United States Court of Appeals
For the First Circuit
No. 11-1474
UPPER BLACKSTONE WATER POLLUTION ABATEMENT DISTRICT,
Petitioner,
v.
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY,
Respondent.
No. 11-1610
CONSERVATION LAW FOUNDATION, INC.
Petitioner,
v.
UPPER BLACKSTONE WATER POLLUTION ABATEMENT DISTRICT,
Intervenor,
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY,
Respondent.
PETITIONS FOR REVIEW OF A FINAL PERMIT DECISION BY THE
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
Before
Lynch, Chief Judge,
Souter, Associate Justice,*
and Stahl, Circuit Judge.
*
The Hon. David H. Souter, Associate Justice (Ret.) of the
Supreme Court of the United States, sitting by designation.
Robert D. Cox, Jr., with whom Douglas T. Radigan,
Bowditch & Dewey, LLP, Fredric P. Andes, and Barnes & Thornburg
LLP, were on brief, for petitioner Upper Blackstone Water Pollution
Abatement District.
Christopher M. Kilian, with whom Anthony N.L. Iarrapino
was on brief, for petitioner Conservation Law Foundation.
Madeline Fleisher, with whom Ignacia S. Moreno, Assistant
Attorney General, U.S. Department of Justice, Environment and
Natural Resources Division, and Samir Bukhari, Ira W. Leighton,
Karen A. McGuire, U.S. Environmental Protection Agency, were on
brief, for respondent.
Donald L. Anglehart on brief for City of Marlborough,
amicus curiae.
David M. Moore, City Solicitor, and Jennifer H. Beaton,
Assistant City Solicitor, on brief for City of Worcester, amicus
curiae.
Karma B. Brown, Brooks M. Smith, Hunton & Williams LLP,
and Nathan Gardner-Andrews on brief for National Association of
Clean Water Agencies, amicus curiae.
August 3, 2012
-2-
LYNCH, Chief Judge. These petitions seek review of
certain effluent limitations imposed by the Environmental
Protection Agency (EPA) in a National Pollutant Discharge
Elimination System (NPDES) permit on the discharges of Upper
Blackstone Water Pollution Abatement District, a sewage treatment
plant located in central Massachusetts.
The District's discharges are into the headwaters of a
polluted river which, in due course, flows into other rivers, and
ultimately empties into Narragansett Bay. The states of
Massachusetts and Rhode Island each have strong interests in the
health of these waters and generally have supported the EPA's
decisions during the permitting process. The District, supported
by its member towns, has an interest in avoiding compliance costs
associated with the permit and has challenged the effluent
limitations as premature and unsupported by the scientific record.
We have stayed enforcement of the permit during this
appeal and while the parties were engaged in settlement
negotiations in a court-sponsored settlement program. We now lift
the stay, deny the petitions, and find no error in the EPA's final
permit decision.
I.
The Blackstone River is a major, interstate freshwater
river which runs south from Worcester, Massachusetts, crosses the
border into Rhode Island, and continues on to Pawtucket Falls.
-3-
There, it reaches sea level, becomes tidal, and changes its name to
the Seekonk River, which, in turn, flows into the Providence
River,1 and ultimately empties into Narragansett Bay. The
Blackstone River provides a significant source of freshwater to the
Bay.
At the peak of the industrial revolution, water-powered
textile mills lined the Blackstone River; dams, millponds, and
canals altered its natural course and halted its flow at points.
Toxic sediments of heavy metals and other industrial waste products
released into the River accumulated behind its many impoundments
and damaged its ecology. Today, industry has moved on; its legacy
remains in leftover dams and the toxic sediments held in place
behind them.
With the discontinuation of industrial river dumping, the
River's health has dramatically improved. Massachusetts and Rhode
Island now seek to put the River to new economic and recreational
uses including tourism, recreation, and commercial fishing. The
new limiting factor, and the subject of dispute in this case, is
not the River's industrial legacy, but sewage treatment. As
population has increased along the River, sewage processing has not
kept apace. An influx of nitrogen and phosphorus from sewage
1
We will refer to the Blackstone River as "the River" on
occasion; the Seekonk and Providence Rivers, by their full names
only; and all three together as "the three rivers."
-4-
treatment plants is causing serious problems for the River's waters
and those downstream.
The Blackstone, Seekonk, and Providence Rivers, and
Narragansett Bay, all suffer from severe cultural eutrophication,
a process fueled by unnaturally high concentrations of nitrogen and
phosphorus. When excessive levels of these chemical nutrients are
introduced into a water system, algae populations rapidly multiply
to nuisance levels. As populations "bloom" and die-off in quick
succession, dead algae accumulate and decompose -- their
nutrient-laden remains further enriching the immediate environment,
thereby perpetuating the eutrophication cycle. Increased rates of
respiration and decomposition deplete the available dissolved
oxygen in the water, threatening other plant and animal life in the
system. When oxygen saturation levels drop below what is needed
by fish and invertebrates to breathe, the waters become host to
fish kills, red tides, and shellfish poisonings, events which can
pose threats to human health as well.
Phosphorus drives cultural eutrophication in freshwater
systems and nitrogen drives the same process in marine waters. The
Blackstone River currently suffers from severe phosphorus-driven
cultural eutrophication. Algae blooms, thick, cloudy waters,
putrid smells, and sudden fish kills periodically contaminate its
waters. The numerous dams and impoundments along the River create
areas of stagnant water where nutrients collect and cultural
-5-
eutrophication flourishes. The toxicity build-up behind the dams
and serious concerns about resuspension of the contaminated
sediments rule out an easy solution to this problem.
Narragansett Bay and the Seekonk and Providence Rivers,
in turn, are each affected by the Blackstone's degraded waters.
Narragansett Bay, the ultimate depository for all the nutrients
carried by the Blackstone, suffers from severe nitrogen-driven
cultural eutrophication. The Seekonk River, which forms the
uppermost part of the Bay, is the most seriously impaired by the
Blackstone's nitrogen loadings.
Conditions in the three rivers and the Bay have been
deteriorating for many years. Increased domestic waste inputs into
the rivers are worsening their nutrient-related problems. Among
the numerous events documented in the record, severely hypoxic
(waters characterized by levels of dissolved oxygen below what is
needed by aquatic organisms to breathe) to nearly anoxic (waters
completely depleted of dissolved oxygen) conditions, along with
associated fish kills, were observed in upper Narragansett Bay,
including the Providence River, in the summers of 2001 and 2002.
August 2003 witnessed one of the Bay's largest fish kills in
history, when more than one million fish died in anoxic water
conditions near East Greenwich, Rhode Island.
The Rhode Island Department of Environmental Management
(RIDEM) has set up response teams which monitor the Bay
-6-
continuously and publish public notices when bacterial or pollution
conditions pose a threat to public health and commercial fishing.2
In recent years, the state has been forced to close down some of
the Bay's beaches and commercial fishing grounds entirely, measures
which damage state tourism and recreation businesses, and which
place the state's commercial fishing and shellfishing industries in
jeopardy.
Recognizing the watershed's growing problems, and
motivated by the desire to improve its resource value,
Massachusetts and Rhode Island have begun implementing
comprehensive plans to rehabilitate the three rivers and the Bay.
These efforts build on decades of work by both government actors
and private groups to study and address nutrient-related problems
in the watershed.
Congress designated the Blackstone River Valley as a
National Heritage Corridor in 1986 for the purpose of recognizing
the historical significance of the River and restoring its
watershed. The EPA formed the Narragansett Bay Project in the
1980s and the Blackstone River Initiative in the 1990s, to study,
2
See Bay Assessment & Response Team, RIDEM,
http://www.dem.ri.gov/bart/index.htm (information regarding
monitoring and closure of Narragansett Bay beaches and fisheries
when bacterial or pollution levels threaten public health) (last
visited Aug. 2, 2012); Office of Water Res., RIDEM,
http://www.dem.ri.gov/programs/benviron/water/shellfsh/index.htm
(information regarding monitoring and closure of shellfishing
grounds) (last visited Aug. 2, 2012).
-7-
among other issues, the impacts of cultural eutrophication on the
water systems. The Governors of Massachusetts and Rhode Island
first signed a Memorandum of Understanding in 1992 to underscore
the two states' commitments to studying and restoring the
watershed. In 1998, President Clinton designated the Blackstone
River an American Heritage River. Bills "[t]o establish the John
H. Chafee Blackstone River Valley National Historic Park" are
currently before both houses of Congress. S. 1708, 112th Cong.
(2011); H.R. 3191, 112th Cong. (2011).
Federal, state, and local governments, businesses, and an
array of outside groups and coalitions have funded and conducted
numerous scientific studies on nutrient-related problems in the
three rivers and the Bay. The EPA considered many of these studies
in setting the 2008 permit limits; just a few of those included in
full in the administrative record in this case are studies
-8-
conducted by Massachusetts3 and Rhode Island,4 as well as the U.S.
Army Corps of Engineers5 and the EPA.6
Although nitrogen and phosphorus end up in the rivers and
the Bay from diverse sources, including storm run-off, agricultural
3
See Fiorentino, Div. of Watershed Mgmt., Mass. Dep't of
Envtl. Prot., Blackstone River Watershed 2003 Biological Assessment
(2006) (comprehensive report on Blackstone River water quality
incorporating historical perspectives and previous studies,
including 2003 biomonitoring survey); Tamul, Div. of Watershed
Mgmt., Mass. Dep't of Envtl. Prot., Blackstone River Watershed 2003
DWM Water Quality Monitoring Data (2005) (biomonitoring survey of
water quality including nitrogen and phosphorus inputs and related
effects); Weinstein et al., Div. of Watershed Mgmt., Mass. Dep't of
Envtl. Prot., Blackstone River Basin 1998 Water Quality Assessment
(2001) (comprehensive evaluation of water quality in Blackstone
River and related tributaries, and specific recommendations for
managing nitrogen- and phosphorus-related water quality problems).
4
See Nixon et al., Anthropogenic Nutrient Inputs in
Narragansett Bay, A Twenty-five Year Perspective: A Report to the
Narragansett Bay Commission and Rhode Island Sea Grant (2005)
(study of nitrogen and phosphorus sewage inputs into Narragansett
Bay over a twenty-year period, with measurements taken in 1975,
1976, 1983, 1991, 1992, 2003, and 2004); Governor's Narragansett
Bay Watershed Planing Comm'n, Nutrient and Bacteria Pollution Panel
Initial Report (2004) (study and management plan for addressing the
problems with cultural eutrophication in the Bay); RIDEM,
Evaluation of Nitrogen Targets and WWTF Load Reductions for the
Providence and Seekonk Rivers (2004) (reporting results of Rhode
Island's TMDL efforts and a management plan for addressing cultural
eutrophication in the Bay).
5
See Wright et al., Dry Weather Water Quality Sampling and
Modeling, Blackstone River Feasibility Study (2004) (for U.S. Army
Corps of Engineers) (study of water quality conditions in
Massachusetts segment of the Blackstone River for future use in
developing a TMDL).
6
See Wright et al., Blackstone River Initiative, Water
Quality Analysis of the Blackstone River Under Wet and Dry weather
Conditions (2001) (for EPA New England) (integrated water quality
study and report on both Massachusetts and Rhode Island segments of
the River and Narragansett Bay).
-9-
fields, and construction sites, sewage treatment facilities are the
primary source of anthropogenic nutrient inputs into the Seekonk
and Providence Rivers and the Bay. Thus, a critical component of
both states' rehabilitation plans has been to impose tighter
limits, under the Clean Water Act (CWA or the "Act"), on the
amounts of nitrogen and phosphorus that sewage treatment facilities
may discharge into the rivers and the Bay.
The CWA was enacted by Congress to address the serious
threats water pollution poses to public health, economic activity,
and the long-term viability of the Nation's water resources. 33
U.S.C. §§ 1252(a), 1313(c)(2)(A). The Act's primary goal is "to
restore and maintain the chemical, physical, and biological
integrity of the Nation's waters." Id. § 1251(a). States and the
federal government share responsibility for achieving this goal.
Id. § 1251(g); Arkansas v. Oklahoma, 503 U.S. 91, 101 (1992).
States have primary responsibility for designating the
ambient water quality of the waters within their territory. 33
U.S.C. § 1313(c)(1), (2)(A). These "water quality standards" are
expressed as "designated uses" of water bodies (such as propagation
of aquatic life, recreation, aesthetics, and use as public water
supply), and as numeric or narrative "criteria," which specify the
amounts of pollutants that may be present in these water bodies
without impairing their designated uses. Id. § 1313(c)(2)(A). In
addition to incorporating state water quality standards, the Act
-10-
also employs federal, technology-based effluent limitations on
individual discharges of pollution into navigable waters. Id.
§§ 1311, 1314(b). State water quality standards generally
supplement these effluent limitations, so that where one or more
point source dischargers, otherwise compliant with federal
conditions, are nonetheless causing a violation of state water
quality standards, they may be further regulated to alleviate the
water quality violation. Id. § 1311(b)(1)(C) ("[T]here shall be
achieved . . . any more stringent limitation, including those
necessary to meet water quality standards . . . established
pursuant to any State law or regulations . . . ."); see also id.
§§ 1311(e), 1312(a), 1313(d)(1)(A), (d)(2), (e)(3)(A).
"[A]ny person" who wishes to discharge "any pollutant"
from a "point source" into the navigable waters must obtain an
NPDES permit. Id. §§ 1311(a), 1342. NPDES permits bring both
state ambient water quality standards and technology-based effluent
limitations to bear on individual discharges of pollution, id.
§ 1342(a)(3), (b)(1)(A), and tailor these to the discharger through
procedures laid out in the Act and in EPA regulations, id. § 1342.
NPDES permits may be administered by the EPA or by an authorized
state or Indian tribe. Id. §§ 1342(b), 1377(e); 40 C.F.R.
§ 123.31. To date, the EPA has authorized forty-six states to
administer their own NPDES permit programs, including Rhode
-11-
Island.7 Massachusetts has not received authorization, and so the
EPA administers NPDES permits in that state.
The CWA also requires states to identify the waters
within their boundaries that fail to meet their designated water
quality standards and rank these in order of priority, taking into
account "the severity of the pollution and the uses to be made of
such waters." 33 U.S.C. § 1313(d)(1)(A). States must then begin
the planning process for bringing these waters into compliance with
water quality standards.8 Id. § 1313(d), (e); 40 C.F.R.
§ 122.44(d)(1).
7
See EPA, NPDES Specific State Program Status,
http://cfpub.epa.gov/npdes/statestats.cfm?view=specific (last
visited Aug. 2, 2012); see also 49 Fed. Reg. 39,063 (Oct. 3, 1984)
(approving Rhode Island's NPDES program).
8
Part of this process requires the development of Total
Maximum Daily Loads (TMDLs) for each pollutant that is responsible
for a violation of water quality standards. 33 U.S.C.
§ 1313(d)(1)(C). A TMDL is a calculation of the maximum quantity
of a pollutant that may be added to a water body from all sources
without exceeding applicable water quality standards including "a
margin of safety which takes into account any lack of knowledge
concerning the relationship between effluent limitations and water
quality." Id. TMDLs take time and resources to develop and have
proven to be difficult to get just right; thus, under EPA
regulations, permitting authorities must adopt interim measures to
bring water bodies into compliance with water quality standards.
Id. § 1313(e)(3); 40 C.F.R. § 122.44(d); see also, e.g., 43 Fed.
Reg. 60,662, 60,665 (Dec. 28, 1978) ("EPA recognizes that State
development of TMDL's and wasteload allocations for all water
quality limited segments will be a lengthy process. Water quality
standards will continue to be enforced during this process.
Development of TMDL's . . . is not a necessary prerequisite to
adoption or enforcement of water quality standards . . . .").
-12-
In some circumstances, discharge into the waters of one
state may cause a violation of water quality standards in a
downstream state. The CWA anticipates conflicts over pollution
discharges between upstream and downstream states. See Milwaukee
v. Illinois, 451 U.S. 304, 325-26 (1981). When an application is
made for a discharge which may affect the water quality of a
downstream state, the EPA is required to notify both the origin
state and the downstream state. 33 U.S.C. § 1341(a)(2). If the
downstream state then determines that the discharge will violate
its water quality standards, it may submit its objections and
request a public hearing. Id.
The Supreme Court has held that the CWA grants the EPA
authority to require in NPDES permits conditions which ensure
compliance with the water quality requirements of downstream
states. Arkansas, 503 U.S. at 105; see 33 U.S.C. § 1341(a)(2)
("[The permitting agency] shall condition such license or permit in
such manner as may be necessary to insure compliance with
applicable water quality requirements. If the imposition of
conditions cannot insure such compliance such agency shall not
issue such license or permit."). EPA regulations have so required
since 1973. See 40 C.F.R. § 122.4(d) ("No permit may be issued .
. . [w]hen the imposition of conditions cannot ensure compliance
with the applicable water quality requirements of all affected
States . . . .").
-13-
In this case, both Massachusetts and Rhode Island have
listed the Blackstone River as "impaired" under Section 1313(d) of
the CWA; Rhode Island has also listed the Seekonk and Providence
Rivers and Narragansett Bay as impaired.
Massachusetts has designated the Blackstone River for
primary and secondary contact uses, including swimming, fishing,
and boating, and as habitat for fish and other wildlife. 314 Mass.
Code Regs. 4.05(3)(b).9 Under Massachusetts' narrative water
quality standards, the Blackstone River must be "free from
pollutants in concentrations or combinations that settle to form
objectionable deposits; float as debris, scum or other matter to
form nuisances; produce objectionable odor, color, taste or
turbidity; or produce undesirable or nuisance species of aquatic
life;" "free from pollutants in concentrations or combinations or
from alterations that adversely affect the physical or chemical
nature of the [river's] bottom, interfere with the propagation of
fish or shellfish, or adversely affect populations of non-mobile or
sessile benthic organisms;" "free from pollutants in concentrations
that are toxic to humans, aquatic life or wildlife," and "free from
nutrients in concentrations that would cause or contribute to
impairment of existing or designated uses," id. at 4.05(3)(b), at
all times, even under low flow conditions, id. at 4.03(3).
9
See Massachusetts Water Quality Designations, available
at http://www.mass.gov/dep/water/laws/regulati.htm#wqual (last
visited Aug. 2, 2012).
-14-
Massachusetts has determined that the Blackstone River
fails to meet state water quality standards. In its testing and
analysis of the contaminants in the River, the Massachusetts
Department of Environmental Protection (MassDEP) has documented
multiple impairments including unknown toxicity, priority organics,
metals, ammonia, chlorine, nutrients, organic enrichment, low
dissolved oxygen, flow and other habitat alterations, pathogens,
suspended solids, turbidity, objectionable deposits, and taste,
odor, and color objections. Watershed Planning Program, Office of
Watershed Mgmt., Massachusetts Year 2006 Integrated List of Waters
81-82 (2006).
Rhode Island has designated the Blackstone, Seekonk, and
Providence Rivers and Narragansett Bay for primary and secondary
contact recreational uses and as habitat for wildlife.10 Rhode
Island's narrative water quality criteria require that all three
rivers and the Bay be free of pollutants in concentrations that
adversely affect the composition of fish and wildlife; adversely
affect the physical, chemical, or biological integrity of the
habitat; interfere with the propagation of fish and wildlife; or
adversely alter the life cycle functions, uses, processes, and
activities of fish and wildlife. With respect to nutrient
pollution, Rhode Island requires that the three rivers and the Bay
10
See Rhode Island Water Quality Regulations, available at
http://www.dem.ri.gov/pubs/regs/regs/water/h2oq10.pdf (last visited
Aug. 2, 2012).
-15-
be free of nutrients "in such concentration that would impair any
[designated uses] . . . or cause undesirable or nuisance aquatic
species associated with cultural eutrophication."
Rhode Island has determined that all three rivers and the
Bay fail to meet its water quality standards. RIDEM has documented
numerous impairments in the Rhode Island segment of the Blackstone
River including ammonia, copper, lead, pathogens, nutrients, low
dissolved oxygen, and biodiversity impacts. Rhode Island monitors
Narragansett Bay and its extensions particularly closely due to
their importance to state industries. It has documented numerous
impairments to the Seekonk and Providence Rivers, including
nutrient pollution, low dissolved oxygen, and excessive algae
growth, and similar impairments to the Bay, including nutrients,
low dissolved oxygen, and pathogens.
In order to address these impairments, Rhode Island has
issued several Rhode Island Pollutant Discharge Elimination System
permits (RIPDES permits) to the major sewage treatment facilities
along the rivers and Bay, which tighten nitrogen effluent
limitations. The two largest treatment facilities, both on the
Providence River, Narragansett Bay Commission Fields Point and
Narragansett Bay Commission Bucklin Point, are designed to
discharge 65 million gallons per day (mgd), and 31 mgd,
respectively. As part of its major nitrogen removal initiative,
RIDEM has issued both facilities nitrogen effluent limitations of
-16-
5.0 mg/L. RIDEM has also set a 5.0 mg/L nitrogen limit for East
Greenwich, a much smaller facility, with an average daily flow of
approximately 1.7 mgd, but which is located on a particularly
impaired portion of Narragansett Bay. The Woonsocket facility,
which, behind the petitioner District in this case, is the second-
largest sewage treatment plant discharging into the Blackstone
River, has been given a 3.0 mg/L nitrogen limit as part of a
consent agreement. Five other much smaller facilities have been
given nitrogen limits of 8.0 mg/L.
As part of this process, in an effort to reduce the
incoming nitrogen into the Bay, Rhode Island also requested and
recommended to the EPA that the nitrogen limits on Massachusetts
dischargers into the Blackstone River be tightened as well. While
nitrogen discharge does not cause cultural eutrophication in the
Blackstone River's fresh-waters, the discharge is swiftly carried
downstream to Rhode Island's saltwater rivers and the Bay, where it
produces severe cultural eutrophication and resulting violations of
Rhode Island's water quality standards.
II.
Against this complex backdrop, the present dispute
arises. The petitioner in this case, Upper Blackstone Water
Pollution Abatement District (the "District"), is the largest
sewage treatment plant along the Blackstone River. It is located
in Millbury, Massachusetts, very near the Blackstone River's
-17-
headwaters. It discharges approximately 34 to 43 mgd of treated
domestic and industrial sewage11 into the River.12
The District's discharge represents approximately seventy
percent of the total municipal wastewater flow into the Blackstone
River, making it the dominant discharger of both nitrogen and
phosphorus into the River's waters.
The District's plant came online in 1976, and only
recently went through its first major upgrade. This comprehensive
upgrade was completed pursuant to an administrative consent order
issued to the District by the EPA after the District had violated
its September 30, 1999, NPDES permit, as modified by an August 3,
2001, Settlement Agreement (the "2001 permit"). The upgrade
involved extensive plant renovations implemented over an eight-year
period, through which the District adapted its facilities to comply
with the 2001 permit's 0.75 mg/L limit on phosphorus, and, although
the permit did not limit nitrogen discharge, a 8.0 - 10.0 mg/L
limit on nitrogen in anticipation of future nitrogen controls.13
11
More than 200 industrial users contribute wastewater to
the District's facilities, thirty-three of which currently qualify
for the "pre-treatment" program under 40 C.F.R. § 403.3(v).
12
This is somewhat below its design flow capacity of 56
mgd. The next largest sewage treatment plant on the Blackstone
River is located in Woonsocket, Rhode Island, and has a design flow
of 16 mgd, and an actual average discharge of 7 mgd.
13
These needed upgrades to the District's aging facility
cost $180 million and resulted in rate increases for the District's
customers. However, even with these upgrades, and as was noted in
the administrative record, relative to other Massachusetts
-18-
On November 8, 2005, while the upgrade was still ongoing,
the District submitted a timely application to the EPA for renewal
of the 2001 permit.14 As part of the permit reissuance process,
the EPA evaluated a variety of factors, including the District's
expected future discharge -- accounting for the upgrade -- and the
state of the receiving waters. The EPA found that all three rivers
and the Bay exhibited severe nitrogen- and phosphorus-driven
cultural eutrophication, and that the District's discharge was the
predominant point source of both phosphorus and nitrogen in the
Blackstone River.
Applying Massachusetts and Rhode Island water quality
requirements, the EPA determined that the District's nitrogen and
phosphorus discharges "will cause, have the reasonable potential to
cause, or contribute to an excursion above" applicable state water
quality standards. 40 C.F.R. § 122.44(d)(1)(i). Based on its
comprehensive analysis of these and the other required factors, the
EPA concluded that lower limits on the District's nitrogen and
phosphorus discharge were necessary to achieve compliance with
residents, the District's ratepayers pay significantly less than
the average sewage rate. For 2011 figures, see Mass. Water Res.
Auth. Advisory Bd., Annual Water and Sewer Retail Rate Survey
(2011), available at http://mwraadvisoryboard.com/wp-content/uploa
ds/2012/01/0-COMBINED-MASTER2.pdf (last visited Aug. 2, 2012).
14
Although it expired in 2006, during the permit reissuance
process that followed, the 2001 permit was administratively
continued, and remained in effect after this court granted the
District's motion to stay the 2008 permit on April 29, 2011.
-19-
state water quality standards. See id. § 122.44(d)(1)(vi).
Because both Massachusetts and Rhode Island employ narrative water
quality criteria for the relevant pollutants, the EPA translated
these into numeric limits under its procedures set out in 40 C.F.R.
§ 122.44(d)(1)(vi).
On March 23, 2007, the EPA published a draft permit that
limited total phosphorus discharge to 0.1 mg/L from April 1 through
October 31, and 1.0 mg/L from November through March, and limited
total nitrogen to 5.0 mg/L from May 1 through October 31, and
imposed a narrative criteria for nitrogen during the remaining
months.
As part of the lengthy, public permitting process that
followed, the EPA published the draft permit and its accompanying
rationale in full, accepted public comments -- extending the time
for these from thirty to sixty-four days -- and held a public
hearing on the permit. See 40 C.F.R. § 124.10. The EPA received
and considered thirty-four sets of written comments from a variety
of stakeholders, interested parties, individuals, and researchers,
including the District, the states of Massachusetts and Rhode
Island, several municipalities, and numerous other organizations.
The EPA responded to each set of comments at length.
On August 22, 2008, the EPA issued the final permit,
which contained the same limits on phosphorus and nitrogen proposed
in the draft permit. In addition, on April 15, 2009, the EPA
-20-
issued a draft permit modification proposing an effluent limitation
for aluminum discharge in order to comply with Massachusetts'
aluminum criterion. After public comment, the EPA issued a final
permit modification adopting this limitation into the final permit.
On September 15, 2008, the District filed a petition for
review of the permit with the EPA's highest adjudicative body, the
Environmental Appeals Board (the "EAB" or "Board"), see 40 C.F.R.
§ 1.25(e), appealing, among other provisions, the permit's
phosphorus, nitrogen, and aluminum15 discharge limits. Seven other
parties also filed petitions for review with the Board: MassDEP,
the Massachusetts towns of Holden, Millbury, and Worcester, the
Conservation Law Foundation (CLF), the Northern Rhode Island
Chapter of Trout Unlimited, and Cherry Valley Sewer District.
MassDEP raised several objections to the methodology employed by
the EPA in setting the nitrogen limit. RIDEM filed an amicus
curiae brief in support of the permit's nitrogen limit, citing the
comparable nitrogen limits it had imposed on similarly situated
Rhode Island facilities. CLF contended that both the nitrogen and
phosphorus limits were too high.
The District challenged multiple aspects of the permit,
re-raising many of the points it had made in its comments on the
draft permit. In particular, the District challenged: (1) the
15
The District filed a separate petition for review of the
aluminum limit, which the Board consolidated with the District's
original petition.
-21-
EPA's decision to tighten the nitrogen and phosphorus limits before
the District had fully completed its facility upgrades and without
more data on nutrient impairment in the Bay; (2) the EPA's refusal
to delay issuance of the permit until a new computer model of the
Blackstone River, then under development by the District, could be
completed, and (3) the EPA's conclusion that the District's
aluminum discharge had a reasonable potential to cause or
contribute to a violation of Massachusetts' water quality criterion
for aluminum.
On May 28, 2010, the Board issued a 106-page decision
upholding the permit, with the exception of a provision that made
several other municipal entities "co-permittees," which the Board
remanded to the EPA for further action. In re: Upper Blackstone
Water Pollution, Abatement Dist., Nos. 08-11 et al., 2010 EPA App.
LEXIS 17 (EAB May 28, 2010). After thorough review of the record
materials, the Board considered and addressed each of the parties'
various objections to the permit's nitrogen, phosphorus, and
aluminum limits. Id. at *188. It found the that the available
science and data concerning both the District's discharge as well
as the quality of the affected waters supported the EPA's judgment
to impose the tighter permit limits on the three chemical elements.
Id. It rejected the argument that the EPA should have delayed the
permit until the District's computer model was complete and
declined to consider some very preliminary outputs from that model
-22-
because the "development and testing of the model ha[d] not been
completed" and were "not utilized in setting [the limits] for this
Permit."16 Id. at *147. On the whole, it found the EPA's actions
reasonable and supported by the record. Id. at *188. After the
Board denied further review, the EPA provided the District with
notice of its final permit decision on April 6, 2011.
On April 29, 2011, the District filed a petition for
review with this court along with an emergency motion for a stay of
the new permit during the pendency of the appeal. This court
granted the stay that same day as to each permit condition cited by
the District in its motion. On May 27, 2011, CLF filed a petition
for review of the new permit, and on June 22, 2011, this court
consolidated CLF's petition with the District's petition for
purposes of briefing and oral argument.
The court received extensive briefing from the District,
the EPA, CLF, and amici curiae17 on the issues in this case. The
District and amicus curiae, the City of Worcester, filed additional
briefing on the HSPF water quality model the District was still in
the process of completing. The District had raised the issue of
16
In 2004, the District began the lengthy process of
developing a Hydrological Simulation Program--Fortran (HSPF)
computer model of the Blackstone River watershed. The model
remained incomplete through the permitting and EAB review process.
17
The court acknowledges the assistance provided by the
amici curiae in this case: City of Marlborough, City of Worcester,
and National Association of Clean Water Agencies.
-23-
the then-unfinished model multiple times during the permitting
process and on review before the EAB, arguing that, once completed,
the model's results might justify a material change in the permit's
conditions. See id. at *147. However, the District could not
estimate when the model would be finished, and the EPA declined to
delay the permit for an indefinite period until the District could
complete its model. Instead, it instructed the District to file a
permit modification request when the model was complete, and
therein submit the model's results for consideration. See 40
C.F.R. §§ 122.62(a)(2), 124.5. The District represented that it
would file such a request.18
III.
In its petition, the District challenges the 2008
permit's effluent limitations for nitrogen, phosphorus, and
aluminum. It argues that key parts of the scientific record before
the EPA were inadequate and unreliable, and that the agency
irrationally based the permit's limitations on this flawed record.
18
By the time of briefing in this case, the preliminary
results of the District's model were available. After reviewing
the parties' arguments as well as the permit modification
mechanism, this court issued an order on January 24, 2012,
directing the parties to participate in a court-sponsored Civil
Appeals Management Program (CAMP). CAMP provided the parties with
an opportunity to resolve the issues in this case more quickly and
easily than proceeding with the appeal. Despite their good-faith
efforts to do so, the parties were unable to resolve their
differences and so informed the court on June 12, 2012. The court
received additional filings from the parties on June 20 and 25,
2012.
-24-
It also argues that the EPA acted irrationally in refusing to delay
the permit until the District could complete both its facility
upgrade, then ongoing, and a new water quality model. CLF supports
the science in the record, but takes issue with the EPA's
interpretation of one report, arguing that a proper analysis of the
report requires a more stringent nitrogen limitation.
The formulation of the 2008 permit's effluent limitations
for the three chemical elements at issue required substantial
scientific and technical expertise. Our review of the EPA's
decision is deferential. See 33 U.S.C. § 1369(b)(1)(F); City of
Pittsfield, Mass. v. EPA, 614 F.3d 7, 10 (1st Cir. 2010). Under
the Administrative Procedure Act, we ask whether the EPA's actions
were "arbitrary, capricious, an abuse of discretion, or otherwise
not in accordance with law." 5 U.S.C. § 706(2)(A).
We will not set aside those actions unless the agency
"has relied on factors which Congress has not intended it to
consider, entirely failed to consider an important aspect of the
problem, offered an explanation for its decision that runs counter
to the evidence before the agency, or is so implausible that it
could not be ascribed to a difference in view or the product of
agency expertise." Motor Vehicle Mfrs. Ass'n v. State Farm Mut.
Auto. Ins. Co., 463 U.S. 29, 43 (1983). We will "uphold a decision
of less than ideal clarity" where it finds support in the record
and has a rational basis. FCC v. Fox Television Stations, Inc.,
-25-
556 U.S. 502, 513-14 (2009) (quoting Bowman Transp., Inc. v.
Ark.–Best Freight Sys., Inc., 419 U.S. 281, 286 (1974)) (internal
quotation mark omitted); Adams v. EPA, 38 F.3d 43, 49 (1st Cir.
1994).
This deference goes to the entire agency action, which
here includes both the EPA's permitting decision and the EAB's
review and affirmance of that decision. See 33 U.S.C.
§ 1369(b)(1); see also 5 U.S.C. §§ 551(13), 704.
Our scope of review is further modulated by the
scientific and technical nature of the EPA's decisionmaking here.
Adams, 38 F.3d at 49; P.R. Aqueduct & Sewer Auth. v. EPA, 35 F.3d
600, 604 (1st Cir. 1994); see also Balt. Gas & Elec. Co. v. Natural
Res. Def. Council, Inc., 462 U.S. 87, 103 (1983) ("[A] reviewing
court must remember that [where the agency] is making predictions,
within its area of special expertise, at the frontiers of science
. . . . as opposed to simple findings of fact, a reviewing court
must generally be at its most deferential."); Coal. for Responsible
Regulation, Inc. v. EPA, Nos. 09-1322 et al., 2012 WL 2381955, at
*7 (D.C. Cir. June 26, 2012) (to be published in F.3d) ("[W]e give
an extreme degree of deference to the agency when it is evaluating
scientific data within its technical expertise." (quoting Am. Farm
Bureau Fed'n v. EPA, 559 F.3d 512, 519 (D.C. Cir. 2009)) (internal
quotation marks omitted)).
-26-
We also defer to the EPA's reasonable interpretation of
the CWA. Fed. Express Corp. v. Holowecki, 552 U.S. 389, 397
(2008). This deference increases where the EPA interprets its own
regulations, Adams, 38 F.3d at 49; generally speaking, the agency's
interpretation will be "controlling unless 'plainly erroneous or
inconsistent with the regulation,'" Auer v. Robbins, 519 U.S. 452,
461 (1997) (quoting Robertson v. Methow Valley Citizens Council,
490 U.S. 332, 359 (1989)).
Below we consider the petitioners' respective arguments,
in turn, and affirm the EPA's decision as to each.
A. The EPA Did Not Commit Error by Issuing the 2008 Permit
Without Waiting for Additional Information
The District argues that the EPA should have waited to
issue the 2008 permit until after the District could complete both
its facility upgrades and its new computer model of the Blackstone
River. Instead of waiting for the "latest and best data," the
District argues, the EPA "rush[ed] to issue the permit" in a
"mechanical desire to reach a rapid conclusion without regard to
whether the result is sound." Br. of Pet'r Upper Blackstone Water
Pollution Abatement District, at 22 (quoting P.R. Sun Oil Co. v.
EPA, 8 F.3d 73, 79 (1st Cir. 1993)) (internal quotation marks
omitted). The District's arguments here fail.
The District first argues that the EPA should have waited
to reissue the new permit until after the District had fully
implemented its facility upgrades to comply with the 2001 permit
-27-
and 2002 administrative consent order. These upgrades reduced the
District's nitrogen and phosphorus discharge down from prior levels
to 8.0 - 10.0 mg/L and 0.75 mg/L, respectively, and the District
argues that the EPA should have first assessed any water quality
gains from these reductions before tightening the limits further
still.
The 2002 consent order was issued under 33 U.S.C. § 1319
for the District's violation of the 2001 permit and established a
schedule for construction of new facilities designed to achieve
compliance with the 2001 permit. As sometimes happens when a
permit requires new construction, the compliance schedule extended
beyond the five-year expiration date of the actual permit. In its
own words, the order provided "a schedule for compliance that the
Director . . . has determined to be reasonable."
The order did not purport to alter the EPA's duties under
the CWA to review and reissue permits every five years. See 33
U.S.C. § 1342(a)(3), (b)(1)(B) (permits "are for fixed terms not
exceeding five years"). EPA regulations provide that no permit's
term may be extended beyond this five-year statutory deadline
except where administratively continued by the EPA during the
permit reissuance process. See 40 C.F.R. § 122.46(a)-(b) ("NPDES
permits shall be effective for a fixed term not to exceed 5 years
. . . . Except as provided in § 122.6, the term of a permit shall
not be extended by modification beyond the maximum duration
-28-
specified in this section."); id. § 122.6(a) ("[T]he conditions of
an expired permit continue in force under 5 U.S.C. § 558(c) until
the effective date of a new permit . . . ."); see also Natural Res.
Def. Council, Inc. v. EPA, 859 F.2d 156, 212-14 (D.C. Cir. 1988)
(per curiam) (upholding continuation provision). Here, neither the
CWA nor EPA regulations allow the District's requested delay.
In addition, the record reflects that in formulating the
2008 permit limits, the EPA found that even with the fully
completed facility upgrades, the District's discharge would still
"cause, have the reasonable potential to cause, or contribute to"
a violation of water quality standards.19 40 C.F.R.
§ 122.44(d)(1)(i). As to nitrogen, the EPA determined that "a
seasonal reduction to no more than 5.0 mg/l is required . . . to
achieve water quality standards," and that "[t]here is no realistic
likelihood . . . that water quality standards could be met with a
less stringent nitrogen limit." As to phosphorus, the EPA found
that the 2001 permit's 0.75 mg/L limit is "inadequate for ensuring
the water quality standards related to the control of
eutrophication."
19
The argument has been raised in the briefing that water
quality in the Blackstone River would be improved by modifying or
removing the many industrial-era dams to allow the River's flow to
move faster. No evidence has been presented to the EPA as to this
proposal, so we may not address it. Moreover, it may be that
since the build-up of toxic sediments behind the dams poses a
serious toxic risk to water quality, this proposal may exacerbate
the River's water quality problems and so should be studied
carefully.
-29-
As to the District's computer model, neither the CWA nor
EPA regulations permit the EPA to delay issuance of a new permit
indefinitely until better science can be developed, even where
there is some uncertainty in the existing data. The five-year term
limit requires the EPA or state permitting authority to re-ensure
compliance with the Act whenever a permit expires and is renewed.
33 U.S.C. § 1342(a)(3), (b)(1)(B); 40 C.F.R. § 122.46(a), (b).
Thus, in regular intervals, the Act requires reevaluation of the
relevant factors, and allows for the tightening of discharge
conditions. The Act's goal of "eliminat[ing]" the discharge of
pollutants by 1985 underscores the importance of making progress on
the available data. 33 U.S.C. § 1251(a)(1).
In this case, the District overstates the availability of
its data during the 2008 permit process. Although it was working
on a computer model during the permitting process, the District did
not present any data from the unfinished model during the sixty-
four-day public comment period, and could not provide an estimated
date for the model's completion. Indeed, it was uncertain during
permitting whether the District would be able to successfully
complete the model at all. Multiple previous attempts by state and
federal actors to develop similar computer models had failed,
leading RIDEM's experts to conclude that the Blackstone watershed
"was too complicated to simulate with available mathematical
models." The EPA took into account this prior experience as well
-30-
as the uncertainty surrounding the District's efforts to develop
the model when it declined to delay issuance of the permit until
some indefinite point in the future.
The EPA also concluded based on the extensive scientific
record before it that even with the District's completed computer
model, there was "no reasonable likelihood that a less stringent
limit will meet [state water quality] standards." This
determination is entitled to deference.
The EPA's decision entailed not only an evaluation of the
sufficiency of the available scientific record, but also a risk
analysis of the consequences of waiting. Nitrogen-based cultural
eutrophication becomes more difficult to address the longer it is
left unchecked. Nitrogen loadings accumulate and persist in water
systems in a way that can exacerbate future water quality problems.
The EPA found that both the severity of the existing water quality
problems, and the potential for aggravated future problems,
"counsel[ed] in favor of imposing a nitrogen limit . . . based on
information currently available." This type of risk assessment is
within the EPA's policymaking discretion, and its judgment here is
entitled to respect. See Ethyl Corp. v. EPA, 541 F.2d 1, 28 (D.C.
Cir. 1976) (en banc).
The District argues that delay was especially warranted
here because the existing science was old and unreliable, and the
District's new model could offer superior information. We
-31-
addressed and rejected a similar challenge in Sur Contra La
Contaminación v. EPA, 202 F.3d 443 (1st Cir. 2000). A community
organization, SURCCo, had challenged the EPA's issuance of a
Prevention of Significant Deterioration permit under the Clean Air
Act as arbitrary and capricious, on the grounds that the analysis
relied upon by the EPA was faulty and the EPA should have required
an alternative analysis to be conducted. We rejected SURCCo's
argument, finding it was rational for the EPA "to prefer its own
model, [and] to reject SURCCo's proposed alternative modeling."
Id. at 448. We also rejected SURCCo's argument that the EPA had
"relied on outdated -- and perhaps incorrect" data, and that the
agency instead should have "relied on more recent data," available
either from the state or from an analysis yet to be conducted when
the permit was issued. Id. at 449. We credited the EPA's response
that it had "no reason to question the continuing validity" of the
data on which it relied. Id. (internal quotation mark omitted).20
20
The District also relies on this court's decision in
Puerto Rico Sun Oil Co. v. EPA, 8 F.3d 73 (1st Cir. 1993), for this
point. However, at issue in that case was the EPA's rebuff of a
state environmental quality board's request to delay the issuance
of an NPDES permit until it could assess a possible mistake it had
made in certifying the permit under 33 U.S.C. § 1341. However, the
EPA ignored the state board's request and issued a final permit
"with no explanation for its refusal to wait." 8 F.3d at 77. This
court held that the EPA's failure to provide any justification for
its actions was arbitrary and capricious. Id. at 78. In this
instance, the EPA articulated its reasons for proceeding with
issuance of the permit.
-32-
In almost every case, more data can be collected, models
further calibrated to match real world conditions; the hope or
anticipation that better science will materialize is always
present, to some degree, in the context of science-based agency
decisionmaking. Congress was aware of this when it nonetheless set
a firm deadline for issuing new permits.
As in many science-based policymaking contexts, under the
CWA the EPA is required to exercise its judgment even in the face
of some scientific uncertainty. The Supreme Court has recognized
this dimension of EPA decisionmaking in the context of the Clean
Air Act. In Massachusetts v. EPA, 549 U.S. 497 (2007), the Court
held that the EPA cannot "avoid its statutory obligation by noting
the [presence of] uncertainty." Id. at 534. If "scientific
uncertainty is so profound that it precludes EPA from making a
reasoned judgment . . . EPA must say so. That EPA would prefer not
to regulate greenhouse gases because of some residual uncertainty
. . . is irrelevant. The statutory question is whether sufficient
information exists to make an endangerment finding." Id.; see also
Miami-Dade County v. EPA, 529 F.3d 1049, 1065 (11th Cir. 2008)
(holding that the "EPA is compelled to exercise its judgment in the
face of scientific uncertainty unless that uncertainty is so
profound that it precludes any reasoned judgment"); Ethyl Corp.,
541 F.2d at 28 ("[R]ecognizing . . . the developing nature of [the
field] . . . . [t]he [EPA] Administrator may apply his expertise to
-33-
draw conclusions from suspected, but not completely substantiated,
relationships between facts, from trends among facts, from
theoretical projections from imperfect data, from probative
preliminary data not yet certifiable as 'fact,' and the like.").
The EPA did not act irrationally here by issuing the permit in the
face of some scientific uncertainty.
Both the CWA and EPA regulations provide for the
incorporation of new information into a permit once it has issued.
The District will have multiple opportunities to submit new
information to the EPA during the lengthy permit compliance
process, which entails a period of close collaboration between the
state, permittee, and EPA. The EPA noted during the permitting
process that it intended "to establish a reasonable schedule for
[the District] to come into compliance with the new nutrient
limits." It has now drafted a compliance schedule, which provides
the District with more than five years to implement the upgrades.
A full twenty-one months are allocated to the District on the front
end of this schedule for conducting testing and investigation into
what measures should be implemented to comply with the permit's
conditions. The District is required to submit its plans for
implementing the new measures to the EPA one year later, and
commence construction five months after that. The EPA has
explained that "it may be appropriate to allow some period of time
to operate the new plant before making a final decision on all
-34-
aspects of additional treatment facilities to enable [the District]
and its consultants to determine the most cost-effective
technologies for achieving the new limits."
In addition to this schedule for compliance, the CWA and
EPA regulations provide procedures for the modification of issued
permits where, in the EPA's view, change is warranted. 33 U.S.C.
§ 1342(a)(4), (b)(1)(C); 40 C.F.R. §§ 122.62(a)(2), 124.5. The EPA
has stated that "if the model being developed for [the District],
together with any other relevant evidence, makes it clear that
alternative limits will result in attainment of water quality
standards, EPA will modify the permit accordingly." The District
has already submitted a permit modification request to the EPA
based on its computer model and additional measurements it has
conducted, and the EPA may consider that request in the normal
course. The modification request is not before us, but both that
procedure and the compliance process are relevant to our evaluation
of the District's more extreme claims that it is being harmed by
the EPA's decision not to delay the 2008 permit.
The EPA did not act arbitrarily here in deciding to issue
the permit when it did.
B. The EPA Did Not Act Arbitrarily in Setting the 2008
Permit Limits
1. Nitrogen Limit
Both the District and CLF challenge the 5.0 mg/L seasonal
nitrogen limit included in the 2008 permit. The District argues
-35-
that the EPA arbitrarily selected this limit based on an unreliable
scientific model and without making the touchstone finding, under
33 U.S.C. § 1311(b)(1)(C), that the limit is "necessary to meet
water quality standards." CLF argues that the EPA drew an
unreasonable inference from the scientific record and, as a result,
set a limit that is too lenient. We affirm the EPA's decision.
a. The District's Challenges
The District first attacks a scientific model the EPA
incorporated into its analysis of the nitrogen-fueled cultural
eutrophication in Narragansett Bay. The District argues that this
model, which was created by the University of Rhode Island's Marine
Ecosystems Research Laboratory (MERL) in the 1980s to simulate
water quality conditions in the Bay,21 is so unreliable and
unrepresentative of actual Bay conditions as to entirely undermine
the EPA's nitrogen analysis.
The MERL model was peer-reviewed and published in a
scientific journal. Oviatt et al., Patterns of Productivity During
21
MERL developed a physical, enrichment gradient model of
the Bay through a series of tank experiments, which were designed
to simulate the impact of nutrient loadings in lower Narragansett
Bay and measure certain effects from these loadings, including
dissolved oxygen impairments and chlorophyll a production. The
experiment used multiple large tanks, each designed to model the
temperature, mixing, turnover, and light conditions of Narragansett
Bay; several tanks were used as controls, with conditions "similar
to a relatively clean Northeast estuary with no major sewage
inputs." Nutrients were added to the remaining tanks at varying
levels designed to reflect a range of sewage-discharge scenarios in
the Bay.
-36-
Eutrophication: A Mesocosm Experiment, 28 Marine Ecol. Progress
Series 69 (1986). It has been used by the EPA, RIDEM, and other
groups to better understand the causal relationship between
nitrogen loadings and cultural eutrophication in Narragansett Bay.
The EPA recently used the model in developing national guidance for
nutrient reduction in water systems.22
RIDEM used the model to set nitrogen effluent limitations
for Rhode Island sewage treatment plants situated along the
Blackstone River and Narragansett Bay. As part of this process,
RIDEM first conducted an extensive water quality study of the
Blackstone, Seekonk, and Providence Rivers and the Bay, sampling
and testing waters at various sites over a two-year period. It
then compared the results of this study with the outputs of the
MERL model, and concluded that both the study's results and the
model showed that higher nitrogen levels led to increased cultural
eutrophication and a less stable system in general. RIDEM
published these results, as well as its rationale for imposing new
nitrogen limits on Rhode Island sewage treatment facilities, in a
2004 Report. RIDEM, Evaluation of Nitrogen Targets and WWTF Load
Reductions for the Providence and Seekonk Rivers (2004) ("RIDEM
Report").
22
See EPA, Nutrient Criteria, Technical Guidance Manual:
Estuarine and Coastal Waters, at 2-11, 2-16 (2001), available at
http://water.epa.gov/scitech/swguidance/standards/criteria/nutrie
nts/marine/index.cfm (last visited Aug. 2, 2012).
-37-
The EPA consulted both the RIDEM Report and the Nutrient
Criteria Technical Guidance Manual in setting the nitrogen limit
for the District's 2008 permit. The record demonstrates that the
EPA carefully analyzed the MERL model during the permitting process
and compared its results with the outputs of water quality
measurements taken from the three rivers and the Bay. The EPA
found that "[b]oth the MERL tank experiments and the data from the
Providence/Seekonk River system confirm a clear correlation between
nitrogen loadings, dissolved oxygen impairment, and chlorophyll a
levels" in those water bodies. Both the MERL model and the field
measurements demonstrated that as nitrogen loadings increase,
dissolved oxygen decreases and chlorophyll a increases, with both
becoming less stable and subject to greater swings at higher levels
of nitrogen. The EPA concluded that the basic causal relationship
demonstrated in the MERL experiments "corresponds to what is
actually occurring in the Providence/Seekonk River system."
All of the parties agree that MERL's physical model did
not perfectly capture Bay conditions. The EPA recognized that the
model's flushing rate was lower than the Bay's natural flushing
rate, which may have caused the model to overestimate the impacts
of nitrogen loadings, while the model's mixing rate was higher than
that in the Bay, which may, on the other hand, have caused the
model to underestimate the impacts of nitrogen loadings. The model
did not generate the precise maximum level of nitrogen loading at
-38-
which Rhode Island's water quality standards would be maintained.
It did, however, generate a range of nitrogen loading scenarios
which the EPA used in calculating the numeric limit in the permit.
The District argues that the discrepancies between the
Bay's actual conditions and the conditions under which the
experiments were conducted are so great as to render the model
wholly unreliable. It argues that based on these "material[]"
differences, and based on the fact that the model cannot predict
the level of nitrogen control needed to meet state standards, it
was irrational for the EPA to consider the model in setting the
2008 permit's nitrogen limit.
The EPA responds that the model provided one source of
"useful information" in a multi-factored analysis and that to the
extent it did rely on the model, it fully accounted for the model's
shortcomings, and ultimately selected a nitrogen limit based on a
less stringent nitrogen loading scenario than the model, considered
alone, would warrant.
Our task is not to engage in a "de novo" evaluation of
what scientific evidence was before the EPA, but to look instead
for whether the EPA engaged in the proper decisionmaking process,
and whether its decision finds support in the record. Motor
Vehicle Mfrs. Ass'n, 463 U.S. at 43; Kennecott v. EPA, 780 F.2d
445, 449 (4th Cir. 1985) (Wilkinson, J.) ("The court best acts as
a check on agency decisionmaking by scrutinizing process and by
-39-
determining whether 'the decision was based on a consideration of
the relevant factors and whether there has been a clear error of
judgment.'" (quoting Citizens to Preserve Overton Park, Inc. v.
Volpe, 401 U.S. 402, 416 (1971))).
Where the agency follows the proper procedures and acts
with a reasonable basis, both its choice of scientific data and
interpretation and application of that data to real world
conditions are entitled to deference. Sur Contra La Contaminacion,
202 F.3d at 448; P.R. Aqueduct & Sewer Auth., 35 F.3d at 604;
see also Coal. for Responsible Regulation, 2012 WL 2381955, at *7.
Although the District singles out the EPA's reliance on
the MERL model, the EPA used many sources of information in
formulating the nitrogen limits, including both Massachusetts and
Rhode Island reports on nitrogen loadings in the Bay, water quality
studies evaluating nitrogen levels and response variables in the
Bay, and national nitrogen guidance.23 One significant source of
23
The EPA consulted multiple sources, including: Deacutis
et al., Hypoxia in the Upper Half of Narragansett Bay, RI, During
August 2001 and 2002, 13 Ne. Naturalist (Special Issue 4) 173
(2006); RIDEM, Plan for Managing Nutrient Loadings to Rhode Island
Waters (2005); Governor's Narragansett Bay Watershed Planing
Comm'n, Nutrient and Bacteria Pollution Panel Initial Report
(2004); RIDEM, Evaluation of Nitrogen Targets and WWTF Load
Reductions for the Providence and Seekonk Rivers (2004); Howes et
al., Massachusetts Estuaries Project: Site-Specific Nitrogen
Thresholds for Southeastern Massachusetts Embayments: Critical
Indicators, Interim Report (2003) (submitted to MassDEP). The EPA
argues, and the record reflects, that the MERL model provided one
source of "useful information" in an analysis that incorporated a
wide variety of information.
-40-
information the EPA examined was Rhode Island's own in-state limits
for nitrogen discharge into the relevant waters. As noted above,
Rhode Island has imposed nitrogen limits equivalent to or stricter
than the District's 5.0 mg/L limit on similarly situated sewage
treatment facilities discharging into the three rivers and the Bay.
Where the EPA did rely on the MERL model, the record
reflects that it fully accounted for the model's shortcomings. The
EPA delved into "the assumptions and methodology used in preparing
the model," Sierra Club v. Costle, 657 F.2d 298, 333 (D.C. Cir.
1981), and, far from ignoring the differences between the model and
real world conditions, it highlighted and responded to these during
the permitting process. Specifically, the EPA found that the
difference in flushing rates tended to overestimate the effects of
nutrient loadings, while the different stratification levels tended
to underestimate those effects. It did not blindly follow RIDEM's
conclusion, based on the model, that a 3.0 mg/L nitrogen limit
might be needed, but, "conscious[] of the limits of its model," id.
at 334, the EPA "chos[e] a nitrogen limit based on a less stringent
loading scenario." See Am. Coke & Coal Chems. Inst. v. EPA, 452
F.3d 930, 943 (D.C. Cir. 2006) (holding that EPA's use of the
challenged model was not arbitrary or capricious in part based on
EPA's extensive efforts to compare model assumptions against
real-world data).
-41-
The District's argument that the MERL model should have
been excluded from consideration entirely is without merit. The
EPA is not limited to models which perfectly replicate real world
conditions. A model does not have to precisely predict the actual
or an average future to increase understanding of a particular
process or the role that different elements play in that process.
The District's objection that the MERL model does not predict the
level of nitrogen control needed misstates and misunderstands the
different roles that scientific models may play in informing
science-based decisions. Here, the EPA states, and the record
reflects, that the MERL model demonstrated the relationship between
nitrogen loading, dissolved oxygen, and chlorophyll a production
for a range of loading scenarios in a water environment similar to
the Bay's.
The EPA also followed the proper procedures for ensuring
that the model received scrutiny not only from the permittee, but
from the scientific community and the public. The EPA highlighted
the model's potential shortcomings in the draft permit documents it
published for public comment. Numerous stakeholders,
organizations, and individuals submitted support for and criticism
of the model. In its detailed and extensive responses to these
comments, the EPA carefully reviewed and responded to each
criticism raised. The EAB further reviewed the EPA Region's
analysis of the model, and found no reason to fault that analysis.
-42-
"[A]dmission of uncertainties where they exist," "public exposure
of the assumptions and data incorporated into the analysis," "the
acceptance and consideration of public comment," and, ultimately,
a decision that reflects the rule of reason, are the structural
features of reasoned, publicly accountable science-based agency
decisionmaking. Sierra Club, 657 F.2d at 334 & n.130; see also
Nat'l Mar. Safety Ass'n v. Occupational Safety & Health Admin., 649
F.3d 743, 752 (D.C. Cir. 2011), cert. denied, 134 S. Ct. 1960
(2012). The EPA incorporated these structural safeguards into its
decisionmaking process.
The EPA's determination, based on its analysis of the
evidence before it as a whole, that a nitrogen limit of 5.0 mg/L
was necessary to achieve Rhode Island's water quality standards was
not a "hunch[] or wild guess[]" but a rational exercise of
judgment. Ethyl Corp., 541 F.2d at 28.
The District's second challenge to the 2008 permit's
nitrogen limit is that the EPA failed to prove that the limit is
either "necessary" or "sufficient" to attain Rhode Island water
quality standards. The CWA requires the EPA to impose certain
types of discharge limitations on point source dischargers,
including publicly owned sewage treatment facilities, such as the
District's, "including those necessary to meet water quality
standards." 33 U.S.C. § 1311(b)(1)(C). The District argues that
the EPA failed to make specific findings either that the nitrogen
-43-
limit is "necessary" to achieve Rhode Island water quality
standards or that it will "in fact" do so.
We reject the first claim, since the EPA expressly found
that the 5.0 mg/L limit was necessary to meet state standards, and
that a higher limit would not achieve those standards. In the Fact
Sheet that accompanied the original draft permit, the EPA found
"[b]ased on the available evidence, including nitrogen loadings
from [the District] and the discharge of the Blackstone River to
the Seekonk River, where the greatest impacts have been measured,
. . . [the] seasonal reduction of nitrogen to no more than 5.0 mg/l
is required at [the District's facility] in order to achieve water
quality standards." The EPA reiterated that this limit was
"necessary" to attain water quality standards at multiple other
points in the draft permit and during the permitting process.
As to the second objection, the District argues that the
EPA never found the nitrogen limit "sufficient" to attain water
quality standards and that a still lower effluent limit may be
needed. The EPA noted in the draft permit's Fact Sheet the
possibility that further monitoring "will demonstrate that
additional pollutant reductions are ultimately needed to meet water
quality standards." This review and potential tightening of the
conditions in NPDES permits is a basic feature of the CWA that the
-44-
District does not dispute.24 See 33 U.S.C. §§ 1251(a), 1313,
1342(b).
The District's argument seems to go to the precision of
the permit's nitrogen limit. But where a complex administrative
statute, like those the EPA is charged with administering, requires
an agency to set a numerical standard, courts will not overturn the
agency's choice of a precise figure where it falls within a "zone
of reasonableness." See, e.g., Nat'l Mar. Safety Ass'n, 649 F.3d
at 752; Solite Corp. v. EPA, 952 F.2d 473, 488 (D.C. Cir. 1991)
(per curiam) (judicial deference is warranted where EPA chooses "a
numerical standard . . . within a 'zone of reasonableness'"
(omission in original) (quoting Small Refiner Lead Phase-Down Task
Force v. EPA, 705 F.2d 506, 525 (D.C. Cir. 1983) (internal
quotation marks omitted))); Kennecott, 780 F.2d at 450 (EPA's
"conclusions with respect to data and analysis need only fall
within a 'zone of reasonableness'" (quoting Reynolds Metals Co. v.
EPA, 760 F.2d 549, 559 (4th Cir. 1985)) (internal quotation mark
omitted)); Hercules, Inc. v. EPA, 598 F.2d 91, 117 (D.C. Cir. 1978)
(holding that within the zone of reasonableness, "the choice of a
24
The District's "sufficiency" argument is actually
directed toward a very different point. The District argues that
in addition to determining whether specific effluent limits will
attain state water quality standards, the EPA must make a companion
determination that the state standards are in fact "attainable,"
and if not, the EPA should work with the state to revise the
standard "to reflect what can be attained." However, the District
does not argue that Rhode Island's water quality standards here are
unattainable, so we reject this argument.
-45-
precise figure is left to EPA"). The nitrogen limit the EPA chose
here is justified by the record and within the zone of
reasonableness.25 The District's challenges to the limit fail.
b. CLF's Challenges
CLF does not challenge the EPA's reliance on either the
RIDEM Report or the MERL model, but attacks the inferences the
agency drew from these sources. CLF argues that because RIDEM
25
The District makes a cursory argument that the EAB
applied an excessively deferential standard in its review of the
2008 permit's nitrogen and phosphorus limits. Since 1980, EPA's
regulations have provided that the EAB's review of NPDES permitting
decisions at the regional level shall be deferential to the EPA
Regions' determinations, see 40 C.F.R. § 124.19(a); see also 45
Fed. Reg. 33,290, 33,412 (May 19, 1980), particularly where these
involve science-based and technical judgments, see In re: NE Hub
Partners, L.P., 7 E.A.D. 561, 567-68 (EAB 1998), review denied sub
nom. Penn Fuel Gas, Inc. v. EPA, 185 F.3d 862 (3d Cir. 1999). This
deference is not unbounded, however; in its review of petitions,
the Board carefully examines the permit decisionmaking process and
the full record. See, e.g., In re: City of Marlborough, Mass.
Easterly Wastewater Treatment Plant, 12 E.A.D. 235, 248-52 (EAB
2005) (remanding permit because "the Region [had] not sufficiently
explained where or how [the compliance finding] is reflected in the
record").
In this case, the EAB exhaustively reviewed the EPA
Region's permitting decision in a thorough and exacting 106-page
opinion. The EAB carefully addressed each of the arguments of the
parties to this appeal, as well as those of seven other entities,
including the states of Massachusetts and Rhode Island. The
Board's opinion, upholding the permit in part, and remanding to the
EPA for further proceedings in part, reviewed the analysis and
methodology employed by the EPA Region in full. The Board's review
of the permit decision on this record was reasonable. To the
extent that amicus curiae, City of Marlborough, makes additional
broader arguments about the EAB's standard of review in this case,
amicus is not a party and we do not engage those arguments.
Downing/Salt Pond Partners, L.P. v. Rhode Island, 643 F.3d 16, 28
(1st Cir. 2011), cert. denied 132 S. Ct. 502 (2011).
-46-
determined in its Report that a 5.0 mg/L nitrogen limit "would not
be acceptable as [a] water quality goal[] for the area," and that
instead, a limit of at least 3.0 mg/L was necessary to ensure
compliance with Rhode Island water quality standards, the EPA acted
irrationally choosing a 5.0 mg/L limit over a more stringent limit.
The EPA responds, and the record reflects, that RIDEM
noted in its Report that "some uncertainty remains regarding
predicted water quality improvements and loading reductions
necessary to meet water quality standards. . . . For these reasons,
evaluation of phased implementation is indicated." RIDEM,
Evaluation of Nitrogen Targets and WWTF Load Reductions for the
Providence and Seekonk Rivers at 27 (2004). RIDEM, in fact, set a
limit of 5.0 mg/L for two sewage treatment facilities located along
the Providence River, both of which are comparable in size to the
District. The EPA took these factors, and many others, including
the additional studies and data we have referenced above, into
account in setting the nitrogen limit.
CLF's argument that the EPA should have interpreted
RIDEM's Report to require a 3.0 mg/L limit amounts to an attack on
the EPA's interpretation and application of the scientific data
before it to real world conditions. We give the EPA substantial
deference in this area. See Coal. for Responsible Regulation, 2012
WL 2381955, at *7; Adams, 38 F.3d at 49; P.R. Aqueduct & Sewer
Auth., 35 F.3d at 604; Kennecott, 780 F.2d at 450. Here, the EPA
-47-
independently analyzed the model and data utilized in RIDEM's
Report and reasonably concluded that certain aspects of each
warranted a slightly higher limit than the Report recommended.
This decision is entitled to deference.
CLF makes an additional argument that the EPA made the
impermissible assumption in setting the nitrogen limit that the
District's discharge will remain below its design flow of 56 mgd.
See 40 C.F.R. § 122.45(b)(1) ("POTW[] [publicly owned treatment
work] effluent limitations . . . shall be calculated based on
design flow."). However, CLF has waived this argument by failing
to present it either to the EPA Region during the permitting
process or during the initial round of briefing before the EAB. 40
C.F.R. § 124.13 ("All persons . . . who believe any condition of a
draft permit is inappropriate . . . must raise all reasonably
ascertainable issues and submit all reasonably available arguments
supporting their position by the close of the public comment period
. . . ."). As a result, the EPA did not have the opportunity to
assess or respond to CLF's objection on the record.
"Simple fairness to those who are engaged in the tasks of
administration, and to litigants, requires as a general rule that
courts should not topple over administrative decisions unless the
administrative body not only has erred but has erred against
objection made at the time appropriate under its practice." United
States v. L. A. Tucker Truck Lines, Inc., 344 U.S. 33, 37 (1952);
-48-
see also Pepperell Assocs. v. EPA, 246 F.3d 15, 27 (1st Cir. 2001).
The waiver rule serves particularly important purposes in the
administrative review context, both in that it accords respect to
the agency decisionmaking process by providing the agency with the
"opportunity to address a party's objections, . . . apply its
expertise, exercise its informed discretion, and create a more
finely tuned record for judicial review," and in doing so, guards
against a system in which regulated parties "simply turn to the
courts as a tribunal of first resort." Mass. Dep't of Pub. Welfare
v. Sec'y of Agric., 984 F.2d 514, 523-24 (1st Cir. 1993). Because
CLF failed to abide by this rule, its argument is waived.
2. Phosphorus Limit
The District objects to the 2008 permit's imposition of
a 0.1 - 1.0 mg/L seasonal limit on its phosphorus discharge. In
the way of brief background to this challenge, the District's 2001
permit limited phosphorus discharge to 0.75 mg/L in order to
address low dissolved oxygen levels, but not cultural
eutrophication, in the Blackstone River. Around the time the 2001
permit issued, the EPA was in the process of studying nutrient-
related issues more closely in water systems across the country.
In 2001, the EPA published a national action plan for the
development and establishment of numeric nutrient criteria as well
-49-
as recommended numeric criteria for most water systems.26 In
conjunction with these ongoing efforts, the EPA specifically noted
during the 2001 permitting process that more stringent phosphorus
limits might be necessary in future permits to address cultural
eutrophication impacts in the Blackstone River.
Subsequently, in order to address the severe and ongoing
phosphorus-driven cultural eutrophication in the Blackstone River,
the EPA incorporated a more stringent phosphorus limit into the
2008 permit. In formulating this limit, the EPA considered the
national and regional guidance criteria and recommended values it
had recently published. See, e.g., Buck et al., Office of Water,
EPA, Nutrient Criteria Technical Guidance Manual: Rivers and
Streams (2000); Office of Water, EPA, Ambient Water Quality
Criteria Recommendations: Information Supporting the Development of
State and Tribal Nutrient Criteria: Rivers and Streams in Ecoregion
XIV (2000) (guidance document on river watersheds in eastern
coastal states, including Massachusetts); Barbour et al., Office of
Water, EPA, Rapid Bioassessment Protocols For Use in Streams and
Wadeable Rivers: Periphyton, Benthic Macroinvertebrates, and Fish
(2d ed. 1999). It also looked at older studies, see Office of
Water Regulations & Standards, Quality Criteria for Water (1986)
26
See Grubbs, Office of Sci. & Tech., Development and
Adoption of Nutrient Criteria into Water Quality Standards (2001),
available at, http://water.epa.gov/scitech/swguidance/standards/
criteria/nutrients/upload/2009_01_21_criteria_nutrient_nutrientsw
qsmemo.pdf (last visited Aug. 2, 2012).
-50-
(the "Goldbook"), and considered site-specific data from studies
conducted after the 2001 permit had issued.
The District argues that the 2008 permit's phosphorus
limit is arbitrary because the EPA considered national guidance on
phosphorus reduction, and other regional and area studies which,
the District argues, have no proven connection to the Blackstone
River.
The EPA did not act irrationally by considering its
national and regional phosphorus guidance criteria in addition to
site-specific data. The guidance documents helped inform the EPA's
background understanding of phosphorus-driven eutrophication and
recommended "a range of ambient phosphorus concentrations that
[would be] sufficiently low to prevent cultural eutrophication" in
river systems similar to the Blackstone. See 40 C.F.R.
§ 122.44(d)(1)(vi).
Of the documents the EPA considered which recommended
specific numeric phosphorus limits, the Nutrient Criteria Technical
Guidance Manual recommends an in-stream phosphorus concentration of
0.01 - 0.09 mg/L, the Ambient Water Quality Criteria
Recommendations for Ecoregion XIV recommends an in-stream
concentration of 0.024 mg/L, and the Goldbook recommends an in-
stream concentration of 0.05 mg/L for any stream entering a lake or
reservoir, and 0.1 mg/L for any stream not discharging into an
impounded waterbody (the EPA noted that the Blackstone River is
-51-
characterized by multiple impoundments). The EPA did not blindly
follow any of these recommended limits, but after examining
additional site-specific data, including local water quality
studies, selected a phosphorus limit designed to ensure an in-
stream concentration of 0.1 mg/L.27
The EPA also analyzed various site-specific phosphorus
load data produced after 2001, including studies conducted by
MassDEP, EPA New England, and the U.S. Army Corps of Engineers.28
To account for the District's phosphorus treatment upgrade,
implemented as part of the 2001 permit and 2002 consent order, the
EPA also examined data collected when the District's phosphorus
discharge was comparable to what it would be with the upgrade. The
EPA examined data collected by MassDEP under low flow conditions in
August of 2003, when the District's average monthly discharge was
0.8 mg/L, very close to the 2001 permit's 0.75 mg/L limit.
See Fiorentino, Div. of Watershed Mgmt., Mass. Dep't Envtl. Prot.,
27
In this instance, the EPA determined that a monthly
average total phosphorus limit of 0.1 mg/L was necessary from April
1 through October 31 in order to ensure an in-stream concentration
of not more than 0.1 mg/L due to the lack of any significant
dilution in the River's waters downstream during these months.
28
See Fiorentino, Div. of Watershed Mgmt., Mass. Dep't of
Envtl. Prot., Blackstone River Watershed 2003 Biological Assessment
(2006); Tamul, Div. of Watershed Mgmt., Mass. Dep't of Envtl.
Prot., Blackstone River Watershed 2003 DWM Water Quality Monitoring
Data (2005); Wright et al., Dry Weather Water Quality Sampling and
Modeling, Blackstone River Feasibility Study (2004) (for U.S. Army
Corps of Engineers); Wright et al., Blackstone River Initiative,
Water Quality Analysis of the Blackstone River Under Wet and Dry
weather Conditions (2001) (for EPA New England).
-52-
Blackstone River Watershed 2003 Biological Assessment (2006).
MassDEP nonetheless observed a "luxuriant algal community" and
measured levels of phytoplankton which were "extremely abundant,
covering virtually the entire river bottom." The EPA reasonably
determined, and the record reflects, that the 2001 permit's 0.75
mg/L phosphorus limit would thus be insufficient to reduce cultural
eutrophication and bring the River into compliance with state water
quality standards.
To the extent the District challenges the precision of
the 2008 permit's numeric limit, we have already recognized that
the EPA's choice of a precise numeric value will be affirmed where
it is within the zone of reasonableness. See Nat'l Mar. Safety
Ass'n, 649 F.3d at 752. The permit's phosphorus limit is within
this zone of reasonableness.
The District also alleges that the EPA was required to
demonstrate both that the phosphorus limit "would have a
substantial impact on the cultural eutrophication of the Blackstone
River" and that it will alleviate not merely cultural
eutrophication but "a specific impairment in designated uses." In
other words, the District argues that any effluent limitation
imposed upon it must cure (or nearly so) the water quality problem.
The CWA quickly disposes of these arguments. The Act's
TMDL and interim planning process both contemplate pollution
control where multiple point sources cause or contribute to water
-53-
quality standard violations. 33 U.S.C. § 1313(d), (e). Under
earlier legislation, including the 1965 Federal Water Pollution
Control Act, when a water body failed to meet its state-designated
water quality standards, pollution limits could not be strengthened
against any one polluter unless it could be shown that the
polluter's discharge had caused the violation of quality standards.
See EPA v. California ex rel. State Water Res. Control Bd., 426
U.S. 200, 202-03 (1976). This standard was ill-suited to the
multifarious nature of modern water pollution and prevented the
imposition of effective controls. Id. In 1972, Congress declared
that the system was "inadequate in every vital aspect," and had
left the country's waterways "severely polluted" and "unfit for
most purposes." S. Rep. No. 92-414, at 3674 (1971). The CWA
rejected the earlier approach and, among other things, introduced
individual pollution discharge limits for all point sources. 33
U.S.C. 1311(b). To maintain state water quality standards, the Act
establishes the TMDL and continuing planning processes, which
target pollution from multiple sources. Id. § 1313(d), (e).
EPA regulations require permitting authorities to include
in NPDES permits conditions which "control all pollutants or
pollutant parameters . . . [that] are or may be discharged at a
level which will cause, have the reasonable potential to cause, or
contribute to an excursion above any State water quality standard,
including State narrative criteria for water quality." 40 C.F.R.
-54-
§ 122.44(d)(1)(i); see also 54 Fed. Reg. 23,868, 23,873 (June 2,
1989). We thus reject the notion that in order to strengthen the
District's discharge limits, the EPA must show that the new limits,
in and of themselves, will cure any water quality problems.
3. Aluminum Limit
Finally, the District challenges the limit placed on
aluminum discharge, arguing that the EPA assembled and then relied
upon an erroneous data set in deriving the limit. Specifically,
the District argues that the EPA should have excluded an "outlier
data point of 344 µg/L" in calculating the District's average daily
aluminum discharge. The EPA responds that the atypical data point
was properly included in its analysis of in-stream concentrations
of aluminum since the District did not demonstrate that the
conditions which led to the high discharge will not reoccur, and
that, in any case, the District has waived its argument.
We find that the District has waived the argument by
failing to raise it during the public comment period of the
permitting process. See 40 C.F.R. § 124.13; L. A. Tucker Truck
Lines, 344 U.S. at 37; Pepperell Assocs., 246 F.3d at 27; Mass.
Dep't of Pub. Welfare, 984 F.2d at 523-24. In its February 27,
2009, comments on the draft modification for aluminum discharge,
the District objected that the EPA "used and relied upon incomplete
and incorrect data and as a result reached incorrect conclusions."
However, this comment was directed at another argument, that the
-55-
EPA had failed to include data from certain years, and did not
state or imply that the EPA should have excluded the 344 µg/L data
point. By failing to give the EPA an opportunity to address the
argument during the permitting process, the District has waived its
claim.
IV.
The District's responsibility for serious pollution
problems in the important waterways of two states is clear, and its
challenge to the 2008 permit has no merit. As the District has
recognized, cost considerations may not be considered by the EPA in
the setting of permit limits to assure compliance with state water
quality standards. 33 U.S.C. §§ 1311(b)(1)(C), 1342(a)(2);
Defenders of Wildlife v. Browner, 191 F.3d 1159, 1163 (9th Cir.
1999); U.S. Steel Corp. v. Train, 556 F.2d 822, 838 (7th Cir.
1977). We trust that the District, as well as the EPA, will now
act with expedition to address these problems.
The District and CLF's petitions are denied. The stay
granted by this court on April 29, 2011, is lifted. No costs are
awarded.
So ordered.
-56-