American Farm Bureau Federation et al v. United States Environmental Protection Agency
Filing
150
MEMORANDUM AND ORDER: 1. Pltfs joint motion for summary judgment 95 isDENIED;2. Dft EPAs cross-motion for summary judgment 99 is GRANTED;3. Dft-Intervenor Municipal Associations Groups crossmotion for summary judgment 103 is GRANTED;4. The clerk of court is directed to enter judgment against Pltfs and in favor of Dft EPA and Dft-Intervenors on all claims.5. The clerk of court is directed to CLOSE this case. Signed by Honorable Sylvia H. Rambo on 09/13/13. (ma)
IN THE UNITED STATES DISTRICT COURT
FOR THE MIDDLE DISTRICT OF PENNSYLVANIA
AMERICAN FARM BUREAU
FEDERATION, et al.,
:
:
:
Plaintiffs
:
:
:
v.
:
:
:
UNITED STATES
:
ENVIRONMENTAL PROTECTION :
AGENCY, et al.,
:
:
Defendants
:
CIVIL NO. 1:11-CV-0067
Judge Sylvia H. Rambo
MEMORANDUM
Presently before the court are several motions for summary judgment
related to an administrative review of the issuance of the Chesapeake Bay Total
Maximum Daily Load for Nitrogen, Phosphorus, and Sediment (“TMDL”, “Bay
TMDL”, or “Final TMDL”). Plaintiffs filed a joint motion for summary judgment
(Doc. 95) and Defendant United States Environmental Protection Agency (“EPA”)
filed a cross-motion for summary judgment (Doc. 99). Some Defendant-Intervenors
filed briefs in support of EPA’s cross-motion (Docs. 102 & 108), and other
Defendant-Intervenors filed a separate cross-motion for summary judgment and
brief in support (Docs. 103 & 104) that largely supplemented EPA’s motion. For
the reasons that follow, Plaintiffs’ motion will be denied, and EPA’s and DefendantIntervenors’ cross-motions will be granted.
I.
Background
Plaintiffs are seeking a declaratory judgment and injunctive relief
against EPA, asking the court to vacate the Final TMDL for the Chesapeake Bay.
Plaintiffs allege that EPA lacked authority under the Clean Water Act (“CWA”), 33
U.S.C. § 1251, et seq., to issue the TMDL; the TMDL is ultra vires; the TMDL is
arbitrary and capricious; and EPA failed to provide adequate public notice and
comment, in violation of the Administrative Procedures Act (“APA”), 5 U.S.C. §
500, et seq. (See Doc. 16, Am. Compl.)
Understanding the legal, procedural, historical, and scientific
complexities of this case requires a detailed recitation of the extensive relevant
background information, including information regarding the parties to this suit, the
complex legal framework established under the CWA, the historical efforts to
improve water quality in the Chesapeake Bay, and the scientific modeling and
calculations utilized by EPA in promulgating the final TMDL. The court will
address each topic ad seriatum before turning to Plaintiffs’ substantive arguments.
A.
The Parties
EPA is the federal agency charged with the administration and
enforcement of the CWA, in accordance with the delegations of authority from
Congress contained in that statute. (Doc. 16 ¶ 18.) On December 29, 2010, EPA
promulgated the Final TMDL for the Chesapeake Bay, which is the subject of this
suit. (Id. ¶ 70.)
The original complaint (Doc. 1) was filed by Plaintiffs American Farm
Bureau Federation and the Pennsylvania Farm Bureau. The American Farm Bureau
Federation is a voluntary general farm organization formed in 1919 to protect,
promote, and represent the business, economic, social, and educational interests of
American farms. (Doc. 16 ¶ 7.) The American Farm Bureau Federation represents
more than 6.2 million member families through member organizations, some of
2
which are located in the 64,000-square mile Chesapeake Bay watershed. (Id. ¶¶ 7,
8.) The Pennsylvania Farm Bureau is a general farm organization that has provided
legislative support, information, and services to Pennsylvania’s farmers and rural
families since 1950. (Id. ¶ 11.) Some of the Pennsylvania Farm Bureau members
have farms located within the Chesapeake Bay watershed. (Id.)
On April 4, 2011, an amended complaint was filed, which also named
as Plaintiffs The Fertilizer Institute, a group that represents the nation’s fertilizer
industry, as well as several non-profit trade associations, to wit: the National Pork
Producers Council, the National Corn Growers Association, the National Chicken
Council, the U.S. Poultry and Egg Association, and the National Turkey Federation.
(Id. ¶¶ 12-17.)
On October 13, 2011, the court granted three motions to intervene.
(Doc. 87.) In those motions, two different groups of intervenors and a separate
municipal association, sought leave to intervene as Defendants in this action. The
first group includes various environmental and public interest groups, to wit: the
Chesapeake Bay Foundation, Inc.; Citizens for Pennsylvania’s Future; Defenders of
Wildlife; Jefferson County Public Service District; Midshore Riverkeeper
Conservancy; and the National Wildlife Federation (collectively, the “CBF Group”).
The second group includes several municipal clean water associations, to wit: the
National Associations of Clean Water Agencies (“NACWA”); the Maryland
Association of Municipal Wastewater Agencies, Inc. (“MAMWA”); and the
Virginia Association of Municipal Wastewater Agencies, Inc. (“VAMWA”)
(collectively, the “Municipal Associations Group”). The final movant was the
Pennsylvania Municipal Authorities Association (“PMAA”). The court granted the
3
motions, finding that the intervenors have a legally cognizable interest in this
litigation that could be adversely affected by the outcome of this litigation. (Id.; Am.
Farm Bureau Fed’n v. EPA, 278 F.R.D. 98 (M.D. Pa. 2011).)
B.
Statutory Framework
In addition to the alleged procedural shortcomings of the TMDL under
the APA, this dispute, at its core, raises questions regarding the proper division of
duties between the states and the federal government under the applicable CWA
statutory framework. Thus, to properly understand the parties’ respective
arguments, it is necessary to provide the framework upon which these claims rest.
This framework will provide context for later analysis of the legal issues
surrounding the Bay TMDL.
The CWA is a comprehensive water quality statute designed “to restore
and maintain the chemical, physical and biological integrity of the Nation’s waters.”
33 U.S.C. § 1251(a); PUD No. 1 of Jefferson Cty. v. Wash. Dep’t of Ecology, 511
U.S. 700, 704 (1994). “A core element of the CWA is a two-step approach to
improving water quality, which delegates certain responsibilities to EPA and others
to the states in furtherance of the Act’s stated purpose of promoting cooperation
between federal and state governments.” Anacostia Riverkeeper, Inc. v. Jackson,
798 F. Supp. 2d 210, 214 (D.D.C. 2011) (citing 33 U.S.C. § 1251(b)). Thus, water
quality restoration and maintenance efforts, as envisioned by the CWA, demand
cooperative federalism and require significant levels of communication and
4
coordination between EPA and the state environmental agencies in the six states1
and the District of Columbia (collectively, “Bay Jurisdictions”).2
Generally, efforts to improve water quality first focus on the
establishment of technology-based limitations on individual discharges into
navigable waters from point sources. 33 U.S.C. § 1311. Point sources are “any
discernable, confined and discreet conveyance . . . from which pollutants are or may
be discharged,” such as any pipe, ditch, channel, or tunnel. 33 U.S.C. § 1362(14).
These sources represent a logical starting point for monitoring and regulating water
contamination because they are easily identifiable sources of contamination. See
Anacostia Riverkeeper, 798 F. Supp. 2d at 214. Pursuant to Section 301 of the
CWA, EPA is to develop effluent limitations based upon “the best available
technology economically achievable” that cap the maximum allowable discharge at
each individual point source. 33 U.S.C. § 1311(b)(1). The primary method used to
implement these limitations is the National Pollution Discharge Elimination System
(“NPDES”). Id. The NPDES is a permit program through which individual entities
that discharge point source pollutants receive permits setting the maximum
discharge levels of a particular contaminant. See id.; see also Sierra Club v.
Meiburg, 296 F.3d 1021, 1024 (11th Cir. 2002) (“The statute gives EPA the
authority to issue permits for point sources, and those permits are to establish
technology-based effluent limitations that incorporate increasingly stringent levels
1
The Bay states include Virginia, Maryland, Delaware, West Virginia, Pennsylvania, and
New York.
2
For ease and the sake of clarity, “Bay Jurisdictions” and “Bay states” may be used
interchangeably by the court, notwithstanding the inclusion of the District of Columbia.
5
of pollution control technology over time.”); Anacostia Riverkeeper, 798 F. Supp.
2d at 214.
In addition to regulating point sources, non-point sources are also
regulated under the CWA. The distinction between point and non-point sources of
pollution is critical to understanding the primary issue in this case, as is evident
from the analysis below. As stated, point sources of pollution emanate from a
discrete conveyance. Non-point sources, meanwhile, are non-discrete sources such
as sediment run-off from agriculture fields or from timber harvesting. See
Pronsolino v. Nastri, 281 F.3d 1123, 1129 (9th Cir. 2002). Unlike point source
pollution, EPA lacks the authority to control non-point source discharges through a
permitting process. Anacostia Riverkeeper, 798 F. Supp. 2d at 214-15 (citing
Defenders of Wildlife v. EPA, 415 F.3d 1121, 1124 (10th Cir. 2005)). Thus, in order
to address water quality concerns from all sources of pollution, the CWA requires
each state to develop water quality standards for interstate waters within its border.
See 33 U.S.C. § 1313(c). These standards supplement the NPDES permitting
process. As stated in PUD No. 1, “these state water quality standards provide ‘a
supplementary basis . . . so that numerous point sources, despite individual
compliance with effluent limitations, may be further regulated to prevent water
quality from falling below acceptable levels.” 511 U.S. at 704 (quoting EPA v. Cal.
ex rel. State Water Res. Control Bd., 426 U.S. 200, 205 n.2 (1976)). Today,
“nonpoint source pollution has become the dominant water quality problem in the
United States, dwarfing all other sources of volume . . . .” Pronsolino v. Marcus
(“Pronsolino I”), 91 F. Supp. 2d 1337, 1338 (N.D. Cal. 2000), aff’d sub nom.
Pronsolino v. Nastri (“Pronsolino II”), 291 F.3d 1123 (9th Cir. 2002).
6
Water quality standards are regulations comprised of: 1) a description
of the designated use or uses of a water body; 2) the criteria necessary to protect the
use or uses; and 3) a statement by the applicable state that the standard will maintain
and protect the existing use and the water quality of the water body. 40 C.F.R. §
131.6. “Designated use” refers to “the use and value of water for public water
supplies, protection and propagation of fish, shellfish and wildlife, recreation in and
on the water, agricultural, industrial, and other purposes including navigation.” 40
C.F.R. § 131.10(a). In other words, water quality standards define the water quality
goals of a particular body of water by setting forth “the use or uses to be made of the
water and by setting criteria necessary to protect its uses.” 40 C.F.R. § 130.3.
Unlike the NPDES, which focuses on mandatory effluent limitations, water quality
standards focus on maintenance of the quality of the receiving water body.
The water quality criteria designed to protect the uses of the water body
may be expressed as numeric criteria, articulating measurable quantities of
pollutants, or as narrative statement, articulating acceptable levels of pollution in
light of the designated use. 40 C.F.R. § 131.11. These state standards must be at
least as protective of water quality as existing uses, 40 C.F.R. § 130.10, and are
subject to EPA review, 33 U.S.C. § 1313(c).
After promulgating such standards, states are primarily responsible for
monitoring progress, and identifying those waters for which the current pollution
controls “are not stringent enough to implement any water quality standard
applicable to such waters.” 33 U.S.C. § 1313(d)(1)(A). This list, known as a state’s
“impaired waters list” or “303(d) list,” identifies waters where the water quality
7
goals have not been attained, deeming those areas “water quality limited segments,”
or “WQLS.”
Finally, we arrive at the TMDL. The inclusion of a water body on a
state’s 303(d) list triggers the statutory requirement to establish a total maximum
daily load, or TMDL, for that water body. 33 U.S.C. § 1313(d)(1)(c); 40 C.F.R. §
130.7(c)(1). As set forth by EPA in its regulations, a TMDL defines the maximum
amount of a pollutant that a body of water can receive from point sources, or waste
load allocations (“WLAs”), and non-point sources, or load allocations (“LAs”). 40
C.F.R. § 130.2. Thus, a total TMDL is the “sum of the individual WLAs for point
sources and LAs for any nonpoint sources and natural background.” Id. § 130.2(I).
Before EPA establishes a TMDL, or approves a state-drafted TMDL, it
determines whether the state has provided a “reasonable assurance” that non-point
LAs will achieve water quality standards.3 (Administrative Record (“AR”)
0000063.) EPA purports to require reasonable assurances “to be sure that WLAs
and LAs established in the TMDL are not based on overly generous assumptions
regarding the amount of non-point source pollution reduction that will occur.”
(AR0000250.)
TMDLs are not self-implementing, but rather are informational tools
utilized by EPA and the states to coordinate necessary responses to excessive
pollution in order to meet applicable water quality standards. See Anacostia
Riverkeeper, 798 F. Supp. 2d at 216 (citing Pronsolino II, 291 F.3d at 1129).
TMDLs provide crucial information for federal, state, and local actors in furtherance
3
TMDLs involving only point sources do not require reasonable assurances because the
NPDES regulatory and permitting program provide the necessary reasonable assurance that the WLAs
will be achieved. (AR0000250.)
8
of the cooperative efforts to improve water quality as envisioned by the CWA. See
id. at 217. Implementation mechanisms are available under other provisions of the
CWA, as well as the Clean Air Act, state laws, federal and state regulations, and
local ordinances. (AR0000043.) States are required to submit their lists of WQLSs
and TMDLs to EPA every two years for approval. 33 U.S.C. § 1313(d)(2); 40
C.F.R. § 130.7(d)(1) & (2). States are also required to establish a priority ranking
for WQLSs based on “the severity of the pollution and the uses to be made of such
waters.” 33 U.S.C. § 1313(d)(1)(C). If EPA disapproves a state’s impaired waters
list, priority rankings, or TMDL, EPA must assume the duty to issue such a list or
TMDL, which are then incorporated into the state’s continuing planning process. 40
C.F.R. § 130.7(d)(2).
During the TMDL planning process, the seven Bay Jurisdictions were
required to submit Watershed Implementation Plans (“WIPs”). The WIPs,
developed pursuant to Section 117(g) of the CWA, 33 U.S.C. § 1267(g),4 provide
roadmaps for how the Bay Jurisdictions will achieve the preliminary target loads for
nitrogen, phosphorus, and sediment allocations necessary to meet the applicable
water quality standards. (AR0000255; AR0023289.) EPA expects that the WIPs
will identify a schedule for accomplishing nutrient and sediment load reductions,
and identify programs and actions to achieve these reductions, such as adopting new
regulatory authorities, improving compliance with existing regulations, securing
4
This section provides, in part: “The Administrator, in coordination with other members of
the Chesapeake Executive Council, shall ensure that management plans are developed and
implementation is begun by signatories to the Chesapeake Bay Agreement . . . .” 33 U.S.C. § 1267
(g)(1).
9
additional resources for cost-share programs, and issuing NPDES permits with more
stringent effluent limits. (See AR000265.)
At this juncture, it is helpful to provide an overview of past Chesapeake
Bay preservation efforts before reviewing the actions taken specifically with regard
to the final TMDL.
C.
Chesapeake Bay Preservation Efforts
The Chesapeake Bay TMDL is not a new or recent idea. Thus, it would
be improper to view the Final TMDL in a vacuum as a single, isolated effort to
restore water quality to the Chesapeake Bay. Rather, it is readily apparent from the
record before this court that the Final TMDL is merely the latest effort in a long line
of efforts dating back several decades to reach that end. In order to understand how
and why this litigation came to be, and in order to determine whether EPA’s actions
were arbitrary and capricious, it is helpful to provide a history of past preservation
and restoration efforts on the Bay.
To begin, it is essential to understand the physical characteristics of the
Chesapeake Bay. The Chesapeake Bay is an estuary in the United States, and it has
been described as one of the most biologically productive ecosystems in the world.
(AR0004682.) The Bay is approximately 200 miles long and between four and 30
miles wide. (Id.) The water surface of the Bay encompasses more than 2,500 square
miles, and the watershed, or drainage basin to the Bay, covers 64,000 square miles
in Virginia, Maryland, Pennsylvania, New York, Delaware, West Virginia, and the
District of Columbia. (Id.) Although huge in surface area, the Bay is relatively
shallow, averaging 28 feet in depth. (Id.) There are 50 major tributaries to the Bay,
the largest of which is the Susquehanna River. (Id.) Other major tributaries include
10
the Potomac, Patuxent, Rappahannock, York, James, and Choptank Rivers, as well
as the West Chesapeake Drainage Area. (Id.) As with all estuarine systems, the
water of the Chesapeake Bay is an ever-changing mixture of salt and freshwater.
(AR0004685.)
The Bay and its watershed add ecological, economic, recreational,
historic, and cultural value to the region. The Bay’s value has been estimated by
economists to exceed $1 trillion. (AR0006969.) More than 500 million pounds of
seafood, including crabs, oysters, and rockfish, are harvested from Bay waters each
year. (Id.; AR0021459.) The Bay supports a diverse ecosystem and is home to
more than 3,600 species of plants, fish, and other animals, and is a key resting
ground for migratory bird species along the Atlantic Flyway. (Id.; AR0005417.) In
light of this, Congress has recognized that the Chesapeake Bay is a “national
treasure and resource of worldwide significance.” (AR0021458.)
Much of the Chesapeake Bay watershed has been dramatically changed
and no longer resembles what Captain John Smith encountered some 400 years ago.
(AR0005509.) Indeed, for over 300 years, the Bay region has supported a number
of growing economic sectors, including forestry, agriculture, and industry.
(AR004691.) Population growth over the past century has dramatically impacted
land use in the Bay watershed. (AR0005417.) Between 1950 and 1980, for
example, the amount of developed land in the Patuxent River Basin in southern
Maryland has risen from approximately three percent to over 35 percent, a number
that is no doubt higher today. (AR0004692.) The intensified agricultural and
forestry activities and urban development have placed a significant strain on the
Bay’s ecological health. (AR0004694.) In 1982, a five-year study concluded that a
11
rapid loss of aquatic life was due to excess nutrient runoff into the Bay, namely
nitrogen and phosphorus. (AR0004725-AR0005374; AR0000050.) By 2009, more
than half of the streams in the Chesapeake watershed were rated “poor” or “very
poor” by the EPA. (AR0005511.) Non-point sources are the primary sources of
pollutants to the Bay, with agriculture alone accounting for 44 percent of the
nitrogen and phosphorus loads, and 65 percent of the sediment loads delivered to the
Bay. (AR0000136.)
Efforts to improve the water quality of the Bay have been ongoing for
more than 30 years. In 1983, the governors of Maryland, Virginia, and
Pennsylvania, as well as the Mayor of the District of Columbia, the chairman of the
Chesapeake Bay Commission, and the EPA Administrator signed the first
Chesapeake Bay Agreement. (AR0000051.) This represented the first multi-state
coordinated effort to restore water quality in the Bay. The signatories to the
Agreement acknowledged the decline of the Bay and agreed “to assess and oversee
the implementation of coordinated plans to improve and protect the water quality
and living resources of the Chesapeake Bay estuarine systems.” (AR0005488AR0005489.)
The signatories to the 1983 Agreement entered into another agreement
in 1987 with the intent of establishing a more comprehensive and coordinated
approach to restoring water quality in the Bay. (AR0000051.) The 1987 Agreement
set a key goal to “reduce and control point and nonpoint sources of pollution to
attain the water quality condition necessary to support the living resources of the
Bay.” (AR0005483.) To accomplish this goal, the signatories set the first numeric
goal for water quality enhancement: a 40 percent reduction in nitrogen and
12
phosphorus entering the Bay by 2000. (Id.) That same year, Congress amended
Section 117 of the Clean Water Act to establish the Chesapeake Bay Program
(“CBP”), directing the CBP to “coordinate state and federal efforts to improve Bay
water quality, to evaluate sediment impacts on the Bay, and to determine the impact
of natural and human-induced environmental changes on the living resources of the
Bay.” 33 U.S.C. § 1267.
In 1991, the CBP reevaluated the progress made toward achieving the
1987 Agreement’s 40 percent nutrient reduction goal. (AR0000051.) The 1991
reevaluation contained an evaluation of progress made to that point in improving
water quality, and it prescribed a detailed quantification of the original narrative
goal. (Id.) As a result of this reevaluation, several amendments to the 1987 Bay
Agreement were made in 1992. (AR0000052; AR0005478.) Based on the 1991
reevaluation, the 1992 amendments included an increased focused on the importance
of the Bay’s tributaries to the goal of water quality restoration. (Id.) The parties to
the 1987 Bay Agreement agreed to develop and begin implementation of tributaryspecific strategies to meet nutrient reduction goals and improve water quality by
August 1993. (AR0005479.)
In 1997, the CBP once again conducted an evaluation to determine
what progress had been made toward the goal set in the 1987 Agreement of a 40
percent reduction by 2000 in nitrogen and phosphorus. (AR0000052.) The
reevaluation indicated significant progress toward the nutrient reduction goals,
finding that between 1985 and 1996, phosphorus loads delivered to the Bay declined
by six million pounds annually, and nitrogen loads declined by 29 million pounds
annually over that same period. (Id.) Nevertheless, the reevaluation recognized that
13
there was no clear improvement in dissolved oxygen levels5 (“DO”), and further
concluded that it would be necessary to speed up implementation of nutrient
reduction strategies if the goal of a 40 percent reduction by 2000 was to be met. (Id.)
In 1998, EPA added the mainstem and tidal tributary waters of the
Chesapeake Bay to Virginia’s Section 303(d) list of impaired waters, thus triggering
the statutory requirement under 33 U.S.C. § 1313(d)(1)(c) for the establishment of a
TMDL for those water bodies. (AR0000063.)
On June 28, 2000, the governors of Maryland, Virginia, and
Pennsylvania, as well as the Mayor of the District of Columbia, the Administrator of
the EPA, and the chairman of the Chesapeake Bay Commission signed the
Chesapeake 2000 Agreement. (AR0000052; AR0005417-AR0005429.) The 2000
Agreement noted that water quality protection and restoration has increasingly
focused on a “tributary approach” and, for the first time, emphasized the regulatory
framework of the CWA (see supra Section I.B) along with the cooperative efforts
by the members of the CBP as the means to address nutrient enrichment problems
within the Bay and its tributaries. (AR0005421-AR0005422; AR0000052AR0000053.) Specifically, the 2000 Agreement set the overall goal of correcting
nutrient- and sediment-related problems in the Bay and its tidal tributaries sufficient
to remove those waters from the list of impaired waters by 2010. (AR0000053;
AR0005422.) To achieve this, the agreement set specific benchmarks and
5
The goal of nutrient (nitrogen and phosphorus) reduction is to increase dissolved oxygen
levels in Bay waters. Excessive nutrients act as fertilizer and encourage the growth of undesirable weed
plants and blue-green algae. When the resulting plant growth dies off, the ensuing decay causes
dissolved oxygen levels in the water to plummet, leading to anoxic conditions. Moreover, algal blooms
severely limit the growth of ecologically desirable submerged aquatic vegetation. (AR0004700AR0004701.)
14
established a cooperative framework between the Bay Jurisdictions (at that point
including Maryland, Virginia, Pennsylvania, and the District of Columbia), whereby
the Bay Jurisdictions would: (1) By 2001, define the water quality conditions
necessary to protect living resources and then assess load reductions for nitrogen,
phosphorus, and sediment for each major tributary; (2) By 2002, complete a public
process to develop and begin implementation of revised tributary strategies to
achieve and maintain the assigned loading goals; (3) By 2003, adopt new or revised
water quality standards consistent with the defined water quality conditions.
(AR0000053; AR0005422.) Once the Bay Jurisdictions adopted these revised
standards, EPA would review the standards. Following EPA’s review, the revised
standards would be the basis for removing the Bay and its tributaries from the
303(d) list of impaired waters. (Id.)
Also in 2000, EPA, Maryland, New York, Pennsylvania, Virginia, and,
for the first time, New York and Delaware, signed a Memorandum of Understanding
(“MOU”). (AR0005415-AR0005416.) This multi-jurisdictional MOU once again
recognized that, despite some progress, the Bay remained on the CWA’s list of
impaired waters and, at least insofar as this court is able to discern, mentions for the
first time the establishment of a TMDL by May 2011, unless the Bay and its
tributaries meet applicable water quality standards by 2010. (Id.) In 2002, West
Virginia signed the MOU, and the parties collectively agreed to work cooperatively
to achieve nutrient and sediment targets to cause the Bay and its tidal tributaries to
be removed from the list of impaired waters. (AR0000053.) The MOU also called
for an “inclusive, open and comprehensive public participation process” and for
collaboration in the development of innovative methods to improve water quality.
15
(Id.) It is worth noting that, as with the 1983, 1987, and 2000 Bay Agreements, the
signatories to the MOU included both the Bay states (Delaware, Maryland, New
York, Pennsylvania, Virginia, West Virginia, District of Columbia) as well as the
federal government (EPA).
In 2003, EPA and the seven Bay Jurisdictions, using the best scientific
information available, established cap loads for nitrogen, phosphorus, and sediment
entering the Bay. (AR0000053; AR0005397.) The goal was that the allocations
would serve as the basis for each state’s tributary strategy, which were scheduled to
be completed in 2004. (Id.) Specifically, the states, the District of Columbia, and
EPA agreed to cap annual nitrogen loads delivered into the Bay’s tidal waters at 175
million pounds and annual phosphorus loads at 12.8 million pounds. (AR0000054;
AR0005397). All parties concurred that attainment of these load reductions would
eliminate the persistent anoxic conditions in the deep waters of the Bay.
(AR0005398.)
As stated, in order to achieve the nitrogen, phosphorus, and sediment
cap loads, the seven Bay Jurisdictions developed their own Chesapeake Bay
tributary strategies. Each tributary strategy outlined river basin-specific
implementation activities to reduce nitrogen, phosphorus, and sediment load from
point and non-point sources with the goal of removing the Bay and its tidal
tributaries from the 303(d) list of impaired waters. (AR0000054.) By way of
example, in December 2004, the Pennsylvania Department of Environmental
Protection issued Pennsylvania’s Chesapeake Bay Tributary Strategy. (AR0024672AR0024798.) In that document, Pennsylvania recognized that in order to meet the
water quality goals set forth in the Chesapeake 2000 Agreement, reductions of 37
16
million pounds of nitrogen per year, 1.1 million pounds of phosphorus per year, and
116,000 tons of sediment per year were necessary. (AR0024674.) In the 119-page
document, numerous strategies and recommendations were proposed to show how
those goals could be met. Such strategies included: (1) limiting wastewater and
industrial discharges through the NPDES permitting process; (2) upgrading sewer
and water infrastructure by using $250 million in new grants and loans that had been
secured; (3) enhancing stormwater management through the NPDES permitting
process; (4) accelerating dam removals and building fish passageways; (5) enacting
extensive new farm management regulations through the $13 million Preserving
Agriculture, Communities, and Rural Environments (“ACRE”) initiative; (6)
expanding the Conservation Reserve Enhancement Program (“CREP”); (7)
increasing forested buffers and wetlands; (8) promoting manure-to-energy programs
by increasing the number of methane biodigesters through programs such as the
Pennsylvania Energy Harvest Grant Program, Alternative Energy Portfolio Standard,
and the First Industries Farm Investment Fund; (9) establishing a market-based
nutrient trading program; (10) securing conservation easements for riparian buffers;
and (11) expanding the $52 million Growing Greener II initiative. (AR0024675AR0024675.) The Tributary Strategy contains a detailed analysis of these strategies
and programs, their goals and expected effectiveness, and a cost table that estimates
the total tributary strategy implementation cost to be $703,318,063.6 (AR0024798.)
6
The purpose of this summary is not to convey a detailed understanding of Pennsylvania’s
Tributary Strategy, which would be well beyond the scope of this memoranda, but rather to give the
reader a sense of the types of issues and level of effort put forth by the states as members of the CBP.
Similar plans were generated by Virginia (AR0024844-AR0024928), West Virginia (AR0024929AR0024981), New York (AR0024581-AR0024671), Maryland (AR0024529-AR0024579), Delaware
(AR0024419-AR0024439), and the District of Columbia (AR0024440-AR0024528).
17
In 2007, the seven Bay Jurisdictions reevaluated their nutrient and
sediment cap loads. (AR0000055.) The 2007 reevaluation found that insufficient
progress had been made toward improving water quality to a level that indicated the
mainstem of the Chesapeake Bay and its tidal tributaries and embayments were no
longer impaired by nitrogen, phosphorus, and sediment pollution. (Id.)
Coordination of the seven Bay Jurisdictions and EPA was
accomplished mainly through the Principal Staff Committee (“PSC”). The PSC
includes the cabinet secretaries of each Bay state’s agricultural, environmental, and
natural resources departments and the EPA Region III Administrator. (AR0000055;
AR0000059-AR0000060.) A management board oversees six implementation
teams, the most crucial of which for the purposes of the TMDL is the Water Quality
Goal Implementation Team (“WQGIT”). (Id.)
At a meeting of the PSC on October 1, 2007, the seven Bay
Jurisdictions and EPA reached consensus that EPA would establish a Bay TMDL
with a target date of 2025 when all necessary pollution control measures would be in
place. (AR0000056.) Specifically, it was agreed that “The Bay watershed TMDLs
will be developed jointly between the six Bay watershed states, the District and EPA
and then established by EPA . . . no later than May 1, 2011.” (Id.; see also Meeting
Summary for the Chesapeake Bay Program Principals’ Staff Committee, October 1,
2007, available at
http://archive.chesapeakebay.net/pubs/calendar/PSC_10-01-07_Minutes_1_9029.pd
f, link provided at AR0000426.)
D.
Drafting the Bay TMDL
18
To date, more than 47,000 TMDLs have been completed throughout the
United States. (AR0000018; Doc. 110 at 14 of 52, n. 3.) The Chesapeake Bay
TMDL, however, is the largest and most complex TMDL thus far. (Id.) Between
2005 and 2010, EPA and members of the CBP met numerous times to evaluate and
agree on approaches to address multiple technical aspects related to developing the
Bay TMDL. (AR0000198.) By this court’s count, 730 CBP committee, team, and
stakeholder meetings took place during that time frame. (See AR0000422AR0000454.) Of those, 639 meetings took place after the October 1, 2007 decision
to have EPA issue the TMDL. Numerous meetings were held with the public
including national, regional, and local stakeholders to discuss issues regarding
development of TMDL models and the use of data. (Id.; AR0000060-AR0000062.)
Throughout 2009 up to the summer of 2010, EPA and the Bay states
developed target loads for nitrogen, phosphorus, and sediment for each state..
(AR0000244.) These targets were developed based on the recognition that an
equitable approach to apportionment of allowable loading among the Bay
Jurisdictions was necessary. (AR0000212.) To that end, on October 23, 2009, the
partners7 met and reached consensus on several principles to guide equitable
allocation, including:
C
The allocated loads should protect the living
resources of the Bay and its tidal tributaries and
result in all segments of the Bay mainstem, tidal
tributaries, and embayments meeting [water quality
standards] for [dissolved oxygen], chlorophyll, and
water clarity.
7
The TMDL refers to the seven Bay Jurisdictions and EPA collectively as “Partners” or “the
Partnership.” (AR0000055.)
19
C
Major river basins that contribute the most to the
Bay water quality problems must do the most to
resolve those problems (on a pound-per-pound
basis).
C
All tracked and reported reductions in nitrogen and
phosphorous loads are credited toward achieving
final assigned loads.
(Id.)
Applying those principles, EPA developed draft nitrogen and
phosphorus target loads. By way of a letter dated November 3, 2009, EPA proposed
those targets to the Bay Jurisdictions. (AR0023289-AR0023293.) On July 1, 2010
and August 13, 2010, EPA issued refined target loads as to nitrogen, phosphorus,
and sediment. (AR0000244; AR0012670-AR0012682.) The Bay Jurisdictions
developed their Phase I WIPs using these revised allocations. (AR0000244.)
Meanwhile, in May 2009, President Obama issued Executive Order
13508, which required seven federal agencies, led by the Administrator of the EPA,
and in consultation with the Bay Jurisdictions, to develop a strategy for addressing
Bay pollution and preserving Bay natural resources. (AR0006953-AR0006960.)
The executive order recognized that “at the current level and scope of pollution
control within the Chesapeake Bay’s watershed, restoration of the Chesapeake Bay
is not expected for many years.” (AR0006953.) In calling for the development of a
Bay strategy, the President acknowledged that a “renewed commitment to
controlling pollution from all sources” is required. (Id.) The President called for a
“new era of shared federal leadership with respect to restoration of the Chesapeake
Bay” and required that the seven federal agencies prepare and submit reports to that
end. (AR0006954.) The President also called for extensive consultation with the
seven Bay Jurisdictions. (AR0006956.)
20
Between 2008 and 2010, EPA provided several letters to the Bay states
explaining its expectations regarding each state’s proposed WIP. (See, e.g.,
AR0000255-AR0000256; AR0023294-AR0023301; AR0023289-AR0023293.) In
those letters, EPA acknowledged the complexities associated with drafting WIPs
and outlined a three-step process in which the WIPs would be drafted: (1) Phase I
WIPs were to be submitted to EPA by September 1, 2010; (2) Phase II WIPs by
November 1, 2011; and (3) Phase III WIPs by 2017. (AR0000256.) EPA would use
the Phase I WIPs to support the development of specific allocations in the draft Bay
TMDL. (Id.) The Phase II and III WIPs will be submitted after the Final TMDL is
established, and will refine the actions and controls implemented to achieve
applicable water quality standards. (Id.)
EPA conducted a “reasonable assurances” evaluation on the states’
draft Phase I WIPs to see if they met expectations, in terms of (1) meeting the state’s
numeric target loads, and (2) providing reasonable assurance that the state’s
proposed source and sector allocations would be met. (AR0000257; AR0024034AR0024054.) EPA found that many of the draft Phase I WIPs did not meet their
target goal and therefore adjusted the allocations accordingly. (AR0000020.) These
adjustments are referred to as “backstop” allocations. EPA then used the states’
draft Phase I WIPs in conjunction with its own backstop allocations, to issue a draft
TMDL. (Id.; AR0023773.) That TMDL was published for a 45-day public
comment period from September 24 to November 8, 2010. (AR0000016.) During
that time, EPA held 18 public meetings in all six states and the District of Columbia.
(AR0000020.) EPA also held 15 webinars in 2010 to keep the public up to date
(AR0000340) and received over 14,000 public comments from individuals as well
21
as agricultural, municipal, and environmental groups. (AR0000341.) EPA reviewed
and responded to each comment, and, where appropriate, incorporated responses to
those comments in developing the Final TMDL. (Id.; AR0000016.)
EPA continued working with the Bay Jurisdictions on strengthening the
WIPs (AR0000025) and, upon receipt of the final Phase I WIPs, found those plans
to be considerably improved compared to the draft WIPs. (AR0000263,
AR0000266.) As a result, EPA was able to significantly reduce the number of
backstop allocations used in the Final TMDL. In the Final TMDL, backstop
measures were provided only in the three following instances: (1) making New
York’s WLA for wastewater sources more stringent (AR0000285-AR0000286); (2)
shifting 50 percent of Pennsylvania’s urban stormwater load that is not currently
subject to NPDES permits from the LA category to the WLA category
(AR0000287); and (3) shifting 75 percent of the pollutant loads that West Virginia
allocated to animal feeding operations that are not subject to NPDES permitting
from the LA category to the WLA category and signaling that EPA is prepared to
designate any animal feeding operations as a source requiring a NPDES permit
(AR0000292). The remainder of the Final TMDL was based on the Bay states’ final
Phase I WIPs, which EPA determined had satisfied the reasonable assurances
analysis. On December 29, 2010, the Final Bay TMDL was issued. (AR0000015AR0003790.) That TMDL set forth allocations of 185.9 million pounds per year
(mpy) of nitrogen (representing a 25 percent reduction from current levels), 12.5
mpy of phosphorus (representing a 24 percent reduction), and 6.45 billion pounds
per year of sediment (representing a 20 percent reduction) among the Bay
Jurisdictions. (AR0000016.) The TMDL requires that all pollution control
22
measures be fully implemented by 2025, with at least 60 percent of the actions taken
by 2017. (Id.; AR0000021.)
E.
Consent Decrees, Settlement Agreements, and Memoranda of
Understanding
The Chesapeake Bay TMDL has also been the subject of considerable
litigation over the years. This was not always the case, however, as state and federal
governments largely ignored their obligations under CWA Section 303(d) in the
years after its passage. Indeed, given the complexities, costs, scientific uncertainties
associated with identifying impaired waters and determining TMDLs for those water
bodies, and perceived difficulties in implementing TMDLs, states were initially
reluctant to undertake such efforts. Likewise, EPA largely ignored CWA Section
303(d) until environmental groups began bringing citizen’s suits against EPA for
inadequately implementing TMDLs as envisioned by the CWA. See, e.g., Scott v.
City of Hammond, 741 F.2d 992 (7th Cir. 1984); Alaska Ctr. for the Env’t v.
Browner, 20 F.2d 981 (9th Cir. 1994); Idaho Sportsmen’s Coal. v. Browner, 951 F.
Supp. 962 (W.D. Wash. 1996). More relevant to the case sub judice are several
consent decrees, MOU’s, and settlement agreements involving the establishment of
a TMDL for Chesapeake Bay and its tributaries. For example, in 1996, the
American Littoral Society and Sierra Club filed suit against EPA due to EPA’s
failure to, inter alia, establish TMDLs for all WQLSs in Delaware, in violation of
the APA and CWA. (AR0012640; Am. Littoral Soc’y v. EPA, Docket No. 96-cv591 (D. Del. 1997).) The consent decree, which was reviewed and approved by the
court and effectively resolved the lawsuit, called for EPA to “establish TMDLs for
the balance of all pollutants for all WQLSs for which Delaware has not established
TMDLs by December 15 of the year following the State’s deadline, except that EPA
23
shall establish all such TMDLs by December 15, 2006.” (AR0012647.) In other
words, the consent decree required EPA to establish TMDLs if Delaware failed to
do so within the 10-year TMDL development schedule, which was attached to the
decree. (AR0000066; AR0012668.)
EPA entered into a similar consent decree in Kingman Park Civic
Assoc. v. EPA, Docket No. 1:98-CV-00758 (D.D.C. June 13, 2000). In that case,
Plaintiffs Kingman Park Civic Association, Friends of the Earth, and the Anacostia
Watershed Society sued EPA alleging that EPA failed to establish TMDLs for all the
District of Columbia’s WQLSs, which constituted a violation of CWA Section
303(d). The court-approved consent decree required EPA to, inter alia, establish
TMDLs for the District’s portions of the tidal Potomac and Anacostia rivers, if not
first established by the District by a certain date. (AR0000066; AR0012502AR0012526.) These rivers are tidal tributaries to the Chesapeake Bay and share
common impairing pollutants (nitrogen and phosphorus). Thus, establishment of
TMDLs on these rivers is directly related to the establishment of the Bay TMDL.
(AR0000066.)
In 1999, EPA entered into another consent decree in American Canoe
Assoc. v. EPA, Docket No. 98-cv-979 (E.D. Va. June 11, 1999). In that case,
Plaintiffs, American Canoe Association, Inc., and the American Littoral Society,
sued EPA as a result of EPA’s failure to establish a TMDL for all waters on
Virginia’s Section 303(d) list. Specifically, Virginia was required to submit its
Section 303(d) list and its TMDL by June 26, 1979. See Am. Canoe Assoc. v. EPA,
30 F. Supp. 908, 913 (E.D. Va. 1998). Virginia had failed to do so, and by the time
the case was filed nearly 20 years later, had still failed to submit a TMDL. In
24
response to EPA’s motion to dismiss, Plaintiffs argued, inter alia, that Virginia’s
failure to submit a TMDL by the 1979 deadline constituted a constructive
submission that no TMDLs are required, and the CWA compelled EPA to
disapprove this position as inadequate and establish a federal TMDL for Virginia’s
WQLSs. EPA disagreed, arguing that its duty to approve or disapprove is triggered
only when the states submit their TMDL to EPA. Thus, EPA argued, because
Virginia submitted no TMDL, there was no duty to disapprove and submit its own
TMDL. Id. at 919. The court disagreed with EPA and declined to dismiss the
plaintiffs’ complaint on this ground, finding that “it seems highly likely that
Congress intended that EPA should be required to act not only when states
promulgate lists that fail to meet the standards set forth in Section 303, but also
when states completely ignore their mandatory statutory responsibilities and fail to
promulgate any list at all.” Id. at 921. The suit eventually settled pursuant to a
court-approved consent decree, which required EPA to establish a Bay TMDL if
Virginia failed to do so by May 1, 2011, in accordance with the schedule established
in the consent decree. (AR0000065; AR0012537-AR0012538.) However, as stated
above, Virginia, as well as all other Bay Jurisdictions, requested in 2007 that EPA
establish TMDLs for nutrient- and sediment-impaired tidal portions of the
Chesapeake Bay. (AR0000056, AR0000065.)
In addition to the above consent decrees which required EPA to
establish nutrient and sediment TMDLs for the Chesapeake Bay, EPA also entered
into an MOU with Maryland which required the Maryland Department of the
Environment to use available resources to establish and submit to EPA a TMDL for
each WQLS identified in Maryland’s 303(d) list by 2008 (“Maryland MOU”).
25
(AR0012626.) This goal was somewhat superceded by the Chesapeake 2000
agreement which targeted 2010 as the year to achieve water quality standards.
(AR0000067.) Accordingly, Maryland and EPA entered into a revised MOU in
September 2004, that extended the schedule for TMDL development to 2011.
(AR0012454-AR0012501.) Regardless, in 2007, Maryland, like all the other Bay
Jurisdictions, requested that EPA take the lead in developing TMDLs for its portion
of the Chesapeake Bay watershed, in essence mooting the Maryland MOU.
(AR0000056; AR0000067.)
Finally, in January 2009, the Chesapeake Bay Foundation and others
filed suit against EPA alleging, inter alia, that EPA has failed to carry out its
nondiscretionary duties under Section 117(g) of the CWA, 33 U.S.C. § 1267(g), to
achieve and maintain the goals of the Chesapeake Bay Agreement. (AR0000067;
AR0012363; Fowler v. EPA, No. 1:09-C-00005-CKK (D.D.C. 2009).) In May,
2010, the parties entered into a settlement agreement whereby EPA was to establish
a nutrient and sediment TMDL for the Bay and its tidal tributaries by December 31,
2010. (AR0000067; AR0012374.) In the case sub judice, EPA asserts that it
established the Final TMDL in part to meet its commitment under that settlement
agreement, and further contends that the establishment of the TMDL is consistent
with EPA’s duties established in the other consent decrees. (AR0000067; Doc. 100
at 18 of 76.)
The above historical recitation of the Bay TMDL development, and the
legal challenges to date, are, without doubt, complicated and confounding.
However, as previously stated, a familiarity with the evolution of the Bay TMDL is
relevant to Plaintiffs’ challenge of the Final TMDL. To simply view the Final
26
TMDL in a vacuum would ignore the bigger question of why this complicated
regulatory procedure has been established in such a manner. With this
understanding of the historical and legal context of the TMDL, the court is better
positioned to address Plaintiffs’ challenges.
F.
Procedural History
Plaintiffs filed their initial complaint on January 10, 2011 (Doc. 1),
followed by an amended complaint (Doc. 16) challenging the Bay TMDL, and
seeking a declaratory judgment and injunctive relief against EPA and requesting that
the court vacate the TMDL. On May 25, 2011, the CBF Group filed a motion to
intervene (Doc. 25) and a brief in support on June 3, 2011 (Doc. 52). Also on May
25, 2011, the Municipal Associations Group filed a motion to intervene and brief in
support. (Docs. 27 & 29.) Plaintiffs filed a consolidated response on June 20, 2011.
(Doc. 57.) Reply briefs were filed by the Municipal Associations Group (Doc. 66)
and the CBF Group (Doc. 67) on July 5 and July 7, 2011, respectively. On June 27,
2011, PMAA filed a motion to intervene (Doc. 59) and brief in support (Doc. 61).
Plaintiffs filed a brief in opposition on July 14, 2011 (Doc. 68), to which a reply
brief was filed on July 28, 2011 (Doc. 70). On October 13, 2011, the court issued a
memorandum and order granting all three motions to intervene. (Doc. 87; Am.
Farm. Bureau Fed’n v. EPA, 278 F.R.D. 98 (M.D. Pa. 2011).)
On August 26, 2011, EPA filed a notice of lodging the administrative
record and a certified index to the record. (Doc. 76.) The administrative record was
filed on September 1, 2011 (Doc. 77) and electronic copies of the record were sent
to the court and the parties.
27
On October 11, 2011, Plaintiffs filed a motion to complete the
administrative record and brief in support, requesting the court add additional
documents to the administrative record. (Docs. 82 & 85.) Briefs in opposition were
filed by EPA (Doc. 88) and Defendant-Intervenors (Doc. 89) on October 28, 2011.
Plaintiffs filed a reply brief on November 14, 2011. (Doc. 91.) On December 18,
2011, the court issued a memorandum and order granting in part and denying in part
the motion, permitting some of the requested documents to be added to the record.
(Doc. 92.)
On January 27, 2012, Plaintiffs filed the instant motion for summary
judgment and brief in support. (Docs. 95 & 96.) On March 27, 2012, EPA filed a
cross-motion for summary judgment and a brief in opposition to Plaintiffs’ motion
for summary judgment. (Doc. 100.) On April 20, 2012, Defendant-Intervenor
PMAA filed a brief in support of EPA’s cross-motion for summary judgment and in
opposition to Plaintiffs’ motion for summary judgment. (Doc. 102.) On April 20,
2012, Defendant-Intervenor Municipal Associations Group filed its own crossmotion for summary judgment and brief in support of that motion and in opposition
to Plaintiffs’ motion for summary judgment. (Docs. 103 & 104.) On April 24,
2012, Defendant-Intervenor CBF Group filed a brief in support of EPA’s crossmotion for summary judgment and in opposition to Plaintiffs’ motion for summary
judgment. (Doc. 108.) On May 21, 2012, Plaintiffs filed a brief in opposition to
EPA’s cross-motion for summary judgment and a reply in support of its own motion
for summary judgment. (Doc. 109.) On June 20, 2012, EPA filed its brief in
support of its cross-motion for summary judgment. (Doc. 110.) Reply briefs were
28
filed by PMAA and the Municipal Associations Group on July 13, 2012. (Docs. 115
& 116.) On July 20, 2012, the CBF Group filed its reply brief.8 (Doc. 122.)
Given the complexities of this case and the volume of documents in the
administrative record, Plaintiffs moved for oral argument on the cross-motions for
summary judgment on August 3, 2012. (Doc. 123.) EPA and DefendantIntervenors opposed the motion. (Doc. 125.) By order dated August 10, 2012, the
court granted the motion and, on October 4, 2012, the court heard oral arguments on
the motions for summary judgment. Following oral argument, the court requested
additional briefing on the issue of agency deference. (Doc. 138.) Briefs on this
issue were filed by Plaintiffs on October 17, 2012 (Doc. 139), and by EPA and the
CBF Group on October 24, 2012 (Docs. 140 & 142). Defendant-Intervenors PMAA
and the Municipal Associations Group abstained from filing additional briefing.
(Docs. 141 & 143.) On November 2, 2012, Plaintiffs filed a reply brief. (Doc.
146.) Thus, all issues having been fully briefed, the motion and cross-motions for
summary judgment are ripe for disposition.
II.
Standard
Under Federal Rule of Civil Procedure 56, summary judgment will be
granted “if the movant shows that there is no genuine dispute as to any material fact
and the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a).
However, unlike the typical summary judgment action, Plaintiffs are seeking judicial
review under the APA. While “[s]ummary judgment is the proper mechanism for
8
Over Plaintiffs’ objections, the court also accepted for consideration the Amicus Curiae
Memorandum filed by the City of Annapolis, Maryland. (Doc. 117.)
29
deciding, as a matter of law, whether an agency action is supported by the
administrative record and consistent with the APA standard of review[,] . . .
[b]ecause . . . ‘the district judge sits as an appellate tribunal’ in such cases, the usual
standard for summary judgment does not apply.” Udin v. Mayorkas, 862 F. Supp.
2d 391, 399-400 (E.D. Pa. 2012) (citing Am. Bioscience, Inc. v. Thompson, 269 F.3d
1077, 1083 (D.C. Cir. 2001); UMPC Mercy v. Sebelius, 793 F. Supp. 2d 62, 67
(D.D.C. 2011)); Occidental Eng’g Co. v. I.N.S., 753 F.2d 766, 770 (9th Cir. 1985)
(explaining that summary judgment in an original district court proceeding “is
appropriate only when the court finds there is no factual issues requiring resolution
by trial”; whereas, summary judgment in a case where the district court is reviewing
the decision of an administrative agency under the APA “is an appropriate
mechanism for deciding the legal question of whether the agency could reasonably
have found the facts as it did.”).
Under the APA, a district court may only hold unlawful and set aside an
agency action if it is “arbitrary, capricious, an abuse of discretion, or otherwise not
in accordance with law.” 5 U.S.C. § 706(2)(A). The court is required to walk a fine
line of conducting a “searching and careful” inquiry into the administrative record to
determine whether the agency’s decision was “based on a consideration of the
relevant factors and whether there has been a clear error of judgment” while, at the
same time, refraining from substituting its own judgment for that of the agency.
Citizens Advisory Comm. on Private Prisons v. Fed. Bureau of Prisons, 197 F.
Supp. 2d 226, 240 (W.D. Pa. 2001) (quoting Citizens to Protect Overton Park, Inc.
v. Volpe, 401 U.S. 402, 415-17 (1971) and Soc’y Hill Towers Owners’ Ass’n v.
Rendell, 201 F.3d 168, 178 (3d Cir. 2000)). Even an agency “decision of less than
30
ideal clarity” should be upheld “if the agency’s path may be reasonably discerned.”
Anacostia Riverkeeper,798 F. Supp. 2d at 222 (quoting Motor Vehicle Mfrs. Ass’n v.
State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983)). At the same time, it is “an
axiom of administrative law that an agency’s explanation of the basis for its decision
must include a rational connection between the facts and the choice made.” Id.
(quoting Bowen v. Am. Hosp. Ass’n, 476 U.S. 610, 626 (1986)). It has been
repeatedly stated that “the focal point for judicial review should be the
administrative record already in existence, not some new record made initially in the
reviewing court.” United States v. Keystone Sanitation Co., 1996 U.S. Dist. LEXIS
22808, *10 (M.D. Pa. Aug. 27, 1996) (quoting Camp v. Pitts, 411 U.S. 138, 142
(1973)). The “whole record” consists of materials either directly or indirectly
considered by the decision maker. Keystone Sanitation, 1996 U.S. Dist. LEXIS
22808, *23 n.6 (citing Wade v. Dole, 631 F. Supp. 1100, 1107 (N.D. Ill. 1986), aff’d
813 F.2d 798 (7th Cir. 1987)); see also Bar MK Ranches v. Yuetter, 994 F.2d 735,
739 (10th Cir. 1993); Ohio Valley Envtl. Coal. v. Whitman, 2003 U.S. Dist. LEXIS
148, *8 (S.D. W.Va. Jan. 6, 2003). “An agency's action is entitled to a presumption
of validity, and the petitioner challenging that action bears the burden of
establishing that the action is arbitrary or capricious.” Forest Guardians v. U.S.
Fish & Wildlife Serv., 611 F.3d 692, 704 (10th Cir. 2010); see also Taggart v.
GMAC Mortg., LLC, 2013 U.S. Dist. LEXIS 113823 *10 n.6 (E.D. Pa. Aug. 12,
2013).
The court owes Chevron deference to the extent that EPA’s actions are
based on an interpretation of statutory language. All parties are in agreement that
the applicable analysis is the two-step analysis set forth in Chevron, USA, Inc. v.
31
Natural Res. Def. Council, 467 U.S. 837, 842 (1984).9 Under Chevron, an agency’s
interpretation is entitled to deference if “Congress delegated authority to the agency
generally to make rules carrying the force of law, and . . . the agency interpretation
claiming deference was promulgated in the exercise of the authority.” United States
v. Mead, 533 U.S. 218, 226-27 (2001). If a court finds Chevron deference applies,
the court must first ask “whether Congress has directly spoken to the precise
question at issue.” Id. “If the intent of Congress is clear, that is the end of the
matter; for the court, as well as the agency, must give effect to the unambiguously
expressed intent of Congress.” Id. at 842-43. Second, a court asks whether, if the
statute is ambiguous, the agency has rendered “a permissible construction.” Id. at
843.
A court is more likely to find the agency’s interpretation permissible if
there is a “complex and highly technical regulatory program,” Robert Wood Johnson
Univ. Hosp. v. Thompson, 297 F.3d 273, 282 (3d Cir. 2002) (citations and
quotations omitted), or if the agency has employed formal procedures, such as notice
and comment rulemaking, see Christensen v. Harris Cnty., 529 U.S. 576, 587
(2000). If Chevron deference applies, the court must defer to the agency’s
interpretation as long as it is reasonably consistent with the statute. See Mead, 533
U.S. at 229. Where a court declines to extend Chevron deference, it may
nonetheless extend the lesser degree of deference set forth in Skidmore v. Swift &
9
Although Plaintiffs did not believe Chevron deference should be applicable “in the context
of an agency interpreting a statute to determine the limits of its own jurisdiction,” Plaintiffs conceded
that the Third Circuit has held that it is applicable. (Doc. 139, n. 3 (citing NE Hub Partners, L.P. v.
CGN Transmission Corp., 239 F.3d 333, 355 (3d Cir. 2001).)
32
Co., 323 U.S. 134 (1944).10 However, the court is in agreement with the parties that
Chevron deference is appropriate in this case. See Pronsolino II, 291 F.3d at 1131
(finding Chevron deference applicable in a TMDL case because the “EPA has the
statutory authority to enact a rule carrying force of law [in a TMDL case because]”
the CWA delegates to EPA the general rule-making authority necessary for the
agency to carry out its functions under the Act).
III.
Discussion
Plaintiffs put forward various arguments in support of vacating the Bay
TMDL. As a preliminary matter, however, the court will address EPA’s argument
that Plaintiffs lack standing to bring suit. The court will then address each of
Plaintiffs’ arguments ad seriatum.
A.
EPA’s Standing Argument
EPA argues for the first time in its memorandum in support of its crossmotion for summary judgment (Doc. 100) that Plaintiffs lack standing to bring suit
under Article III of the Constitution. Article III limits the federal courts to
adjudication of actual “[c]ases” and “[c]ontroversies.” U.S. Const. Art. III, § 2, cl.
1. “Standing circumscribes the federal judicial power by requiring a litigant to show
that it is entitled to have the court decide the merits of its case.” Am. Auto. Ins. Co.
v. Murray, 658 F.3d 311, 317 (3d Cir. 2011) (citing Allen v. Wright, 468 U.S. 737,
750-51 (1984)). The three constitutional elements of standing are: (1) an “injury in
fact,” that is, a concrete and particularized invasion of a legally protected interest
10
Under Skidmore, a court defers to an agency’s position according to its persuasiveness.
323 U.S. 139-140. Factors relevant to this analysis include the agency’s expertise, care, consistency, and
formality, as well as the logic of the agency’s position. Id.
33
that is actual or imminent, not conjectural or hypothetical; (2) causation, the
showing of a fairly traceable connection between the alleged injury in fact and the
alleged conduct of the defendant; and (3) redressability, that is, “it must be ‘likely,’
as opposed to merely ‘speculative,’ that the injury will be ‘redressed by a favorable
decision.’” Id. (quoting Lujan v. Defenders of Wildlife, 504 U.S. 555, 560-61
(1992)).
The import of EPA’s argument in this regard is that Plaintiffs failed to
submit affidavits or set forth any evidence to establish the requisite elements of
representational standing. In response to this argument, Plaintiffs, in their brief in
opposition to EPA’s cross-motion, attached 13 declarations which contain
statements pertaining to the various Plaintiffs’ standing in this case. (Docs. 109-1 –
109-13.) EPA nevertheless argues in its reply brief that such affidavits must be
submitted “at the first appropriate point in the review proceeding,” in this case
Plaintiffs’ opening brief, and that Plaintiffs produced no reason to excuse the
untimely filing of the declarations. (Doc. 110 at 10 of 52 (citing Communities
Against Runway Expansion, Inc. v. FAA, 355 F.3d 678, 684 (D.C. Cir 2004)
(“CARE”).)
The court rejects EPA’s argument. First, while it is true that at the
summary judgment stage, a plaintiff may not base standing on mere allegations and
must set forth by affidavit or other evidence of “specific facts” to support standing,
see Sierra Club v. EPA, 292 F.3d 895, 900 (D.C. Cir. 2002), there is no hard and fast
rule that failure to attach said affidavits to an opening brief automatically divests a
plaintiff of Article III standing, and thus the court of subject matter jurisdiction.
Indeed, in CARE, a case relied upon by EPA, the court readily excused the plaintiff’s
34
belated submission of declarations regarding standing, where, like here, the
declarations were not filed concurrent with the opening brief. 355 F.3d at 685. The
court found that the supplemental declarations clearly demonstrated “injury in fact”
sufficient for standing purposes, and further found that the agency was not
prejudiced because it was permitted to respond to the declarations. Id.
Here, too, EPA’s prejudice is limited because it was able to respond to
Plaintiffs’ declarations in its reply brief, wherein EPA merely restates its
untimeliness argument. (Doc. 110 at 10-11 of 52.) Moreover, the affidavits were
not filed egregiously late in the proceedings. Indeed, the first time standing was
even challenged was in EPA’s memorandum in support of its cross-motion. This
was not a case where a preliminary motion to dismiss for lack of standing was filed.
Nor was this a case where a plaintiff did not submit any evidence regarding standing
until its final brief, thus requiring post-argument supplementation, as in Sierra Club.
Rather, the court finds Plaintiffs’ standing to seek review of this administrative
action to be self-evident. As stated by the court in Sierra Club,
[I]n many, if not most cases the petitioner’s standing to
seek review of administrative action is self-evident; no
evidence outside the administrative record is necessary for
the court to be sure of it. In particular, if the complainant
is “an object of the action . . . at issue” – as is the case
usually in review of a rulemaking and nearly always in
review of an adjudication – there should be “little question
that the action or inaction has caused him injury, and that
judgment preventing or requiring the action will redress
it.”
Sierra Club, 292 F.3d at 899-900 (quoting Lujan, 504 U.S. at 561-62.) Here, the
declarations make clear that Plaintiffs are the object of the action, and that they will
incur economic injury if required to comply with the TMDL. For example, some of
the Plaintiffs are point source dischargers and are NPDES permit holders who will
35
need to comply with the standards set forth in the Bay TMDL. (See, e.g., Doc. 1091, Shafer Decl. ¶¶ 6, 8, 10 (“Pilgrim’s Pride [a member of the U.S. Poultry & Egg
Association] has determined that in order to comply with the individual wasteload
allocations in the TMDL, it will need to change its wastewater treatment operations
[and] employ new treatment technology [which will be] prohibitively expensive for
most businesses and municipalities.”); Doc. 109-2, Igli Decl. (“Each Tyson
[(member of U.S. Poultry & Egg Association)] facilit[y] operates a wastewater
treatment plant [and] holds a [NPDES] permit. [To meet the requirements of the Bay
TMDL] will require additional costs to achieve compliance. Tyson also expects
increases in operating costs associated with all facilities.”).) Other declarations
contain similar concerns and declarations of economic injuries. (See Doc. 109-3,
Behrer Decl. ¶¶ 12-14 (describing economic harm from changes to dairy operations
that a farm needs to undertake to comply with “aggregate” WLAs); Doc. 109-4,
Kettler Decl. ¶¶ 11-14, 16 (describing costs of complying with the Final TMDL as
including engineering and planning work to develop stormwater pollution control
plans, installation of systems to treat and control stormwater, and implementation
and maintenance measures); Doc. 109-5, Sowers Decl. ¶¶ 8-15 (same); Doc. 109-6,
Herz Decl. ¶¶ 5-8 (describing how the Final TMDL will increase the cost of
complying with discharge limitations and result in reduced fertilizer sales); Doc.
109-7, Doggett Decl., ¶¶ 4, 6 (describing how pollutant allocations in the Final
TMDL will increase already significant costs associated with nutrient management
plans for corn farms).) It has long been held that economic injuries are a sufficient
36
basis for standing. See Clinton v. City of N.Y., 524 U.S. 417, 434-33 (1998); see
also Sierra Club v. Morton, 405 U.S. 727, 733-34 (1972).11
Moreover, based on a review of the declarations, the court has no
trouble concluding that the causation and redressability requirements are also
satisfied. CBF argues that the Final TMDL is not the cause of Plaintiffs’ injuries
because the TMDL is not self-implementing; rather, the TMDL is implemented by
the states. This view is too restrictive. A TMDL, while only informational in
nature, is, in this case, the product of extensive collaboration between the Bay states
and EPA. The TMDL was used by the states to construct the WIPs, which are the
cause of the alleged injuries. Moreover, EPA’s role is not purely passive: EPA is
the permitting authority for point source pollution, via NPDES permitting. In short,
if the court were to adopt CBF’s argument, then ostensibly no party, including
environmental groups, regulated entities, or trade groups, would ever have standing
to challenge a TMDL in federal court as either insufficiently protective or in excess
of EPA’s regulatory authority. Yet, many courts have asserted jurisdiction over
such cases. See, e.g., Natural Res. Def. Council, Inc. v. Muszynski, 268 F.3d 91 (2d
Cir. 2001); Dioxin/Organochlorine Ctr. v. Clarke, 57 F.3d 1517 (9th Cir. 1995);
Anacostia Riverkeeper, 798 F. Supp. 2d 210. Accordingly, the court finds that
Plaintiffs’ alleged imminent injuries are fairly traceable to the Final TMDL, and thus
causation is satisfied.
11
Defendant-Intervenor CBF also argues that the alleged injuries are speculative and not
sufficiently imminent. (Doc. 108 at 17 of 39.) While it is true that the concept of imminence is elastic,
it should not be “stretched beyond the breaking point” by only alleging an injury “at some indefinite time
in the future.” Lujan, 504 U.S. at 564 n. 2. The court does not find that the alleged economic injury is
so speculative. Plaintiffs, through the various declarations, state that, if the Final TMDL is upheld, then
economic injury will result as a result of the reduced nutrient allocations. Although this injury might not
be present or immediate, it is sufficiently imminent for the purposes of Article III standing.
37
Lastly, the court finds that the requested relief, vacatur of the Final
TMDL, would likely relieve Plaintiffs of their alleged injuries. Redressability is
closely related to traceability, except that traceability looks backward (did the
defendants cause the harm?), while redressability looks forward (will a favorable
decision alleviate the harm?). See Toll Bros., Inc. v. Twp. of Readington, 555 F.3d
131, 143 (3d Cir. 2009) (citing Lujan, 504 U.S. at 560-61). Plaintiffs need not show
that a favorable decision will certainly redress their members’ injuries, only that it is
likely to do so. Id. Here, based on the above, the court is satisfied that Plaintiffs
have met the redressability prong, as vacatur of the TMDL will likely alleviate at
least some of Plaintiffs’ economic concerns.
In short, Plaintiffs have satisfied the test for Article III standing: injuryin-fact, causation, and redressability. That the declarations providing evidence in
support of Plaintiffs’ standing were not filed in the opening brief, but rather in their
opposition brief to Plaintiffs’ cross-motion, is not fatal to Plaintiffs’ standing.
B.
Bay TMDL is not an unlawful federal implementation plan
In this issue of first impression, Plaintiffs argue that the Final TMDL
represents an unlawful federal implementation because it impedes on the states’
rights to implement the TMDL as each state sees fit. Plaintiffs assert that, while
EPA may issue a TMDL, EPA has no authority to implement a TMDL. Plaintiffs
further argue that only Congress can grant EPA authority to implement TMDLs, and
no executive order, consent decree, or MOU can expand EPA’s authority. Plaintiffs
point to the level of detail of TMDL allocations, as well as EPA’s backstop
measures as evidence of unlawful federal implementation measures. For the reasons
set forth below, the court agrees with Plaintiffs that TMDL implementation
38
responsibilities primarily fall to the individual states, but disagrees that the Final
TMDL represents an unlawful implementation plan.
1. CWA §§ 303 (33 U.S.C. § 1313) and 117 (33 U.S.C. § 1267)
As explained in detail above, the CWA sets forth a step-by-step
approach to restoring impaired waters. See supra Section I.B. To review, the CWA
first requires the establishment of water quality standards. 33 U.S.C. § 1313(a) &
(c). Second, states are required to identify waters that do not meet those standards
(the “303(d) list”). 33 U.S.C. § 1313(d). Third, states must establish TMDLs,
subject to EPA approval, for those waters at levels necessary to achieve the
standards. Id. Finally, states are required to submit plans (the “continuing planning
process”), subject to EPA approval, which are designed to achieve the water quality
standards. 33 U.S.C. § 1313(e).
Plaintiffs concede that, while states have primary responsibility for
establishing water quality standards, 303(d) lists, and TMDLs, EPA is authorized to
take action in the event of state inaction or insufficient action. See 33 U.S.C. § 1313
(c) (authorizing EPA to establish water quality standards where it is determined that
the state standards are inconsistent with the CWA); 33 U.S.C. § 1313(d)
(authorizing EPA to establish TMDLs where it is determined that the proposed state
TMDL will not achieve water quality standards); 33 U.S.C. § 1313(e) (authorizing
EPA to review each state’s continuing planning process and disapprove of any state
permit program for any state that does not have an approved continuing planning
process). (Doc. 96 at 40 of 81.) Plaintiffs nevertheless argue that states have
exclusive authority over the final step in this process: implementation of the TMDL
39
allocations. (Id.) Having reviewed the applicable statutory provisions, the court
finds that this argument is overbroad.
TMDL implementation, as is evident from the analysis below, is an
amorphous term. Practically speaking, TMDL “implementation” is divided between
EPA and states. See Sierra Club v. Meiburg, 296 F.3d 1021 (11th Cir. 2002). For
example, point-source discharges are regulated through the federal NPDES
permitting regime, with TMDLs incorporated into the effluent and technology-based
limitations. 33 U.S.C. §§ 1311(b)(1)(C), 1342(d)(3); 40 C.F.R. §
122.44(d)(1)(vii)(B). These permits may be issued either by EPA, or by states with
EPA-approved NPDES programs. However, even where EPA has delegated
permitting authority to the states, EPA retains the right to include additional limits in
NPDES permits when necessary to ensure achievement of water quality. 33 U.S.C.
§ 1312(a), 1342(a). Non-point source regulation, however, is generally left to the
states. 33 U.S.C. § 1329.12 Nevertheless, EPA can influence state implementation
by providing grant money for state non-point source pollution management
programs. 33 U.S.C. §§ 1311(b)(1)(C), 1342(d)(2). Thus, to say that
implementation is left exclusively to the states would be an overstatement.
Plaintiffs’ exclusivity argument is based on CWA Section 303(e),
which requires that states prepare a “continuing planning process” (“CPP”). 33
U.S.C. § 1313(e)(2). These CPPs include TMDLs for pollutants, as well as effluent
limitations and standards, revision procedures, and adequate implementation
12
This section requires states to prepare a non-point source management plan, 33 U.S.C. §
1329(a), and a management program that identifies “best management practice and measures,” 33 U.S.C.
§ 1329(b). “EPA exercises authority over these programs and must approve them.” Meiberg, 296 F.3d
at 1026. Once a management program is approved, EPA may make grants to the states to allow them to
implement the plan. 33 U.S.C. § 1329(h).
40
measures including a schedule for compliance. Id. at 1313(e)(3). Plaintiffs note that
Section 303(e) does not confer backstop authority to EPA or permit EPA to
otherwise take over state implementation plans. (Doc. 96 at 40-42 of 81.) Plaintiffs
contrast this section with Section 303(c) and (d) which, as state above, permit EPA
to establish water quality standards and TMDLs, respectively, wherever state efforts
fall short. (Id. at 41 of 81.)
After reviewing Section 303(e), the court agrees that EPA is not
authorized to establish or otherwise take over TMDL implementation plans.
However, here again, it would go too far to say that EPA has no role in developing
state implementation plans. In fact, EPA is required to review and approve or
disapprove each state’s CPP, and, once its process has been approved, occasionally
review it to ensure that it stays consistent with the Act. 33 U.S.C. § 1313(e)(2).
Thus, here too, EPA has supervisory authority. EPA’s supervisory authority is
consistent with the CWA’s requirement that EPA “ensure that management plans are
developed and implementation is begun by signatories to the Chesapeake Bay
Agreement to achieve and maintain . . . the nutrient goals of the Chesapeake Bay
Agreement . . . [and] the water quality requirements necessary to restore living
resources to the Chesapeake Bay ecosystem.” 33 U.S.C. § 1267(g). Nevertheless,
Plaintiffs are correct that Section 303(e) stops short of giving EPA authority to enact
its own implementation plan where it has determined that the state’s effort has fallen
short. EPA may not, for example, dictate to a state what measures the state must
undertake to reduce pollution from a particular source.
In the end, the parties do not have any real dispute in this regard. Both
Plaintiffs and EPA acknowledge that EPA is authorized under the CWA to take
41
action regarding water quality standards and establishment of TMDLs if the states’
efforts fall short. (See Doc. 96 at 40 of 81; Doc. 100 at 15-16 of 76.) It is further
undisputed, despite Plaintiffs’ claim that implementation is left exclusively to the
states, that EPA’s implementation authority is limited to its authority over NPDES
permitting for point sources, and providing or withholding grant money to
encourage implementation for non-point sources.13 It is logical for states to retain
control over implementation of non-point pollution regulation because non-point
pollution control measures often involve local land use and zoning decisions,
activities which are generally within the well-protected province of state and local
government.
Nevertheless, Plaintiffs claim that EPA unlawfully impinged on the Bay
states’ implementation authority when issuing the Final TMDL. Specifically,
Plaintiffs argue that EPA violated the CWA because (1) the Final TMDL contains
detailed allocations rendering the TMDL tantamount to an implementation plan, (2)
EPA unlawfully imposed “backstop” adjustments, and (3) EPA unlawfully locked-in
those allocations by establishing a federal timeline for implementation and reserving
exclusive authority to revise them. Plaintiffs further claim that EPA violated the
CWA by requiring “reasonable assurances” of the Bay states’ WIPs and by
providing allocations for “upstream” states (the headwater jurisdictions of New
York, Pennsylvania, and West Virginia). The court will address each argument.
13
In Pronsolino II, the court noted that this sort of “carrot-and-stick” approach is central to
attaining acceptable water quality without direct federal regulation of non-point sources of pollution.
291 F.3d at 1127. The court further explained that the advantage of the “intricate scheme is that the
CWA leaves to the states the responsibility of developing plans to achieve water quality standards if the
statutorily-mandated point source controls will not alone suffice, while providing federal funding to aid
in the implementation of the state plans.” Id. at 1128 (citations omitted).
42
2. Detailed Allocations
Plaintiffs initially argue that the high level of detail in the TMDL’s
allocations constrains the states implementation powers. (Doc. 96 at 29 of 81.)
Specifically, Plaintiffs state that “[a]lthough implementation of TMDLs involves
difficult policy choices concerning land use and regulation that are left to the states
under the CWA, EPA’s allocations in the TMDL micro-manage implementation by
dictating the distribution of loadings among numerous source categories and even
individual sources throughout the watershed.” (Id. at 28 of 81.) Plaintiffs note that
EPA established annual and daily WLAs for specific sectors, such as regulated
agriculture, regulated stormwater, and wastewater and for 478 individual permitted
facilities throughout the seven Bay Jurisdictions. (Id. at 28-29 of 81.) At oral
argument, Plaintiffs argued that “the amount of regulation of individual sources, the
extent of those allocations is unprecedent[ed].” (Notes of Transcript (“Tr.”) 36.)14
The court does not find that the level of detail associated with
allocations renders the TMDL a de facto implementation plan. Plaintiffs themselves
seemingly concede this point, stating that “EPA violated the CWA not by
referencing detailed allocations in the TMDL, but by locking those allocations in,
establishing a federal timeline for implementation, and reserving exclusive authority
to revise them.” (Doc. 109 at 23 of 56) (emphasis added.) Nevertheless, at oral
argument, Plaintiffs reiterated their argument that by first dividing the allocations in
WLAs and LAs and then further dividing allocations among various sectors,
14
The court reporter provided the court with a preliminary copy of the transcript of
proceedings on October 4, 2012. A final transcript has not been requested by any party. Any citations
to the transcript refer to the court’s copy of the preliminary transcript and, therefore, may contain
different pagination to a final transcript, in the event that one is requested.
43
including agriculture, stormwater, wastewater, forest, non-tidal atmospheric
deposition, onsite septic and urban, EPA was essentially divesting the states of the
ability to split the allocations as they saw fit, thus restricting the scope of their
implementation powers.15 (Tr. 32; Doc. 96 at 29 of 81.)
As explained below, the court disagrees that allocations, by virtue of
their level of detail, have converted the Final TMDL into an unlawful federal
implementation plan.
a. WLAs and LAs
Plaintiffs argue that the regulation establishing a TMDL as the sum of
WLAs from point sources and LAs from non-point sources, see 40 C.F.R. § 130.2, is
in violation of the CWA because the CWA only authorizes EPA to establish the
total maximum daily load. (Doc. 96 at 21 of 81; Doc. 139 at 6 of 12.) In other
words, Plaintiffs interpret the CWA to authorize EPA to establish a single total load
for a state for a particular pollutant, but not to authorize EPA to allocate that total
load or otherwise determine how the total load is to be achieved. (Doc. 139 at 6 of
12.) Thus, the question, which appears to be an issue of first impression, is whether
EPA exceeded its authority under the CWA by defining a TMDL as “[t]he sum of
15
Here again, Plaintiffs seemingly contradicted themselves by subsequently stating at oral
argument that the detail of the allocation is not at issue in the following exchange:
THE COURT: You don’t object to the detail of the allocations, you object to the
fact that you claim they’re locked in?
MR. SCHWARTZ: That’s correct. If they’re not locked in, then that actually
changes the equation.
(Tr. 7.)
44
the individual WLAs for point sources and LAs for non point sources and natural
background.” 40 C.F.R. § 130.2. The court finds that it did not.
Plaintiffs’ argument in this regard is limited to the reservations of state
authority in 33 U.S.C. § 1251(b) and § 1370. Section 1251(b) provides as follows:
(b) Congressional recognition, preservation, and protection
of primary responsibilities of States. It is the policy of the
Congress to recognize, preserve, and protect the primary
responsibilities and rights of States to prevent, reduce, and
eliminate pollution, to plan the development and use . . . of
land and water resources, and to consult with the
Administrator in the exercise of authority under this Act.
It is the policy of Congress that the States manage the
construction grant program under this Act and implement
the permit programs under sections 402 and 404 of this
Act. It is further the policy of Congress to support and aid
research relating to the prevention, reduction, and
elimination of pollution, and to provide Federal technical
services and financial aid to State and interstate agencies
and municipalities in connection with the prevention,
reduction, and elimination of pollution.
33 U.S.C. § 1251(b). Section 1370 provides:
§ 1370. State Authority
Except as expressly provided in this Act, nothing in this
Act shall (1) preclude or deny the right of any State or
political subdivision thereof or interstate agency to adopt
or enforce (A) any standard or limitation respecting
discharges of pollutants, or (B) any requirement respecting
control or abatement of pollution; except that if an effluent
limitation, or other limitation, effluent standard,
prohibition, pretreatment standard, or standard of
performance is in effect under this Act, such State or
political subdivision or interstate agency may not adopt or
enforce any effluent limitation, or other limitation, effluent
standard, prohibition, pretreatment standard, or standard of
performance which is less stringent than the effluent
limitation, or other limitation, effluent standard,
prohibition, pretreatment standard, or standard of
performance under this Act; or (2) be construed as
impairing or in any manner affecting any right or
jurisdiction of the States with respect to the waters
(including boundary waters) of such States.
45
33 U.S.C. § 1370. Thus, Section 1251 contains broad Congressional policy
statements, and Section 1370 requires states to meet minimum federal effluent
limitations standards. Interestingly, although both sections address state authority,
neither section addresses the issues of TMDLs and whether a TMDL may consist of
both WLAs and LAs. There is nothing in these sections that explicitly prohibits
defining a TMDL as the sum of WLAs and LAs. At the same time, the CWA does
not expressly define a TMDL as a sum of WLAs and LAs, instead describing a
TMDL as the “the total maximum daily load, for those pollutants which the
administrator identified under Section 304(a)(2) [33 U.S.C. § 1314(a)(2)] . . . to
implement the applicable water quality standards.” 33 U.S.C. § 1313(d)(1)(C).16
Clearly, the CWA leaves room for interpretation when defining a
TMDL. EPA, while conceding that there is no statutory mention of WLAs and LAs,
has determined that it is necessary to include both WLAs and LAs in order to
effectuate the overall goal of achieving water quality standards. Specifically, EPA
has determined that:
Although section 303(d)(2) of the Act does not specifically
mention either WLAs or LAs, it is impossible to evaluate
whether a TMDL is technically sound and whether it will
be able to achieve standards without evaluating component
WLAs and LAs and how these loads were calculated.
Thus, it is necessary for EPA to review and approve or
16
CWA § 304(a)(2) simply states that “[t]he Administrator, after consultation with
appropriate Federal and State agencies . . . shall develop and publish information . . . on the
identification of pollutants suitable for maximum daily load measurement correlated with the
achievement of water quality objectives.” 33 U.S.C. § 1314. This is not at issue here because Plaintiffs
are not challenging suitability of regulating nitrogen, phosphorus, or sediments to achieve water quality
objectives. In any event, EPA issued its final identification of pollutants subject to TMDL regulation on
December 28, 1978, wherein EPA did not identify any specific pollutants by name, but simply identified
“all pollutants, under proper technical conditions, as being suitable for the calculation of total maximum
daily loads.” 43 Fed. Reg. 60662.
46
disapprove a TMDL in conjunction with component WLAs
and LAs.
50 Fed. Reg. 1775 (Jan. 11, 1985).
Plaintiffs argue that this interpretation is not entitled to Chevron
deference because there is no ambiguity in the statutory language at issue. (Doc.
139 at 3-6.) In support, Plaintiffs reiterate that the CWA authorizes EPA to
establish the total maximum daily load, but does not otherwise authorize EPA to
allocate that total load among sources, and the statute’s silence on the matter does
not prove ambiguity. (Id.) The court disagrees.
As stated above, under the Chevron standard, where the court finds that
the statute is open or ambiguous – that is, if Congress left a “gap” for the agency to
fill – then this court must uphold the agency’s interpretation so long as it is
“reasonable.” Moreover, that silence does not prove ambiguity is besides the point
because the court in Chevron held that where “the statute is silent or ambiguous
with respect to a specific issue, the question for the court is whether the agency’s
answer is based on a permissible construction of the statute.” 467 U.S. at 843.
The court finds the statutory provisions at issue are precisely the type
that Congress intended to leave to EPA for interpretation. For one, there is no
question that the calculation of a TMDL is a “highly technical, specialized
interstitial matter that Congress does not often decide itself, but delegates to
specialized agencies to decide.” Zuni Pub. Sch. Dist. v. Dep’t of Educ., 550 U.S. 81,
90 (2007). Second, there is no dispute that Congress was silent as to the precise
variables attributable to a TMDL, defining a TMDL only as the load necessary “to
implement the applicable water quality standards.” 33 U.S.C. § 1313(d)(1)(C).
Third, the statutory language itself supports EPA involvement in interpreting the
47
statute. See 33 U.S.C. § 1313(d)(1)(c) (defining a TMDL as “the total maximum
daily load, for those pollutants which the Administrator identified under Section
304(a)(2) [33 U.S.C. § 1314(a)(2)] . . . to implement the applicable water quality
standards.” (emphasis added)); see also 33 U.S.C.§ 1314(a)(2) (“The Administrator .
. . shall develop and publish . . . the identification of pollutants suitable for
maximum daily load measurement . . . .” (emphasis added)).
EPA’s position also finds support in the courts. Although neither the
parties nor the court has been able to identify any case where a plaintiff specifically
challenged EPA’s definition of a TMDL as the sum of WLAs and LAs plus natural
background, courts have nevertheless cited to that provision numerous times without
issue. For example, in Anacostia Riverkeeper v. Jackson, 798 F. Supp. 2d 210
(D.D.C. 2011), the court set aside a TMDL for the Anacostia River that sought to
reduce sediments and total suspended solids (TSS) by 85 percent. The court found
that while the proposed reduction was designed to ensure some of the water quality
standards for the river’s designated uses – namely protection of plant and animal life
– it did not consider the reduction required to protect all the water’s designated uses,
which also included water contact recreation (e.g., swimming), secondary contact
recreation (e.g., boating), and aesthetic enjoyment. The TMDL at issue contained
both WLAs and LAs. The court noted that:
A core requirement of any TMDL is to divide sources of
contamination along the water body by specifying load
allocations, or LAs, to predict inflows of pollution from
particular non-point sources; and to then set[] wasteload
allocations, or WLAs, to allocate daily caps among each
point source of pollution.
Id. at 248-49. Neither the court, nor any party, took issue with this requirement.
48
In Pronsolino I, the district court resolved a challenge to the TMDL for
the Garcia River in northern California. 91 F. Supp. 2d 1337. In that case, the
Pronsolinos, landowners in the Garcia River watershed, challenged the TMDL on
the grounds that the CWA does not authorize EPA to determine a TMDL for rivers
and waters polluted only by non-point sources. The plaintiffs argued that the TMDL
requirements of Section 303(d) were reserved exclusively for point sources, largely
because that section makes no mention of non-point sources.17 The court disagreed,
holding that a TMDL is applicable to “all pollutants,” which includes both point and
non-point sources. The court looked to Ninth Circuit case law, wherein the court
found support for the defendant’s view that TMDLs are applicable to both point and
non-point sources. Id. at 1348-49 (quoting Alaska Ctr. for the Env’t v. Browner, 20
F.3d 981, 985 (9th Cir. 1994) (“Congress and EPA have already determined that
establishing TMDLs is an effective tool for achieving water quality standards in
waters impacted by non-point source pollution.”); and Dioxin/Organochlorine Ctr.
v. Clarke, 57 F.3d 1517, 1520 (9th Cir. 1995) (“A TMDL defines the specific
maximum amount of a pollutant which can be discharged or ‘loaded’ into the waters
at issue from all combined sources. Thus a TMDL represents the cumulative total of
‘load allocations’ which are in turn best estimates of the discrete loading attributed
to nonpoint sources, natural background sources, and individual wasteload
allocations . . . , that is, specific portions of the total load allocated to individual
17
As previously stated, Section 303(d) requires states to identify and compile a list of waters
for which certain “effluent limitations” are not stringent enough to meet applicable water quality
standards (e.g., the aforementioned 303(d) list). 33 U.S.C. § 1313(d)(1). Effluent limitations pertain
only to point sources. This was the basis of the plaintiff’s challenge, which is distinguishable from
Plaintiffs’ challenge in this case, wherein EPA’s ability to allocate as between WLAs and LAs is
challenged on the basis that the CWA only authorizes EPA to set the allocation as a single, cumulative
number.
49
point sources.”)). The court concluded that, “[i]n the face of these statements, it
would be difficult for a district court within the Ninth Circuit to hold that TMDLs
were not required for listed rivers and waters harmed only by nonpoint pollution.”
Id. at 1349.
The court also looked to the legislative history of Section 303(d),
finding that although the legislative history focused on effluent limitations for point
sources, it also recognized that “non-point sources of pollution are a major
contribution to water quality problems.” Id. at 1350. The court interpreted this
statement as Congress’ recognition that non-point pollution would also be required
to meet water quality standards. Id.
Additionally, the court looked to CWA Section 304(a)(2) which states
that TMDLs are obligatory only for those “pollutants” which the Administrator
identifies under Section 304(a) as suitable for calculation. As stated, EPA identified
“all pollutants” as being suitable for TMDL calculation. See supra note 16. The
question, then, was whether sediment, the non-point source at issue, constituted a
“pollutant” even though it was not identified as a pollutant under the Act’s
definition of “pollutant,” which states “[t]he term ‘pollutant’ means dredged spoil,
solid waste, incinerator residue, sewage, garbage, sewage sludge, munitions,
chemical wastes, biological materials, radioactive wastes, heat wrecked or discarded
equipment, rock, sand, cellar dirt[,] and industrial, municipal, and agricultural waste
discharged into water.” 33 U.S.C. § 1362; 40 § C.F.R. 122.2. Nevertheless, the
court found that sediment was a “pollutant,” in part based on Ninth Circuit case law,
and in part based on legislative history, which referenced sediment as a “pollutant.”
91 F. Supp. 2d at 1351 (citing S. Rep. No. 92-414, 92 Cong. 1st Sess. 52
50
(1971)(“sediment, often associated with agricultural activities, is by volume our
major pollutant . . . .” )). The court also looked to other sections in the CWA which
referred to “non-point sources” as “pollutants.” Id. at 352 (citing CWA §§ 105(d),
304(e), and 305(b)(1)(E)). Thus, the court concluded that the “operative language of
the Act . . . expressly treated pollutants as emanating from nonpoint sources” and
thus non-point sources were “pollutants” subject to TMDLs. Id. at 1351-52.
In holding that TMDLs applied to non-point sources of pollution (as
well as point sources), the court stated broadly that this view was in line with the
“comprehensive fabric” of the Act. Id. at 1352. Further, any remaining doubt, the
court stated, should be eliminated by deference to EPA’s reasonable construction of
the statute. Id.
On appeal, the Ninth Circuit Court of Appeals affirmed the district
court’s holding. The court initially found that EPA’s interpretation of the statutory
requirements of Section 303(d) [33 U.S.C. § 1313(d)] as being applicable to waters
that receive water from point sources, non-point sources, or both, is entitled to
Chevron deference. The court reasoned that “Congress entrusted to the EPA the
responsibility of approving or disapproving § 303(d)(1) lists, bestowing upon it the
discretion that comes with such responsibility; the EPA has specialized experience
regarding the CWA which this court lacks; and the agency has consistently
interpreted the provisions at issue.” 291 F.3d at 1134. The court, after analyzing the
language and structure of Section 303(d) and the structure of the statutory scheme as
a whole, found EPA’s interpretation of Section 303(d) to be “entirely reasonable.”
Id. at 1139.
51
Returning to Plaintiffs’ argument here, the court simply can not find
any basis to conclude that EPA’s interpretation of a TMDL as the sum of WLAs and
LAs is unreasonable. EPA has defined a TMDL as the sum of WLAs and LAs (plus
background) since 1985. See 50 Fed. Reg. 1774. Since then, more than 47,000
TMDLs have been completed throughout the United States. (AR0000018; Doc. 110
at 14 of 52 n.2.) Of those, more than 25,000 were issued or approved by EPA and
contained WLAs and LAs. Now, 28 years later, Plaintiffs challenge this
interpretation of TMDLs arguing, for the first time, that “EPA’s interpretation that
the ‘total’ load somehow derives from the ‘allocations’ is unreasonable . . . .” (Doc.
139 at 9 of 12.) Nevertheless, every case that this court has identified that touched
on this issue has said that a TMDL appropriately pertains to both point and nonpoint source allocations. See, e.g., Pronsolino I, 91 F. Supp. 2d 1337; Pronsolino II,
291 F.3d 1123; see also Anacostia Riverkeeper, 798 F. Supp. 2d 210; Meiburg, 296
F.3d at 1025 (“As should be apparent, TMDLs are central to the Clean Water Act’s
water-quality scheme because . . . they tie ‘together point-source and non-point
source pollution issues in a manner that addresses the whole health of the water.’”).
In so concluding, courts have (1) parsed the language of Section 303(d), (2)
analyzed the legislative history of Section 303(d), and (3) analyzed the statutory
scheme as a whole. The court finds this analysis persuasive and exhaustive. As
stated, EPA’s interpretation is entitled to Chevron deference, because the CWA does
not precisely define a TMDL, the definition of which is complex and technical. In
the end, the court finds EPA’s allocation of a TMDL as between WLAs and LAs to
be entirely reasonable, and consistent with Congress’s goals of establishing an “allcompassing program of water pollution regulation” and to establish a
52
“comprehensive long-range policy for the elimination of water pollution.” Arkansas
v. Oklahoma, 503 U.S. 91, 107 n.12 (quoting Milwaukee v. Illinois, 451 U.S. 304,
318 (1981)).
b. Sector and Individual Source Allocations
Plaintiffs’ unlawful implementation argument does not end there. As
stated, Plaintiffs challenge EPA’s allocations not only because they divide the
allocations into WLAs and LAs, but also because EPA further allocated among
various sectors including WLAs and LAs for agriculture, stormwater, wastewater,
forest, non-tidal atmospheric deposition, onsite septic, and urban. (Doc. 96 at 29 of
81.) Plaintiffs further point out that EPA also established annual and daily WLAs
for 478 individual permitted facilities. (Id.) It is this level of detail that Plaintiffs
argue results in unlawful “micro-manage implementation.” (Id. at 28 of 81.)18
There is no denying that the Chesapeake Bay Final TMDL is much
more than a single number for nitrogen, phosphorus, and sediment total load
allocations. Indeed, the TMDL itself is several thousand pages (see AR0000015AR0003790), but many of those pages are dedicated to information regarding source
identification, watershed and land use descriptions, modeling application, and
implementation approaches. The total nitrogen, total phosphorus, and total
suspended sediment allocations themselves can be found in Tables 9-1, 9-2, and 9-3
of the TMDL, respectively, and span 33 pages, each page containing approximately
30 allocations. (AR000029-AR0000348.) Thus, it would be misleading to suggest
18
Here again, this argument is asserted notwithstanding Plaintiffs’ various contradictory
statements that it is not the level of detail in the TMDL that amounts to implementation, but the fact that
the allocations are “locked-in.” See supra note 15.
53
that the TMDL is not highly detailed and complex. However, the court does not
find that this level of detail unlawfully crosses the line into TMDL implementation.
EPA’s regulations instruct that WLAs and LAs should be assigned to
“one of [the water body’s] existing or future point sources of pollution.” 40 C.F.R.
§§ 130.2 (g) & (h). In Anacostia Riverkeeper, the court noted that “[a] core
requirement of any TMDL is to divide sources of contamination along the water
body by specifying load allocations, or LAs, to predict inflows of pollution from
particular non-point sources; and then to set[] wasteload allocations, or WLAs, to
allocate daily caps among each point source of pollution.” 798 F. Supp. 2d at 24849 (emphasis added). Thus, neither the regulations, nor the court in Anacostia
Riverkeeper, anticipated that a TMDL would consist of only a single number. The
regulations provide that a TMDL include allocations to point and, if necessary, nonpoint sources of pollution, rather than be devised at a later stage of post-TMDL
implementation. See id. at 216 (“In addition to setting a maximum daily level of
pollution, EPA regulations require TMDLs to allocate contaminant loads among
point and non-point sources of pollution.”). The court in Anacostia Riverkeeper
went on to explain that:
Total pollutant load established by a TMDL are
incorporated into the NPDES permit system, which is a
key step in the enforcement of those load limits. Absent
specification of WLAs for individual point sources in the
TMDL, therefore, the task of breaking down the total
pollutant load – represented by a single number – into
individual allocations is effectively delegated to NPDES
permit writers. To the extent multiple permit writers
oversee a single water body, such delegation risks either
failure to implement the TMDL through overly-generous
individual allocations that, in the aggregate, exceed total
load limits, or over-enforcement of the TMDL through the
setting of unnecessarily harsh individual allocations
developed out of fear of under-enforcement. To minimize
54
these risks, EPA reasonably determined that specific WLAs
should be developed at the stage when both the State and
Agency are evaluating the health of an entire water body –
i.e., when developing the TMDL – because the designers of
the TMDL can more easily take into account all point
sources and attempt to divvy up acceptable pollution levels
among them.
Id. at 249-50 (emphasis added). In other words, the court concluded that EPA
reasonably assigned allocations to individual point sources. To do otherwise, i.e., to
simply give a number to an entire municipal sewer system, consisting of multiple
sources of point source pollution, and then letting multiple permit writers attempt to
attain that allocation, does not make sense because, as the court pointed out, the
individual permit writers would lack the coordination required to effectively “divvy
up acceptable pollution levels among [the sources].”
The court finds the Anacostia Riverkeeper court’s reasoning persuasive,
and, if that reasoning holds true regarding a municipal sewer system draining into a
single water body (as in Anacostia Riverkeeper), it is all the more true here, where
six states and the District of Columbia all drain into the Chesapeake Bay. To merely
set a number, and then let the states, permit writers, and other groups within each
state “duke it out” would not only be impractical, but would also be inconsistent
with the CWA’s foundational principle, which is that the burdens of eliminating
pollution in the Nation’s water is one to be shared among federal, state, and local
authorities. Id. at 250 (citing Friends of the Earth v. EPA, 346 F. Supp. 2d 182, 203
(D.D.C. 2004)).
In addition, it would be misleading to say that EPA was the sole author
of the TMDL. Rather, the allocations were devised largely by the states in their
WIPs. The process included considerable back-and-forth between EPA and the Bay
55
states. To reiterate, on November 3, 2009, EPA, following meetings with the Bay
states, devised proposed target loads for nitrogen and phosphorus (AR0023289AR0023293), followed by revised target loads as to nitrogen, phosphorus, and
sediment. (AR0000244; AR0012670-AR0012682.) States used these targets to
begin drafting their WIPs. EPA communicated with the Bay states during this
phase, and set deadlines and expectations to guide the drafting process. (See, e.g.,
AR0000255-AR0000256; AR0023294-AR0023301; AR0023289-AR0023293.)
EPA and the Bay states worked together to improve the successive draft WIPs. The
Final TMDL was, in all but three instances wherein EPA substituted backstop
allocations, based on the Bay states’ Phase I WIPs. Thus, as EPA argues in its reply
brief “in all but three cases (the so-called “backstops”), EPA’s TMDL allocations
were informed by the state’s WIPs, not the other way around.” (Doc. 110 at 25 of
52) (emphasis in original).
Plaintiffs argue, however, that the WIP drafting process was not so
cooperative, and that EPA exerted pressure over the states that amounted to
coercion. In support, Plaintiffs offer two slides from EPA presentations, each
containing a single-panel comic. Plaintiffs suggest that these comics portray EPA’s
coercive attitude toward the WIP drafting process. One comic contains the headnote
“It’s a new day for restoring local streams, rivers and the Chesapeake Bay.”
(AR0032986.) The illustration depicts a classroom with students, a teacher, and a
caged tiger in the rear of the classroom. The caption reads “Well, Timmy, it looks
like you’ve just earned yourself 10 minutes in the cage with Mr. Whiskers.” The
second comic depicts two men, one holding a ball and chain attached to his ankle.
(AR0027660.) The caption states “You dropped the ball, You must have known
56
there would be consequences.” Plaintiffs also point to two emails that, in their view,
further illustrate how EPA exerted pressure over the states. In one email, an EPA
employee wrote to two employees of the Virginia Department of Conservation and
Natural Resources, inquiring whether Virginia had “a better understanding of what
needs to be in the WIPs and how EPA will judge adequacy of WIPs.” (Doc. 85-2.)19
In the other email, an EPA employee stated, “It’s important to stress that in the
absence of significant revisions to the discussion topics we’ve had at our meeting
EPA will be forced to retain these backstop allocations in the final TMDL.” (Doc.
85-6.) Finally, Plaintiffs list, without explaining, other threats, including: “(a)
promulgating federal numeric nutrient standards, (b) requiring unreasonable
additional point source reductions, (c) engaging in increased federal enforcement
activity, (d) withholding grant money to states for reasons not intended by Congress,
all because it did not agree with a state’s WIP.” (Doc. 96 at 33 of 81 (citing
AR0024032-33).)
There is no doubt that EPA conveyed its expectations during the WIP
drafting process and further conveyed the possibility of using backstop measures
where the states did not meet EPA’s expectations. EPA expected each Bay state’s
Phase I WIP to: (1) meet the state’s numeric target loads; and (2) provide
“reasonable assurance” that the state’s proposed source and sector allocations would
be met. (See Doc. 100 at 32 of 76.) Further, there is no dispute that, if EPA
determined that the states’ efforts fell short, it would substitute its own backstop
19
These emails were not part of the original administrative record, but were added to the
record when the court granted in part and denied in part Plaintiffs’ motion to complete the administrative
record. (Doc. 92.) The emails were attached to Plaintiffs’ memorandum in support of its motion to
complete the administrative record as exhibits. (See Docs. 85-2 and 85-6.)
57
measures. As Plaintiffs point out, the states did not always agree with EPA’s
backstop allocations. (See Doc. 109 at 30 of 56.) The question, then, is whether this
arrangement amounted to unlawful coercion, or was the result of collaborative,
cooperative federalism. The court finds in favor of the latter.
As several commentators have recognized, cooperative federalism can
be, at times, messy and cumbersome. See Robert L. Fischman & Jaelith Hall-Rivera,
A Lesson for Conservation from Pollution Control Law: Cooperative Federalism for
Recovery Under the Endangered Species Act, 27 Colum. J. Envtl. L. 45, 79 (2002)
(“[D]espite its sometimes messy and redundant framework, cooperative federalism
has proven to be one of the most enduring characteristics of pollution control law
over the past three decades.”); Philip J. Weiser, Federal Common Law, Cooperative
Federalism, and the Enforcement of the Telecom Act, 76 N.Y.U.L. Rev. 1692, 1693
(2001) (“Cooperative federalism regulatory programs, which combine federal and
state authority in creative ways, strike many courts and commentators as a messy
and chaotic means of generating federal law.”) It is unavoidable that states and the
federal government will occasionally disagree. Here, the federal government had
oversight of the states’ WIP drafting efforts. EPA worked with the states to ensure
that the proposed allocations were sufficient to achieve water quality standards. The
states had the first opportunity to determine the allocations necessary to achieve
water quality standards. EPA then reviewed the proposed allocations, approving
some while disapproving others. EPA inserted backstop allocations where
necessary, and remanded the draft WIPs to the states for further analysis and
revision. The states then submitted a final Phase I WIP. In short, EPA incorporated
the states’ allocations in all but three instances.
58
The parties have wildly different interpretations of this process. While
Plaintiffs view the process as “threatening” and “coercive” (see Doc. 96 at 17-19 of
81), EPA describes the process as a “collaborative process that synergistically
developed the TMDL allocations” and as being “both efficient and a model of ‘good
government’ in action” (Doc. 110 at 23 of 52 n.9). Although there may be a fine
line between collaboration and coercion, the court finds this framework to be more
indicative of collaboration. The purpose of the revision process and the insertion of
backstops was to strengthen the WIPs to ensure attainment of water quality
standards through the use of both federal and state resources and expertise. The
court is not convinced that the portions of the record identified by Plaintiffs rise to
the level of coercion. Indeed, the record is replete with numerous communications
that demonstrate discussion, debate, and negotiation between the federal and state
government, not coercion.
Complete unanimity between the states and EPA in resolving all the
complex issues involved here is likely impossible. Disagreements between the
states and the federal government regarding some of the allocations necessary to
achieve water quality standards was to be expected, and the debate and discussions
that ensued were of nature that is required in a cooperative federalism scheme.
Moreover, although Plaintiffs believe that this process was coercive, it is noteworthy
that no state has filed suit challenging the TMDL, let alone alleged that their
participation in the TMDL drafting process was a result of coercion. In short, the
court concludes that the inclusion of sector and individual source allocations is
consistent with the CWA and relevant caselaw. Moreover, the court finds that most
of the individual allocations were provided by the states, not EPA, through the use
59
of CWA’s cooperative federalism scheme. Thus, the record, when viewed as a
whole, does not support a finding that the framework of federal and state interaction
was coercive in nature so as to render the TMDL an unlawful federal
implementation plan.
3. Backstop Adjustments
Plaintiffs contend that EPA unlawfully overrode state decisions on
TMDL implementation by substituting backstop adjustments to the Bay states’
WIPs. (Doc. 96 at 50 of 81.) As stated, EPA adopted the allocations in the state
WIPs in all but three instances, which were as follows: (1) making New York’s
WLA for wastewater sources more stringent (AR0000285-AR0000285); (2) shifting
50 percent of the urban stormwater load that is not currently subject to NPDES
permits from the LA category to the WLA category (AR0000287); and (3) shifting
75 percent of the pollutant loads that West Virginia allocated to animal feeding
operations that are not subject to NPDES permitting from the LA category to the
WLA category and signaling that EPA is prepared to designate any animal feeding
operations as requiring a NPDES permit (AR0000292). Plaintiffs argue that these
measures are binding on the states, and that the CWA does not authorize EPA to
take such actions. (Doc. 109 at 25 of 56.)
The primary flaw in Plaintiffs’ argument is that the CWA contains
several provisions that support EPA’s backstop authority. For instance, under the
broad language of Section 117(g), EPA, in coordination with members of the
Chesapeake Bay Executive Council, is charged with “ensur[ing] that management
plans are developed and implementation is begun by signatories to the Chesapeake
Bay Agreement to achieve and maintain . . . the nutrient goals of the Chesapeake
60
Bay Agreement . . . [and] the water quality requirements necessary to restore living
resources to the Chesapeake Bay ecosystem.” 33 U.S.C. § 1267(g) (emphasis
added). Moreover, Section 303 gives EPA oversight over the waters identified and
the loads established in the TMDL. Specifically, Section 303(d) states:
Each State shall submit to the Administrator from time to
time, . . . for his approval the waters identified and the
loads established [for those waters requiring a TMDL].
The Administrator shall either approve or disapprove such
identification and load . . . . If the Administrator
disapproves such identification and load, he shall . . .
identify such waters in such state and establish loads for
such waters as he determines necessary to implement water
quality standards applicable to such waters and upon such
identification and establishment the State shall incorporate
them into its current plan.
33 U.S.C. § 1313(d)(2).
The court finds that the backstop measures were properly used in
instances where EPA disapproved of the state-submitted allocations, and, consistent
with its responsibilities under Section 303(d), and its broad responsibilities of
ensuring the nutrient goals of the Chesapeake Bay Agreement are achieved under
Section 117(g), substituted its own allocations calculated to achieve applicable
water quality standards. Plaintiffs do not dispute that these backstops were
necessary to achieve water quality standards. Moreover, as explained below, the
court finds that neither the backstop measures nor the WLAs or LAs are binding on
the states. In short, EPA’s actions of reviewing the states’ proposed WIP allocations
and substituting its own allocations where necessary did not violate the CWA.
4. Reasonable Assurances
In determining whether a state’s proposed allocations were adequate,
EPA required “reasonable assurances” from the state that LAs will be achieved and
61
applicable water quality standards will be met.20 (Doc. 96 at 51 of 81; AR0000250.)
Plaintiffs argue, here again, that “the ‘reasonable assurance’ requirement is simply
an attempt by EPA to unlawfully insert itself into TMDL implementation.” (Id.)
Plaintiffs contend that this requirement lacks any basis in the CWA and is therefore
ultra vires. In support, Plaintiffs note that Congress blocked EPA’s previous
attempt to implement revised TMDL regulations that incorporate a “reasonable
assurance” requirement. (Doc. 109 at 33 of 56) (citing 68 Fed. Reg. 13608, 13609
(Mar. 19, 2003).)
The court does not find that the reasonable assurances requirement was
an unlawful exercise of authority by EPA under the CWA. First, unlike Plaintiffs’
other arguments, the court is hard-pressed to see precisely how this argument relates
to implementation. For example, if EPA determines that a state has not met its
burden of providing reasonable assurances, EPA may substitute a backstop
allocation. (See Doc. 100 at 33 of 76.) This substitution gives way to the plausible,
but rejected argument that EPA’s backstop allocations cross the line into
implementation. However, the mere practice of setting a standard upon which the
proposed allocations are judged is not, by itself, implementation. The standard does
not require the states to undertake any particular implementation effort. Rather, the
court finds that the “reasonable assurance” standard was an attempt by EPA to
clarify the basis upon which the proposed allocations are judged. Moreover, as
stated above, Section 303(d) requires that a TMDL be “established at a level
necessary to implement the applicable water quality standards . . . .” 33 U.S.C. §
20
EPA’s reasonable assurance requirement was first published in a 1991 guidance
document, and was later reiterated in a 1997 guidance document. (Doc. 100 at 52 of 76; AR0022979AR0022980).
62
1313(d)(1). EPA’s reasonable assurances requirement appears to be consistent with
this provision.
It bears repeating that a TMDL is an informational document, not an
implementation plan. However, TMDLs provide crucial information for federal,
state, and local actors in furtherance of the cooperative efforts to improve water
quality as envisioned by the CWA. See Anacostia Riverkeeper, 798 F. Supp. 2d at
216-17. Here, where the target water body is drained by a multi-state watershed,
cooperation and coordination are all the more crucial to achieving the statutory goal
of achieving water quality standards. To the extent that TMDLs guide the states’
implementation process,21 it is essential that the allocations contained therein be
reasonably calculated to achieve those goals. This point was recognized by the court
in Anacostia, when it stated that WLAs and LAs can be “developed at the stage
when both the State and the [EPA] are evaluating the health of an entire water body
– i.e., when developing the TMDL, because the designers of the TMDL can more
easily take into account all point sources and attempt to divvy up acceptable
pollution levels among them.” 798 F. Supp. 2d at 250. To this end, the reasonable
assurances requirement helps to inform the TMDL writer of the proper setting of
pollutant allocations so that the TMDL equation is properly budgeted. This is true
because WLAs are determined, in part, on the expectations of pollution reductions
from LAs. If LAs are not fully achieved, water quality standards will not be met.
21
The TMDL/WIP dichotomy makes clear that development of the TMDL and WIPs are
guided by each other. For example, “[d]raft Phase I WIPs were developed and submitted to EPA . . .
[who] used them to support the development of specific allocations in the draft Bay TMDL.”
(AR0000256.) In turn, the Phase II WIPs will “subdivide the allocations provided in the Bay TMDL at
an increasingly finer scale.” (Id.) Lastly, “EPA will consider whether modifications to the Chesapeake
Bay TMDL are necessary and appropriate on the basis of developments or changes in the jurisdictions’
[Phase II and III] WIPs.” (Id.)
63
The WLAs contained in an ineffectual TMDL will themselves be ineffectual and will
therefore be useless as a NPDES permitting guide. On the other hand, where EPA
determines reasonable assurances exists, greater loadings can be allocated to point
sources. (AR000251.) Thus, the requirement of reasonable assurances allows a
TMDL writer to decide how to apportion loadings between point and non-point
sources under the TMDL cap.
In short, the court finds that nothing here runs afoul of the CWA.
Rather, the reasonable assurances requirement is a practical measure that has a basis
in Section 303(d) and 117(g) (requiring EPA to ensure that management plans and
implementation are meeting the Bay’s nutrient goals).22 This requirement does not
violate the TMDL/WIP dichotomy, nor does it unlawfully impinge on the states’
rights to make decisions regarding the implementation of TMDL allocations.
5. Allocations Are Not Binding
Plaintiffs additionally argue that the TMDL is much more than just an
informational tool, but rather it creates unlawfully binding, “locked-in” allocations.
Specifically, Plaintiffs state that “EPA violated the CWA not by referencing detailed
allocations in the TMDL, but by locking those allocations in, establishing a federal
timeline for implementation, and reserving exclusive authority to revise them.”
22
The court is cognizant that broad policy declarations can not be used to justify every
action. See Rodriguez v. United States, 480 U.S. 522, 525-26 (1987) (“[N]o legislation pursues its
purposes at all costs. Deciding what competing values will or will not be sacrificed to the achievement
of a particular objective is the very essence of legislative choice – and it frustrates rather than effectuates
legislative intent simplistically to assume that whatever furthers the statute’s primary objective must be
the law.”). Nevertheless, in the face of no countervailing provisions explicitly or implicitly requiring or
prohibiting a certain action, any action that is consistent with policy declarations and otherwise lawful
should be upheld.
64
(Doc. 109 at 23 of 56.) Having examined each of Plaintiffs’ points, the court
disagrees that the allocations are “locked-in.”
The primary basis for Plaintiffs’ argument rests in 40 C.F.R. §
122.44(d)(1)(vii)(B), which provides, that effluent limits in permits for point sources
be “consistent with the assumptions and requirements of any available wasteload
allocation for the discharge prepared by the State and approved by EPA pursuant to
40 CFR 130.7.” Said another way, NPDES permits must contain effluent limits that
are consistent with applicable WLAs in a TMDL. Plaintiffs further point to the
language of the TMDL itself which states that TMDL allocations may only be
revised with the approval of EPA as further evidence that the allocations are binding
on the Bay Jurisdictions. (Doc. 109 at 21 of 56 (citing AR0000332-AR0000333
(“EPA would consider a request by the jurisdictions to propose such a revision to the
TMDL following appropriate notice and comment. Alternatively, a jurisdiction
could propose to revise a portion(s) of the Bay TMDL that applies within its
boundaries (including, but limited to specific WLAs and LAs) and submit those
revisions to EPA for approval. If EPA approved any such jurisdiction-submitted
revisions, those revisions would replace their respective parts in the EPA-established
Bay TMDL framework.”)).)
In essence, the parties dispute the amount of flexibility the Bay states
retain to adjust allocations as point source permits are issued and non-point source
pollution control measures are implemented. Plaintiffs claim these allocations are
“in ink” and can only be changed by EPA. (Doc. 109 at 24 of 56.) Upon closer
review, the court disagrees that the allocations are so permanent.
65
First, it is evident from the language of the Bay TMDL that a state is not
powerless to effectuate a revision or modification in TMDL allocations. Indeed, a
state is free to propose modifications and submit them to EPA for review. EPA
established this framework, recognizing that:
[N]either the world at large nor the Bay watershed is static.
In a dynamic environment like the Bay watershed, during
the next 15 years change is inevitable. It may be possible
to accommodate some of those changes within the existing
TMDL framework without the need to revise it in whole, or
in part.
(AR0000332.) Thus, the TMDL framework anticipates future modifications which
can originate from either EPA or the states. That EPA gets final approval makes
sense, given that EPA had final approval over the original allocations during the
drafting process outlined and approved above. An alternative scenario, where states
retain the flexibility to change the allocations as they see fit, would render the
TMDL allocations essentially meaningless, and would be inconsistent with CWA
Section 117(g) which requires EPA to ensure that management plans are developed
and implementation is begun in order to achieve and maintain the Bay’s nutrient
goals. 33 U.S.C. § 1267(g).
Second, as recognized by the TMDL, and by the EPA Environmental
Appeals Board, “WLAs are not permit limits per se; rather they still require
translation into permit limits . . . . [W]hile [40 C.F.R. § 122.44(d)(1)(vii)(B)]
require[s] consistency, [it does] not require that permit limitations that will finally be
adopted by a final NPDES permit be identical to any of the WLAs that may be
provided in a TMDL.” (AR0000332; In re City of Moscow, 10 E.A.D. 135, 2001
WL 988721 (July 27, 2001) (emphasis in original).) Accordingly, in some
circumstances, a state may write a NPDES permit limit that is different from the
66
WLA, provided that it is consistent with the operative assumptions underlying the
WLA. (AR0000332.)
Other provisions in the TMDL provide for additional flexibility to the
states. For example, inevitably, new or increased loadings of nutrients or sediments
will occur that are not specifically accounted for in the TMDL. The TMDL
contemplates such an occurrence, and permits these loadings, provided that the
increases are offset by reductions and credits generated by other sources pursuant to
offset programs developed and implemented by the states and subject to periodic
review by EPA. (AR0000329-AR0000331.) Additionally, the TMDL supports the
use of water quality trading programs that permit point and non-point sources to
trade pounds of phosphorus or nitrogen, provided such trading does not result in
exceedances of water quality standards and is otherwise consistent with the CWA
and applicable regulations. (AR0000331.) Thus, the individual sources are free to
trade pollution amounts without the need to revise or adjust the TMDL allocations.
With all these considerations in mind, it is apparent the TMDL allocations are not set
in stone to the extent suggested by Plaintiffs. The court, therefore, rejects Plaintiffs’
assertion that “a state has no flexibility to reallocate pollutant loadings or from
nonpoint to point sources” and that there are no circumstances in which “permit
writers can include less stringent permit limits.” (Doc. 109 at 20 of 56 (emphasis in
original).)
Plaintiffs also argue that the TMDL is illegally binding with respect to
non-point sources, because “EPA can coerce state action through threats to withhold
grant funding.” (Id.) No party disputes that the states retain primary responsibility
for non-point pollution source control, and that EPA may influence state action
67
through the grant program. See supra note 12; 33 U.S.C. § 1329. In Pronsolino II,
the Ninth Circuit, in upholding the Garcia River TMDL, stated that the TMDL did
not invade California’s implementation plan because “California chose both if and
how it would implement the Garcia River TMDL.” 291 F.3d at 1140. The court
explained that “[s]tates must implement TMDLs only to the extent that they seek to
avoid losing federal grant money; there is no pertinent statutory provision otherwise
requiring implementation of § 303 plans or providing for their enforcement.” Id. In
other words, nothing requires states to “uncritically and mechanically” implement
each and every TMDL allocation. Rather, states are free to choose whether or not
they decide to do so, subject only to the risk of losing federal grant money. While
the district court in Pronsolino I noted that such a withholding may seem like
“coercive threats,” especially to states that previously received and relied upon
federal grant money, the framework nevertheless “is not direct federal regulation”
but rather state regulation, albeit “influenced by incentives established by Congress
and the agency charged with protecting the environment.” 91 F. Supp. 2d at 1355.
While recognizing the fine line between incentivizing and coercion, the
court is content that the grant program does not coerce state action. Plaintiffs do not
allege any specific instance of federal coercion, opting instead to challenge the
framework as whole. However, the court concludes, as did the courts in Pronsolino I
and II, that the prospect of losing federal grant money does not make TMDLs
“binding” or invade in the states’ planning process.
Finally, the court must also address Plaintiffs’ argument that the
establishment of a federal timeline violates the CWA. (Doc. 109 at 23 of 56.) As
stated above, the TMDL requires that all pollution control measures be fully
68
implemented by 2025, with at least 60 percent of the actions taken by 2017.
(AR0000016, AR0000021.) Plaintiffs argue that “when EPA locked-in those
allocations and deadlines, it exceeded its CWA authority by invading state
implementation planning.” (Doc. 109 at 26 of 56.) However, Plaintiffs’
characterization of these deadlines as “EPA’s deadlines” is misleading. A closer
look at the record reveals that EPA and the Bay Jurisdictions reached a consensus
regarding the target dates. At a meeting of the PSC on October 1, 2007, the seven
Bay Jurisdictions and EPA reached consensus that, by 2025, all necessary pollution
control measures would be in place. (AR0000056.) Accordingly, the record
supports a conclusion that the timeline at issue was established by the Bay
Partnership, which undermines the position that the timeline was a unilateral federal
dictate from EPA.
In short, the court concludes that, because the 2025 implementation
target was established jointly by the Bay Partnership, and because the states retain
sufficient flexibility to change the allocations, the TMDL does not violate the CWA
by impermissibly “locking-in” the TMDL allocations.
6. Upstream States
Plaintiffs argue that EPA’s authority was limited to establishing the
Final TMDL “to implement the tidal Bay [water quality] standards,” which include
those standards adopted by Maryland, Virginia, Delaware, and the District of
Columbia. (Doc. 96 at 58 of 81 (emphasis added).) However, Plaintiffs maintain
that EPA does not have authority to set allocations for the headwater jurisdictions of
Pennsylvania, New York, and West Virginia. Plaintiffs reason that EPA’s authority
is derivative of the states’ authority under 33 U.S.C. § 1313(d)(2). Because states
69
have no authority to allocate pollutant loadings for water bodies and sources outside
their boundaries when establishing TMDLs under Section 1313(d)(2), EPA similarly
lacks authority to do so. Plaintiffs believe that, if EPA’s interpretation of the CWA
is adopted, an untenable precedent would be established whereby any downstream
state (e.g., Louisiana) could establish a TMDL with allocations to sources in
upstream states (e.g., the other 31 upstream states in the Mississippi River Basin).
The court rejects this argument.
Section 303(d) of the CWA does not expressly address what happens
when a multi-state water body is impaired. See 33 U.S.C. § 1313(d). Indeed, there is
no on-point precedent that establishes what happens when waters that are impaired
overlap state boundaries. Likewise, there is no precedent that establishes precisely
how to reduce water pollution loadings to an interstate water body impaired by
pollutants from seven different states. These scenarios implicate obvious federalism
concerns, some of which have already been addressed by the court. The history of
the Bay TMDL, as outlined above, represents the Partnership’s efforts to resolve
these issues without upsetting the balance of federal-state control established by the
CWA. The question remains, however, whether EPA has the authority to issue
allocations not only to the tidal states, but to the upstream states as well. The court
finds that it does.
Although nothing in the CWA specifically authorizes EPA to take this
holistic, or watershed approach, it is equally true that nothing in the CWA prohibits
such an approach. In the legislative history to the CWA, Congress recognized and
anticipated a watershed-wide approach by stating “the Chesapeake Bay is an
ecosystem that ignores State boundaries” and that implementation “will require a
70
partnership between the Federal Government and the individual states.” (Doc. 100 at
19 of 56) (citing Leg. History of Water Qaulity Act of 1987 at 1473-74 (1988).)
This watershed-wide approach also appears to be consistent, if not
specifically authorized by CWA Section 303(d), which requires TMDLs to be
established for impaired waters “at a level necessary to implement the applicable
water quality standards.” 33 U.S.C. § 1313(d)(1)(C). The accomplishment of this
task, however, raises practical questions pertaining to the equitable distribution of
the burden of reducing pollutant loads. If the court were to adopt Plaintiffs’ stance,
then the tidal states of Maryland, Virginia, and Delaware would be responsible for
reducing their pollution loadings to achieve water quality standards, notwithstanding
significant contributions from upstream states. Pennsylvania, for example, is
responsible for the largest portion of nitrogen loads to the Bay, accounting for 44
percent of the total, and is the second highest contributor of sediment, accounting for
32 percent of the total. (AR0000108-AR00000109.) Thus, to pin the hopes of
attaining the statutorily-mandated goal of achieving water quality standards on the
three tidal states would not only be inequitable, but also impractical and likely
impossible.
Application of the TMDL to upstream states also finds support within
EPA’s regulations. For example, under EPA’s regulations, WLAs and LAs must
reflect the “portion of a receiving water’s loading capacity that is allocated to one of
its existing or future [point or non-point] sources. . . .” 40 C.F.R. §§ 130.2 (g) & (h)
(emphasis added). In support of watershed-wide allocations, EPA interpreted the
reference to “its” point and non-point sources to mean all watershed sources – from
tidal as well as upstream sources – contributing to nutrient loading and
71
sedimentation of the Bay. (See Doc. 100 at 55 of 76.) EPA’s interpretation of its
own regulation is entitled to deference, unless “plainly erroneous or inconsistent
with the regulation.” Auer v. Robbins, 519 U.S. 452, 461 (1997) (citing Robertson v.
Methow Valley Citizens Council, 490 U.S. 332, 359 (1989)). The court finds EPA’s
interpretation not only meets this deferential standard, but is otherwise entirely
reasonable, considering that upstream sources unquestionably contribute pollutants
to the Bay.
As with most issues involved here, there is scant judicial precedent to
guide the court’s analysis. However, in Arkansas v. Oklahoma, the United States
Supreme Court addressed a similar issue in the context of a NPDES permit
challenge. 503 U.S. 91 (1992). In that case, the city of Fayetteville, Arkansas,
applied for a NPDES permit to discharge sewage into a tributary to the Illinois River
at a point upstream from the Oklahoma border. Id. at 95. EPA issued the permit,
conditioned upon the outcome of a study then underway. Id. If that study indicated
that more stringent limitations were necessary to ensure compliance with
Oklahoma’s water quality standards, the permit would need to be modified to
incorporate those limits. Id. Oklahoma challenged the permit, arguing that the
upstream discharge violated Oklahoma’s water quality regulations, which, as they
pertained to the Illinois River, were stringent in light of Oklahoma’s designation of
that river as a “scenic river.” Id. at 95-96. Arkansas, meanwhile, argued that the
CWA does not require an Arkansas point source to comply with Oklahoma
standards. Id. at 97.
The Court began its analysis by recognizing that interstate waters issues,
particularly scenarios wherein a downstream state objects to the introduction of
72
pollutants to a waterway by an upstream state, have been a “font of controversy”
since the founding of our Nation. Id. at 98. The Court further recognized, as this
court does here, that the CWA “anticipates a partnership between the States and the
Federal Government, animated by a shared objective: ‘to restore and maintain the
chemical, physical, and biological integrity of our Nation’s waters.’” Id. at 101
(quoting 33 U.S.C. § 1251(a)). The Court then framed the issues as follows: “First,
does the Act require the EPA, in crafting and issuing a permit to a point source in
one State, to apply the water quality standards of downstream States? Second, even
if the Act does not require as much, does the Agency have the statutory authority to
mandate such compliance?” Id. at 104.
The Court found that EPA’s position – that in issuing the NPDES
permit for a source in Arkansas, EPA was required by the CWA and its regulations
to also comply with Oklahoma’s downstream water quality standards – was a
reasonable exercise of EPA’s statutory discretion. The Court reasoned that, although
the CWA does not require that upstream discharges comply with downstream water
quality standards, the CWA also does not limit EPA’s authority to mandate such
compliance. The Court further reasoned that “[t]he application of state water quality
standards in the interstate context is wholly consistent with the Act’s broad
purposes” of restoring our Nation’s waters. Id. at 106-107 (citing 33 U.S.C. §
1251(a).) Thus, this case supports the proposition that EPA has authority to regulate
upstream pollution sources in order to achieve downstream water quality standards.
This conclusion is also consistent with the Supreme Court’s recognition that “the
Clean Water Act vests the EPA and the States broad authority to develop long-range,
areawide programs to alleviate and eliminate existing pollution.” Id. at 108. This
73
holding is persuasive to the court’s conclusion here that EPA has the authority to set
TMDL allocations for upstream states in order to achieve downstream water quality
standards.
In short, the court endorses the holistic, watershed approach used here.
This approach receives ample support in the CWA, its legislative history, and
Supreme Court precedent. Although Plaintiffs propose alternative methods of
regulating upstream sources,23 the existence of these alternatives does not render
EPA’s present approach unreasonable or unlawful. Rather, the court finds the
approach to be consistent with the CWA, and practical in terms of attaining a full
and fair contribution by all major source sectors and coordinated participation of all
states in the watershed.
7. Conclusion as to implementation arguments
In the end, the court is tasked with determining precisely what is
“implementation” in this context. As stated above, implementation is not an easily
discernable term. Webster’s Dictionary does not provide much guidance, defining
“implementation” as “the act of implementing or the state of being implemented.”
The Oxford Online Dictionary is also of little help, defining implementation as “the
process of putting a decision or plan into effect; execution.” By far the most helpful
definition comes from the Eleventh Circuit’s statement in Meiburg, referring to an
“implementation plan” as “a formal statement of how the level of pollutant can be
brought down or kept under the TMDL.” 296 F.3d at 1030. Clearly, this TMDL is
not an implementation plan because it contains only allocations, and no formal
23
For example, Plaintiffs suggest that rather than establishing a watershed-wide TMDL,
EPA could regulate upstream sources by objecting to inadequate NPDES permits, or establishing
separate upstream water quality standards and TMDLs.
74
statement of how the allocations are to be achieved. Indeed, the TMDL is silent as to
methodology, strategy, and other implementations measures. Rather,
implementation, in this regard, is left correctly to the states.24 Furthermore, the states
retain sufficient flexibility within this framework regarding the TMDL allocations to
support the conclusion that the allocations are not binding. In the end, the states are
still free to choose both if and how they will implement the TMDL allocations,
regardless of the level of detail in those allocations, rendering Plaintiffs’ federalism
concerns unfounded.
The parties argue the import of the numerous consent decrees,
settlement agreements, and memoranda of understanding outlined above. See supra
Section I.E. The parties also argue the import of the states’ consent to EPA’s
establishment of a Bay TMDL on behalf of the states. EPA contends that the states’
consent coupled with the consent decrees, MOUs, settlement agreements, as well as
President Obama’s executive order, provide supplemental sources of authority for
EPA’s issuance of the Final TMDL. (See Doc. 100 at 19 of 76.) Plaintiffs,
meanwhile, contend that neither state consent nor a consent decree or MOU justify
an ultra vires action or supplant the provisions of the CWA. (Doc. 109 at 29 of 56)
(citing Meiburg, 296 F.3d at 1034.) The court agrees that none of these
24
The individual Jurisdictions themselves seemingly recognize this flexibility in their Phase
I WIPs. Virginia’s Phase I WIP, for example, states:
Virginia . . . reserves the right to adjust this [implementation] plan based on new
information . . . . [W]e will continue to work with EPA, stakeholders, and the
public to ensure that our implementation improves water quality in a manner that
is sensible, fair and cost effective as this process unfolds over the next 15 years.
(AR0026675.) Similar reservations appear in other states’ Phase I WIPs. (See, e.g., AR0025149,
AR0026456, AR0026460.)
75
supplemental sources can unilaterally expand Congressionally-bestowed powers.
However, because the court concluded that EPA’s actions in this matter were
authorized under the CWA without considering these supplemental sources of
authority, the court need not opine further on this issue. The court will note,
however, that EPA’s actions of establishing a watershed-wide TMDL appear to be
consistent with the consent decrees, MOUs, and settlement agreements identified
above, as well as the President’s executive order.
Having determined that EPA did not act ultra vires by unlawfully
invading the states’ rights to implementation, and that the Final TMDL was
otherwise consistent with the CWA, the court will now turn to Plaintiffs’ other
arguments raised under the APA.
C.
Alleged Procedural Violations Under the APA
Plaintiffs’ arguments alleging procedural violations are two-fold. First,
Plaintiffs argue that the 45-day public comment period was insufficient because it
did not give the public adequate opportunity to meaningfully participate in the
rulemaking process. (See Doc. 96 at 59 of 81.) Second, Plaintiffs contend that key
information and documentation regarding the models used was unavailable during
the comment period. The court rejects both arguments.
1. The 45-day Public Comment Period Was Adequate
The APA requires EPA to provide notice of its proposed rulemaking
adequate to afford “interested parties a reasonable opportunity to participate in the
rule making through submission of written data, views, or arguments . . . .” 5 U.S.C.
§ 553(c). “The required publication or service of a substantive rule shall be made
not less than 30 days before its effective date . . . .” 33 U.S.C. § 553(d). The
76
purpose of the public comment period is to allow interested individuals the
opportunity to communicate information, concerns, and criticisms to EPA during the
rule-making process. See Conn. Light & Power Co. v. NRC, 673 F.2d 525, 530
(D.C. Cir. 1982). During that period, EPA “must provide sufficient factual detail
and rational for the rule to permit interested parties to comment meaningfully.” Fla.
Power & Light Co. v. United States, 846 F.2d 765, 771 (D.C. Cir. 1988).
The court does not find the 45-day public comment period to be
unreasonable. For one, it exceeds the statutory minimum requirement of a 30-day
period. 33 U.S.C. § 553(d). Thus, EPA did more than was statutorily required by
the APA. Moreover, although the technical complexities of the regulations and
issues raised here might have warranted a longer public comment period, to suggest
that public participation in this process was limited to 45 days belies the record. As
outlined above, efforts to improve the water quality of the Chesapeake Bay date back
more than three decades, and the TMDL drafting process has been ongoing for more
than a decade. See supra Sections I.C. & D. Over that decade, numerous meetings
were held wherein EPA encouraged public participation and accepted public input.
From 2005-2010 alone, 730 CBP committee, team, and stakeholder meetings were
held. (See AR0000422-AR0000454.) As EPA points out, some of the Plaintiffs
participated in the committee meetings and were involved in the drafting process.
(Doc. 100-3 at 37-38; AR0000432.) Nothing in the record suggests that Plaintiffs
could not avail themselves of these opportunities for participation.
Plaintiffs also fail to state specifically how they were harmed by the 45day comment period, other than to claim generally that the comment period was
insufficient to “allow the public to understand – let alone evaluate – how EPA
77
arrived at the allocation scheme in the Draft and Final TMDL.” (Doc. 96 at 59 of
81.) Despite this assertion, the record shows that Plaintiffs submitted 141 comments,
many of which addressed the issues challenged here. (Doc. 100 at 60 of 75;
AR0029851.) To EPA’s credit, a team of EPA specialists reviewed and responded to
the more than 14,000 comments, including the 141 comments submitted by
Plaintiffs. (AR0000341.) The comments were considered in the establishment of
the Final TMDL. (Id.; AR0000016.) EPA also held 18 public meetings and 15
webinars during the comment period. (AR0000020; AR0000339-AR0000340.)
Based on this, as well as the fact that the Final TMDL is the product of an open
process spanning more than a decade, the court is unable to find the 45-day public
comment period unreasonable. Simply put, Plaintiffs either participated, or had the
opportunity to participate, in the drafting process in a meaningful way. Thus, the
court concludes that EPA’s actions were not arbitrary and capricious. It is also
worth noting that a longer comment period would likely violate the terms of the
settlement agreement in Fowler v. EPA, No. 1:09-C-00005-CKK (D.D.C. 2009),
which required that the Final TMDL be established by December 31, 2010. This
conclusion also finds support in relevant caselaw. See, e.g., N. Am. Van Lines, Inc. v
Interstate Commerce Comm’n, 666 F.2d 1087, 1092 (7th Cir. 1981) (finding a 45day comment period to be adequate for new regulations issued by the Interstate
Commerce Commission, noting that “[o]nce an agency has fulfilled its statutory
requirement governing a § 553 rulemaking, its decision may not be subjected to any
additional procedural restraints”); Conn. Light & Power Co., 673 F.2d at 534
(approving a 30-day comment period, notwithstanding “the technical complexity of
the regulations”); Omnipoint Corp. v. Fed. Commc’ns Comm’n, 78 F.3d 620, 629
78
(D.C. Cir. 1996) (approving a 7-day comment period, in part due to a Congressional
mandate to implement the regulations “without administrative or judicial delays”).25
2. EPA Provided Adequate Documentation Regarding
Modeling
Plaintiffs’ second procedural objection is that they were deprived of key
modeling information during the public comment period. Specifically, Plaintiffs
contend that EPA withheld documentation regarding three core models underlying
the Final TMDL: Scenario Builder, the Watershed Model, and the water quality and
sediment transport model (“WQSTM”).
As stated above, during a public comment period, an agency must
provide sufficient factual background to give interested parties an opportunity to
meaningfully comment on the proposed rule. See Fla. Power & Light Co., 846 F.2d
at 771. “When the basis for a proposed rule is a scientific decision, the scientific
material which is believed to support the rule should be exposed to the view of
interested parties for their comment.” United States v. Nova Scotia Food Prods.
Corp., 568 F.2d 240, 252 (2d Cir. 1977); see also Prometheus Radio Project v. Fed.
Commc’ns Comm’n, 373 F.3d 372, 412 (3d Cir. 2004). However, a regulation is not
automatically invalidated even when notice-and-comment errors are committed by
the agency; the party asserting error has the burden of demonstrating prejudice to its
ability to effectively comment on the proposed rule. See 5 U.S.C. § 706 (“In
[reviewing an agency action], the court shall review the whole record or those parts
25
EPA further argues that Plaintiffs’ argument regarding the alleged procedural deficiency
of the 45-day public comment period is deficient because 5 U.S.C. § 553(c) applies to agency
“rulemaking” and the TMDL is an “informal adjudication” as opposed to a “rule.” (See Doc. 100 at 60
of 76.) Because the court is able to reject Plaintiffs’ argument based on a reading of the APA and
applicable caselaw, the court need not decide the “adjudication” versus “rule” distinction in this context.
79
of it cited by a party, and due account shall be taken of the rule of prejudicial
error.”); see also Pers. Watercraft Indus. Ass’n v. Dept. of Commerce, 48 F.3d 540,
544 (D.C. Cir. 1995); AARP v. Equal Emp’t Opportunity Comm’n, 390 F. Supp. 2d
437, 461 (E.D. Pa. 2005). In order to prevail, Plaintiffs must “indicate with
reasonable specificity what portions of the documents it objects to and how it might
have responded if given the opportunity.” See AARP, 390 F. Supp. 2d at 461 (citing
Pers. Watercraft Indus. Ass’n, 48 F.3d at 544). This rule is not without reason. As
is evident from this case, administrative procedures are often lengthy and complex
and to vacate an administrative action due to any procedural error would be extreme.
Accordingly, “[a]s incorporated into the APA, the harmless error rule requires the
party asserting error to demonstrate prejudice from the error.” First Am. Disc. Corp.
v. CFTC, 222 F.3d 1008, 1015 (D.C. Cir. 2000). With these precepts in mind, the
court will analyze Plaintiffs’ arguments as to each of the implicated models.
a. Scenario Builder
Scenario Builder, as described in the Final TMDL, “is a standalone data
pre-processor for the Phase 5.3 Chesapeake Bay Watershed Model.” (AR0000179.)
The model is designed to estimate sediment and nutrient loads from land use
activities and to “facilitate parameterization of those sources for watershed model
scenarios to be run through the Bay Watershed Model.” (Id.) In essence,
information from Scenario Builder is inputted to the Watershed Model, which then
simulates fate, transport, and delivery of those pollutants to the Bay.
Plaintiffs argue that key components of the model were withheld from
the public during the comment period. Specifically, Plaintiffs argue that only a
single document describing how the model was developed was provided. (Doc. 96 at
80
62 of 81) (citing AR0000954-AR0000955; AR0001321; AR0001527-AR0001529).)
Plaintiffs believe that EPA’s failure to disclose key documents, or delayed release of
documentation, limited their ability to fully analyze the technical science under the
TMDL and comment in a meaningful way.
A review of the record reveals that Plaintiffs had access to more than
just a single document regarding Scenario Builder. For example, the draft TMDL,
which was made available on or about September 24, 2010 (the beginning of the
public comment period), contained a “live” link, providing “[a]dditional information
related to Scenario Builder and its application in Bay TMDL development . . .”
(AR0023947.) EPA represents that the link directed users to a September 2010
publication titled Estimates of County-Level Nitrogen and Phosphorus Data for Use
in Modeling Pollutant Reduction, Documentation for Scenario Builder Version 2.2,
which is attached to Plaintiffs’ brief in opposition to EPA’s cross-motion. (Doc. 110
at 34 of 52; AR0023947-AR0023948; see also Doc. 109-15.)26 This 129-page
document describes how Scenario Builder was used, and covered, in detail, the
mathematical functions, sources of data, key tables of data, and summaries of other
data used in Scenario Builder. Furthermore, in response to requests for more
information, EPA provided additional documents regarding Scenario Builder about
one week before the close of the public comment period. That documentation
included the Scenario Builder code, which was provided on October 29, 2010, and
additional information regarding supporting databases was provided on November 15, 2010. (Doc. 100 at 64 of 76; AR0000929.)
26
Although the court is unable to access the link, Plaintiffs have not refuted EPA’s
representation, and the publication will be considered for all intents and purposes.
81
With regard to the delayed disclosure of the additional requested
documents, courts have held that:
[I]t is not per se improper for EPA to add evidence to the
record at the end of or close to the end of the comment
period. EPA may sometimes be able to show that the late
entry did not foreclose an opportunity for ‘meaningful
public comment.’ For example, it might be proper for EPA
to develop new evidence in order to respond to a particular
comment, so long as it gives the commenter an opportunity
to reply to the new evidence.
Small Refiner Lead Phase-Down Task Force v. EPA, 705 F.2d 506, 541 (D.C. Cir.
1983.) Furthermore, “[a]gencies may develop additional information in response to
public comments and rely on that information without starting anew, unless
prejudice is shown.” Pers. Watercraft Indus. Ass’n, 48 F.3d at 544 (internal
quotations and citations omitted).
Here, although the disclosures were made late in the comment period,
Plaintiffs have failed to show, with “reasonable specificity,” or any specificity for
that matter, how they were prejudiced. For example, Plaintiffs fail to suggest what
they might have told EPA if delayed information was disclosed earlier. See id. Nor
do Plaintiffs suggest that the information they did receive was defective. See id.
Even now, having received and reviewed all the disclosed information, Plaintiffs
remain unable to make a reasonably specific showing of prejudice. Instead,
Plaintiffs argue more generally, stating that EPA’s actions are per se improper given
“the critical importance of the [three models].” (Doc. 109 at 44 of 56.) As stated
above, courts have rejected this generalized argument.
Plaintiffs instead hitch their wagon to their belief that, in the Third
Circuit, “a regulated party automatically suffers prejudice when members of the
public . . . are denied access to the complete public record.” (Doc. 109 at 45 of 56
82
(quoting Hanover Potato Prods. v. Shalala, 989 F.2d 123, 130 n.9 (3d Cir. 1993)
(emphasis added)).) However, Plaintiffs’ reliance on Hanover Potato is misplaced.
Initially, the court notes that the quoted language is from a footnote and is clearly
dicta. Plaintiffs have not identified any court that has followed Hanover Potato for
this proposition; nor has the court’s independent research revealed any other case
holding that an incomplete public record automatically results in prejudice.
Moreover, the Hanover Potato case is distinguishable from the case at bar. First, the
underlying facts are distinguishable. The facts underlying that case involved a
request for the administrative record on which the Food and Drug Administration
based its regulatory decision to ban sulfites as applied to “fresh” potatoes. In the
underlying case, the district court granted summary judgment in favor of Hanover,
finding that the Food and Drug Administration (“FDA”) acted arbitrarily and
capriciously by not making the entire record available for public inspection.
Specifically, the FDA admitted that the 83-volume record previously certified “was
not the true and complete administrative record.” 989 F.2d at 126. The court then
certified a new administrative record wherein 63 percent of the new record had never
been disclosed to the public. The district court understandably found prejudice in
light of the incomplete public record. 989 F.2d at 126 n.5. Thus, Hanover Potato
did not involve a TMDL, or any other CWA or environmental regulation. Second,
the case is procedurally distinguishable. On appeal, the Third Circuit affirmed the
district court’s decision. Following the appeal, Hanover moved the district court for
attorney’s fees. The district court denied that motion on the ground that Hanover
was not prejudiced by FDA’s omissions because Hanover did not review the
administrative record during the comment period. In the case relied upon by
83
Plaintiffs here, the appellants were appealing the district court’s denial of attorney’s
fees under the Equal Justice to Act law, 28 U.S.C. § 2412. 989 F.2d 123. Thus, the
Third Circuit was not resolving an administrative review under the APA, but rather
was deciding the merits of an attorney fees petition. In short, the court will follow
the well-settled and well-reasoned rule that a regulation is not automatically
invalidated even where notice-and-comment errors are committed by the agency
unless the party asserting error satisfies its burden of demonstrating prejudice to its
ability to meaningfully comment on the proposed rule. See 5 U.S.C. § 706; see also
AARP, 390 F. Supp. 2d at 461; Pers. Watercraft Indus. Ass’n, 48 F.3d at 544.
The court in Hanover Potato also stated that one of the purposes of the
public comment period “was to give the public the opportunity to participate in the
rule-making process.” 989 F.2d 130 n.9 (citing Conn. Light & Power Co. 673 F.2d
at 530.) This court agrees, and it appears as though this purpose has been fulfilled.
As Plaintiffs point out, Scenario Builder has been in development since 2003. (Doc.
96 at 61 of 81.) By this court’s count, 730 CBP committee, team, and stakeholder
public meetings took place between 2005 and 2010, some of which were attended by
Plaintiffs. (See AR0000422-AR0000454; see also Doc. 108-5.) The Scenario
Builder model was discussed at several of these meetings (see, e.g., AR0000433AR0000434; Docs. 100-11, 100-12, & 100-13) and this process was capped with a
45-day public comment period. Thus, the court finds that the public was given
ample opportunity to participate in the Scenario Builder development process.
Finally, Plaintiffs infer that they have been improperly blamed for not
being able to identify how they would have commented differently had they received
adequate information. (See Doc. 109 at 45 of 56) (“EPA resorts to blaming us for
84
not demonstrating how we would have commented differently . . . .”) This, however,
is precisely the burden that courts have placed on the objecting party. See Pers.
Watercraft Indus. Ass’n, 48 F.3d at 544 (“The party objecting has the burden of
indicating with reasonable specificity . . . how it might have responded if given the
opportunity.” (internal quotations and citations omitted)); Small Refiner Lead PhaseDown Task Force, 705 F.2d at 540-41 (“It is also incumbent upon a petitioner
objecting to the agency’s late submission of documents to indicate with ‘reasonable
specificity’ . . . how it might have responded if given the opportunity.” (internal
quotations and citations omitted)).
In short, the court finds that Plaintiffs have failed to meet their burden
of establishing how they were prejudiced by the alleged failure to disclose key
documents regarding Scenario Builder, and further finds that members of the public,
including Plaintiffs, were provided with a meaningful opportunity to participate in
the TMDL drafting and comment process. Thus, the court concludes that EPA’s
actions related to the disclosure of Scenario Builder documentation were not
arbitrary and capricious.
b. Watershed Model
The Phase 5.3 Community Watershed Model simulates loading and
transport of nitrogen, phosphorus, and sediment from pollutant sources throughout
the Bay watershed and provides loading estimates resulting from various
management scenarios. (AR0000171.) Plaintiffs argue that the documentation
provided for the model was outdated as it pertained to an earlier Phase 5 version that
was created in 2008, two years prior to the re-calibrated Phase 5.3 version. (Doc. 96
at 63 of 81.) Plaintiffs contend that this documentation was of little value, because
85
the public did not have access to information describing how the current model was
developed, calibrated, and applied. (Id.) Plaintiffs further contend that EPA failed
to disclose estimated nutrient transport factors and edge-of-stream nutrient targets
for conservation cropland, information which is important in calibrating watershed
models. (Doc. 96 at 51, 52 of 81.)
EPA concedes that it did not provide the public with Phase 5.3
documentation during the public comment period, but reasons that it could not have
done so because the final application of the Watershed Model was not completed
until after public comments were reviewed and final decisions regarding the Model
were made by the Bay Partnership. (Doc. 100 at 62 of 76.) Consequently, EPA
argues that production of complete documentation was impossible during the public
comment period. (Id.) EPA further concedes that the documentation provided
during the public comment period did not include estimated nutrient transport factors
or edge-of-stream nutrient targets. However, EPA argues that Plaintiffs have made
no efforts to show specifically how this lack of information resulted in prejudice.
Turning to the administrative record, there is no dispute that the Phase
5.3 Watershed Model itself and the supporting information necessary to run the
model (i.e., the model code) were available to the public during the public comment
period. (See Doc. 96 at 63 of 81; Doc. 110 at 31 of 52.) Moreover, there is no
dispute that EPA provided documentation regarding the Phase 5 Watershed Model.
(Doc. 96 at 63 of 81.) Nevertheless, Plaintiffs contend that because EPA did not
provide any information regarding how the current model was developed, calibrated,
and applied, the availability of the model itself was of little value. (Id.)
86
The court’s review of the record reveals that there was ample
information available during the public comment period that explained the Phase 5.3
Watershed Model. For example, the draft TMDL explained the purpose and
importance of the Watershed Model, and its overall purpose in the modeling
framework. (See AR0023922-AR0023962.) More specifically, the draft TMDL
described the Chesapeake Bay watershed water quality network and explained that
“[d]ata from [this network] have been used to develop, calibrate and verify the Phase
5.3 Chesapeake Bay Watershed Model . . . .” (AR0023929.) The draft TMDL also
described the Phase 5.3 Watershed Model as an “open source model” or a
“community model” in which “input data [including precipitation information, point
source discharges, atmospheric deposition, and land use] are all available to the
public,” allowing end users to actually use the model. (AR0023948.) The draft
TMDL further provided a detailed description regarding the development and
calibration of the Phase 5.3 Watershed Model. (AR0023948-AR0023957.) In
addition to that information, the draft TMDL provided links to the model itself, as
well as to information that included further details regarding model inputs and
explaining how certain loading was calculated. (See, e.g., AR0023948,
AR0023951.) It is not apparent from Plaintiffs’ briefs or from oral argument
precisely how this information was deficient, and this court is ill-equipped to
conduct its own technical review of the Watershed Model. In short, the court
concludes that information provided during the public comment, including the model
itself, the code, and supporting documentation was sufficient, and does not support a
finding that EPA’s actions were arbitrary and capricious.
87
Even if the court did find inadequacies, it is not readily apparent
precisely how Plaintiffs were prejudiced by such inadequacies. As stated above,
Plaintiffs have the burden of showing with reasonable specificity how it might have
responded if given the opportunity. Plaintiffs point out that they now have access to
the edge of stream nutrient target information as a result of an “errata” which is part
of the administrative record. (See Doc. 96 at 66 of 81 (citing AR0014689).) Even
so, Plaintiffs still fail to identify how they would have responded differently. Rather,
Plaintiffs once again argue that the failure to have access to the complete record
automatically resulted in prejudice (Doc. 96 at 66-67 of 81 (citing Hanover Potato,
989 F.2d 130 n.9)), an argument that the court already considered and rejected.
c. WQSTM
Plaintiffs assert similar arguments regarding the WQSTM. Plaintiffs
argue that full and complete documentation regarding the WQSTM was made
available only after the close of the public comment period, and that the draft TMDL
acknowledges that the WQSTM was “in preparation” during that time. (Doc. 96 at
67 of 81.) Plaintiffs further argue that the documentation otherwise provided was
outdated, as it refers to an earlier version of the WQSTM. (Id.)
EPA argues, once again, that formal documentation for the WQSTM
could not be completed until after the public comment period when all decisions by
the Partnership were finalized, giving due consideration to the comments. (Doc. 100
at 66 of 76.) Nevertheless, EPA contends that Plaintiffs had access to all necessary
information for public comment. EPA explains that “[t]he WQSTM is composed of
a series of linked and nested models including: hydrodynamic model, estuarine water
column model, sediment transport model, sediment/water interface and flux model,
88
underwater Bay grasses model, bottom sediment dwelling community model, filter
feeder model, phytoplankton model, and zooplankton model.” (Id.) EPA further
explains that only the sediment transport model was altered after 2002, and therefore
any documentation provided was current for all component models except the
sediment transport model. Plaintiffs retort that the WQSTM “fundamentally” and
“dramatically” changed between 2005 and 2010. (Doc. 96 at 67 of 81; Doc. 109 at
41 of 56.) Unfortunately, the citations provided by the parties are of little help to the
court.27
The court must reject Plaintiffs’ arguments for two reasons. First, the
court finds that the public was given ample opportunity to participate in the rule
making process. As with the other models, the WQSTM was in development for
years and was discussed during several public meetings. (See, e.g., AR0000433AR0000434; Doc. 100, Exs. K & L.) This process was capped with a 45-day public
comment period in which numerous documents were provided to the public that
explained how the sediment transport component of the WQSTM was applied in
developing sediment load allocations. (See, e.g., AR0023991; AR0024008AR0024012; AR0024015-AR0024016; AR0024317-AR0024384; AR0024374AR0024384.) As to the adequacy of the documentation, it is readily apparent that
27
Plaintiffs, for example, ask us to “compare” a 373-page document titled The Chesapeake
Bay Eutrophication Model (July 2004) (AR0015530-AR0015903) with a 227-page document titled The
2010 Chesapeake Bay Eutrophication Model: A Report of the US Environmental Protection Agency
Chesapeake Bay Program (December 2010) (AR0016176-AR0016403). Without further explanation, it
is nearly impossible for this court to identify precisely how these documents indicate a fundamental
change in the model. EPA, for its part, cites to the same documentation to support the proposition that
all models, except for the sediment transport model, remained the same. Here again, same problem. See
N.W. Nat’l Ins. Co. v. Baltes, 15 F.3d 660, 662-63 (7th Cir. 1994) (“District judges are not
archaeologists. They need not excavate masses of papers in search of revealing tidbits – not only
because the rules of procedure place the burden on the litigants, but also because their time is scarce.”).
89
the draft TMDL references several documents that provide an explanation of the
WQSTM. (See AR0024131.) Although one of these, a document titled The
Chesapeake Bay Water Quality and Sediment Transport Model (2010), is labeled as
“in preparation” (Id.), other documents, including, but not limited to, documentation
for the 2002 WQSTM were provided. (Id.; AR0015530-AR0015903). The court is
unclear precisely what information Plaintiffs required beyond what was provided.
Second, Plaintiffs again failed to identify, with any specificity, how they might have
responded to the final documentation. The final documentation, which was
published along with the Final TMDL in December 2010 (see AR0016176AR0016403), has now been available for review for nearly three years.
Nevertheless, Plaintiffs still do not point to any specific information and proffer how
they would have responded had it been available during the public comment period.
Rather, they merely point out differences between the preliminary documentation
and the final documentation, which fails to sustain Plaintiffs’ burden of
demonstrating prejudicial error. Accordingly, the administrative record does not
support a finding that EPA acted arbitrarily and capriciously in this regard.
D.
Alleged Modeling Flaws
Plaintiffs’ final arguments relate to EPA’s alleged reliance on flawed
models and flawed data inputs. Plaintiffs raise several arguments contending that the
Final TMDL is arbitrary and capricious on the basis that EPA used models to support
TMDL allocations beyond their predictive capabilities. (See Doc. 96 at 68 of 81.)
The court will first set forth the standard for judicial review of an agency’s use of
analytic modeling before addressing each argument in turn.
90
A model “is an abstraction from and simplification of the real world.”
Small Refiner Lead Phase-Down Task Force, 705 F.2d at 535. “Administrative
agencies have undoubted power to use predictive tools.” Id. Under the arbitrary
and capricious standard set forth in the APA, a court’s “deference to the agency is
greatest when reviewing technical matters within [the agency’s] expertise. In
particular, the choice of scientific data and statistical methodology to be used is best
left to the sound discretion of the [EPA].” Nat’l Ass’n of Metal Finishers v. EPA,
719 F.2d 624, 657 (3d Cir. 1983), rev’d on other grounds sub nom. Chem. Mfrs.
Ass’n v. NRDC, 470 U.S. 116 (1985); see also Kennecott v. EPA, 780 F.2d 445, 449
(4th Cir. 1985) (“Once the agency has been found to follow the prescribed course of
procedure, its choice of scientific data and statistical methodology is entitled to
respect.”). As to data gathering, “EPA typically has wide latitude in determining the
extent of data-gathering necessary to solve a problem” and a court “generally
defer[s] to an agency’s decision to proceed on the basis of imperfect scientific
information, rather than to ‘invest the resources to conduct a perfect study.’” Sierra
Club v. EPA, 167 F.3d 658, 662 (D.C. Cir. 1999) (citing Am. Iron & Steel Inst. v.
EPA, 115 F.3d 979, 1004 (D.C. Cir. 1997) (per curiam)). However, EPA’s
discretion, while broad, is not infinite, and an agency’s choice of model will be
rejected if it “bears no rational relationship to the reality it purports to represent.” Id.
(quoting Columbia Falls Aluminum Co. v. EPA, 139 F.3d 914, 923 (D.C. Cir. 1998)).
As before, the burden is on Plaintiffs to prove that EPA’s actions were arbitrary and
capricious. Forest Guardians, 611 F.3d at 704.
91
1. Flawed Models
Plaintiffs first challenge the models used by EPA as being insufficient
for allocations established at the sub-watershed level. Plaintiffs argument targets
specifically the Watershed Model, contending that, although the Watershed Model
was appropriate for TMDL development on a regional scale, the model was “not
appropriate for development and implementation of TMDLs at the local watershed
scale.” (Doc. 96 at 70 of 71 (citing AR0015016-AR0015017).) Plaintiffs further
argue that EPA acknowledged, but largely ignored, a peer review by the Chesapeake
Bay Program’s Scientific and Technical Advisory Committee (“STAC”) that
concluded that the Watershed Model was insufficient to support management
decisions at the local watershed scale. Id.
EPA defends the use of the Watershed Model, noting that local
allocations were not established solely through EPA’s use of the Watershed Model,
but by the Bay states in their WIPs. (Doc. 100 at 69 of 76.) EPA points out that,
through a collaborative effort, EPA and the Bay states first developed nitrogen,
phosphorus, and sediment allocations at the river basin level – a scale approved by
STAC’s peer review – and then continued to work with the states to develop
proposed allocations at smaller levels using a combination of, inter alia, modeling
results, programmatic implementation capabilities, monitoring data, and land use
information. (Id. at 69-70.) Nevertheless, Plaintiffs maintain that the Bay states
used the Watershed Model when developing their WIP allocations at the local
watershed level notwithstanding STAC’s assessment that the Watershed Model was
not capable of supporting TMDL implementation at that level. (Doc. 109 at 48 of
56.)
92
The record does not support a finding that EPA’s actions were arbitrary
and capricious. First, the record is clear that the individual WLAs and LAs were, in
all but three instances, provided by the Bay states via their respective WIPs, and
were not derived solely from the Watershed Model. See supra Section III.B.2.b.
Second, it is not per se improper for the Bay states to use the Watershed Model to
assist in developing local watershed allocations. In fact, as explained above, the
model was developed as a “community” model, allowing end users, such as
watershed researchers, TMDL model developers, and implementation plan
developers, to use the model in whatever way they deemed proper. (AR0000181.)
The record shows that the Bay states used a variety of sources in developing local
allocations, as recommended by the Partnership in its January 28, 2009 document
titled Response of the Modeling Subcommittee to the Second STAC Review of the
Phase 5 Community Watershed Model. (AR0014964-AR0014974.) In that
document, the Modeling Subcommittee acknowledged that inputs for the Wastershed
Model are at the county level, and stated that “in some cases, the best approach for a
local TMDL exercise would be to use appropriate elements of the Phase 5
[Watershed] Model with augmentation of local-scale land use and monitoring data
when this is available or can be set up.” (AR0014967-AR0014968.) The
Subcommittee continued, stating
[T]he use of [the] Phase 5 [Watershed Model] for local
TMDLs has the merit of the best available information
consistently applied at the local scale. The alternative local
approach is incorporation of additional local data at a more
localized scale into a separate model, but that has the
tradeoff of inconsistent analyses among different local
jurisdictions. Given the tradeoffs of the relative merits of
the two approaches, we believe the local allocations should
be evaluated on a case-by-case basis, and this is what our
State partners are doing.
93
(AR0014968) (emphasis added.) A review of the record confirms that, in setting
local allocations, the Bay states used a variety of data including land use
information, annual data on agricultural conservation practices implemented by
farmers, stormwater best management practices, and current treatment technologies
at wastewater discharge facilities. (See Doc. 110 at 45, 46 of 110 (citing
AR0024982-AR0025421; AR0025422-AR0025524; AR0025525-AR0026300;
AR0026301-AR0026392; AR0026393-AR0026671; AR00266720-AR0026812;
AR0026813-AR0026962; AR0000250-AR0000261; AR0005397-AR0005405; and
AR0012888-AR0012937).)
In light of the record, the court finds no support of Plaintiffs’ argument
that EPA stretched the Watershed Model’s capabilities too far. While it appears that
the Watershed Model is not calibrated to set local allocations, it is also apparent that
the Model was used in conjunction with a number of other local factors that states
also considered in drafting their local allocations, which were decided on a case-bycase basis. Thus, Plaintiffs have failed to meet their burden of showing that there
was no rational relationship between the use of the Watershed Model and the
development of local allocations.
2. Flawed Data
Plaintiffs next argue that EPA’s reliance on flawed data renders the
Final TMDL arbitrary and capricious. Plaintiffs point to several data inputs that they
contend were erroneously used to determine loading estimates. For example,
Plaintiffs argue that EPA improperly estimated that 50 percent of the cultivated
cropland in the Bay watershed employed conventional tillage while the other 50
percent used conservation tillage. (Doc. 96 at 73 of 81; AR0014637.) In support,
94
Plaintiffs point to a U.S. Department of Agriculture Natural Resource Conservation
Service (“NRCS”) draft report dated October 2010, which estimated that 88 percent
of the 4.38 million acres of cultivated cropland in the Bay watershed employed
conservation tillage, while only seven percent used conventional tillage, with the
remaining five percent using a mix of both practices. (AR0032862; Doc. 98-4.)
According to Plaintiffs, correction of this single factor would have significantly
changed the modeled pollutant loadings from these areas. (See Doc. 96 at 74 of 81
(displaying chart showing a more than eight million pound per year difference in
nitrogen loading when the different figures are used).)
Nevertheless, the court must give substantial deference to EPA so long
as EPA provides a rational basis for its use of data. EPA states that it used data
provided by the United States Department of Agriculture (“USDA”)-funded
Conservation Technical Information Center at Purdue University (“USDA data”)
because it was more detailed and comprehensive than the data used by the NRCS.
(Doc. 100 at 71 of 76.) EPA explained that
[t]he conservation tillage data, as well as the agricultural
portion of the CBP Watershed Model, is based in part on
USDA county-level agricultural census data from
thousands of farms from 1982-2007. . . . On the other
hand, the NRCS data on conservation tillage[] was based
on surveys of a sample of approximately 200 farms located
across the Bay watershed, covered only four years, and
provided information only at the scale of four large
watersheds – Susquehanna River, Potomac River, upper
Chesapeake Bay, and lower Chesapeake Bay – for the
entire Chesapeake Bay basin.
(Doc. 100 at 71, 72 of 76 (citing AR0029737; AR0000184-R0000187).)
Accordingly, it is clear that EPA considered the NRCS data, but ultimately rejected it
in favor of USDA data. (Id.; see also Doc. 89-1.) There is also evidence on the
95
record that EPA worked with USDA to discuss the differences in EPA’s and
USDA’s modeling efforts. (AR0029735-AR0029738; AR0029752-AR0029759.)
The court must defer to EPA’s use of data, even if that data is imperfect,
unless the data bears no rational relationship to the reality it purports to represent.
See Sierra Club, 167 F.3d 662. Based on the record highlighted above, it is clear
that EPA had a rational basis for the data used. Accordingly, the court can not
conclude that Plaintiffs have satisfied their burden of showing that EPA’s data
choice was arbitrary and capricious and the court must defer to the Agency’s
expertise. See Nat’l Ass’n of Metal Finishers,719 F.2d at 657; In re Three Mile
Island Alert, Inc., 771 F.2d 720, 737 (3d Cir. 1985) (giving deference to Nuclear
Regulatory Commission’s decision to rely on earlier studies of the health effects of
the TMI-2 accident, stating “we believe this is the kind of scientific determination
over which ‘a reviewing Court must generally be at its most deferential.’” (quoting
Balt. Gas & Elec. Co. v. Natural Res. Def. Council, Inc., 462 U.S. 87, 97 (1983)));
Pa. Dep’t of Envtl. Res. v. EPA, 932 F.2d 269, 272 (3d Cir. 1991) (“On the merits of
EPA’s refusal to consider the updated . . . data, we defer to its expertise.” (citing
EPA v. Nat’l Crushed Stone Ass’n, 449 U.S. 64, 83 (1980))).
Plaintiffs’ argument that EPA used improper assumptions regarding
agricultural runoff suffers the same fate. Plaintiffs contend that EPA improperly
assumed that 15 to 21 percent of all manure at animal feeding operations is left on
impervious surfaces and managed in such a way that it runs off into Bay tributaries.
(Doc. 96 at 76 of 81.) Here again, EPA did consider and address concerns regarding
EPA’s manure management data (see AR0001535-AR001550), and explained that
“EPA’s data reflect reductions in nutrients due to natural processes such as runoff
96
flows from feeding operations to streams, and that only a portion of nitrogen and
phosphorous contained in the 15-21% of manure losses actually enters adjacent
streams.” (Doc. 100 at 73 of 76 (citing AR0016176-AR0016403).) Thus, for the
same reasons cited above, the court will defer to EPA’s use of this data and
concludes that Plaintiffs have failed to demonstrate that EPA’s actions in this regard
were arbitrary and capricious.
IV.
Conclusion
Notwithstanding the expansive administrative record, and the
complexity of the numerous issues implicated herein, the court’s scope of review in
this case is relatively narrow. In accordance with the deferential standards
applicable to a court’s review of an agency’s actions, this court must give EPA’s
interpretation of the CWA and its use of scientific models and data due deference in
light of EPA’s scientific and technical expertise. Plaintiffs are charged with the
heavy burden of showing that the issuance of the Bay TMDL was arbitrary and
capricious, and that EPA’s use of modeling and data bore no rational relationship to
the realities they purport to represent. Having carefully considered Plaintiffs’
arguments, and the applicable portions of the administrative record related thereto,
the court concludes that Plaintiffs have failed to meet this burden. The court further
concludes that the procedures established to ensure public participation in the TMDL
drafting process were sufficient to withstand scrutiny under the APA.
In closing, the court offers the following. The ecological and economic
importance of the Chesapeake Bay is well-documented. As the largest estuary in the
United States, the Chesapeake Bay is essential for the well-being of many living
97
things. (See AR0024989.) The record demonstrates extensive efforts on behalf of
the Bay Partnership to protect this important resource. And yet, nutrient pollution
and sedimentation remain a critical concern. Relevant to the legal challenges sub
judice, the record reveals that the Partnership undertook significant efforts to
preserve the framework of cooperative federalism, as envisioned by the CWA, and
that EPA did not unlawfully infringe on the Bay states’ rights because the CWA is an
“all-compassing” and “comprehensive” statute that envisions a strong federal role for
ensuring pollution reduction. See Pronsolino I, 91 F. Supp. 2d at 1341; 33 U.S.C. §
1267(g). Indeed, considering the numerous complexities of regulating an interstate
water body, EPA’s role is critical to coordinating the Bay Jurisdictions’ efforts to
ensure pollution reduction. In short, the court concludes that the framework
established by the Bay Partnership in developing the Bay TMDL is consistent with
the provisions of the CWA and APA. Accordingly, the court will grant Defendant
EPA’s and Defendant-Intervenor Municipal Associations Group’s cross-motions for
summary judgment and will deny Plaintiffs’ motion for summary judgment. An
appropriate order will issue.
S/Sylvia H. Rambo
United States District Judge
Dated: September 13, 2013.
98
IN THE UNITED STATES DISTRICT COURT
FOR THE MIDDLE DISTRICT OF PENNSYLVANIA
AMERICAN FARM BUREAU
FEDERATION, et al.,
:
:
:
Plaintiffs
:
:
:
v.
:
:
:
UNITED STATES
:
ENVIRONMENTAL PROTECTION :
AGENCY, et al.,
:
:
Defendants
:
CIVIL NO. 1:11-CV-0067
Judge Sylvia H. Rambo
ORDER
In accordance with the accompanying memorandum of law, it is
HEREBY ORDERED as follows:
1.
Plaintiffs’ joint motion for summary judgment (Doc. 95) is
DENIED;
2.
Defendant EPA’s cross-motion for summary judgment (Doc. 99)
is GRANTED;
3.
Defendant-Intervenor Municipal Associations Group’s crossmotion for summary judgment (Doc. 103) is GRANTED;
4.
The clerk of court is directed to enter judgment against Plaintiffs
and in favor of Defendant EPA and Defendant-Intervenors on all
claims.
5.
The clerk of court is directed to CLOSE this case.
S/Sylvia H. Rambo
United States District Judge
Dated: September 13, 2013.
Disclaimer: Justia Dockets & Filings provides public litigation records from the federal appellate and district courts. These filings and docket sheets should not be considered findings of fact or liability, nor do they necessarily reflect the view of Justia.
Why Is My Information Online?