Center for Regulatory Reasonableness, Inc. v. United States Environmental Protection Agency ( 2019 )

UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA
CENTER FOR REGULATORY
REASONABLENESS,
Plaintiff,
v.
UNITED STATES ENVIRONMENTAL
PROTECTION AGENCY, et al.
Civil Case No. 16-1435 (RJL)
Defendants,
and
MINNESOTA POLLUTION CONTROL
AGENCY, and MINNESOTA CENTER
FOR ENVIRONMENTAL ADVOCACY,
\_/V\./V\-/\/\/VV\/V\/\./\/\_/WVVV
Defendant-Intervenors.
§€@
MEM_QRANDUM oPINloN
(March§? , 2019 [Dia. ## 38, 421
The Center for Regulatory Reasonableness (“CRR”) is a coalition of municipal and
industrial entities from around the United States lthat seeks to ensure that regulatory
requirements are scientifically founded, publicly vetted, and cost-conscious Compl. at
‘H ll [Dkt. # l]. On behalf of its Minnesota members, CRR brought this action against the
Environmental Protection Agency and Region V of the Agency (collectively, “EPA” or
“Agency”) challenging EPA’S (l) approval under the Clean Water Act (“CWA” or “Act”)
of certain Water quality criteria promulgated by the Minnesota Pollution Control Agency
(“MPCA”), and (2) refusal to Withdravv approval in response to CRR’s administrative
petition for reconsideration Icz’. at jj 2. CRR alleges that EPA’s approval and denial of
reconsideration were arbitrary and capricious and/or otherwise in violation of the
Administrative Procedure Act (“APA”), 5 U.S.C. § 706(2). Id. at W 88~112. MPCA and
the Minnesota Center for Environmental Advocacy 4(“l\/ICEA”) have intervened to defend
the challenged EPA actions. See [Dl33 U.S.C. § 125 l(a)(Z).
A State’s water quality criteria quantify the amount of pollutants that may be present
in a waterbody while still protecting the designated uses. See 40 C.F.R. § 13 l.l l; JA387-
881. The criteria must be at least “sufficient to protect the designated uses,” 40 C.F.R.
§§ 131.6(c), l3l.1l(a)(1), but States are expressly authorized to adopt standards “more
stringent” than those required by the CWA and accompanying regulations, 33 U.S.C.
§ 1370', 40 C.F.R. § l3l.4(a). Water quality criteria “come in two varieties: specific
numeric limitations on the concentration of a specific pollutant in the water” and “more
general narrative statements applicable to a wide set of pollutants.” Am. Paper ]nst., 996
F.Zd at 349; 40 C.F.R. §§ 131.3(b); id. § 131.11. Criteria “must be based on sound
scientific rationale and must contain sufficient parameters or constituents to protect the
designated use.” 40 C.F.R. § 131.ll(a)(1). States that promulgate numeric criteria “should
. [e]stablish” them “based on:” EPA guidance, EPA guidance modified to refiect
conditions at the site, or “[o]ther scientifically defensible methods.” Id. § 131.1 l(b).
B. EPA Review and Approval Process
While the CWA allocates to the States the primary authority to develop water
quality standards, the States nonetheless must submit all new and revised standards to EPA
l This Memorandum Opinion cites to the portions of the administrative record filed
with the Court in the parties’ joint appendix, the pages of which are Bates stamped “JA_.”
3
for approval or disapproval. 33 U.S.C. § 1313(€)(2)_(3). EPA must base its approval or
disapproval “on the requirements of the Act as described in [40 C.F.R.] §§ 131.5 and 131.6,
and, with respect to Great Lal40 C.F.R. § 131.21(b).
During the time period relevant here, § 131.5(a) required EPA to consider whether the
State: (1) adopted water uses consistent with the CWA’s requirements; (2) adopted criteria
that protect the designated uses; (3) followed its own legal procedures for revising or
adopting standards; (4) for standards that do not include uses specified in CWA
§ 10l(a)(2), based its criteria on appropriate technical and scientific data and analyses ; and
(5) provided a submission that meets the requirements outlined in 40 C.F.R. § 131.6, and
for Great Lakes States, the requirements of 40 C.F.R. part 132.
As to the final § 131.5(a) factor, under § 131.6, a State’s submission must contain:
(a) designated uses consistent with CWA §§ 101(a)(2) and 303(0)(2); (b) the methods used
and analyses conducted to support revisions to existing standards; (c) “water quality criteria
sufficient to protect the designated uses”; (d) an anti-degradation policy consistent with
§ 131.12; (e) certification from the State’s Attorney- General or other State legal authority
that the standards were duly adopted under state law; and (f) where standards do not include
designated uses specified in CWA § 101(a)(2), general information that will aid EPA in
determining the adequacy of the scientific basis, as well as general policies that may affect
application and implementation And for Great Lakes States, to satisfy 40 C.F.R. part 132
“[f]or pollutants listed in Table 5 of” part 132, the State must “[a]pply any methodologies
and procedures acceptable under 40 CFR part 131 when developing water quality criteria.”
40 c.F.R. § 132.4(g)(1).
lf EPA determines that a State’s water quality standards and submission meet the
foregoing requirements, EPA must approve the standards within 60 days after Submission.
33 U.S.C. § 1313(0)(3); 40 C.F.R. § l3l.5(b). Upon approval by EPA, the standards
become effective under the CWA. 40 C.F.R. § 131._21(0).
C. Eutrophication, Numeric Nutrient Criteria, and Stressor-Response Analysis
Since the late 1990s, EPA has required States to adopt water quality criteria
addressing eutrophication in surface waters (e.g., lal40 C.F.R.
§ 123.25(a); Minn. Stat. 115.03, subd. l(a). l\/Iinnesota had previously designated rivers
2 See JA443-44 & Table 3-1.
3 Ecoregions are regions within which differing ecological conditions or other
considerations merit the development of distinct sets of numeric criteria See, e.g., JA30.
6
and streams for aquatic life protection, see, e.g., JA16, 185, 343-44, 387, 532, and the
Criteria “d[id] not include new or revised use designations,” JA531.
ln deriving its River Eutrophication Criteria to protect aquatic life, “MPCA
followed a process consistent with the four-step process set forth in EPA’s Stressor-
response Guidance.” JA534.
Step One. Using field data, empirical observation, and scientific literature, MPCA
developed conceptual models for eutrophication in Minnesota’s rivers and Streams.
JA534~38; see also JA188, 189 Fig. 2. l\/IPCA’s conceptual model was consistent with the
model provided in the EPA Stressor-Response Guidance JA537; see also JA187-91.
According to the model, as the levels of phosphorus and nitrogen rise, the growth of aquatic
plants and algae is stimulated-depicted as increased chlorophyll ot-which impacts the
aquatic environment by, inter olia, reducing dissolved oxygen (“DO”) concentration,
causing increased fluctuations in DO levels (i.e., diel DO flux), and increasing demand for
oxygen, which can be measured as biochemical oxygen demand (“BOD”)§. JA188-90. As
MPCA’s model illustrates, these environmental impacts (and others, like pH fluctuations)
stress or even kill sensitive aquatic plants and animals JA188_90.
Step Two. l\/IPCA identified variables representing concepts from its model that
connect the causal values (like phosphorus) and environmental responses (like loss of
sensitive aquatic life). JA539~45. From the identified variables, l\/IPCA selected those
that are also identified as commonly used in the EP_A Stressor-Response Guidance: total
phosphorus, DO flux, BOD, and other response metrics that gauge the aquatic health of
Minnesota’s rivers. JA539-4l. l\/lPCA conducted statistical and other data analyses to
7
determine that total phosphorus levels are significantly related to increased plant and algae
growth (measured as chlorophyll ot), JA541 & Table 1V.2; see also JA221 Fig. l9a, and
such growth is significantly related to chemical changes in the water (detectable as, inter
alia, increased BOD5 and diel DO flux), which are indicative of eutrophication, JA541 &
Table lV.2; see also JA221~29 & Figs. 26, 27. After additional analysis, l\/lPCA identified
14 biological response measures of aquatic life harm (e.g., percentage of sensitive fish) that
strongly correlated to the response variables (e.g., diel DO flux, BOD§) and determined the
ranges of response values that correlate with harm. JA541_43; see also 232-39. These
findings validated the conceptual model as well as the causal and response variables that
l\/lPCA had identified. JA541-43. ln particular, MPCA confirmed that heightened levels
of phosphorus directly cause increases in diel DO flux and BOD§, which, again, relates
directly to increased plant and algae growth. JA544; see also JA22l-31. However,
because MPCA’s analyses showed that DO flux and BOD5 increases can result from other
factors (e.g., water temperature or shade) in addition to phosphorus, l\/lPCA decided that
DO flux and BOD5 should be incorporated in the water quality criteria as constituent parts
of a multi-variable standard rather than as standalone measures of eutrophication JA544~
45. ln other words, MPCA took the view that a'waterbody should not be considered
impaired unless both its total phosphorus level and one (or more) of diel DO flux or BOD5
(or another response variable) exceed the numeric standards JA544-45.
Step Three. MPCA conducted analyses to determine whether to divide the State
into ecoregions with distinct corresponding numeric nutrient criteria. JA545~48. Relying
in part on the EPA Stressor-Response Guidance, l\/lPCA concluded that three ecoregional
8
criteria distinctions were appropriate due to demonstrated regional variances in baseline
phosphorus concentrations and the levels of phosphorus at which certain aquatic life
suffered harm. JA546-48. Having identified the appropriate ecoregions, MPCA, using
statistical analyses recommended in the EPA Stressor-Response Guidance, assessed the
relationships in each ecoregion between the causal and response variables it had identified
JA548-52. MPCA found that aquatic life is harmed'over a range of phosphorus levels (the
causal variable), and it located the points or “thresholds” at which the maximum amount
of harm was caused per unit change in diel DOlflux, BOD§, and the other response
variables JA551-53. l\/lPCA then selected the appropriate numeric values for each of the
variables lt determined that “a mean or median statistic would be under protective because
the concentration threshold [at which harm is observable] would be exceeded for
approximately half of the biological metrics.” JA548, 555; see also JA243. Accordingly,
MPCA decided to set the numeric criteria at the 25th percentile of the harm thresholds it
had identified for each variable in each ecoregion. JA548, 55 5; see also JA243.
Step Four. Using the methods recommended in the EPA Stressor-Response
Guidance, MPCA validated the relationships it had observed and confirmed JA556_57;
see also JA492*98. MPCA also considered implementation issues, which were addressed
by its development of ecoregion-specific values and its dual-indicator approach, both of
which would help to ensure that the State’s water quality standard would not be exceeded
absent actual phosphorus-driven eutrophication. JA556-57. l\/IPCA documented its data
and analytic work throughout the process JA557. '
ln their final form, the River Eutrophication Criteria consist of numeric limitations
on total phosphorous and each of four measurable responses that are indicative of
eutrophication: diel DO flux, BOD§, pH, and chlorophyll ot. JA533. Under the Criteria, a
l\/linnesota waterbody is deemed impaired if the waterbody exceeds simultaneously the
numeric value for both total phosphorus ana’ one or more of the response variables ]a’.
Thus, the Criteria effectively “consist of four separate dual-pollutant criteria”: (l) total
phosphorous + chlorophyll ag (2) total phosphorous + DO flux; (3) total phosphorous +
BOD§; and (4) total phosphorous + pH. JA533. Additionally, the Criteria are distinguished
by ecoregion, with different numeric values for waterbodies in the Northern, Central, and
Southern parts ofthe State. JA533; see also l\/linn. R. 7050.0222, subp. 2, 2b, 3, 3b, 4, 4b.
E. EPA Approval and Reconsideration of the River Eutrophication Criteria
ln August 2014, l\/lPCA submitted the rule adopting the River Eutrophication
Criteria to EPA for approval. JAS30.4 In additionto the rule, the submission included a
certification that the rule was adopted in accordance with State legal procedures,
documentation of MPCA’s methods and analyses-for developing the rule, and general
policies applicable to l\/linnesota standards JA3~5, 530-32; see generally JA148~77. The
4 MPCA’s 2014 submission was far from its first interaction with EPA regarding
the River Eutrophication Criteria. ln 2009, l\/lPCA provided to EPA Region V a draft
technical support document summarizing its findings and proposed criteria for the causal
and response variables JA400. Region V sent the draft to EPA headquarters which
retained independent subject matter experts for peer review. JA400. Several reviewers
provided positive comments regarding l\/lPCA’s technical work and scientific approach
JA128 (“basic approach is sound”); JA131 (“one of the most comprehensive technical
documents related to nutrient criteria development that l have yet seen”); JA141
(“scientifically defensible criteria”).
10
rule adopting the Criteria did not amend Minnesota’s prior approval of aquatic life as a
designated use for rivers and streams, and it didnot amend the State’s existing anti-
degradation policy. JA531~32.
Over the ensuing months, EPA assessed the l\/lPCA’s process for developing the
River Eutrophication Criteria. EPA evaluated MPCA’s conceptual model, its data and
analyses, and its scientific conclusions pursuant to the Stressor-Response Guidance.
JA535-62. Based on its analysis, EPA determined that MPCA had a “sound scientific
rationale” for its conclusions at each step in the Stressor-Response Guidance process and
for the multi-indicator values it selected, and that the. Criteria protect the designated aquatic
life use for l\/linnesota rivers and streams JA535-62. ln January 2015, EPA approved the
Criteria and released a 61-page document detailing the basis for its decision. J'A528~88.
On December 10, 2015, CRR submitted an administrative petition to EPA
requesting that the Agency reconsider and withdraw its approval JASS9¢99.5 CRR
challenged the MPCA’s use of(l) BOD5 based on alleged testing issues and because BOD5
itself is not directly toxic; (2) diel DO flux because it allegedly is not scientifically accepted
and the selected DO flux value was too stringent; and (3) different numeric criteria for
different ecoregions. JA590. In June 2016, EPA denied CRR’s reconsideration request in
a 13-page response, which explained in detail how the issues raised in CRR’s
5 This was not the first formal challenge to the River Eutrophication Criteria. See
MPCA l\/lem. in Opp’n to CRR Mot. for Summ. J. at 11-12 [Dkt. # 41] (describing
challenges in Minnesota Court of Appeals and l\/linnesota Office of Administrative
Hearings raising similar issues to the present case).
ll
reconsideration petition were addressed in the Agency’s 2015 approval of the Criteria.
JA634-46.
On July 12, 2016, CRR brought this action under the APA challenging EPA’s
January 2015 approval of the Criteria’s total phosphorous, diel DO flux, and BOD5
components and the Agency’s 2016 denial of CRR’s petition for reconsideration [Dkt.
# l].6 EPA filed its answer on September 16, 2016, [Dkt. # 5], and a corrected certified
index to the administrative record on December 13, 2016, [Dkt. # 12]. MCEA and MPCA
moved to intervene as defendants on December 2, 2016 and April 13, 2017, respectively
[Dkt. ## l 1, 22]. And on January 20, 2017, CRR moved to supplement the administrative
record, [Dkt. # 14], which EPA opposed, [Dkt. # 15]. On September 25, 2017, l permitted
MCEA and MPCA to intervene and denied CRR’s challenge to the administrative record.
Minute Grder (Sept. 25, 2017). On July 20, 2018, CRR moved for summary judgment,
CRR Mot. for Summ. J. [Dkt. # 38],7 which MCEA and l\/lPCA separately opposed on
October 17, 2018, [Dkt. ## 40, 41]. On the same day, EPA filed a combined cross-motion
for summary judgment and opposition to CRR’s motion for summary judgment and
supporting brief. EPA Cross-l\/lot. and Opp’n [Dkt. # 42]; EPA Br. in Supp. of Cross-l\/lot.
6 While EPA actions are often directly reviewed in the Circuit Court, “[t]he approval
of water quality standards is initially reviewed by the district courts under the
Administrative Procedure Act.” Cz`ly of All)aqaerqae v. Browner, 97 F.3d 415, 421 n.7
(10th Cir. 1996).
7 Along with its summary judgment motion, CRR filed a 15-page Statement of
Material Undisputed Facts, [Dkt. # 38-3], which it references in its brief. CRR’s statement
of facts is improper under the Local Rules, as this is a “case[ j in which judicial review is
based solely on the administrative record.” LCvR 7(h)(2). l\/ly review is limited to the
record before the EPA at the time it made its challenged decisions
12
and Opp’n [Dkt. # 42-1].8 On December 19, 2018, CRR filed a combined reply in support
of its motion for summary judgment and response in opposition to EPA’s cross-motion for
summary judgment CRR Reply and Resp. Br. [Dkt. # 47]. On l\/Iarch 13, 2019, EPA filed
its reply in support of its cross-motion for summary judgment EPA Reply Br. [Dkt. # 51].
And on l\/larch 27, 2019, the parties filed their joint appendix pursuant to Local Civil Rule
7(n). [Dkt. # 52].
LEGAL STANDARD
At the summary judgment stage, courts typically apply the familiar standard in
Federal Rule of Civil Procedure 56(a), granting judgment “‘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.” But courts assessing agency actions “sit[ ] as an appellate tribunal.” Am.
Bz`oscz`ence, Inc. v. leompson, 269 F.3d 1077, 1083 (D.C. Cir. 2001). In such cases, the
complaint “actually presents no factual allegations but rather only arguments about the
legal conclusion to be drawn about the agency action.” Rempfer v. Sharfstez`n, 583 F.3d
860, 865 (D.C. Cir. 2009) (internal quotation marks `omitted). ln other words, “[t]he entire
case on review is a question of law.” Am. Bioscience, 269 F.3d at 1083 (internal quotation
marks omitted).
Accordingly, while summary judgment remains “the proper mechanism” for
resolving the lawfulness of an agency’s action, Loma Lz`na’a Um`v. Mecl. Ctr. v. Sel)elz'as,
684 F.Supp.2d 42, 52 (D.D.C. 2007), the reviewing court is limited to deciding whether,
8 EPA filed a corrected supporting brief on December 10, 2018. [Dkt. # 45-1].
13
as a matter of law, the evidence in the administrative record supports the agency’s decision,
Ciz‘z`zensfor Responsz'bz`lily & Etl/zz‘cs in Waslz. (“CREW”) v. SEC, 916 F.Supp.Zd 141, 145
(D.D.C. 2013). The agency already has resolved any factual issues in making its decision
See Statterz`rzg Foancl. ofAm. v. Sprz`nger, 498 F.Supp.2d 203, 207 (D.D.C. 2007). As such
“the function of the district court is to determine whether or not as a matter of law the
evidence in the administrative record permitted the agency to make the decision it did.” Ia’.
(quoting Occz'a’enlal Eng’g Co. v. ]NS, 753 F.2d 766, 769~70 (9th Cir. 1985)).
Under the APA’s “default standard” of review, which applies absent a specified
standard in the organic or other governing statute, “'[a] court must set aside agency action
it finds to be ‘arbitrary, capricious, an abuse of discretion, or otherwise not in accordance
with law.”’ Toaras Recora's, [nc. v. DEA, 259 F.3d 731, 736 n. 10 (D.C. Cir. 2001) (quoting
5 U.S.C. § 706(2)(A)); see Nataral Res. Def. Coancl`l, Inc. v. EPA, 16 F.3d 1395, 1400 (4th
Cir. 1993) (“NRDC”) (applying “arbitrary and capricious” standard to EPA approval of
state water quality standards). “The ‘arbitrary and capricious’ standard of review as set
forth in the APA is highly deferential,” and the Court must therefore “presume the validity
of agency action.” Am. Horse Prot. Ass ’n v. Yeatter, 917 F.2d 594, 596 (D.C. Cir. 1990).
The Court must be satisfied only that the agency has “‘examine[d] the relevant data and
articulate[d] a satisfactory explanation for its action including a rational connection
between the facts found and the choice made.”’ Alpharma, Inc. v. Leavz'tt, 460 F.3d l, 6
(D.C. Cir. 2006) (internal quotation marks omitted). “[T]he focal point for judicial review
should be the administrative record already in existence, not some new record made
initially in the reviewing court.” Camp v. Pz`trs, 411 U.S. 138, 142 (1973); see also Ass ’n
14
ofPrz`vate Sector Colls. & Um`vs. v. Dancan, 681 F;3d 427, 441 (D.C. Cir. 2012) (review
is “limited to assessing the record that was actually before the agency”).
ANALYSIS v
l\/ly task in this case “is to scrutinize the EPA’s activity to determine whether the
record reveals that a rational basis exists for its decision” to approve the River
Eutrophication Criteria. NRDC, 16 F.3d at 1400. lScrutiny, of course, does not mean a
substantive reexamination of the EPA’s technical and scientific analyses “[T]he CWA is
a lengthy and complex statute and . . . its mandate and policy often require the evaluation
of sophisticated data.” Ia’. To paraphrase our Circuit Court, l am not by training or
experience a chemist, biologist or statistician, Ethyl Corp. v. EPA, 541 F.2d l, 36 (D.C.
Cir. 1976) (en banc), nor am l equipped to “sit as a scientific body, meticulously reviewing
all data under a laboratory microscope,” NRDC, 16 F.3d at 1400. 1 am a “reviewing court
exercising [my] narrowly defined duty of holding agencies to certain minimum standards
of rationality.” Ethyl Corp., 541 F.2d at 36. l\/Iy role is to ensure that the EPA has “fully
and ably explain[ed] its course of inquiry, its analysis, and its reasoning, and show[n] that
a rational connection exists between its decision-making process and its ultimate decision.”
NRDC, 16 F.3d at 1400; see Envz'ron. Def. Fana', Inc. v. EPA, 489 F.2d 1247, 1252 (D.C.
Cir. 1973) (court is “confronted with a problem in.administrative law, not in chemistry,
biology, medicine, or ecology,” and must review agency’s decision, not “make the same
decision” itself).
As such, l am mindful of the particular deference owed to the EPA’s “scientific
determinations,” as they are “presumed to be the product of agency expertise.” Franks v.
15
Salazar, 816 F.Supp.2d 49, 55 (D.D.C. 2011) (quoting Balt. Gas & Elec. Co. v. Nataral
Res. Def. Coancz`l, [nc., 462 U.S. 87, 103 (1983)) (alteration omitted). That deference is
“great” when an agency decision is “based upon highly complex and technical matters,”
and it has been described as “extreme” when the agency is “evaluating scientific data within
its technical expertise.” West Vz'rgz`m`a v. EPA, 362 F.3d 861, 867~68 (D.C. Cir. 2004)
(internal quotation marks omitted). 1 am not, however, a rubber stamp, and mere “technical
complexity” does not abrogate an agency’s duty “to consider all relevant factors and to
identify the stepping stones to its final decision.” Sierra Clab v. Costle, 657 F.2d 298, 332
(D.C. Cir. 1981); see also NRDC, 16 F.3d at 1401. Judicial deference ends where neglected
evidence and conclusory or inadequate explanationsbegin. See, e.g., Genaz‘ne Parts Co. v.
EPA, 890 F.3d 304, 312 (D.C. Cir. 2018) (“agency cannot ignore evidence that undercuts
its judgment” or “minimize such evidence without adequate explanation”).
With these principles in mind, l turn to CRR’s challenge under APA § 706(2) to
EPA’s approval under the CWA of l\/linnesota’s River Eutrophication Criteria and the
Agency’s denial of CRR’s administrative petition for reconsideration thereof.9
9 The parties do not address our Circuit’s “general rule . . . that an agency’s denial
of a petition for reconsideration is not subject to judicial review” absent some exception
like new evidence or changed circumstances AT&T Corp. v. FCC, 363 F.3d 504, 507-08
(D.C. Cir. 2004). lt may be that an exception applies here, given the additional materials
that CRR submitted to EPA in its reconsideration petition. See CRR l\/lot. for Summ. J. at
22-24 (detailing CRR’s post-approval FOIA requests and consultation with outside
experts). But even if an exception applies, “a court will reverse an agency’s denial of
reconsideration only in the most extraordinary circumstances . . . and only if the agency
has engaged in the clearest abuse of discretion.” AT&T Corp., 363 F.3d at 509 (internal
quotations and citation omitted)). ln the end, the potential for differential standards of
review is immaterial; EPA’s reconsideration denial survives even under the less deferential
arbitrary and capricious standard See Kennecotz‘ C_orp. v. EPA, 684 F.2d 1007, 1012-13
16
A. EPA’s Approval Determination Under 40 C...FR § 131.5(21)
While EPA review and approval are important components of the CWA scheme,
“Congress clearly intended the EPA to have a limited, non-rulemaking role in the
establishment of water quality standards by states.” Pennaco Energy, Inc. v. EPA, 692
F.Supp.2d 1297, 1312 (D. Wyo. 2009) (citing Am. Wz`lcllana’s v. Browner, 260 F.3d 1192,
1194 (10th Cir. 2001)). As such, the States maintain “primary responsibility for
establishing appropriate water quality standards,” and the “EPA sits in a reviewing capacity
of the state-implemented standards with approval and rejection powers only.” NRDC, 16
F.3d at 1399. Here, then, my APA review is not of the decision to adopt the River
Eutrophication Criteria. lnstead, l must review EPA’s review of the Criteria to “determine
whether the record reveals that a rational basis exists for” EPA’s approval, id. at 1401, or
whether, as CRR contends, “EPA rubber stamped the l\/IPCA submission, even in the face
of serious deficiencies,” CRR l\/lot. for Summ. J. at 22.
While “[t]he CWA provides no fixed criterion that clearly delineates when approval
is required,” Sanz'tary Bcl. ofClzarleston v. Wheeler, --- F.3d ---, 2019 WL 1119586, at *4
(4th Cir. Mar. 12, 2019), EPA’s determination is circumscribed: it must “‘be based on the
requirements of the Act as described in [40 C.F.R. §§] 131.5 and 131.6, and, with respect
to GreatLakes States . . . 40 CFR Part 132.” 40 C.F.R. § l31.21(b). lt is clear from the
(D.C. Cir. 1980) (considering EPA’s denial of reconsideration petition together with
substantive review of EPA action); Natz`onal Envz`ron'meatal Dev. Ass ’n ’s Clean Az'r Projecz
v. EPA, 686 F.3d 803, 809_13 (D.C. Cir. 2012) (same).
17
administrative record before me that EPA more than adequately considered the relevant
factors and provided a reasoned basis for its approval decision.
Section 131.5 of the implementing regulations required EPA to consider five factors
in reviewing the River Eutrophication Criteria for approval or disapproval. EPA first had
to assess whether the Criteria adopt designated water uses that are consistent with the
CWA. Ia’. § 131.5(a)(1); see NRDC, 16 F.3d at 1400 (in adopting water quality standards,
“states must first classify the uses for which the water is to be protected”). Minnesota had
previously designated aquatic life uses for rivers and streams, see, e.g., JA16, 343~44, 532,
which is plainly consistent with the Act, see 33 U.S.C. § 1251(a)(2). As the Criteria did
not designate any additional uses for EPA to assess, JA531, § 131.5(a)(1) was not relevant
to EPA’s review. Similarly, § 131.5(a)(4) is confined to standards that “do not include the
uses specified in [CWA § 101(a)(2)],” 40 CFR 131.5(a)(4)--i.e., the protection and
propagation of fish, shellfish, and wildlife, 22 U.S.C. § 1251(a)(2). As the Criteria were
adopted to protect aquatic life uses, EPA was not required to consider § l31.5(a)(4).
EPA was, however, required to consider whether l\/linnesota followed appropriate
legal procedures 40 C.F.R. § 131.5(a)(3). lt did so. l\/linnesota’s submission included a
letter from an l\/IPCA attorney detailing the State’s compliance with applicable state legal
procedures for adopting water quality standards JA3-5, and EPA determined based on the
letter “that l\/linnesota followed its legal procedures” when it adopted the River
Eutrophication Criteria, JA532. Section 131.5(a)(3) therefore was satisfied
Next, under § 131.5(a)(5), EPA had to assess whether l\/linnesota’s submission
complied with the requirements set out in § 131.6 and, if applicable, the requirements in
18
40 C.F.R. part 132, which apply to the Great Lakes States 40 C.F.R. § l3l.5(a)(5); see
supra p. 4. The administrative record makes clear that l\/lPCA’s submission satisfied
§ 131.6. As noted above, l\/linnesota’s preexisting aquatic life use designation comports
with the CWA, and the submission included appropriate legal certification See 40 C.F.R.
§§ 131.6(a), (e). Additionally, Minnesota had previously approved an antidegradation
policy, JA531-32, and CRR does not contend that a new policy was required here. 40
C.F.R. § 131.6(d). MPCA provided substantial technical support documentation in its
submission describing its methods and analyses E.g., JA178»266; see 40 C.F.R.
§ 13 1.6(b). For the reasons stated below in relation to § 131.5(a)(2), l\/lPCA’s water quality
criteria are sufficient to protect the designated uses See 40 C.F.R. § 131.6(0). And finally,
MPCA’s submission included the general policies applicable to Minnesota standards that
may affect the implementation and application of the Criteria. JA148-53, 174-77; see 40
C.F.R. § l3l.6(f). As for 40 C.F.R. part 132, while l\/linnesota is a Great Lakes State,
because the nutrient pollutants at issue here (phosphorus, BOD and DO) are contained in
Table 5 of part 132, MPCA was permitted to “[a]pply any methodologies and procedures
acceptable under 40 C.F.R. part 131.” 40 C.F.R. § 132.4(g)(1).10
Having dispensed with the formal box checking, 1 arrive at the heart of EPA’s
review. Under § 131.5(a)(2), the Agency was required to determine whether the River
Eutrophication Criteria protect the designated water uses To do so, EPA had to satisfy
itself that the River Eutrophication Criteria are “based on sound scientific rationale” and
10 MPCA’s use of the EPA Stressor-Response Guidance process constitutes an
“[o]ther scientifically defensible method” in compliance with § 131.1 1(b)(l)(iii).
19
contain “sufficient parameters or constituents to protect” the designated uses Icl.
§ 131.11(a)(1); NRDC, 16 F.3d at 1402 (EPA’s “duty, under the CWA and the
accompanying regulations, is to ensure that the underlying criteria, which are used as the
basis of a particular state’s water quality standard are scientifically defensible and
protective of the designated uses”). As the Criteria are numerical, they also “should” be
“based on” EPA guidance, EPA guidance modified to reflect conditions at the site, or on
other scientifically defensible methods 40 C.F.R. § 131.11(b); NRDC, 16 F.3d at 1400.
Even a cursory review of the administrative record in this case reveals that EPA
fulfilled its obligation under the CWA to ensure that the River Eutrophication Criteria are
based on sound science and will protect aquatic life. EPA issued a 6l-page explanatory
approval document, see JA528~88, roughly half of which was devoted to detailing EPA’s
careful scientific assessment of the four-step Stressor-Response Guidance process that
l\/lPCA undertook to develop the Criteria, see JA532-62. As EPA explained that process,
“if properly applied, provides a sound scientific rationale for developing nutrient criteria
that are protective of aquatic life designated uses.” JA534 (“proper u’se” of the EPA
Stressor-Response Guidance “can satisfy the requirement of 40 CFR 131.1 1(a) that criteria
must be based on ‘sound scientific rationale’ and ‘protect the designated use”’). For the
reasons summarized below, EPA concluded that l\/IPCA did, in fact, properly apply the
four-step EPA Stressor-Response Guidance.
First, EPA found that MPCA’s conceptual model was “markedly similar to EPA’s
conceptual method,” which EPA developed “based on an extensive review of the relevant
scientific literature.” JA535. lndeed, l\/lPCA’s model was itself based on published
20
scientific research, including many of the same sources relied on by EPA in its Stressor-
Response Guidance JA537. MPCA’s model appropriately observed that (1) increased
nutrient concentrations can adversely impact aquatic systems by altering the chemical
habitat and disrupting the food web; (2) the eutrophication process begins with increases
in sestonic algal and microbes; (3) “chlorophyll ot, daily DO flux, pH, and BOD5 are key
indicators of the initial response” in the eutrophication process ; and (4) fluctuations in plant
and bacterial production can impact aquatic food resources and water chemistry. JA537.
Next, EPA found that MPCA’s “use of exploratory data analysis” was in step with
EPA’s Guidance recommendations JA540. ln brief, l\/lPCA: (1) selected response
variables that reflect the concepts and processes underlying its conceptual model and are
“significantly correlated to ecological response, determined through measures of aquatic
community health”; (2) chose response values that capture the impact on a “diverse array
of aquatic life”; and (3) used appropriate graphical and quantitative techniques to perform
its data analyses JA541. EPA’s approval document includes a detailed and highly
technical assessment of MPCA’s analyses, which supports the foregoing JA541-45.
Third, EPA evaluated MPCA’s additional stressor-response relationship analyses
and its derivation of criteria. JA545. EPA noted that its Stressor-Response Guidance
“recommends the classification of waterbodies by ecoregion to account for other
environmental variables,” JA545, an approach that l\/IPCA followed when it “divided the
state into three nutrient ecoregions” and ‘°conducted the statistical analyses for each
ecoregion separately in order to determine defensible eutrophication indicator values,”
JA546. MPCA’s detailed rationale and analysis for the variations in nutrient values across
21
the State, JA546-47, was “based on sound scientific rationale, JA547. Specifically, it was
reasonable for l\/lPCA to conclude that baseline phosphorus levels would decrease moving
from the fertile prairieland in southwest l\/linnesota to the more barren, rocky soil in the
northeast. J'A548 (noting conformity with relevant scientific literature and EPA
recommendations).
Additionally, EPA conducted a thorough, 9-page assessment of the statistical tools
that MPCA used to determine the statistically-significant thresholds underlying the
selected criteria. JA548-56. The assessment included probing l\/lPCA’s selection of a 25th
percentile threshold concentration JA548, 555. Ultimately, EPA concluded that l\/lPCA’s
approach for deriving the threshold was “based on sound scientific rationale,” that the
chosen numeric criteria “demonstrate significant relationships to designated use protection
(aquatic community health protection),” and the statistical tools used by l\/lPCA-e.g.,
linear and quantile regression, changepoint analyses-wwere acceptable “for deriving
thresholds to determine the concentrations of [total phosphorus], chlorophyll ot, DO flux,
and BOD necessary to ensure that aquatic health is protected.” JA556.
Finally, EPA explained that it was satisfied with MPCA’s validation efforts, which
included (1) assessing the threshold results against independent estimates of the same
indicators ; (2) addressing implementation issues through use of ecoregional derivation and
multiple indicators that both must be exceeded for impairment; and (3) documenting
extensively and transparently its criteria derivation methodology and validation efforts in
its technical support documents and public rulemaking JA556~62.
22
ln sum, because “MPCA followed a process consistent with” the EPA Stressor-
Response Guidance “to derive its eutrophication criteria,” JA534, and because l\/lPCA’s
work at each step stood up to EPA’s thorough and careful assessment under the applicable
regulatory standards, JA535-62, EPA concluded that the Criteria “are based on sound
scientific rationale and protective of l\/linnesota’s aquatic life use designations” “in
accordance with 40 CFR 131.5(a)(2) and 131.11(a),” JA562.ll
ln short, this is a classic example of a case warranting deference to EPA on scientific
and technical matters within its sphere of expertise As the Fourth Circuit observed in a
closely related context, “the technical questions embedded within the EPA’s review of state
water quality standards require the sort of scientific judgment that is the hallmark of agency
discretion.” leeeler, F.3d ---, 2019 WL 11195_86, at *3. The Agency “br[ought] its
own understanding of the” relevant scientific and methodological research and literature
“to bear,” and “[t]here is nothing in this record to suggest that the EPA” or l\/lPCA “used
the [Stressor-Response Guidance process] in a cursory or pretextual fashion.” Ia’. at *4,
"‘9. Moreover, “EPA adequately documented and explained its reasons for approving [the]
water quality standards in” its approval document “and did not merely rubber-stamp [the]
proposed standard.” NRDC, 16 F.3d at 1402. ln sum, EPA exercised the requisite
“independent judgment” under the CWA “as to whether the state’s proposed standards are
‘based on sound scientific rationale’ and are actually capable of meeting the environmental
ll Additionally, as noted above, l\/lPCA’s compliance with the Stressor-Response
Guidance satisfies the “[o]ther scientifically defensible method” standard in 40 C.F.R.
§ 131.11(b)(1)(iii). Cf CRR Reply and Resp. Br. 12-16.
23
ends that have been identified for each body of water.” leeeler, --- F.3d ---, 2019 WL
1119586, at *4 (quoting 40 C.F.R. § 131.11(a)(1)).
Undeterred, CRR nevertheless takes issue with EPA’s performance of its review
obligations under § 131.5(a)(2) and § 131.11(a)(1). ln CRR’s view, MPCA’s use ofBOD5
and diel DO flux as response criteria for determining waterbody impairment is problematic
for a variety of reasons, any one of which renders EPA’s approval decision arbitrary and
capricious CRR l\/lot. for Summ. J. at 31~41. Several of these criticisms are meritless on
their face. F or example, there is no legal support for CRR’s contention that “the
identification of a reliable aquatic life impairment threshold” for nutrient criteria “is
required to set a scientifically defensible standard.” Ia’. at 31 (emphasis added).
Compelling States to pinpoint the threshold value at which impairment results from a given
nutrient indicator-assuming that is even feasible given the demonstrably indirect impact
of nutrient pollutantslz-would suggest that the States must set numeric nutrient criteria
just shy of the point of impairment_i.e., the highest nutrient pollutant level before aquatic
life is harmed CRR and its members might prefer that such a requirement exist, but it does
12 See, e.g., JA635_36. The indirect effects of nutrients on aquatic life disposes of
CRR’s related argument that the State must demonstrate a direct causal relationship
between each nutrient criteria and aquatic life harm.' E.g., CRR Reply and Resp. Br. at 9_
17, 19-24, 26~28. As EPA explained in its reconsideration response, “[u]nlike pollutants
that are directly toxic, nutrients impact aquatic organisms indirectly,” and, therefore, “the
approach used for deriving criteria for toxic pollutants (measuring the exposure that
directly causes an adverse impact . . .) does not work for nutrients.” JA635-36. ln any
case, l\/lPCA did find a causal relationship between phosphorous, eutrophication, and harm
to aquatic lif`e. See JA638 (discussing MPCA’s finding “that water bodies with both
elevated BOD5 [or diel DO flux] and [total phosphorous] tended to exhibit aquatic life
impacts consistent with nutrient pollution as predicted by the conceptual model”).
24
not, and if it did it would conflict with the States’ prerogative under the CWA to adopt
“standards more stringent than required,” 40 C.F.R. § 131.4(a) (emphasis added)-
meaning States are expressly empowered to adopt criteria substantially below any
hypothetical “impairment threshold.” See NRDC, 16 F.3d at 1405 (rejecting plaintiff’ s
unsupported argument that “states have an obligation under the CWA or its accompanying
regulations to adopt a single numeric criterion for dioxin that protects against all
identifiable effects to . . . aquatic life”).l3
CRR argues at length that BOD5 and diel DO flux are inadequate nutrient indicators
because neither independently causes eutrophication and because both are “confounded
parameters,” meaning factors other than eutrophication can cause the values to elevate.
E.g., CRR l\/lot. for Summ. J. at 31-35. CRR’s criticisms of these response indicators do
not persuade me that EPA’s approval was arbitrary or otherwise unlawful, “[a]nd this is
not just because of the deference [1 must] give to EPA when it evaluates ‘scientific data
within its technical expertise.”’ Eclz`sorz Elec. lastitate v. EPA, 391 F.3d 1267, 1270 (D.C.
Cir. 2004) (quoting City of Waukesha v. EPA, 320 F.3d 228, 247 (D.C. Cir. 2003)). lt is
also because CRR’s arguments lack merit As to the former critique, it is abundantly clear
on this record that EPA and MPCA agree that neither BOD5 nor DO flux are per se drivers
of eutrophication For that precise reason, the River Eutrophication Criteria establish a
dual-indicator regime, under which a waterbody is not impaired unless it exceeds the values
13 For the same reasons, there is no merit to CRR’s argument that States are
restricted to establishing criteria that are “necessary” to protect the designated uses See
CRR Reply and Resp. Br. at 10~11. '
25
for both total phosphorus arza’ BOD5 or DO flux _(or another response variable). This
approach is well supported: as EPA summarized in its reconsideration response, l\/lPCA’s
“extensive water quality and biological monitoring and statistical analyses showed that,
when [total phosphorus] and BOD5 [or DO flux] are found in water bodies at elevated
levels, aquatic life are adversely affected.” JA639, 641.
As to CRR’s latter concern about “confoundedness,” EPA explained in detail in its
reconsideration response that l\/lPCA “accounted for site-specific factors (identified in the
letter from CRR as ‘confounding factors’) by adopting combined criteria, with stressor and
response components that both must be exceeded” and “dividing the state into three
different ecoregions for purposes of the eutrophication standards” JA638; see also JA644~
46 (rejecting CRR’s suggestion that a separate .“confounding factors analysis” was
necessary); MPCA l\/lem. in Opp’n to CRR l\/lot. for Summ. J. at 17-20 (describing in detail
the foundation in the administrative record that supports EPA’s conclusion that MPCA
addressed CRR’s “confounding factor” concern).` To the extent that CRR’s position
reflects an alternative school of thought on this issue, “[h]appily, it is not for the judicial
branch to undertake comparative evaluations of conflicting scientific evidence.” Nataral
Res. Def. Coancz`l v. EPA, 824 F.2d 1211, 1216 (D.C. Cir. 1987).14
14 CRR also claims that DO flux cannot be _a “standalone aquatic life impairment
metric,” as the minimum DO amount and not the DO “flux”_which measures the
difference between the maximum and minimum daily DO concentration_is the relevant
DO variable. CRR 1\/lot. for Summ. J. at 37~40. As MPCA explains in its brief, see MPCA
l\/lem. in Opp’n to CRR Mot. for Summ. J. at 20, this criticism was raised during the state
administrative hearing process, and MPCA conducted additional analysis that confirmed
the negative impact of DO flux on aquatic life, see JA324_25. 1 cannot conclude that
EPA’s approval was unreasonable on this basis
26
What remains is CRR’s challenge to MPCA’s selection and use of ecoregional
criteria CRR’s argument is, in sum and substance, that l\/lPCA lacked a rational basis for
setting different numerical nutrient criteria for the causal and response indicators across
each of the three chosen ecoregions CRR l\/lot. for Summ. J. at 41»43. As detailed above,
EPA’s approval document includes an extensive explanation of why it found l\/lPCA’s
selection and use of ecoregional criteria to be founded in sound scientific rationale. JA545_
48. 1 “simply [am] not in a position to second-guess the technical decision by
administrative experts” on this issue. NRDC, 16 F.3d at 1404.
ln practical terms, CRR seeks judicial reconsideration ofthe scientific and technical
underpinnings of the River Eutrophication Criteria__i.e., CRR wants me to afford EPA no
deference in scrutinizing its approval decision and to disregard any deference that EPA
afforded MPCA in conducting its review. Courts, however, have roundly refused to
conduct clefacto a’e novo review under a statutory scheme that contemplates structural
deference See, e.g., NRDC, 16 F.3d at 1401 (rejecting argument “that EPA, as well as the
district court, had a duty under the CWA to assert a more dominant role in the review
process”); Browrier, 97 F.3d at 426 (rejecting argument “that the EPA was required to
reject” water quality standards “unless the EPA had established its own record based on a
sound scientific rationale”). lndeed, “[i]n view ofthe vigorous federalism of the Clean
Water Act, the EPA may not tell a state how to achieve its water quality standards.”
Deferza’ers of Wz`lcllz`fe & Forest Gaaralz`ans v. EPA, No. CIV 02-150, 2004 WL 7337744, at
’1‘17 (D.N.l\/l. l\/lay 21, 2004) (internal quotation marks omitted). And vis-a-vis the courts,
27
“the EPA is not simply a fact-finder or special master . . . , providing us with the data we
need to reach our ownjudgment.” Wheeler, 2019 WL 1119586, at ’1‘10.
Nothing in the administrative record before me suggests that EPA’s decisions to
approve Minnesota’s River Eutrophication Criteria and to deny reconsideration were
arbitrary, capricious, or contrary to law. Quite the opposite, the record here reflects
precisely the sort of cooperative federalism that the CWA envisions
CONCLUSION
Thus, for all of the foregoing reasons, EPA’s motion for summary judgment is
GRANTED, and CRR’s motion for summary judgment is DENIED. An order consistent
with this l\/lemorandum Opinion is separately and contemporaneously issued herewith
RICHA-
United States District Judge
28