TAMM, Circuit Judge:
These two appeals, consolidated for purposes of argument and decision, are taken from the action of the Administrator of the Environmental Protection Agency [EPA] in setting "standards of performance" for new or modified stationary sources of pollution pursuant to the mandate of § 111 of the Clean Air Act, as amended [Act], 42 U.S.C. § 1857c-6 (1970). Among the stationary sources for which standards were set are sulfuric acid plants, subject to challenge in No. 72-1072, and coal-fired steam generators, subject to challenge in No. 72-1079. See 40 C.F.R. §§ 60.1, et seq. Even when limited to the scope of review prescribed by the Supreme Court in Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402, 91 S.Ct. 814, 28 L.Ed.2d 136 (1971), we find that the Administrator's action as to certain aspects of the standards must be remanded for further proceedings; as to the bulk of the standards, however, we find that the Administrator has acted properly within the scope of his authority and not in abuse of his discretion.
On March 31, 1971, pursuant to the requirement of § 111 of the Act, 42 U.S.C. § 1857c-6 (1970),1 the EPA published a list of categories of stationary air pollution sources which significantly contribute to the endangerment of public health and welfare. The list included steam generators, incinerators, sulfuric acid plants, nitric acid plants, and portland cement plants. 36 Fed.Reg. 5931 (March 31, 1971). Thereafter, on August 17, 1971, proposed regulations were published establishing federal standards of performance for each new source category on the list. 36 Fed.Reg. 15704 (August 17, 1971). Issued concurrent with the proposed regulations were documents entitled "Background Information for Proposed New-Source Performance Standards" and "Summaries of Test Data," which set forth the justification for the new proposed standards. In December of 1971, after receiving and evaluating more than 200 comments from interested parties,2 the EPA published final regulations, 40 C.F.R. part 60, incorporating only minor changes from the proposed regulations. 36 Fed.Reg. 24876 (December 23, 1971).
On January 21 and 24, 1972, Essex Chemical Corp., et al. (No. 72-1072), Portland Cement Association (No. 72-1073), and Appalachian Power Co., et al. (No. 72-1079), petitioned for review in this court pursuant to § 307(b)(1) of the Act, 42 U.S.C. § 1857h-5(b)(1) (1970),3 challenging the standards set by the EPA for sulfuric acid plants, portland cement plants, and coal-fired steam generators, respectively. Shortly thereafter this court issued its decision in Kennecott Copper Corp. v. EPA, 149 U.S.App.D.C. 231, 462 F.2d 846 (1972), concerning a national secondary ambient air quality standard promulgated by the EPA pursuant to § 109(b) of the Act, 42 U.S.C. § 1857c-4(b) (1970), wherein a remand was ordered so that the EPA might "supply an implementing statement that will enlighten the court as to the basis on which [the Administrator] reached the . . . standard . . . ." Id. at 850. In light of the Kennecott Copper decision the EPA subsequently published a "Supplemental Statement in Connection with Final Promulgation," 37 Fed.Reg. 5767 (March 21, 1972), in order to avoid the problems presented in Kennecott Copper and thus "ensur[e] the rapid conclusion of judicial review of the validity of the [stationary source] standards." Id.4
On June 29, 1973, another panel of this court issued an opinion remanding the record to the EPA for further proceedings regarding the standards of performance for portland cement plants, Portland Cement Association v. Ruckelshaus, 158 U.S.App.D.C. ___, 486 F.2d 375 (1973). While the records in the two cases sub judice are substantially different from that in Portland Cement and consequently engender differing conclusions as to the legality of the standards, several issues are so similar and so conclusively dealt with by the Portland Cement decision that they can be treated summarily by this court in its determinations today.
1. NEPA Impact Statement
One issue raised and extensively briefed in both No. 72-1072 and No. 72-1079 is the consequence of the Administrator's failure to file an "impact statement" pursuant to § 102(2)(c) of the National Environmental Policy Act of 1969 [NEPA], 42 U.S.C. § 4332(2)(c) (1970).5 Petitioners here allege (as was alleged in Portland Cement) that the EPA, merely because it is an environmentally oriented agency, is not exempt from the NEPA provision that in "major Federal actions significantly affecting the quality of the human environment," all Federal agencies are subject to the requirement that the responsible official file "a detailed statement . . . on . . . the environmental impact of the proposed action." Petitioners pointedly note that an impact statement, if filed, would have required the EPA to consider factors such as alternative adverse environmental effects and cost/benefit analyses to a considerably more significant degree than that which the record shows actually occurred.
The Portland Cement panel, per Judge Leventhal, extensively analyzed the considerations pertaining to a general, limited, or non-existent exemption for the EPA from the NEPA impact statement requirements. See Portland Cement, supra, 486 F.2d 379-387. The panel concluded, leaving for another time the consideration of whether the EPA was completely exempt from the rigors of an impact statement,6 that the EPA was exempt at least in its action in promulgating the § 111 standards. The determination was founded upon a "proper" construction of the § 111 mandate that the "standard of performance" reflect "the best system of emission reduction" and require the Administrator to take into account "the cost of achieving such reduction." As "[t]hese criteria require the Administrator to take into account counter-productive environmental effects of a proposed standard,7 as well as economic costs to the industry," the decision that an impact statement is not required in this specific instance strikes a "workable balance between some of the advantages and disadvantages of full application of NEPA." Id. at 385, 386. The court stated:
What is decisive, ultimately, is the reality that, section 111 of the Clean Air Act, properly construed, requires the functional equivalent of a NEPA impact statement.
Id. at 16. Finding no good reason to divert from or expand upon the logic of the Portland Cement decision, we adhere to the position that no NEPA impact statement need be filed by the Administrator in making his § 111 determinations.8
2. Opacity Standard
The standards promulgated by the Administrator restrict "acid mist" discharge to no greater than "10 percent opacity" in sulfuric acid plants and "particulate matter" discharge to no greater than "20 percent opacity" in coal-fired steam generators.9 See 40 C.F.R. §§ 60.83(b), 60.42(b). Opacity is defined as the "degree to which emissions reduce the transmission of light and obscure the view of an object in the background." 40 C.F.R. § 60.2(j). A 10 percent opacity standard was established for the portland cement stationary sources of pollution and the Portland Cement court, impressed both by the arbitrariness of the test—it is essentially a subjective test on the part of trained inspectors who attempt to judge the per cent of opacity present in a smoke plume—and the persuasive challenges on the part of the petitioners,10 found that further consideration and explanation of the efficiency and objectivity of the test was necessary prior to its adoption as a standard:
[I]t is one thing to use a method of testing to observe possible violations of a standard; it is another to constitute that method as the standard itself. If the opacity test is to be a standard, and if violations can result in enforcement actions without further testing, the standard must be consistent with the statute and congressional intent.
Portland Cement, supra, 486 F.2d 401. Similar opacity standards are challenged in both suits before us today and the principles requiring further consideration are unchanged. The record must be remanded for additional consideration and explanation by the Administrator regarding the reasonableness of the opacity standards.11
3. Startup, Shutdown, Equipment Malfunction
Petitioners in both No. 72-1072 and No. 72-1079 raise objections in varying degrees regarding the EPA's failure to provide that lesser standards, or no standards at all, should apply when the stationary source is experiencing startup, shutdown, or mechanical malfunctions through no fault of the manufacturer.12 Petitioners allege that nowhere in the record has there been a showing that the standards can be equalled during periods of less than normal operation, when plant efficiency and consequently pollution emissions vary significantly from the norm. On August 18, 1972, EPA Acting Administrator Fri admitted that "[a]t the time of promulgation, it was known that even in the case of sources which ordinarily complied with the standards and which followed proper operating and maintenance procedures, there could on occasion occur malfunctions or other events (process startups and shutdowns) during which emissions might temporarily exceed the standards."13 37 Fed.Reg. 17214 (August 25, 1972). Concurrent with that statement regulations were proposed providing a framework within which unavoidable excesses of the standard would normally not be considered violations of the regulatory guidelines (so long as certain reporting requirements were fulfilled). The identical issue was raised in Portland Cement and the court there found the challenge persuasive enough to merit a remand. The significance of the proposed regulation was fully recognized:
If the EPA adopts, or intends to adopt, this proposed regulation, it may take the attendant flexibility into account, on remand, as pertinent to the manufacturers' objections, even though the new regulation has been proposed in a proceeding with a different docket number and caption.
Portland Cement, 486 F.2d 399. We agree that such variant provisions appear necessary to preserve the reasonableness of the standards as a whole and that the record does not support the "never to be exceeded" standard currently in force. Thus, we remand for further consideration of this issue, noting that the proposed regulation should play an integral role in any reconsideration.
The focal point of our analysis of the EPA's standard setting must be the statutory statement defining "standard of performance." Section 111 of the Act, 42 U.S.C. § 1857c-6(a)(1) (1970) states:
The term "standard of performance" means a standard for emissions of air pollutants which reflects the degree of emission limitation achievable through the application of the best system of emission reduction which (taking into account the cost of achieving such reduction) the Administrator determines has been adequately demonstrated.
It is the system which must be adequately demonstrated and the standard which must be achievable. This does not require that a sulfuric acid plant be currently in operation which can at all times and under all circumstances meet the standards; nor, however, does it allow the EPA to set the standards solely on the basis of its subjective understanding of the problem or "crystal ball inquiry." See Portland Cement, supra, 486 F.2d 391. An adequately demonstrated system is one which has been shown to be reasonably reliable, reasonably efficient, and which can reasonably be expected to serve the interests of pollution control without becoming exorbitantly costly in an economic or environmental way. An achievable standard is one which is within the realm of the adequately demonstrated system's efficiency and which, while not at a level that is purely theoretical or experimental, need not necessarily be routinely achieved within the industry prior to its adoption.14
In subjecting the Administrator's actions to judicial review we apply a test of reasonableness, wherein we are "not empowered to substitute [our] judgment for that of the agency" but must consider whether "the decision was based on a consideration of the relevant factors and whether there has been a clear error of judgment." Citizens to Preserve Overton Park v. Volpe, supra, 401 U.S. at 416, 91 S.Ct. at 824. The judgment of the Administrator is to be weighted against his statutory function and limitations, the record searched to determine if indeed his decisions and reasons therefore are themselves reasoned, and at that point our function terminates. Our "expertise" is not in setting standards for emission control but in determining if the standards as set are the result of reasoned decisionmaking. Yet, even this limited function requires that we foray into the technical world to the extent necessary to ascertain if the Administrator's decision is reasoned. While we must bow to the acknowledged expertise of the Administrator in matters technical we should not automatically succumb thereto, overwhelmed as it were by the utter "scientificity" of the expedition.
A vast majority (97 per cent) of sulfuric acid plants in the United States produce sulfuric acid by means of a "contact" process, involving these three steps:
(1) Elemental sulfur or other sulfur-bearing raw material is burned to obtain sulfur dioxide gas [SO(2)];
(2) Sulfur dioxide gas is converted into sulfur trioxide [SO(3)], upon contact with a catalyst, generally vanadium pentoxide;
(3) Sulfur trioxide gas is then absorbed in sulfuric acid [H(2)SO(4)], through combination with the water in a 98 to 99 percent sulfuric acid solution with which it is circulated in an absorption tower.
Any unconverted sulfur dioxide gas passes through demister equipment (to eliminate acid mist, basically sulfuric acid mist formed by the combination of sulfur trioxide and water vapor), and escapes out the stack into the atmosphere. The regulations permit sulfur dioxide emissions up to a level of 4.0 lbs. per ton of sulfuric acid produced, and acid mist emissions up to 0.15 lb. per ton of sulfuric acid produced, as measured by methods approved in the regulations. See 40 C.F.R. §§ 60.82, 60.83. Generally those plants burning an elemental sulfur feedstock—about 68 per cent of all sulfuric acid plants in the United States—are subject to greater quality control, higher efficiency, and consequently experience less emission problems than those burning only sulfur-bearing raw materials.15 The latter, known variously as "recycle," "spent acid," and "wet gas" plants, serve an important environmental function, however, for among the raw materials used for feedstocks are wastes from petroleum refineries and other chemical processes.16
There are two types of sulfuric acid plants in general use today, "single" and "dual" absorption. The terms "single" and "dual" refer to whether the plant uses one or two stages in steps #2 and #3 of the contact process. Dual absorption plants pass the sulfur dioxide gas which remains after a first conversion attempt through a second catalytic converter and absorption tower, thus generating a higher conversion efficiency and minimizing the ultimate gas emissions into the atmosphere. The dual absorption system is relatively new, and although it has been used extensively in Europe it has only recently emerged in the United States as a viable method of sulfuric acid production. At the time of EPA testing in preparation for the standards only one dual absorption plant was in operation in the United States, an elemental sulfur burning plant owned by the American Cyanamid Company.17
The EPA standards are applied to all types of contact process sulfuric acid plants, whether single or dual absorption and whether designed for use with elemental or recycle feedstock.18 The standards are based on information and data derived from (1) inspections and stack tests of existing facilities; (2) consultations with operators, designers, and state and local control officials; and (3) review of available literature on the subject. The EPA must rely upon the dual-absorption system as the adequately demonstrated system which will meet the standards for new elemental sulfur burning plants, and on a sodium sulfite-bisulfite scrubbing system as the adequately demonstrated system which will meet the standards for the new recycle and modified existing plants which become subject to the standards.19 The acid mist standard is based on a system of "high-efficiency acid-mist eliminators, including both fiber demisters and wire and tube electrostatic precipitators."20
1. Elemental Sulfur Feedstock—Sulfur Dioxide Standards
Petitioners do not object to the EPA's selection of dual absorption as the "best system of emission reduction" which has been "adequately demonstrated" for use in elemental sulfur burning plants. They state in their Reply Brief:21
Petitioners . . . object to EPA's implications that industry is unwilling to utilize the best control technologies. Industry is agreeable to using dual absorption technology, which EPA finds best for the 80% of the industry that uses elemental sulfur feedstocks.
It is the designation of 4.0 lbs. of sulfur dioxide per ton of sulfuric acid produced as the "limitation achievable" through use of the dual absorption system that strikes the discordant note. Petitioners assert that the standard is unrealistically high, not based on the evidence in the record, and hence arbitrary and unreasonable. The bulk of the comments received by the EPA from interested parties—generally industrial companies subject to the controls—suggest a standard of 6.5 lbs./ton as a more reasonable alternative.
The EPA bases the 4.0 lbs./ton standard on three separate sources of information: (1) EPA supervised tests at the Warners Plant of the American Cyanamid Co. on February 17 and 18, 1971; (2) Available technical literature providing information on several of the European dual absorption plants; (3) American Cyanamid's own supervised tests at the Warners Plant.22
The results of the EPA supervised tests are impressive. Three independent test runs measured sulfur dioxide emissions of less than 2.0 lbs./ton. Yet for purposes of our review the relevancy of the test results is at best minimal, for as petitioners correctly point out the test data clearly indicate that during the tests acid production in the plant was only at approximately 52% of capacity. Since the regulations specifically provide that performance tests "shall be conducted while the affected facility is operating at or above the maximum acid production rate at which such facility will be operated," 40 C.F.R. § 60.85(b), and as nothing appears in the record to indicate any direct proportional relationship between emissions and plant production levels,23 we are hard pressed to find much support for the EPA standards on the basis of these test results. A similar vice affects the relevancy of the literature regarding the European plants, for while purporting to show impressive emission controls—three of the plants reportedly average emissions of less than 3.0 lbs./ton—no data are provided indicating at what percentage capacity the various plants were operating when these results were obtained.24
Filed with the comments of the American Cyanamid Co. were summaries of many of the company's own tests of its Warners Plant,25 including the test data which prompted the EPA to state in the Background Information for Proposed New-Source Performance Standards [Background Document], J.A. at 57:
Testing by the operator showed full-load SO(2) emissions to be consistently less than 4.0 pounds of SO(2) per ton of acid, with emissions below 3.0 pounds per ton much of the time.
While the test results hardly support the EPA's allegation of full-load emissions "consistently less" than 4.0 lbs./ton, they do suffice to convince us of the reasoned decision-making of the Administrator. The Warners Plant is designed for a conversion efficiency (percentage of SO(2) conversion to SO(3)) of 99.5, which amounts to a design expectancy of sulfur dioxide emissions in the 6.5 lbs./ton range.26 The test results, however, both as reported by the EPA and as clarified and expanded in American Cyanamid's data, show that on three occasions when production was at or near full capacity emission levels of only 2.7, 3.5, and 4.0 lbs./ton were recorded. In fact, the average of the nineteen readings taken when the plant was near full capacity is approximately 4.6 lbs./ton. In sum, the proposed standard was exceeded on two occasions, equalled on another, and nearly equalled on the average of nineteen different readings. Keeping in mind Congress' intent that new plants be controlled to the "maximum practicable degree,"27 we find that the 4.0 lbs./ton standard based on a dual absorption system for new elemental sulfur burning plants is the result of the exercise of reasoned discretion by the Administrator and cannot be upset by this court.
The petitioners additionally challenge what they feel to be an inadequate consideration of economic costs in the selection of the 4.0 lbs./ton standard. Essentially what they seek is a cost-benefit analysis comparing their desired 6.5 lbs./ton standard with the standard ultimately adopted.28 Again, we affirmatively defer to the Portland Cement decision, supra, 486 F.2d 387:
However desirable in the abstract, such a requirement would conflict with the specific time constraints imposed on the Administrator. The difficulty, if not impossibility, of quantifying the benefit to ambient air conditions, further militates against the imposition of such an imperative on the agency. Such studies should be considered by the Administrator, if adduced in comments, but we do not inject them as a necessary condition of action. (Footnote omitted.)
The record amply supports the EPA's contention that it has taken economic costs into account in selecting the best system, whether that system be dual absorption, the sodium sulfite-bisulfite scrubber, or acid mist controls.29 It is not unlikely that industry and the EPA will disagree on the economic costs of these various control techniques. We have no desire or special ability to settle such a dispute, but we do find that the Administrator has considered economic costs as required by § 111 of the Act, 42 U.S.C. § 1857c-6(a)(1) (1970), and that his consideration is reasoned.30
2. Recycle Plants—Sulfur Dioxide Standard
In contrast to the petitioners' position concerning elemental sulfur feedstock plants and the dual absorption system, their challenge regarding recycle plants is directed more to the sodium sulfite-bisulfite scrubber as the "best system . . . adequately demonstrated" than to whether the 4.0 lbs./ton standard is "achievable" thereunder.
The standard as applied to recycle plants must be analyzed through the utilization of a sodium sulfite-bisulfite scrubbing system, for there exists no data in the record supporting the proposition that any other system, dual absorption or otherwise, has been adequately demonstrated for use in such plants.31 A scrubbing system is separate from the acid producing process and differs from dual absorption in that it achieves emission reduction not by directly recycling unconverted sulfur dioxide gas back into the system, but by extracting the sulfur dioxide gas out of the exit gas stream through exposure to a sodium sulfite solution. This produces sodium bisulfite which upon heating yields sulfur dioxide gas (then fed back into the acid producing system) and sodium sulfite (then returned to combine again with sulfur dioxide in the exit gas stream). In order to retain high efficiency levels in the scrubbing system it is necessary to continuously add new solution and to drain off part of the old solution as "purge."32 The purge is collected in significant amounts, up to 52 lbs./ton of acid produced, and thus presents noteworthy disposal problems.33
Despite petitioners' protestations concerning the "ideal" conditions under which the EPA conducted its tests, we find adequate support for the conclusion that the standard is achievable in the results of tests conducted at the Olin Chemicals Co. plant in Paulsboro, New Jersey. The tests were conducted under EPA supervision on March 26 and 27, 1971, when the Paulsboro plant was the only active recycle plant in the United States equipped with a sodium sulfite-bisulfite scrubber.34 The results are impressive, reflecting sulfur dioxide emission levels of 2.59 and 2.85 lbs./ton, well below the standard. Equally important, however, is the fact that the tests were conducted when the plant was operating at or near full capacity.
As we have earlier noted, § 111 of the Act, 42 U.S.C. § 1857c-6(a)(1) (1970), requires that when the Administrator makes his "best system" and "achievable limitation" determination that he "take into account counter-productive environmental effects."35 We cannot conclude on the basis of the record before us that the Administrator met this obligation in promulgating a standard to be applied to recycle acid plants. The record evidence indicates that the standard is achievable only through use of a sodium sulfite-bisulfite scrubber, yet no consideration of the significant land or water pollution potential resulting from disposal of the 52 lbs./ton liquid purge byproduct is apparent. A 700 ton/day sulfuric acid plant operating at capacity will produce nearly twenty tons of the purge waste in only one day of operation, and while the scrubber will effectively cut air emissions from a 50 to 100 lbs./ton level its "counter-productive environmental effects" cannot be ignored.36 Judge Leventhal's statement in Portland Cement, supra, 486 F.2d 386 that "we cannot imagine that Congress intended that `best' could apply to a system which did more damage to water than it prevented to air," seems particularly apropos.
We realize that the only apparent alternative to the scrubbing system is no control at all.37 Since recycle plants amount to only about twenty per cent of the country's sulfuric acid plants, and since collectively the sulfuric acid plants in this country account for only about two per cent of the sulfur dioxide emission, such an alternative (at least in selected instances) might prove best for the environment as a whole.38 We have no way of knowing; it is something the EPA is uniquely suited to determine. We do know, however, that the feeble statements in the Background Document that "[m]ethods for disposing of these products will have to be considered by plant operators," and "process designers are investigating several means of handling these wastes," are poor substitutes for the reasoned consideration of this problem that is required.39 The record is thus remanded to the Administrator for further consideration and explanation of the adverse environmental effects of requiring a 4.0 lbs./ton standard for recycle acid plants.40
3. Acid Mist Emissions
Our review of the record leads us to conclude that ample evidence exists, in terms of both technical literature and results of EPA conducted tests, to satisfy us of the reasoned nature of the EPA's acid mist standard.41 The standard, 0.15 lb. of sulfuric acid mist per ton of acid produced, is deemed achievable through the use of either electrostatic or fiber precipitator systems, both of which are certainly adequately demonstrated.42 Petitioners' objections regarding the result oriented definition of acid mist (i. e., whatever the prescribed method for testing detects, as opposed to the "accepted definition" of "liquid particulate matter exclusive of gas") is considered, but we cannot say that the Administrator was arbitrary in promulgating such a definition.43
Reviewing the coal-fired steam generator standards promulgated for particular matter, sulfur dioxide, and nitrogen oxides, see 40 C.F.R. § 60.40 et seq., and subjecting them to the various tests and scopes of inquiry discussed above, we find that with but one exception the record evidence supports the conclusion that they are the result of reasoned decision-making.44 The evidence, including tests of prototype and full-scale control systems, considerations of available fuel supplies, literature sources, and documentation of manufacturer guarantees and expectations, convinces us that the systems proposed are adequately demonstrated, that cost has been taken into consideration, and that the emission standards are achievable.45 See 42 U.S.C. § 1857c-6(a)(1) (1970). Although in the instance of the sulfur dioxide standards and the lime slurry scrubbing system the EPA tests give results indicating that a presently installed unit approaches rather than achieves the 1.2 lbs./million B.t.u. heat input standard,46 when the results are considered in conjunction with the prototype testing data and the predictions and guarantees of domestic equipment manufacturers for plants under construction,47 we cannot say that the standard represents "a clear error of judgment." See Citizens to Preserve Overton Park v. Volpe, supra, 401 U.S. at 416, 91 S.Ct. 814.
A lime slurry scrubbing system is the anticipated "best system" for use in coal-fired steam generators which burn coal of such a high sulfur content that separate control mechanisms are necessary to meet the standard.48 This scrubbing system, like the sodium sulfite-bisulfite system used in sulfuric acid plants, produces significant quantities of sludge byproduct which present substantial disposal problems. The counter productive environmental effects of the system were noted in comments to the EPA,49 but the Administrator's consideration of the issue, at least as it appears on the record, is insufficient. The only statement in the Background Information for Proposed New-Source Performance Standards concerning such is merely an admission of the problem:
Lime-scrubbing systems are essentially throwaway processes that produce significant quantities of solid waste. For a 3.0-percent-sulfur coal, the additional wastes are roughly equal to the ash generated from burning coal.50
Consequently, the record is remanded for further consideration and explanation by the Administrator of the adverse environmental effects of requiring a 1.2 lbs./million standard for those coal-fired steam generator plants which must use a lime slurry scrubbing system as the only means of achieving the standard.51
Accordingly, the record in No. 72-1072 and No. 72-1079 is remanded to the Administrator of the Environmental Protection Agency for further proceedings not inconsistent with this opinion.