Mision Industrial, Inc. v. Environmental Protection Agency and Russell Train, Administrator , 547 F.2d 123 ( 1976 )

547 F.2d 123

9 ERC 1604, 7 Envtl. L. Rep. 20,096

MISION INDUSTRIAL, INC., et al., Petitioners,
v.
ENVIRONMENTAL PROTECTION AGENCY and Russell Train,
Administrator, Respondents.

No. 75-1377.

United States Court of Appeals,
First Circuit.

Argued Feb. 10, 1976.
Decided Dec. 10, 1976.

David Schoenbrod, New York City, and Richard Ayres, Washington, D. C., with whom Edward Frank, Ross Sandler, New York City, Arthur Gutekunst, Brooklyn, N. Y., and Marcia Tompkins, Washington, D. C., were on brief, for petitioners.

Lee R. Tyner, Atty., Dept. of Justice, and Jeffrey O. Cerar, Atty., E. P. A., with whom Peter R. Taft, Asst. Atty. Gen., Robert V. Zener, Gen. Counsel, E. P. A., and Alfred T. Ghiorzi, Atty., Dept. of Justice, Washington, D. C., were on brief, for respondents.

Before COFFIN, Chief Judge, McENTEE and CAMPBELL, Circuit Judges.

LEVIN H. CAMPBELL, Circuit Judge.

1

Mision Industrial and other environmental groups and individuals seek review1 of a decision made by the Acting Administrator of the Environmental Protection Agency (EPA) approving a revision to the air implementation plan for the Commonwealth of Puerto Rico. The challenged revision substitutes for the approach used in the original plan a new and different strategy for controlling sulfur dioxide (SO 2) emissions.

2

Puerto Rico's original implementation plan was submitted by the Environmental Quality Board of Puerto Rico (EQB), and approved by the EPA Administrator, in 1972. The plan was intended to attain primary and secondary ambient air quality standards for all criteria pollutants, including SO 2, by April 1975. Its principal means for reducing SO 2 emissions was to compel industrial users to burn low sulfur content fuels. Sulfur content by weight in fuel was limited to 1 percent, except in the municipality of San Juan where the limit was 0.5 percent.

3

The present revision was proposed and a public hearing held in Puerto Rico before the EQB in 1974, see 42 U.S.C. § 1857c-5(a)(2). Following the hearing, the revision was formally adopted by the Commonwealth of Puerto Rico, and submitted to the EPA Administrator for his approval. After inviting notice and comment, the Administrator approved most of it. 40 Fed.Reg. 42191-94 (Sept. 11, 1975).

4

Under the revision, limitations continue to be placed upon the amount of sulfur in fuel but the limits are tailored to each industrial source. Instead of a single (except in San Juan) island-wide percentage limitation on sulfur content, the revision allocates different maximum sulfur-in-fuel percentages to different sources, the amounts ranging from .01 percent to 3.1 percent depending on the source's location. To arrive at these percentages, the EQB used a computer model of atmospheric dispersion which predicted the SO 2 concentrations which would result in the ambient air at numerous hypothetical receptor sites from the burning of a given percentage of sulfur in the fuel at each regulated source.

5

* Section 110(a)(3)(A) of the Clean Air Act provides that "(t)he Administrator shall approve any revision of an implementation plan applicable to an air quality control region if he determines that it meets the requirements of paragraph (2) and has been adopted by the State after reasonable notice and public hearings." 42 U.S.C. § 1857c-5(a)(3)(A). The " requirements of paragraph (2)" are the eight general conditions applicable to original implementation plans; thus for a revision to qualify for agency approval, it "is subject only to the condition that (it) satisfy the general requirements applicable to original implementation plans." Train v. National Resources Defense Council, Inc., 421 U.S. 60, 80, 95 S. Ct. 1470, 1482, 43 L. Ed. 2d 731 (1975).

6

Petitioners contend that the present revision fails to meet the above criteria in several ways. They claim that (1) the notice and public hearing afforded by Puerto Rico were deficient in that the EQB did not, before the hearing, make available certain key data necessary to understand the proposed revision; (2) the revision will not achieve national primary and secondary ambient air quality standards within the statutory time, section 110(a)(2)(A), 42 U.S.C. § 1857c-5(a)(2)(A); (3) the revision does not utilize emission limitations as required under section 110(a)(2)(B), 42 U.S.C. § 1857c-5(a)(2)(B); and (4) the revision fails to provide assurances concerning funding and personnel necessary for carrying out the plan, section 110(a)(2)(F)(i), 42 U.S.C. § 1857c-5(a)(2) (F)(i). We consider each of these contentions in turn.

7

a. Reasonable Notice and Hearing.

8

Before approval of the revision, the EPA Administrator had to determine that it was adopted by the "state" (a term which includes, for these purposes, the Commonwealth of Puerto Rico) "after reasonable notice and hearing". 42 U.S.C. § 1857c-5(a)(2); 40 C.F.R. 51.4. When determining if there was compliance with this requirement, the Administrator had before him the same objection Mision Industrial has tendered to us, that the hearing was inadequate because information vital to understanding the revision was withheld prior to and during the hearing. The Administrator resolved this issue against Mision Industrial. He stated in his approval of the revision that "(a)fter reviewing the above material, the Administrator has determined that the information which was made available to the public by EQB prior to the public hearings was adequate to describe in detail the proposed revision to the Puerto Rico implementation plan and the probable effects of that revision. Adequate opportunity for meaningful public participation was assured by EQB prior to the public hearing." 40 Fed.Reg. 42192 (Sept. 11, 1975). For reasons to be stated we accept this ultimate finding, although we do not accept the adequacy of the EQB's showing in regard to the availability of the computer print-out discussed below.

9

It is petitioners' principal complaint that they were given a run-around when they tried to see the computer print-out showing the basis for the sulfur-in-fuel limitations assigned to each source under the plan.2 At the hearing, two witnesses for Mision Industrial testified under oath that the EPA responded to a prehearing request for the print-out by referring them to the EQB, which, when asked for the final results of the computer run, responded that it did not have the data but that petitioners should ask the Fuel Office. The Fuel Office, in response to two separate queries, stated that it did not have the computer results but that either the EPA or the EQB had them. Petitioners contend that they did not see the material until after the Commonwealth's hearing had been held.

10

We find it disturbing that these specific allegations about requests and responses were never expressly faced or explained by the EQB or other Commonwealth officials. The lack of specifics strongly suggests that the EQB has no real answer. The Executive Director of the EQB merely testified at the hearing that the computer print-out was available and had always been available at the Board, and that any interested person had access to it "with due protection to prevent the disappearance of any of this material, since it is unique". The Executive Director subsequently certified as part of a more general certification required by EPA regulations, that the print-out had been available, and the hearing examiner made a catch-all finding that material "pertinent to the amendment" had been available. When the EPA later made inquiry by letter, the General Counsel of the EQB repeated the general assurance that at the time of the hearings, EQB had made available for public inspection a copy of the computer print-out.

11

EPA now argues that the Administrator could accept these official statements, and denies he had any duty to go behind them, principally, so EPA argues, because state officials are entitled to a presumption of regularity in their actions. However, the presumption of regularity is rebuttable; and even according it maximum weight, we think it would be irrational in these circumstances to accord the presumption greater credit than the testimony of Mision's witnesses given the EQB's total failure to confront or rebut the particulars of their stories. EQB never sought to explain why the witnesses were not informed of the purported availability of the print-out, and why even after their testimony at the hearing the witnesses were not invited to inspect the print-out. On the state of this record, we think it cannot sensibly be assumed that the computer print-out was publicly available prior to the hearing. Indeed, the Administrator avoided making an express finding that the print-out was available. Accordingly, we must decide whether the pre-hearing unavailability of the computer print-out deprived the petitioners of so vital a component of the proposed revision as to materially impair their ability to comment at the public hearing, and the ability of the hearing itself to serve the purpose Congress intended. We conclude that it did not.

12

Petitioners appear to have received sufficient information to learn the proposed control technique (variation in the permitted sulfur content in fuel depending on the location of the source), the technology used to arrive at each assigned limitation (the diffusion model formulas), and the conclusions reached by the application of the computer model (the actual assigned limitations).3 The missing computer print-out was from the program used to arrive at each assigned sulfur limitation. Its principal use would be to verify that the limitations chosen would result, under the formulas, in SO 2 concentrations of acceptable levels to meet ambient air standards. Petitioners do not now contend, in light of their present familiarity with the data contained in the print-out, that the results were inconsistent with the method used. The print-out was generally available during the comment period while the revision was pending before the Administrator, and we do not see that petitioner's perceptions and arguments against the revision were fortified or altered in any way as a result of the availability of the computer print-out. To the contrary, petitioners assert that the print-out proved useless to them, turning out merely to be a miscollated massive stack of paper. We conclude, therefore, that had the data been available, it would not have affected the arguments at the public hearing, which centered on the alleged unsuitability of the diffusion model and the source-by-source approach.

13

This is not to condone the EQB's failure to allow responsible members of the public to inspect the print-out. Whether defined as part of the actual revision itself or merely a work paper, the print-out was the type of relevant documentation which should have been made available before the public hearing so as to maximize the opportunity for intelligent comment and debate. The EPA so recognized and indeed, at least in theory, so did the EQB. Where documents of this sort are withheld, especially in circumstances suggesting something less than good faith, it is not members of the public, such as petitioners, who must bear the laboring oar in proving prejudice; rather prejudice should be assumed unless its absence can be fairly inferred from the record. Cf. United States v. Honneus, 508 F.2d 566, 572 (1st Cir. 1974), cert. denied, 421 U.S. 948, 95 S. Ct. 1677, 44 L. Ed. 2d 101 (1975). Here we think the latter is the case. We can see no basis for a finding that the presence of the print-out would have materially affected petitioners' presentation of views at the hearing. We accordingly uphold the Administrator's finding of substantial compliance with the Act's notice and public hearing requirement.

14

b. Meeting the Ambient Air Standards.

15

Under section 110(a)(2)(A), any approved plan must be determined by the Administrator to provide for attainment of the national primary air quality standard as expeditiously as practicable, but within no more than three years, and attainment of the secondary standards within a reasonable time. The Administrator determined that under the Puerto Rico revised plan, "(p)rotection of the annual national ambient air quality standard is assured through the conservative estimate of an 80 percent load factor on an annual basis when estimating source emissions." 40 Fed.Reg. 42191 (Sept. 11, 1975). Petitioners challenge the Administrator's approval, arguing on at least four grounds that the predicting methodology used in drafting the plan permitted too great a likelihood for error and that as a result the plan fails to provide for attainment of national air quality standards.

16

First, petitioners object generally to the methodology used by the EQB, arguing that the 20 percent margin for error built into the diffusion model is smaller than the model's admitted potential for error. The EPA does in fact concede a possible random error as high as 150 percent for the annual average of pollutant emissions and 200 percent for short-term concentrations. 40 Fed.Reg. 42193 (Sept. 11, 1975). The Administrator explains, however, that these errors are the extreme, occur only as isolated incidents, and are just as likely to result in overprediction as underprediction. He concluded that the conservative measures taken by EQB were sufficient to decrease the possibilities of pollution underprediction.4

17

Petitioners next contend that Puerto Rico's rough terrain and the resulting terrain turbulence have an effect in actual conditions which is not properly accounted for by the Puerto Rico diffusion model. For this reason also, a higher margin for error is said to be a necessary part of the model. The EPA, on the other hand, considers the model to have been applied more conservatively than if terrain turbulence were somehow compensated for.

18

Petitioners object that the model did not use on-site weather conditions as a basis for its projections but used instead weather data gathered at only three locations on the island. EPA responds that it would be impossible to set up a site at every hypothetical monitoring point and that the EQB, where appropriate, assumed the worst weather conditions as a basis for its predictions. In addition, the record indicates that the EPA took an active role in assuring that weather data was properly applied by asking the EQB specifically for an explanation of how the worst case condition was derived. Only after the Administrator was satisfied with the EQB methodology did he determine that the weather assumptions used in the Puerto Rico model were appropriate.

19

Finally, Mision Industrial claims that it was error to approve a revised plan based on an uncalibrated model, that no real life data has been used in arriving at or calibrating the accuracy of the predicted relationship between sulfur-in-fuel and air quality. The EPA counters that calibration on incomplete data is not good practice and that the agency supports Puerto Rico's decision to rely on theoretical data in the light of conservative assumptions and calculations applied to compensate for the lack of precision. 40 Fed.Reg. 42193 (Sept. 11, 1975).

20

Petitioners' criticisms go to the heart of the methodology used in the revised air implementation plan. Their concerns were communicated in detail to the EPA Administrator during the federal comment period. He in turn appears to have considered their views and objections and he determined, without at least obvious unreason, that the computer model was a satisfactory predictive tool on which to base Puerto Rico's revision. See id. at 42191-94. This is an area where EPA's "expertise is heavily implicated", Sierra Club v. EPA, 176 U.S.App.D.C. 335, 540 F.2d 1114, 1131 (1976), cert. filed, 45 U.S.L.W. 3346-47 (Nov. 9, 1976), and we may not substitute our judgment for that of the Administrator. His determinations in the foregoing must therefore stand.

21

c. Emission Limitations.

22

Before approving an air quality implementation plan or revision, the Administrator must determine that it "includes emission limitations . . . and such other measures as may be necessary to insure attainment and maintenance of (the) primary or secondary standard . . . ." 42 U.S.C. § 1857c-5(a)(2)(B). Petitioners contend that because the Puerto Rico revised plan permits varying sulfur content in fuel depending on a source's geographical location, the effect is to disperse the pollutants rather than to control them by "emission limitations". The Administrator did not address this argument in his approval of the revision, but he argues to this court that control of the sulfur content in fuel is an "emission limitation" under the statute and the applicable case law.

23

Section 110(a)(2)(B) requires only there be provision for "emission limitation" in a clear air implementation plan. The exact type of limitation is left to the discretion of the state which develops the plan. Train v. National Resources Defense Council, Inc., supra, 421 U.S. at 79, 95 S. Ct. at 1481. So long, therefore, as the technique employed by the Puerto Rico revision is based on an "emission limitation" that is adequate to achieve compliance with national air quality standards, the Administrator is not authorized to reject it in favor of another technique. See id. at 80, 95 S.Ct. at 1482.

24

Emission limitations have been defined by the Supreme Court as " regulations of the composition of substances emitted into the ambient air from such sources as power plants, service stations, and the like." Id. at 78, 95 S.Ct. at 1481. Thus, a plan must regulate the amount of a given pollutant (here SO 2) included in the emission from a source. Big Rivers Electric Corp. v. EPA, 523 F.2d 16, 21-22 (6th Cir. 1975), cert. denied, 425 U.S. 934, 96 S. Ct. 1663, 48 L. Ed. 2d 175 (1976).

25

The EPA contends that the sulfur-in-fuel limitations satisfy the requirements under this definition and we do not disagree. Setting a limit on permissible sulfur content restricts the amount of that pollutant in the ongoing emissions of each industrial source. The method continuously regulates the amount of sulfur dioxide that will be contained in the smoke emitted from each plant. The fact that the specified limits vary between plants does not make the technique any the less dependent on the deployment of emissions limitations. See generally Train v. National Resources Defense Council, Inc.,supra, 421 U.S. at 78, 95 S. Ct. at 1481; Kennecott Copper Corp. v. Train,526 F.2d 1149, 1155 (9th Cir. 1975), cert. denied, 425 U.S. 935, 96 S. Ct. 1665, 48 L. Ed. 2d 176 (1976); Big River Electric Corp. v. EPA, supra, 523 F.2d at 21; National Resources Defense Council, Inc. v. EPA, 489 F.2d 390 (5th Cir. 1974), rev'd in part on other grounds sub nom. Train v. National Resources Defense Council, Inc., 421 U.S. 60, 95 S. Ct. 1470, 43 L. Ed. 2d 73 (1975). The Administrator was not in error in determining that the plan complied with section 110(a)(2)(B).d. Necessary Assurances.

26

Petitioners contend that the Administrator was mistaken in his determination that the revision provided suitable assurances of sufficient personnel and resources to carry out the plan. See 42 U.S.C. § 1857c-5(a)(2)(F)(i); 40 C.F.R. 51 & App.K. In the revision as originally presented at the public hearing, mention was not made of personnel and resource needs nor of how these would be met. Assurances along these lines came later, under prodding of the Administrator. In petitioners' view, these belated express assurances are inadequate. They are seen as merely rescrambling the same personal and resources provided for the entire plan originally an unrealistic response, say petitioners, given the revision's greater complexity to administer. But we are unable to say that the Administrator abused his discretion in deeming the assurances finally received as adequate. See National Resources Defense Council, Inc. v. EPA, 478 F.2d 875, 884 (1st Cir. 1973). The adequacy of "necessary assurances" is largely an administrative question within the agency's expertise. The Administrator treated the question with seriousness; he rejected an initial submission from the EQB; and while in describing his acceptance of the later submission he unfortunately misstated the number of man hours per year that the EQB had represented would be necessary to insure implementation of the revision, we cannot say on the basis of this isolated error that he did not exercise an informed judgment.5

II

27

Beyond claiming, as discussed above, that the revision did not meet certain of the eight conditions set forth in paragraph (2) of section 110(a), petitioners argue that by approving this revision the Administrator abdicated his duty to promote the general purpose of the Clean Air Act, which is to "protect and enhance the quality of the Nation's air resources so as to promote the public health and welfare and the productive capacity of its population", 42 U.S.C. § 1857(b)(1). Petitioners contend that, in approving a plan which permits an increase in SO 2 emissions in an area with a sulfate problem the Administrator has failed to comply with his own expressed policy of limiting such emissions. Further, petitioners decry the underlying theory of the revision, saying that the variable sulfur-in-fuel limitations will lead to increased pollution where the air is presently cleaner than national ambient standards. This is said to violate the nondegradation policy of the Clean Air Act. See generally Sierra Club v. EPA, supra, 540 F.2d at 1124-31. We address each argument in turn.

28

a. Sulfur Dioxide Emissions and the Sulfate Problem.

29

Petitioners contend that the Administrator abused his discretion because approval of the Puerto Rico plan failed to conform to his announced policy of "limiting" sulfur emissions in areas of high sulfate concentrations.6 The EPA, in its approval of the plan, did not address this issue, but did note, in response to the more general assertion which we discuss last, that any increase in sulfur dioxide emissions was permissible under the recent EPA regulation pertaining to the prevention of significant deterioration of air quality. 40 Fed.Reg. 42193 (Sept. 11, 1975). More to the point, he now explains that while the EPA is much concerned about the health hazards associated with sulfates and is continuing to study and evaluate the problem, the agency has done as much as can be done until enough is known by scientists about the formation of sulfates to establish appropriate national ambient air limitations. Meanwhile, the most that can be accomplished is to minimize emissions of SO 2, one known precursor of sulfates. For the present, having determined that the Puerto Rico revision proposal met the requirements of section 110(a)(2) of the Clean Air Act, the Administrator could not disapprove the revision on the basis of his concern for sulfates, no standard for sulfates having as yet been promulgated. See Train v. National Resources Defense Council, Inc., supra, 421 U.S. at 80, 95 S. Ct. at 1482. See also Plan for Arcadia, Inc. v. Anita Assoc., 501 F.2d 390, 392 (9th Cir.), cert. denied, 419 U.S. 1034, 95 S. Ct. 517, 42 L. Ed. 2d 309 (1974) (court cannot order the promulgation of particular Clean Air Act regulations). We agree with the Administrator, who has both expertise and considerable discretion in the matter. While we sympathize with petitioners' frustration at the Administrator's failure in his plan approval to discuss the Puerto Rico sulfate problem, he determined that the plan complied with both the Act and with his own regulations as currently in effect. This judgment is not clearly wrong, and we cannot overturn his approval of the revision on the basis of our own attempted lay judgments concerning how best to deal with the sulfate problem.

30

b. Compliance with the Nondegradation Policy.

31

Petitioners claim, finally, that the very nature of the Commonwealth's plan is based on dispersion of SO 2 pollutants rather than their reduction with the inevitable result that some air will become more polluted than it is already thereby contravening the policy of the Clean Air Act to prevent significant deterioration in the quality of existing clean air. See Sierra Club v. EPA, supra, 540 F.2d at 1124-31; Sierra Club v. Ruckleshaus, 344 F. Supp. 253, 256 (D.D.C.), aff'd by an equally divided Court sub nom. Fri v. Sierra Club, 412 U.S. 541, 93 S. Ct. 2770, 37 L. Ed. 2d 140 (1973). The premise of their argument would seem to be that the source-by-source approach provides an economic incentive for building new plants in areas of presently clean air where higher sulfur content fuel may be burned. Even assuming that the revision would have such a tendency, we hold that the Puerto Rico revised plan complies with EPA nondegradation regulations and that the regulations as here applied satisfy the Clean Air Act policy of preventing the significant deterioration of clean air. See Sierra Club v. EPA, supra, 540 F.2d at 1131-32.

32

We begin by stating that we are persuaded by Judge Wright's able analysis in Sierra Club v. EPA, supra, 540 F.2d at 1124-30, that implicit in the "protect and enhance" language of section 101(b)(1) of the Clean Air Act, 42 U.S.C. § 1857(b)(1), is a "clear understanding (by Congress) that the Act embodie(s) a pre-existing policy of nondeterioration of air cleaner than the national standards." Sierra Club v. EPA, supra, 540 F.2d at 1124. It was in response to this mandate as interpreted by the courts, see Sierra Club v. Ruckleshaus, supra, that the EPA promulgated nondegradation regulations setting standards for increases in pollution considered not to affect significantly existing air quality. In Class II areas (including all of Puerto Rico), a limited increase in sulfur dioxide and particulate matter occurring after January 1, 1975, is permissible as "insignificant". 40 C.F.R. 52.21(c)(2)(i). See Sierra Club v. EPA, supra, 540 F.2d at 1119-20. As the EPA explained when it promulgated the regulations, "Class II applie(s) to areas in which deterioration normally accompanying moderate well-controlled growth would be considered insignificant", 39 Fed.Reg. 42510 (Dec. 5, 1974). This approach by the EPA has recently withstood multiple challenges in Sierra Club v. EPA, supra, 540 F.2d at 1131-32, where in response to the argument that allowing pollution increases in Class II, as well as Class III, areas was permitting significant deterioration of air quality, the court stated that "the significance of deterioration of air quality should be determined by a qualitative balancing of clean air considerations against the competing demands of economic growth, population expansion, and development of alternative sources of energy." Id. at 1132. We see no reason to disagree with the D. C. Circuit's thoughtful analysis. Thus, the only inquiry remaining is whether the Administrator made a reasoned determination that the Puerto Rico revised plan conformed to EPA nondegradation regulations.

33

In approving the Puerto Rico revised implementation plan, the Administrator directly addressed the issue. He conceded that at some EQB receptor sites the Class II increment for SO 2 might be exceeded, but pointed out that the nondegradation regulations themselves, 40 C.F.R. 52.21(d), exempt from EPA review modifications which result from switching to a higher sulfur content fuel. 40 Fed.Reg. 42193 (Sept. 11, 1975).7 In any event, the increased emissions would be counted against that allowable for the area. See Sierra Club v. EPA, supra, 540 F.2d at 1129-30. As the Administrator explained, in those areas in which the Class II increment is partially used up, "review of new sources and modifications under 40 CFR 52.21 will be conducted so as to assure that the Class II increment is not violated", and that in those areas where the increment has already been absorbed the review of modifications and new sources will "assure that there will be no increase in the ambient air quality values of sulfur oxides due to the source identified in 40 CFR 52.21." 40 Fed.Reg. 42193 (Sept. 11, 1975).

34

The Administrator has determined that the revised plan conforms to the nondeterioration regulations and has given assurances that through review of new sources and modifications the incremental increase in SO 2 pollution throughout the island will not exceed that permitted for a Class II area. We find no basis for overturning that determination and, as we find that the EPA classification scheme gives effect to the Clean Air Act's purpose of preserving and enhancing air quality, we hold that the Puerto Rico revision satisfies the mandate of the Clean Air Act.

35

Petition for review denied, and the Approval of the Revision to the Puerto Rico Implementation Plan is affirmed.

1

Our jurisdiction exists under section 307(b)(1) of the Clean Air Act Amendments of 1970 (the Act), 42 U.S.C. § 1857h-5(b)(1)

2

Petitioners also complain that the materials made available prior to the hearing did not add up to the proposed revision, being self-serving rhetoric and a confused compilation of additions and deletions to the original plan. There is some force to this, the materials not being as well organized nor as complete as would have been desirable. But we think the available materials were minimally adequate to support the Administrator's conclusion that they described "in detail the proposed revision . . . and (its) probable effect. . . ." The most serious omission was the print-out, discussed in the text. Petitioners argue that since some significant details were added only after the public hearing, at the EPA's request, the materials made available beforehand did not amount to the actual, proposed revision as required by EPA regulations but merely a fragment thereof. 40 C.F.R. 51.4(a)(2). The question is one of degree. A plan or revision cannot have undergone a public hearing if it was so incomplete when the hearing was held as to lack the substance of the final approved plan. On the other hand, "(a) hearing is intended to educate an agency to approaches different from its own; in shaping the final rule it may and should draw on the comments tendered." South Terminal Corp. v. EPA, 504 F.2d 646, 659 (1st Cir. 1974). We think the essentials of the revision as ultimately accepted by the Administrator were sufficiently disclosed in the materials before the public at the time of the public hearing

3

Petitioners received the following materials in advance of the EQB hearing: (1) proposed amendments to the Regulation for Control of Atmospheric Pollution, which included the formulas used in the computer diffusion model and an appendix listing the maximum allowed percentage of sulfur-in-fuel for each regulated source; and (2) amendments to the State Implementation Plan of the Commonwealth of Puerto Rico listing additions to and deletions from the original plan and presenting the revision's source-by-source approach including, in general terms, the method to be used to effectuate it

4

Petitioners also argue that the revision, unlike the original plan, fails to take into account sulfur pollution from sources other than industrial plants. The Administrator considered these additional sources to be adequately compensated for by the 20 percent margin for error built into the model. 40 Fed.Reg. 42193 (Sept. 11, 1975)

5

Petitioners point out that the Commonwealth's proposed allocation of resources was not stated as part of the proposed revision itself at the time of the public hearing. Without endorsing the omission, we do not see it as fatal to the hearing or to the Administrator's later acceptance of the revision. Nothing in the record indicates that revision proponents used the omission to enhance their position. To the contrary, the absence of positive assurances at that time would have reinforced petitioners' position that enforcement personnel and resources beyond those already committed were not contemplated

6

According to an EPA position paper, sulfates (including sulfuric acid, ammonium bisulfate, neutral metallic sulfates, absorbed SO 2 and sulfites) are thought to contribute to increased respiratory disease. Although it is known that SO 2 emissions can be oxidized into sulfates through the catalytic effects of particulate matter, photochemical smog, ammonia and other agents, neither the relative importance nor the role of these agents in the conversion process is fully understood. It has apparently been shown, however, that in a 24-state region of the Eastern United States, there is a correlation between high sulfur dioxide emissions and high atmospheric sulfate concentration

7

This exemption is an accommodation of the Energy Supply and Environmental Coordination Act of 1974, 15 U.S.C. § 791 et seq. enacted to minimize dependence on imported oil. See Sierra Club v. EPA, supra, 540 F.2d at 1129