Kennecott Copper Corp. v. Train

OPINION Before TUTTLE,* KOELSCH and BROWNING, Circuit Judges. BROWNING, Circuit Judge: Kennecott Copper Corporation1 petitions for review of an order of the Environmental Protection Agency (EPA) rejecting a portion of the State of Nevada’s implementation plan under the Clean Air Act2 relating to control of sulfur dioxide (SO2), and substituting provisions formulated by EPA. The problem arises from a single source of SO2 emissions in Nevada — Kennecott’s copper smelter at McGill in White Pine County.3 EPA based its order upon an interpretation of the Clean Air Act which requires that national air quality standards be met by continuous emission limitations to the maximum extent possible, and that intermittent controls and dispersion systems by used only when continuous emission controls are not economically feasible. This court denied Kennecott’s request for a temporary injunction against enforcement of the substitute plan promulgated by EPA but expedited Kennecott’s appeal. We affirm EPA’s order. I Section 109 of the Clean Air Act, 42 U.S.C. § 1857c-4, requires EPA to promulgate national primary and second [1151] ary air quality standards.4 Section 110 of the Act, 42 U.S.C. § 1857c-5, applicable to existing sources of pollutants such as Kennecott’s McGill smelter,5 provides that the states must devise plans to implement, maintain, and-enforce these national standards. EPA must approve state implementation plans if they are adopted after reasonable notice and hearing, and meet other specified requirements. EPA must disapprove any state plan that does not comply with the statute, and propose and adopt a plan of its own. 42 U.S.C. § 1857c-5(c)(1). EPA approved the provisions of Nevada’s implementation plan relating to control strategy except those involving control of S02 at Kennecott’s McGill smelter, the sole stationary source of this pollutant in the Nevada Intrastate Air Quality Control Region.6 Nevada submitted an amended plan.7 EPA rejected the state’s amendments, and proposed,8 and eventually adopted,9 a plan of its own. The amended Nevada plan provided for a 60 percent reduction of S02 emissions from the McGill smelter by installation of a plant to convert S02 to sulfuric acid. When weather conditions are so adverse that the 60 percent reduction in emissions resulting from operation of the acid plant would not be sufficient to maintain national air quality standards, the Nevada plan provided for reducing the level of production at the smelter. EPA rejected the Nevada plan on the ground that an 86 percent reduction of S02 emissions from the McGill smelter was required to achieve air quality standards. EPA recognized that it was not presently economically feasible to install an acid plant or other constant emission control that would reduce S02 emission from the McGill smelter more than the 60 percent contemplated by the Nevada plan. The EPA plan therefore provided that, as an interim measure, Kennecott might use continuous emission reduction technology capable of reducing emissions by 60 percent, together with such other controls (including reducing production and use of a tall stack) as might be needed to maintain national air standards.10 However, EPA’s plan also provided that, until full compliance with national air quality standards is achieved entirely by means of continuous emission reduction,11 Kennecott must undertake a research program to improve continuous emission control technology,12 and must adopt such improved technology as it becomes available for use at the McGill smelter on an economically feasible basis.13 [1152] Kennecott’s basic position is that EPA is not authorized to require continuous emission reduction techniques in preference to intermittent controls or other methods for dispersion, or dilution, of pollutants. Kennecott contends that EPA must approve a state implementation plan that provides for any combination of continuous emission controls and alternative control systems devised by the state, so long as the state plan will attain and maintain national air quality standards within the statutory time periods. This view of the statute underlies Kennecott’s opposition to EPA’s requirement that Kennecott engage in research to develop and apply constant emission control technology.14 It is also the principal source of Kennecott’s objection to EPA’s determination that an 86 percent reduction in S02 emissions from the McGill smelter is required to meet national air quality standards. EPA’s calculation rests upon measurements made prior to the installation of a new 750-foot tall smokestack at McGill. Kennecott submitted data to EPA indicating that the new tall stack resulted in a 92 percent decrease in ground level S02 concentrations.15 EPA refused permanent credit for this reduction because, in EPA’s view, the statute requires usé of continuous emission reduction technology, as opposed to dispersion techniques, whenever economically feasible.16 [1153] II EPA bases the requirement of constant emission controls upon section 110(a)(2)(B) of the Clean Air Act, 42 U.S.C. § 1857c-5(a)(2)(B).17 This subsection provides that EPA shall approve a state implementation plan if “it includes emission limitations, schedules, and timetables for compliance with such limitations, and such other measures as may be necessary to insure attainment and maintenance of such primary or secondary standard, including, but not limited to, land-use and transportation controls; . ” (emphasis added). EPA reads the phrase “as may be necessary” as modifying only “such other measures,” and not “emission limitations.” In EPA’s view, measures other than emission limitations are therefore permissible only if “necessary” to achieve applicable air quality standards; such “other measures” are not “necessary” if economically feasible emission limitation technology is available. EPA supports its interpretation of section 1857c-5(a)(2)(B) by references to. the language and legislative history of the Clean Air Act Amendments of 1970. EPA relies upon the fact that an option to utilize intermittent controls or tall stacks carries the potential for evasion of the intent of Congress that emission limitations be included in implementation plans. EPA invokes the policy of nondegradation of the quality of the nations’s air, implied from the Act.18 EPA also draws support from the Supreme Court’s decision in Train v. Natural Resources Defense Council, Inc., 421 U.S. 60, 95 S.Ct. 1470, 43 L.Ed.2d 731 (1975), and from the provisions and legislative history of the Energy Supply and Environmental Coordination Act of 1974. The Courts of Appeals for the Fifth and Sixth Circuits have adopted the interpretation of section 1857c-5(a)(2)(B) urged by EPA. Natural Resources Defense Council, Inc. v. EPA, 489 F.2d 390 (5th Cir. 1974), reversed in part on other grounds sub nom., Train v. Natural Resources Defense Council, Inc., 421 U.S. 60, 95 S.Ct. 1470, 43 L.Ed.2d 731 (1975); Texas v. EPA, 499 F.2d 289, 311 (5th Cir. 1974); and Big Rivers Electric Corp. v. EPA, 523 F.2d 16 (6th Cir. 1975). In Natural Resources Defense Council, the Fifth Circuit held that EPA may approve tall stack dispersion techniques in a state implementation plan “only (1) if it is demonstrated that emission limitation regulations included in the plan are sufficient standing alone, without the dispersion strategy, to attain the standards; or (2) if it is demonstrated that emission limitation sufficient to meet the standard is unachievable or infeasible, and that the state has adopted regulations which will attain the maximum degree of emission limitation [1154] achievable.” 489 F.2d at 410 (emphasis in original).19 We agree, in general, with the Fifth Circuit’s analysis (489 F.2d at 403-09) of the history and contents of the Clean Air Act Amendments of 1970 leading to this conclusion.20 In Big Rivers Electric Corp., the Sixth Circuit extended the Fifth Circuit’s decision to intermittent emission control systems, holding that section 1857c-5(a)(2)(B) authorizes EPA to reject state plans providing for the use of intermittent emission control systems without a showing that constant emission controls are unavailable.21 The Sixth Circuit based its conclusion upon the Fifth Cir [1155] cuit’s analysis of the Clean Air Act Amendments of 1970, and upon the Supreme Court’s discussion of section 1857c — 5(a)(2)(B) in Train v. Natural Resources Defense Council. Train parses section 1857c-5(a)(2)(B) as EPA’s interpretation suggests (421 U.S. at 67, 95 S.Ct. 1476), and adopts the position that a state plan must include provisions regulating the composition of substances emitted from a given source. 421 U.S. at 78, 95 S.Ct. at 1481. The Sixth Circuit noted that since composition means “kind and amount,” Train recognizes that a state plan must limit the amount of a pollutant. 523 F.2d at 21-22. We agree with the Sixth Circuit that Train adds significant support to EPA’s interpretation of section 1857c-5(a) (2)(B) as expressing a preference for emission limitations.22 Intermittent control systems (such as those restricting production, or utilizing less polluting fuels, during periods of adverse weather) do limit the amount of pollutant emitted while such controls are being applied. However, the reliability and enforceability of such controls is questionable; they may not be implemented when they are in fact needed.23 Moreover, there is no assurance that temporary reductions in emissions resulting from such controls will not be balanced, or even exceeded, by an increase in the amount of pollutant emitted when weather conditions improve and production is increased to make up for prior losses, or more polluting fuels are again used. Thus, intermittent controls, like tall stacks, may only disperse the pollutant rather than reduce it. Tall smokestacks disperse a pollutant through greater quantities of air; intermittent control systems disperse pollutant peaks through longer periods of time. Neither assures a reduction in the quantity of the pollutant eventually emitted.24 Under section 1857c-5(a)(2)(B), EPA may require that assurance. • We therefore agree with the Sixth Circuit that EPA acted within its authority in refusing to accept intermittent controls as adequate compliance with the statute. Kennecott presses one argument, based upon the 1970 amendments, not addressed by either circuit. Section 1857c-5(a)(2)(B) is applicable to existing [1156] sources of pollutants; section 1857c-6(b)(1)(B) governs new sources. Congress intended the standard applicable to the latter to be the more stringent. Kennecott argues that EPA’s interpretation of section 1857c-5(a)(2)(B) would make the new source and existing source standards the same. Kennecott is mistaken. The new source standard requires “the degree of emission limitation achievable through the application of the best system of emission reduction” without regard to air quality standards. 42 U.S.C. § 1857c-6(a)(l). In contrast, existing sources, under EPA’s view of section 1857c-5(a)(2)(B), need only reduce emissions to the extent necessary to comply with air quality standards. Moreover, existing sources, as opposed to new sources, need adopt only such systems of emission reduction as are economically feasible, so long as interim compliance with air quality standards may be achieved by “other measures.” These differences are real. The record discloses that technology capable of reducing S02 emissions by 80 percent is available and is now installed at two Kennecott smelters. Such technology has not been required at McGill only because the required investment would be uneconomic at that smelter. See note 13. Ill Congress again amended the Clean Air Act in 1974. The Energy Supply and Environmental Coordination Act of 1974 added a new section 119 to the Act, 42 U.S.C. § 1857c-10. The provisions of this amendment assume that section 1857c-5(a)(2)(B) has the meaning attributed to it by the Fifth and Sixth Circuits and by EPA. The legislative history adds further significant support to this interpretation. The 1974 Amendments were a response to the 1973 oil embargo and resulting energy crisis. They provide, among other things, for temporary suspension of emission limitations where fuels necessary to compliance become unavailable, or a plant burning oil or gas converts to coal. A predecessor to the 1974 Amendments was adopted during the first session of the 93d Congress, but was vetoed.25 The legislative history of the aborted statute throws light on the proper construction of section 1857c-5(a)(2)(B). As it passed the House, the statute contained a provision (originating in the so-called Murphy Amendments) implying that the Clean Air Act did not require emission limitations, and permitting the permanent use of intermittent controls at certain emission sources.26 [1157] This provision did not survive the Conference Committee. As adopted, the Conference substitute permitted use of alternative or intermittent controls only by pollution sources that converted to combustion of coal, and then only under such conditions as “would require incremental steps toward compliance by utilization of low sulfur coal or coal by-products, or by continuous emission reduction systems.”27 The 1973 legislation assumed the correctness of EPA’s construction of section 1857c-5(a)(2)(B). The same pattern was repeated with respect to the bill adopted in the next session of the 93d Congress and signed by the President — again Congress assumed that the Clean Air Act is to be construed as requiring the use of continuous emission reduction systems if available, and. again Congress rejected an affirmative effort to modify the statute to alter this construction. In March, 1974, Russell E. Train, Administrator of EPA, transmitted a proposed bill to the Speaker of the House of Representatives which, in pertinent part, was the same in substance as the statute eventually adopted.28 Mr. Train also transmitted two proposals advanced by other agencies of the Executive branch which, he wrote, “I do not support.” One of these proposals would have rejected the nondegradation policy, collaterally relevant here.29 The other, directly relevant, [1158] would have amended section 1857c — 5 expressly to bar the interpretation of subsection (a)(2)(B) adopted by the Fifth and Sixth Circuits and EPA, and to require the interpretation urged by Kennecott. The proposed bill would have added a new subsection to section 1857c-5 providing that nothing in that section was to be construed “to preclude use of alternative or intermittent control measures” which are reliable and enforceable and “will permit attainment and maintenance of the national ambient air quality standards.”30 Mr. Train described the Administration’s proposed bill as a victory over the two proposals favored by other executive agencies, which, he said, would “significantly weaken the Clean Air Act.” He pointed out that these two proposals were submitted to Congress by the Administration only as issues to be considered, not as recommended changes. He noted that the proposal regarding intermittent controls, rejected by the Administration, was intended “to allow the use of intermittent control strategies as a permanent method for achieving compliance with stationary source emission standards.” Mr. Train reaffirmed EPA’s contrary position, stating, “As in the past, EPA will contend that the intermittent controls can be used only as an expedient, temporary control measure"31 EPA’s position prevailed in Congress as it had within the Executive branch. The 1974 Amendments did not include the proposal approving general permanent use of intermittent controls to achieve national air quality standards,32 or the proposal rejecting the policy of nondegradation. Intermittent controls were allowed only in carefully limited situations, and then only temporarily. Plants converting from oil or natural gas to coal might obtain an extension of compliance dates but only on condition that they enter into binding contracts for a long term supply of coal that would enable them to meet emission requirements, or contract for continuous emission reduction systems necessary to enable them to achieve the required degree of emission reduction. EPA was empowered to impose such “interim requirements” as were “reasonable and practicable,” but the required degree of emission reduction was to be achieved no later than a date certain set by the statute. 42 U.S.C. § 1857c — 10(c)(2)(B) & (C). The required degree of emission limitation was fixed by reference to the state implementation plan applicable to the particular plant, but no one suggested that the statutory scheme might be frustrated because such a plan might seek to attain air quality standards by means other than emission reduction. On the contrary, Congress clearly understood that existing law required that air quality standards be achieved by emission reduction, and that other control measures were to be employed only temporarily. Thus, in the course of presenting the Conference Report on the 1974 Amend [1159] ments to the Senate, Senator Muskie, Chairman of the Subcommittee on Environmental Pollution and manager of the bill in the Senate, pointed out that “intermittent control strategies are permitted as an interim measure applicable to coal conversion” if EPA determined they were enforceable.33 After expressing his personal doubt as to the enforceability of intermittent controls, Senator Muskie said: It is these doubts that lead me to underscore the fact that no one should view limited application of enforceable strategies related to this legislation as a precedent for future legislation or as a reinterpretation of the requirements of the existing law which bar the application of intermittent control strategies as a substitute for emission limitations.34 These legislative materials offer convincing evidence that the 93d Congress interpreted the 1970 Amendments to the Clean Air Act to require that air quality standards must be met by continuous emission reduction controls so far as pos [1160] sible.35 The 1974 Amendments confirm this reading of the earlier statute by strong implication. In these circumstances, the rule that “[s]ubsequent legislation declaring the intent of an earlier statute is entitled to significant weight” is appropriately applied. NLRB v. Bell Aerospace Co., 416 U.S. 267, 275, 94 S.Ct. 1757, 40 L.Ed.2d 134 (1974); see Red Lion Broadcasting Co. v. FCC, 395 U.S. 367, 381, 89 S.Ct. 1794, 23 L.Ed.2d 371 (1969); South Terminal Corp. v. EPA, 504 F.2d 646, 669-70 (1st Cir. 1974); cf. Rath Packing Co. v. Becker, 530 F.2d 1295, 1312 (9th Cir. 1975). IV Kennecott spent $4,000,000 to implement the Nevada plan, but suspended further construction on the $30,000,000 project pending completion of this review proceeding. Kennecott states that the McGill smelter will only support the expenditures required by the Nevada plan, and that the smelter must close if Kennecott is required, in addition, to conduct research to devise new constant control technology which would be economically feasible at McGill. EPA’s plan requires the same immediate capital outlay as the Nevada plan with which Kennecott has said it can comply. Contrary to Kennecott’s contention, EPA’s plan would not expose Kennecott to the risk that it would be required to make further capital investments for the installation of uneconomic continuous emission control systems in the future. As EPA has undertaken to assure Kennecott (see note 13), under EPA’s interpretation of section 1857c-5(a)(2)(B), EPA could not compel Kennecott to install additional emission reduction systems at McGill unless it were economically feasible for Kennecott to do so. The order of the Environmental Protection Agency is sustained. Footnotes [1] The State of Nevada intervened in support of Kennecott. [2] 42 U.S.C. § 1857 et seq. Jurisdiction of this court is under 42 U.S.C. § 1857h-5(b)(l) (regarding petitions for review or orders of the administrator promulgating plans under § 1857c-5(c)). [3] For those not acquainted with the copper smelters’ problem of controlling sulfur dioxide emissions, a few facts will help put the matter in perspective^] Most copper produced in the U. S. is obtained from sulfide ores. The average copper concentrate contains about one ton of sulfur for each ton of copper, and in the smelting process this sulfur is driven off, largely in the form of sulfur dioxide. Each ton of sulfur produces two tons of sulfur dioxide which is emitted in gas streams of varying concentrations. Muth, Origins and Current Status of Sulfur Oxide Emission Standards for Nonferrous Smelters, reprinted in Implementation of the Clean Air Act Amendments of 1970, Hearings before the Subcomm. on Air & Water Pollution of the Senate Comm, on Public Works, 92d Cong., 2d Sess., at 553 (1972) [hereinafter cited as 1972 Oversight Hearings], About 90% of the sulfur dioxide is captured by a smelter’s gas collection system and is vented into the atmosphere through smokestacks. The remaining approximately 10% escapes the smelter as “fugitive emissions.” See Brief Of Petitioner Kennecott at 5 n. 1. Senator Eagleton explained the importance of sulfur oxide emissions from copper smelters as follows: In Clean Air Act implementation, the copper industry has a significant role. Smelters annually pour 4 million tons of sulfur oxide (S02) [sic] into the atmosphere. The S02 from copper smelters constitutes one-ninth of the total S02 problem with steam generators for electric power the largest source of S02 pollution — 20 million tons annually. SO„ causes about $8.5 billion damages annually 1972 Oversight Hearings, supra, at 435. [4] The “primary standard” fixed by EPA is that “requisite to protect the public health.” 42 U.S.C. § 1857c-4(b)(l). The “secondary” standard “shall specify a level of air quality [which] is requisite to protect the public welfare from any known or anticipated adverse effects associated with the presence of [an] air pollutant in the ambient air.” 42 U.S.C. § 1857c-4(b)(2). [5] The principal question presented in this case would not arise if the McGill smelter were a proposed facility rather than an existing one. Section 111 of the Clean Air Act, 42 U.S.C. § 1857c-6, provides that a new source of pollutant emissions must utilize the best available technology to reduce emissions without regard to air quality standards. [6] 37 Fed.Reg. 10879 (May 31, 1972). [7] 39 Fed.Reg. 38104-05 (Oct. 29, 1974). [8] Id. [9] 40 Fed.Reg. 5511 (Feb. 6, 1975). [10] 40 C.F.R. § 52.1475(e)(7); 40 Fed.Reg. 5515 (Feb. 6, 1975). [11] 40 C.F.R. § 52.1475(e)(15); 40 Fed.Reg. 5517 (Feb. 6, 1975). [12] 40 C.F.R. § 52.1475(e)(13) & (15)(i)(b); 40 Fed.Reg. 5516 & 5517 (Feb. 6, 1975). [13] In the preamble to its Nevada S02 regulations EPA stated; [T]he selective use of Supplementary Control Systems (systems which limit pollutant emissions during periods when meteorological conditions are conducive to ground level concentrations in excess of the National Ambient Air Quality Standards) to attain and maintain the ambient standards is consistent with the Clean Air Act when the only alternatives are permanent production curtailment, shutdown, or delays in the attainment of the national standards. . EPA will approve such measures as intermittent production curtailment and use of [1152] dispersion techniques, including tall stacks, as an addition to available constant control measures, until such time as the treatment of weak gas streams can be accomplished through reasonable retrofit control techniques. Evaluation of the availability of constant control techniques which may be developed in the future will be made using the same criteria as were used in developing these regulations. Namely, [Supplementary Control Systems] will continue to be allowed where permanent production curtailment, shutdown or delays in attainment of national standards are the only other alternatives. 40 Fed.Reg. 5508, 5510 (Feb. 6, 1975). [14] This is the only ground upon which Kennecott challenges EPA’s requirement that Kennecott undertake a research program to develop improved continuous emission control technology and adopt such technology when it is economically practicable. Kennecott does not argue, for example, that even if EPA may require use of continuous emission reduction controls in preference to intermittent controls or controls that dilute pollutants, EPA is nonetheless limited to imposing requirements that are practicable in light of the state of the art, and of the economy, when the implementation plan is promulgated. Kennecott appears to concede that if EPA can require that continuous emission reduction controls be used so far as practicable, EPA may also impose reasonable conditions to assure the development and adoption of such controls in the future. Nor does Kennecott argue that the research effort required by EPA is an unreasonable one; nor could it, since the nature and scope of the program has yet to be defined. [15] 40 Fed.Reg. 5509 (Feb. 6, 1975) (col. 3). [16] EPA apparently allows full credit for the diluting effect of smokestacks in existence before the first implementation plans were filed in January 1972. For stacks built after that date, but before the decision in Natural Resources Defense Council, Inc. v. EPA, 489 F.2d 390 (5th Cir. 1974), EPA allows credit for the effect of a stack no greater than two and a half times the height of the facility and nearby obstructions. See Air Programs Office, EPA, Region IV, Evaluation of the Georgia Control Strategy, June 21, 1974, revised Mar. 31, 1975, at 2-3; Brief of Respondent at 41-42. Kennecott was given credit for the dispersion effect of a stack two and a half times the height of its McGill smelter. The lawfulness of this allowance is not challenged and is not before us. We also need not consider whether EPA acted lawfully in allowing full credit for the actual dispersal effect of tall stacks built before the initial implementation plans were filed. The considerations that might justify these exceptions to the general requirement that emission reduction techniques be used so far as practicable are not applicable to Kennecott’s new stack at McGill. Kennecott also challenges EPA’s calculation of the percentage reduction in emissions required at McGill because EPA used a “linear rollback method” which EPA itself had labeled inappropriate in some circumstances. As we understand the record, the inadequacy acknowledged by EPA is that the method may not take account of higher levels of pollution that occur at ground points other than those at which monitoring stations are located. This deficiency favors the pollutor. Furthermore, meteorological data for air dispersion modeling is unavailable for McGill. Brief for Respondent at 43^14. The rollback method cannot be characterized as arbitrary or capricious in [1153] light of existing scientific knowledge. Texas v. EPA, 499 F.2d 289, 301 (5th Cir. 1974). See Air Quality & Stationary Source Emission Control, Report by the Commission on Natural Resources, National Academy of Sciences, National Academy of Engineering, Natural Research Council, prepared for Sen. Comm, on Public Works, Sen. Doc. 94 — 4, 94th Cong., 1st Sess., at 242-45 (1975) [hereinafter cited as Air Quality Report ]. [17] EPA also argues that the policy stated in § 101(b)(1) of the Act, 42 U.S.C. § 1857(b)(1), “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,” authorizes EPA to restrict the use of intermittent controls with respect to S02 because of the potentially dangerous effects of aerosol sulfuric acid and sulfates, the end products of atmospheric S02 reactions even at S02 concentration levels below those permitted by national air quality standards. The dangers and uncertainties are outlined in Environmental Protection Agency, Health Consequences of Sulfur Oxides: A Report from CHESS, 1970-1971, at 18 & 20 (May 1974); Air Quality Report, supra note 16, at XVII-XVIII, 532-33. It is unnecessary to consider this possible alternative source of authority. • [18] The nondegradation policy requires that existing air quality be maintained even if degradation would not reduce the quality of the air below national standards. Natural Resources Defense Council, Inc. v. EPA, 489 F.2d 390, 408 (5th Cir. 1974). This court has recognized that the Clean Air Act intends such a policy. Natural Resources Defense Council, Inc. v. EPA, 507 F.2d 905, 913 (9th Cir. 1974). [19] Prior to the Fifth Circuit’s decision in Natural Resources Defense Council, EPA vacillated in its interpretation of the statute “[i]n the face of strong industry • and Administration pressure.” Ayres, Enforcement of Air Pollution Controls on Stationary Sources Under the Clean Air Amendments of 1970, 4 Ecology L.Q. 441, 457-58 (1975). Thereafter EPA consistently urged this construction, as it has in this case. [20] Kennecott’s attempt to distinguish the Fifth Circuit decision on the ground that the Georgia plan relied entirely upon dispersion techniques is unavailing; the Georgia plan, like that of Nevada, included the use of substantial emission reduction measures. 489 F.2d at 409. Kennecott contends that the Fifth Circuit erred in reading the words “emission limitations” in § 1857c-5(a)(2)(B) to exclude emission dispersion and intermittent controls. Any other reading would render the words “other measures” surplusage. The Fifth Circuit’s reading also best serves the congressional purpose to protect and enhance the quality of the nation’s air. A summary of the 1970 amendments presented to the Senate by Senator Muskie distinguishes between emission limitations and “interim control measures.” 116 Cong.Rec. 42384 (1970) (col. 3). Senator Muskie later expressed the view that emission reductions are required by the Act. 119 Cong. Rec. 19183 (1973) & 120 Cong.Rec. S 10409 (daily ed. June 12, 1974). William Ruckelshaus, EPA’s first Administrator, indicated that “emission limitations” meant emission reductions. 1972 Oversight Hearings, supra note 3, at 265, 266. Despite Kennecott’s contrary arguments, we think the Fifth Circuit properly viewed the provisions §§ 1857c-6, 1857c-7, and 1857c-6(d), as supportive of a construction of § 1857c-5(a)(2)(B) reflecting a preference for continuous emission reductions. These provisions do not point unerringly to that result, of course, but the supporting inferences the Fifth-Circuit drew from them are justified. We do not attach the same significance the Fifth Circuit does to the use of the words “emission standard or limitation” in the statement of the authorized subject matter of citizen enforcement actions in § 1857h-2(a)(l). It may be argued (as the Fifth Circuit does) that since citizen enforcement actions are limited to emission limitations, Congress must have intended that such limitations be used whenever possible; but it may also be argued (as Kennecott does) that Congress must have intended the term “emission limitations” to include all types of controls. Both arguments assume that the term has the same meaning in § 1857c-5(a)(2)(B) as it has in § 1857h-2(a)(l). However, if the overall purpose of Congress would be better served by construing the term to include intermittent controls and tall stacks in § 1857h-2(a)(l) and to exclude them in § 1857c-5(a)(2)(B), there is no reason this reading cannot be adopted. Bailey v. United States, 360 F.2d 113, 116 (9th Cir. 1966); Grand Lodge of Int’l Ass’n of Machinists v. King, 335 F.2d 340, 344 (9th Cir. 1964). The use of “other measures” in § 1857c-5(a)(2)(B) and its omission in § 1857h-2(a)(l) supports such a construction. Senator Muskie, who clearly thought § 1857c-5(a)(2)(B) expressed a preference for emission reduction (120 Cong. Rec. S 10409 (daily ed. June 12, 1974)), also thought a private citizen could sue to enforce provisions requiring intermittent controls. Id. Kennecott’s argument that use of a tall stack reduces ground level concentrations, or “dilutes” the pollutant and therefore is not inconsistent with a policy of nondegradation of air quality, ignores the undeniable fact that a tall stack does nothing to reduce emissions because it introduces all of the pollutant into the " atmosphere. See 1972 Oversight Hearings, supra note 3, at 638 (explanation of tall stacks by Mr. Crawford). Kennecott admits that a “Summary of the Provisions of Conference Agreement on the Clean Air Amendments of 1970” (see 116 Cong.Rec. 42384 (Dec. 18, 1970)) submitted to the Senate by Senator Muskie “seem[s] to reflect a view that all sources should be subject to emission limitations.” Petitioner’s Brief at 39. This conceded, it is not circular (as Kennecott suggests) to argue that § 1857c-5(a)(2)(B) must be interpreted to prefer emission limitations in order to prevent evasion of the intent of Congress, since state plans might otherwise rely wholly on less expensive intermittent controls and dispersion techniques to achieve air quality standards. [21] The Fifth Circuit stated in dicta that intermittent control systems were dispersion techniques subject to interdictions similar to those imposed on tall stacks. 489 F.2d at 394 n. 2. [22] Kennecott argues EPA cannot require Kennecott to develop and adopt improved methods of emission reduction in view of the Supreme Court’s statement in Train that “so long as the national standards are being attained and maintained, there is no basis in the present Clean Air Act for forcing further technological developments.” 421 U.S. at 91, 95 S.Ct. at 1487. The Supreme Court was referring to modification of source-by-source emission limitations fixed by the state. EPA must approve such modifications so long as the state plan as a whole provides for a mix of emission limitations from all sources sufficient to meet national air quality standards. The Train holding is that EPA may not raise the emission limitation fixed by the state for a particular source to force technological improvement as to that source so long as the emission limitations set by the state for all sources are sufficient to meet national air quality standards. The Supreme Court was not addressing the question presented in this case; that is, whether EPA may seek to force development of technology to the extent necessary to enable national standards to be met by emission limitation controls, rather than by other measures. [23] See 37 Fed.Reg. 15095 (July 27, 1972); 38 Fed.Reg. 25698 (Sept. 14, 1973). See also 120 Cong.Rec.S 10409 (daily ed. June 12, 1974) & 119 Cong.Rec. 19183 (1973) (col. 2) (remarks by Senator Muskie). [24] As an EPA staff paper stated (119 Cong.Rec. 19190 (1973) (col. 1) (emphasis original)): Depending on the circumstances, [an intermittent control system] may or may not reduce the average long-term emissions. If plant operation is curtailed during poor dispersion conditions, then it may be increased during good conditions to make up for the lost production. Average emissions would be about the same with or without [an intermittent control system] for this situation. If clean fuel is used to reduce emissions during poor dispersion conditions, then average emissions will be reduced somewhat. If fuel with higher sulfur content is used during good conditions, then average emissions could be greater with [an intermittent control system]. It must be concluded, therefore that although [an intermittent control system] employs temporary emission limitation, the long-range control method is that of taking advantage of good dispersion rather than emission reduction. [25] 10 Wkly. Comp, of Pres.Doc. 289 (1974) (Veto of S. 2589). [26] Murphy Amendments to H.R. 11450, § 201 (particularly proposed § 119(a)(2)(A)(iii) & (B)(i) of the Clean Air Act): Compare 119 Cong.Rec. 41776 (1973) (col. 1) with id. 41163 (1973) (col. 2-3). Prior to adoption of the Murphy Amendments the proposed statute provided that the Administrator of EPA could temporarily suspend any stationary source fuel or emission limitation “if the Administrator finds-— . (iii) that such person has been placed on a schedule which provides for the use of methods which the Administrator determines will assure continuing compliance with the stationary source fuel or emission limitation as soon as practicable (but no later than June 30, 1979), which schedule shall include increments of progress toward compliance with such limitation by such date. (B)(i) Any schedule under subparagraph (A)(iii) shall include a date by which a contractual obligation shall be entered into for an emission reduction system. . 119 Cong.Rec. 41160 (1973) (col. 3) (emphasis added). Congressman Murphy’s amendment modified this provision to read as follows: (iii) that such person has been placed on a schedule which provides for the use of methods which the Administrator determines will assure continuing compliance with a national primary ambient air quality standard as soon as practicable (but no later than June 30, 1979), which schedule shall include increments of progress toward compliance with such standard by such date. (B)(i) Any schedule under subparagraph (A)(iii) shall include, if necessary to meet a national primary air quality standard, a date by which a contractual obligation shall be entered into for an emission reduction system. . [1157] 119 Cong.Rec. 41776 (1973) (col. 1) (emphasis added). Congressman Murphy declared that “[n]othing in section 110 of the Clean Air Act mandates the use of any particular method of achieving air quality, such as scrubbers or low sulfur fuel.” Id. at 41163 (1973) (col. 3). These views were shared by Congressman Jones of Alabama. Id. at 41146 and Congressman Nelsen, id. at 41164 (col. 2). But see note 34 infra. In passing the House, the Murphy Amendment survived a counter amendment by Congressman Symington who recognized the implications of Congressman Murphy’s proposal. 119 Cong.Rec. 41714 (1973). [27] The report of the Conference Committee states (Conf.Rep., S.Rep.No.93-663, 93d Cong., 1st Sess. 83-84 (1973)): The House-passed bill would have permitted the use of so-called intermittent or alternative control strategies as a means of meeting ambient air quality standards if such strategies were determined by the Administrator to be reliable and enforceable. This permission would have applied to both existing sources not affected directly by the energy emergency and sources required to convert to coal under the emergency legislation. The Senate bill would have permitted revision of existing implementation plans to require use of continuous emission reduction systems on any fuel-burning stationary sources affected by shortages of fuels, suspensions or conversions. The conference agreement does not include either of the foregoing broad provisions. Instead, the conferees decided to limit the application of this provision to those sources which convert to combustion of coal as a result of the energy emergency. The conference substitute requires these converting sources to come into compliance with all plan requirements by 1979 (or 1980, if a postponement is obtained under ' section 110(f)) in accordance with a schedule which meets requirements of regulations of EPA. These requirements would require incremental steps toward compliance by utilization of low sulfur coal or coal by-products, or by continuous emission reduction systems to permit the combustion of high sulfur coal (or coal with high ash content) in compliance with such plan requirements. [28] Letter dated Mar. 22, 1974, from Russell E. Train, Administrator of EPA, to Hon. Carl T. Albert, Speaker of the House of Representatives, attached to H.R.Rep.No.93-1013, 2 U.S. Code Cong. & Admin.News, 93d Cong., 2d Sess., at 3298 (1974). The Administration proposal appears at 4 BNA Env.Rptr. 2004-09 (1974) (current developments). [29] This proposal read: A BILL Be it enacted by the Senate and House of Representatives of the United States of America in Congress assembled, SIGNIFICANT DETERIORATION Section 101(b)(1) of the Clean Air Act is amended to read as follows: “(1) to protect and enhance the quality of the Nation’s air resources by establishing, achieving, and maintaining national ambient air quality standards, standards of performance for new stationary sources, and national emission standards for hazardous air pollutants so as to promote the public health and welfare and the productive capacity of the Nation, but nothing in this Act is intended to require or authorize the establishment by the Administrator of standards more stringent than primary and secondary ambient air quality standards;”. 4 BNA Env.Rptr. 2010 (1974) (current developments). [30] The proposal read: A BILL Be it enacted by the Senate and House of Representatives of the United States of America in Congress assembled, INTERMITTENT OR ALTERNATIVE CONTROL MEASURES Section 110 of the Clean Air Act is amended by adding subsection (h) which reads as follows: “(h) Nothing in this section shall be construed to preclude use of alternative or intermittent control measures which the Administrator determines are reliable and enforceable and which he determines will permit attainment and maintenance of the national ambient air quality standards.” 4 BNA Env.Rptr. 2010 (1974) (current developments). [31] 4 BNA Env.Rptr. 1927 (1974) (current developments). [32] In addition to the bill set out in note 30, Congress had before it a proposal offered by Congressman Nelsen (H.R. 13464) to amend 42 U.S.C. § 1857c-5(a)(2)(B) to provide that intermittent controls were acceptable as permanent alternatives to emission limitations. 120 Cong. Rec. H 1780, E 1376-78 (daily ed. Mar. 13, 1974). The bill died in committee. [33] 120 Cong.Rec. S 10409 (daily ed. June 12, 1974) (1st col.) (emphasis added). [34] 120 Cong.Rec. S 10409 (daily ed. June 12, 1974) (3d col.) (emphasis added). See also 120 Cong.Rec. S 8014 (daily ed. May 14, 1974). The statement of Senator Baker, ranking minority member of the subcommittee, in support of the Conference Report, reflects the same understanding that emission limitations must be met by continuous reduction systems (120 Cong.Rec. S 10426 (daily ed. June 12, 1974) (1st col.)): I am concerned about the conference report provision that powerplants unable to obtain sufficient low sulfur coal or coal alternatives to meet emission limitations applicable under the law must undertake to obtain continuous emission reduction systems which are capable of meeting these limitations by 1979 while burning high sulfur coal. Although the term “continuous emission reduction system” is broad enough to encompass a broad range of technology, I foresee the possibility that certain specific solutions to the problem of sulfur oxide emissions might receive undue emphasis. For this reason, I want to emphasize that the term is meant to indicate any technology involving advanced techniques of combustion of coal — such as the fluidized-bed process — or after-treatment of combustion gases — for example flue gas desulfurization, better known as scrubber technology. Senator Baker’s statement echoes the view of John Sawhill, a subsequent Administrator of the Federal Energy Office, stated in a May 20, 1974, letter to Chairman Jennings Randolph of the Senate Public Works Committee: Specifically, we are concerned with the provisions of section 119(b)(2)(B) that require that plants scheduled to convert must be committed to a compliance schedule that provides a date by which the source must enter into contracts for low sulfur coal or scrubbers. This provision is coupled with section 119(b)(2)(C) that requires plants granted suspensions to come into compliance with emission regulations in a state implementation plan that are in effect on the date of enactment of these amendments. The requirement concerning contracts for low sulfur fuel or scrubbers would effectively preclude the use of intermittent control systems as an alternative method for achieving compliance. If the Administration’s proposal to permit use. of intermittent control systems, contained in our March 22 amendments to the Clean Air Act, is adopted, this section of H.R. 14368 would have to be amended to conform with it. 120 Cong.Rec. S 10414 (daily ed. June 12, 1974). (Both the Federal Energy Office and EPA claimed to represent the Administration, despite their inconsistent position on whether to permit permanent use of intermittent controls. The Energy Office’s proposal was, however, only an alternative.) Congressman Nelsen had earlier supported the Murphy Amendments, see note 26, and had introduced legislation to amend 42 U.S.C. § 1857c-5(a)(2)(B) to permit intermittent controls, see note 32. He noted that the “total impact of [new section 119] and the Clean Air Act provisions is to lock in the technology of scrubber systems . . ..” 120 Cong.Rec. H 5002 (daily ed. June 11, 1974). Congressman Nelsen nonetheless voted for the bill because Congress was committed to reexamination of his concerns regarding intermittent controls. Id. See also 120 Cong.Rec. S 8018 (daily ed. May 14, 1974) (comments by Senator Randolph) (“[Section 119] would clarify EPA’s authority to impose interim requirements that both protect air quality and insure appropriate efforts are taken to secure continuous emission control systems or conforming fuels”). Senator Randolph mentioned the Fifth Circuit’s Natural Resources Defense Council decision, and was critical only of that portion relating to the granting of variances. [35] Kennecott’s entire response to these materials is a footnote suggestion that the 1974 Act and its legislative history “must be construed in the context of the oil shortage it was intended to deal with, which gave rise to the necessity of permitting the EPA to impose [intermittent control] strategies upon states against their wishes and in violation of their state emission limitations despite the Clean Air Act’s command that states be permitted to opt for stricter emission limitations than would be required to meet national ambient air quality standards” (emphasis in original). Reply Brief of Petitioner Kennecott at 14 n. 1. Giving this circumstance full weight, it does not diminish the significance of the compelling showing that Congress interpreted 42 U.S.C. § 1857c-5(a)(2)(B) as EPA urges us to interpret it. Accord, Air Quality Report, supra note 16, at 213.