Mobil Oil Corporation (Board decision, September 29, 1994)

Plain-English summary

Mobil Oil Corporation operated a petroleum refinery in Paulsboro, New Jersey. During regeneration of a sulfur-recovery unit, the refinery released sulfur dioxide above the permitted mass limit but did not report the release to the local emergency planning committee until March 22, 1990. The Environmental Appeals Board held that a Clean Air Act permit did not exempt a release that exceeded the permit limits, and that Mobil should have used available data and technical resources to report the release by March 17. The Board reduced the penalty from $75,000 to $30,250, assessing $8,250 for the first day and $5,500 for each of the next four days.

Decision snapshot

• Cited authorities: 42 U.S.C. §§ 11004, 11045, 9601, and 9603; 40 C.F.R. §§ 22.27 and 22.31.
• Outcome: The finding of an EPCRA reporting violation was affirmed, and the civil penalty was reduced from $75,000 to $30,250.
• Key point: A federally permitted release exemption does not cover an air release that exceeds the applicable permit limits, and emergency reporting must be pursued with reasonable urgency.

Full text (EPA EAB public release)

490 ENVIRONMENTAL ADMINISTRATIVE DECISIONS
IN RE MOBIL OIL CORPORATION
EPCRA Appeal No. 94-2
FINAL DECISION
Decided September 29, 1994
Syllabus

Mobil Oil Corporation (Mobil) appeals an interlocutory order of the Presiding Officer denying Mobil’s motion to dismiss a complaint filed by U.S. EPA Region II, and an initial decision assessing a civil penalty against Mobil of $75,000 for violation of the emergency release reporting provisions of § 304 of the Emergency Planning and Community Right-To- Know Act (EPCRA), 42 U.S.C. § 11004. The Region's complaint alleged that Mobil violated EPCRA § 304 by failing to provide immediate notice to the local emergency planning com- mittee (LEPC) of a release of sulfur dioxide to the air from Mobil’s Paulsboro, New Jersey, refinery. Mobil did not report the release to the LEPC until ten days after the release. Mobil sought dismissal of the complaint on the grounds that the release was a “federally permitted release” under EPCRA § 304(a)(2)(A), because Mobil has a Clean Air Act State Implementa- tion Plan permit authorizing certain releases of sulfur dioxide, although the release at issue was above the permit’s emission limits. EPCRA § 304(a)(2)(A) exempts “federally permitted releases” (as defined in § 101(10) of the Comprehensive Environmental Response, Compen- sation, and Liability Act (CERCLA)) from EPCRA reporting requirements. The Presiding Of- ficer denied Mobil’s motion and, following an evidentiary hearing, held that Mobil could have reported the release three days earlier than it did, and therefore Mobil violated EPCRA
§ 304. The Presiding Officer imposed a total civil penalty of $75,000 against Mobil for the violation ($25,000 for each day reporting was delayed).

Held: First, we affirm the decision of the Presiding Officer denying Mobil’s motion to dismiss the complaint on the grounds that the release of sulfur dioxide was exempt from EPCRA reporting requirements as a “federally permitted release.” We agree with the Presid- ing Officer that the exemption is properly read as being limited to releases in conformance with permit and regulatory requirements. This interpretation is strongly supported by the legislative purposes and histories of EPCRA and CERCLA.

Second, with respect to the Presiding Officer's penalty assessment, we conclude that, with proper diligence, Mobil could have reported the release five days earlier than the actual report was made. The evidence demonstrates that Mobil did not give emergency reporting sufficient priority following the release, but instead focused its resources on continuing its usual operations. We further conclude that the Presiding Officer's penalty assessment was not consistent with the EPCRA Penalty Policy in certain respects. We assess a civil penalty against Mobil of $8,250 for the first day of violation, and a penalty of $5500 for each of the additional four days, for a total penalty of $30,250.

VOLUME 5

MOBIL OIL CORPORATION 491

Before Environmental Appeals Judges Nancy B. Firestone,
Ronald L. McCallum, and Edward E. Reich.

Opinion of the Board by Judge Reich:

Mobil Oil Corporation (Mobil) appeals the Initial Decision of the Presiding Officer assessing a penalty against Mobil in the amount of
$75,000 in connection with an enforcement action brought by U.S. EPA Region II. Region II’s complaint alleged that Mobil violated the emer- gency release reporting provision of § 304 of the Emergency Planning and Community Right-To-Know Act (EPCRA), 42 U.S.C. § 11004.' Re- gion II also appeals, seeking imposition of a penalty in the amount of
$250,000, as originally proposed in the complaint. Region II contends that Mobil violated EPCRA § 304 by failing to immediately report a release of sulfur dioxide in excess of the EPCRA “reportable quantity” (RQ) from Mobil’s Paulsboro, New Jersey, facility on March 12, 1990.

While the case was before the Presiding Officer, Mobil filed a motion to dismiss, or in the alternative for an accelerated decision, on the grounds that the March 12 release was a “federally permitted re- lease” within the meaning of § 101(10)(H) of the Comprehensive En- vironmental Response, Compensation, and Liability Act (CERCLA), 42 U.S.C. § 9601 et seq., and therefore exempt from EPCRA reporting. EPCRA incorporates the “federally permitted release” exemptions of CERCLA, which exempt certain releases from CERCLA liability and emergency notification requirements. See EPCRA § 304(a)(2)(A) and CERCLA §§ 103(a), 107(a).?In an Interlocutory Order dated September 30, 1992, the Presiding Officer rejected Mobil’s argument. Following an evidentiary hearing, the Presiding Officer rendered an Initial Decision on December 27, 1993, assessing a civil penalty against Mobil in the amount of $75,000. These appeals followed. The parties presented oral argument to the Board on the issues raised on appeal on July 27, 1994.4

• Section 304 of EPCRA requires that immediate notice of certain releases of “extremely hazard-
  ©us substances” be provided to the local emergency planning committee for the affected area and to the State emergency planning commission for the affected State. EPCRA § 304(b)(1).

*CERCLA was enacted in 1980. Pub. L. 96-510. EPCRA was enacted as title III of the Superfund Amendments and Reauthorization Act of 1986 (SARA). Pub. L. 99-499.

• Former Chief Administrative Law Judge (ALJ) Henry B. Frazier III entered the Interlocutory
  Order ruling on Mobil’s motion. Senior ALJ Gerald Harwood presided at the hearing, and rendered the Initial Decision.

“In addition to the appellate briefs received from the parties, the Board received and considered briefs from amici curiae Ashland Petroleum Company, Division of Ashland Oil, Inc; Borden Chemi- cals and Plastics; the Mid-America Legal Foundation; and the Society of the Plastics Industry, Inc.

VOLUME 5

492 ENVIRONMENTAL ADMINISTRATIVE DECISIONS

Mobil does not dispute that the March 12, 1990 release occurred, nor that it reported the release ten days after the incident. Rather, Mobil contends that the release was a “federally permitted release” exempt from EPCRA reporting requirements, and alternatively that the release report it made was consistent with the requirements of EPCRA. In addition, both Mobil and Region II challenge the Presiding Officer's penalty determination; Mobil seeks a substantially reduced penalty and Region II seeks the $250,000 penalty it originally requested in the complaint. For the reasons discussed below, we conclude that a viola- tion of EPCRA did occur and that the imposition of a penalty for Mobil’s violation of EPCRA § 304 is appropriate. However, we disagree with the Presiding Officer's penalty assessment, and instead assess a total penalty of $30,250.

I. BACKGROUND

The facts underlying the Region's complaint are essentially undis- puted and are set forth in detail in the Initial Decision. Briefly, a part of Mobil’s oil refining process includes the recovery of elemental sul- fur from the refining of crude petroleum. The Paulsboro refinery con- tains a Sulfur Recovery Unit Complex consisting of two Sulfur Recov- ery Units (SRUs). The SRUs use a “Claus reaction” process to convert hydrogen sulfide in fuel gas to recoverable sulfur and water. The reac- tion process is aided by a catalyst consisting of aluminum oxide spheres. Over a period of time, hydrocarbons and sulfur compounds accumu- late in the catalyst and impair the reaction process. Approximately once a year, the catalyst in each SRU is “regenerated” through a com- bustion process that oxidizes the hydrocarbon and sulfur deposits. Only one SRU is regenerated at a time; refinery operations can and do continue with one SRU in operation while the other SRU is being regenerated. The regeneration process consists of a “burn” period to oxidize the deposits on the catalyst, followed by removal of the old catalyst and replacement with fresh catalyst. The “burn” phase of the process causes sulfur dioxide, carbon dioxide and water to be emitted from an incinerator stack. It was during such a regeneration event for the unit known as “SRU-3” that the release at issue occurred.

Mobil has a Clean Air Act State Implementation Plan (SIP) operat- ing permit issued by the State of New Jersey that establishes sulfur dioxide emission limits for the SRU-3 incinerator stack. The permit limits sulfur dioxide concentration in stack emissions to an hourly average of 15,000 parts per million (ppm) by volume at standard con- ditions and 540 pounds per hour by mass.

VOLUME 5

MOBIL OIL CORPORATION 493

At the time of the release, Mobil had in place certain systems for

monitoring process parameters at SRU-3, including the concentration (but not mass) of sulfur dioxide emissions. According to Mobil, there are no State or federal requirements mandating the use of such moni- toring systems; Mobil’s monitoring of sulfur dioxide concentration was voluntary. The Region has not disputed this. Sulfur dioxide concentra- tion and temperature data from the stack were recorded by Mobil’s Process Monitoring System (PMS). The PMS system provided sulfur dioxide concentration data in six-minute, hourly, and daily averages. In addition, at the time of the release, Mobil had implemented a new Process Information (PI) system that provided sulfur dioxide concen- tration data in minute-by-minute as well as hourly averages.

Mobil initiated the regeneration process at SRU-3 on the morning of March 12, 1990. Around noon, the refinery received an odor com- plaint from persons working at an adjacent Mobil facility. Mobil ceased regeneration operations, conducted a review of available process data, and then notified the State Emergency Response Commission (SERC) of the receipt of the odor complaint within one and one-half hours of receiving the complaint. Mobil states that this report was made solely pursuant to a State law that required immediate reporting of any inci- dent that might give rise to an odor or nuisance complaint off-site, and was not made because of EPCRA requirements. According to the record, the Mobil employees who reported the odor were sent for medical evaluations and released the same day. Tr. at 424-26.

Immediately after receiving the odor complaint, Mobil’s Environ- mental Compliance Supervisor, Richard Rodack, examined the PMS six-minute and hourly-average concentration data. Those data showed that sulfur dioxide concentrations prior to the odor complaint were below permit limits.° On the basis of this initial data review, Mr. Rodack decided that no permit exceedance had occurred. However, this initial determination, based on concentration data alone, did not adequately account for mass emissions. Because mass emissions depend in part on air and fuel flow, and air flow during the regeneration process is much higher than normal, compliance with the mass limit could not be

*The New Jersey SERC is the “State emergency planning commission” for purposes of EPCRA
§ 304 emergency notification.

• According to Mr. Rodack’s testimony at the hearing, the hourly-average data showed that
  sulfur dioxide concentrations were about two-thirds the allowed limit of 15,000 ppm for the two- hour period preceding the odor complaint. Although the six-minute average data for the 12-minute Period after noon showed concentrations in excess of 15,000 ppm, Mr. Rodack did not believe permit limits had been exceeded because the regeneration process was being shut down at that time, causing the flow volume to decrease drastically, and therefore the volumetric air flow would be going to zero. Tr. at 446-47; 450.

VOLUME 5

494 ENVIRONMENTAL ADMINISTRATIVE DECISIONS
estimated based on the relationship between mass and concentration levels that exists during normal operation.

More specifically, mass sulfur dioxide emissions are a function of the sulfur dioxide concentration and the flow stream volume, includ- ing three air flows (main, trim, and draft) and two fuel gas flows (to the catalyst regenerator and incinerator). The parties agree that in or- der to determine sulfur dioxide emissions on a mass pounds-per-hour basis, it is necessary to perform a relatively simple calculation utilizing sulfur dioxide concentration data, and air and fuel flow data. However, the only flow stream data immediately available to Mobil at the time of the release were main air and trim air flows; the draft air and fuel data necessary for an exact calculation were not immediately known.

Despite Mr. Rodack’s initial belief that emission limits for sulfur dioxide had not been exceeded, Mobil undertook additional review of the March 12 emissions. According to Mr. Rodack, this additional re- view was undertaken because the complaining personnel at the adja- cent Mobil facility had been affected by odor emanating from the stack, and because Mobil wanted to explore what happened during the process and whether there may have been a permit exceedance. In Mr. Rodack’s words, “[t]his was a non-steady-state process. It was not a typical operation.” Tr. at 451. Mobil directed Kim Murphy, the pro- cess engineer who was providing technical assistance during the re- generation, to calculate the mass sulfur dioxide emissions for each of the two hours preceding receipt of the odor complaint.

Ms. Murphy eventually prepared three sets of calculations. Although the record is unclear as to the length of time spent by Ms. Murphy in calculating the emissions, she testified that she believed she began making the initial set of calculations on the day of the release. Tr, at

515. For the initial calculations, Ms. Murphy used only the PI hourly-
     average sulfur dioxide concentration data in conjunction with data on the two known flow constituents (main air and trim air), Ms. Murphy's results, using the known air flows and hourly PI data, showed emis- sions well within permit limits for both hours. However, according to Mobil’s expert witness at trial, if Ms. Murphy had used the available PMS system’s 6-minute concentration averages, instead of hourly aver- ages from the newer PI system, the calculation would have shown emissions very close to the permit limit, even without the addition of the other flow constituents. Tr. at 718-21. It is undisputed that the inclusion of the remaining unknown flow constituents in the calcula-

7 The 6-minute averages were used by the expert himself in recreating Mobil's calculations. Id. They were also part of Mr. Rodack’s initial data review. VOLUME 5

MOBIL OIL CORPORATION 495
tion could only serve to increase the level of mass sulfur dioxide emissions. See Mobil’s Brief at 13.

Because the first calculation indicated no permit exceedance, Mobil restarted the regeneration process. After assisting with the regenera- tion process, Ms. Murphy undertook to refine her first calculations to include the three remaining unknown flows (draft air, and fuel gas to the incinerator and regenerator), since mass sulfur dioxide emissions are dependent upon the total flow volume. Tr. at 517. Estimating these flows involved researching the plant’s energy usage and making calcu- lations concerning the draft air entering the unit. Id. at 518. Ms. Murphy again relied on the PI hourly-average data in performing the second calculation. The second calculation, utilizing all air flows, showed that sulfur dioxide emissions for the ten o’clock hour on March 12 were 557 pounds per hour (17 pounds over the permit limit) and 538 pounds per hour for the eleven o'clock hour. While the administrative record on the precise timing of the second calculation is not clear, it appears to have been completed approximately seven or eight days after the release. Oral Arg. Tr. at 9-10.* Although these data indicated a permit exceedance for the ten o’clock hour, no report was made by Mobil at this time.

Because the second calculation suggested a permit exceedance, Ms. Murphy retrieved minute-by-minute sulfur dioxide concentration data from the PI system to further refine her calculations. Ms. Murphy prepared a third calculation using the minute-by-minute data. The record indicates that these data were retrieved on March 21, 1990. Respondent's Hearing Ex. 3. Ms. Murphy’s third calculation, completed on March 22, 1990, showed emissions of 680 pounds per hour for the ten o'clock hour on March 12 and 873 pounds per hour for the eleven o’clock hour, substantially in excess of permit limits.’ Mobil reported the re- lease to the Local Emergency Planning Committee (LEPC) on that date. On March 29, 1990, Mobil submitted a follow-up report as required by EPCRA § 304(c). That report stated that the estimated quantity of sulfur

®The Region agrees that the second calculation was completed on the seventh or eighth day, Region II's Reply Brief at 10.

°The disparity between the second and third calculations is likely attributable to the fact that

the third calculation included data that had been ignored as “over-range” by the PI system in pro- ducing the hourly-average data. See Tr. at 575, 585. VOLUME 5

496 ENVIRONMENTAL ADMINISTRATIVE DECISIONS
dioxide emitted on March 12 was “473 Lbs. above the permitted level.” Respondent’s Hearing Ex. 6.”

Ms. Murphy was the only Mobil engineer engaged in making the calculations necessary to determine whether the permit had been ex- ceeded on March 12. She testified that she was also responsible from March 12 forward for ensuring that the regeneration process was com- pleted and SRU-3 brought back on-line. Tr. at 586. According to her testimony, completing the “burn” phase of the regeneration typically takes from 24 to 36 hours, while the catalyst replacement typically takes two or three days. Id. at 507; 587. During the regeneration pro- cess, Ms. Murphy was working 12-hour days in order to bring SRU-3 back on-line. Id. at 587. She testified that it would have been “possible” for her supervisor to assign another engineer to continue the regenera- tion while she pursued the emissions calculations, but that was not done. Id.

II. THE INITIAL DECISION

The Presiding Officer concluded that by failing to report the re- lease until March 22, 1990, Mobil had not given “immediate” notice as required by EPCRA. He determined that if Mobil’s process engineer, Ms. Murphy, had utilized the available PMS 6-minute concentration data in performing her first calculation, it would have been apparent that the emissions were so near the permit limit that it was not clear that no exceedance had occurred, particularly in light of the fact that adding in the unknown flow rates would only increase the mass emis- sions rate. The Presiding Officer suggested that Mobil had sufficient data to enable it make a preliminary report to the LEPC on March 12, but he also recognized that EPCRA provides no clear basis for assess- ing a penalty for failure to make a preliminary report. He noted that although the preliminary data did not demonstrate an exceedance, Mobil was not justified in waiting ten days to report the release. He noted that there was no evidence that Mobil acted with any particular urgency in completing the calculations. The Presiding Officer found that it was reasonable to conclude that Mobil could have completed

The 473 pound figure represents the sum of the amounts by which the calculated emissions (680 and 873 pounds) exceeded the 540 pound permit limit for each hour. Mobil’s expert witness at tial, Professor Pablo Debenedetti, performed a retroactive analysis of the March 12 release and concluded that Mobil had exceeded its permit by only 47.4 pounds. Tr. at 714, The Region’s own expert, Howard Schiff, performed calculations showing that Mobil exceeded its permit by about 75 pounds for the first hour examined, and about 60 pounds for the second. id. at 171-72. We find, however, that Mobil’s March 29, 1990, written follow-up report noting a release of 473 pounds of sulfur dioxide provides the best guide for determining the amount of the release. There is no evi- dence that Mobil ever sought to amend its release report to reflect any different amount on the basis of any subsequent analyses of the data.

VOLUME 5

MOBIL OIL CORPORATION 497
the necessary calculations within seven days of the release. Thus, in the Presiding Officer's view, Mobil’s release report was made three days later than necessary.

In assessing a penalty for the three-day reporting delay, the Pre- siding Officer relied on EPA’s Final Penalty Policy for Sections 302, 303, 304, 311 and 312 of [EPCRA] and Section 103 of [CERCLA], June 13, 1990 (Penalty Policy). Utilizing the penalty matrix set forth in the Policy," the Presiding Officer classified the violation as “Level I” in extent, and “Level A” as to gravity. These are the highest extent and gravity levels available on the matrix. Based on the penalty matrix, this classification results in a maximum penalty of $25,000 per day. The Presiding Officer found it appropriate to assess the full penalty for each of the three days the report was delayed, and thus assessed a total penalty of $75,000 against Mobil. He concluded that “[the per day assessment is a reasonable deterrent against a repetition of this or a similar violation.” Initial Decision at 22.

III. DISCUSSION

EPCRA provides that:

If a release of an extremely hazardous substance re-

ferred to in [§ 302] occurs from a facility * * * and such

release is not subject to the notification requirements

under [CERCLA] the owner or operator of the facility

shall immediately provide notice as described in sub-

section (b) of this section, but only if the release—
(A) is not a federally permitted release as defined
in [CERCLA § 101(10)],

‘The Penalty Policy contains two matrices for calculating a base penalty that take into account four statutory penalty factors reflecting characteristics of the violation: its “nature, circumstances, extent and gravity.” Once a base penalty amount is determined, it may then be adjusted upward or downward to take account of the other statutory penalty factors reflecting characteristics of the violator. The “nature” of the violation, ie,, the type of statutory requirement violated, determines which of the two matrices applies. The “gravity” of the violation, measured by the amount of the chemical that was involved, is reflected on the horizontal axis of the matrix. ‘The “extent” of the violation, measured by the amount of deviation from the statutory requirement, is reflected on the vertical axis. Once the violation is assigned to the cell in the matrix where the axes intersect, the “circumstances” of the violation are evaluated as a basis for determining a specific penalty amount within the range of penalty amounts indicated in that cell. The circumstances take into account the likelihood of exposure to hazard and the adverse effect of the violation on implementing the stat- ute. Jn re Great Lakes Div. of Nat'l Steel Corp., EPCRA Appeal No. 93-3, n. 25 (EAB, June 29, 1994).

VOLUME 5

498 ENVIRONMENTAL ADMINISTRATIVE DECISIONS
(B) is in an amount in excess of a quantity which
the Administrator has determined (by regulation)
requires noticel.]
re ee ee
Unless and until superseded by regulations establish-
ing a quantity for an extremely hazardous substance
described in this paragraph, a quantity of 1 pound shall
be deemed that quantity the release of which requires
notice as described in subsection (b) of this section.
EPCRA § 304(a)(2). Sulfur dioxide was included on the list of “ex- tremely hazardous substances” published by the Administrator pursu- ant to EPCRA § 302(a)(2). See Extremely Hazardous Substances List and Threshold Planning Quantities; Emergency Planning and Re- lease Notification Requirements; Final Rule, 32 Fed. Reg. 13,378 (Apr. 22, 1987). Sulfur dioxide is not a “hazardous substance” for purposes of CERCLA emergency notification requirements, although it has been proposed for inclusion under CERCLA. Designation of Extremely Haz- ardous Substances as CERCLA Hazardous Substances; Proposed Rule, 54 Fed. Reg. 3388 (Jan. 23, 1989). Further, the Agency has not promul- gated a final RQ for sulfur dioxide; therefore the one-pound statutory default RQ remains applicable for EPCRA reporting purposes. The Agency has proposed an RQ for sulfur dioxide of 100 pounds, but that rulemaking has not been completed. Reportable Quantity Adjustments; Proposed Rule, 54 Fed. Reg. 35,988 (Aug. 30, 1989). Upon learning of a release of an EPCRA “extremely hazardous substance” at or above the designated RQ, a facility must give immediate notice to the Local Emergency Planning Committee and to the State Emergency Planning Commission of any State likely to be affected by the release.’ EPCRA
§ 304(b)(1).
A. "Federally Permitted Release” Exemption
EPCRA exempts certain releases from reporting requirements, and specifically incorporates the “federally permitted release” exemptions of CERCLA § 101(10). EPCRA § 304(a)(2)(A). CERCLA § 101(10) defines “federally permitted releases” for purposes of exemption from CERCLA liability and CERCLA § 103 notification requirements. CERCLA § 101(10) sets forth 11 definitions of “federally permitted release” that are spe- cific to the particular permit program at issue. With respect to air re- leases, § 101(10)(H) exempts: 2p this case, as previously noted, Mobil gave timely notice to the State although it did so to comply with a State nuisance law rather than EPCRA. Supra, Part 1. ‘Therefore, the only violation alleged relates to reporting to the LEPC. VOLUME 5

MOBIL OIL CORPORATION 499
[A]ny emission into the air subject toa permit or control regulation under section 111, section 112, Title I part C, Title I part D, or State implementation plans submitted in accordance with section 110 of the Clean Air Act (and not disapproved by the Administrator of the Envi- ronmental Protection Agency), including any schedule or waiver granted, promulgated, or approved under these sections[.] CERCLA § 101(10)(H) (internal citations omitted) (emphasis added). One other § 101(10) subsection exempts releases “subject to” a per- mit; other program-specific subsections of § 101(10) exempt releases “in compliance with” a permit," “identified in a permit or permit ap- plication,” or “authorized under” certain federal or State programs."

Mobil contends that the various program-specific subsections of § 101(10) differentiate between releases “in compliance with” a permit and releases “subject to” a permit, and that the statutory exemption for air releases applies regardless of whether an emission is in compliance with the limits stated in the permit. As the Presiding Officer noted in his September 30, 1992, interlocutory order on Mobil’s motion, “liln other words, Respondent argues that the exception applies to any emission subject to a permit or control regulation, not just to the sub- set of such emissions that are in compliance with a permit or control regulation.” Interlocutory Order at 15. Mobil reasserts on appeal its contention that the March 12 sulfur dioxide release, being “subject to” the terms of an operating permit issued pursuant to New Jersey’s State Implementation Plan, was a “federally permitted release” exempt from EPCRA reporting requirements, regardless of whether the release was in compliance with or exceeded the permit limits.

Following a comprehensive and thoughtful analysis of the issue, the Presiding Officer rejected Mobil’s argument. Upon consideration of the guiding principles of statutory construction applied by the Presid-

CERCLA § 101(10)(B) (Clean Water Act National Pollutant Discharge Elimination System (NPDES) where release results “from circumstances identified and reviewed and made part of the public record * * * and subject toa condition of such permit.”) (emphasis added).

¥ Id. §§ 101(10XA) (NPDES); (D) (Clean Water Act § 404); (E) (Solid Waste Disposal Act); (F) (Marine Protection, Research, and Sanctuaries Act); (J) (releases to publicly owned treatment works); (K) (releases of source, special nuclear, or byproduct material)

°5 Jd, § 101(10XC) (discharges under NPDES permit where discharge is “identified in a permit or permit application”).

6 Id. §§ 101(10)(G) (underground injections under Safe Drinking Water Act); (D (injection of

materials related to development of crude oil or natural gas supplies). VOLUME 5

500 ENVIRONMENTAL ADMINISTRATIVE DECISIONS

ing Officer, and the structure and history of § 101(10), we are per- suaded that the Presiding Officer's conclusion was correct. We set forth below a brief summary of the Presiding Officer’s opinion, followed by our supplemental analyses of particular factors that we believe bear emphasis as providing support for the Presiding Officer’s conclusion. The Presiding Officer first analyzed the meaning of “federally permitted release” under the two-part test of Chevron U.S.A. Inc. v. NRDC, 467 U.S. 837 (1984). Applying the first prong of the Chevron test, the Presiding Officer considered whether the intent of Congress concerning the meaning of the “federally permitted release” exemp- tion for air releases was clear and unambiguous, in particular with respect to Congress’ use of the phrase “subject to” in § 101(10)(H). The Presiding Officer applied familiar and well-settled canons of statutory construction in his analysis. He found “persuasive” Mobil’s argument that Congress’ use of different phrases in different subsections of the same statute (“in compliance with” versus “subject to”) must be given due effect.'” However, the Presiding Officer concluded that application of that canon of construction did not resolve the issue, because the phrase “subject to” is inherently ambiguous. He further concluded that a “conflicting tool of statutory construction” ran contrary to Mobil’s argument—the principle that when the literal meaning of a term would “compel an odd result” then the legislative history must be examined to determine Congress’ intent. Interlocutory Order at 21. Because of these conflicting rules of construction, the Presiding Officer deter- mined that “Congressional intent should be divined by examining the legislative history and design of the act.” Id. at 22 (citing, e.g., DuBois

v. Thomas, 820 F.2d 943, 948-49 (8th Cir. 1987)).

The Presiding Officer quoted at length from the legislative history of CERCLA § 101(10).'* He then observed that “none of this legislative history offers clear illumination on the answer to the question of whether Congress intended a difference between the terms ‘in compli- ance with’ and ‘subject to’ as they are used in the definition of ‘feder- ally permitted release.” In the Presiding Officer’s view, selected por- tions of the legislative history could be read as supporting either Mobil’s or the Region’s position. Id. at 24,

"'Interlocutory Order at 21 (citing Russello v. United States, 464 US. 16 (1983)). See also Chi- cago v. Environmental Defense Fund, __ U.S. , 114 S.Ct. 1588, 1593 (1994) (“It is generally Presumed that Congress acts intentionally and purposely’ when it ‘includes particular language in one section of a statute but omits it in another.” (citing Keene Corp. v. United States, 308 US. , 113 S.Ct, 2035, 2040 (1993),

*S. Rep. No. 848, 96th Cong., 2d Sess, 49 (1980).

VOLUME 5

MOBIL OIL CORPORATION 501

Because of the lack of clarity in Congress’ meaning, the Presiding Officer proceeded to the second prong of the Chevron analysis: whether the Agency’s interpretation, as expressed in four Notices of Proposed Rulemaking (NPRMs) under CERCLA,” represented a permissible con- struction of the statute. However, the Presiding Officer questioned whether Chevron deference should apply to the Agency’s interpreta- tion, in view of the fact that the Agency had never promulgated a final rule or definitive construction of the phrase. Because of the “dubious- ness of applying Chevron deference to the Agency’s interpretation,” the Presiding Officer applied the analysis set forth in Skidmore v. Swift
& Co., 323 U.S. 134 (1944). Interlocutory Order at 32. Under Skidmore, the weight to be given an Agency’s interpretation depends upon the thoroughness of the Agency’s evaluation of the issue, the validity of the Agency’s reasoning, and consistency with other pronouncements. 323 U.S. at 140.

The Presiding Officer found that EPA had given the “federally permitted release” exemption issue for Clean Air Act releases careful consideration. He noted that the general issue of federally permitted releases was the subject of four Agency NPRMs, and that the Agency had sought input by way of public comments. He determined that the Agency’s position in the case against Mobil was consistent with its July 1988 NPRM, in which the Agency stated that in order for the Clean Air Act “federally permitted release” exemption to apply, the emission

The Agency first issued a NPRM relating to the “federally permitted release” issue in 1983. 48 Fed. Reg. 23,552 (May 25, 1983). The NPRM expressed uncertainty as to the intended scope of the exemption, particularly “the extent to which emissions from permitted stationary sources, if they contain CERCLA-designated substances, qualify for the federally permitted release exemption.” Id. at 23,557.

The Agency issued its second NPRM relating to the federally permitted release issue in 1985. 50 Fed. Reg. 13,456 (April 4, 1985). In that NPRM, the Agency stated only that “[d]ue to the complex- ity of the issues involved, the Agency has decided to study the scope of this exemption further: today’s rule does not resolve the “federally permitted release’ issue.” Id. at 13,458.

In 1988, the Agency issued a NPRM that proposed to interpret the provisions of CERCLA § 101(10), including the exemption for air releases. 53 Fed. Reg. 27,268 (July 19, 1988). In this NPRM, the Agency said that § 101(10)(H):

[Clannot be read broadly to cover any and all types of air emis-

sions. ** * [Flor the exemption to apply, the emission must be in

compliance with the applicable permit or control regulation. Id. at 27,273 (emphasis added).

‘The Agency issued a supplemental NPRM in 1989 that “clarified” that the Agency “would ex- empt as federally permitted under CERCLA section 101(10)H) only those air releases in compliance with a CAA permit or control regulation that specifically identifies the hazardous substance and is specifically designed to limit or eliminate the emission of that hazardous substance.” 54 Fed. Reg. 29,306 Guly 11, 1989).

VOLUME 5

502 ENVIRONMENTAL ADMINISTRATIVE DECISIONS

must be in compliance with an applicable permit or control regula- tion. Interlocutory Order at 34 (citing 53 Fed. Reg. at 27,273).?° The Presiding Officer further determined that the Agency’s position was not inconsistent with its earlier NPRMs concerning the scope of the exemption. The Presiding Officer found the reasoning underlying EPA’s interpretation, as expressed in the 1988 NPRM, to be “unassailable,” particularly in light of the purposes of CERCLA and EPCRA as reflected in the legislative history of each statute. Id. at 35.7!

The Presiding Officer explained that a major purpose of the CERCLA
§ 103 notification requirement is to notify the government of releases of hazardous substances that may harm public health or the environ- ment, and that may require immediate response. He noted language in the Senate report which stated that the exemptions provided in § 101(10) “are not to operate to create gaps in actions necessary to protect the public or the environment.” Id. at 36.

The Presiding Officer then considered the purpose of EPCRA as explained in the legislative history. According to a Senate report on the draft legislation, one purpose was to provide for immediate local no- tification of releases, since CERCLA reporting to the federal National Response Center did not guarantee that local authorities would receive the swift notification necessary to respond effectively to an emergency. See S. Rep. No. 99-11, 99th Cong., Ist Sess. 8 (1985). The Presiding Officer noted that air releases were of particular concern to the draft- ers of EPCRA, in light of incidents of catastrophic air releases at Bhopal, India, and Institute, West Virginia. Interlocutory Order at 39.

The Presiding Officer found that the broad interpretation of the exemption language urged by Mobil created a “gap” in the government's

• While the Presiding Officer did not specifically focus on the subsequent 1989 NPRM, we
  note that that NPRM reiterated the Agency's position that air releases must be “in compliance with” a Clean Air Act permit in order to fall within the exemption. See supra n. 19.

2In the 1988 NPRM, the Agency stated that a “straightforward interpretation” and the “plain language” of the statute led to the conclusion that a release in excess of permit levels is “not “in compliance with’ the permit and cannot be “federally permitted.” 53 Fed, Reg, at 27,269. The Agency expressed the view that its interpretation was “essential to ensure adequate protection of public health and the environment.” Id. It explained that because of differences in release notification requirements under various permit programs, the goals of CERCLA notification — immediate noti- fication of releases to a central office — might not be served by reliance on other permit programs or permit conditions. Id. The Agency also stated that because the legislative history of § 101(10)(H) referred to air controls designed to limit or eliminate hazardous air emissions, the exemption for ait releases could not be read broadly to cover air emissions not in compliance with a permit or control regulation. Id. at 27,273, According to the Agency, a broad interpretation of the exemption would exempt “dangerous episodic releases” from CERCLA reporting, and would also be inconsistent with the purposes of EPCRA, which was enacted specifically to address harmful air releases. Id. VOLUME 5

MOBIL OIL CORPORATION 503
ability to respond to air releases of hazardous substances. He noted that neither the Clean Air Act, regulations under the Act, nor Mobil’s permit included immediate reporting requirements of the type intended by Congress when it created the “federally permitted release” exemp- tion under CERCLA. The Presiding Officer found that Mobil’s interpre- tation was “directly contrary to EPCRA’s legislative purpose.” Inter- locutory Order at 42. Accordingly, he rejected Mobil’s interpretation because it would leave the “public and the environment dangerously unprotected from harm inherent to toxic chemical air releases.” Id. at

43. In contrast, the Presiding Officer found EPA’s interpretation to be
    “eminently reasonable and consistent with the purposes of both CERCLA and EPCRA.” Id. He therefore concluded that Mobil’s releases were not exempt from the requirements of EPCRA and CERCLA.

On appeal, Mobil reiterates its argument that the meaning of “sub- ject to” as used in CERCLA § 101(10)(H) is plain and unambiguous, and encompasses releases that exceed permitted emissions limits. Mobil emphasizes that Congress used the phrase “in compliance with” in other § 101(10) subsections, and that Congress’ use of the two distinct phrases must be given due effect.

As the Presiding Officer observed, the statutory language of
§ 101(10)(H) offers no insight as to the scope of the exemption created for air releases “subject to” the specified air permit programs or control regulations. We agree with the Presiding Officer that the phrase “sub- ject to” is inherently ambiguous, and has little independent meaning. Moreover, simply viewing the context of subsection (H) in light of the surrounding subsections does little to clarify the ambiguity. Thus, we must look to the legislative history of § 101(10) and the purposes of CERCLA and EPCRA to determine the appropriate meaning.

We recognize, as did the Presiding Officer, the familiar rule that when “Congress includes particular language in one section of a stat- ute but omits it in another section of the same Act, it is generally presumed that Congress acts intentionally and purposely in the dispar- ate inclusion or exclusion.” Chicago v. Environmental Defense Fund,
US. at , 114 S.Ct. at 1593; Russello v. United States, 464 U.S. 16, 23 (1983). Thus, we must presume that there is a reason why “subject to” appears in some subsections of § 101(10), and why Congress used other phrases in other subsections. See supra n. 13-16. However, Mobil’s argument adds an additional presumption to this familiar canon: that when Congress uses dissimilar language the phrases chosen can never under any circumstances have similar meaning, even though they may be used in different contexts for different purposes.

VOLUME 5

504 ENVIRONMENTAL ADMINISTRATIVE DECISIONS

In this instance, Mobil’s argument is flawed. Although it is pre- sumed that Congress uses different phrases intentionally and purposely, that canon alone does not necessarily resolve the specific meaning Congress intended to ascribe to the program-specific language in the various § 101(10) subsections. The terms used by Congress in the § 101(10) subsections (“in compliance with,” “subject to,” “identified in,” “authorized under”) are not statutorily-defined terms having particular and differentiated regulatory meaning. Compare with Chicago, __ U.S. at ___, 114 S.Ct. at 1588 (RCRA § 3001(i) exemption for “treating,” “storing,” “disposing of,” or “otherwise managing” certain hazardous wastes did not also exempt “generating” of hazardous wastes, where the regulatory terms used described specific and defined activities, and the defined term “generating” was excluded from § 3001) but expressly made part of a different exemption.).

Based upon our review of the statutory language and the pur- poses of EPCRA and CERCLA, as well as EPCRA’s and CERCLA’s legis- lative history, we believe that Congress’ use of the different terms stems from Congress’ efforts to accommodate the wide range of regu- latory methods and control measures falling under the broad rubric “federally permitted” for eleven different regulatory schemes, and that use of the “subject to” term in § 101(10) had purely pragmatic pur- poses. For instance, “subject to” was incorporated in subsection (B) (the exemption for certain discharges into waters of the United States) simply because that language was drawn verbatim from Clean Water Act § 311. See S. Rep. No. 848 at 47. With respect to the exemption in subsection (H) (for air releases), Congress took note of the fact that “{iJn the Clean Air Act, unlike some other Federal regulatory statutes, the control of hazardous air pollutant emissions can be achieved through a variety of means,” including emissions limits, technology require- ments, work practices and the like. Id. at 49. Significantly, the Clean Air Act “federally permitted release” exemption is the only § 101(10) pro- vision that expressly includes “waivers” from regulatory obligations as a means of exemption. § 101(10)(H) (“federally permitted release’ means * * * any emission into the air subject to a permit or control regulation under [the Clean Air Act] * * * including any schedule or waiver granted, promulgated, or approved under these sections.”) (em- phasis added). Thus, the phrase “subject to” certainly may be given a full meaning without accepting Mobil’s interpretation of the phrase, and Mobil’s contention that its broad reading of “subject to” is the only logical reading is not supported. Certainly Congress’ decision to use the phrase “subject to” rather than “in compliance with” in a context where a waiver specifically authorizes non-compliance with Clean Air Act requirements is logical. Contrary to Mobil’s contention, it is simply not apparent that Congress, by using the phrase “subject to” in § VOLUME 5

MOBIL OIL CORPORATION 505,
101(10)(H), intended to adopt a substantially broader scope for the “federally permitted release” exemption as applied to the air program than as applied to other programs, encompassing releases that are not in conformance with permits or other Clean Air Act requirements. Rather, as discussed below, the legislative purposes and histories of both EPCRA and CERCLA suggest that only those air releases in con- formity with the Clean Air Act’s requirements are exempted from EPCRA and CERCLA reporting obligations and liabilities as “federally permit- ted releases.”

The Presiding Officer concluded, and we agree, that it is appropri- ate to examine the statutory purposes of EPCRA and CERCLA in order to aid our interpretation of the scope of the “federally permitted re- lease” exemption. See Interlocutory Order at 35 (quoting Cabel v. Markbam, 148 F.2d 737, 739 (2d Cir.), aff'd 326 U.S. 404 (1945) (“lelven though the words used, even in their literal sense, are the primary, and ordinarily the most reliable, source of interpreting the meaning of any writing, it is one of the surest indexes of a mature and developed jurisprudence not to make a fortress out of a dictionary; but to remem- ber that statutes always have some purpose or object to accomplish, whose sympathetic and imaginative discovery is the surest guide to their meaning.”)). The statutory purposes are particularly relevant when the terms to be interpreted do not have a specific technical or defined meaning. Burnet v. Chicago Portrait Co., 285 U.S. 1, 6 (1932) (where a term used in a statute is not “technical or artificial,” then “the sense in which it is used in a statute must be determined by reference to the purpose of the particular legislation.”); American Mining Congress v. EPA, 824 F.2d 1177, 1185 (D.C. Cir. 1987) (anaylsis of statutory term “discarded” requires more than resort to everday meaning of term; term must be considered in context of legislative purpose, citing Burnet, 285 U.S. at 6). In short, because the words “subject to” can be construed narrowly to only cover releases in conformity with Clean Air Act requirements or more broadly to include non- conforming air releases, we would be remiss in our interpretive duties if, as Mobil suggests,? we ignored the purposes underlying

• Mobil contends that “It}he meaning of a particular provision in a detailed statute is not dis-
  Placed by reference to broad purposes of the legislation as a whole.” Appellant's Brief at 42 (citing Board of Governors of the Fed. Reserve Sys. v. Dimension Fin. Corp., 474 U.S. 361 (1986); American Mining Congress v. EPA, 824 F.2d 1177, 1185 (D.C. Cir. 1987)). We recognize that it may be inappro- Priate to rely on general statutory purposes to supplant the plain meaning of statutory language. As noted, however, the specific meaning of the terms at issue here is not clear, Further, in the cases cited by Mobil the courts were evaluating attempts to define statutory terms where the intent of Congress in using the terms was reasonably clear. See Dimension Fin. Corp, 474 US. 361 (Federal Reserve Board's attempts to enact regulation expanding explicit and detailed statutory definition of “bank”); American Mining Congress, 824 F.2d 1177 (EPA's attempt to broadly define “discarded” under RCRA where Congress intended narrower meaning). In addition, American Mining Congress specifically recognized the relevance of legislative purpose to statutory interpretation. Id. at 1185.

VOLUME 5

506 ENVIRONMENTAL ADMINISTRATIVE DECISIONS
enactment of EPCRA and CERCLA.”

Based upon our review, we can find no persuasive support for Mobil’s contention that Congress intended to create the sweeping exemption for air releases propounded by Mobil. It is obvious that EPCRA was intended to build on CERCLA’s emergency reporting provisions by ensuring that, in addition to notification to federal authorities, the State and local authori- ties best able to respond to an emergency release received immediate notification of a release.“In enacting EPCRA, Congress expressed particu- lar concern for prompt notification of releases of hazardous substances to the air. See, e.g., 131 Cong. Rec. 23,947 (Sept. 17, 1985) (comments of Senator Byrd regarding delayed local notification of hazardous air release in Institute, West Virginia), id. at 24,339-356 (Sept. 19, 1985) (comments of Senator Heinz relating to catastrophic air releases in West Virginia and Bhopal, India); id. at 23,947 (Sept. 17, 1985) (comments of Senator Lautenberg con- cerning need to minimize risks associated with chemical releases to air).

Based on the particular concern for air releases that prompted enact- ment of EPCRA, we conclude that in the context of emergency reporting,

See Friedrich v. City of Chicago, 888 F.2d 511, 514 (7th Cir. 1989) vacated on other grounds US. __, 111 8.Ct. 1383 4991):

When a court can figure out what Congress probably was driv- ing at and how its goal can be achieved, it is not usurpation — it is interpretation in a sense that has been orthodox since Aristotle
— for the court to complete (not enlarge) the statute by reading it to bring about the end that the legislators would have speci- fied had they though about it more clearly or used a more per- spicuous form of words.

8. Rep. No. 99-11, 99th Cong,, Ist Sess. 8 (1985) (“One problem that has emerged [with
   CERCLA], however, is that notification of the [National Response Center] may not be relayed quickly enough back to the State and local authorities who must provide the first line of emergency re- sponse. The reported bill corrects the problem by requiring immediate direct notification of State and local emergency response officialsl.]").

• The Agency recognized this specific concern for air releases in promulgating its interim final
  rule under EPCRA:

Particularly after the Bhopal, India disaster of December, 1984,

it became clear that substances which are highly acutely toxic

and have a high potential for becoming airborne posed a spe-

cial problem for emergency response.
Emergency Planning and Community Right to Know Programs, Interim Final Rule, 51 Fed. Reg. 41,570 (Nov. 17, 1986). Indeed, EPCRA's list of “extremely hazardous substances” was developed as. part of the Agency's Chemical Emergency Preparedness Program initiated under CERCLA and the Toxic Substances Control Act, and announced as part of the Agency's Air Toxics Strategy for ad- dressing air releases of toxic substances. Id. at 41,572. VOLUME 5

MOBIL OIL CORPORATION 507
conformity with Clean Air Act regulatory requirements is the construction Congress apparently envisioned when it incorporated the “federally per- mitted release” exemption of CERCLA § 101(10)(H) into EPCRA without change. Any other construction would be wholly at odds with the pur- pose of EPCRA. To adopt Mobil’s argument that any noncomplying air release triggers the exemption so long as the pollutant released is ad- dressed in some way in a permit or other Clean Air Act requirement would mean that potentially significant air releases would be exempt from EPCRA reporting obligations, regardless of the extent of the noncompliance or resulting environmental harm. Had Congress intended to create this radi- cally different and extreme exemption for air releases, it would have said so in much more explicit terms. In light of Congress’ concern for air releases, it is unlikely that Congress would have carried forward the CERCLA “federally permitted release” language without change unless it felt that this language would also serve the purposes of EPCRA. There is nothing to suggest that in enacting EPCRA, Congress would have created a much broader exemption for the very type of release that motivated its enactment. Had Congress believed that the exemption would apply in the manner advanced by Mobil, it could easily (and likely would) have re- vised the exemption or created other exemption language, much in the way it used language different from CERCLA in defining other key terms.”

Moreover, the legislative purpose and history of CERCLA § 101(10) also contradict Mobil’s claim. As Mobil’s counsel conceded at oral argu- ment, Congress’ focus in creating the “federally permitted release” ex- emption in CERCLA (from which the EPCRA exemption derives) was on the liability aspects of CERCLA.” In that context, Congress wanted to avoid subjecting releases that were expressly allowed by other federal regulatory programs to the CERCLA liability scheme.” Its discussion of the issue was framed purely in terms of “compliance” with permit

>For example, the definition of “facility” under EPCRA is quite different from that under CERCLA.
Compare EPCRA § 329(4) with CERCLA § 101(9).

• Counsel for Mobil stated: “You have to start with the fact that it’s [glot a different liability
  scheme, as well. Because that’s where they were starting from. And that’s why (CERCLA §] 107()) is. in there. This is, after all, a liability provision.” Oral Arg, Tr. at 24.

2] is important to remember that CERCLA established strict and retroactive liability for releases of hazardous substances, without regard to whether a release was otherwise prohibited by law. Be- cause of the far-reaching scope of CERCLA liability, Congress believed that it was appropriate to pro- vide some relief from liability where a release was specifically authorized by a federal permit or regu- latory program. See S. Rep. No. 848 at 46; see also Babich, Restructuring Environmental Law, 19 ELR 10,057 (Feb. 1989) (“To reconcile its new, liability-based approach with existing regulatory programs, Congress carved an exception to liability under CERCLA for “federally permitted releases’ of hazard- ous substances. Under this exception, polluters are not liable for contamination authorized by permits issued under federal, and some state, environmental laws. The provision encourages polluters to obtain and comply with permits, enhancing the effectiveness of the regulatory system.”).

VOLUME 5

508 ENVIRONMENTAL ADMINISTRATIVE DECISIONS
requirements.” Nothing that Congress said or did suggests that non- complying facilities should also be exempt from CERCLA liability. Exemption from CERCLA reporting obligations simply flowed from the liability exemption, and the scope of the exemption should be viewed in that light.

Our conclusion that a release “subject to” Clean Air Act regu- latory requirements must be in conformance with those require- ments in order to be exempt from EPCRA and CERCLA emergency reporting provisions is consistent with the positions expressed in the Agency’s NPRMs addressing the scope of the exemption. How- ever, our determination is based on our independent review and analysis of the issue. Moreover, since counsel for Mobil acknowl- edged that Mobil had actual knowledge of the Agency’s interpreta- tion of § 101(10)(H), and that it acted in accordance with that interpretation in reporting the March 12 release, Oral Arg. Tr. at 19,

» For example, the first paragraph of the Senate report addressing the “federally permitted release” exemption states as follows:

Adoption of an express Federal liability mechanism for releases pursuant to certain Federal permit systems seemed inappropri- ate at this time. Such a proposition was particularly troublesome for permits under section 402 of the Clean Water Act because the Congress had excluded them from the Act's section 311 li- ability provisions in 1978. The rule of common law is that com- pliance with a permit is not a defense to liability. Moreover, Congress has never said or suggested that a Federal permit amounts to a license to create threats to public health or the environment with legal immunity. However, in view of the large sums of money spent to comply with specific regulatory pro- grams, liability for federally permitted releases ought to be de- termined based on the facts of each individual case. Therefore, the reported bill authorizes response to federally permitted re- leases, but requires costs to be assessed against the permit holder under the liability provisions of other laws, not S. 1480. S. Rep. No. 848 at 46 (emphasis added).

Indeed, there is evidence that even where Congress used different language in the § 101(10) subsections, Congress envisioned that regulatory conformity would provide the benchmark. CERCLA
§ 101(10)(G) exempts releases “authorized under” federal or State underground injection control programs and omits any reference to “compliance” with those programs, yet the legislative history makes clear that “compliance” is necessary in order to obtain the exemption. S. Rep. No. 848 at 49 (This provision exempts a release to the substrata of the Earth * * * if such release is authorized and in compliance with the conditions of such authorization granted under the Underground Injection Control Program of the Safe Drinking Water Act.”) (emphasis added).

VOLUME 5

MOBIL OIL CORPORATION 509
the facts here simply do not support a due process claim predi- cated on lack of fair notice, as Mobil has asserted.>” #!

B. “Immediacy” of Mobil’s Report
The duty to report under EPCRA arises as soon as the facility

personnel have knowledge that a reportable release has occurred, or should know of such a release. See In re Genicom Corp., EPCRA-III- 057 (ALJ, July 16, 1992), aff'd In re Genicom Corp., EPCRA No. 92-2 (EAB, Dec. 15, 1992); 52 Fed. Reg. 13,378, 13,393 (Apr. 22, 1987) (explaining that “knowledge” of a release under EPCRA includes con- structive knowledge as well as actual knowledge). Further, “knowl- edge” does not necessarily mean conclusive knowledge of the exact quantity of a release. As the Presiding Officer stated in Genicom:

What is at issue is when did Genicom have enough

information that it could reasonably be said that it knew

that the releases were at or above reportable quantities

even though it did not know the exact quantities re-

leased. A company should be given some latitude about

how it interprets the information it has. At some point,

however, the nature of the information can be such that

the failure to give notice is indicative of the company

» Because the Board serves as the final decisionmaker for the Agency, the concepts of Chev- ron and Skidmore deference do not apply to its deliberations. See, e.g., In re City of Detroit, TSCA Appeal No. 89-5, n. 8 (CJO, July 9, 1991); In re Louisville Gas & Electric Co., NPDES Appeal No. 81-

3. GO, Sept. 24, 1981) (decisionmaking units of Agency are not bound by standards of judicial re-
   view of Agency action). Because we have not applied a deferential standard of review to the Agency's interpretations, it is unnecessary for us to address Mobil's contention that in accordance with the D.C. Circuit's decision in Kelley v. EPA, 15 F.3d 1100 (D.C. Cir. 1994), the Agency has no authority to issue interpretive or legislative regulations under the liability provisions of CERCLA, or that defer- ence to the Agency’s interpretation under Skidmore v. Swift is inappropriate because Skidmore deference requires that the Agency demonstrate consistency and thoroughness in its interpretation, and the Agency's NPRMs in this case have not been totally consistent.

31 Mobil makes one additional argument to support its contention that it did not violate EPCRA reporting requirements in this case. In a footnote in its reply brief to Region I's brief on appeal, Mobil contends that the March 12, 1990 release did not trigger EPCRA reporting requirements because re- portable releases under EPCRA are determined on a facility-wide rather than point-of-release basis, and therefore facility-wide emissions limits must be aggregated in order to determine whether a report is required. Mobil’s Reply Brief at 6, n.4. Mobil argues that because the facility contained five permitted sources of sulfur dioxide emissions with an aggregate permitted emissions limit of 3000 pounds per hour, the March 12 release did not exceed “national ambient air standards.” Id.

Mobil’s argument was prompted by the amicus brief of Ashland Petroleum Company, and is premised on the fact that the terms “facility” and “release” are defined differently under CERCLA and EPCRA. However, this is not an argument that was made by Mobil when the case was before the Presiding Officer; Mobil raises the issue for the first time on appeal. Accordingly, the Board can, and does, exercise its discretion to decline to consider the argument now. See In re ALM Corp., TSCA No. 90-4 at 11.12 (EAB, Oct. 11, 1991).

VOLUME 5

510 ENVIRONMENTAL ADMINISTRATIVE DECISIONS
not knowing the requirements or being hostile or indif-
ferent to them, rather than of any uncertainty that a
release in reportable quantities had taken place.
Genicom, EPCRA-III-057 at 13.

Mobil contends that it did not become aware that a reportable emission had occurred until its process engineer, Ms. Murphy, com- pleted the third and final set of calculations on March 22. The Region contends that all data necessary to determine the existence of a report- able release were available to Mobil on March 12. The dispute thus centers on the availability and choice of data necessary to perform calculations regarding the sulfur dioxide emissions, and the time re- quired to complete the calculations. In addition, Mobil contends that the Presiding Officer erroneously imposed “preliminary reporting” and monitoring requirements upon Mobil in concluding that Mobil could have reported the release before March 22.

We agree with the Presiding Officer's conclusion that Mobil could have reported the release before March 22, but we reject the Region’s contention that Mobil had all data necessary on March 12 to confirm that a reportable release had occurred. Mobil had available on the day of the release data in the form of six-minute sulfur dioxide average concentrations from the PMS system that, when combined with data on the known air flows, showed that Mobil was very close to (but not over) its permit limit. The inclusion of the remaining unknown flows would only serve to increase the emissions volume (although the ex- tent of the increase was unknown on March 12). Although the 6-minute data were available to Mobil, and were in fact examined by Mr. Rodack immediately following the release, Mobil’s process engineer did not compare the results of her first calculation using the hourly-average data with the six-minute data, nor did she examine the one-minute average PI data until the second iteration of her calculations indicated a permit exceedance. In view of the apparently unusual occurrence— an odor complaint—prompting the data examination, and the poten- tial that emergency notification requirements had been triggered, pru- dent practice would dictate careful scrutiny of all available data. Such scrutiny would have suggested that Mobil was precipitously close to its permit limit, and, at a minimum, that immediate, urgent further investigation was necessary.

Mobil emphasizes that the missing flow data were needed to con- clusively determine that a reportable release had occurred, and that the length of time necessary to determine how to estimate these data justified the delay in reporting. While we agree that such data were VOLUME 5

MOBIL OIL CORPORATION 511
relevant, Mobil should have taken all action necessary to expedite the gathering of this information. Although Mobil eventually determined that these flows comprised only about 10% of the total flow volume, it did not know this until after the data had been gathered. The unknown flows could have proven to be much more significant to the emissions volume, yet Mobil did nothing to ensure that they were collected as speedily as possible.

Moreover, we agree with the Presiding Officer that there is little evidence that Mobil acted with any particular urgency in completing the investigation and calculations necessary to finally determine that a permit exceedance had occurred. Mobil assigned the task of complet- ing the necessary research and calculations to the process engineer whose primary responsibility was the time-consuming task of com- pleting the regeneration of SRU-3 and bringing the unit back on-line. According to Ms. Murphy’s testimony, this work consumed the bulk of the five days following the release (approximately two days to com- plete the burn, and two to three days to change out the catalyst). See Tr. at 507; 587. Mobil confirmed at oral argument that Ms. Murphy focused initially on completing the regeneration process and only then shifted back to calculating the magnitude of the release. Oral Arg. Tr. at 10. Mobil acknowledged that its primary concern was getting the SRU back on-line because it had only one SRU running during the regeneration and thus did not have any back-up capability if anything happened to that unit. As previously noted, however, the shutdown of SRU-3 did not in itself prevent the refinery from operating. See Oral Arg. Tr. at 9.

While completing the regeneration process may have been impor- tant to Mobil, this should not have taken precedence over its legal obligations. If Ms. Murphy was preoccupied with getting SRU-3 back on-line, it is reasonable to have expected that, during this time, an- other Mobil engineer be assigned the task of gathering the unknown flow data necessary to refine and complete the calculations. The addi- tional flow data do not appear to have been exclusively within Ms. Murphy’s zone of expertise in fact, she testified that she had to resort to textbooks and plant energy usage data to develop the calculations. Mr. Rodack was himself a registered professional engineer, and he supervised other engineers at the refinery. Tr. at 419-421.

In light of the evidence as a whole, it appears that, while acknowl- edging that a permit exceedance may have occurred, Mobil opted to carry on business as usual and perform its investigation of the release in due course. However, Mobil cannot, consistent with EPCRA, shield itself from gaining knowledge of a reportable release by putting the

VOLUME 5

512 ENVIRONMENTAL ADMINISTRATIVE DECISIONS

investigation of a release on a slow track while it carries on normal plant operations. Having the data to suggest on March 12 that its per- mit limit may have been exceeded, and having apparently formed some belief that further investigation was necessary, it was incumbent on Mobil to act expeditiously to confirm or negate the fact of a report- able release. We recognize that EPCRA does not require reporting be- fore a facility has some degree of certainty that a reportable release has occurred. See Genicom, EPCRA III-057, at 13. What we reject is the notion that facilities are free to place the acquisition of certainty on a timetable that is convenient for the facility.

Even if we accept Mobil’s claim that it gathered all required data and performed all calculations as quickly as possible, it is undisputed that Mobil’s second set of calculations did show a permit exceedance for the ten o’clock hour, and that those calculations were completed on the seventh or eighth day following the release. Oral Arg. Tr. at 9- 10, Region II’s Reply Brief at 10. Yet Mobil delayed reporting until it could complete a third set of calculations to confirm the exceedance. According to Ms. Murphy, the third calculation was made using differ- ent sulfur dioxide concentration data than the first two calculations because she was concerned about the validity of the data used in the first two calculations. Tr. at 569. However, as noted in Genicom, such exactitude is unnecessary for purposes of gaining knowledge of a reportable release; once Mobil’s process engineer confirmed a likely permit exceedance in her calculations on the seventh or eighth day following the release, Mobil was not justified in further delaying its report to the LEPC while it fine-tuned its calculations, possibly in an attempt to show that a violation did not in fact occur. See Genicom at 13.

Based on the evidence adduced in this case, we conclude that Mobil had sufficient knowledge of a permit exceedance to report the release on March 19 or 20, rather than waiting until March 22, based solely on the limited resources it applied to the permit exceedance determination. However, we further conclude that had Mobil employed in a timely manner additional technical resources (for example, by assigning another engineer to perform the calculations), it could have completed the investigation and calculations necessary to confirm the permit exceedance even earlier than March 19 or 20. Since Ms. Murphy apparently completed the first set of calculations by no later than the morning of the second day and appears to have spent approximately two or three days in performing the second calculation following completion of the regeneration,* we conclude that Mobil could have,

2 As noted, Ms. Murphy testified that she believed she began the first calculation on the day of the release, Tr. at 515. The regeneration required about 24 to 36 hours to complete the “burn” phase, and about two to three days to replace the catalyst. Tr. at 507, 587.

VOLUME 5

MOBIL OIL CORPORATION 513
with proper diligence, completed the second calculation by no later than March 17, five days earlier than the actual report was made. We thus conclude that Mobil could, and should, have had knowledge sufficient to enable it to report the release to the LEPC at least two days earlier than the date upon which the Presiding Officer concluded the report could have been made.

We also reject Mobil’s other claims of error by the Presiding Of- ficer, Mobil contends that the Presiding Officer erred by suggesting that it was “reasonable” to expect Mobil to make a preliminary report of the release before Mobil had gained knowledge that the release had actually exceeded permit levels. It is true that the Presiding Officer stated that:

In sum, where the preliminary data that {were] avail-
able at the time of the release showed a release so
close to the reportable quantities, it could be found
that Mobil should have at least made the preliminary
oral report to the LEPC on March 12.
Initial Decision at 20. However, the Presiding Officer went on to say that: While this interpretation of EPCRA would be reason- able, it is not so clear from the wording of the statute as to provide a basis for assessing the large penalty that the EPA proposes. Id.\t is clear from the Initial Decision that the failure to provide a “preliminary report” did not form the basis for the Presiding Officer’s determination that Mobil violated EPCRA. Rather, as explained above, the finding of a violation was based on the Presiding Officer's consid- eration of the data available to Mobil concerning the release, and the time needed to make the calculations necessary to establish that a reportable release had occurred. Our decision is based entirely on these considerations as well. Accordingly, the correctness of the Pre- siding Officer's discussion of any preliminary reporting obligation under EPCRA is not an issue that needs to be resolved in this appeal. Mobil makes a similar argument with respect to the Presiding Officer’s conclusion that “Mobil should at least have taken reasonable Die have been forced to use approximate timeframes due to the vagueness of Ms. Murphy's recollections, but our calculations are, if anything, generous to Mobil. is not entirely clear from the Initial Decision how the Presiding Officer selected seven days from the date of the release as the reporting trigger. VOLUME 5

514 ENVIRONMENTAL ADMINISTRATIVE DECISIONS
steps to insure that it would be able to report promptly in the event that an emission in reportable quantities did occur.” Initial Decision at

17. Mobil argues that the Presiding Officer erred by imposing a “moni-
    toring” requirement on Mobil, because such requirements are not part of EPCRA. Mobil further argues that imposition of such requirements violates the “fair notice” requirements of due process. However, the Presiding Officer acknowledged that “Mobil is given great latitude as to what is reasonable, of course, since there are no monitoring re- quirements.” Id. at 17-18 (emphasis added).** Moreover, the Presiding Officer’s conclusions regarding the existence of an EPCRA violation did not hinge on a failure to preplan or monitor, but on a failure to utilize all data readily available to Mobil by virtue of the processes it did have in place, and to complete expeditiously the calculations nec- essary to determine whether the release must be reported. Because the Presiding Officer did not base his ultimate conclusions on any failure to preplan or monitor (nor do we), and the Region concedes that EPCRA imposes no monitoring requirements, we reject Mobil’s claim of error.® C. Calculation of the Penalty Amount Mobil raises several objections to the Presiding Officer’s applica- tion of the EPCRA Penalty Policy. Mobil contends that the Presiding Officer failed to properly address the specific statutory penalty criteria, but made only a “cursory” analysis under the Penalty Policy. On this point, we disagree. EPCRA § 325(b)(1)(C) provides that in determining the amount of any Class I penalty assessed for a violation of § 304, the presiding officer shall consider the “nature, circumstances, extent and gravity of the violation or violations and, with respect to the violator, ability to pay, any prior history of such violations, the degree of culpability, economic benefit or savings (if any) resulting from the violation, and other matters as justice may require.” As we have explained: At the oral argument of this appeal, counsel for the Region unequivocally agreed that EPCRA imposes no monitoring requirements: JUDGE FIRESTONE: Let's see if we can clarify this first. [Are] there any monitoring obligations * * * established by EPCRA? MS. BINDER: No. EPCRA requires immediate emergency report- ing. It does not tell a facility how it’s going to get to the point where it has knowledgel.]

Oral Arg. Tr. at 43-44 (emphasis added).

While it is clear that EPCRA does not impose any monitoring requirement, we express no opinion as to whether “preplanning” requirements fall within the scope of “monitoring” require- ments and thus are similarly beyond the scope of EPCRA.

VOLUME 5

MOBIL OIL CORPORATION 515

The Agency has issued penalty policies to create a frame-

work whereby the decisionmaker can apply his discre-

tion to the statutorily-prescribed penalty factors, thus

facilitating the uniform application of these factors. * * *

As we stated in the Genicom decision, a penalty policy

‘reasonably implements the statutory criteria, with a

range of penalties to reflect differing circumstances.’

*** Therefore, a presiding officer may properly refer to

such a policy as a means of explaining how he arrived

at his penalty determination. * * * Agency regulations

require that a presiding officer consider any penalty

policy issued under the Act, although they do not man-

date that he adhere to it.
In Re Great Lakes Div. of Nat'l Steel Corp., EPCRA Appeal No. 93-3, at 23-24 (citations omitted). A Presiding officer may satisfy his or her duty of articulating the basis for a penalty by referring to the Penalty Policy and explaining how the facts fit the policy. See In re Sandoz, Inc., RCRA (3008) No. 85-7 at 8 n.11 (CJO, Feb., 27, 1987). In this instance, the Presiding Officer did not err because he placed his analysis within the framework of the Penalty Policy and did not “walk through” each of the statutory factors.

If the penalty assessed by a presiding officer falls within the range of penalties determined through proper application of a penalty policy, the Board will not usually substitute its judgment for that of the presid- ing officer. Great Lakes, at 24. However, for the reasons explained below, we conclude that the penalty assessed by the Presiding Officer in this case was not consistent with the Penalty Policy in certain re- spects, and must be revised.

The Presiding Officer accepted the Agency’s recommendation that the violation should be classified as Level I in “extent” (the highest level), due to Mobil’s failure to notify the LEPC within two hours after knowledge of the release. We agree that this determination is consis- tent with the EPCRA statutory penalty factors and the Penalty Policy. See Penalty Policy at 11 (Level I is appropriate for violations of § 304 where there has been “{nlo notification to the appropriate SERC(s) and LEPC(s) within 2 hours after the owner or operator had knowledge of

the release unless extenuating circumstances existed that prevented notification.”) (emphasis added). Although Mobil did provide prompt notification of the release to the SERC, EPCRA mandates immediate notification to the local emergency response authorities as well. In- deed, notification to local authorities is of paramount importance, since it is local personnel who are likely to be in the best position to provide VOLUME 5

516 ENVIRONMENTAL ADMINISTRATIVE DECISIONS

any immediate response necessary. Because, as explained above, we conclude that Mobil did not report the release to the LEPC until five days after it should have had knowledge sufficient to apprise it of a reportable release, the violation is appropriately categorized as “Level I.”

The Presiding Officer classified the “gravity” level of the violation as Level A (also the highest level) because, in accordance with the Penalty Policy, the amount of the release was greater than 10 times the RQ. While this determination was based on a literal reading of the Penalty Policy, we believe some adjustment is appropriate in the unique circumstances of this case.

The Penalty Policy makes clear that the purpose of assigning a gravity level to the violation is to account for the relative hazard to human health and the environment posed by the release. The Penalty Policy attempts to do this by making the gravity level a function of the amount by which the release exceeded the RQ for the released substance:

The RQ scale itself is a relative measure of the hazards

posed by the chemical and therefore the potential threat

to human health and the environment; the lower the

RQ, the greater the potential threat to human health

and the environment. The greater the amount released

over the RQ, the greater the potential for the need for

immediate notification.
Penalty Policy at 16. In general, this is a very sound approach. How- ever, in the case of sulfur dioxide, the one-pound RQ does not truly reflect any technical determination as to the potential threat posed by the chemical; it is merely the statutory default amount applicable to all substances for which no independent RQ has been promulgated. Al- though it is necessary to use the one-pound RQ as the statutory trigger for the reporting obligation, some flexibility can and should be utilized in assessing a civil penalty to more closely approximate the actual threat posed by the violation.

As noted above, the Agency has proposed, but not promulgated, an RQ of 100 pounds for sulfur dioxide based on the Agency’s assess- ment of the risks posed by sulfur dioxide releases, The proposed RQ, in contrast with the statutory default amount, provides a scientific basis for gauging the potential harm caused by Mobil’s release, and represents the Agency’s most recent technical judgment of record con- cerning its assessment of that potential harm. Using the proposed value of 100 pounds as the RQ, the appropriate gravity level for the 473 pound release reported by Mobil is Level C: “The amount released was VOLUME 5

MOBIL OIL CORPORATION 517
greater than 1, but less than or equal to 5 times the RQ.” Applying extent Level I and gravity Level C yields a base penalty range under the matrix of $6,600-$8,250. Penalty Policy at 20.%”

Upon arriving at a range, the “circumstances” of the violation are considered to arrive at the specific penalty within the range. According to the Penalty Policy, “{clircumstances refers to the potential conse- quences of the violation.” Penalty Policy at 18. "Consequences” has a broader meaning than actual harm to human health and the environ- ment, and includes “the adverse effect noncompliance has on the statu- tory or regulatory purposes or procedures for implementing the CERCLA
§ 103/EPCRA program.” Id. In this instance, the emergency notification requirements of EPCRA are fundamental to the integrity of the pro- gram, and we believe that the substantial reporting delay in this in- stance merits a penalty amount at the maximum end of the range, or
$8,250, for at least the first day of violation.

We must next consider whether it is appropriate to assess a per-day penalty for each of the five days reporting was delayed, and if so at what amount. Mobil contends that a per-day penalty is inappropriate because there was no “continuing violation.” In Mobil’s view, a failure to report cannot continue beyond one day, and only a single violation can occur once a facility has knowledge of a release and fails to immediately report it. Mobil’s reasoning is illogical in view of EPCRA’s purpose: what is re- quired under EPCRA is reporting at the earliest possible time, and each day that passes between when a report could have been made and when it is actually made is a continuing violation. See In re All Regions Chemical Labs, Inc., CERCLA 1-88-1089 (ALJ, Dec. 1, 1989) (assessing per-day pen- alty for violations of EPCRA § 304); aff'd, CERCLA/EPCRA Appeal No. 90- 1 (JO, July 2, 1990); aff'd, 932 F.2d 73 (1st Cir. 1991).

In accordance with the Penalty Policy, “one reason to use a per day assessment is to create incentives for violators to return to compliance as expeditiously as possible.” Penalty Policy at 21. The Policy suggests that one way of doing this is to assess the base penalty for a single day, and a smaller per-day penalty from the date of the initial violation until the date of compliance.* See also All Regions Chemical Labs, CERCLA

¥ We agree with the Region that the statutory RQ remains the RQ for sulfur dioxide until a different RQ is promulgated by the Agency. As noted, unless and until changed, the statutory default amount of 1 pound will remain the value that triggers EPCRA reporting requirements for sulfur dioxide. We emphasize that our use of the proposed 100 pound RQ in the penalty calculation simply reflects that the use of the RQ in that context is only as a surrogate measure of environmental harm and the Penalty Policy should be implemented with that in mind.

*Cases involving “continuing harm” to human health or the environment may be appropriate for the assessment of the full base penalty for every day that the violation continues. Id.

VOLUME 5

518 ENVIRONMENTAL ADMINISTRATIVE DECISIONS

1-88-1089 (assessing $20,000 penalty for first day of noncompliance, and $400 per day for each of 144 days of noncompliance thereafter). We believe that this approach is reasonable, and merits application in this case, because of our conclusion that Mobil failed to give EPCRA reporting the priority it deserved in the days that lapsed between the release and the actual report, but instead placed emphasis on completing the regeneration of SRU-3. This failure was especially egregious since the refinery continued to be operational without SRU-3.

Mobil contends that if a per-day penalty is assessed, we should assess a per-day penalty no greater than $400. Mobil cites All Regions and an example noted in the Penalty Policy as support for this conten- tion. The example noted in the Penalty Policy is not set in any particu- lar factual context, and provides little guidance here. See Penalty Policy at 21. With respect to the All Regions case, the presiding officer did not explain how he arrived at the $400 per-day penalty, Setting a relatively low per-day penalty amount may have been deemed appropriate con- sidering the number of days (144) for which the penalty was assessed.” In this case, given the lack of urgency Mobil assigned to its efforts to determine its reporting obligation, we conclude that a per-day penalty of two-thirds of the base penalty, or $5500, is appropriate. Accord- ingly, we assess a total base penalty of $30,250 ($8,250 for the first day and $5500 for each of the four days thereafter) against Mobil for viola- tion of EPCRA § 304 emergency reporting requirements.

We next turn to the consideration of penalty factors relating to the violator to see if any adjustment in the base penalty is appropriate. See Penalty Policy at 22. Mobil contends that several factors support a downward adjustment of the base penalty: (1) Mobil’s argument that the Agency’s interpretation of the federally permitted release exemp- tion is not an authoritative interpretation; (2) the immediate “odor nuisance” report to the State; (3) Mobil’s “initiative” in evaluating whether a reportable release had occurred; (4) the absence of a history of violations regarding release reporting, and (5) Mobil’s voluntary control and monitoring of process parameters.

In our view, none of these arguments provides a justification for a reduction under the facts presented. First, Mobil did not rely on the existence of any release exemption in evaluating the release or delay- ing reporting of the release. As explained above, the report to the SERC on the basis of the odor complaint was compelled by law and does not satisfy EPCRA’s mandate that local authorities receive imme- diate notification of a release. As explained in our consideration of the

»* The base per-day penalty totalled $57,600 for 144 days. Id.

VOLUME 5

MOBIL OIL CORPORATION 519
timeliness of Mobil’s report, we conclude that Mobil demonstrated no special “initiative” in its post-release efforts to evaluate whether a re- port should be made. In accordance with the Penalty Policy, “history of violations” is considered only for upward adjustments of the base penalty. Penalty Policy at 24. Finally, Mobil’s voluntary control and monitoring of process parameters did little to ameliorate the release or expedite reporting in this instance. In sum, we find this case readily distinguishable from All Regions, in which a 10% penalty reduction was allowed due to the respondent's efforts to insure that cleanup of the release was accomplished promptly and properly.”

IV. CONCLUSION

For the reasons explained above, we conclude that the Presiding Officer correctly determined that the release of sulfur dioxide from Mobil’s Paulsboro, NJ, facility on March 12, 1990, was not a “federally permitted release” exempt from EPCRA reporting requirements. We agree that Mobil could have reported the release to the LEPC before March 22, 1990, and we further find that Mobil could have made its report on March 17, 1990. Accordingly, for violation of EPCRA § 304(a)(2), a civil penalty of $30,250 is assessed against Mobil Oil Cor- poration pursuant to § 325(b)(1) of EPCRA. Payment of the entire amount of the civil penalty shall be made within sixty (60) days of service of this final order (unless otherwise agreed to by the parties), by cashier’s check or certified check payable to the Treasurer, United States of America, and forwarded to:

EPA Region II

Regional Hearing Clerk
P.O. Box 360188M
Pittsburgh, PA 15251

So ordered.

Mobil also argues that the substantial penalty proposed by Region I reflected “punitive ani- mus” because an EPA staff person expressed a need to “send a message” to Mobil. Because we have rejected the Region’s proposed penalty, we find it unnecessary to address this allegation.

VOLUME 5