Anne W.
Cottingham entered an appearance.

Before: SENTELLE , Chief Judge, TATEL and GRIFFITH ,
Circuit Judges.

Opinion for the Court filed by Chief Judge SENTELLE .

SENTELLE , Chief Judge: Four states, an Indian community,
and a number of environmental groups petition this Court for
review of a Nuclear Regulatory Commission (“NRC” or
“Commission”) rulemaking regarding temporary storage and
permanent disposal of nuclear waste. We hold that the
 3

rulemaking at issue here constitutes a major federal action
necessitating either an environmental impact statement or a
finding of no significant environmental impact. We further hold
that the Commission’s evaluation of the risks of spent nuclear
fuel is deficient in two ways: First, in concluding that permanent
storage will be available “when necessary,” the Commission did
not calculate the environmental effects of failing to secure
permanent storage—a possibility that cannot be ignored.
Second, in determining that spent fuel can safely be stored on
site at nuclear plants for sixty years after the expiration of a
plant’s license, the Commission failed to properly examine
future dangers and key consequences. For these reasons, we
grant the petitions for review, vacate the Commission’s orders,
and remand for further proceedings.

I. Background

This is another in the growing line of cases involving the
federal government’s failure to establish a permanent repository
for civilian nuclear waste. See, e.g., In re Aiken County, 645
F.3d 428, 430–31 (D.C. Cir. 2011) (recounting prior cases). We
address the Commission’s recent rulemaking regarding the
prospects for permanent disposal of nuclear waste and the
environmental effects of temporarily storing such material on
site at nuclear plants until a permanent disposal facility is
available.

After four to six years of use in a reactor, nuclear fuel rods
can no longer efficiently produce energy and are considered
“spent nuclear fuel” (“SNF”). Blue Ribbon Commission on
America’s Nuclear Future, Report to the Secretary of Energy
10–11 (2012). Fuel rods are thermally hot when removed from
reactors and emit great amounts of radiation—enough to be fatal
in minutes to someone in the immediate vicinity. Id. Therefore,
the rods are transferred to racks within deep, water-filled pools
 4

for cooling and to protect workers from radiation. After the fuel
has cooled, it may be transferred to dry storage, which consists
of large concrete and steel “casks.” Most SNF, however, will
remain in spent-fuel pools until a permanent disposal solution is
available. Id. at 11.

Even though it is no longer useful for nuclear power, SNF
poses a dangerous, long-term health and environmental risk. It
will remain dangerous “for time spans seemingly beyond human
comprehension.” Nuclear Energy Inst., Inc. v. Envtl. Prot.
Agency, 373 F.3d 1251, 1258 (D.C. Cir. 2004) (per curiam).
Determining how to dispose of the growing volume of SNF,
which may reach 150,000 metric tons by the year 2050, is a
serious problem. See Blue Ribbon Commission, supra, at 14.
Yet despite years of “blue ribbon” commissions, congressional
hearings, agency reports, and site investigations, the United
States has not yet developed a permanent solution. That failure,
declared the most recent “blue ribbon” panel, is the “central flaw
of the U.S. nuclear waste management program to date.” Id. at
27. Experts agree that the ultimate solution will be a “geologic
repository,” in which SNF is stored deep within the earth,
protected by a combination of natural and engineered barriers.
Id. at ix, 29. Twenty years of work on establishing such a
repository at Yucca Mountain was recently abandoned when the
Department of Energy decided to withdraw its license
application for the facility. Id. at 3. At this time, there is not
even a prospective site for a repository, let alone progress
toward the actual construction of one.

Due to the government’s failure to establish a final resting
place for spent fuel, SNF is currently stored on site at nuclear
plants. This type of storage, optimistically labeled “temporary
storage,” has been used for decades longer than originally
anticipated. The delay has required plants to expand storage
pools and to pack SNF more densely within them. The lack of
 5

progress on a permanent repository has caused considerable
uncertainty regarding the environmental effects of temporary
SNF storage and the reasonableness of continuing to license and
relicense nuclear reactors.

In this case, petitioners challenge a 2010 update to the
NRC’s Waste Confidence Decision (“WCD”). The original
WCD came as the result of a 1979 decision by this court
remanding the Commission’s decision to allow the expansion of
spent-fuel pools at two nuclear plants. Minnesota v. NRC, 602
F.2d 412 (D.C. Cir. 1979). In Minnesota, we directed the
Commission to consider “whether there is reasonable assurance
that an off-site storage solution [for spent fuel] will be available
by . . . the expiration of the plants’ operating licenses, and if not,
whether there is reasonable assurance that the fuel can be stored
safely at the sites beyond those dates.” Id. at 418. The WCD is
the Commission’s determination of those risks and assurances.

The original WCD was published in 1984 and included five
“Waste Confidence Findings.” Briefly, those findings declared
that: 1) safe disposal in a mined geologic repository is
technically feasible, 2) such a repository will be available by
2007–2009, 3) waste will be managed safely until the repository
is available, 4) SNF can be stored safely at nuclear plants for at
least thirty years beyond the licensed life of each plant, and 5)
safe, independent storage will be made available if needed.
Waste Confidence Decision, 49 Fed. Reg. 34,658, 34,659–60
(Aug. 31, 1984). The Commission updated the WCD in 1990 to
reflect new understandings about waste disposal and to predict
the availability of a repository by 2025. See Waste Confidence
Decision Review, 55 Fed. Reg. 38,474, 38,505 (Sept. 18, 1990).
The Commission reviewed the WCD again in 1999 without
altering it. See Waste Confidence Decision Review: Status, 64
Fed. Reg. 68,005, 68,006–07 (Dec. 6, 1999).
 6

In 2008, the Commission proposed revisions to the Waste
Confidence Findings, and, after considering public comments,
made revisions in 2010. Waste Confidence Decision Update, 75
Fed. Reg. 81,037 (Dec. 23, 2010). That decision, under review
in this case, reaffirmed three of the Waste Confidence Findings
and updated two. First, the Commission revised Finding 2,
which, as of 1990, expected that a permanent geologic
repository would be available in the first quarter of the twenty-
first century. As amended, Finding 2 now states that a suitable
repository will be available “when necessary,” rather than by a
date certain. Id. at 81,038. In reaching that conclusion, the
Commission examined the political and technical obstacles to
permanent storage and determined that a permanent repository
will be ready by the time the safety of temporary on-site storage
can no longer be assured. Id.

Finding 4 originally held that SNF could be safely stored at
nuclear reactor sites without significant environmental effects
for at least thirty years beyond each plant’s licensed life,
including the license-renewal period. Id. at 81,039. In revising
that finding, the Commission examined the potential
environmental effects from temporary storage, such as leakages
from the spent-fuel pools and fires caused by the SNF becoming
exposed to the air. Concluding that previous leaks had only a
negligible near-term health effect and that recent regulatory
enhancements will further reduce the risk of leaks, the
Commission determined that leaks do not pose the threat of a
significant environmental impact. Id. at 81,069–71. The
Commission also found that pool fires are sufficiently unlikely
as to pose no significant environmental threat. Id. at 81,070–71.
As amended, Finding 4 now holds that SNF can be safely stored
at plants for at least sixty years beyond the licensed life of a
plant, instead of thirty. Id. at 81,074. In addition, the
Commission noted in its final rule that it will be developing a
plan for longer-term storage and will conduct a full assessment
 7

of the environmental impact of storage beyond the sixty-year
post-license period. Id. at 81,040. Based on the revised WCD,
the Commission released a new Temporary Storage Rule
(“TSR”) enacting its conclusions and updating its regulations
accordingly. See Consideration of Environmental Impacts of
Temporary Storage of Spent Fuel after Cessation of Reactor
Operation, 75 Fed. Reg. 81,032 (Dec. 23, 2010); 10 C.F.R.
§ 51.23(a). Petitioners challenge the amended 10 C.F.R.
§ 51.23(a) based on both Finding 2 and Finding 4.

II. The Commission’s Obligations Under NEPA

The National Environmental Policy Act of 1969 (“NEPA”),
42 U.S.C. § 4321 et seq., requires federal agencies such as the
Commission to examine and report on the environmental
consequences of their actions. NEPA is an “essentially
procedural” statute intended to ensure “fully informed and well-
considered” decisionmaking, but not necessarily the best
decision. Vermont Yankee Nuclear Power Corp. v. NRDC, 435
U.S. 519, 558 (1978). Under NEPA, each federal agency must
prepare an Environmental Impact Statement (“EIS”) before
taking a “major Federal action[] significantly affecting the
quality of the human environment.” 42 U.S.C. § 4332(2)(C).
An agency can avoid preparing an EIS, however, if it conducts
an Environmental Assessment (“EA”) and makes a Finding of
No Significant Impact (“FONSI”). See Sierra Club v. Dep’t of
Transp., 753 F.2d 120, 127 (D.C. Cir. 1985); see also Theodore
Roosevelt Conservation P'ship v. Salazar, 616 F.3d 497, 503–04
(D.C. Cir. 2010) (explaining NEPA procedures in detail). The
issuance or reissuance of a reactor license is a major federal
action affecting the quality of the human environment. See New
York v. Nuclear Regulatory Comm’n, 589 F.3d 551, 553 (2d Cir.
2009).
 8

The parties here dispute whether the WCD itself constitutes
a major federal action. To petitioners, the WCD is a major
federal action because it is a predicate to every decision to
license or relicense a nuclear plant, and the findings made in the
WCD are not challengeable at the time a plant seeks licensure.
The Commission contends that because the WCD does not
authorize the licensing of any nuclear reactor or storage facility,
and because a site-specific EIS will be conducted for each
facility at the time it seeks licensure, the WCD is not a major
federal action. To the Commission, the WCD is simply an
answer to this court’s mandate in Minnesota to ensure that plants
are only licensed while the NRC has reasonable assurance that
permanent disposal of the resulting waste will be available. The
Commission also contends that the WCD constitutes an EA
supporting the revision of 10 C.F.R. § 51.23(a), and because the
EA found no significant environmental impact, an EIS is not
required.

We agree with petitioners that the WCD rulemaking is a
major federal action requiring either a FONSI or an EIS. The
Commission’s contrary argument treating the WCD as separate
from the individual licensing decisions it enables fails under
controlling precedent.

We have long held that NEPA requires that “environmental
issues be considered at every important stage in the decision
making process concerning a particular action.” Calvert Cliffs'
Coordinating Comm., Inc. v. Atomic Energy Comm'n, 449 F.2d
1109, 1118 (D.C. Cir. 1971). The WCD makes generic findings
that have a preclusive effect in all future licensing decisions—it
is a pre-determined “stage” of each licensing decision. NEPA
established the Council on Environmental Quality (“CEQ”)
“with authority to issue regulations interpreting it.” Dep’t of
Transp. v. Public Citizen, 541 U.S. 752, 757 (2004). The CEQ
has defined major federal actions to include actions with
 9

“[i]ndirect effects, which are caused by the action and are later
in time or farther removed in distance, but are still reasonably
foreseeable.” 40 C.F.R. §§ 1508.8, 1508.18; Public Citizen, 541
U.S. at 763; see also Andrus v. Sierra Club, 442 U.S. 347, 358
(1979) (holding that the CEQ’s NEPA interpretations are
entitled to substantial deference); accord CTIA-Wireless Ass’n
v. FCC, 466 F.3d 105, 115 (D.C. Cir. 2006). It is not only
reasonably forseeable but eminently clear that the WCD will be
used to enable licensing decisions based on its findings. The
Commission and the intervenors contend that the site-specific
factors that differ from plant to plant can be challenged at the
time of a specific plant’s licensing, but the WCD nonetheless
renders uncontestable general conclusions about the
environmental effects of plant licensure that will apply in every
licensing decision. See 10 C.F.R. § 51.23(b).

Petitioners’ argument continues by suggesting that the
WCD lacks an EIS and must be reversed on that basis. Not
necessarily. No EIS is required if the agency conducts an EA
and issues a FONSI sufficiently explaining why the proposed
action will not have a significant environmental impact. Public
Citizen, 541 U.S. at 757–58. Though we give considerable
deference to an agency’s decision regarding whether to prepare
an EIS, the agency must 1) “accurately identif[y] the relevant
environmental concern,” 2) take a “hard look at the problem in
preparing its EA,” 3) make a “convincing case for its finding of
no significant impact,” and 4) show that even if a significant
impact will occur, “changes or safeguards in the project
sufficiently reduce the impact to a minimum.” Taxpayers of
Michigan Against Casinos v. Norton, 433 F.3d 852, 861 (D.C.
Cir. 2006) (internal quotation omitted). An agency’s decision
not to prepare an EIS must be set aside if it is “arbitrary,
capricious, an abuse of discretion, or otherwise not in
accordance with law.” Public Citizen, 541 U.S. at 763 (quoting
5 U.S.C. § 706(2)(A)).
 10

III. Availability of a Permanent Repository

With these NEPA obligations in mind, we turn to the
Commission’s conclusion that a permanent repository for SNF
will be available “when necessary.” In so concluding, the
Commission examined the historical difficulty—now measured
in decades rather than years—in establishing a permanent
facility. See, e.g., Waste Confidence Decision Update, 75 Fed.
Reg. at 81,049. Though a number of commenters suggested that
the social and political barriers to building a geologic repository
are too great to conclude that a facility could be built in any
reasonable timeframe, the Commission believes that the lessons
learned from the Yucca Mountain program and the Blue Ribbon
Commission on America’s Nuclear Future will ensure that,
through “open and transparent” decisionmaking, a consensus
would be reached. Id. Further, the Commission noted that the
Nuclear Waste Policy Act mandates a repository program,
demonstrating the continued commitment and obligation of the
federal government to pursue one. The scientific and
experiential knowledge of the past decades, the Commission
explained, would enable the government to create a suitable
repository by the time one is needed. Id.

A.

Petitioners argue that the Commission’s conclusion
regarding permanent storage violates NEPA in two ways: First,
it fails to fully account for the significant societal and political
barriers that may delay or prevent the opening of a repository.
Second, the Commission’s conclusion that a permanent
repository will be available “when necessary” fails to define the
term “necessary” in any meaningful way and does not address
the effects of a failure to establish a repository in time.
Petitioners further contest the Commission’s claim that the
WCD constitutes an EA for permanent disposal, let alone the
 11

EIS they contend is required here.

The Commission responds by contending that it “candidly
acknowledged” the societal and political challenges, and crafted
the WCD to account for those risks. Overcoming political
obstacles is not the responsibility of the Commission, it
contends, and the NRC’s conclusion that institutional obstacles
will not prevent a repository from being built is entitled to
substantial deference. The Commission contends that the
selection of a precise date for Finding 2 is not required by NEPA
or any other laws governing the NRC, and the Commission used
the “when necessary” formulation as far back as 1977. See
NRDC v. Nuclear Regulatory Comm’n, 582 F.2d 166, 170, 175
(2d Cir. 1978).

As for examining the environmental effects of failing to
establish a repository, the Commission contends that the WCD
is an EA supporting the revision of 10 C.F.R. § 51.23(a). No
EIS is necessary regarding permanent disposal because, the
Commission argues, the WCD is not a major federal action, and
conducting an EIS for this issue would be the sort of “abstract
exercise” the Supreme Court declined to require in Baltimore
Gas and Electric Company v. NRDC, 462 U.S. 87, 100 (1983).
Further, the Commission’s existing “Table S-3” already
considers the environmental effects of the nuclear fuel cycle
generally and found no significant impacts. Therefore, the
Commission believes, no EIS is required.

B.

The Commission’s “when necessary” finding is already
imperiled by our conclusion that the WCD is a major federal
action. We hold that the WCD must be vacated as to its revision
to Finding 2 because the WCD fails to properly analyze the
environmental effects of its permanent disposal conclusion.
 12

While we share petitioners’ considerable skepticism as to
whether a permanent facility can be built given the societal and
political barriers to selecting a site, we need not resolve whether
the Commission adequately considered those barriers.
Likewise, we need not decide whether, as the Commission
contends, an agency’s interpretation of the political landscape
surrounding its field of expertise merits deference. Instead, we
hold the WCD is defective on far simpler grounds: As we have
determined, the WCD is a major federal action because it is used
to allow the licensing of nuclear plants. See supra Part II.
Therefore, the WCD requires an EIS or, alternatively, an EA
that concludes with a finding of no significant impact. The
Commission did not supply a suitable FONSI here because it did
not examine the environmental effects of failing to establish a
repository.

Even taking the Commission’s word that the WCD
constitutes an EA for the permanent storage conclusion, see
Waste Confidence Decision Update, 75 Fed. Reg. at 81,042, the
EA is insufficient because a finding that “reasonable assurance
exists that sufficient mined geologic repository capacity will be
available when necessary,” id. at 81,041, does not describe a
probability of failure so low as to dismiss the potential
consequences of such a failure. Under NEPA, an agency must
look at both the probabilities of potentially harmful events and
the consequences if those events come to pass. See, e.g.,
Carolina Envtl. Study Grp. v. U.S., 510 F.2d 796, 799 (D.C. Cir.
1975). An agency may find no significant impact if the
probability is so low as to be “remote and speculative,” or if the
combination of probability and harm is sufficiently minimal.
See, e.g., City of New York v. Dep’t of Transp., 715 F.2d 732,
738 (2d Cir. 1983) (“The concept of overall risk incorporates the
significance of possible adverse consequences discounted by the
improbability of their occurrence.”). Here, a “reasonable
assurance” that permanent storage will be available is a far cry
 13

from finding the likelihood of nonavailability to be “remote and
speculative.” The Commission failed to examine the
environmental consequences of failing to establish a repository
when one is needed.

The Commission argues that its “Table S-3” already
accounts for the environmental effects of the nuclear fuel cycle
and finds no significant impact. Not so. Table S-3, like the
Commission itself, presumes the existence of a geologic
repository. Therefore, it cannot explain the environmental
effects of a failure to secure a permanent facility. The
Commission also complains that conducting a full analysis
regarding permanent storage would be an “abstract exercise.”
Perhaps the Commission thinks so because it perceives the
required analysis to be of the effects of the permanent repository
itself. But we are focused on the effects of a failure to secure
permanent storage. The Commission apparently has no long-
term plan other than hoping for a geologic repository. If the
government continues to fail in its quest to establish one, then
SNF will seemingly be stored on site at nuclear plants on a
permanent basis. The Commission can and must assess the
potential environmental effects of such a failure.

IV. Temporary On-Site Storage of SNF

In concluding that SNF can safely be stored in on-site
storage pools for a period of sixty years after the end of a plant’s
life, instead of thirty, the Commission conducted what it
purports to be an EA, which found that extending the time for
storage would have no significant environmental impact. See
Waste Confidence Decision Update, 75 Fed. Reg. at 81,074.
This analysis was conducted in generic fashion by looking to
environmental risks across the board at nuclear plants, rather
than by conducting a site-by-site analysis of each specific
nuclear plant. Two key risks the Commission examined in its
 14

EA were the risk of environmental harm due to pool leakage and
the risk of a fire resulting from the fuel rods becoming exposed
to air. See id. at 81,069–71. We conclude that the
Commission’s EA and resulting FONSI are not supported by
substantial evidence on the record because the Commission
failed to properly examine the risk of leaks in a forward-looking
fashion and failed to examine the potential consequences of pool
fires.

A.

Petitioners challenge the finding of no significant impact on
two bases: First, petitioners argue that a generic analysis is
simply inappropriate and that the Commission was required to
look at each plant individually. A site-by-site analysis is
necessary, petitioners argue, because the risks of leaks and fires
are affected by site-specific factors such as pool configuration,
leak detection systems, the nature of SNF stored in the pool, and
the location of the pool within the plant. Overall, petitioners
argue that NEPA requires the Commission to fully analyze the
environmental effects of on-site storage, and a generic analysis
cannot fulfill that statutory mandate.

Second, petitioners argue that even if generic analysis is
appropriate, the Commission’s generic EA in this case was
insufficient. They maintain that the Commission did not
adequately account for leaks from on-site pools because the
Commission only looked at past leaks to see if they caused
environmental damage, rather than examining the risks of future
leaks. Also, as petitioners point out, the Commission’s own
studies have shown that previous leaks “did, or potentially
could, impact ground-water resources relative to established
EPA drinking water standards.” NRC, Liquid Radioactive
Release Lessons Learned Task Force Final Report 13 (2006).
Petitioners also argue that the Commission’s analysis of the
 15

effects of pool fires was deficient because the Commission
declined to examine the consequences of pool fires due to the
low probability of such an occurrence. In petitioners’ view, the
Commission could only avoid examining the consequences of
pool fires in a full EIS if it found the risk so low as to be
“remote and speculative”—a finding the Commission did not
make. Finally, Petitioners contend that the Commission
completely failed to look at non-health environmental factors
such as effects on the Prairie Island Indian Community’s
homeland, which is located near one of the plants governed by
the rule.

The Commission responds by stating that its examination of
past leaks properly demonstrated that the potential for
environmental harm from leakage is negligible. The
Commission argues that the effects of past leaks have been
shown to be quite minimal, and the Commission’s leakage task
force has recommended twenty-six specific measures to
minimize the risk even further. Also, the NRC exercises
oversight over the pools and will ensure that they do not become
unsafe over the sixty-year period. With regard to fires, the
Commission contends that it engaged in an “exhaustive
consideration” of the risk and found that such an event is
extremely unlikely. In the Commission’s view, a site-by-site
analysis of pool-fire risk is unnecessary because the
Commission relied on studies which accounted for all of the
variations cited by petitioners and essentially looked at the most
dangerous combinations of site-specific factors. Even looking
to a worst-case scenario, the Commission says, the risk of fires
was still extremely low.

Responding to petitioners’ argument that the Commission
failed to determine that the risk of fires was “remote and
speculative,” the Commission suggests that it did not dismiss the
risk out of hand as “remote and speculative” but rather examined
 16

it thoroughly and found it to be so low that the consequences
could not possibly overcome the low probability. Therefore, the
Commission did not need to conduct a full EIS for pool fires.
Finally, the Commission argues that petitioners did not raise the
issue of non-health impacts during the rulemaking, and thus they
cannot raise that issue on petition now.

B.

Both the Supreme Court and this court have endorsed the
Commission’s longstanding practice of considering
environmental issues through general rulemaking in appropriate
circumstances. See, e.g., Baltimore Gas, 462 U.S. at 100 (“The
generic method chosen by the agency is clearly an appropriate
method of conducting the hard look required by NEPA.”); see
also Minnesota, 602 F.2d at 416–17. Though Baltimore Gas
dealt with the nuclear fuel cycle itself, which is generally
focused on things that occur outside of individual plants, we see
no reason that a comprehensive general analysis would be
insufficient to examine on-site risks that are essentially common
to all plants. This is particularly true given the Commission’s
use of conservative bounding assumptions and the opportunity
for concerned parties to raise site-specific differences at the time
of a specific site’s licensing. Nonetheless, whether the analysis
is generic or site-by-site, it must be thorough and
comprehensive. Even though the Commission’s application of
its technical expertise demands the “most deferential” treatment
by the courts, Baltimore Gas, 462 U.S. at 103, we conclude that
the Commission has failed to conduct a thorough enough
analysis here to merit our deference.

1.

The Commission admits in the WCD Update that there have
been “several incidents of groundwater contamination
 17

originating from leaking reactor spent fuel pools and associated
structures.” 75 Fed. Reg. at 81,070. The Commission brushes
away that concern by stating that the past leaks had only a
negligible near-term health impact. Id. at 81,071. Even setting
aside the fact that near-term health effects are not the only type
of environmental impacts, the harm from past leaks—without
more—tells us very little about the potential for future leaks or
the harm such leaks might portend. The WCD Update seeks to
extend the period of time for which pools are considered safe for
storage; therefore, a proper analysis of the risks would
necessarily look forward to examine the effects of the additional
time in storage, as well as examining past leaks in a manner that
would allow the Commission to rule out the possibility that
those leaks were only harmless because of site-specific factors
or even sheer luck. The WCD Update has no analysis of those
possibilities other than to say that past leaks had “negligible”
near-term health effects. Id. A study of the impact of thirty
additional years of SNF storage must actually concern itself with
the extra years of storage.

The Commission also notes that a taskforce has made
recommendations for improvements to spent-fuel pools, which
the NRC “has addressed, or is in the process of addressing.” Id.
But those improvements are thus far untested, and we have no
way of deferring to the Commission’s conclusion that they will
ensure the absence of environmental harm. Finally, the
Commission refers to its monitoring and regulatory compliance
program as a buffer against pool degradation. Id. That
argument is even less availing because it amounts to a
conclusion that leaks will not occur because the NRC is “on
duty.” With full credit to the Commission’s considerable
enforcement and inspection efforts, merely pointing to the
compliance program is in no way sufficient to support a
scientific finding that spent-fuel pools will not cause a
significant environment impact during the extended storage
 18

period. This is particularly true when the period of time covered
by the Commission’s predictions may extend to nearly a century
for some facilities.

Despite giving our “most deferential” treatment to the
Commission’s application of its technical and scientific
expertise, we cannot reconcile a finding that past leaks have
been harmless with a conclusion that future leaks at all sites will
be harmless as well. The Commission’s task here was to
determine whether the pools could be considered safe for an
additional thirty years in the future. That past leaks have not
been harmful with respect to groundwater does not speak to
whether and how future leaks might occur, and what the effects
of those leaks might be. The Commission’s analysis of leaks,
therefore, was insufficient.

2.

Even though the Commission engaged in a more substantial
analysis of fires than it did of leaks, that analysis is plagued by
a failure to examine the consequences of pool fires in addition
to the probabilities. Petitioners, citing Limerick Ecology Action,
Inc. v. Nuclear Regulatory Commission, 869 F.2d 719, 739 (3d
Cir. 1989), argue that the Commission could only avoid
conducting an EIS if it found the risk of fires to be “remote and
speculative.” The Commission, citing Carolina Environmental
Study Group v. United States, 510 F.2d at 799, argues that it did
not need to examine the consequences of fires because it found
the risk of fires to be very low.

We disagree with both parties. As should be clear by this
point in our opinion, an agency conducting an EA generally
must examine both the probability of a given harm occurring
and the consequences of that harm if it does occur. Only if the
harm in question is so “remote and speculative” as to reduce the
 19

effective probability of its occurrence to zero may the agency
dispense with the consequences portion of the analysis. See
Limerick Ecology Action, Inc., 869 F.2d at 739. But, contra
petitioners, the finding that the probability of a given harm is
nonzero does not, by itself, mandate an EIS: after the agency
examines the consequences of the harm in proportion to the
likelihood of its occurrence, the overall expected harm could
still be insignificant and thus could support a FONSI. See
Carolina Envtl. Study Grp., 510 F.2d at 799 (“Recognition of
the minimal probability of such an event is not equatable with
nonrecognition of its consequences.”). Here, however, the
Commission did not undertake to examine the consequences of
pool fires at all. Depending on the weighing of the probability
and the consequences, an EIS may or may not be required, and
such a determination would merit considerable deference. C.f.,
City of New York, 715 F.2d at 751–52 (deferring to an agency’s
weighing of a “catastrophic” harm against an “infinitesimal
probability”). But unless the risk is “remote and speculative,”
the Commission must put the weights on both sides of the scale
before it can make a determination.

3.

As for petitioners’ remaining argument that the Commission
did not consider non-health environmental effects, we agree
with the Commission that petitioners did not properly raise those
issues in the rulemaking. Petitioners essentially present two
non-health impacts: decrease in property values and risk of harm
to the Prairie Island Indian Community’s homeland. The Tribe
did mention its small size and close proximity to the Prairie
Island Nuclear Generating Plant, but it did not assert specifically
how it might be harmed by either the rulemaking itself or the
licensing the rulemaking enables. With regard to property
values, petitioners point to a study considering the economic
impact of the Indian Point plant. But that study actually
 20

assumes a diminution in values caused by current plant
operation and simply extends it mathematically—it in no way
asserts whether or how any harm to property values might occur
nor how that harm is related to a change in the physical
environment. Petitioners’ failure to raise these objections to the
agency waives them. See Public Citizen, 541 U.S. at 764. We
note, as did the Supreme Court in Public Citizen, that primary
responsibility for compliance with NEPA lies with the
Commission, not petitioners; nonetheless, the non-health effects
alluded to here are not “so obvious that there is no need for a
commentator to point them out.” Id. Given, however, that we
are invalidating the Commission’s conclusions as a whole,
petitioners will have the opportunity to properly raise and clarify
these concerns on remand.

* * *

Overall, we cannot defer to the Commission’s conclusions
regarding temporary storage because the Commission did not
conduct a sufficient analysis of the environmental risks. In so
holding, we do not require, as petitioners would prefer, that the
Commission examine each site individually. However, a
generic analysis must be forward looking and have enough
breadth to support the Commission’s conclusions. Furthermore,
as NEPA requires, the Commission must conduct a true EA
regarding the extension of temporary storage. Such an analysis
must, unless it finds the probability of a given risk to be
effectively zero, account for the consequences of each risk. On
remand, the Commission will have the opportunity to conduct
exactly such an analysis.

V. Conclusion

We recognize that the Commission is in a difficult position
given the political problems concerning the storage of spent
 21

nuclear fuel. Nonetheless, the Commission’s obligations under
NEPA require a more thorough analysis than provided for in the
WCD Update. We note that the Commission is currently
conducting an EIS regarding the environmental impacts of SNF
storage beyond the sixty-year post-license period at issue in this
case, and some or all of the problems here may be addressed in
such a rulemaking. In any event, we grant the petitions for
review, vacate the WCD Update and TSR, and remand for
further proceedings consistent with this opinion.

So ordered.