Cleveland Electric Illuminating Co. v. Environmental Protection Agency

EDWARDS, Circuit Judge. This court now has before it 23 petitions involving 32 companies filed against the United States Environmental Protection Agency which levy a variety of complaints against the federal agency's imposition of a sulfur dioxide (SO2) pollution control plan for industrial discharges into Ohio’s ambient air. The issues, which have been extensively briefed and argued, divide into general legal and procedural complaints which might be applicable to any one of the petitioners and a wider variety of specific complaints about the application of the EPA controls to particular power-generating or industrial plants. The cases dealt with in this opinion1 present the major general issues. Other individual cases, in addition to presenting one or more of the general issues, also present specific issues of fact. These are reserved pending a review of and reports on the factual disputes between the United States EPA and the individual petitioners. The major issues dealt with in this opinion are: 1) intervenor, the State of Ohio, claims that this court should disapprove the federal plan as irrational and arbitrary and rely upon Ohio to come forward with a more rational plan sometime in the future; 2) petitioners claim that the EPA SO2 plan should be remanded for hearings because the informal rulemaking hearings employed by EPA under 5 U.S.C. § 553 (1970 & Supp. V 1975) were inadequate; and 3) petitioners claim that the major model employed by the United States Environmental Protection Agency in establishing specific emission limitations for particular plans is invalid both intrinsically and as applied. This model is termed the “Real-Time Air-Quality-Simulator Model” (hereinafter RAM). THE HISTORY OF THIS LITIGATION The United States Congress has been wrestling with the problem of pollution of the ambient air since 1955. See Act of July 14, 1955, Pub.L. No. 84-159, 69 Stat. 622. The original act has now been amended many times. It now is cited as the Clean Air Act and has been codified in 42 U.S.C. §§ 1857-18571 (1970 & Supp. V 1975).2 The prior history of litigation concerning sulfur dioxide emission controls in this — 1153 — court is set forth in Buckeye Power, Inc. v. EPA, 481 F.2d 162 (6th Cir. 1973) (Buckeye Power # 1) and Buckeye Power, Inc. v. EPA, 525 F.2d 80 (6th Cir. 1975) (Buckeye Power #2). National air quality standards for sulfur dioxide, one of the most important pollutants of the ambient air, were set by EPA in 1973 as follows: § 50.4 National primary ambient air-quality standards for sulfur oxides (sulfur dioxide). The national primary ambient air quality standards for sulfur oxides measured as sulfur dioxide by the reference method described in Appendix A to this part, or by an equivalent method, are: (a) 80 micrograms per cubic meter (0.03 p.p.m.) — annual arithmetic mean. (b) 365 micrograms per cubic meter (0.14 p.p.m.) — Maximum 24-hour concentration not to be exceeded more than once per year. § 50.5 National secondary ambient air quality standards for sulfur oxides (sulfur dioxide). The national secondary ambient air quality standard for sulfur oxide measured as sulfur dioxide by the reference method described in Appendix A to this part, or by any equivalent method is 1,300 micrograms per cubic meter (0.5 p.p.m.) maximum 3-hour concentration not to be exceeded more than once per year.[3] Ambient Air Standards (Primary & Secondary), 40 C.F.R. §§ 50.4, 50.5 (1976). The federal Clean Air Act program which produced these standards is based primarily upon the adverse effect which air pollution has upon human life and health. Acute episodes of high pollution have clearly resulted in mortality and morbidity. Often the effects of high pollutant concentrations in these episodes have been combined with other environmental features such as low temperatures or epidemic diseases (influenza) which may in themselves have serious or fatal consequences. This has sometimes made it difficult to determine to what extent pollution and temperature extremes are responsible for the effects. Nevertheless, there is now no longer any doubt that high levels of pollution sustained for periods of days can kill. Those aged 45 and over with chronic diseases, particularly of the lungs or heart, seem to be predominantly affected. In addition to these acute episodes, pollutants can attain daily levels which have been shown to have serious consequences to city dwellers. * * * * * * There is a large and increasing body of evidence that significant health effects are produced by long-term exposures to air pollutants. Acute respiratory infections in children, chronic respiratory diseases in adults, and decreased levels of ventilatory lung function in both children and adults have been found to be related to concentrations of S02 and particulates, after apparently sufficient allowance has been made for such confounding variable as smoking and socioeconomic circumstances. Rail, Review of the Health Effects of Sulfur Oxides, 8 Env’tal Health Perspectives 97, 99 (1974). It appears that present national air quality standards have been set with little or no margin of safety. Adverse health effects are set forth in the two following charts; and the minimal or nonexistent margins of safety are vividly portrayed below: — 1154 — [[Image here]] The major sources of sulfur dioxide pollution of the ambient air are coal-fired plants — exemplified by power plants operated by some of the petitioners in this case. — 1155 — Two other facts should be added from the extensive technical record in this case before we turn to the specific legal issues. The first is that sulfur dioxide emitted from plant stacks reacts with other elements in the atmosphere to form sulfuric acid mist and various suspended sulfates which are in fact the irritants which adversely affect human health. T. Lewis, M. Amdur, M. Fritzhand & K. Campbell, Toxicology of Atmospheric Sulfur Dioxide Decay Products 17 (1972). The second important fact is that these derivatives from sulfur dioxide tend to be airborne for days. They affect areas at great distances downwind, even when in the original sulfur dioxide form they were emitted from a high power plant stack. Rail, Review of the Health Effects of Sulfur Oxides, 8 Env’tal Health Perspectives 97, 106 (1974). The cases considered in this opinion represent the fourth time this sulfur dioxide control problem has been before this court. In the instant cases a hearing was held November 14, 1976, at which numerous arguments were advanced concerning petitioners’ claims that they had been deprived of an adequate opportunity to comment upon the EPA sulfur dioxide standards. In particular they complained about not having an opportunity in advance to comment upon the EPA’s use of the RAM model. The hearing resulted in the entry of an order by this court remanding all of these petitions to the EPA for reopening of the record to allow presentation of additional objections, corrections, and comment. The order provided in part as follows: On receipt and consideration of the thirty-six (36) above-styled petitions for review attacking the emission standards for the State of Ohio applicable to sul-phur dioxide (SO2) promulgated by respondents Environmental Protection Agency (EPA) and Train on August 27, 1976; and On receipt and consideration of respondent’s motion to consolidate such petitions for hearing — and certain petitioners’ objections thereto — and said motion to consolidate having been granted by this Court; and On receipt and consideration of certain petitioners’ motions for stay of the respondent’s orders pending this Court’s review; and On receipt and consideration of respondent EPA’s motion to hold a prehearing conference and certain concurrences therein, and such prehearing conference having been held after due notice to all parties; and On inspection and consideration of petitioners’ motions, briefs, and oral arguments and noting that many of them deal with claims of due process violations in respondent EPA’s closing of the administrative record without further opportunity on the part of petitioners to present comment or evidence deemed by them to be essential to a just result, Now therefore this Court, in the interest of as expeditious judicial disposition of this complex litigation as possible, hereby, sua sponte, extends to all petitioners in this consolidated proceeding a stay of enforcement of said orders of respondents EPA and Train, subject to the following conditions: No petitioners shall be permitted to submit any new emission, process or air quality data. Comments relating to clerical or computational errors shall be permitted. Whenever possible, petitioners shall make consolidated submissions to the Agency. All submissions shall be made by petitioners no later than January 14, 1977, and the Agency shall prepare an appropriate response and shall amend the subject regulations if and as necessary within an additional 60 days therefrom. The stay granted herein will terminate twenty-one (21) days after respondent EPA files with this Court the response called for above, unless otherwise directed by this Court. Pursuant to the order of this court, the effective date of the SO2 regulations was June 17, 1977. EPA has advised this court, however, that it has not begun enforcement proceedings in relation to any sources involved directly in this litigation. — 1156 — Although this court’s order allowed the petitioners 60 days for presentation of additional evidence and comment, EPA sua sponte extended the remand period briefly. Promptly upon notice that EPA had filed its response to petitioners’ objections and comments resulting from the remand, this court scheduled two full days of hearings on these cases for purposes of as quick an adjudication at the appellate level as might be achieved. DISPOSITION OF THE GENERAL ISSUES 1. The State of Ohio’s Petition On July 13, 1977, the State of Ohio belatedly moved for leave to intervene in this proceeding. Its motion attacked the EPA sulfur dioxide emission control plan as having an adverse impact on the Ohio coal industry, and the Ohio economy as a whole. The motion also asserted that the State was developing a sulfur dioxide plan which would eliminate excessive abatement requirements which Ohio perceived to exist in the federal regulations. This court granted the motion for leave to intervene and has considered the brief and the reply brief { filed by Ohio. Under this first disposition heading we consider only Ohio’s suggestion I that this court reject the United States ! Environmental Protection Agency’s sulfur ! dioxide control plan and rely upon Ohio’s ¡implied promise to promulgate a state sulfur dioxide plan sometime in the future, j We reject this suggestion on the basis of a record of delay and default which has left Ohio in the position of being the only major industrialized state lacking an enforceable plan for control of sulfur dioxide. It was clearly the intention of Congress to have a plan for control of sulfur dioxide emissions in place in all states in need of such control by the year 1972. Clean Air Act §§ 109(a), 110(a), 42 U.S.C. §§ 1857c-4(a), 5(a) (1970 & Supp. V 1975). It was equally clearly the intention of Congress that the preferred mechanism for establishment of such a plan was through the establishment and operation of a state environmental protection agency. Section 107(a), 42 U.S.C. § 1857c-2(a) (1970). On January 30,1972, Ohio did submit a plan for approval by the Administrator of the United States Environmental Protection Agency under Section 110 of the Act and the Administrator approved that plan. That approval, however, was challenged in this court on the ground that such approval required a federal rulemaking hearing prior to the required approval by the federal Administrator. Among other claims laid before this court in that petition was an attack on the sulfur dioxide control scheme contained in the Ohio plan, claiming “there is presently no technologically feasible method of removing from their coal burning emissions an amount of sulfur sufficient to meet the standards.” See Buckeye Power, Inc. v. EPA, 481 F.2d 162, 167 (6th Cir. 1973). It was also petitioners’ contention in that same litigation that they had not been allowed to document these claims of impossibility before the federal Administrator prior to his approval of the state plans. On analysis of these arguments, this court vacated the approval of the Ohio state plan and remanded the case to the Agency for compliance with Section 4 of the Administrative Procedure Act, 5 U.S.C. § 553 (1970 & Supp. V 1975), which requires adherence to informal rulemaking procedures. Before the hearing could be held which was called for in Buckeye Power # 1, supra, the governor of Ohio, on August 27, 1972, “withdrew” the sulfur dioxide portion of the state EPA plan. At that point Ohio began work on a new plan for sulfur dioxide control. On May 30, 1974, the second plan was submitted to the United States EPA for approval. It had, however, been challenged at the state level and was partially vacated on procedural grounds by the Ohio Environmental Board of Review on September 12, 1974, and for a second time on July 16,1975, the governor of Ohio withdrew the Ohio plan to control sulfur dioxide. The Clean Air Act, as amended, provides in part: (c)(1) The Administrator shall, after consideration of any State hearing record, promptly prepare and publish proposed regulations setting forth an implementation plan, or portion thereof, for a State if— (A) the State fails to submit an implementation plan for any national am — 1157 — bient air quality primary or secondary standard within the time prescribed. Clean Air Act § 110(c)(1), 42 U.S.C. § 1857c-5(c)(1) (Supp. V 1975). Clearly, the State of Ohio has failed to submit an implementation plan for sulfur dioxide for which a national ambient air quality primary standard has been prescribed. Equally clearly, five years have now elapsed beyond the date when such an implementation plan was called for under the Clean Air Act. Under these circumstances, we find no warrant, consistent with the purposes of the federal legislation, for giving heed to Ohio’s petition for further delay. Intervenor Ohio’s other objections to the United States EPA’s sulfur dioxide control plan will be considered under Part 3 of the disposition section of this opinion. 2. The Additional Remand and Cross-Examination Issue The leading brief in this series of cases filed on behalf of the utilities opens its argument for remand as follows: The Most Appropriate Manner To Resolve The Multitude Of Issues Raised Is A Remand To The EPA With Directions To Hold Further Hearings To Reconsider The Significant Issues; Given The Nature Of This Rulemaking, Any Remand Should Incorporate Procedural Safeguards Such As Right To Cross-Examine Or Question EPA. Admittedly, there is no statutory requirement that EPA afford the regulated the opportunity to confront its decision makers through adjudicative-type hearings. See Buckeye Power, Inc., supra. [Buckeye Power, Inc. v. EPA, 481 F.2d 162 (1973)]. However, this EPA promulgation contains so many specific findings and actions that the normal comment period has not been sufficient to expose and evaluate all of the important facts before this Court. EPA has promulgated emission limits specific to a plant, has applied specific diffusion models specific to a plant, and has reached specific conclusions regarding economics specific to a plant. Each decision is based on fact upon fact and conclusion upon conclusion. In essence and in operation, this plan and its formulation smack of the issuance of an order as defined by EPA. In Buckeye Power # 1, a panel of this court, prior to the first remand of the sulfur dioxide problem for federal hearings, considered the question as to whether or not adjudicatory hearings (including cross-examination) were required. The opinion rejected this suggestion with the following reasoning: However, as heretofore noted, the petitioners herein do not simply request a remand with instructions to adhere to the informal rulemaking dictates of Section 553 of the APA; they also request a full-scale evidentiary hearing before the Administrator to adjudicate their complex and intricate claims of high cost-benefit, technological infeasibility and resource unavailability. We cannot accept this position. Administrative rulemaking which is to be preceded by extensive hearings where “[a] party is entitled to present his case or defense by oral or documentary evidence, to submit rebuttal evidence, and to conduct such cross-examination as may be required for a full and true disclosure of the facts . . . ” (5 U.S.C. § 556(d) (1967)) is required only when the last sentence of Section 553(c) of the APA applies. This section provides: “When rules are required by statute to be made on the record after opportunity for an agency hearing, sections 556 and 557 of this title apply instead of this subsection.” (Emphasis added). (5 U.S.C. § 553(c) (1967)). (Sections 556 and 557 of the APA outline the requirements for extensive, adjudicatory-type hearings.) Thus, when a statutory provision directing certain agency action states that such action shall be “made on the record after opportunity for an agency hearing,” then, and only then, is the agency required to have full-scale adjudicatory hearings prior to rulemaking. * * * — 1158 — There is no provision that, in approving or disapproving proposed state pollution-abatement plans pursuant to 42 U.S.C. § 1857c-5(a)(2) (1973 Supp.), the Administrator shall make a determination “on the record after an opportunity for an agency hearing.” Furthermore, when Congress intended that actions of the Administrator of the EPA be preceded by adjudicatory-type hearings it either specifically outlined the type of hearings, (42 U.S.C. § 1857f-5a(c)(1) (1973 Supp.)), or invoked the determination “on the record” provision of 5 U.S.C. § 553(c) (1967). See 42 U.S.C. § 1857c-5(f)(2) (1973 Supp.); 42 U.S.C. § 1857f-5(b)(2)(B) (1973 Supp.). On these grounds alone we would reject the petitioners’ argument that the Administrator is required to have full-scale adjudicatory-type hearings prior to acceptance of the state plans. Buckeye Power, Inc. v. EPA, 481 F.2d 162, 172-73 (6th Cir. 1973). (Footnote omitted.) More importantly, we note the following discussion of the issue now before us in the unanimous opinion in United States v. Allegheny-Ludlum Steel Corp., 406 U.S. 742, 92 S.Ct. 1941, 32 L.Ed.2d 453 (1972), wherein the Supreme Court of the United States said: This Court has held that the Administrative Procedure Act applies to proceedings before the Interstate Commerce Commission. Minneapolis & St. Louis R. Co. v. United States, 361 U.S. 173, 192, 80 S.Ct. 229,4 L.Ed.2d 223 (1959). Appellees claim that the Commission’s procedure here departed from the provisions of 5 U.S.C. §§ 556 and 557 of the Act. Those sections, however, govern a rulemaking proceeding only when 5 U.S.C. § 553 so requires. The latter section, dealing generally with rulemaking, makes applicable the provisions of §§ 556 and 557 only “[w]hen rules are required by statute to be made on the record after opportunity for an agency hearing . . ..” The Esch Act, authorizing the Commission “after hearing, on a complaint or upon its own initiative without complaint, [to] establish reasonable rules, regulations, and practices with respect to car service 49 U.S.C. § l(14)(a), does not require that such rules “be made on the record.” 5 U.S.C. § 553. That distinction is determinative for this case. “A good deal of significance lies in the fact that some statutes do expressly require determinations on the record.” 2 K. Davis, Administrative Law Treatise § 13.08, p. 225 (1958). Sections 556 and 557 need be applied “only where the agency statute, in addition to providing a hearing, prescribes explicitly that it be ‘on the record.’ ” Siegel v. Atomic Energy Comm’n, 130 U.S.App.D.C. 307, 314, 400 F.2d 778, 785 (1968); Joseph E. Seagram & Sons, Inc. v. Dillon, 120 U.S.App.D.C. 112, 115 n.9, 344 F.2d 497, 500 n.9 (1965). Cf. First National Bank of McKeesport v. First Federal Savings & Loan Assn., 96 U.S.App.D.C. 194, 225 F.2d 33 (1955). We do not suggest that only the precise words “on the record” in the applicable statute will suffice to make §§ 556 and 557 applicable to rulemaking proceedings, but we do hold that the language of the Esch Car Service Act is insufficient to invoke these sections. Because the proceedings under review were an exercise of legislative rulemaking power rather than adjudicatory hearings as in Wong Yang Sung v. McGrath, 339 U.S. 33, 70 S.Ct. 445, 94 L.Ed. 616 (1950), and Ohio Bell Telephone Co. v. Public Utilities Comm’n, 301 U.S. 292, 57 S.Ct. 724, 81 L.Ed. 1093 (1937), and because 49 U.S.C. § l(14)(a) does not require a determination “on the record,” the provisions of 5 U.S.C. §§ 556 and 557 were inapplicable. This proceeding, therefore, was governed by the provisions of 5 U.S.C. § 553 of the Administrative Procedure Act, requiring basically that notice of proposed rulemaking shall be published in the Federal Register, that after notice the agency give interested persons an opportunity to participate in the rulemaking through appropriate submissions, and that after consideration of the record so made the agency shall incorporate in the rules adopted a concise general statement of their basis and purpose. The “Findings” and “Conclusions” embodied in the Com — 1159 — mission’s report fully comply with these requirements, and nothing more was required by the Administrative Procedure Act. United States v. Allegheny-Ludlum Steel Corp., supra at 756-58, 92 S.Ct. at 1950. (Footnote omitted.) Further, in United States v. Florida East Coast R. Co., 410 U.S. 224, 238, 93 S.Ct. 810, 35 L.Ed.2d 223 (1973), the Supreme Court reiterated and reinforced its decision in United States v. Allegheny-Ludlum Steel Corp., supra. In the Florida East Coast R. Co. case the Court held: Section 553 excepts from its requirements rulemaking devoted to “interpretative rules, general statements of policy, or rules of agency organization, procedure, or practice,” and rulemaking “when the agency for good cause finds . that notice and public procedure thereon are impracticable, unnecessary, or contrary to the public interest.” This exception does not apply, however, “when notice or hearing is required by statute”; in those cases, even though interpretative rulemaking be involved, the requirements of § 553 apply. But since these requirements themselves do not mandate any oral presentation, see Allegheny-Ludlum, supra, it cannot be doubted that a statute that requires a “hearing” prior to rule-making may in some circumstances be satisfied by procedures that meet only the standards of § 553. The Court’s opinion in FPC v. Texaco, Inc., 377 U.S. 33, 84 S.Ct. 1105,12 L.Ed.2d 112 (1964), supports such a broad definition of the term “hearing.” Similarly, even where the statute requires that the rulemaking procedure take place “on the record after opportunity for an agency hearing,” thus triggering the applicability of § 556, subsection (d) provides that the agency may proceed by the submission of all or part of the evidence in written form if a party will not be “prejudiced thereby.” Again, the Act makes it plain that a specific statutory mandate that the proceedings take place on the record after hearing may be satisfied in some circumstances by evi-dentiary submission in written form only. We think this treatment of the term “hearing” in the Administrative Procedure Act affords a sufficient basis for concluding that the requirement of a “hearing” contained in § l(14)(a), in a situation where the Commission was acting under the 1966 statutory rulemaking authority that Congress had conferred upon it, did not by its own force require the Commission either to hear oral testimony, to permit cross-examination of Commission witnesses, or to hear oral argument. United States v. Florida East Coast R. Co., supra at 240-41, 93 S.Ct. at 818. Taking those precedents into account, it seems clear to us that the legislative-type hearings conducted by the United States EPA concerning the Ohio S02 control plan were consistent with the provisions of the Clean Air Act and the Administrative Procedure Act, and we further conclude that the hearings are not inconsistent with the due process clause of the Fourteenth Amendment. As pointed out in the quotation from Buckeye Power # 1, supra, Congress did not insert into the Clean Air Act the language requiring the Administrator to make determinations “on the record after an opportunity for an agency hearing” which the Supreme Court has held to trigger the requirement of an adjudicative hearing. And if there was a legitimate due process complaint arising from the fact that petitioners had not had a chance to comment upon the RAM model as employed by United States EPA in its Ohio S02 control plan, we believe it was surely cured by this court’s remand for reopening of the administrative record and United States EPA’s-reconsideration thereafter. We note, as petitioners encourage us to, that some cases in other circuits hold that it is the importance and complexity of the issues decided by the administrative agency which should determine the kind of hearing procedures required rather than any formal classification of the process as either rule-making or adjudicatory. See Appalachian Power Co. v. EPA, 477 F.2d 495, 500-01 — 1160 — (4th Cir. 1973); Walter Holm & Co. v. Hardin, 145 U.S.App.D.C. 347, 449 F.2d 1009, 1015 (1971). Typically, however, it is important and complex problems which Congress assigns to administrative agencies. Thus far neither Congress nor the Supreme Court has elected to adopt such a flexible standard or to assign exclusive responsibility for the choice of agency hearing procedures to the federal courts. Several petitioners also argue that this Circuit should follow the example of the Ninth Circuit in Bunker Hill Co. v. EPA, 572 F.2d 1286 (9th Cir. 1977) (decided July 5, 1977), to the extent of remanding the proposed sulfur dioxide control standards to allow cross-examination of United States EPA’s experts and additional comment thereafter. In contrast to our remand order of November 12, 1976, the Ninth Circuit did provide for cross-examination pertaining to what it termed a highly complex and technical issue concerning the technological feasibility of the use of sulfur burners to effect control of Bunker Hill’s lead smelter emissions. While we believe that cross-examination of an administrative agency’s experts is not a required or normal part of informal rulemaking under Section 553, we do not exclude the possibility that a case may be presented to this court wherein remand for cross-examination about disputed facts will prove both logical and necessary. We do not, however, find any legal requirement or practical need for any more hearings, with or without cross-examination, in order to answer the three major general issues posed in the instant cases. Petitioners have had ample opportunities to present their views to the agency. A full record has been written. There has already been an inordinate delay of five years longer than Congress contemplated. 3. The RAM Model The petitioners in these cases center most of their criticisms upon the United States EPA’s use of the Real-Time Air-Quality-Simulation Model (“RAM”) which was employed by the agency in preparation of the Ohio sulfur dioxide control plan. RAM is a dispersion model which evaluates the interaction of a variety of facts in order to make predictions concerning the contribution to the pollution of the ambient air by specific plants. Its formula takes into account the capacity of each plant on a stack-by-stack basis and adds thereto smokestack height, surrounding terrain, and weather conditions. The model is operated on the assumption that the plants concerned operate 24 hours a day at full capacity and predictions are made for every day of the year. The ultimate standards are set according to the predicted second-worst day in terms of pollution results shown. In comparison to all other prior methods of controlling pollution, RAM starts with a solid, ascertainable data base. This is the established design capacity of the power plants in question related to the sulfur content of the fuel used by each. From these factors the “emissions data” for each plant are developed. When stack height, wind, weather, terrain, land use, etc., are figured in, the RAM model has the additional value of allowing its user to predict with considerable accuracy the relative contributions of specific power plant stacks to the points of maximum concentration of pollution of the ambient air. The RAM model was actually developed as a result of United States EPA’s public hearings on the proposed plan for Ohio after five days of hearings on said proposed plan in Columbus, Cleveland, Cincinnati, and Steubenville at which petitioners involved in this current litigation were given an opportunity (which most accepted) to appear, testify, or submit comments. At those hearings the major source of criticism from industries, including some of the present petitioners, was that the plan then under consideration did not determine limitations by individual stacks to a sufficient degree. EPA in its brief in this case compares the “rollback” model employed in the preparation of the first Ohio plan to dispersion models like RAM, which is now the source of present controversy: Unlike the rollback model, the dispersion models used in developing the pro — 1161 — mulgated plan allow a determination of the cause-effect relationship between the SO2 emissions of the pollution sources in an area and the resulting ambient air quality. Therefore, it is possible to determine the proportion by which each source must reduce emissions to meet ambient standards. With the use of the rollback model, in contrast, each source’s emissions in the region, whether or not they contributed to a pollution problem, were required to be reduced. Through dispersion modeling, emission limitations can now be set with increased precision. Overeontrol is minimized, so that the plan will still insure attainment and maintenance of the air quality standards, but at a much reduced cost to the sources. This is most clearly demonstrated by comparing emission limitations for power plants under the various plans. Power plants account for approximately 80% of the sulfur dioxide emissions in the State. However, achievement of this added precision requires a massive analytical task. Tremendous amounts of data are required for each source analyzed. In addition to the emissions data for each source, dispersion modeling requires detailed information on all the factors that affect the dispersion of emissions. These include the height of the source’s stack (or usually stacks), the spatial orientation of the sources to each other, the topography of the area and the effects it will have on dispersion, and, of crucial importance, detailed weather data for the area. All this information is needed so that the computer analysis reflects actual conditions. For example, a gaseous pollutant emitted over a grassy field will disperse much differently than if the pollutant is emitted over a large urban area. There the dispersion will be affected not only by the local weather conditions but also by the greater turbulence caused by the different types of surface areas and heat sources throughout a city. EPA goes on to point out that there are more than 1,000 point sources in the State of Ohio and more than 2,000 area sources, and that in relation to emission data, United States EPA utilized (among other sources) the data base on sulfur dioxide required to be reported to the State of Ohio under Ohio Rev.Code Ann. §§ 3704.03(1), 3704.05(C) (Page 1971 & 1976 Supp.). It is, of course, no part of the responsibility of this court to determine whether the RAM model represents the best possible approach to determining standards for the control of sulfur dioxide emissions. Our standard of review of the actions of United States EPA is whether or not the action of the agency is “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” Clean Air Act Amendments of 1977, Pub.L. No. 95-95, § 305(a), 91 Stat. 775 (to be codified as 42 U.S.C. § 7607(d)(9)(A)). Thus, we are required to affirm if there is a rational basis for the agency action and we are not “empowered to substitute [our] judgment for that of the agency.” Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402, 416, 91 S.Ct. 814, 824, 28 L.Ed.2d 136 (1971). Our review of this record convinces us that we cannot properly hold that United States EPA’s adoption of the RAM model for predicting sulfur dioxide emissions and for fixing maximum levels of sulfur dioxide emissions by specific sources was arbitrary and capricious or beyond the agency’s authority under the Clean Air Act. The factors cited below support EPA’s argument that the RAM model is supported by sufficient evidence so that EPA’s adoption cannot be held arbitrary and capricious: 1) United States EPA’s use of the “rollback” model — the principal basis of its first plan on which five days of public hearings were conducted in Ohio — was strenuously objected to by representatives of many of the present petitioners because it was not source-specific and, as a consequence, tended to require more stringent sulfur dioxide controls than would be required if plant capacity, fuel, population, smokestack height, wind and climate were all taken into account. Thus John R. Martin, of Smith & Singer Meterologists, Inc., commented on behalf of Ohio utilities on the first United States EPA plan as follows: More sophisticated modeling is necessary in all seven of the urban counties — 1162 — that use the proportional rollback. In this way, the Federal air quality standards can be attained without unnecessary SO2 emission restrictions being imposed upon sources that do not contribute to an SO2 problem. * * * * * * We recommend that new strategies be tested which will more fairly identify and control SO2 sources that create SO2 problems. Similarly, Dr. Howard M. Ellis, of Enviro-plan, Inc., said on behalf of Ohio power plants: [I]n developing an SO2 control program for this plant, Region V did not consider economically efficient alternatives to constant uniform emission standards — alternatives such as utilizing a supplementary control system to achieve air quality standards or using separate S02 emission standards by stack in accordance with each stack’s contribution to ground-level SO2 concentrations. Separate emission standards by stack can reduce considerably the cost of achieving air quality standards . 2) EPA responded to these arguments favorably by devising and adopting the RAM model which did employ all of these source-specific factors. 3) Further, as shown on the following charts, the United States EPA 1976-1977 SO2 control plan (principally based upon the RAM and MAX-24 models) shows less stringent regulation on a county-by-county basis when compared to the Ohio SO2 control plans originally promulgated in 1972 and 1974. In addition, when the comparison is limited to petitioners involved in this litigation, but including all of their facilities which were subjected to RAM modeling (and which are identified in this record), we find the plan slightly less strict on a facility comparison basis than the Ohio 1972 plan by a count of 24 to 17, and slightly more strict than the Ohio 1974 plan by a count of 23 to 20. These comparisons do not, of course, necessarily demonstrate RAM’s accuracy. Rather, the comparison with Ohio’s previous plans (based upon the earlier rollback model which was used and accepted nationwide) tends simply to show that the choice of RAM modeling lay within administrative discretion. RELATIVE STRINGENCY OF US EPA REGULATIONS AND PREVIOUSLY PROMULGATED REGULATIONS Ohio EPA Ohio EPA 1. US EPA 1976-77 1972 1974 urban RAM regs are: regs for: regs for: stricter than 14 20 of petitioners’ facilitiesa less strict than 21 17 " the same as 1 1 " ambiguousb compared with 6 4 " 2. US EPA 1976-77 rural RAM regs are: stricter than 3 3 " less strict than 3 3 " the same as 0 0 " ambiguousb compared with 0 0 " 3. US EPA 1976-77 regs (all models) are: stricter than 4 7 Ohio countiesc less strict than 40 35 " the same as 0 0 " ambiguousb compared with 24 26 " 4. US EPA 1976-77 regs (all models) are: stricter than 22 32 of petitioners’ facilitiesa less strict than 50 43 " the same as 1 1 " ambiguousb compared with 14 11 " a Including facilities to the regulation of which petitioners do not object. b I.c., stricter for some stacks or facilities and less strict for others; or employing different units of measurement, rendering comparison impossible; or insufficient data available for meaningful comparison. c Twenty other counties contain no point sources of SO2 emissions. All comparisons are based on the data set out in Appendices A, B and C. 4) While this court has currently before it some 32 petitioners protesting the United States EPA’s plan for SO2 emission control for Ohio, it must be remembered that Ohio is estimated to have over 1,000 point sources and over 2,000 area sources of S02 pollution. 5) The RAM model is a general formula which can be applied to many individual sources of pollution to derive specific estimates of SO2 emission rates for each. It employs a wider, more complete and more accurate data base than any prior model yet employed in devising a sulfur dioxide control strategy for a state or county. The crucial data with which the RAM model starts are the design capacity figure, plus the fuel sulfur content, from which is computed the SO2 emission rate for each of the heating or power plants sought to be con — 1163 — trolled. Thus at the outset the RAM model starts with ascertainable specific figures for each source where disputes can be resolved by inspection of the equipment or fuel concerned. Many of the additional components such as stack height, wind direction, physical relationship of sources to each other, and topography of the area are similarly ascertainable as matters of fact. With the enormous financial stakes involved in this litigation, every effort to avoid disputes about the accuracy of the data base should be made. This record shows that United States EPA’s design of the RAM model was brought about at least in large part by Ohio industry’s requests for greater specificity and hence lower costs of compliance with National Air Quality Standards. 6) While there may yet be developed (and hopefully will be) a better method of establishing a control strategy for sulfur dioxide emissions than the RAM model, no one has yet come forward with such. Nor do peti- , tioners point to any such. > This is not to ignore that petitioners do cite Enviroplan’s claims of a superior model termed Air Pollution Evaluation System. This record shows, however, that United States EPA asked for the Enviroplan model and was refused, and is now refused the operative details of that model on the grounds of proprietary interest. While such withholding may be both defensible as a matter of law and understandable as a matter of economics, this court cannot consider Enviroplan’s model as available technology until and unless it is fully disclosed and evaluated by United States EPA — the agency charged by Congress with making these decisions. 7) We recognize that this record does not present positive proofs of the accuracy of RAM’s predictions. Thus far technology has not developed foolproof methods for validating predictions concerning pollution of the ambient air absent years of collection of monitoring data with far more monitors and far more personnel than have thus far been available. Obviously, also, the monitor locations and the receptor sites for the RAM predictions must correspond. We find such identity of monitor locations and receptor sites available in this administrative record for the City of Dayton. The EPA Appendix contains: 1) RAM model computer printouts showing predicted second-highest 24-hour concentrations of sulfur dioxide for several receptor sites, and the location of those sites (EPA Appendix, Vol. IV, at 85-94, Certified Index XIII.EEE.1.a.3.); 2) Air quality data for 1972-76 at several Dayton sulfur dioxide monitors (EPA Appendix, Vol. IV, at 61, 79, Certified Index XV.K.2.r. and s.); and 3) The locations for the Dayton monitors (EPA Appendix, Vol. IV, at 95-96, Certified Index XII.B.4.a.(1)(d)). The following chart displays the data contained in these documents: Second-Highest 24-hr Concentration Predicted Actual Monitor Headings (micrograms per cubic meter) Site by RAM (micrograms No. per cubic meter) 1972 1973 1974 1975 1976 1 195 • • 219 • • 2 201 73 438 181 163 81 3 83 0 ° 117 62 57 4 109 0 0 151 109 17 5 161 .57 198 0 68 41 6 207 0 13 66 110 75,.. Our analysis of these data4 shows that the yearly second-highest concentration of SO2 pollution (for a 24-hour average) actually recorded on available monitors exceeded the RAM model prediction for each location once in a five-year period at five out of six locations. This analysis certainly falls short of showing RAM’s predictive perfection. But it certainly tends to show that the EPA’s use of RAM, if conservative, cannot be held to be arbitrary and capricious. See Sierra Club v. EPA, 176 U.S. App.D.C. 335, 540 F.2d 1114, 1136 (1976), cert. granted on different issues, 430 U.S. 953, 97 S.Ct. 1597, 51 L.Ed.2d 802 (1977). See also Mision Industrial, Inc. v. EPA, 547 F.2d 123, 128-29 (1st Cir. 1976). We recognize that petitioners presented Enviroplan’s study on RAM predictions and — 1164 — existing monitor results for the three counties in which Akron, Toledo and Canton are located, and that they contend that the study’s results show gross overpredictions by RAM. Reevaluation of RAM predictions, however, showed that most discrepancies were occasioned by data errors factored into the original RAM predictions for these three counties. We have also considered the argument based on the Hamill study of RAM’s application to St. Louis and Enviroplan’s subsequent study thereof. While this latter commentary must be taken into account, when weighed against all other record evidence it fails to convince us that United States EPA’s use of RAM was arbitrary and capricious. 8) Finally, as we pointed out at the beginning of this opinion, S02 emissions have a direct impact upon the health and the lives of the population of Ohio — particularly its young people, its sick people, and its old people. If the RAM model did overpredict emission rates, such a conservative approach in protection of health and life was apparently contemplated by Congress in requiring that EPA plans contain “emission limitations . . . necessary to insure attainment and maintenance” of national ambient air standards. 42 U.S.C. § 1857c-5(a)(2)(B) (1970). (Emphasis added.) In summary, we hold that United States EPA’s adoption and employment of the RAM model as its general working tool was based upon informal rulemaking which satisfied the requirements of both the Clean Air Act and the Administrative Procedure Act, and the due process requirements of the United States Constitution. Further, the record indicates that the Administrator’s action in promulgating the sulfur dioxide control regulations for Ohio through use of the RAM model was a rational choice which was well within the discretion committed to him and his agency. We decline petitioners’ requests to set the disputed orders aside on the ground that they are arbitrary and capricious. OTHER ISSUES Somewhat half-heartedly the leading brief for the utilities attacks the United States EPA plan for S02 controls in Ohio as excessively costly and asserts that the satisfactory operation of Flue Gas Desulfurization machinery (“scrubbers”) has not been demonstrated. We note that the United States EPA control strategy for Ohio does not rely heavily upon Flue Gas Desulfurization. (EPA estimates — and petitioners do not dispute — that only six utilities will choose this compliance route.) Alternatives to installation of “scrubbers” are the purchasing and use of low sulfur coals or the employment of coal cleaning or blending techniques. There is no doubt, of course, that S02 controls will indeed be costly. EPA estimates capital costs for Ohio industry of well over half a billion dollars and annual costs of 171 million dollars. It also projects these costs as requiring a 3% increase in annual electric bills for the consumers who will ultimately pay them — and who will also breathe the less polluted air. Basically the choice of economic burden versus continued deterioration of the air we breathe was made by Congress. In this litigation no issue is raised concerning Congress’ power to do so. We have genuine doubt that this court has the power to review what we regard as petitioners’ slightly disguised economic and technological infeasibility arguments. See generally Union Electric Co. v. EPA, 427 U.S. 246, 265-66, 96 S.Ct. 2518, 49 L.Ed.2d 474 (1976).5 Since this issue does not appear to be definitely resolved as to a United States EPA-designed implementation plan (such as we deal with here), see Union Electric Co. v. EPA, supra at 261 n.7, 96 S.Ct. 2518, we observe that if we did have such power, we would conclude that the technical record compiled in the agency proceeding provides ample support for the economic and technological feasibility of — 1165 — the SO2 control strategies which United States EPA has promulgated for Ohio. Petitioners, Cleveland Electric Illuminating Co. and Dayton Power & Light Co., enter objections to the classification of three power plants as “urban.” Inspection of the geographic location of these plants and of the population distribution in the near vicinity convinces us that the United States EPA. classification cannot properly be termed arbitrary and capricious. Power plant pollution of the ambient air is no respecter of municipal boundary lines. We note petitioners’ objection to the use of the urban dispersion coefficients employed in the RAM model. EPA defends their use by citing the St. Louis study and by noting that no better or more accurate coefficients are available. Since to this observation petitioners reply by asking for more study, we conclude that study should progress while the purposes of the Clean Air Act are being served rather than by indefinite postponement of EPA’s mandated task of reducing SO2 pollution in Ohio. No other material issues are presented. One petition pending before this court from the Northern Ohio Lung Association attacks United States EPA’s failure to promulgate a separate implementation plan for the “secondary standards” for the ambient air. This petition will be the subject of separate consideration. Similarly, this opinion does not govern any petitions where the RAM model was not used. We do not decide any specific fact disputes raised by any petitioner as to plants other than those treated in this opinion.6 Decision of these cases will follow. For the reasons stated above, the decision of the Administrator in imposing the S02 control plan is affirmed subject to the reservations indicated above. No costs are allowed since important public questions are involved. APPENDIX A COMPARISON OF PROMULGATED REGULATIONS CONCERNING SULFUR DIOXIDE EMISSION LIMITATIONS FOR THE STATE OF OHIO Footnote list: a. Source: Ohio Environmental Protection Agency Reg. EP-11-14 (AP-3-14), effective July 17,1972 (available from Anderson Publishing Co., Cincinnati, Ohio). See Appendix B. b. Source: Ohio Environmental Protection Agency Reg. EP-11-13, effective Feb. 1, 1974 (available from Anderson Publishing Co., Cincinnati, Ohio). See Appendix C. c. Source: 41 Fed. Reg. 36,324-40 (Aug. 27, 1976); 41 Fed. Reg. 42,455-56 (Nov. 30, 1976); 42 Fed. Reg. 27,588-93 (May 31,1977). d. All figures represent lbs. S02 per million British Thermal Units (mBTU) heat input, unless otherwise specified. e. Located in Morgan and Washington Counties. f. Optional compliance equation omitted from chart. g. No present objection to regulation of this facility. h. Not regulated. i. Excluding sources subject to New Source Performance Standards. j. Subject to New Source Performance Standards. k. Additional limitations, specified in units other than lbs. S02/mBTU heat input, omitted from chart. m. Information not available. — 1166 — [[Image here]] — 1167 — [[Image here]] — 1168 — [[Image here]] — 1169 — [[Image here]] — 1170 — [[Image here]] — 1171 — [[Image here]] — 1172 — [[Image here]] — 1173 — [[Image here]] — 1174 — [[Image here]] Appendix B to follow. — 1175 — APPENDIX B OHIO EPA 1972 REGULATIONS EP-11-14 (AP-3-14) Restrictions on emission of sulfur dioxide from the use of fuel. (B) Emission limitations. (1) No person shall cause, suffer, allow, or permit the emission of sulfur compounds caused by the combustion of fuel in fuel-burning equipment from any stack or chimney in excess of the quantity set forth in Figure III. (2) All persons located within air control regions classified as Priority I Regions shall attain or exceed that degree of emission reduction specified by Curve P-1 by the effective date of this regulation. (3) All persons located within air quality control regions classified as Priority II Regions shall attain or exceed that degree of emission reduction specified by Curve P-2 by the effective date of this regulation. (4) All persons located within air quality control regions classified as Priority III Regions shall attain or exceed that degree of emission reduction specified by Curve P-3 by the effective date of this regulation. (5) All persons located within air quality control regions classified as Priority II or III Regions shall attain or exceed, no later than July 1, 1975, that degree of emission reduction specified by Curve P-1. (Adopted July 6,1972; effective July 17,1972.) — 1176 — [[Image here]] — 1177 — EP-11-06 (AP-3-06) Classification of regions. CLASSIFICATION BY REGIONS (See Figure IV at end of EP-11) FEDERAL PRIORITY CLASSIFICATION — BY POLLUTANT — FEDERAL REGISTER, MAY 31, 1972 AIR QUALITY CONTROL REGION FEDERAL NUMBER PARTICULATES SOx CO NOx PhtOx Cincinnati (Ky.-lnd.) 079 I II III I I Cleveland (Akron, Canton, etc.) 174 I I III I 1 Columbus 176 I III III I 1 Dayton 173 I II III I 1 Mansfield-Marion 175 II II III III III Marietta (W. Va.) 179 I II III III III Northwest Ohio 177 II I III III III Portsmouth-lronton (Ky.-W.Va.) 103 I III III III III Sandusky 180 III III III III III Steubenville (W. Va.) 181 I I III III III Toledo (Michigan) 124 I I III I I Wilmington-Chillicothe-Logan 182 III III III III III Youngstown 178 I II III III III Zanesville 183 II IA III III III — 1178 — [[Image here]] Appendix C to follow. — 1179 — APPENDIX C OHIO EPA 1974 REGULATIONS EP-11-13 Restrictions on emission of sulfur dioxide from use of fuel. (B) Emission limitations. (1) No person shall cause, permit, or allow the emission of sulfur oxides from any facility as defined in subsection (A)(2) of this regulation in excess of the quantity set forth in Table II (Figure V following EP-11) for the county in which the source is located.. (2) No person shall cause, permit, or allow the emission of sulfur oxides in excess of 1.0 pounds per one million BTU of heat input from any new facility made up of one or more new sources with a combined capacity of 100 million BTU per hour or more for which a Permit to Install or Permit to Construct was not obtained before February 1, 1974. New sources with a combined capacity of less than 100 million BTU per hour shall comply with the emission limitations in paragraph (1) of this subsection. (Adopted January 21,1974; effective February 1,1974.) TABLE II (See Figure V at end of EP-11) COUNTY SULPHUR DIOXIDE ALLOWABLE EMISSIONS Allen 1.0 pounds of sulfur dioxide Ashtabula per million BTU of heat input Columbiana Cuyahoga Jefferson Lucas Summit Trumbull COUNTY SULPHUR DIOXIDE ALLOWABLE EMISSIONS Belmont 1.6 pounds of sulfur dioxide Butler per million BTU of heat input Coshocton Erie Hamilton Henry Lake Lawrence Lorain Mahoning Monroe Montgomery Pickaway Richland Tuscarawas Washington Wayne Franklin 3.2 pounds of sulfur dioxide ..Hancock per million BTU of heat input Hocking Holmes Knox Licking Morgan Ross Seneca Shelby — 1180 — COUNTY SULPHUR DIOXIDE ALLOWABLE EMISSIONS Athens Auglaize 4.0 pounds of sulfur dioxide per million BTU of heat input Carroll Clinton Defiance Fairfield Fayette Greene Hardin Harrison Highland Huron Jackson Marion Mercer Miami Muskingum Paulding Sandusky Union Vinton Williams Wood All other counties: 4.8 pounds of sulfur dioxide per million BTU of heat input — 1181 — [[Image here]] Footnotes [1] This decision dismisses the objections to the regulations that apply to the following facilities: (a) Cleveland Electric Illuminating Co. — all facilities. (b) Dayton Power & Light Co. — Montgomery County facilities only. (c) Ohio Edison Co. — Lorain County facilities only. (d) Toledo Edison Co. — all facilities. (e) The Timken Co. — all steam generating units. (f) White-Westinghouse Corp. — all facilities. (Although there was some confusion on this point in the briefs, the record makes clear that White-Westinghouse’s Franklin County facility is subject to the RAM model. See EPA Final Technical Support Document at IV-57.) (g) Standard Oil Co. of Ohio — Lucas County steam generating units. (h) Interlake, Inc. — all steam generating units. (i) Coulton Chemical Corp. — all steam generating units. [2] The Clean Air Act was originally enacted in 1963, Pub. L. No. 88-206, 77 Stat. 392. It was amended in relatively minor ways three times during the following six years. Pub. L. No. 89-272, 79 Stat. 992 (1965); Pub. L. No. 89-675, 80 Stat. 954 (1966); Pub. L. No. 90-148, 81 Stat. 485 (1967). The Act’s present form, however, is derived from amendments adopted in 1970 and subsequently. Clean Air Act Amendments of 1970, Pub. L. No. 91-604, 84 Stat. 1676, as amended, Pub. L. No. 92-157, 85 Stat. 464 (1971); Pub. L. No. 93-319, 88 Stat. 246 (1974); Pub. L. No. 95-95, 91 Stat. 685 (1977). The Act is being recodified as 42 U.S.C. §§ 7401-7626. [3] The EPA plan for Ohio presently under con- • sideration contains no separate implementation plan for secondary air quality standards. This issue is raised in the petition of the Northern Ohio Lung Association and will be considered and decided subsequently. [4] Petitioners object with vehemence to EPA’s reference in its brief to its discussion of these data as a “study,” calling our attention to the fact that the comparison was made by EPA attorneys after the closing of the record. Clearly, however, the raw data to which we have referred and which we have inspected were and are .available in the administrative record of this case. [5] United States EPA’s S02 control plan for Ohio has as its goal the attainment of national air quality standards. It does not seek to exceed them. See Note, The Clean Air Act: "Taking a Stick to the States,” 25 Cleve.St.L. Rev. 371, 405 (1976). [6] See note 1, supra.