23CA2143 Center Biological v Dept Public Health 05-22-2025 COLORADO COURT OF APPEALS
Court of Appeals No. 23CA2143 City and County of Denver District Court Nos. 21CV30049 & 21CV30886 Honorable Sarah B. Wallace, Judge
Center for Biological Diversity, Colorado Latino Forum, and Sierra Club, Plaintiffs-Appellants, v.
Colorado Department of Public Health and Environment, Air Pollution Control Division, and American Petroleum Institute Colorado,
Defendants-Appellees.
JUDGMENT AFFIRMED
Division I
Opinion by JUDGE BROWN
J. Jones and Yun, JJ., concur
NOT PUBLISHED PURSUANT TO C.A.R. 35(e)
Announced May 22, 2025
Allison N. Henderson, Crested Butte, Colorado; Jeremy McKay, Denver, Colorado; Ryan Maher, Washington, D.C., for Plaintiffs-Appellants
Phillip J. Weiser, Attorney General, Laura Terlisner Mehew, Senior Assistant Attorney General, Julia La Manna, Assistant Attorney General, Denver, Colorado, for Defendant-Appellee Colorado Department of Public Health and Environment, Air Pollution Control Division
Williams Weese Pepple & Ferguson PC, Jennifer L. Biever, John H. Bernetich, Dale T. Ratliff, Denver, Colorado, for Defendant-Appellee American Petroleum Institute Colorado
¶1 Center for Biological Diversity, Colorado Latino Forum, and Sierra Club (collectively, the Public Interest Groups) appeal the district court’s judgment affirming the decision of the Colorado Air Pollution Control Division (Division) to issue a general construction permit, General Permit 11 (GP11), to regulate routine or predictable emissions (ROPE) from oil and gas operations. The Public Interest Groups contend that the Division’s decision to issue GP11 was arbitrary or capricious or contrary to law because (1) GP11’s conditions are not practically enforceable, and (2) GP11 allows sources to exceed the national ambient air quality standards (NAAQS) set by the Environmental Protection Agency (EPA). We affirm.
I. Regulatory Framework
¶2 The Clean Air Act (CAA) “establishes a cooperative-federalism framework” to prevent and control air pollution. WildEarth Guardians v. Extraction Oil & Gas, Inc., 457 F. Supp. 3d 936, 941 (D. Colo. 2020); 42 U.S.C. § 7402. One of the CAA’s primary goals is “to protect and enhance the quality of the [n]ation’s air resources so as to promote the public health and welfare.” 42 U.S.C. § 7401(b)(1). To that end, the CAA directs the EPA to publish a list
2
of air pollutants with emissions that “cause or contribute to air pollution which may reasonably be anticipated to endanger public health or welfare.” 42 U.S.C. § 7408(a)(1)(A). The CAA then requires the EPA to promulgate NAAQS for such pollutants in designated air quality control regions across the country. 42 U.S.C. §§ 7409(a), 7410(a)(1).
¶3 Each state is responsible for “assuring air quality” within its geographic boundaries by developing and submitting to the EPA a state implementation plan (SIP) to achieve, maintain, and enforce the NAAQS in each air quality control region within the state. 42 U.S.C. §§ 7407(a), 7410(a)(1). Among other things, the SIP must include “enforceable emission limitations and other control measures, means, or techniques . . . as may be necessary or appropriate” and regulations addressing “the modification and construction of any stationary source . . . as necessary to assure that [NAAQS] are achieved.” § 7410(a)(2)(A), (C).
¶4 Within the Colorado Department of Public Health and Environment (CDPHE), two sub-departments administer Colorado’s EPA-approved SIP: the Air Quality Control Commission (Commission) and the Air Pollution Control Division (Division). See
3
§§ 25-7-103(2), (7), -104(a), -114.4(1), -114.5, C.R.S. 2024; Dep’t of Pub. Health & Env’t Reg. 3, 5 Code Colo. Regs. 1001-5 (Regulation 3); WildEarth Guardians, 457 F. Supp. 3d at 941. The Commission is responsible for developing rules and regulations regarding the construction, operation, and permitting of stationary sources of air pollutants. § 25-7-114.4(1) (“The commission shall promulgate . . . regulations . . .for the orderly and effective administration of construction permits.”). The Division then implements the rules and regulations promulgated by the Commission. See §§ 25-7-114.2, -114.4(1)-(2), -114.5, C.R.S. 2024. A “[s]tationary source” is defined as “any building, structure, facility, or installation which emits or may emit any air pollutant.” § 25-7-103(23); accord Regulation 3, pt. A, § I.B.52; § 7411(a)(3).
¶5 As detailed in CDPHE Regulation 3, Colorado’s New Source Review (NSR) program governs the permitting of stationary sources of air pollutants. See Regulation 3, pt. D; WildEarth Guardians, 457 F. Supp. 3d at 941. The program requires a stationary source1
1 Regulation 3 applies to any source that did not commence
construction or operation before February 1, 1972. Dep’t of Pub. Health & Env’t, 5 Code Colo. Regs. 1001-5, pt. B, § I.A.
4
to have or obtain a valid construction permit before (1) constructing or substantially altering any building, facility, structure, or installation (with exceptions not relevant here); (2) installing any machine, equipment, or device; or (3) commencing the conduct, performance, or operation of any such activity. § 25-7-114.2; Regulation 3, pt. B, §§ II.A, III.I.2.
¶6 A stationary source can be classified as a “[m]ajor source,” a “[m]inor source,” or a “[s]ynthetic minor source.” § 25-7-114(3), (6), C.R.S. 2024; Regulation 3, pt. A, § I.B.30, I.B.31. A source’s classification is determined by its potential to emit (PTE), defined as “the maximum capacity of [the] stationary source to emit a pollutant under its physical and operational design.” § 25-7-114(4); Regulation 3, pt. A, § I.B.43; accord 40 C.F.R. § 51.166(b)(4) (2024).
¶7 A source is a “major source” if its PTE is above certain thresholds. § 25-7-114(3)(a)-(c); Regulation 3, pt. A, § I.B.30; accord 42 U.S.C. § 7479(1); 40 C.F.R. § 51.165(a)(1)(iv)(A) (2024). A source is a “[m]inor source” if it “does not qualify as a major source.” Regulation 3, pt. A, § I.B.31. Major sources are subject to greater regulatory requirements, see generally 40 C.F.R. §§ 51.165, 51.166, while minor sources are subject to “only the barest of
5
requirements,” Sierra Club v. EPA, 964 F.3d 882, 886 (10th Cir. 2020) (citation omitted).
¶8 A “[s]ynthetic minor source” is “any source which would otherwise meet the definition of major source for any pollutants but for the existence of enforceable emission limitations contained in the permit or regulation applicable to that source.” § 25-7-114(6). In other words, a source may “voluntarily lower emissions to avoid major-source requirements,” WildEarth Guardians, 457 F. Supp. 3d at 942, by implementing physical or operational limitations on the capacity of the source to emit pollutants, see 40 C.F.R. § 51.166(b)(4); Regulation 3, pt. B, § II.A.7. Such limitations can include, for example, “air pollution control equipment and restrictions on hours of operation or on the type or amount of material combusted, stored, or processed.” Regulation 3, pt. A, § I.B.43; accord 40 C.F.R. § 51.166(b)(4).
¶9 The Division can issue two types of construction permits: an individual construction permit, which is unique to a source and requires analysis of that source’s particular operations and emissions, or a general construction permit, which is issued to cover numerous similar sources. § 25-7-114.2; Regulation 3, pt. A,
6
§ I.B.16, I.B.27, pt. B, § III.B, III.I. The EPA has approved the use of general permits as they “provide for emission limitations in a one-time permitting process, and thus avoid the need to issue separate permits for each source.” Off. of Enf’t & Compliance Assurance, EPA, Guidance on Enforceability Requirements for Limiting Potential to Emit through SIP and § 112 Rules and General Permits 3 (1995) (1995 EPA Guidance).
II. Procedural History
¶ 10 In 2019, the Commission decided that routine or predictable emissions (ROPE) associated with certain maintenance and processing activities at oil and gas facilities should no longer be exempt from permitting requirements. Instead, these emissions should be subject to permitting under Colorado’s SIP. See Regulation 3, pt. F, § I.DDD. The Commission also determined that the new ROPE permitting requirements would apply to existing sources, not just newly constructed or modified sources. Id.
¶ 11 In August 2020, the Division proposed GP11 to regulate ROPE and initiated a public comment period. The Center for Biological Diversity timely submitted comments raising the same concerns it raises on appeal. In December 2020, the Division issued GP11.
7
¶ 12 The following month, the Public Interest Groups filed a complaint seeking judicial review of the Division’s decision to issue GP11 under the State Administrative Procedure Act (APA). See §§ 24-4-106(2), 25-7-120(1), C.R.S. 2024. The American Petroleum Institute Colorado (API) intervened in the action as a defendant. After a hearing, the district court issued a written order affirming the Division’s decision to issue GP11.
III. Public Interest Groups’ Challenges to GP11
¶ 13 The Public Interest Groups contend that the Division’s decision to issue GP11 was arbitrary or capricious or contrary to law because (1) GP11’s conditions are not practically enforceable, and (2) GP11 allows the construction of sources that cause an exceedance of NAAQS. We conclude that the Division’s decision was neither arbitrary or capricious nor contrary to law.
A. Standard of Review
¶ 14 “Our review of a district court’s decision in a proceeding under the [APA] is de novo.” Farmer v. Colo. Parks & Wildlife Comm’n, 2016 COA 120, ¶ 12. We sit in the same position and apply the same standard of review as the district court. Gessler v. Grossman, 2015 COA 62, ¶ 9, aff’d, 2018 CO 48. We must “hold unlawful and
8
set aside” an agency action if, as argued here, the action is arbitrary or capricious or otherwise contrary to the law. § 24-4-106(7)(b)(I), (VIII), (IX). If we perceive no such error, we must “affirm the agency action.” § 24-4-106(7)(a).
¶ 15 “In applying this standard, we presume the validity and regularity of administrative proceedings and resolve all reasonable doubts as to the correctness of administrative rulings in favor of the agency.” Grossman, ¶ 11. Ultimately, “[t]he burden is on the party challenging the agency action to overcome the presumption that the agency’s acts were proper.” Wildwood Child & Adult Care Program, Inc. v. Colo. Dep’t of Pub. Health & Env’t, 985 P.2d 654, 655 (Colo. App. 1999).
¶ 16 To conclude that an agency’s decision is arbitrary or capricious, “we must determine that no substantial evidence exists in the record to support the agency’s decision.” Grossman, ¶ 39. “There must be a clear error of judgment, and we may not substitute our judgment for that of the agency.” Id. Indeed, the agency, not this court, “has the task of weighing the evidence and resolving any conflicts.” Chostner v. Colo. Water Quality Control Comm’n, 2013 COA 111, ¶ 24. “So long as the agency decision
9
reflects conscientious effort to reasonably apply legislative standards to particular administrative proceedings, it is neither arbitrary nor capricious.” Moya v. Colo. Ltd. Gaming Control Comm’n, 870 P.2d 620, 624 (Colo. App. 1994).
¶ 17 We also review de novo the agency’s interpretations of law. Citizens for Clean Air & Water in Pueblo & S. Colo. v. Colo. Dep’t of Pub. Health & Env’t, 181 P.3d 393, 396 (Colo. App. 2008). But, “[a]s part of our de novo review, ‘we may consider and defer to an agency’s interpretation of its own enabling statute and [of] regulations the agency has promulgated.’” Gessler v. Colo. Common Cause, 2014 CO 44, ¶ 7 (quoting Bd. of Cnty. Comm’rs v. Colo. Pub. Utils. Comm’n, 157 P.3d 1083, 1088 (Colo. 2007)); see Chostner,
¶ 24 (“[A]though we are not bound by an agency decision that misapplies or misconstrues the law, we defer to an agency’s interpretation of the statute or regulation it is charged with administering.” (citing El Paso Cnty. Bd. of Equalization v. Craddock, 850 P.2d 702, 705 (Colo. 1993))). An agency’s interpretation is “most useful to the court when the subject involved calls for the exercise of technical expertise which the agency possesses.” Craddock, 850 P.2d at 705.
10
B. General Permit 11 (GP11)
¶ 18 GP11 is a general minor source construction permit that authorizes and regulates ROPE from discrete activities conducted at certain oil and gas industry facilities. ROPE are only one component of the entire set of emissions from oil and gas operations. GP11 does not address non-ROPE emissions, which must be permitted through other means. ROPE consist mostly of volatile organic compounds (VOCs).
¶ 19 Because ROPE are produced by routine activities that are generally short in duration, sources registered under GP11 need not control ROPE. Instead, to comply with GP11’s emission limits, operators can choose to minimize ROPE or reduce the frequency of ROPE-producing activities. Still, a major source can register under GP11 as a synthetic minor source by choosing to control ROPE — that is, by reducing VOCs through air pollution control equipment such as combustion devices (flares or enclosed combustion devices), vapor recovery units, or catalysts. Either way, GP11 requires sources to aggregate emissions from facility-wide ROPE activities to determine whether they meet certain emission limits.
11
¶ 20 If an operator chooses to control ROPE with a combustion device, GP11 allows the operator to claim a destruction and removal efficiency (DRE) of 95%. A DRE of 95% means that 95% of the VOCs from ROPE will be destroyed by the device before being released into the atmosphere. However, the combustion of VOCs creates nitrogen oxides (NOx) as a byproduct. NOx emissions are measured as nitrogen dioxide (NO2) in NAAQS. 40 C.F.R. § 50.11(a)-(b) (2024). To ensure that sources using combustion devices to control ROPE achieve 95% DRE, GP11 imposes certain design, operation, monitoring, and recordkeeping requirements on operators.
¶ 21 For design, GP11 requires that any pollution control equipment used to comply with the permit’s emission limits be “adequately designed and sized to achieve the control efficiency rates and to handle reasonably foreseeable fluctuations in gas quantity and/or composition during emitting activities.” Combustion devices used to control ROPE must also “have a design destruction efficiency of at least 98%” and “be equipped with and operate an auto-igniter.”
12
¶ 22 For operations, GP11 requires that any pollution control equipment “be operated and maintained consistent with manufacturer specifications and good engineering and maintenance practices.” It also requires that the manufacturer specifications be kept on file for review by the Division.
¶ 23 For monitoring, GP11 requires that any combustion devices “have no visible emissions during normal operation, and be designed so that an observer can, by means of visual observation from the outside of the enclosed combustion device, or by other means approved by the [D]ivision, determine whether it is operating properly.” It also requires the operator to “monitor proper operation of the air pollution control equipment . . . at least once each day that [such] equipment is in use to control the emitting activities registered under” GP11 and identifies certain “[i]ndications of improper operation.”
¶ 24 And for recordkeeping, GP11 requires that, “[f]or each separate air pollution control equipment, the owner or operator must keep an air pollution equipment maintenance log,” displaying the date, time, and nature of any maintenance activity. It requires that any “[i]mproper operation of air pollution control equipment” be
13
documented, including a description of the problem and its resolution, the date range the equipment was inoperable, and the “uncontrolled mass of emissions during the downtime.” GP11 also requires operators to record “actual emissions from all emitting activities” and, for synthetic minor sources that use equipment to control emissions, to calculate such emissions on a “rolling twelve (12) month” basis.
C. Practical Enforceability
¶ 25 The Public Interest Groups contend that the Division’s decision to issue GP11 is arbitrary or capricious or contrary to law because the conditions that allow a source to control ROPE through combustion devices are not enforceable as a practical matter. In particular, the Public Interest Groups argue that (1) GP11 allows sources to assume 95% DRE despite evidence in the administrative record showing that combustion control devices can have much lower control efficiencies; (2) the Division did not account for variables that affect control efficiency; (3) only testing can verify that emissions remain below the permit’s limits; and (4) in crafting GP11’s conditions, the Division relied on EPA guidance regarding
14
practical enforceability that has been undermined by recent EPA decisions.
¶ 26 We begin by reviewing the requirement that permit conditions allowing a source to limit PTE be “practically enforceable.” Next, we conclude that substantial record evidence supports the Division’s determination that GP11’s conditions — under which a major source can elect to control ROPE through combustion devices in order to register as a synthetic minor source — are practically enforceable. See Grossman, ¶ 39. We then address and reject each of the Public Interest Groups’ contentions because they effectively ask us to reweigh the evidence and substitute our judgment for the Division’s, something we cannot do. See id.
1. GP11’s Control Conditions Must Be Practically Enforceable
¶ 27 As an initial matter, the parties agree that permits allowing sources to limit PTE must include “practically enforceable permit conditions,” Regulation 3, pt. B, § II.A.4, that impose “physical or operational limitation[s] on the capacity of the source to emit a pollutant, including air pollution control equipment,” Regulation 3, pt. A, § I.B.43. See also § 24-7-114(6) (defining synthetic minor source as one constrained by “enforceable emission limitations
15
contained in the permit or regulation applicable to that source”); WildEarth Guardians, 457 F. Supp. 3d at 959 (conditions must be “legally and practicably enforceable by a state or local air pollution control agency” (quoting EPA, Release of Interim Policy on Federal Enforceability of Limitations on Potential to Emit 3 (1996))). Thus, the conditions in GP11 that apply to “[c]ombustion devices, vapor recovery units, catalysts, or other division-approved air pollution control equipment used to reduce emissions (from the emitting activities registered under this general permit) below” applicable limits must be “practically enforceable.” Regulation 3, pt. B, § II.A.4.
¶ 28 Notably, “practically enforceable” is not defined by Regulation 3 or any relevant Colorado statute. To interpret the term, the Division relies on the 1995 EPA Guidance regarding practical enforceability. It provides that a general permit’s provisions must “specify (1) a technically accurate limitation and the portions of the source subject to the limitation; (2) the time period for the limitation (hourly, daily, monthly, annually); and (3) the method to determine compliance including appropriate monitoring, record keeping, and reporting” and must “(4) identify the categories of sources that are
16
covered by the [permit]; (5) where coverage is optional, provide for notice to the permitting authority of the source’s election to be covered by the [permit]; and (6) recognize the enforcement consequences relevant to the [permit].” 1995 EPA Guidance at 6.
¶ 29 The 1995 EPA Guidance clarifies that general permits must identify and apply to categories of sources “specifically or narrowly enough so that specific limits and compliance monitoring can be identified and achieved by all sources.” Id. It further provides that “monitoring” may refer to “many different types of data collection, including continuous emission or opacity monitoring, and measurements of various . . . [p]arameters of processes or control devices (e.g. temperature, pressure drop, fuel usage) and record keeping of parameters that [have] been limited.” Id. at 9. It notes that “[c]ontinuous emissions monitoring, especially in the case of smaller sources, is not required.” Id. And it provides that general permits “can allow for generic control efficiencies where technically sound and appropriate, depending on the extent of the application and ability to monitor compliance with resultant emission limits.” Id. at 6.
17
2. Division’s Determination that GP11’s Conditions are Practically Enforceable Was Not Arbitrary or Capricious or Contrary to Law
¶ 30 The Division’s determination that GP11’s conditions, allowing sources to elect to control ROPE through combustion devices, are practically enforceable is supported by the foregoing EPA guidance, the unique characteristics of ROPE, and the technical evidence in the administrative record.
¶ 31 Consistent with the 1995 EPA Guidance, GP11 identifies and applies to a narrow category of sources, activities, and emissions; it governs a limited number of specific ROPE-producing activities conducted only at certain “oil and gas industry well production facilit[ies] . . . and the associated . . . well(s).” GP11 also identifies a narrower subcategory of sources that elect the permit’s option to control ROPE through pollution control equipment such as combustion devices. Only the latter subcategory must comply with “practically enforceable” conditions to register as synthetic minor sources under the permit.2
2 Because the focus of the Public Interest Groups’ arguments is on GP11’s conditions related to combustion devices, ours is as well.
18
¶ 32 Given the unique characteristics of ROPE as discrete episodic or sporadic emissions that occur in connection with routine maintenance and processing activities at oil and gas facilities, the Division determined that operators are unlikely to opt to control ROPE at all. The Division also considered the types of devices an operator might use to control ROPE and how to ensure proper operation of those devices. Due to the short-term, variable nature of ROPE and the types of devices used to control it, the Division determined that performance testing was not a feasible means of monitoring control efficiency. See, e.g., 40 C.F.R. § 60.5413(d)(2) (2024) (requiring testing for combustion devices in “three one-hour (or longer) test runs”). The Division reasoned that if operators were required to conduct performance tests on combustion devices used to control ROPE, they would likely need to supplement the material being combusted and run the devices for longer periods than necessary, which would lead to increased pollution.
¶ 33 Instead, to ensure that operators choosing to control ROPE through combustion devices would meet emission limits, the Division implemented the design, operation, monitoring, and recordkeeping conditions set forth in GP11 and detailed above. The
19
Division concluded that, when these combustion devices are operated and maintained properly, as required by GP11’s conditions, they achieve DRE of 95% or higher. Therefore, allowing sources to claim 95% DRE was “technically sound and appropriate.” 1995 EPA Guidance at 6. And the Division’s conclusion is supported by substantial evidence in the administrative record, including reports, control device manufacturer specifications, and dozens of performance test results. See Grossman, ¶ 39.
¶ 34 We perceive no clear error in the Division’s judgment. See id. Rather, the Division’s decision “reflects conscientious effort to reasonably apply” the practical enforceability standard to a narrow category of sources that elect to control a unique type of emissions through combustion devices. Moya, 870 P.2d at 624. And given the Division’s technical expertise, its decision is entitled to substantial deference. See Citizens for Clean Air, 181 P.3d at 397. Thus, we conclude that the Division’s decision was neither arbitrary or capricious nor contrary to law.
20
3. Public Interest Groups’ Arguments a. Contrary Evidence in the Record
¶ 35 The Public Interest Groups contend that the Division’s decision is arbitrary or capricious because GP11 allows sources to assume 95% DRE, despite record evidence demonstrating that combustion control devices can have significantly lower control efficiencies. They argue that, without testing, GP11’s conditions are not enforceable as a practical matter because there is no reliable way to measure the pollution emitted by the permitted source.
¶ 36 To support this argument, the Public Interest Groups point to a test of combustion flares at Bonanza Creek’s Wetco Farms A-4 well pad (hereinafter Wetco Farms Test). This test yielded DRE results of 73.92%, 73.80%, and 73.85% for one flare and 91.21%, 90.06%, and 89.34% for another flare. The Division counters that it is unclear whether the Wetco Farms Test involved ROPE or whether the tested combustion devices were subject to GP11’s design, operation, monitoring, or recordkeeping requirements. The Division and the API also argue that the conditions of the Wetco Farms Test were not representative of normal operating conditions. They point
21
to evidence that a follow-up test conducted at the same well pad produced a DRE of over 99%.
¶ 37 The Public Interest Groups also emphasize findings from a report involving EPA Region 8 and the Wyoming Department of Environmental Quality (EPA-Wyoming Report), which found that combustion flare efficiency varied dramatically “depending on the operational setup.” Michael Stovern et al., EPA, Region 8, Measuring Enclosed Combustion Device Emissions Using Portable Analyzers – Phase 1: Test Summaries 9 (May 14, 2020), https://perma.cc/RBT2-9547. But the Division notes that it is unclear whether the emissions measured in the EPA-Wyoming Report were ROPE or analogous to ROPE. The Division also argues that some of the DRE results below 95% were due to insufficient or variable gas flow rates and improper operation. The Division explains that these issues are not relevant in this case because ROPE are “manually-triggered bursts of gas over a short period of time,” and GP11 requires that combustion devices be adequately designed to handle fluctuations in gas and operated consistent with manufacturer specifications.
22
¶ 38 Still, the Public Interest Groups argue that the Division has failed to account for variables such as weather, altitude, damage during shipping, equipment installation, improper construction, wear and tear over time, fluctuations in fuel and waste streams, and different temperatures needed to destroy different VOCs. They cite a technical comment on a different permit to support their position. See Ranajit Sahu, Technical Comments on the Proposed CDPHE Permit No.20AD0062 for Haugen #1-30, https://perma.cc/8MSV-ALT3. But, as the Division points out, that permit involved a permanent emissions-generating activity (a separator); it did not involve ROPE or consider the conditions in GP11 that allow sources to opt to control ROPE.
¶ 39 Even if the parts of the record the Public Interest Groups highlight support their claim, they do not negate the other substantial evidence on which the Division based its decision. The Division provides a rational explanation for discounting the Wetco Farms Test, the EPA-Wyoming Report, and the technical comments and, instead, relying on the dozens of other tests that show combustion devices operating at or above 98% DRE. See FCC v. Fox Television Stations, Inc., 556 U.S. 502, 513 (2009) (Under the
23
“narrow” arbitrary or capricious standard of review, “we insist that an agency ‘examine the relevant data and articulate a satisfactory explanation for its action.’”) (citation omitted). And we decline to reweigh the evidence, as the Public Interest Groups essentially ask us to do. See Chostner, ¶ 24. In the end, the Public Interest Groups have not persuaded us that the design, operation, monitoring, and reporting requirements in GP11 are inadequate. See Wildwood Child & Adult Care Program, Inc., 985 P.2d at 655.
b. Recent EPA Decisions
¶ 40 The Public Interest Groups also argue that the Division’s reliance on the 1995 EPA Guidance to conclude that GP11’s conditions are practically enforceable has been undermined by two recent EPA orders granting objections to specific operating permits. See Bonanza Creek Energy Operating Co., Petition No. VIII-2023-11 (EPA Jan. 30 2024) (final order on petition) (Bonanza Creek Order); DCP Operating Co. LP, Platteville Nat. Gas Processing Plant, Petition No. VIII-2023-14 (EPA Apr. 2, 2024) (final order on petition) (Platteville Order). In both cases, the petitioners (including two of the Public Interest Groups) challenged the Division’s issuance of permits before the EPA in part because the permits “[u]njustifiably
24
[a]ssume” a 95% control efficiency for control devices, without adequate testing, monitoring, and reporting “[d]espite [e]vidence to the [c]ontrary.” Bonanza Creek Order at 9; see also Platteville Order at 7. The petitioners made arguments similar to those raised by the Public Interest Groups in this appeal, including that “only site-specific, periodic testing” can ensure compliance. Bonanza Creek Order at 9-10; see also Platteville Order at 8. The EPA concluded that the existing permit records did not sufficiently explain how the permit conditions “assure[d] compliance with the requirements to achieve 95 percent VOC control efficiency” applicable to the specific units, facilities, and activities at issue. Bonanza Creek Order at 15; see also Platteville Order at 9. The EPA directed CDPHE to revise the permit records to address the petitioners’ arguments. Bonanza Creek Order at 15; see also Platteville Order at 13.
¶ 41 The Division and the API argue that we should not consider these EPA orders because they were not available to the Division when it issued GP11 and are not part of the administrative record. See Brighton Pharmacy, Inc. v. Colo. State Pharmacy Bd., 160 P.3d 412, 417 (Colo. App. 2007) (“Agency actions are to be reviewed
25
based solely upon the record made before the agency.”). But see Sierra Club, 964 F.3d at 893 n.9 (taking judicial notice of documents published on EPA’s website). To be sure, since the EPA issued these two orders in 2024, the Division could not have been aware of them or considered them when issuing GP11. They are not part of the administrative record and were not available to the district court when it affirmed the Division’s decision. For these reasons, we hesitate to even review the EPA orders as part of this APA judicial review proceeding. But even if we did consider them, they fail to demonstrate how the Division’s reliance on the 1995 EPA Guidance is misguided or that the conditions in GP11 are not practically enforceable to control ROPE generated from the limited activities GP11 covers.
¶ 42 First, the Public Interest Groups have not established that the standard the EPA applied to evaluate the major source permits in those orders applies to evaluate the practical enforceability of conditions in minor source permits like GP11. See Bonanza Creek Order at 6-8 (identifying each source at issue as a “major source under title V”); Platteville Order at 6 (same). As noted, major source permits are subject to stricter regulation, while minor source
26
permits are not. See Sierra Club, 964 F.3d at 886; Luminant Generation Co. v. EPA, 675 F.3d 917, 922 (5th Cir. 2012) (for minor sources, the CAA requires only that the SIP regulate the construction of a stationary source “as necessary to assure that [NAAQS] are achieved” and “includes no specifics regarding the structure or functioning of minor NSR programs” (first quoting § 7410(a)(2)(C); and then quoting Operating Permit Programs and Flexible Air Permitting Rule, 74 Fed. Reg. 51418, 51421 (Oct. 6, 2009))). Indeed, the petitioners argued that “title V permits must contain ‘sufficiently reliable’ procedures for determining compliance and ‘periodic monitoring sufficient to yield reliable data from the relevant time period that are representative of the source’s compliance with the permit.’” Bonanza Creek Order at 10 (first quoting 42 U.S.C. § 7661c(b); and then quoting 40 C.F.R. § 70.6(a)(3)(i)(B) (2024)); see also Platteville Order at 11. And part of the EPA’s rationale for concluding that the permit records were insufficient was that title V permits must “set forth . . . monitoring . . . requirements to assure compliance with the permit terms and conditions,” Bonanza Creek Order at 13 (first quoting § 7661c(c); and then citing 40 C.F.R. § 70.6(c)(1)), and that “[t]he
27
rationale for the selected monitoring requirements must be clear and documented in the permit record,” id. (citing 40 C.F.R. § 70.7(a)(5) (2024)); see also Platteville Order at 11-12. It is not clear how these standards relate to practical enforceability.
¶ 43 Second, in each case, the EPA did not conclude that the permit conditions were not practically enforceable to “assure compliance” with the presumed 95% control efficiency. Instead, the EPA determined that the permit record was inadequate to explain how the permit conditions assured compliance. See Bonanza Creek Order at 14; see also Platteville Order at 12. In other words, uncertainty remains about whether the conditions at issue in those cases are, in fact, sufficient. And notably, the permit records are not part of the administrative record before us, so we have no way to compare what the EPA considered inadequate there with what the Division relied on to issue GP11. Under these circumstances, we are not persuaded that the EPA orders demonstrate that the Division’s decision regarding the conditions in GP11 — a minor source permit regulating limited ROPE-producing activities — is arbitrary or capricious.
28
¶ 44 Because there is competent evidence supporting the Division’s decision, see Grossman, ¶ 39, and because we defer to the Division’s interpretation of its own regulations based on its technical expertise, see Craddock, 850 P.2d at 705, we conclude that the Public Interest Groups have not met their burden to show the Division’s issuance of GP11 was arbitrary or capricious or contrary to law, see Wildwood Child & Adult Care Program, Inc., 985 P.2d at 655.
D. Exceedance of NO2 NAAQS
¶ 45 The Public Interest Groups contend that the Division’s decision to issue GP11 is contrary to law because (1) the self-executing nature of GP11 allows a source to begin construction once it has submitted an application, but before the Division has determined that it will not exceed NO2 NAAQS; (2) the Division failed to explain how it would determine whether a source is exceeding emission limits; and (3) the public is unable to submit comments or obtain judicial review as required under state and federal law when a permit goes into effect upon registration. We are not persuaded.
29
1. Applicable Law
¶ 46 The EPA has approved the use of general permits broadly, see 40 C.F.R. § 49.156 (2024); 1995 EPA Guidance, and specifically as part of Colorado’s SIP, see Approval and Promulgation of Air Quality Implementation Plans; Colorado; New Source Review and Prevention of Significant Deterioration, 59 Fed. Reg. 42500 (Aug. 18, 1994) (approving modifications to Colorado’s SIP). The EPA has explained that “[t]he purpose of a general permit is to simplify the permit issuance process for similar facilities so that a reviewing authority’s limited resources need not be expended for site-specific permit development for such facilities.” Review of New Sources and Modifications in Indian Country, 76 Fed. Reg. 38748, 38767 (July 1, 2011). According to the EPA, “general permits offer a cost-effective means of issuing permits and provide a quicker and simpler alternative mechanism for permitting minor sources than the site-specific permitting process.” Id.; see also Approval and Promulgation of Air Quality Implementation Plans; North Dakota; Revisions to Air Pollution Control Rules, 84 Fed. Reg. 11646, 11647 (Mar. 28, 2019) (“The EPA has a well-established, longstanding position that the use of general permits for construction of . . .
30
minor sources[, including synthetic minor sources,] is appropriate under the CAA.”).
¶ 47 To implement general permitting in Colorado, the Commission was charged with promulgating “[p]rocedures for issuing general permits after notice and an opportunity for hearing, covering numerous similar sources.” § 25-7-114.4(1)(p). The regulations authorize the Division to “issue a general construction permit covering numerous similar sources to a source that would otherwise be required to obtain a construction permit.” Regulation 3, pt. B, § III.I.1. General construction permits “shall comply with all applicable requirements, including notice and opportunity for public participation where warranted for such sources.” Id. Such permits “shall undergo statewide public notice.” Regulation 3, pt. B, § III.I.7.
¶ 48 The Division must also “enforce compliance with the . . . terms and conditions of any permit” issued under Colorado’s SIP. § 25-7-115(1)(a), C.R.S. 2024. Enforcement mechanisms include cease-and-desist orders, injunctions, and civil penalties. See §§ 25-7-112(1), -113, -115, -121, -122, C.R.S. 2024.
31
2. Registration and Certification under GP11
¶ 49 Under GP11’s terms, a source initially receives conditional certification that takes effect on the date the source submits a complete registration request. This means a source can commence construction and operation of ROPE-emitting activities as soon as the Division receives its completed registration.
¶ 50 GP11 provides that, within ninety days of the date the Division receives the complete registration, it must approve or deny the registration request in writing. See Regulation 3, pt. B, § III.I.4 (requiring the Division to review and certify or deny each application for coverage under a general permit). The Division may also deny or revoke registration and require a source to apply for an individual construction permit if either of the following occurs: (1) “[a] change has occurred in the availability of control technology or practices for the control or abatement of air pollutants applicable to the source,” or (2) “[c]ircumstances have changed since the time of the request,” and “the source is no longer appropriately controlled under the general construction permit.” Regulation 3, pt. B, § III.I.3.c.(i). GP11 also provides that, if a source improperly registers under it, the operator “accepts the liability of commencing” construction and
32
emitting activities, and the Division may take enforcement action against the operator.
¶ 51 For minor sources permitted under GP11, compliance with emission limits is determined on an annual calendar-year basis. Minor sources must keep a compliance record on site or at a local office. A synthetic minor source’s emission limits compliance is determined on a rolling twelve-month basis. A synthetic minor source is required to calculate actual emissions each month and keep a record on site or at a local office.
3. Division’s Determination that GP11 Sources Will Not Cause an Exceedance of NAAQS
¶ 52 At the outset, we note that the Public Interest Groups argue that the Division’s decision is contrary to law because “GP11 allows the construction of sources of air pollution that cause or contribute to violations of NAAQS set by EPA.” (Emphasis added.) But under Regulation 3, the Division must grant a construction permit if it finds that the “proposed source or activity will not cause an exceedance of any [NAAQS].” Regulation 3, pt. B, § III.D.1.c (emphasis added). That is the standard we apply. See § 25-7-114.5(7)(a)(III) (The Division shall grant a construction
33
permit if it finds that “the source or activity will meet any applicable ambient air quality standards and all applicable regulations.”).
¶ 53 On the merits, the Public Interest Groups argue that, because a source can begin construction and emitting activities once it registers under GP11, the Division cannot determine whether the source will cause an exceedance of NAAQS before construction starts. They frame this issue as a “straightforward temporal problem” resulting from the self-executing nature of GP11. And they point to a computer modeling analysis of a well pad in Bighorn, which they claim “is under the emission caps in GP11, and yet the modeling demonstrates that the source will cause violations of the NAAQS for nitrogen oxides.”
¶ 54 The Public Interest Groups’ argument overlooks the fact that the Division already determined that the sources eligible to register under GP11, categorically, will not exceed NAAQS by engaging in the permitted emitting activities. The Division explains that, because VOCs that occur from ROPE are intermittent and typically of short duration and because operators are unlikely to opt to control ROPE through combustion devices, it used a “qualitative method” to analyze potential NO2 emissions rather than
34
quantitative “modeling.” Using its technical expertise and judgment, the Division concluded,
While a small small amount of [NO2] may be periodically generated where an operator chooses to use a combustion device to control emissions, the amount generated by control of ROPE is currently expected to be well below the levels at which the [D]ivision would require a source to perform modeling to demonstrate compliance.
The Division has cited information from the administrative record that supports each step in its reasoning. Thus, we conclude that the Division’s decision to issue GP11 is not contrary to law because registration under GP11 is available to only those sources that the Division predetermined would not cause an exceedance of the NO2 NAAQS.
¶ 55 Admittedly, the significance of the Bighorn well pad modeling remains unclear to us. The Division asserts that the modeling is not representative of NOx emissions from controlling ROPE. The Division points out that the constant emission rate from that well pad was significantly higher than the emissions the Division would normally expect from sources controlling ROPE. The Division also notes that monitoring data from 2016 to 2020 showed no
35
exceedance of NO2 NAAQS across the state. Given this lack of clarity in the record, we defer to the Division’s technical expertise. See Craddock, 850 P.2d at 705.
4. Enforcement
¶ 56 The Public Interest Groups contend that the Division has not explained how it will identify sources that violate GP11 or exceed NAAQS, arguing that the Division is obligated to prevent the construction of violating sources rather than applying a “retroactive fix” through post hoc enforcement mechanisms. But the Public Interest Groups provide no authority requiring that the Division individually preapprove every source that registers under GP11. Indeed, such a requirement would undermine the purpose of general permitting to simplify the permitting process and avoid the need for the Division to conduct site-specific approval.
¶ 57 GP11 authorizes sources to register only if they meet all the permit conditions. It sets emission limitations and requires sources to maintain compliance records for Division review. For sources that elect to control ROPE, the permit requires additional recordkeeping. If a source violates the permit, the Division has access to all mechanisms of enforcement available under the law.
36
See §§ 25-7-112, -113 (cease and desist orders); § 25-7-115 (general enforcement authority); § 25-7-121 (injunctions); § 25-7-122 (civil penalties). We are not convinced that the Division’s reliance on these enforcement methods undermines its initial determination that sources authorized to register under GP11 will not cause an exceedance of NO2 NAAQS.
5. Public Comment
¶ 58 Finally, the Public Interest Groups contend that a member of the public cannot submit comments or obtain judicial review, even if they have evidence that a source permitted under GP11 caused an exceedance of NAAQS. But the applicable statutes and regulations require notice and an opportunity for the public to participate before general permits are issued. See § 25-7-114.4(1)(p); Regulation 3, pt. B, § III.I.1. Indeed, the Center for Biological Diversity participated and submitted comments for the Division’s consideration before GP11 was issued.
6. Concerns with General Permitting
¶ 59 Although the Public Interest Groups acknowledge that general permits are part of Colorado’s SIP and insist they are not challenging general permitting broadly, the Division and the API
37
argue, and we tend to agree, that the Public Interest Groups attack features common to all general permits. The “temporal problem[s]” the Public Interest Groups highlight regarding when the Division determines a permitted source will not cause an exceedance of NAAQS, when the Division will enforce permit violations, and when the public can participate, exist with any general permit and are not unique to GP11.
¶ 60 On this record, we conclude the Public Interest Groups have not met their burden to show that the Division acted contrary to law. See Wildwood Child & Adult Care Program, Inc., 985 P.2d at 655.
IV. Disposition
¶ 61 We affirm the district court’s judgment affirming the Division’s issuance of General Permit 11.
JUDGE J. JONES and JUDGE YUN concur.