Summit Petroleum Corp. v. United States Environmental Protection Agency

690 F.3d 733, 42 Envtl. L. Rep. (Envtl. Law Inst.) 20167, 2012 WL 3181429, 75 ERC (BNA) 1129, 2012 U.S. App. LEXIS 16345

• Court: Court of Appeals for the Sixth Circuit
• Decided: August 7, 2012
• Docket: 09-4348, 10-4572
• Status: Published

Opinions

SUHRHEINRICH, J., delivered the opinion of the court, in which CLAY, J., joined. MOORE, J. (pp. 751-57), delivered a separate dissenting opinion.

OPINION

SUHRHEINRICH, Circuit Judge.

This case arises from a final action of the Environmental Protection Agency (EPA) determining that a natural gas sweetening plant and various sour gas production wells commonly owned by Petitioner Summit Petroleum Corporation (Summit) and separately located within an area of approximately forty-three square miles constitute a single stationary source under the EPA’s Clean Air Act Title V permitting program. Specifically at issue is the EPA’s conclusion that Summit’s facilities satisfy the regulatory requirement of being “located on ... adjacent properties” because, although physically independent, they are “truly interrelated.” Summit, together with Amici American Petroleum Institute and American Exploration and Production Counsel, argues that the EPA’s determination that the physical requirement of adjacency can be established through mere functional relatedness is unreasonable and contrary to the plain meaning of the term “adjacent.” We agree. For these and other reasons fully set forth herein, we VACATE the EPA’s final determination and REMAND this case to the EPA to determine whether Summit’s sweetening plant and sour gas wells are sufficiently physically proximate to be considered “adjacent” within the ordinary, i.e., physical and geographical, meaning of that requirement.

I. Background

A. Summit’s Natural Gas Production Facilities

Summit is a producer of natural gas that owns and operates a natural gas sweetening plant in Rosebush, Michigan. Summit’s plant “sweetens” the “sour” gas from approximately one hundred sour gas production wells by removing hydrogen sulfide so that the gas can be used. Summit owns all of the production wells and the subsurface pipelines that connect each of the wells to the sweetening plant. The wells themselves are located over an area of approximately forty-three square miles at varying distances from the plant — from five.hundred feet to eight miles away — and Summit does not own the property between the individual well sites or the property between the wells and the plant. None of the well sites share a common boundary with each other, nor do any of the well sites share a common boundary with Summit’s production plant. Flares work as part of the plant operations by burning off natural gas waste to relieve pressure on the gas collection equipment. The closest flare is located approximately one half-mile from the plant, while the remaining flares are each over one 'mile away.

The sweetening plant and the majority of Summit’s gas production wells and flares are located within the territory of Michigan’s Saginaw Chippewa Indian Tribe’s Isabella Reservation. The sweetening plant, gas production wells, and flares emit sulfur dioxides and nitrous oxides, air pollutants subject to regulation under the Clean Air Act (CAA). The plant alone emits, or has the potential to emit, just under one hundred tons of these pollutants per year. Each flare and each well site emits, or has the potential to emit, much lower amounts of pollutants. However, if the emissions of sulfur dioxide from the plant and any one production well were to be combined, they would exceed one hundred tons of pollutants per year.

B. The EPA’s Title V Permitting Program

In 1970, Congress passed the CAA, 42 U.S.C. §§ 7401-7671q, “to protect and enhance the quality of the Nation’s air resources so as to promote the public health and welfare and the productive capacity of its population....” 42 U.S.C. § 7401(b)(1) (2006). In 1990, Congress enacted Title V of the CAA (Title V), 42 U.S.C. §§ 7661-7661f, establishing an operating permit program to regulate stationary sources of air pollution. Rather than imposing new substantive air quality control requirements, Title V supplements the CAA by requiring regulated sources of air pollution to obtain operating permits that include emission limitations, standards, monitoring requirements, compliance schedules, and other conditions necessary to assure compliance with the CAA. See 42 U.S.C. §§ 7661a(a), 7661c(a); Ohio Pub. Interest Research Grp., Inc. v. Whitman, 386 F.3d 792, 794 (6th Cir.2004). Although “[sjtates and local governments bear ‘primary responsibility’ ” in administering Title V regulations, Whitman, 386 F.3d at 794 (citing 42 U.S.C. § 7401(a)(3)), certain circumstances permit or require the EPA itself to establish and promulgate Title V regulations in lieu of the EPA-approved state plans. In the absence of an EPA-approved plan regulating Indian territory located within a given state, for example, the EPA is authorized to implement a federal air quality control plan within that jurisdiction. Id. § 7601(d)(4). In this case, the EPA has asserted jurisdiction to regulate Indian territory within the State of Michigan, including the Saginaw Chippewa Indian Tribe’s Isabella Reservation. See, e.g., Approval and Promulgation of Air Quality Implementation Plans; Michigan; PSD Regulations (PSD Regulations), 73 Fed. Reg. 53,366 (Sept. 16, 2008) (to be codified at 40’ C.F.R. pt. 52) (“Michigan is not authorized to carry out its Federally approved air program in ‘Indian Country,’ as defined in 18 U.S.C. 1151.”).

Title V requires every “major source” of air pollution to obtain a Title V operating permit. 42 U.S.C. § 7661a(a). The EPA’s Title V plan defines a major source to include “any stationary facility or source of air pollutants which directly emits, or has the potential to emit, one hundred tons per year of any pollutant,” including nitrous oxides and sulfur dioxides (the pollutants emitted by Summit’s operations). See 42 U.S.C. § 7602©; see also 40 C.F.R. § 71.2. The EPA defines a “stationary source” as “any building, structure, facility, or installation which emits or may emit a regulated [air] pollutant.” 40 C.F.R. § 52.21(b)(5). Multiple pollutant-emitting activities, such as Summit’s sweetening plant and sour gas production wells, can be aggregated together and considered a “building, structure, facility, or installation,” i.e., a single stationary source, under Title V only if they: (1) are under common control; (2) “are located on one or more contiguous or adjacent properties”; and (3) belong to the same major industrial grouping. 40 C.F.R. § 71.2.¹ If the pollutant-emitting activities fail to satisfy any one of these three criteria, they are considered separate stationary sources and their emissions cannot be aggregated to meet the major source threshold for which a Title V permit is required.

C. Major Source Determination Correspondence Between Summit and the EPA

In January 2005, Summit, together with the Michigan Department of Environmental Quality (MDEQ), submitted a request to the EPA to determine whether Summit’s facilities met the definition of a Title V major source of air pollution.² Summit’s request correctly noted that emissions from its sweetening plant were insufficient to constitute it a major source, but “if potential emissions from the sour gas production wells are included with the gas sweetening plant emissions, the aggregated sources combined may be considered a single major source for criteria pollutants, namely nitrogen oxides and sulfur dioxides.” Summit’s request suggested that aggregation of its facilities would be improper for two principal reasons. First, Summit’s wells were “located at great distances from its production facility on entirely different tracts, leases and surface sites,” making them neither contiguous nor adjacent to one another. Second, the EPA had not distinguished between the definition of a “facility” found in its hazardous air pollutant (HAP) regulations from the definition of a “facility” for criteria pollutant (Title V) purposes. The EPA’s HAP regulations specifically considered aggregation of multiple facilities in the oil and gas industry and concluded that aggregation was improper: “Emissions from any oil or gas exploration or production well, with its associated equipment, and emissions from any pipeline compressor or pump station shall not be aggregated with emissions from other similar units, whether or not these units are in a contiguous area under common control....” See Article II, Part 55, Air Pollution Control, of the Michigan Natural Resources and Environmental Protection Act, 1994 PA 451, as amended (Act 451), and 1980 AACS R 336.1211 (Rule 211) (emphasis added).

The EPA responded to Summit’s source determination request on April 26, 2007, confirming that it was asserting jurisdiction to regulate Summit’s facilities within the Isabella Reservation but stating that it was “unable to conclude if the wells and plant constituted a single source for Title V purposes.” The EPA distinguished its Title V and HAP regulations, maintaining that pollutant-emitting facilities could be aggregated to constitute a single stationary source under Title V if they operated under common control, .were located on contiguous or adjacent properties, and belonged to the same industrial grouping. It then identified the proper procedure for determining a single stationary source in the oil and gas industry, outlined in an EPA guidance memorandum issued by Acting Assistant Administrator William Wehrum:

1) Identify the surface sites,

2) Evaluate whether each individual surface site qualifies as a separate stationary source, or if the emissions activities should be grouped together to form a stationary source (i.e., if their emission activities are under common control, belong to the same major industrial grouping, and are contiguous or adjacent), and

3) Aggregate two or more stationary sources if they are under common control and are located in proximity of each other.³

Memorandum from William L. Wehrum, Acting Assistant Adm’r, to Reg’l Adm’rs IX (Jan. 12, 2007) available at http://www. eenews.net/public/25/12769/features/ documents/2009/10/13/document_pm_02. pdf (last visited May 16, 2012) (Wehrum Memorandum).

The EPA found that based on the information provided by Summit’s request, its wells and production plant were undisputedly commonly owned, part of the same industrial grouping, and not located on contiguous surface sites. Accordingly, the agency concluded that it “must evaluate the adjacency, (i.e., the nearness or closeness) of the sour gas wells to the sweetening plant” to determine whether Summit’s facilities constituted a single stationary source subject to regulation. Although it noted that “because the production wells supply the gas to the gas sweetening plant, we believe that the sites meet the common sense notion of a plant,” the EPA found that the Wehrum Memorandum “makes clear that proximity is the most informative factor in making a source determination involving oil and gas activities.” Finally, the EPA requested additional information — a map of Summit’s facilities and a schematic of its processes that showed the connection between Summit’s “exploration and production activities” — to assist it in reaching a decision.

Summit provided the requested information in a letter dated April 24, 2008. In the letter, Summit agreed that under the Wehrum Memorandum, proximity was the most crucial factor in making source determinations for the oil and gas industry. Summit reminded the EPA of the Wehrum Memorandum’s additional comments regarding the proximity factor, e.g., the EPA’s position that it was unreasonable to aggregate wells and production field activities that “occur over large geographic distances with the downstream processing plant into a single stationary source.” See id. Summit submitted that the concepts of contiguousness and adjacency should assess “whether the land associated with the pollutant-emitting activity is connected to, or is nearby, land associated with another pollutant-emitting activity.” In support, Summit pointed again to the Wehrum Memorandum, which advised that “[a] reviewing authority can consider two surface sites to be in close proximity if they are physically adjacent, or if they are separated by no more than a short distance (e.g., across a highway, separated by a city block or some similar distance).” Id. Sum-, mit noted that its wells and production plant were separated by much larger distances than those contemplated by the Wehrum Memorandum and reiterated its request that the EPA not consider them a single stationary source under Title V.

In July 2009, the EPA responded that even with Summit’s supplemental materials, it was again unable to conclusively determine whether Summit’s facilities com stituted a single stationary source. The EPA addressed Summit’s emphasis on the Wehrum Memorandum and stressed that physical proximity was not the only factor in adjacency determinations, stating that “generally, the EPA determines whether two facilities are ‘adjacent’ based on a ‘common sense’ notion of a source and the functional interrelationship of the facilities, rather than simply on the physical distance between the facilities.” The EPA concluded by requesting that Summit contact it as soon as possible to further “discuss the requirements of the ... Title V permit applications for sources located on tribal lands.”

The parties engaged in a conference call in July 2009. According to Summit, the EPA indicated that it might no longer be using the Wehrum Memorandum as guidance in Summit’s source determination. The EPA said it would investigate the status of the Wehrum Memorandum and gather information on permit conditions for other sources before making a final source determination.

1. The EPA determines Summit’s facilities to be a single and major source under Title V.

On September 8, 2009, EPA issued its determination that Summit’s sweetening plant and sour gas wells constituted a single stationary source, and were therefore a major source, under Title V. The EPA did not explain the basis for its decision, save for citing to its prior April 26, 2007 and April 16, 2009 letters to Summit. Shortly after the EPA issued its decision, it forwarded Summit a September 22, 2009 guidance memorandum prepared by the EPA’s Office of Air and Radiation Assistant Administrator Gina McCarthy entitled “Withdrawal for Source Determinations for Oil and Gas Industries” (McCarthy Memorandum). See Memorandum from Gina McCarthy, Assistant Adm’r, to Reg’l Admr’s Regions I-X (Sept. 22, 2009), available at http://www.eenews.net/public/ 25/12769/features/doeuments/2009/10/13/ doeument_pm_01.pdf (last visited May 16, 2012). In pertinent part, the McCarthy Memorandum reasoned that the simplified approach of the Wehrum Memorandum, with its emphasis on physical proximity in the determination of what constituted a stationary source, was no longer appropriate. The McCarthy Memorandum called permitting authorities to instead “rely foremost on the three regulatory criteria for aggregating emissions from multiple sources, ... (1) whether the activities are under the control of the same person (or person under common control); (2) whether the activities are located on one or more contiguous or adjacent properties; and (3) whether the activities belong to the same industrial grouping.” Id. (citing 40 C.F.R. § 52.21(b)(6)). The guidance memorandum nevertheless conceded that stationary source determinations should be made on a case-by-ease basis and that “in some cases, ‘proximity’ may serve as the overwhelming factor in a permitting authority’s source determination decision.” Id.

In response to the EPA’s September 8, 2009 determination, Summit filed its initial Petition for Review with this Court on November 4, 2009. On January 5, 2012, Summit sent a letter to the EPA requesting an administrative stay of the deadline the EPA had imposed on Summit to obtain a Title V operating permit so that Summit could submit additional information for the EPA’s consideration. The EPA granted Summit’s request and the parties submitted a joint motion for abeyance to this Court on March 4, 2010. This Court granted the motion on March 18, 2010, holding the case in abeyance and vacating the briefing schedule to allow the parties time to resolve their pending administrative issues.

On October 18, 2010, the EPA sent Summit a final letter confirming that even with Summit’s supplemental information, the EPA considered Summit’s facilities a single stationary source, and thus a major source, under Title V. The EPA explained that the McCarthy Memorandum directed it to consider all three regulatory factors. On the issue of the facilities’ adjacency, the EPA indicated that it had never established a specific distance between pollutant-emitting entities at which they were no longer adjacent. Id. The EPA reiterated that, historically, factors such as the “nature of the relationship between the facilities” and the “degree of interdependence between them” had been important to the question of whether two facilities were adjacent. The agency discussed its view that Summit’s plant, wells, and flares worked together as a single unit that “together produced a single product” and found that Summit had failed to provide any evidence to demonstrate that the emissions sources were not “truly interdependent.” The EPA thus concluded that given this functional interrelationship, Summit’s facilities “should not be considered separate emission sources.”

Summit filed a second Petition for Review in this Court on October 18, 2010. The EPA opposed, and this Court denied, a motion to intervene filed by American Petroleum Institute (API). This Court granted the unopposed motions of API and American Exploration and Production Council (AXPC) (collectively Amici) to submit amicus briefs in the case.

II. Analysis

A. Standard of Review

Section 307(b) of the CAA, 42 U.S.C. § 7607(b), does not specify a standard for judicial review of an EPA final action. Accordingly, this Court reviews the EPA’s source determination under the judicial review provisions of the Administrative Procedure Act (APA). 5 U.S.C. § 706(2)(A). See Alaska Dep’t of Envtl. Conservation v. E.P.A., 540 U.S. 461, 496-97, 124 S.Ct. 983, 157 L.Ed.2d 967 (2004). The APA requires us to credit an agency action unless it is “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). Where, as here, we review an agency’s interpretation of its own regulation and not its application of a statute passed by Congress, we must defer to the agency’s interpretation unless it is “plainly erroneous or inconsistent with the regulation.” Auer v. Robbins, 519 U.S. 452, 461, 117 S.Ct. 905, 137 L.Ed.2d 79 (1997) (internal quotation marks and citations omitted). We afford an agency’s interpretation no deference, however, if the language of the regulation is unambiguous, for doing so would “permit the agency, under the guise of interpreting a regulation, to create de facto a new regulation.” Christensen v. Harris Cnty., 529 U.S. 576, 588, 120 S.Ct. 1655, 146 L.Ed.2d 621 (2000). If the regulation is ambiguous and deference is due, we have still noted that “deferential” review is not inconsequential, Ky. Waterways Alliance v. Johnson, 540 F.3d 466, 474 (6th Cir.2008), and we must be satisfied that the agency’s action minimally involved a “rational connection between the facts found and the choice made.” Motor Vehicle Mfrs. Ass’n of the U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983).

The issue in the instant case is relatively straightforward. Summit, the EPA, and Amici agree that the EPA has properly exercised jurisdiction to regulate the tribal lands on which Summit’s facilities are located. They agree that Summit’s production plant and sour gas wells are commonly owned and share a major industrial grouping, satisfying two parts of the three-part test for a stationary source under the EPA’s Title Y permitting program. All parties further agree that if Summit’s plant and wells are aggregated into a single stationary source, the total emissions from this source would exceed the threshold required for a stationary source to be considered a major source, and Summit would be required to obtain a Title V permit to regulate its emissions of criteria pollutants. Finally, the parties agree that Summit’s production plant and sour gas wells are not located on “contiguous,” i.e., bordering, properties. Undisputedly, the sole disputed issue before us is whether Summit’s facilities are “adjacent” to one another, thus converting them into a stationary source, and a major source, under Title V.

Together with Amici, Summit argues that the EPA’s conclusion that Summit’s activities are “adjacent” because they are “truly interrelated” is unreasonable, and urges this Court to vacate the EPA’s determination that Summit’s plant and wells constitute a single stationary source. Specifically, Summit argues that the term “adjacent” is unambiguous, and that EPA’s interpretation of it defies its plain and ordinary meaning. We agree.

B. As used in the EPA’s Title V plan, the term “adjacent” is unambiguous.

Summit argues that the term “adjacent,” both generally and as it appears in the EPA’s regulatory requirement that aggregated activities be “located on contiguous or adjacent properties,” is unambiguous. The EPA disagrees, submitting, inter alia, that the term “adjacent” is unquestionably ambiguous because the EPA has never defined a specific physical distance by which it is defined or with which it is simultaneous. See Resp’t Br. at 30 (connecting the premise that “adjacent” is ambiguous because it lacks a definite distance with the conclusion that “[a]s applied to plant operations that vary enormously across the different industries regulated by the [CAA], [the term “adjacent”] is at minimum ambiguous as to whether emissions units that are part of an integrated industrial operation and located 500 feet to roughly eight miles apart may be considered to be located on ‘adjacent’ properties.... ”).

We conclude that both the dictionary definition and etymological history of the term “adjacent,” as well as applicable case law, support Summit’s position.

1. The dictionary definition of “adjacent” implies physical proximity.

This Court, and others as well, have often consulted dictionaries to ascertain the meaning of words. See, e.g., Terrell v. United States, 564 F.3d 442, 451 (6th Cir.2009) (citing MCI Telecomms. Corp. v. AT & T Co., 512 U.S. 218, 225, 114 S.Ct. 2223, 129 L.Ed.2d 182 (1994)). Our research satisfies us that dictionaries agree that two entities are adjacent when they are “[c]lose to; lying near ... [n]ext to, adjoining.” American Heritage Dictionary of the English Language, available at www. ahdictionary.com (search “adjacent”) (last visited May 16, 2012). The EPA does not cite, nor could we locate, any authority suggesting that the term “adjacent” invokes an assessment of the functional relationship between two activities. See, e.g., Merriam-Webster Dictionary, available at www.merriam-webster.com (search adjacent”) (last visited May 16, 2012) (“not distant: nearby <the city and adjacent suburbs >; having a common endpoint or border <adjacent lots> ...; immediately preceding or following); Oxford Dictionaries, available at www.oxforddictionaries. com (search “adjacent”) (last visited May 16, 2012) (“next to or adjoining something else; adjacent rooms; the area adjacent to the fire station ”).

Even the etymology of the term belies the EPA’s position that the term’s definition is even partially based on the contextual relationship, as opposed to the geographic location, of two activities. Here, our study reveals that the word “adjacent” originated in the early fifteenth century from the combination of syllables “ad” and “jet.” Online Etymology Dictionary, available at www.etymonline.com (search “adjacent”) (last visited May 16, 2012). “Ad” is a “prefix expressing direction toward or in addition to, ... with regard to, [or] in relation to ... space or time .... ” Id. (search “ad-”) (last visited May 16, 2012) (emphasis added). As incorporated in “adjacent,” the verb “jet” implies an action resulting in a certain physical resting place for an object: “to throw, cast ... with notion of ‘to cast (oneself) down.’ ” Id. (search: “adjacent” and “jet”) (last visited May 16, 2012). Thus, quite literally, two things are adjacent if they have been “throw[n], cast ... down ... in relation to [one another] ... in space.... ” See generally Online Etymology Dictionary, supra.

The EPA argues that though they are not incorrect, dictionaries provide an incomplete definition of “adjacent,” and that the functional interrelationship of two facilities must be considered because physical distance is meaningless without context. See Resp’t Br. at 27 (“It is only by examining the relative context that a reasonable determination can be made as to whether two objects are ‘close’ or ‘proximate’ to one another.”). The notion that adjacency expands and contracts with context, true as it is, must not be overextended. In the first instance, it is certainly correct that two states could be adjacent to one another in the context of a country, just as two houses could be adjacent in the context of a neighborhood. It is noted, however, that the contextual element of the overall space within which one asks whether two things are adjacent does not fundamentally change the physical and geographical nature of the question. The EPA makes an impermissible and illogical stretch when it states that one must ask the purpose for which two activities exist in order to consider whether they are adjacent to one another. Resp’t Br. at 26-27 (“A distance of one mile between two properties may be considered sufficiently small to be close to each other if one is driving between them; not so if one is trying to throw a softball from one to the other.”). Whether the distance between two facilities enables a given relationship to exist between them is immaterial to the concept of adjacency — it merely answers the question of whether a certain activity can or cannot occur between two locations that were, and will continue to be regardless of whether they host the activity, physically distant or physically adjacent.

Case law supports this understanding.

2. Case law supports the idea that adjacency relates only to physical proximity.

Several cases specifically address whether the term “adjacent” is ambiguous. While some courts conclude that “adjacent” is ambiguous in the limited sense of lacking an abstract definition, there is common recognition of the fact that adjacency is a purely physical and geographical, even if case-by-case, determination. United States v. St. Anthony R.R. Co. is an illustrative example. The EPA claims support in the case because of its statement that: “[adjacency] must be defined with reference to the context, at least to some extent[].” United States v. St. Anthony R.R. Co., 192 U.S. 524, 530, 24 S.Ct. 333, 48 L.Ed. 548 (1904). Ironically, the EPA disregards the context of .the Supreme Court’s assertion, which undisputedly cabined any purported ambiguity to purely physical dimensions: “[a]s the word [adjacent] is frequently uncertain and relative as to its meaning,” the Court reasoned, “it might ... be regarded as more extended when used with reference to a large object than with reference to a comparatively small one. In other words, it must be defined with reference to the context, at least to some extent.” Id. See also, e.g., Padula v. Lally, No. 160313, 1994 WL 16193907, at *2 (Mass.Land Ct. Feb. 9, 1994) (noting that all definitions of the word “adjacent” put before the court “suggest that ... two objects must be in reasonable proximity”); Syufy Enters, v. Home Ins. Co. of Ind., No. 94-0756, 1995 WL 129229, at *2 (N.D.Cal. Mar. 21, 1995) (stating that the “‘ordinary and popular’ reading of the term ‘adjacent’ denotes a sense of physical proximity”).

The most persuasive authority comes from the Supreme Court’s decision in Rapanos v. United States, 547 U.S. 715, 126 S.Ct. 2208, 165 L.Ed.2d 159 (2006). Similar to this case, Rayanos concerned a regulation enacted by an administrative agency, the Army Corps of Engineers (Corps), interpreting Congress’ legislation in the Clean Water Act (CWA). Id. at 723-24, 126 S.Ct. 2208. Specifically at issue in Rayanos was a provision in the CWA prohibiting . the discharge of pollutants in “navigable waters,” which the CWA further defined as “the waters of the United States, including the territorial seas.” Id. at 723, 126 S.Ct. 2208 (citing 33 U.S.C. § 1362(7)). In its regulations implementing the CWA, the Corps interpreted “the waters of the United States” to include, in pertinent part, “[w]etlands adjacent to ... waters [and tributaries] (other than the waters that are themselves wetlands[ ]).” Id. at 724, 126 S.Ct. 2208 (citing 33 C.F.R. § 328.3(a)(7)). In the action that was the impetus for suit, the Corps interpreted their own regulation to exercise jurisdiction over a private owner’s “sometimes saturated” parcel of land located eleven to twenty miles from the nearest body of navigable water, a proximity it deemed sufficiently “adjacent” to constitute it part of “the waters of the United States.” Id. at 720-21, 126 S.Ct. 2208.

The Sixth Circuit deferred to the Corps’ interpretation of its regulation, but the Supreme Court reversed, holding that deference was not due to “the Corps’ definition of ‘adjacent,’ which ... has been extended beyond reason....” Id. at 746, 126 S.Ct. 2208. In swift refutation of the dissent’s view that adjacency should be interpreted expansively, the majority held that the Court had previously used “ ‘adjacent’ and ‘adjoining’ interchangeably ” and had only previously concluded that a wetland was adjacent to a navigable body of water “ ‘[ ][b]ecause’ it ‘actually abutfted] on a navigable waterway.’ ” Id. at 747, 126 S.Ct. 2208 (some emphasis added). Finally, the Court noted that “[h]owever ambiguous the term may be in the abstract, ‘adjacent’ ... is not ambiguous between ‘physically abutting’ and merely ‘nearby.’ ” Id. at 748, 126 S.Ct. 2208.

Together with the ordinary and dictionary definitions of the term “adjacent,” the Rapanos decision, and similar case law, points clearly toward the conclusion that the regulatory requirement that aggregated activities be “located on contiguous or adjacent properties” is unambiguous in the context in which it is here considered. For the foregoing reasons, we adopt this position.

C. The EPA’s interpretation of the term “adjacent,” to which no deference is owed, runs contrary to its plain meaning.

Having determined that the word “adjacent” is unambiguous, we apply no deference in our review of the EPA’s interpretation of it. Christensen, 529 U.S. at 588, 120 S.Ct. 1655. Instead, we must test the agency’s interpretation against the plain meaning of its unambiguous regulation. Here, we conclude that the EPA’s interpretation of the requirement that activities be “located on contiguous or adjacent properties,” i.e., that activities can be adjacent so long as they are functionally related, irrespective of the distance that separates them, undermines the plain meaning of the text, which demands, by definition, that would-be aggregated facilities have physical proximity. The mere fact that in a limited sense, the term “adjacent” is not susceptible to abstract definition does not render it unclear, for “however ambiguous the term may be in the abstract, ‘adjacent’ ... is not ambiguous between ‘[physically proximate]’ and merely ‘[functionally related].’ ” See Rapanos, 547 U.S. at 748, 126 S.Ct. 2208. In light of this conclusion, we hold that the EPA has interpreted its own regulatory term in a manner unreasonably inconsistent with its plain meaning and vacate the EPA’s stationary source determination hereby directing the EPA to reassess the aggregation of Summit’s facilities under the ordinary understanding of its requirement that Summit’s plant and wells be located on adjacent, i.e., physically proximate, properties.

1. The EPA’s unreasonable interpretation is not entitled to heightened deference merely because it is longstanding.

The EPA opposes this conclusion, claiming that because it has an established history of supplementing the traditional definition of adjacency with the concept of activities’ functional relatedness, we must review its interpretation with heightened deference. Resp’t Br. at 31-32 (citing Alaska Dep’t of Envtl. Conservation, 540 U.S. at 487, 124 S.Ct. 983 (stating that courts “normally accord particular deference to an agency interpretation of ‘longstanding’ duration”) (quoting North Haven Bd. of Ed. v. Bell, 456 U.S. 512, 522 n. 12, 102 S.Ct. 1912, 72 L.Ed.2d 299 (1982))). The EPA’s appeal to the “longstanding duration” principle is unsurprising, since even a cursory overview demonstrates that the agency has rarely, if ever, considered physical proximity the sine qua non of two pollutant-emitting activities being “adjacent” to one another. See, e.g., Letter From Edward E. Reich, EPA to Clyde B. Eller, EPA Region IX (May 16, 1980), available at www.epa.gov/region7/air/nsr/ nsrmemos/shelhpdf (last visited May 16, 2012) (two facilities 1.8 miles apart and connected by pipeline are a single source because neither facility produces products independently); Letter from Edward E. Reich, EPA to Steve Rothblatt, EPA Region V (June 30, 1981), available at www. epa.gov/region7/air/title5/t5memos/defsrce 2.pdf (last visited May 16, 2012) (two plants one mile apart are a single source because they are connected by railroad and operated together); Letter from Robert G. Kellam to Richard R. Long, EPA Region VIII (Aug. 27, 1996), available at http://www.epa.gov/region07/air/nsr/ nsrmemos/abntpdf (last visited May 16, 2012) (landfarm and brewery six miles apart are single source because the land-farm operation is integral to the brewery operation); Letter from Richard R. Long, EPA Region VIII to Lynn R. Menlove, Utah Dept, of Envtl. Quality (Aug. 8, 1997), available at www.epa.gov/region07/ air/nsr/nsrmemos/util-atl.pdf (last visited May 16, 2012) (pump station and production operations 21.5 miles apart and on opposite sides of a lake are a single source because connective channel transports raw materials between them and facilities have a support relationship); Letter from Cheryl L. Newton, EPA Region V, to Donald Sutton, Illinois EPA (Mar. 13, 1998), available at www.epa.gov/region7/air/nsr/ nsrmemos/acme.pdf (last visited May 16, 2012) (steel mill facilities 3.7 miles apart and separated by Lake Calumet, landfills, the 1-94 expressway, and the Little Camulet River are a single source because “USEPA considers that the close proximity of the sites, along with the interdependency of the operations and their historical operation as one source ...”); Letter from Richard Long, EPA Region VIII, to Dennis Myers, Colorado Air Pollution Control Div. (Apr. 20, 1999), available at http://www.epa.gov/region07/air/nsr/ nsrmemos/amersoda.pdf (last visited May 16, 2012) (“distance alone” does not preclude a mine and plant thirty-five to forty miles apart but connected by pipeline being considered a single source given “integral connectedness of the[] facilities”); Forest Gil, et al. (2001) (EPA guidance that adjacency analysis should include operational, not geographical, inquiries, e.g., whether the facilities are situated to enable working integration, whether materials and people are routinely transported between them, and whether the production process is split in any way between the facilities); Letter from Pamela Blakley, EPA Region V to Don Smith, Minn. Pollution Control Agency (Mar. 23, 2010), available at http://www.epa.gov/region7/air/nsr/ nsrmemos/single.pdf (last visited May 16, 2012) (facilities separated by three miles without any physical connection are a single source because materials are transported between them daily). See also Letter from Winston A. Smith, Region VI, to Randy C. Poole, Mecklenburg Cnty. Dept. of Envtl. Protection (May 19, 1999), available at http://www.epa.gov/region07/ air/nsr/nsrmemos/wel999.pdf (last visited May 16, 2012) (two bulk gas terminals separated by .9 miles are separate sources because although close in proximity, they are not functionally related).

Though the EPA has previously considered the functional relationship between activities in assessing whether they lie on “contiguous or adjacent properties,” its request of increased deference to this interpretation rests on the premise that some measure of deference is in fact owed to the EPA’s interpretation in the first instance — a premise we reject in light of the unambiguousness of the regulation at issue. See Christensen, 529 U.S. at 588, 120 S.Ct. 1655. Here again, we rely on authority from the Supreme Court in Rápanos:

In a curious appeal to entrenched executive error, the dissent contends that ‘the appropriateness of the Corps’ 30-year implementation of the Clean Water Act should be addressed to Congress or the Corps rather than to the Judiciary.’ Surely this is a novel principle of administrative law — a sort of 30-year adverse possession that insulates disregard of statutory text from judicial review. It deservedly has no precedent in our jurisprudence.

Rapanos, 547 U.S. at 752, 126 S.Ct. 2208 (internal citations omitted). Like Rapanos, we conclude that an agency may not insulate itself from correction merely because it has not been corrected soon enough, for a longstanding error is still an error. We thus reject the EPA’s argument to the contrary and have little hesitation in taking this opportunity — the first our Court has been given in the EPA’s own history of “entrenched executive error” — to vacate the agency’s unreasonable interpretation of its Title V permitting plan. Id.

D. Assuming arguendo the term “adjacent” is ambiguous, the EPA’s interpretation is inconsistent with the regulatory history of the EPA’s Title Y plan and the EPA-issued guidance memorandums.

Although we consider the unambiguousness of the regulatory term “adjacent” to be quite clear, we conclude that our holding would be the same even if we were to accept the EPA’s argument that its regulation is ambiguous.

Where an agency interprets its own ambiguous regulation, we credit that interpretation with great deference. Auer, 519 U.S. at 461, 117 S.Ct. 905. Indeed, only if the agency’s interpretation is plainly erroneous or inconsistent will we reject it. Id. Because even deferential review is not inconsequential, however, see Ky. Waterways Alliance, 540 F.3d at 474, we will not grant administrative agencies unbridled discretion to interpret the terms of their own regulations in any conceivable manner. See generally Auer, 519 U.S. at 461, 117 S.Ct. 905; Motor Vehicle Mfrs. Ass’n, 463 U.S. at 42-43, 103 S.Ct. 2856. Here, even if we were to find that the word “adjacent” was ambiguous, we would still find that the EPA’s interpretation of its regulatory requirement that Title V single stationary sources are located on “contiguous or adjacent properties” was entirely inconsistent with: (1) the regulatory history of its Title V permitting plan; and (2) its own guidance memorandums regarding the application of its Title V regulations to the oil and gas industry. We would thus conclude that EPA’s interpretation was unreasonable under Auer, and our decision in this case would be the same.

1. The EPA’s interpretation is inconsistent with the regulatory history of its Title V plan.

On the issue of the EPA’s regulatory history interpreting Title V, we first note the incongruity between the EPA’s refusal to include a functional relationship test in its single stationary source analysis and its current position that its analysis cannot be completed without it.

In order for multiple activities to be together considered a single “building,” “structure,” “facility,” or “installation” within the meaning of the EPA’s Title V plan, the activities must: (1) be under common control; (2) be located on contiguous or adjacent properties; and (3) belong to the same industrial grouping. 40 C.F.R. § 71.2. The EPA first established this stationary source test for use in its Prevention of Significant Deterioration (PSD) program. See 40 C.F.R. § 52.21(b)(6) (“Building, structure, facility, or installation means all of the pollutant-emitting activities which belong to the same industrial grouping, are located on one more contiguous or adjacent properties, and are under the control of the same person (or persons under common control).... ”). It developed the test in the wake of the Alabama Power Co. v. Costle decision, a 1979 challenge to the EPA’s former PSD regulations, adjudicated by the United States Court of Appeals for the D.C. Circuit. See Alabama Power Co. v. Costle, 636 F.2d 323 (D.C.Cir.1979). Assessing the EPA’s then-existing stationary source test, the Alabama Power Court advised that “[t]o allow an entire plant or other appropriate grouping of industrial activity to be subject as a single unit ... EPA should devise regulatory definitions of the terms ‘structure,’ ‘building,’ ‘facility,’ and ‘installation’ to provide for the aggregation, where appropriate, of industrial activities according to considerations such as proximity and ownership.” Id. at 397. The court also required that the EPA make its aggregation test sufficiently predictable to give plausibly regulated entities “explicit notice as to whether (and on what statutory authority) the EPA construes the term [stationary] source....” Id.

In the preamble to the 1980 amendments to the final PSD rules, the EPA recognized that Alabama Power required a new definition of a single stationary source to, inter alia, “approximate a common sense notion of a ‘plant[ ]’ [and] ... avoid aggregating pollutant-emitting activities that as a group would not fit within the ordinary meaning of ‘building,’ ‘structure,’ ‘facility,’ or ‘installation.’ ” Requirements for Preparation, Adoption, and Submittal of Implementation Plans; April 4, 2012 Approval and Promulgation of Implementation Plans, 45 Fed. Reg. 52,676 (Aug. 7,1980) (to be codified at 40 C.F.R. pts. 51, 52, and 124). Consistent with this understanding, the EPA declined to adopt the court’s proposed dual-prong proximity/ownership test, finding that this “definition [of a single source] would fail to approximate a common sense notion of a ‘plant,’ since in a significant number of cases it would group activities that ordinarily would be regarded as separate.” 45 Fed. Reg. at 52,695. The EPA therefore set out to add a third criteria which would narrow the field of aggregated sources to those more in line with the “common sense notion of a plant.” Id.

The functional relationship test, which EPA admittedly now uses to determine whether activities are adjacent to one another, was one of the additional criteria the EPA considered and rejected during its post-Aia&ama Power rule-making process. First, the EPA “asked for comment on whether factors other than proximity and control, such as the functional relationship of one activity to another, should be used [in stationary source determinations].” 45 Fed. Reg. at 52,694. Then, after due consideration, the EPA rejected the option of adding a third requirement (in addition to ownership and proximity) that the activities comprising a single stationary source be functionally related to one another. 45 Fed. Reg. at 52,695. The EPA specifically found that assessing whether activities were sufficiently functionally related to constitute a single source “would be highly subjective” and would make “administration of the definition substantially more difficult, since any attempt to assess those interrelationships would have embroiled the Agency in numerous, fine-grained analyses.” Id. In lieu of the rejected functional interrelationship test, the EPA elected to incorporate the “industrial grouping” criterion, a factor which asks whether the multiple activities are engaged in the same type of business, i.e., operate under the same SIC code. See supra n.2; 40 C.F.R. § 71.2.

a. The EPA’s mis-characterization of its own regulatory history is unconvincing.

The EPA argues that its prior rejection of the proposal that “functional relatedness” be a third prong of its stationary-source test is immaterial because its decision not to use this as a stand-alone factor is not inconsistent with its use of the concept as a supplement to its adjacency analysis. In at least two ways, neither Title Vs regulatory history nor common sense supports the EPA’s position.

First, the fact that the EPA considered adding to the tests of ownership and proximity with a test of functional relatedness belies its current contention that the factors of proximity and functional relatedness are one in the same. The EPA admits that it retained Alabama Power’s suggested ownership and proximity test, seeking only a third test to reduce the number of single stationary sources to those representative of a common plant. If the adopted proximity requirement— articulated by the EPA as the requirement that activities be “located on contiguous or adjacent properties” — suggested a measure of the activities’ functional relatedness, there would have been no need to consider adopting this already-included test as a third prong in the EPA’s single source analysis. Clearly, at least at the time the EPA adopted its stationary source regulation, the EPA understood adjacency in geographical, rather than operational, terms.

Second, the EPA’s decision not to employ a functional relatedness test was categorical and unqualified. Far from advocating the use of a functional relatedness assessment as part but not all of the stationary source analysis, the EPA expressed the sweeping conclusion that any reference to the operational relationship between activities “would be highly subjective,” would “make administration of the definition substantially more difficult,” and would burden “the Agency in numerous, fine-grained analyses.” 45 Fed. Reg. 52,-695. This broad condemnation cuts heavily against the EPA’s argument that the use of any assessment of functional relatedness is reasonable (even in its own view). Ultimately, the EPA has failed to convince us why the criterion it considered, and then strongly rejected, is now indispensable to its ability to make adequate stationary source determinations. The foregone conclusion, given the absence of notice and comment procedures amending the EPA’s Title V plan to include an assessment of activities’ functional relatedness, is that the EPA’s current test is in practice completely inconsistent with the wording of its own regulation.

2. The EPA’s interpretation is inconsistent with its own guidance memorandums.

We next note that just as the EPA’s present interpretation of adjacency is inconsistent with the regulatory history of its Title V permitting plan, so too is it inconsistent with the very EPA-issued guidance memorandums on which the EPA relies in the instant case. As discussed supra, the EPA’s first two letters to Summit cited the Wehrum Memorandum as primary authority in its decision-making process. The Wehrum Memorandum had particular relevance to the EPA’s single source determination for Summit — it was issued to guide permitting authorities specifically on the application of Title V to the aggregation of pollutant-emitting activities in the oil and gas industry. The Wehrum Memorandum rejected the notion that geographically distant oil and natural gas activities could be considered “contiguous” or “adjacent:”

[F]or this industry, we do not believe determining whether two activities are operationally dependent drives the determination as to whether two properties are contiguous or adjacent, because it would embroil the Agency in precisely the fine-grained analysis we intended to avoid, and it would potentially lead to results which do not adhere to the common sense, notion of a plant.... Given the diverse nature of the oil and gas activities, we believe that proximity is the most informative factor in making source determinations for these industries.

Wehrum Memorandum, supra. We need not strain to conclude that the EPA’s source determination for Summit, in which the EPA admittedly found that aggregation of Summit’s sweetening plant and sour gas wells was appropriate despite their physical separation, is patently inconsistent with this guidance.

While the EPA withdrew the Wehrum Memorandum and issued the McCarthy Memorandum shortly before the final iteration of its single source determination for Summit, the McCarthy Memorandum does not support the EPA’s analysis, either. To be sure, the McCarthy Memorandum criticized the Wehrum Memorandum’s emphasis on physical proximity, noting that such was not “a sufficient endpoint in the decision-making process.” McCarthy Memorandum, supra. Apart from revoking this prior guidance, however, the McCarthy Memorandum promotes a neutral and plain meaning application of the EPA’s Title V plan, directing permitting authorities to “rely foremost on the three regulatory criteria” for aggregating emissions from multiple sources, i.e., “(1) whether the activities are under the control of the same person (or person under common control); (2) whether the activities are located on one or more contiguous or adjacent properties; and (3) whether the activities belong to the same industrial grouping.” Id. (citing 40 C.F.R. § 52.21(b)(6)). As discussed supra, the McCarthy Memorandum concedes that at the end of any case-by-case analysis, proximity may well be the overwhelming and determinative factor. If the McCarthy Memorandum says anything about including supplemental and unexpected factors in the Title V stationary source analysis, it cautions against it.

There is some dispute between the parties as to which of the EPA’s letters declaring Summit’s plant and wells a single stationary source was the EPA’s “final agency determination.” Summit argues that the EPA’s October 2010 correspondence was a mere reiteration of its final determination in September 2009; thus, Summit contends, the Wehrum Memorandum applied to the EPA’s decision and its emphasis on physical proximity in the adjacency analysis should have caused Summit’s facilities to be considered separate emissions sources. Ultimately, whether the Wehrum Memorandum had been superceded by the McCarthy Memorandum at the time of the EPA’s final determination would not impact our decision. Although the Wehrum Memorandum made clear the near-singular importance of the physical proximity of two facilities in the EPA’s stationary source analysis, the McCarthy Memorandum supports Summit’s position with its focus on applying the regulatory text in Title V decisions. In either case, then, the EPA-issued guidance memorandums support the error in the EPA’s interpretation of its own regulatory term.⁴

E. Arguments by Amicus API support our conclusion.

Our decision finds support in many of the arguments raised in the brief of Amicus API.⁵ First, as a member of the oil and gas industry itself, API urges us to consider the fact that the EPA’s interpretation of adjacency to include a measure of functional relationship is subjective and unpredictable, undermining the Alabama Power Court’s mandate that the EPA’s test give “explicit notice as to whether (and on what statutory authority) the EPA construes the term [stationary] source.... ” See Alabama Power, 636 F.2d at 397. Relatedly, API criticizes the EPA for supporting its source determination with the concept of a “common sense notion of a plant,” reasoning that this general principle is clearly meant to constrain, rather than enlarge, the EPA’s definition of a single stationary source.

API also argues that the EPA’s interpretation has led in this case, and will continue to lead, to absurd source determinations within the industry. In a context where oil and gas producers often have little say over where emission points are located and nearly every facility is connected to one another by pipeline (and thus, in a sense, functionally related), API argues, the aggregation of activities without regard to their physical proximity is inherently unreasonable. Indeed, with the concept of a “common sense notion of a plant” construed as facilitating the aggregation of operationally related but physically distant activities, there is little foreseeable limit to the aggregation of emission points spread out literally across the country. Moreover, API cautions that source determinations dependent on activities’ functional relatedness will be quickly outdated as oil and gas producers add and remove emission points from their production systems, potentially requiring nearly constant renewed source determination requests and decisions.

We are particularly struck by API’s final observation — that the EPA’s source determination in this case is an ironic showcase of the very fears that caused the agency not to adopt a functional relatedness test for source determinations in the first instance. From Summit and MDEQ’s joint source determination request in January 2005 to the final iteration of the EPA’s decision, nearly five years passed. The parties engaged in at least twenty-five conference calls and exchanged “a small mountain of paper” within this period. Certainly, the cost for Summit to produce the data and schematics requested by the EPA, and the cost for the EPA to distill and understand the same, were high in terms of both monetary and capital resources. As API points out, this process produced exactly the “fine-grained” and administratively burdensome result the EPA sought to avoid in its drafting of its Title V stationary source test, see 45 Fed. Reg. at 52,695, and is strong evidence that the EPA’s interpretation of its own regulation was unreasonable.

III. Conclusion

For the reasons stated, we VACATE the final agency determination of the EPA aggregating Summit’s sour gas wells and sweetening plant into a single and major source under its Title V permitting plan. We REMAND this case to the EPA for a reassessment of Summit’s Title V source determination request in light of the proper, plain-meaning application of the requirement that Summit’s activities be aggregated only if they are located on physically contiguous or adjacent properties.

1 . A group of stationary sources belong to the same industrial grouping if they share a two-digit Standard Industrial Classification (SIC) code, a code used to determine the type of activity in which the facilities are engaged. 40 C.F.R. § 71.2. The SIC manual is available at http://www.osha.gov/pls/imis/sic_manual. html.

2 . 40 C.F.R. § 71.3 provides that an owner or operator of an emission source may submit a written request to the EPA administrator for a determination of the applicability of Part 71 to a particular facility. At the time of the parties' determination request, Summit was operating its plant under a Permit to Install (No. 632-821) issued by MDEQ under Michigan's Air Pollution Control Act, Part 55 of the Natural Resources and Environmental Protection Act, M.C.L. § 324.5501 ef seq. (Part 55), and Michigan’s CAA State Implementation Plan, 1980 AACS R. 336.1201.

3 . Before this Court and, it appears, in practice, the EPA styles and regards its guidance memorandums as non-binding, advisory documents. Both the Wehrum Memorandum and the McCarthy Memorandum, discussed infra, express that notwithstanding the suggestions for conducting Title V stationary source analyses contained therein, actual determinations are to be made on case-by-case, situational, and particularized bases.

4 . Summit additionally argues, as it has since its initial correspondence with EPA, that the definition of a "facility” in EPA's HAP regulations cuts against aggregation of even adjacent facilities in the oil and gas industry. Summit specifically cites the Preamble to Subpart HH, Section V of the Significant Comments and Changes to the Proposed Standards of HAP Regulations, within which EPA states that "EPA believes it would not be reasonable to aggregate emissions from surface sites that are located on the same lease, but are great distances apart.” Summit notes that the Preamble "makes no mention of distinguishing HAP emissions from criteria pollutant emissions [under Title V],” arguing that the same non-aggregation standard should apply to pollutant-emitting sources regulated under Title V. Pet’r Br. at 35. The Wehrum Memorandum would seem to support Summit's position as its three-step procedure for determining a stationary source cites to EPA's HAP regulations. Still, this is not Summit’s strongest argument. As even Summit concedes, "HAPs are not at issue in this case.” Pet'r Br. at 36. Furthermore, the Senate committee discussing the new definition of a "source” within EPA's HAP regulations specifically cautioned that any definition would refer only to HAPs. See S.Rep. No. 101-228 at 150-51 (1989), 1990 U.S.C.C.A.N. 3385, 3535-36 reprinted in 5 Legislative History at 8490-91 (cautioning that the definition would “apply only in the context of this section and should not be confused with other meanings of the term 'major source’ in parts C (prevention of significant deterioration) or D (non-attainment) of the Act.”). Because Michigan's Title V regulations are also not at issue in this case, Summit's similar appeal to the incongruity between EPA's Title V plan and Michigan's plan (which precludes aggregation of oil and gas facilities despite contiguousness or adjacency) also fails.

5 . While we credit Amicus AXPC with raising several persuasive arguments, as well, they have substantial overlap with the positions taken by Summit and have thus already been addressed.