United States District Court, D. Rhode Island
PHASE II FINDINGS OF FACT AND CONCLUSIONS OF
LAW
William E. Smith, Chief Judge
I.
Introduction
Dioxin
and other toxic chemical pollution at the Centredale Manor
Restoration Project Superfund Site (“Centredale
Site” or “Site”) in North Providence, Rhode
Island, has led to going on ten years of litigation over
which parties are responsible and what is the appropriate
remedy. The Court divided the litigation into three phases.
(See Eighth Revised Case Management Order, ECF No.
295.)[1]At the close of trial in Phase I, the Court
found Emhart jointly and severally liable for the release of
dioxin at the Site. The Court has now concluded Phase II of
the trial[2] and must provide findings of fact and
conclusions of law addressing the following two issues: (1)
whether the Environmental Protection Agency's
(“EPA”) remedy-selection process was arbitrary,
capricious, or otherwise not in accordance with law; and (2)
whether Emhart had sufficient cause to refuse to comply with
EPA's June 10, 2014 Administrative Order. The necessary
contributions, if any, of third-party defendants will be
addressed in Phase III of the trial. The Court provided a
comprehensive background discussion and procedural history of
this case in its Phase I opinion and need not repeat it here.
(See Phase I Findings of Fact and Conclusions of Law
(“Phase I Findings”), ECF No. 405.) However,
since issuing the Phase I Findings, there have been three new
and important developments in the case relating to the Phase
II litigation. First, the parties have come to an agreement
regarding the United States' past response costs in light
of the Court's findings in Phase I. (See
Stipulation Regarding United States' Past Response Costs,
ECF No. 444.) As such, the Court need not determine the
amount of past response costs.
Second,
the Government moved to limit the scope of judicial review
during Phase II. (Gov't Mot. to Limit Disc. and Scope of
Review, ECF No. 415.) The Government argued that, because
judicial review under CERCLA is strictly limited to the
administrative record, Emhart should be prohibited from
presenting any evidence or arguments not found in the
administrative record. Emhart opposed the Government's
motion, arguing that review under CERCLA is not so limited.
(Emhart Obj., ECF No. 416.) The Court denied the
Government's motion without prejudice, thereby allowing
Emhart to conduct discovery and present the objected to
evidence and arguments at trial. (Order Den. Mot. to Limit
Disc. and Scope of Review 6, ECF No. 421.) However, the Court
has reserved its judgment on the ultimate admissibility of
the challenged evidence and arguments to this point.
(Id. at 6.)
Third,
and lastly, the Government moved to exclude certain portions
of testimony provided by Emhart's expert, Mr. Jeffrey
Loureiro. (Gov't Mot. to Exclude Certain Test. of Jeffrey
Loureiro, ECF No. 447.) The Government argued that
significant portions of Mr. Loureiro's opinions had not
been disclosed in his expert report in violation of Rule
26(a)(2) of the Federal Rules of Civil Procedure. According
to the Government, the admission of Mr. Loureiro's
testimony would be “highly prejudicial . . . given the
United States' inability to explore such opinions through
discovery or to counter them effectively through rebuttal
testimony.” (Id. at 2.) In light of the
Government's concerns, the Court decided to allow the
testimony of Mr. Lourerio but provided the Government an
opportunity to conduct additional discovery and to designate
and utilize rebuttal experts to address the concerns
highlighted in the Government's motion. The Court
therefore denied that motion as moot. (Text Order of
10/07/2016.)
In this
decision, the Court will set forth its findings of fact and
conclusions of law with regards to the question of remedy,
and address along the way the various issues reserved to this
point. See Fed.R.Civ.P. 52(a)(1). The decision
begins with an overview of the CERCLA process by which EPA
may choose a response action, as well as the standards of
judicial review for challenges to a response action. Next the
Court provides specific findings of fact, starting with a
history of EPA action at the Site, and then moving into the
topics which provide the bases for Emhart's challenge.
After each topic the Court provides conclusions of law. II.
Remedy Selection Congress enacted the Environmental Response,
Compensation, and Liability Act (“CERCLA”), 42
U.S.C. § 9601 et seq., “in response to
the serious environmental and health risks posed by
industrial pollution.” United States v.
Bestfoods, 524 U.S. 51, 55 (1998). To address those
risks, CERCLA “grants the President broad power to
command government agencies and private parties to clean up
hazardous waste sites.” Id. (quoting Key
Tronic Corp. v. United States, 511 U.S. 809, 814
(1994)). These cleanup efforts are called “response
actions, ” and may require both “the cleanup or
removal of released hazardous substances” as well as
other “remedial action . . . to prevent or minimize the
release of hazardous substances.” 42 U.S.C. §
9601(23)-(25).
EPA
acts on the President's behalf in choosing the
appropriate response action. See Exec. Order No.
12580(b)(1), (g). In doing so, EPA must work within the
framework provided by CERCLA and the National Contingency
Plan (“NCP”). See 42 U.S.C. 9604(a)(1)
(requiring EPA to act “consistent with the national
contingency plan”); National Contingency Plan, 40
C.F.R. § 300, et seq. In addition, EPA has
established various guidance documents to assist in the
CERCLA process. While these guidance documents are
non-binding on EPA, they do represent EPA's collective
wisdom as to best practices.[3] The steps required by CERCLA and
the NCP, and implemented with the assistance of EPA guidance
documents, are outlined below.
A.
National Priorities List
The
first step in the CERCLA process is placing a site on the
National Priorities List (“NPL”). See 42
U.S.C. § 9605(a)(8)(b); 40 C.F.R. § 300.425. A site
is appropriately included on the NPL if, for example, EPA
determines that a hazardous substance poses “a
significant threat to public health.” Id.
§ 300.425(c)(3)(ii). A notice and comment period is
required before a site is officially placed on the NPL.
Id. § 300.425(d)(5). Once the notice and
comment process is complete, and if EPA determines that NPL
listing is still appropriate, EPA may begin the process of
developing a response action for the site.
B.
Remedial Investigation and Feasibility Study
EPA is
required to conduct a remedial investigation
(“RI”) and feasibility study (“FS”)
before choosing a response action. The end goal of the RI/FS
process is “to assess site conditions and evaluate
alternatives to the extent necessary to select a
remedy.” Id. § 300.430(a)(2). The first
step is the RI, which seeks to “collect data necessary
to adequately characterize the site for the purpose of
developing and evaluating effective remedial
alternatives.” Id. § 300.430(d)(1). EPA
is given significant leeway to develop a RI process specific
to the site. But, at a minimum, EPA must conduct a field
investigation (i.e., collect site-specific data) and create a
baseline risk assessment. Id.
The
data collected during the field investigation includes the
physical characteristics of the site and the hazardous
material present as well as the exposure pathways through
which the hazardous material may affect human health and the
environment. Id. § 300.430(d)(2)(i)-(vii). Part
of this process involves estimating the reasonable maximum
exposure that is likely to occur for both current and
potential future land use at the site. (EPA, Interim
Final Risk Assessment Guidance for Superfund
(“IFRAGS”), vol. I, Emhart579-24.) EPA
then uses this data in the baseline risk assessment to
understand the extent to which hazardous materials pose a
threat to human health and the environment, as well as what
would be “acceptable exposure levels” for the
site going forward. 40 C.F.R. § 300.430(d)(4).
At this
point, EPA transitions from the RI process of collecting data
to the FS process of finding a remedy.[4] “The
national goal of the remedy-selection process is to select
remedies that are protective of human health and the
environment, that maintain protection over time, and that
minimize untreated waste.” Id. §
300.430(a)(1)(i). As an aid to developing potential remedies
that accomplish this overarching goal, EPA first establishes
what are labeled Preliminary Remediation Goals
(“PRGs”) targeting “acceptable exposure
levels that are protective of human health and the
environment.” Id. § 300.430(e)(2)(i).
These PRGs must, among other things, comply with federal and
state environmental laws[5] and limit the lifetime cancer risk from
carcinogenic exposure. Id. §
300.430(e)(2)(i)(A)(1)-(2). EPA then develops a range of
response alternatives that may achieve those PRGs.
In
order to choose among the response options for the site, the
various alternatives are initially screened using three
criteria: (1) effectiveness; (2) implementability; and (3)
cost. Id. § 300.430(e)(7)(i)-(iii). At this
stage, alternatives that provide “significantly less
effectiveness” or are “technically or
administratively infeasible” may be eliminated.
Id. § 300.430(e)(7)(i)-(ii). Additionally,
“[c]osts that are grossly excessive compared to the
overall effectiveness of alternatives may be considered as
one of several factors used to eliminate alternatives.”
Id. § 300.430(e)(7)(iii).
Alternatives
that survive the initial culling proceed to the
“detailed analysis of alternatives.” Id.
§ 300.430(e)(9). This requires “an assessment of
individual alternatives against each of nine evaluation
criteria and a comparative analysis that focuses upon the
relative performance of each alternative against those
criteria.” Id. § 300.430(e)(9)(ii). The
nine criteria used by EPA to compare alternatives are: (1)
overall protection of human health and the environment; (2)
compliance with federal and state environmental laws (i.e.,
“ARARs”[6]); (3) long-term effectiveness and
permanence; (4) reduction of toxicity, mobility, or volume
through treatment; (5) short-term effectiveness; (6)
implementability; (7) cost; (8) state acceptance; and (9)
community acceptance. See id. §
300.430(e)(9)(iii)(A)-(I). Once sufficient information has
been gathered such that EPA can compare the alternatives
based on the nine evaluation criteria, EPA may proceed with
remedy selection.
C.
Remedy Selection
During
remedy selection EPA places the nine evaluation criteria into
three categories. The first category, labeled
“threshold criteria, ” are the criteria
“that each alternative must meet in order to be
eligible for selection.” Id. §
300.430(f)(1)(i)(A). The two threshold criteria are overall
protection of human health and the environment and compliance
with ARARs (criteria (1) and (2), above).[7]
Once
EPA has screened out alternatives that do not meet the
threshold criteria, EPA then compares the remaining
alternatives based on the second category of criteria,
labeled “primary balancing criteria.”
Id. § 300.430(f)(1)(i)(B). These include
long-term effectiveness and permanence, reduction of
toxicity, mobility, or volume through treatment, short-term
effectiveness, implementability, and cost (criteria (3)-(7),
above). Id. As the label suggests, these
“primary balancing criteria” are balanced against
one-another. In doing so, the NCP provides certain
preferences. For instance, a remedy must be
“cost-effective” in that “its costs are
proportional to its overall effectiveness.”
Id. § 300.430(f)(1)(ii)(D). In addition,
“balancing shall emphasize long-term effectiveness and
reduction of toxicity, mobility, or volume through treatment,
” thereby focusing on “permanent solutions . . .
to the maximum extent practicable.” Id. §
300.430(f)(1)(ii)(E). However, in the end, the NCP does not
dictate exactly how the primary balancing criteria should be
weighed. Instead, the NCP provides EPA with considerable
discretion to select a remedy that “reflect[s] the
scope and purpose of the actions being undertaken and how the
action relates to long-term, comprehensive response at the
site.” Id. § 300.430(f)(a).
The
last category EPA must consider is the “modifying
criteria.” Id. § 300.430(f)(1)(i)(C). The
modifying criteria include state and community acceptance
(criteria (8) and (9), above). Because the state and
community provide formal input during the notice and comment
period for the proposed plan, state and community acceptance
is typically only considered, at this stage, “to the
extent that information is available during the FS.”
EPA, A Guide to Preparing Superfund Proposed Plans,
Records of Decision, and Other Remedy Selection Decision
Documents, Emhart518-33. However, “after public
comment is received on the Proposed Plan, ” the
modifying criteria will be “fully considered.”
Id. This means that, “[i]n the final balancing
of trade-offs between alternatives upon which the final
remedy selection is based, modifying criteria are of equal
importance to the balancing criteria.” Id.
This does not provide the state or community the power to
veto a selected remedy[8]; their input is simply considered along
with the other balancing criteria.
After
completing the “detailed analysis of
alternatives” using the nine criteria, EPA chooses a
remedy through a “two-step process.” 40 C.F.R.
§ 300.430(f)(1)(ii). First, EPA “identifies a
preferred alternative and presents it to the public in a
proposed plan, for review and comment.” Id.
The proposed plan functions not only to “supplement the
RI/FS, ” but also to “provide the public with a
reasonable opportunity to comment on the preferred
alternative for remedial action, as well as alternative plans
under consideration, and to participate in the selection of
remedial action at a site.” Id. §
300.430(f)(2). To accomplish this, the proposed plan must,
among other things, “[p]rovide a brief summary
description of the remedial alternatives evaluated in the
detailed analysis” as well as “[i]dentify and
provide a discussion of the rationale that supports the
preferred alternative.” Id. The public is then
given “a reasonable opportunity, not less than 30
calendar days, for submission of written and oral
comments” as well as an “opportunity for a public
meeting.” Id. § 300.430(f)(3)(i)(C)-(D).
After
receiving public input on the proposed remedy, the second
step for remedy selection requires EPA to “review the
public comments and consult with the state (or support
agency) in order to determine if the [proposed plan] remains
the most appropriate remedial action for the site or site
problem.” Id. § 300.430(f)(1)(ii). The
NCP anticipates that public comments may provide “new
information or points of view” that prompt EPA
“to modify aspects of the preferred alternative or
decide that another alternative provides a more appropriate
balance.” Id. § 300.430(f)(4). To the
extent EPA makes any significant changes to the remedy, those
changes must be documented.[9] EPA must then make the final
remedy selection. Id. § 300.430(f)(4)(i).
Just as
EPA is not required to remove all uncertainty at the RI/FS
stage regarding the conditions at the site, EPA is also not
required to provide complete details of the final remedy at
the selection stage. The NCP envisions that EPA will fill in
the details of the final remedy during the implementation (or
“remedial design”) phase and that the final
remedy may require modifications. See generally id.
§ 300.435. The NCP also allows EPA to reserve decisions
regarding how to handle certain portions of the remedy until
remedial design. See, e.g., id. §
300.825(a)(1)-(2). This process is discussed below in the
“Remedial Design and Remedial Action” section.
D.
Documentation of the Remedy-Selection Process
After
selecting the final remedy, EPA must “establish an
administrative record that contains the documents that form
the basis for the selection of a response action.” 40
C.F.R. § 300.800. An important piece of the
administrative record is the Record of Decision
(“ROD”), which includes “all facts,
analyses of facts, and site-specific policy determinations
considered” by EPA in selecting the final remedy.
Id. § 300.430(f)(5)(i). The ROD is essentially
EPA's justification for its decision, explaining, for
example, “[h]ow the selected remedy is protective of
human health and the environment” and “provides
overall effectiveness proportional to its costs.”
Id. § 300.430(f)(5)(ii)(A), (D). As part of
this justification, the ROD also must include a
“responsiveness summary, ” which is “a
written summary of significant comments, criticisms, and new
relevant information submitted during the public comment
period and the lead agency response to each issue.”
Id. § 300.430(f)(3)(i)(F). Lastly, if EPA
chooses to reserve certain decisions for a later date, it
may, “[w]here appropriate, provide a commitment for
further analysis.” Id. §
300.430(f)(5)(iii)(D).
The
administrative record is not a static document. Even after
the final remedy is selected, EPA is responsible for updating
the administrative record where necessary. For example, if
the ROD did not address a portion of the response action or
reserved certain decisions until the implementation phase,
EPA must document those later decisions in the administrative
record. Id. § 300.825(a)(1). Additionally, if
EPA decides to modify the final remedy during remedial
design, it must document those changes either through an
“explanation of significant differences” or a ROD
amendment, as appropriate. Id. §
300.825.[10]
E.
Remedial Design and Remedial Action
The
final step is “the development of the actual design of
the selected remedy and implementation of the remedy through
construction.” Id. § 300.435(a). The NCP
labels this as the “remedial design/remedial action
(RD/RA) stage.” Id. While the ROD establishes
the final remedy, the NCP leaves it to the RD/RA stage for
EPA to determine the remedy's final design. The
“initial building block in developing” the final
design is the information contained in the RI/FS, but EPA
guidance also envisions that additional “data
acquisition” and “sample analysis” will be
necessary during RD/RA. EPA, Scoping the Remedial
Design, Emhart516-2. EPA then recommends going through
multiple design phases before coming to a “final
design” and beginning construction. Id. at
0001. Once a final design is complete, EPA must - as a final
notice to the public - “issue a fact sheet and provide,
as appropriate, a public briefing prior to the initiation of
remedial action.” 40 C.F.R. § 300.435(c)(3). EPA
may then implement the remedy, and as long as the
“remedial action objectives and remediation goals in
the ROD” are accomplished, the CERCLA response action
process is largely complete. Id. §
300.435(f)(1); 42 U.S.C. § 9621(d) (stating that the
response action must “attain a degree of cleanup . . .
which assures protection of human health and the
environment”).
If
strict adherence to the final design proves unworkable at any
point, EPA guidance provides for significant
“flexibility” to account for “any
constraining factors of the particular site.” EPA,
Scoping the Remedial Design, Emhart516-1. The NCP
also foresees that “[a]dditional work” may be
“needed as a result of such unforeseen situations as
newly discovered sources, types, or quantities of hazardous
substances.” 40 C.F.R. § 300.435(e)(1)(i). Because
the “chief task” of RD/RA is “to achieve
the goals of the Record of Decision . . . in a timely manner,
” EPA, Scoping the Remedial Design,
Emhart516-1, as opposed to blind adherence to any particular
design, EPA is permitted to change the remedial design at any
point. The NCP simply requires that sufficient notice and
opportunity to comment is provided to the
public.[11]
F.
Unilateral Administrative Order
The
parties responsible for the release of the hazard materials
at the site are liable for the costs associated with the
response action. See 42 U.S.C. § 9607(a). EPA
can either complete the response action and seek
reimbursement from the responsible parties or require the
responsible parties to implement the response action
themselves. Where EPA determines that “there may be an
imminent and substantial endangerment to the public health or
welfare or the environment, ” EPA is authorized to
issue a unilateral administrative order (“UAO”)
requiring the responsible parties to implement the response
action “as may be necessary to protect public health
and welfare and the environment.” Id. §
9606(a).
If a
responsible party “willfully violates, or fails or
refuses to comply with” the UAO, EPA may seek an order
“in the appropriate United States district court to
enforce” the UAO. Id. § 9606(b)(1).
Additionally, if the district court finds that the
responsible party refused to comply with the UAO without
“sufficient cause, ” the responsible party is
subject to daily fines during the period of non-compliance as
well as treble damages for any work EPA performed at the
site. Id. § 9606(b)(1), 9607(c)(3). By
regulation, the daily fine amount is $37, 500 per day for
every day of non-compliance between December 6, 2013 and
November 2, 2015, and $54, 789 per day thereafter. 40 C.F.R.
§§ 19.2, 19.4.
III.
Scope of Review
Having
described the remedy-selection process as outlined by CERCLA
and the NCP, the Court will now provide its findings of fact
and conclusions of law regarding EPA's remedy-selection
process at the Site. Before doing so, however, the Court must
first determine what evidence and arguments it will consider,
as well as the standard of review. As discussed previously,
the Government moved, pre-trial, to limit the scope of
discovery and judicial review. According to the Government,
the Court should not consider evidence and arguments not
contained in the administrative record when making its
determination about the appropriateness of EPA's chosen
remedy. The Court denied the Government's motion without
prejudice, permitting Emhart to take discovery and present
evidence and arguments at trial not contained in the
administrative record. The Court “reserve[d] ruling on
the admissibility of any particular extra-record evidence
until the time of trial.” (Order Den. Mot. to Limit
Disc. and Scope of Review 6.) The Government has reiterated
its arguments at trial and in its post-trial briefs, and
Emhart has again responded. (See Gov't
Post-Trial Brief 8-13; Emhart Post-Trial Reply Brief 1-22.)
The two
questions the Court must answer are: (1) whether the Court
should consider evidence not contained in the administrative
record; and (2) whether the Court should consider
Emhart's arguments that were not made during the notice
and comment period.
A. What
Evidence Should the Court Consider?
CERCLA
provides the following limitation to judicial review:
“In any judicial action under this chapter, judicial
review of any issues concerning the adequacy of any response
action taken or ordered by the President shall be limited
to the administrative record.” 42 U.S.C. §
9613(j)(1) (emphasis added). CERCLA reiterates this
limitation when describing the applicable standard of review:
“In considering objections raised in any judicial
action under this chapter, the court shall uphold the
President's decision in selecting the response action
unless the objecting party can demonstrate, on the
administrative record, that the decision was arbitrary
and capricious or otherwise not in accordance with
law.” Id. § 9613 (j)(2) (emphasis
added).[12] As these provisions make clear,
“[u]nder CERCLA, judicial review normally is limited to
the administrative record as it existed at the time of the
challenged agency action.” United States v. JG-24,
Inc., 478 F.3d 28, 33-34 (1st Cir. 2007); see also
Murphy v. Comm'r of Internal Revenue, 469 F.3d 27,
31 (1st Cir. 2006) (“[T]he Supreme Court has
consistently stated that review of administrative decisions
is ordinarily limited to consideration of the decision of the
agency . . . and of the evidence on which it was
based.”) (internal quotations omitted). This ensures
that the Court “take[s] into account ‘neither
more nor less information than did the agency when it made
its decision.'” Linemaster Switch Corp. v.
EPA, 938 F.2d 1299, 1305 (D.C. Cir. 1991) (quoting
Walter O. Boswell Mem'l Hosp. v. Heckler, 749
F.2d 788, 792 (D.C. Cir. 1984)).
However,
while generally confining the scope of judicial review to the
administrative record, CERCLA also provides that
“[o]therwise applicable principles of administrative
law shall govern whether any supplemental materials may be
considered by the court.” 42 U.S.C. § 9613(j)(1).
Typically, “[c]ourts require a strong showing of bad
faith or improper behavior before ordering the
supplementation of the administrative record.” Town
of Norfolk v. U.S. Army Corps of Eng'rs, 968 F.2d
1438, 1458-59 (1st Cir. 1992); see also JG-24, Inc.,
478 F.3d at 34 (“Normally, we do not allow
supplementation of the administrative record unless the
proponent points to specific evidence that the agency acted
in bad faith.”). This exception provides no assistance
to Emhart, as there is no evidence that EPA crafted a remedy
for the Site in bad faith or based on an improper motive, and
EPA's “designation of the Administrative Record,
like any established administrative procedure, is entitled to
a presumption of administrative regularity.” Nw.
Bypass Grp. v. U.S. Army Corps of Eng'rs, No. CIV
06-CV-00258-JAW, 2007 WL 1498912, at *2 (D.N.H. May 14, 2007)
(quoting Bar MK Ranches v. Yuetter, 994 F.2d 735,
740 (10th Cir. 1993)).
Emhart
must therefore find some other legal avenue if it is to
successfully inject evidence outside the administrative
record into this proceeding. Outside of a showing of bad
faith, the First Circuit recognizes two other
“exceptions to the rule against supplementation.”
Nw. Bypass Grp., 2007 WL 1498912, at *2.
First,
“supplementation may be proper when . . . there is a
record so inadequate that it prevents judicial review.”
Id. at *2-3 (citing Murphy, 469 F.3d at
31). This applies in very limited circumstances. For
instance, the Supreme Court has recognized that, “where
there are [no formal] administrative findings that were made
at the same time as the decision, . . . it may be that the
only way there can be effective judicial review is by
examining the decisionmakers themselves.” Citizens
to Pres. Overton Park, Inc. v. Volpe, 401 U.S. 402, 420
(1971) (citing Shaughnessy v. Accardi, 349 U.S. 280
(1955)). This applies where the administrative agency's
decision cannot be discerned or justified on the record.
See Camp v. Pitts, 411 U.S. 138, 142-43 (1973)
(reviewing whether “there was such failure to explain
administrative action as to frustrate effective judicial
review”). Therefore, supplemental materials are
typically unnecessary when the administrative agency's
judgment is based on a substantial record. See Nw. Bypass
Grp., 2007 WL 1498912, at *3 (finding an administrative
record “more than sufficient to allow for judicial
review” because the administrative record was
“hefty, 3, 233 pages over seven volumes, with documents
spanning from 1989 to 2006”). Where a substantial
record is available, even if the administrative agency's
explanation is “curt, ” supplemental materials
are unnecessary as long as the explanation “indicate[s]
the determinative reason for the final action taken.”
Pitts, 411 U.S. at 143.
The
second exception to the “rule against
supplementation” applies where “additional
testimony by experts” will “aid to understanding
highly technical, environmental matters.” Nw.
Bypass Grp., 2007 WL 1498912, at *2 (quoting Valley
Citizens for a Safe Env't v. Aldridge, 886 F.2d 458,
460 (1st Cir. 1989)). The Court may even allow
“additional factual evidence as an aid to
understanding.” Valley Citizens for a Safe
Env't, 886 F.2d at 460. The necessity of such
evidence “is discretionary with the reviewing
court.” Id. However, it is important to note
that the purpose of such evidence is “simply to help
[the Court] understand matters in the agency record.”
Id. Therefore, the Court still “looks first
and foremost at the record before the agency.”
Id.; see also Olsen v. United States, 414
F.3d 144, 155 (1st Cir. 2005) (“The focal point for
judicial review should be the administrative record already
in existence, not some new record made initially in the
reviewing court.”) (quoting Pitts, 411 U.S. at
142); United States v. Dravo Corp., No. 8:01CV500,
2003 WL 21434761, at *3 (D. Neb. June 20, 2003) (reviewing an
EPA response action and refusing to “permit
supplementation of the record absent a showing that it is
explanatory and not intended to invoke new material into the
case”).
In this
case, given the extensive record compiled by EPA, as well as
EPA's documentation of its decision-making process in the
ROD, the Court does not find the record “so
inadequate” as to require supplemental materials.
However, the subject matter involved certainly falls under
the umbrella of “highly technical, environmental
matters” where the Court has discretion to consider
“additional testimony by experts” and
“additional factual evidence” as an aid to
understanding the administrative record. Nw. Bypass
Grp., 2007 WL 1498912, at *2 (quoting Valley
Citizens for a Safe Env't, 886 F.2d at 460). The
Court will therefore consider the expert testimony presented
by both parties. With that said, in considering the expert
testimony, the Court acknowledges that it must remain focused
“first and foremost” on the administrative
record, Valley Citizens for a Safe Env't, 886
F.2d at 460, as “the focal point for judicial review
should be the administrative record already in existence, not
some new record made initially in the reviewing court.”
Olsen, 414 F.3d at 155 (quoting Pitts, 411
U.S. at 142).
B. What
Arguments Should the Court Consider?
Having
established what evidence is admissible in this case, the
Court must determine which of Emhart's arguments the
Court will consider. According to EPA, the Court should not
consider any of Emhart's arguments that were not
previously submitted during the official notice and comment
period on the proposed remedy. Emhart, on the other hand,
urges the Court to consider all of its arguments, including
those made for the first time at trial.
There
is limited precedent directly addressing this issue in the
context of CERCLA. Generally speaking, arguments not made
before an administrative agency are waived when a court
reviews the administrative agency's decision:
[w]e have recognized in more than a few decisions, and
Congress has recognized in more than a few statutes, that
orderly procedure and good administration require that
objections to the proceedings of an administrative agency be
made while it has opportunity for correction in order to
raise issues reviewable by the courts.
United States v. L.A. Tucker Truck Lines, Inc., 344
U.S. 33, 36- 37 (1952). This requirement is commonly referred
to as “issue exhaustion.” Sims v. Apfel,
530 U.S. 103, 107 (2000). As the First Circuit has explained,
“this rule preserves judicial economy, agency autonomy,
and accuracy of result by requiring full development of
issues in the administrative setting to obtain judicial
review.” Pepperell Assocs. v. EPA, 246 F.3d
15, 27 (1st Cir. 2001) (citing Northern Wind, Inc. v.
Daley, 200 F.3d 13, 18 (1st Cir. 1999)).
The
doctrine of issue exhaustion was adhered to relatively
recently by the First Circuit in Upper Blackstone Water
Pollution Abatement Dist. v. EPA, 690 F.3d 9,
18 (1st Cir. 2012). That case involved a sewage company
(“petitioner”) appealing a limitation imposed on
it by EPA under the National Pollutant Discharge Elimination
System. After notice and comment on the limitation,
petitioner challenged EPA's decision in the First
Circuit. Petitioner sought review of, among other things,
“the limit placed on aluminum discharge, arguing that
the EPA assembled and then relied upon an erroneous data set
in deriving the limit.” Id. at 33. The First
Circuit refused to consider that argument, however, because
it had not been raised during the notice and comment period.
The
First Circuit explained that, by regulation, petitioner was
required to “raise all reasonably ascertainable issues
and submit all reasonably available arguments supporting
their position by the close of the public comment period . .
. .” Upper Blackstone, 690 F.3d at 30 (quoting
40 C.F.R. § 124.13). In light of this statutory
requirement, the court found that petitioner had
“waived the argument by failing to raise it during the
public comment period of the permitting process. . . . By
failing to give the EPA an opportunity to address the
argument during the permitting process, [Upper Blackstone]
has waived its claim.” Id. (citing several
cases, including L.A. Tucker Truck Lines, 344 U.S.
33, and Pepperell Assocs., 246 F.3d 15).
Emhart
argues that Upper Blackstone is distinguishable from
this case because, unlike the regulation at issue in
Upper Blackstone, CERCLA and its
implementing regulations do not contain a clear issue
exhaustion provision. The Court disagrees.
A
determination of whether issue exhaustion applies to an
administrative process “requires careful examination of
the characteristics of the particular administrative
procedure provided.” Sims, 530 U.S. at 112-13
(O'Connor, J., concurring) (quoting McCarthy v.
Madigan, 503 U.S. 140, 146 (1992)). And while the
regulation reviewed by the First Circuit in Upper
Blackstone may have been slightly more definitive than
the language in CERCLA, the Court finds that CERCLA and its
implementing regulations, when taken as a whole, clearly
require interested parties to present arguments to EPA before
bringing those issues before a federal court.
After
identifying the “preferred” remedy, EPA is
required to “present[] it to the public in a proposed
plan, for review and comment.” 40 C.F.R. §
300.430(f)(1)(ii). This ensures that the public has “a
reasonable opportunity to comment on the preferred
alternative for remedial action, as well as alternative plans
under consideration, and to participate in the selection of
remedial action at a site.” Id. §
300.430(f)(2). At the completion of the notice and comment
period, EPA is required to create a “responsiveness
summary, ” which is “a written summary of
significant comments, criticisms, and new relevant
information submitted during the public comment period and
the lead agency response to each issue.” Id.
§ 300.430(f)(3)(i)(F). The responsiveness summary is
then placed in the administrative record. Id.
Importantly
though, the responsiveness summary does not address comments
made outside the public comment period. See Id.
§ 300.825. As is mandated by CERCLA's implementing
regulations, the responsiveness summary will not address
comments made “after the close of the public comment
period” unless the comments: (1) “contain
significant information not contained elsewhere in the
administrative record file”; (2) the information
“could not have been submitted during the public
comment period”; and (3) the information
“substantially support[s] the need to significantly
alter the response action.” Id. §
300.825. Therefore, unless an issue raised after the public
comment period falls under this exception, that issue will
not be responded to in the responsiveness summary or included
in the administrative record. And it is in this context that
CERCLA explicitly limits judicial review to the information
contained in the administrative record. See 42
U.S.C. § 9613(j)(1)-(2).
This
statutory and regulatory scheme, when viewed as a whole,
requires parties to make all of their known and available
arguments regarding the merits of a remedy to EPA during the
notice and comment period in the first instance. Only then,
after EPA has had the opportunity to provide its response in
the administrative record, may a federal court review
EPA's decision. See, e.g., JG-24, Inc.,
478 F.3d at 33-34 (“Under CERCLA, judicial review
normally is limited to the administrative record as it
existed at the time of the challenged agency action.”);
Arco v. Travelers Ins. Co., 730 F.Supp. 59, 69 (W.D.
Mich. 1989) (“[R]efusal to participate in this
administrative process [under CERCLA] essentially allows the
EPA a free-reign in dictating response methods since judicial
review is limited to the administrative record.”). To
allow Emhart to make arguments it could have made directly to
EPA for the first time on judicial review would frustrate
this administrative scheme by depriving EPA of the
opportunity to address Emhart's arguments in the first
instance on administrative review. Upper Blackstone,
690 F.3d at 30.
However,
while judicial review is generally limited to arguments
presented during the notice and comment period, the Court
also recognizes that there are some narrow exceptions to this
rule. For instance, the Court of Appeals for the District of
Columbia Circuit has recognized that EPA must justify
“key assumptions” in its analysis, regardless of
whether a party specifically objects during the notice and
comment process:
EPA has a preexisting duty to examine key assumptions as part
of its affirmative burden of promulgating and explaining a
non-arbitrary, non-capricious rule and therefore . . . must
justify that assumption even if no one objects to it during
the comment period.
Oklahoma Dep't of Envtl. Quality v. EPA, 740
F.3d 185, 192 (D.C. Cir. 2014) (quoting Appalachian Power
Co. v. EPA, 135 F.3d 791, 818 (D.C. Cir.
1998)).[13]
While
the First Circuit has not directly addressed this narrow
exception, the Court agrees with the District of Columbia
Circuit that issue waiver cannot absolve EPA of its
responsibility to explain the key assumptions that underpin
its remedy and that contain obvious mistakes. Therefore,
while the Court will generally not consider Emhart's
arguments that were not presented to EPA during the notice
and comment period, the Court will consider several obvious
issues relating to key assumptions that formed the basis of
EPA's selected remedy.
C.
Standard of Review
With
the bounds of admissible evidence and argument established in
the preceding sections, the Court turns to the applicable
standard of review. The Court will uphold EPA's decision
“unless the objecting party can demonstrate, on the
administrative record, that the decision was arbitrary and
capricious or otherwise not in accordance with law.” 42
U.S.C. § 9613(j)(2). The “law” with which
EPA must comply in selecting a remedial action is primarily
found in CERCLA and the NCP. EPA's decisions made within
that legal framework will qualify as “arbitrary and
capricious” if EPA fails to “examine the relevant
data and articulate a satisfactory explanation for its
action.” FCC v. Fox Television Stations, Inc.,
556 U.S. 502, 513-14 (2009) (quoting Motor Vehicle Mfrs.
Ass'n of United States, Inc. v. State Farm Mut. Auto.
Ins. Co., 463 U.S. 29, 43 (1983)).
This is
a “narrow” standard of review. Id.
EPA's explanation need only be “plausible in light
of the record as a whole” and “supported by
substantial evidence in the record.” Leahy v.
Raytheon Co., 315 F.3d 11, 17 (1st Cir. 2002) (citations
omitted). Furthermore, in reviewing the evidence, the Court
is not permitted “to substitute its judgment for that
of the agency.” Fox Television Stations, 556
U.S. at 513-14 (quoting Bowman Transp., Inc. v.
Arkansas-Best Freight Sys., Inc., 419 U.S. 281, 286
(1974)). Particularly when the Court reviews “a purely
factual question within the area of competence of an
administrative agency . . . and when resolution of that
question depends on ‘engineering and scientific'
considerations, ” the Court must “recognize the
relevant agency's technical expertise and experience, and
defer to its analysis unless it is without substantial basis
in fact.” Browning-Ferris Indus. of S. Jersey, Inc.
v. Muszynski, 899 F.2d 151, 160 (2d Cir. 1990) (quoting
Federal Power Commission v. Florida Power & Light
Co., 404 U.S. 453, 463 (1972)).
While
certainly deferential, the arbitrary and capricious standard
is not a shibboleth by which EPA may completely avoid
judicial scrutiny. As the Supreme Court has explained, an
action may qualify as arbitrary and capricious where the
administrative agency: (1) “relied on factors which
Congress has not intended it to consider, ” (2)
“entirely failed to consider an important aspect of the
problem, ” (3) “offered an explanation for its
decision that runs counter to the evidence before the agency,
or is so implausible that it could not be ascribed to a
difference in view or the product of agency expertise,
” or (4) seeks to have its action upheld based on
“post hoc rationalizations.” Motor
Vehicle Mfrs. Ass'n of U.S., Inc., 463 U.S. at 43.
Courts also review whether an administrative agency has
treated similar cases with “apparent irrational
discrimination.” Puerto Rico Sun Oil Co. v.
EPA, 8 F.3d 73, 78 (1st Cir. 1993) (citing Green
Country Mobilephone, Inc. v. FCC, 765 F.2d 235 (D.C.
Cir. 1985)). Put simply, the Court must review EPA's
analysis to ensure that it is “rational” and
“makes sense.” Penobscot Air Servs., Ltd. v.
FAA, 164 F.3d 713, 720 (1999) (quotations, citations,
and brackets omitted).
Emhart's
arguments as to why EPA's actions qualify as arbitrary
and capricious or otherwise not in accordance with law
address several aspects of EPA's processes as well as the
merits of the selected remedy itself. Each of Emhart's
arguments are addressed below.
IV.
Case Specific Findings of Fact and Conclusions of Law
A.
Background
1.
Findings of Fact
In 1996
dioxin was discovered in fish collected from the
Woonasquatucket River. (See Phase I Findings
12.)[14] An EPA investigation of the surrounding
area - later labeled as the Centredale Manor Restoration
Project Superfund Site - ensued. The Site covers a three-mile
stretch of the Woonasquatucket River, which includes a
nine-acre peninsula that has been identified as the
“Source Area” of the Site's hazardous
substances. (Id. at 10-11.) The Source Area contains
two elderly housing facilities (Brook Village and Centredale
Manor) and is bounded to the north by Smith Street, to the
south by Allendale Pond, to the west by the Woonasquatucket
River, and to the east by the “tailrace, ” a
remnant of a narrow body of water used for water power by the
mills that once occupied the peninsula. (Id.)
After
passing by the Source Area, the Woonasquatucket River leads
to the remainder of the Site. The river first runs into
Allendale Pond, a .65-mile-long dammed pond that spans
fifteen acres and has depths ranging from .5 to ten
feet.[15] (ROD, US1444-12, 30.) The river next
runs into Lyman Mill Pond, a .85-mile-long dammed pond
spanning twenty-four acres with depths similar to Allendale
Pond. (Id.) In between Allendale and Lyman Mill
Ponds is the Oxbow Area, a forty-acre forested wetland
habitat adjacent to the river channel. (U.S. Army Corps of
Eng'rs, Oxbow Area Report, US1227-0005.) There
are also various abutting residential and commercial
properties throughout the Site. (Phase I Findings
10-11.)[16]
Ultimately,
EPA determined that the entire Site, not just the Source
Area, was polluted by a variety of contaminants, including
dioxins (2, 3, 7, 8-TCDD, in particular), volatile organic
compounds, polychlorinated biphenyls, semi-volatile organic
compounds, polycyclic aromatic hydrocarbons, and various
metals. (Phase I Findings 11-12.) This determination led EPA
to list the Site on the National Priorities List of Superfund
sites in 2000. (Id.)
The
Site, in short, is complicated. It contains several types of
toxic hazardous waste, and the Site's size and diversity
require EPA to consider various types of flora, fauna,
physical features, and human uses. As part of the cleanup
effort EPA has required several removal and remedial actions.
These involved the construction of three interim protective
caps as well as one RCRA[17]cap over portions of the Source
Area[18]; reconstruction of the Allendale Dam and
restoration of Allendale Pond to prevent further downstream
migration of contaminants; excavation and removal of one
hundred cubic yards of soil from eleven areas along Allendale
and Lyman Mill Ponds; and erection of fences along the
residential properties adjacent to the Site in order to
prevent access to the contamination. (ROD, US1444-15-16.) To
varying degrees, Emhart has participated in each of these
removal actions. (Phase I Findings n.110.)
Additionally,
EPA has pursued a “comprehensive” remedial action
at the Site that will address “all current and
potential future risks caused by soil, sediment, groundwater
and surface water contamination.” (ROD, US1444-6.) The
first step in this process (after listing the Site on the
NPL) was the remedial investigation (“RI”).
EPA's RI[19] utilized Site-specific data and modeling
to characterize the nature and extent of contamination at the
Site. (ROD, U.S. 1444-16-17.) The culmination of this process
was the RI report released in 2005. (ROD, U.S. 1444-16-17;
see also RI, US1098.) The RI included an assessment
of the risks to human health (the Baseline Human Health Risk
Assessment, or “BHHRA”)[20]and ecology (the Baseline
Ecological Risk Assessment, or
“BERA”)[21]posed by the Site.
Using
the information collected during the RI, EPA developed
remediation goals that, if achieved, would likely mitigate
the risks to human health and the environment identified in
EPA's risk assessment. (Feasibility Study, US1254-73-84.)
EPA conducted an extensive feasibility study
(“FS”) in order to determine which remedial
alternative would best achieve those targets.[22] The FS
divides the Site into five “action areas, ” and
provides potential remedial actions for each. (FS,
US1254-10-14.) As is required by the NCP, the FS includes
both an initial screening as well as a detailed analysis of
the remedial alternatives for each action area. (FS,
US1254-147-321.)
Based
on the analysis in the FS, EPA drafted a Proposed Remedial
Action Plan (“PRAP”) in the fall of 2011.
(See PRAP, U.S. 1328.) This version of the remedy
did not last long though. Soon after its publication, EPA
released a nationwide change to its non-cancer toxicity value
for dioxin. Since dioxin is present at the Site, EPA was
forced to issue a “Technical Memorandum” updating
the BHHRA, cleanup levels, and FS for the Site. (See
2012 Technical Mem., US1392.) These findings required several
changes to the PRAP in the form of a PRAP Amendment.
(See PRAP Amendment, US1393.) While much of the PRAP
went unchanged, the PRAP Amendment did require, among other
things, an expanded cleanup area at the Site. These changes
were published in July, 2012.
Both
the PRAP and PRAP Amendment were subject to notice and
comment after their publication. The notice and comment
period on the PRAP and PRAP Amendment went from November 14,
2011 to March 2, 2012, and July 19, 2012 to September 17,
2012, respectively. (ROD, US1444-24-25.) During that time EPA
participated in public hearings and also accepted comments
from a variety of sources, including Emhart. (See
Emhart Comments on PRAP, US1383; Emhart Comments on PRAP
Amendment, US1418.)
On
September 28, 2012, with the public comment period complete,
EPA issued its Record of Decision (“ROD”)
explaining the remedial action plan. (See ROD,
US1444.) The ROD provides EPA's justification for the
chosen remedial action as well as a “Responsiveness
Summary” that addresses significant public comments
submitted to EPA on the PRAP and PRAP Amendment. In the end,
the plan outlined in the ROD was substantially similar to the
plan provided for in the PRAP and PRAP Amendment.
The
remedy, as described in the ROD, contains the following basic
characteristics. In the Source Area the ROD requires removal
and off-Site treatment or disposal of waste material,
installation of a RCRA C cap[23] over existing surfaces, and
relocation of underground utilities into clean corridors.
(ROD, US1444-7.) Sediment and floodplain soil in Allendale
and Lyman Mill ponds will also be excavated. However, unlike
in the Source Area, the majority of the excavated material in
and around the ponds will be placed in a confined disposal
facility (“CDF”) near the Site, leaving only a
small portion of the excavated material (estimated at
approximately ten percent of the total) to be shipped
off-Site[24] for disposal or treatment. (Id.
at 7.) A thin-layer cover will be placed over the remaining
contaminated areas in the Oxbow wetland area and, if
necessary, over remaining contaminated sediment in the
Woonasquatucket River. (Id.) Going forward, the ROD
requires continuous institutional controls that limit certain
activities at the Site, such as construction and use of
groundwater. (Id.) Additionally, long-term
maintenance and monitoring is required, including of the CDF,
damns, sediment, water, and biota located at the Site.
(Id.) In total, EPA estimates that the remedial
action will cost approximately $104, 600, 000. (Id.
at 213.)
Lastly,
the ROD commits that EPA will collect additional information
during the remedial design phase. For instance, EPA will have
to determine the location of the CDF. (ROD, US1444-6.) EPA
must also collect additional soil samples to determine the
precise amount of excavation required. (ROD, US1444-304.) It
is possible that the remedy will change based on this
information.[25]However, EPA maintains that it has the
capacity to adequately address such changes through, for
example, an amendment to the remedial action plan or an
“explanation of significant
differences.”[26]
EPA has
compiled an extensive administrative record to document its
remedy-selection process. The administrative record includes
essentially all documents related to the development,
creation, and implementation of the remedial action. The next
step in EPA's process is to create a more in-depth design
of the remedial action plan and implement it.[27] However,
before the final design and implementation of the remedy
occurred, Emhart challenged EPA's proposed remedy on
several grounds.
2.
Conclusions of Law
As a
general matter, the Court finds that EPA followed the basic
steps mandated by CERCLA and the NCP in developing its
remedial action for the Site. These legal requirements were
previously outlined by the Court. See supra Section
II (“Remedy Selection”). However, Emhart argues
that several of EPA's individual actions and decisions
along the way were either arbitrary, capricious, or not in
accordance with CERCLA or the NCP such that the remedy cannot
withstand judicial scrutiny. The Court addresses each of
Emhart's arguments below.
B.
Excavation and Treatment of Soil and Sediment
1.
Findings of Fact
a.
Excavation of Allendale and Lyman Mill Ponds
EPA's
remedy calls for significant excavation of the sediment in
Allendale and Lyman Mill Ponds. The goal of excavation is to
achieve dioxin levels of approximately fifteen parts per
trillion. (FS, US1254-75.) In order to determine how much
excavation will be required to meet this target, EPA took
sediment samples from both Allendale and Lyman Mill Ponds.
Based on these samples, EPA estimates the average excavation
depth in Allendale Pond required to achieve the target dioxin
level is 2.2 feet. (ROD, U.S. 1444-170.) For Lyman Mill Pond,
the average excavation depth required to achieve EPA's
target is 2.7 feet. (Id. at 170.) Both of these
estimates assume .25 feet of over-excavation will occur.
(Id.)
The
data set used by EPA to come up with these estimates included
250 data records from Allendale Pond (ranging from 0.5 to
twelve feet in depth) and 160 data records from Lyman Mill
Pond (ranging from 0.5 feet to four feet in depth). (ROD,
US1444-303; see also RI, US1098-29; FS, Tables G-3
and G-4, US1254-1458-1477.) The data most heavily relied upon
by EPA were core samples taken in 2003 and 2005, which
included ten sediment cores taken from Allendale Pond and
sixteen sediment cores taken from Lyman Mill Pond. (FS,
US1254-363-64.) Each core includes multiple soil samples all
of which were collected with the specific objective of
detecting the vertical extent of contamination in the ponds.
(Id. at 42, 679.) To that end, EPA conducted
laboratory analysis using high resolution mass spectroscopy,
a highly accurate method that can detect dioxin in parts per
trillion. (Dr. Medine Test., Trial Tr. vol. 12, 13:16-14:18,
ECF No. 495.)
Emhart's
expert, Mr. Loureiro, testified that these samples were
inadequate because only a small portion was taken at depths
greater than one foot.[28] Furthermore, Mr. Loureiro pointed out
that, in certain instances, dioxin levels were detected at
levels above fifteen parts per trillion at depths greater
than EPA's anticipated excavation depths.[29] As such, Mr.
Loureiro opined that the available data was insufficient to
accurately estimate the vertical extent of the contamination
and, in turn, the amount of excavation that will be necessary
to achieve EPA targets.
EPA
agrees that additional sampling is needed. As EPA explained
in the Feasibility Study, “[t]he proposed cleanup areas
or remedial footprints are conceptual and more precise
cleanup footprints will be developed during the remedial
design. For example, additional coring will need to be
performed at Allendale and Lyman Mill Ponds to confirm the
vertical extent of the contamination.” (FS,
US1254-327-28.) For this reason, EPA committed itself in the
ROD to “perform[ing] additional sampling and analysis
closer to the time of remediation to confirm the sediment
cleanup depth and volume.” (ROD, US1444-304.)
Given
the uncertainty as to the vertical extent of the
contamination, the exact amount of excavation required will
likely differ from the amount estimated in the ROD. Precision
in this area is likely impossible, however. As Mr. Loureiro
pointed out, “certainly in my experience with
excavation of all types, even a robust data set aren't
adequate to actually describe the conditions you run into in
the field.” (Mr. Loureiro Test., Trial Tr. vol. 6,
19:5-8.) This means that this component of the cost of the
remedy is to some extent uncertain and could potentially be
more expensive.
Emhart
did raise this general issue during the notice and comment
period on the PRAP. (Emhart PRAP Comments US1383-8
(“EPA fails to adequately define the volume of soil and
sediment requiring excavation.”); see also id.
at 51.) EPA responded in the ROD's “Responsiveness
Summary” by explaining that, in its opinion, sufficient
data had been collected to provide excavation estimates,
particularly in light of the 0.25-foot over-excavation
allowance built into the estimate. (ROD, US1444-303-04.) EPA
also noted that additional sampling will be done during
remedial design in order to refine those estimates.
(Id.) Lastly, to the extent that the proposed level
of excavation does not achieve target dioxin levels, EPA
suggests that “a 6-inch soil cover on the sediment
bottom” could be used where “additional
excavation is not feasible.” (Id.)
b.
Excavation of the Oxbow Area and the Floodplain Soil of
Allendale and Lyman Mill Ponds
In
February 2012, after publication of the PRAP, EPA established
a nation-wide non-cancer toxicity value for dioxin. This new
information forced EPA to reevaluate its remedial design for
the Site. EPA did so by issuing a Technical Memorandum on the
Impact of Dioxin Reassessment that updated EPA's human
health risk assessment and feasibility study. (2012 Technical
Memorandum, US1392.) The analysis of each remedial action
alternative did not change significantly. (See ROD,
US1444-347 (“Impacts resulting from these changed
conditions are presented in EPA's May 2012 Technical
Memorandum and are generally consistent with evaluations
presented in . . . the FS.”).)
However,
EPA did determine that certain areas, not previously
identified in the PRAP, would require excavation. These areas
were primarily located in the Oxbow Area and floodplain soil
around Allendale and Lyman Mill Ponds. (See PRAP
Amendment, US1393-002.) EPA's expanded excavation plan
was documented in the PRAP Amendment and later adopted in the
ROD. (PRAP Amendment, US1393-7-9; ROD, US1444-176-192.)
Emhart submitted comments on the PRAP Amendment in which it
argued that EPA had not collected sufficient Site-specific
data in order to adequately characterize contamination at the
Site. (See Emhart Comments on PRAP Amendment,
US1418-7-9.)[30]
With
regards to the floodplain soils, EPA determined which areas
would require excavation using two types of Site-specific
data: soil samples and FEMA floodplain maps. The soil samples
were collected during the remedial investigation on the
eastern shore floodplains[31] of the Allendale and Lyman
Mill Pond reaches. EPA collected 226 samples in total, 212 of
which detected some level of dioxin. (ROD, US1444-345.) Of
those 212 samples detecting dioxin, “there are
approximately 100 sampling locations where floodplain
residential-use soil samples have [dioxin] concentrations
greater than cleanup levels.” (Id.; see
also 2012 Technical ...