United States Court of Appeals
         FOR THE DISTRICT OF COLUMBIA CIRCUIT



Argued November 7, 2019                 Decided July 28, 2020

                         No. 18-1325

                       MERITOR, INC.,
                        PETITIONER

                              v.

           ENVIRONMENTAL PROTECTION AGENCY,
                     RESPONDENT


       On Petition for Review of Agency Action of the
       United States Environmental Protection Agency


    Catherine E. Stetson argued the cause for petitioner. With
her on the briefs were Mitchell P. Reich, Heidi B. Friedman,
and Joel Eagle.

    James R. Bieke, C. Frederick Beckner III, Peter C.
Tolsdorf, Steven P. Lehotsky, and Michael B. Schon were on
the brief for amici curiae the National Association of
Manufacturers, et al., in support of petitioner. Linda E. Kelly
entered an appearance.

    John Thomas H. Do, Attorney, U.S. Department of Justice,
argued the cause for respondent. With him on the brief was
Jonathan D. Brightbill, Principal Deputy Assistant Attorney
General. Jon M. Lipshultz, Attorney, entered an appearance.
                                2
   Before: MILLETT and KATSAS, Circuit Judges, and
SENTELLE, Senior Circuit Judge.

    Opinion for the Court filed by Circuit Judge MILLETT.

     MILLETT, Circuit Judge: The Environmental Protection
Agency maintains a National Priorities List that identifies those
hazardous waste sites in most urgent need of cleanup based on
the threat that they pose to public and environmental health and
to the public welfare. In 2018, the EPA added the Rockwell
International Wheel & Trim facility and surrounding areas to
the National Priorities List. Meritor, Inc., which has assumed
Rockwell International Corporation’s environmental liabilities
for the facility, challenges the listing as arbitrary, capricious,
and contrary to governing regulations. Meritor’s main
objection is that the EPA failed to take sufficient account of an
already installed sub-slab depressurization system in
determining the hazardousness of the site. Because the EPA’s
decision was reasonable and consistent with the governing
regulatory provisions, we deny the petition for review.

                                I

                                A

     Congress enacted the Comprehensive Environmental
Response, Compensation, and Liability Act of 1980
(CERCLA), 42 U.S.C. §§ 9601 et seq., “to address the growing
problem of inactive hazardous waste sites throughout the
United States.” Eagle-Picher Indus., Inc. v. EPA, 

, 925 (D.C. Cir. 1985). CERCLA directs the EPA to
develop “criteria for determining priorities among releases or
threatened releases” of hazardous waste into the environment.
42 U.S.C. § 9605(a)(8)(A). Based on those determinations, the
EPA must maintain a National Priorities List.

§ 9605(a)(8)(B). That List “identifies those hazardous-waste
sites considered to be the foremost candidates for
environmental cleanup” based on “the relative risk or danger
they pose to the public health, public welfare, or the
environment.” CTS Corp. v. EPA, 

, 55 (D.C. Cir.
2014).

    The EPA has developed a set of regulations, known as the
Hazard Ranking System, that it uses to evaluate whether, and
to what degree, a site poses a risk to the environment or to
human health and welfare. See generally 40 C.F.R. Part 300,
App. A.

     Under that scheme, the EPA evaluates up to four
“pathways” through which hazardous substances can migrate:
(1) groundwater, (2) surface water, (3) air, and (4) soil
exposure. 40 C.F.R. Part 300, App. A § 2.1. The soil-exposure
pathway includes analysis of the extent to which hazardous
substances intrude from the subsurface.

subsurface intrusion component evaluates the emanation (or
potential emanation) of noxious vapors from the soil into
occupied buildings.

subsurface intrusion
factor is the central focus of this case.

     In evaluating each pathway, the EPA weighs three metrics:
(1) the “likelihood of release” of hazardous waste into the
environment, (2) the “waste characteristics” of those
substances (such as the quantity, toxicity, mobility, persistence,
capacity to degrade, or bioaccumulation potential), and (3) the
“targets” of the hazardous waste, meaning who will suffer
exposure, whether humans, animals, natural resources, or
sensitive environments. 40 C.F.R. Part 300, App. A §§ 2.3–
2.5.

     Based on those metrics, the EPA assigns a numerical value
to each pathway, which it then converts into a score between 0
                               4
and 100 for the site as a whole. 40 C.F.R. Part 300, App. A
§ 2.1.1, 2.1.2 & Table 2-1. Sites scoring 28.5 or higher may be
added to the List. 83 Fed. Reg. 46,408, 46,409 (Sept. 13, 2018).

    Once a site is added to the National Priorities List, it
becomes eligible for remedial action financed by the EPA’s
Superfund Program. CTS 

; see also 42
U.S.C. § 9611. If the EPA chooses to conduct a cleanup, it may
seek reimbursement from parties who are potentially
responsible for contributing to the hazard, known as
“Potentially Responsible Parties.” General Elec. Co. v.
Jackson, 

, 114 (D.C. Cir. 2010) (citing 42 U.S.C.
§§ 9604(a), 9607(a)(4)(A)).

                               B

     From 1966 to 1985, the automotive division of Rockwell
International Corporation manufactured wheel covers at a
facility in Grenada, Mississippi, which we shall refer to as the
Rockwell Facility. The Rockwell Facility borders a residential
neighborhood, as well as a creek and agricultural land. The
wheel-cover manufacturing process produced hazardous
substances, including toluene, trichloroethylene (TCE), and
cis-1,2-dichloroethene (DCE), which were stored on site. A
1994 investigation revealed that those operations had led to a
plume of toluene and TCE collecting in the soil and
groundwater under and around the Rockwell Facility.

     Exposure to toluene or TCE can cause a range of health
impairments. Short-term inhalation exposure can result in
central nervous system dysfunction, leading to headaches,
dizziness, confusion, and fatigue. Long-term exposure can
cause even more serious health repercussions. Among other
things, TCE has been shown to be carcinogenic to humans and
                                5
chronic exposure to high levels of toluene can cause ataxia and
cerebral atrophy.1

     In 1985, Rockwell International sold the Rockwell Facility
to another company (that is not a party to this suit). Years later,
Rockwell International spun off its automotive division into a
separate corporation called Meritor, Inc. As a result, while
“Meritor never owned or operated the [Rockwell] Site[,]”
J.A. 48 n.5, it contractually “assumed various environmental
obligations in certain areas of the [Rockwell] Site[.]” J.A. 52.
As Meritor confirmed at oral argument, those environmental
obligations include assuming Rockwell International’s legal
liabilities with respect to the Rockwell Facility. Oral Arg. Tr.
4:25–5:3 (“Meritor * * * took on the liabilities of Rockwell[,]”
including those associated with “the Rockwell [S]ite.”).

     More recent studies of the Rockwell Facility demonstrate
the continued presence of hazardous waste, which has in turn
harmed air quality in the area. J.A. 12–13, 17–18. In October
2016, the EPA’s study of indoor air quality in the Rockwell
Facility’s main production building revealed elevated
concentrations of toluene, TCE, and DCE. In April 2017,
Meritor commissioned a study that found heightened levels of
toluene and TCE beneath the surface.



    1
       See United States EPA, Toluene (2016), https://www.epa.gov/
sites/production/files/2016-09/documents/toluene.pdf; United States
EPA, Trichloroethylene (2016), https://www.epa.gov/sites/
production/files/2016-09/documents/trichloroethylene.pdf.; see also
Agency for Toxic Substances & Disease Registry, 1,2-
Dichloroethene (1996), https://www.atsdr.cdc.gov/phs/phs.
asp?id=462&tid=82 (discussing the potential adverse health
effects of cis-1,2-dichloroethene).
                               6
     That same year, Meritor installed a sub-slab
depressurization system below the Rockwell Facility’s main
building. The depressurization system was designed to reduce
the intrusion of contaminated air into the building by creating
a pressure differential between the building and the underlying
soil. Despite improvements in air quality following the
installation of this system, the degree of contamination within
the main building continued to exceed ambient levels.
J.A. 585–586.

     On September 13, 2018, after going through notice and
comment rulemaking, the EPA added the Rockwell Facility
and surrounding areas (“Rockwell Site”) to the National
Priorities List based on the hazardous subsurface intrusion of
toluene, TCE, and DCE. 83 Fed. Reg. at 46,411.

     In evaluating the subsurface intrusion component of the
soil exposure pathway, the EPA considered the usual three
factors: the “likelihood of release” of hazardous waste, the
“waste characteristics” of those substances, and the “targets”
of such waste. 40 C.F.R. Part 300, App. A §§ 2.3, 2.4, 2.5.

     With respect to the “likelihood of release” factor, the EPA
either relies on an actual “observed exposure” or measures the
“potential for exposure.”        40 C.F.R. Part 300, App. A
§§ 5.2.1.1.1, 5.2.1.1.2. If the EPA documents an observed
exposure, the “likelihood of release” is automatically assigned
a maximum value of 550.

the EPA found
multiple instances of “observed exposure” based on indoor air
samples taken in October 2016 and January 2017, before the
sub-slab depressurization system was operative. So EPA
assigned the maximum value of 550.

    The EPA assigned a “waste characteristics” score of
56/100. That score was based on two elements: (1) the level
of toxicity/degradation of the substances in question, and
                               7
(2) the hazardous waste quantity, which relates to the quantity
of hazardous material found in regularly occupied structures.
40 C.F.R. Part 300, App. A §§ 2.4.1, 2.4.2, 5.2.1.2.

     Finally, the EPA analyzed the “targets” factor, which
accounts for populations and sensitive environments located
near the contaminated area. The EPA began by selecting an
“appropriate benchmark” for sorting workers into two tiers.
Level I applies to individuals who occupy structures where the
concentration of hazardous substances equals or exceeds the
health benchmark. 40 C.F.R. Part 300, App. A § 5.2.1.3.1 &
Table 5-20. Level II applies to individuals who are in
structures or subunits where there is an observed exposure, but
where the concentration of hazardous substances falls below
the benchmark.

focused on the 217 full-time employees
working at the Rockwell Facility at the time of the agency’s
decision. Applying residential exposure assumptions, the EPA
adopted a benchmark of 0.4 µg/m3 for cancer risk based on
TCE exposure and 2.0 µg/m3 for non-cancer toxicological
risk.2 Sorting workers into these two categories yielded a score
of 707.33 for the “targets” category.

      Based on a complicated formula that we need not navigate,
the EPA translated those three subcomponent scores into an
overall hazard score of 50 for the Rockwell Site. That meant
that based on the subsurface intrusion component alone, the
Rockwell Site was well above the 28.5 threshold score for
listing, and the EPA added it to the National Priorities List.
Given that outcome, the EPA had no occasion to analyze the


    2
     The EPA measures the concentration of toluene, TCE, and
DCE in micrograms per cubic meter (μg/m3).
                                8
Rockwell Site’s potential listing under the air, surface water, or
groundwater pathways.

                               II

      We have jurisdiction to review the EPA’s listing decisions
under 42 U.S.C. § 9613(a). Meritorʼs admission that it has
taken on former site owner Rockwell International’s liability
with respect to the Rockwell Site, Oral Arg. Tr. 4:25–5:3, is
sufficient to establish the company’s standing to challenge the
listing as a possible Potentially Responsible Party. See Mead
Corp. v. Browner, 

, 155 (D.C. Cir. 1996) (finding
standing where the corporation’s status as a former owner of
the property in question “would provide a plausible basis for a
claim that it was a [Potentially Responsible Party]”); see also
CTS 

.

     In reviewing the EPA’s listing decision, we borrow the
Administrative Procedure Act’s standard of review. Genuine
Parts Co. v. EPA, 

, 311 (D.C. Cir. 2018)
(reviewing listing decision under the APA standard because
“CERCLA does not specify a standard of review applicable to
EPA’s NPL listing decisions”). As a result, the EPA’s decision
will be set aside only if it was “arbitrary, capricious, an abuse
of discretion, or otherwise not in accordance with law.” Carus
Chem. v. EPA, 

, 441 (D.C. Cir. 2005) (quoting 5
U.S.C. § 706(2)(A)).

     In conducting this review, we afford the EPA “significant
deference” with respect to the “highly technical issues
involved[.]” Carus 

(quoting Bradley
Mining Co. v. EPA, 

, 1359 (D.C. Cir. 1992)). We
are also mindful that the National Priorities List is meant to be
a “rough list of priorities, assembled quickly and
inexpensively.”

972 F.2d at
1359); see also Eagle-Picher Indus., Inc. v. EPA, 

,
                               9
911 (D.C. Cir. 1985) (The “major purpose” of the National
Priorities List and the Hazard Ranking System “is to identify,
quickly and inexpensively, sites that may warrant further action
under CERCLA.”).

                              III

     In adding the Rockwell Site to the National Priorities List,
the EPA followed the well-established Hazard Ranking System
process. The validity of that regulatory framework is not in
dispute. Oral Arg. Tr. 7:15–8:8, 44:10–17. Meritor also does
not challenge the Hazard Ranking System’s pathway-based
approach, nor does it take issue with the inclusion of a
subsurface intrusion component within the soil exposure
pathway. Instead, Meritor argues that the EPA improperly
applied the Hazard Ranking System to the Rockwell Site by
(1) failing to account for the company’s mitigation efforts,
(2) relying on improper benchmarks when analyzing the
“targets” component, and (3) failing to follow the Hazard
Ranking System’s tiered approach to analyzing the “waste
characteristics” component. None of these arguments succeed.

                               A

     At the outset, Meritor argues that the EPA acted arbitrarily
and capriciously by evaluating the Rockwell Site based on
measurements taken before the sub-slab depressurization
system was installed.

     This court previously “rejected the argument that the EPA
must consider the effects of remedial measures in scoring a site
under” the 1982 Hazard Ranking System. Eagle-Picher Indus.,
Inc. v. EPA, 

, 149 (D.C. Cir. 1987); see also
Linemaster Switch Corp. v. EPA, 

, 1306–1307
(D.C. Cir. 1991). But, according to Meritor, the “express
language of the” current Hazard Ranking System strips away
                               10
the EPA’s discretion to disregard remedial measures. Meritor
Br. 22, 24–26; Reply Br. 8; see also Oral Arg. Tr. 7:15–25.
Rather, Meritor argues, the operative Hazard Ranking System
regulations require consideration only of toxic emissions
occurring after the mitigation system was installed.

     In support of this position, Meritor points to two portions
of the Hazard Ranking System that expressly account for the
presence of mitigation measures. Meritor Br. 24 (citing 40
C.F.R. Part 300, App. A §§ 5.2.1.1.2.1 & Table 5-12,
5.2.1.3.2.3 & Table 5-21). But the EPA had no need to reach
or to rely on either of those provisions in this case.

     First, Meritor is correct that, when assessing the potential
for exposure (in the course of determining the likelihood of a
release of hazardous waste), the EPA assigns each building a
structural containment value between 0 and 10 based on a
number of factors. 40 C.F.R. Part 300, App. A § 5.2.1.1.2.1 &
Table 5-12. One of those factors is whether a mitigation
system has been installed.

had no occasion to evaluate the potential for
exposure (and so to consider Meritor’s installation of a sub-slab
depressurization system) because the agency documented an
actual, observed exposure at the site. See 40 C.F.R. Part 300,
App. A § 5.2.1.1.1. Sensibly enough, the regulations do not
require calculating the “potential” for exposure when the
reality of actual exposure has already been documented. Res
ipsa loquitur. Nor do the regulations factor in abatement
efforts when evaluating whether there has been an observed
exposure. See

2-3, 5.2.1.1.1. That direct
observation is why, under the regulations, the EPA
automatically assigned the maximum score of 550 for the
“likelihood of release” component without regard to mitigation
measures. See

     Second, the Hazard Ranking System requires the EPA,
when it assesses the “population within an area of subsurface
contamination” component of the “targets” metric, to weight
the number of occupants based on whether the building has a
mitigation system installed. 40 C.F.R. Part 300, App. A
§ 5.2.1.3.2.3 & Table 5-21; J.A. 4.

     But in this case, the EPA never relied on the “population
within an area of subsurface contamination” factor, so it had no
occasion to look at mitigation measures through that lens. And
by excluding those points from the computation, the EPA’s
analysis necessarily resulted in a lower or equal overall score
for the “targets” metric. See 40 C.F.R. Part 300, App. A § 5.2
Table 5-11. Meritor can hardly complain about the EPA’s
failure to look at other factors that would have, at best, left the
Rockwell Site’s final score right where it was and, at the worst
for Meritor, resulted in an even higher score.

     Third, Meritor backs off in its reply brief from its initial
assertion that the listing decision’s failure to consider the
mitigation measure violated the “the express language of the
[Hazard Ranking System],” Meritor Br. 22. Instead, Meritor
argues in reply that the regulations’ sporadic references to
mitigation systems in some factors implicitly mandate the
consideration of mitigation systems at every step and for every
factor in the analysis, Reply Br. 4.

     But that would amend rather than apply the existing
regulatory scheme. Nothing in the text of the regulations
imposes such a pervasive requirement to factor in mitigation
efforts. To the contrary, the Hazard Ranking System’s
selective inclusion and omission of mitigation systems as a
consideration suggests “that the omission” of any reference to
mitigation systems in other “context[s] was deliberate.”
Council for Urological Interests v. Burwell, 

, 221
                               12
(D.C. Cir. 2015) (applying that principle in a statutory
construction case); see also Yonek v. Shinseki, 

,
1359 (Fed. Cir. 2013) (“Where an agency includes particular
language in one section of a regulation but omits it in another[,]
[courts] generally presume[] that the agency acts intentionally
and purposely in the disparate inclusion or exclusion.”)
(formatting modified); Atlas Tel. Co. v. Oklahoma Corp.
Comm’n, 

, 1265 (10th Cir. 2005) (similar).

    It certainly was not arbitrary and capricious for the EPA to
apply its regulations as written.

                                B

     Meritor next argues that the EPA acted arbitrarily by
relying on a residential health benchmark when evaluating the
“targets” metric.

     The Hazard Ranking System’s categorization of occupied
structures is multilayered. Bear with us.

     The EPA designates any structure with an observed
exposure as Level I if the hazardous substance’s concentration
inside equals or exceeds the “appropriate benchmark[.]” 40
C.F.R. Part 300, App. A § 5.2.1.3.1. If the concentration of a
hazardous substance triggers an observed exposure, “but do[es]
not equal or exceed the appropriate benchmark[,]” the agency
designates the structure as Level II.

     The relevant benchmarks are the “health-based
benchmarks from Table 5-20.” 40 C.F.R. Part 300, App. A
§ 5.2.1.3.2. That Table directs the agency to use the

    3
      This same measurement scheme applies to partitioned subunits
within a structure. 40 C.F.R. Part 300, App. A § 5.2.1.3.1.
                                  13
“[s]creening concentration for cancer corresponding to” a one-
in-a-million risk of cancer.

5-20. It also
instructs the EPA to select a benchmark “for noncancer
toxicological responses” caused by oral or inhalation exposure.

categorizes each building that has an
observed exposure as either Level I or Level II, it divides the
number of occupants in each structure by three if they are full-
time workers or by six if they are part-time workers. 40 C.F.R.
Part 300, App. A §§ 5.2.1.3.2.1, 5.2.1.3.2.2. The resulting
number is then multiplied by 10 for the Level I category.

Level II figure is not multiplied by ten.

modified Level I and Level II values are
summed, along with the “population within the area(s)” rating,
to calculate the “population” score for the “targets” metric.

selecting the appropriate Hazard Ranking System
cancer and non-cancer risk benchmarks, the EPA relies on an
exposure scenario “consistent with a residential
individual * * * across all * * * pathways[,] as this is most
protective.” J.A. 648. Such residential health benchmarks are
based on the assumption that the occupants are exposed to the
hazardous substance “24 hours per day” and “350 days per
year” for a duration of 26 years. J.A. 652. Applying this
approach, the EPA selected TCE benchmarks of 0.4 μg/m3 for



     4
       The “population within the area(s)” subcomponent is typically
added to the modified Level I and Level II values to obtain the overall
population score. 40 C.F.R. Part 300, App. A § 5.2.1.3.2.4. The
EPA did not, however, calculate that subcomponent for the Rockwell
Site, making that score effectively zero.
                                14
cancer risk and 2.0 μg/m3 for the risk of a toxicological
response.5

     In Meritor’s view, the Hazard Ranking System’s mandate
that the EPA select an “appropriate benchmark,” 40 C.F.R.
Part 300, App. A §§ 5.2.1.3.1, 5.2.1.3.2, requires the agency to
adopt site-specific exposure assumptions for cancerous and
toxic health risks. Meritor Br. 38–41. In particular, Meritor
argues, the EPA should have used an industrial, rather than
residential, health benchmark because the employees did not
reside at the Rockwell Facility full time. Rather, they worked
“8-hour shifts” for “five to seven days per week.” J.A. 41. So
Meritor advocates for the assumption that workers are exposed
8 hours per day and 250 days per year for 25 years. Meritor
Br. 37 (citing J.A. 652).

     Meritor’s argument is unsupported by either the text or the
structure of the Hazard Ranking System.

     First, nothing in the text of the Hazard Ranking System
expressly instructs the EPA to use site-specific exposure
assumptions on the front end of the process when it is selecting
the appropriate health benchmarks. The regulation, instead,
instructs the EPA to use “health-based benchmarks” as set out
in Table 5-20. 40 C.F.R. §§ 5.2.1.3.1, 5.2.1.3.2 & Table 5-20.
That Table directs the EPA to consider both toxicological and
carcinogenic benchmarks and, for cancer-related risks,
establishes a one-in-a-million risk of cancer as the appropriate
threshold.

regulation thus makes
no mention of site-specific characteristics, such as residential

    5
      The EPA found that the DCE and toluene concentrations in the
main building did not exceed the selected residential benchmarks, so
the choice between a residential and an industrial benchmark was of
no consequence as to those substances.
                              15
or industrial use. Nor does it mandate that the EPA adopt any
particular exposure assumptions.

     Second, the Hazard Ranking System accounts for the
lower exposure faced by workers relative to full-time residents
on the back end of its calculation. Once the occupants are
sorted into either Level I or Level II, the EPA divides the
number of people by three if they are full-time workers and by
six if they are part-time workers. See 40 C.F.R. Part 300,
App. A §§ 5.2.1.3.2.1, 5.2.1.3.2.2. This weighting accounts for
the fact that full-time workers and part-time workers spend
approximately one-third or one-sixth of the day at work
respectively. UNITED STATES EPA, TECHNICAL SUPPORT
DOCUMENT FOR U.S. EPA’S FINAL RULE: ADDITION OF A
SUBSURFACE INTRUSION COMPONENT TO THE HAZARD
RANKING SYSTEM 62 (2016) (“2016 TECHNICAL SUPPORT
DOCUMENT”).

     So at bottom, because the Hazard Ranking System already
accounts for the workers’ reduced hours of exposure relative to
residents, the EPA reasonably relied on residential health
exposure assumptions when selecting the appropriate health
benchmarks. In fact, because the Hazard Ranking System
requires those divisions by three or six regardless of which
health benchmark is used, 40 C.F.R. Part 300, App. A
§§ 5.2.1.3.2.1, 5.2.1.3.2.2, if the EPA had used an industrial
health benchmark as Meritor proposes, it would have twice
reduced the “targets” score based on worker status. Nothing in
the Hazard Ranking System endorses, much less mandates,
such double discounting.

    Beyond that, the EPA’s use of residential exposure
assumptions is reasonable.

    For starters, when making listing decisions, the EPA
uniformly uses residential exposure assumptions because it
                               16
favors a more “conservative (i.e. protective) approach[.]”
J.A. 650. Erring on the side of caution in evaluating the
toxigenic effects of hazardous substances on people is not
unreasonable. See Board of Regents of the Univ. of Wash. v.
EPA, 

, 1219 (D.C. Cir. 1996) (“It is not on its face
unreasonable for the EPA to strike the balance by erring on the
side of over-inclusion at the listing stage and on the side of
under-inclusion at the remedial phase[.]”); see also Carus

(The National Priorities List is meant
to be a “rough list of priorities, assembled quickly and
inexpensively.”) (internal quotation marks omitted).

     In addition, when selecting health benchmarks, the EPA
generally uses uniform residential exposure assumptions
across all sites and pathways. J.A. 648–650. That ensures that
the Hazard Ranking System properly measures the “relative
rather than absolute risk” of different hazardous waste sites in
identifying the priority sites. 40 C.F.R. Part 300, App. A § 1.0;
see also 42 U.S.C. § 9605(a)(8)(B) (National Priorities List is
intended to identify sites that are a “priorit[y]” for remedial
action).

    Finally, by using the more conservative benchmark, the
agency accounts for both present and possible “future land-use
conditions.” J.A. 650 (The EPA assumes “long-term/chronic
exposures” because that is “the reasonable maximum
exposure * * * expected to occur under both current and future
land-use conditions.”).

     Because the EPA has reasonable policy reasons for
starting out with a residential health benchmark, and its
analysis properly adjusted the Rockwell Site’s score to account
for workers’ reduced hours of exposure, the agency’s use of a
residential health benchmark in calculating the “targets” score
passes muster.
                              17
                               C

     Finally, Meritor argues that the EPA incorrectly calculated
the “waste characteristics” component of the subsurface
intrusion pathway. The “waste characteristics” score has two
components: “toxicity/degradation” and “waste quantity.” 40
C.F.R. Part 300, App. A §§ 2.4.2, 5.2.1.2. Meritor’s challenge
aims at the latter—EPA’s measurement of the waste quantity.
Meritor Br. 47–51.

    The Hazard Ranking System creates a tiered approach to
calculating the quantity of waste in structures.

    Under Tier A, the EPA calculates the “mass of constituents
found in [a] regularly occupied structure[] where [an] observed
exposure has been identified.” 40 C.F.R. Part 300, App. A
§§ 2.4.2.1.1, 5.2.1.2.2. If the EPA is able to estimate the
quantity of waste under this method with “reasonable
confidence,” the waste quantity inquiry ends there.

5.2.1.2.2. If not, the EPA turns to
Tier B.

    Under Tier B, the EPA must calculate the “flow-through
volume” of the structure. 40 C.F.R. Part 300, App. A
§§ 2.4.2.1.1, 2.4.2.1.2, 5.2.1.2.2. That is the amount of
hazardous substances “that flows into the structure from the
subsurface.” 2016 TECHNICAL SUPPORT 

.

     If that computation proves infeasible, the EPA moves to
Tier C, where it must estimate the volume of occupied portions
of the structure. 40 C.F.R. Part 300, App. A §§ 2.4.2.1.3,
5.2.1.2.2; see also 2016 TECHNICAL SUPPORT 

. From this, the agency can estimate the “possible
amount of hazardous substances” in occupied areas of the
                                18
building. See 2016 TECHNICAL SUPPORT 

.

     And if all else fails, the EPA turns to Tier D, which allows
the agency to derive a waste quantity score based on the
structure’s floor area. 40 C.F.R. Part 300, App. A §§ 2.4.2.1.3,
2.4.2.1.4, 5.2.1.2.2. The EPA applied Tier D to calculate the
waste quantity in the Main Building at the Rockwell Site.

     Meritor claims that the EPA should have used Tier A to
calculate waste quantity once the sub-slab depressurization
system was installed. This was so, the company argues,
because the sub-slab depressurization system ensured that the
concentration of TCE, DCE, and toluene stayed within a
narrow range. That, in turn, eliminated the high variance in
concentrations that the EPA cited as an obstacle to calculating
the mass of constituents with “reasonable confidence,” as
Tier A requires. J.A. 35. Meritor also argues that the EPA
erroneously stated that it needed air flow data to calculate the
mass of hazardous substances under Tier A.

     Both of those arguments are forfeited because Meritor did
not raise them before the EPA. The law is settled that those
who challenge a National Priorities List placement “must
present their claims clearly and specifically to the agency
before raising them in a petition for review.” See CTS 

(quoting Kent County v. EPA, 

, 399
(D.C. Cir. 1992)); see also Honeywell Int’l, Inc. v. EPA, 

, 449, 451 (D.C. Cir. 2004); Linemaster 

.

      Meritor does not dispute its duty to present its challenges
first to the EPA. It argues instead that it sufficiently raised its
objection to the Tier D analysis by informing the agency that
“it had violated rules ‘requiring EPA to consider an active
vapor mitigation system * * * when calculating the hazardous
                              19
waste quantity.’” Reply Br. 24 (omission in original) (quoting
J.A. 48 n.7); see also Oral Arg. Tr. 31:5–23.

     That argument, though, only advocated for consideration
of the sub-slab depressurization system somewhere in the
process of the “waste characteristics” analysis. It did not put
the EPA on notice of Meritor’s specific objection to the tier
used in the agency’s waste quantity analysis. See CTS 

(“[P]arties opposing NPL listing must present
their claims clearly and specifically to the agency before
raising them in a petition for review.”); Carus 

(Parties seeking to challenge agency action must first
raise their objections with “reasonable specificity during the
period for public comment[.]”) (internal quotation marks
omitted).

                              IV

    For all of those reasons, the petition for review is denied.

                                                    So ordered.