These challenges to the several EPA effluent limitation regulations were combined for this consideration. The cases were held pending the development of supplementary regulations. There follows a consideration of each challenge in a separate section. We have included some record references for the purpose of clarity and further explanation.
The many points and issues raised in each case have been considered; however, not all have been discussed in this opinion. Thus only the significant or determinative issues have been written on.
Appellate review of the regulations in these several appeals requires a substantial inquiry and probing of the administrative agency’s action in accordance with Citi*1236zens To Preserve Overton Park v. Volpe, 401 U.S. 402, 91 S.Ct. 814, 28 L.Ed.2d 136. In reviewing EPA effluent limitations, we must examine whether the facts underlying EPA action are adequately developed and disclosed. American Petroleum Institute v. EPA, 540 F.2d 1023 (10th Cir.). Otherwise stated, three questions should be addressed (assuming the statute and requisite procedures are satisfied): first, whether the EPA explained the facts and policy concerns relied on in making its decision; second, whether these facts have some basis in the record; and third, whether these facts and policy considerations could lead a reasonable person to the same judgment the Agency reached. Weyerhaeuser Co. v. Costle, 191 U.S.App.D.C. 309, 590 F.2d 1011 (D.C. Cir.).
The court’s function includes neither reweighing the available evidence nor substituting its judgment for the Agency’s. BASF Wyandotte Corp. v. Costle, 598 F.2d 637 (1st Cir.). When available technological data and research are unfamiliar or untried, the Agency necessarily enjoys broad discretion. BASF, 598 F.2d at 650. And as we have previously observed in construing the Act, “the guiding star is the intent of Congress to improve and preserve the quality of the Nation’s waters. All issues must be viewed in the light of that intent.” American Petroleum Institute, 540 F.2d at 1028.
It is fundamental that an agency explain the facts and policy concerns underlying its decisions and conclusions. Courts “are no longer content with bare administrative ipse dixits based on supposed administrative expertise.” Appalachian Power Co. v. EPA, 477 F.2d 495 (4th Cir.). Such explanations must appear in the record and may not be supplied in the form of after-the-fact rationalizations. National Crushed Stone Ass’n v. EPA, 601 F.2d 111, 118 (4th Cir.). At the same time agencies need not supply comprehensive explanations and record citations for each and every conclusion. Seacoast Anti-Pollution League v. Costle, 597 F.2d 306, 311-12 (1st Cir.). These rules are to ensure satisfaction of due process requirements and meaningful public participation in rulemaking, not to straitjacket agency proceedings. BASF Wyandotte Corp. v. Costle, 598 F.2d 637 (1st Cir.). In addition, “the primary purpose of the explanation requirement . . is to facilitate appellate review of administrative decisions.” Seacoast, 597 F.2d at 312.
NO. 78-1608
RANCHERS EXPLORATION AND DEVELOPMENT CORP.
Ranchers challenged EPA’s effluent limitations for the uranium, radium, and vanadium mining industry. However, the only regulation that appears to be specifically challenged in Ranchers’ brief is the total radium limitation for mine discharge. There are two major issues raised by Ranchers. First, it contends that EPA has not provided an adequate data base or adequate explication of its reasoning or analysis. The second argument, that EPA lacks authority to promulgate such limitations, has been rendered academic by a recent decision of this court. 43 Fed.Reg. 29776.
Ranchers did not participate at the rule-making level, and EPA urges that it should not be allowed to seek court review of its challenges. Ranchers acknowledges that it did not so participate, but argues that it can participate in this appeal. It contends that it can challenge EPA’s action as long as the issue was raised by some party during rule-making. There are authorities which hold that as long as issues are raised by some party at the administrative level, they can be raised by another party on review. Wilson & Co. v. United States, 335 F.2d 788 (7th Cir.); Hennesey v. Securities & Exchange Comm’n, 285 F.2d 511 (3d Cir.). “The reason for the rule that such questions should not be raised initially in the court of review is that the administrative agency ought to have the opportunity to rule on the questions in the first instance.” Wilson, at 794. In the recent case of ASARCO, Inc. v. EPA, 188 U.S.App.D.C. 77, 578 F.2d 319 (D.C.Cir.), the District of Columbia Circuit held that the Sierra Club could participate in proceedings even though it did not par*1237ticipate at the rulemaking level. It said: “The issue raised by Sierra was thus not only raised and considered in the proceedings below; it was a substantial part of the fundamental issue in those proceedings.”
It does appear that several challenges to the proposed regulations were raised by different parties during rulemaking, and these challenges did concern adequacy of data sampling and other issues which Ranchers now raises. We will thus consider Ranchers’ arguments. These relate basically to the adequacy of the data base. In any event, it is apparent that development of an adequate data base is not so much an “issue” as it is a requirement that must be met in all instances. 33 U.S.C. § 1314(b). Further, considering the purpose behind the rule that issues must first be raised at the administrative level, it is apparent that EPA was well aware of the problem of developing an adequate data base and so no one was deprived of the opportunity to correct the deficiency during rulemaking.
As discussed, Ranchers claims that the data base is inadequate, and as support for this contention it cites the fact that the Interim Final Regulations were suspended for what it asserts was an inadequate data base. However, the fact that the data base for the Interim Regulations was inadequate has little, if anything, to do with adequacy of data base for the Final Regulations. The record made after suspension of regulations is the key to whether the data base was adequate. The record shows that after the suspension of the Interim Final Regulations, further data was gathered by the EPA and its contractors. The EPA had originally contracted with Calspan Corporation to do the monitoring, sampling, and testing, and also to make the studies of the literature relating to the mining and milling industries. It also prepared detailed cost studies for the economic contractor. The American Mining Congress also gathered data on the subject. Drafts of a Development Document were prepared. After the suspension additional data was accumulated. The EPA engaged another contractor, Jacobs Engineering, to make tests and obtain data relating to the mining and milling of uranium. The Jacobs findings were submitted to the EPA and to Calspan. The contractor for the uranium petitioners, Hazen Research, participated in these discussions and review of the Jacobs report which had been generally circulated in the industry. There were further meetings with the American Mining Congress and several uranium companies. The industry views were sought. The Final Regulations were the product of this process. Ranchers did not choose to participate in any of the proceedings. Its objections before us on appeal constitute a statistical exercise without sufficient consideration of the actual circumstances existing where and when the data was originated.
The control of Radium 226, according to this record, is obtained by barium chloride coprecipitation to reduce the dissolved radium, and to promote the settling of resultant suspended solids and other suspended solids. The record demonstrates the testing done by Calspan and Jacobs of facilities using barium chloride. Extensive data was so collected and furnished by others, and examined by all concerned. The record makes it abundantly clear that the data base for the Regulations applicable to the uranium mining and milling industries was adequate.
It is also apparent from the record that the EPA’s explanation of the course it followed is adequate also. We have elsewhere in this opinion considered the standards which must be followed in that regard. When these standards are here applied, we must reject Ranchers’ objections on this point. In this connection we consider an explanation to include the application of the statistical material in reaching the conclusion. This may or may not be a “calculation” (National Crushed Stone Ass’n v. EPA, 601 F.2d 111 (4th Cir.)), and need not be a separate step by step process.
Ranchers urges a variation on the basic argument and on the use of data or performance levels, and this relates to the use of averages or different levels of the technological performance in the industry. The *1238EPA here refers to the “average of the best existing performance.” This court recently held in American Petroleum Institute v. EPA, 540 F.2d 1023 (10th Cir.), that in promulgating 1977 effluent limitations, the EPA can base its regulations on results from plants using the best technology, rather than the average of the industry. Tanners’ Council of America, Inc. v. Train, 540 F.2d 1188 (4th Cir.); American Frozen Food Institute v. Train, 176 U.S.App.D.C. 105, 115, 539 F.2d 107, 117 (D.C.Cir.). Also in Weyerhaeuser Co. v. Costle, 191 U.S.App.D.C. 309, 590 F.2d 1011 (D.C.Cir.), the EPA based its limitations on “pollution control achieved at the ‘average’ of the best mills.”
Ranchers maintains that EPA did hot consider the cost and economic impacts of the total radium limitation, as required by 33 U.S.C. § 1314(b)(1)(B). However, Ranchers does not deny that EPA considered fully the economic impact of coprecipitation treatment and total suspended solids in uranium processing, and that because of the nature of the treatment process if these limitations are met, then total radium limitations in processing will also be met. Therefore, the total radium limitations set by EPA will impose no additional costs at all, and EPA has thus met the requirement of consideration of cost.
Even if the above explanation by EPA is not valid, case law generally has given EPA wide leeway in its analysis of costs. This court in American Petroleum Institute v. EPA, 540 F.2d 1023 (10th Cir.), found that EPA had complied with the requirement of cost-benefit analysis when capital investment for the industry overall was examined. The First Circuit in BASF Wyan-dotte Corp. v. Costle, 598 F.2d 637 (1st Cir.), characterized the obligation of cost considerations as follows:
“The obligation the Act imposes on EPA is only to perform a limited cost-benefit balancing to make sure that costs are not ‘wholly out of proportion’ to the benefits achieved.”
The court also noted:
“We are not convinced that the duty to ‘include consideration of cost in relation to benefit’ imposed on EPA by this clause of § 1314(b)(1)(B) is significantly different from the duty imposed by the same subsection to ‘take into account’ certain other factors [such as non-water quality impact].”
The court went on to state that even though there may be valid criticisms of EPA estimates and even though estimates may be inaccurate, EPA’s duty is to develop no more than a “rough idea” of cost to industry.
We must hold that adequate considerations have been given to the cost aspect of the problem. Also we must hold that the Regulations, as they pertain to the subject of Ranchers’ petition, are reasonable under the standard of review to be applied.
NOS. 76-1241, 78-1894
HECLA MINING COMPANY
In Hecla’s petition for reconsideration filed December 12, 1978, EPA was requested to withdraw the copper and cyanide effluent limitation for froth flotation mills. The cyanide limitations were withdrawn on March 1, 1979 (44 Fed.Reg. 546), but the copper limitations were not. We are thus here concerned with the copper limitations only. (43 Fed.Reg. 29774).
Posthearing Data:
The contentions of the parties as to the Hecla’s second petition for reconsideration seeking withdrawal of the copper limitations, together with the Werthman and Griffith affidavits seeking to supplement the record, come late as posthearing, postrulemaking, and postappeal explanations of the positions of the respective parties. There are very good and practical reasons why the parties should be limited to the rulemaking proceedings for the development of the issues on the record. See Weyerhaeuser Co. v. Costle, 191 U.S.App.D.C. 309, 590 F.2d 1011 (D.C.Cir.), and American Frozen Food Institute v. Train, 176 U.S.App.D.C. 105, 539 F.2d 107 (D.C.Cir.).
Hecla contends that the requirement for comment has been met because during sus*1239pension of the Interim Final Regulations in 1976, EPA’s contractor became “aware” of Hecla’s problems with formation of eop-per/cyanide complex. Thus Hecla urges that EPA cannot claim Hecla failed to alert them to the problem. However, no case has been found to support the proposition that “awareness” of a problem by an agency is sufficient to waive the requirement for specific comment by an industry at the rule-making level. A strict construction of the standard of review would probably preclude Hecla’s challenge on the issue of copper-cyanide formation. Indeed, if the problem was not clearly raised at the agency level there would be no record of agency action and no way for the court to review the rulemaking process.
As to the posthearing Werthman affidavit submitted by the EPA we have recently noted that in reviewing EPA action under the Clean Water Act, “[a]fter the fact rationalization by counsel in brief and argument does not cure noncompliance by the agency with the stated principles.” American Petroleum Institute v. E. P. A., 540 F.2d 1023, 1029 (10th Cir.). The EPA must make clear its analysis and reasoning at the rulemaking level, and its grounds for action must be clearly disclosed in the record. Hooker Chemicals & Plastics Corp. v. Train, 537 F.2d 620, 633 (2d Cir.). In short, the EPA should not use facts and information gained after rulemaking on which to support its limitations.
However, there is in the record (J.A.II, pp. 1747-48) a letter of October 12, 1976, from Calspan to Hecla relating to the possibility of the formation of the copper-cyanide complex. Hecla sometime later responded (J.A.II, p. 1700), and made further comments on the nontoxic nature of the chemical complex. Thus the beginning, at least, of the discussion of this issue appears in the record. The Werthman affidavit contains new data and an analysis to support the copper limitation and the copper-cyanide complex. It appears to be in part an after-the-fact explanation of agency action; however, we will consider it. Thus Hecla’s arguments on this point were partly considered in the rulemaking period and partly in the posthearing exchanges. We have considered the point or issue in its several aspects, but find nothing in it separately or together with the other points to demonstrate the regulations to be arbitrary and unreasonable. There appears to be a problem but again, if the data and further experience demonstrate that the mill is fundamentally different in this respect, a variance may be applied for. 43 Fed.Reg. 29777, § 440.81.
The Mine-Mill Code:
Hecla advances as a reason for a late challenge to EPA action on the copper limitations that Hecla was denied an opportunity to comment on EPA’s conclusions because interested parties were not advised as to which plants would be exemplaries in the final Development Document. Indeed, Hecla claims that mills (listed by number) as exemplary are not the mills that EPA now lists in its brief as exemplary. Mines and mills are identified in EPA Development Documents by number at the request of the American Mining Congress. Hecla was provided with a mine-mill code in previous litigation to which Hecla was a party so that it could have identified mills mentioned as exemplary. In addition, there is no evidence that Hecla requested the names of exemplary mills at the rulemaking level, or that EPA would have denied this request.
The EPA Mine/Mill Code Index is in a supplement to the Development Document dated December 7, 1976. Thus it appears that the list of nineteen mines and mills (two in Idaho, where Hecla’s mine is located) was available to Hecla and there is no indication that it would not have been furnished if requested (J.A. Ill 2932-33 Attachment). In the Development Document, five mills are listed as exemplary for lead and zinc ores: Nos. 3101, 3102, 3103, 3105, and 3108 (J.A. 560, Dev.Doc. 492). EPA’s brief does indeed state that mills 3105 and 3108 are not exemplary. But the Development Document lists mill 3108 as exemplary for wastewater flow and treatment and effluent characteristics (Dev.Doc. 505-6).
*1240In the “Summary of Mine/Mill Operations Employing Exemplary Wastewater Treatment,” Tables VII 56 and 57 summarize mine/mill operations in all metal categories with treatment technology exclusive of zero discharge. Mills 3108 and 3109 are changed in the same table that was admitted later as supplementary evidence because it is printed more clearly than the original table. The mine/mill index was thus available to Hecla for inspection to determine the identity of exemplary facilities. However, it does appear that EPA’s designation of exemplary facilities is inconsistent and somewhat misleading; however, not to a degree to serve as a basis for the position Hecla asserts. Certainly Hecla was skillful enough to cope with such a matter.
Denial of Notice:
A related question is whether EPA’s inconsistency in listing exemplary mills constitutes a denial of notice and the opportunity to comment. This court very recently stated: “As long as a party to an administrative proceeding is reasonably apprised of the issues in controversy and is not misled, the notice is sufficient.” Savina Home Industries v. Secretary of Labor, 594 F.2d 1358 (10th Cir.). “Agency action should be upheld where the path of administrative proceedings may reasonably be discerned even if the demarcations are of less than ideal clarity.” Hooker Chemicals & Plastics Corp. v. Train, 537 F.2d 620, 632 (2d Cir.), citing Bowman Transportation, Inc. v. Arkansas-Best Freight System, Inc., 419 U.S. 281, 95 S.Ct. 438, 42 L.Ed.2d 447 (1974).
It thus appears that the fact that mills were not listed precisely in the final Development Document did not deprive Hecla of meaningful participation in rulemaking such that remand would be necessary. Indeed, the fact that Hecla made no comment about or request for exemplaries at the rulemaking level renders its argument even less persuasive.
The Explanation:
As a somewhat separate issue, Hecla contends that the effluent limitations should be remanded for reconsideration because of failure to explain how the limitations were developed and what they were based on. As mentioned, adequate explanation of agency action is indeed necessary to the validity of rulemaking. A review of the record with special attention to the final Development Document discloses an adequate explanation of the basis for the regulations. Obviously there is not a step by step progression within the several documents, but taken together with the related material and tables, we must hold that the explanation is adequate within the applicable standards. As hereinabove mentioned the grounds for the EPA action must be disclosed in the record. Hooker Chemicals & Plastics Corp. v. Train, 537 F.2d 620 (2d Cir.).
A brief reference to the EPA explanation should be made. The Development Document discusses in general the process waste water characteristics (J.A. 294, Dev.Doc. 241), and selection of pollution parameters based on stated factors (J.A. 432, Dev.Doc. 379). The toxic effects of copper are discussed, and it is noted that: “In hard water, the toxicity of copper salts is reduced by the precipitation of copper carbonate or other insoluble compounds.” (J.A. 444, Dev.Doc. 391).
Tables VII — 56 and 57 summarize mine/mill operations, and tabulate the effluent concentrations achieved by a number of mills (J.A. 647, Dev.Doc. 579). The' BPCTCA for lead and zinc ores is set out in a somewhat cursory fashion in the Development Document, and Table IX-6, p. 739, sets out the levels of effluent reduction obtainable by application of the tailing pond technology recommended (J.A. 802, Dev.Doc. 736). Chemical Analysis Data for lead/zinc mines and mills 3101, 3104 and 3110 are set out at J.A. Ill 2928. The tables set out parameters, monitoring periods, frequency of observation (usually at least each month), number of observations (from twenty-one to one hundred), and mean standard deviation and range in mg/1.
The explanation of basis for copper limitations is adequate and well within prevailing standards. Data was set forth in detail, *1241appropriate technology is noted, and conclusions are derived therefrom.
Costs and Subcategorization:
The record demonstrates an adequate consideration of costs by EPA. Treatment costs for lead and zinc category mills was considered in the Development Document (J.A. 678, Dev.Doc. 610-14). The twenty-one known major mills were taken into account, and a hypothetical mill was used as the representative. Specific monetary figures were quoted for several categories of both capital investment and operating costs. An “Economic Analysis of Proposed Effluent Guidelines,” dated October 1975 is found at J.A. 980. “Economic Analysis of Effluent Guidelines, The Ore Mining and Dressing Industry,” May 1977 is found at J.A. 989. This analysis is quite detailed and takes into consideration many different factors. Calspan Technical Report: Cost-Data Development and Economic Analysis Supplement B-2, April 18, 1975, is at J.A. Ill 2508. This is also a very detailed analysis, treating such diverse areas as processing systems and taxes and contractor’s fees. In addition, costs for implementation of proposed waste water treatment standards for each type of mining is discussed. Twenty-five pages are devoted to lead/zinc mining and milling (J.A. Ill 2601). Cost considerations were thus more than adequate.
We also conclude that there was no failure by EPA to subcategorize. The various elements including, of course, the nature of waste water were considered and evaluated. See Dev.Doc. p. 159. EPA found subcate-gorization of milling unnecessary because: “The only ore concentrations process currently practiced in the lead/zinc industry is froth flotation.” (J.A. 213, Dev.Doc. 160). Characteristics of process waste water from lead and zinc ores were considered (J.A. 294, Dev.Doc. 241). The characteristics of lead and zinc waste were illustrated by data from three mills.
We must conclude that the regulations herein challenged by Hecla must be upheld. We find no arbitrary and capricious action by the EPA in the adoption of the regulations.
NOS. 75-1878 AND 78 — 1560
KENNECOTT COPPER CORPORATION
Petitioner challenges the effluent limitations for the base and precious metals subcategory as they apply to froth flotations mills and mine drainage facilities. Petitioner’s brief raises many issues; the principal contentions, however, are that the limitations are unsupported by the data, that EPA failed to explain how the limitations were developed, and that EPA inadequately considered the limitations costs. (43 Fed. Reg. 29774).
We have held today that the limitations for the base and precious metals subcategory are supported by adequate and accurate data in the record. See Gulf & Western portion of this opinion. Recently we also answered another of petitioner’s arguments. In United States v. Earth Sciences, Inc., 599 F.2d 368 (10th Cir.), we upheld EPA’s authority to promulgate limitations regulating mining activities.
The record contains EPA explanations for its conclusions. The Development Document fully explains the factors considered and the methodology used in subcategorization. Dev.Doc. 147-77. The choice of subcategorization methodology lies within the Agency’s sound discretion. American Petroleum Institute v. EPA, 540 F.2d 1023 (10th Cir.), and this methodology was unchallenged during rulemaking.
There is no mystery, as petitioner complains, concerning the reasons behind EPA’s choice of mines for data gathering. The Development Document explains that EPA sought mines and milling facilities thought to employ exemplary pollution treatment technology:
“Based upon information gathered as part of the assembly of a data base, examination of NPDES permits and permit applications, surveys by trade associations, and examination of texts, journals, and the literature available on treatment practices in the industry, selection of mining and milling operations which were *1242thought to embody exemplary treatment practice was made for the purpose of sampling and verification, and to supplement compiled data.” Dev.Doc. 15.
The term “exemplary” presents no serious interpretation obstacle as it places all interested persons on notice EPA sought facilities that employed the best pollutant-reduction technology, the most effective methods of “restorpng] and maintainpng] the chemical, physical, and biological integrity of the Nation’s waters.” 33 U.S.C. § 1251(a) (1976) (congressional declaration of Act’s goals and policy). This is consistent with holdings in support of EPA’s discretion to use a standard, which averages the results from the best facilities. See, e. g., E. I. duPont de Nemours & Co. v. Train, 430 U.S. 112, 131, n. 21, 97 S.Ct. 965, 51 L.Ed.2d 204; American Petroleum Institute, 540 F.2d at 1034; Hooker Chemicals & Plastics Corp. v. Train, 537 F.2d 620, 632 (2d Cir.). And it comports with legislative intent. See S.Rep. No. 92-414, at 50 (1971), U.S.Code Cong. & Admin.News 1972, p. 3668.
Here EPA selected two of its exemplary mines because pond design maximized the potential for settling of solids by using the full area available for the settling pond. One of these mines constructed a dike in its tailings pond to reduce the impact of wind, and two other mines use multipond systems to enhance the settling process. Kenne-cott’s tailings pond consists of 250 acres on an available area of 2,500 acres. Using the available area to build a larger pond would increase retention time and maximize the potential for settling of solids. These differences in technology implementation will obviously affect the amount of pollutants discharged and the selection of exemplary plants. Effluent results from mine to mine seem to vary for numerous reasons.
This case differs from National Crushed Stone Ass’n v. EPA, 601 F.2d 111 (4th Cir.), where the court found EPA’s development document failed to “discuss the calculation process by which the agency arrived at the monthly average limit.” Id. at 118. The record here provides sufficient explanations of EPA’s calculation process: “For the ore mining and dressing industry, this level of technology [BPT] is based on the average of the best existing performance by facilities of various sizes, ages, and processes within each of the industry’s subcategories.” Dev.Doc. 723. See generally id. at 723-83. Furthermore, this is not a case where EPA has relied on critical data known only to the agency. See Portland Cement Ass’n v. Ruckelshaus, 158 U.S.App.D.C. 308, 486 F.2d 375 (D.C.Cir.). Nor does petitioner’s statistical analysis refute EPA’s data.
EPA provides an adequate record explanation for its selection of exemplary mines. The limitations promulgated reflect the average of the values produced by best performance. Indeed, the limitations are less stringent than the precise average of the best figure.
The legal standard by which we determine the adequacy of costs consideration is whether “EPA made a serious, careful, and comprehensive study of the costs which compliance will impose on the industry.” American Petroleum Institute, 540 F.2d at 1038. There is ample evidence in the record that EPA has satisfied this requirement. EPA’s contractor performed a plant by plant cost analysis for numerous flotation mills in the base and precious metals subcategory, J.A. 2507 — 2707, which was later updated, J.A. 1880, 1884-87. An additional EPA contractor, Arthur D. Little, Inc., performed a comprehensive analysis of the regulation’s macroeconomic effects in terms of prices, production, and so forth. Accordingly, petitioner’s challenge of EPA’s cost consideration is without merit.
Petitioner also challenges the storm runoff provisions on the basis that they include “nonpoint” as well as “point sources.” We consider the challenge moot in light of EPA’s Notice of Clarification concerning storm runoff provisions.
The Clarification, dated February 2, 1979, has as its purpose:
“. . . [T]o make it clear that those [storm runoff] provisions do not apply to diffuse storm water and runoff, but apply only to point source discharges.
“The regulations are not intended to require the operator to collect and con*1243tain diffuse storm runoff which would not otherwise be collected in or does not otherwise drain into a point source.” (Emphasis added). 44 Fed.Reg. 7,953-54.
This is in complete harmony with our holding that notwithstanding its lack of authority to regulate “nonpoint sources,” EPA has full authority to regulate any “point source” discharge. United States v. Earth Sciences, Inc., 599 F.2d 368 (10th Cir.). We are satisfied that EPA’s storm runoff provisions apply only to “point sources” as authorized by Congress.
Kennecott urges us to strike down these provisions because they are ambiguous, arguing that it is unable to determine whether certain of its facilities are “point sources” or not. These are determinations to be made in the first instance in the context of a permit proceeding. It is sufficient here to note that EPA is entirely within its authority in regulating storm runoff that falls within a “point source.” Congress has defined “point source” as “any discernible, confined and discrete conveyance, including but not limited to any pipe, ditch, channel, tunnel, conduit, well, discrete fissure, container, rolling stock, concentrated animal feeding operation, or vessel or other floating craft, from which pollutants are or may be discharged.” 33 U.S.C. § 1362(14) (1976). Congress has purposefully phrased this definition broadly. This is as it should be given its contemplated applicability to literally thousands of pollution sources. To cast such definitions in absolute, unequivocal terms would be unrealistic, if not altogether impossible. As we observed in American Petroleum Institute, 540 F.2d at 1032: “On the road to attainment of the no discharge objective some flexibility is needed.”
Careful examination of petitioner’s remaining arguments has persuaded us that they are without merit. Accordingly, the BPT regulations are affirmed.
NO. 76-1288
GULF & WESTERN INDUSTRIES, INC.
Petitioner challenges EPA’s effluent limitations for the base and precious metals subcategory which are set forth at 43 Fed. Reg. 29,775. This challenge arises from profound differences between exemplary mines EPA relied upon in formulating the limitations and petitioner’s Friedensville mine in western Pennsylvania. The exemplary mines are largely free of serious geologic fractures or faults, Dev.Doc. 242, and have water flows that range from 144,000 to 835,000 gallons per day, J.A. 1807. The formations at Friedensville are extensively folded and fractured, J.A. 1854-56, and the water flow is over 44,000,000 gallons per day, J.A. 1807. The petitioner contends EPA’s data fail to support the effluent limitations promulgated. Petitioner hinges its position on the assertion that EPA ignored the Friedensville situation, thereby rendering EPA’s assumptions inaccurate and data unrepresentative.
Comments were submitted to EPA on October 26, 1976, wherein petitioner informed the Agency of the unusually high water flow at Friedensville. See J.A. 1808. In response EPA requested its contractor, Calspan, to revise a cost estimate previously prepared for the Friedensville mine. This cost estimate was revised to reflect the new information submitted by petitioner and received by EPA on December 3, 1976. J.A. 1875 — 79. Indeed, EPA requested its contractor to review the literature pertaining to the geology of petitioner’s mine. The contractor returned a report dated January 19, 1977, which observed that serious geological conditions may exist, but that certainty regarding these conditions would require extensive tests. J.A. 1854-58.
It is obvious that EPA was fully aware of petitioner’s situation. EPA took pains to consider these facts. Thus the data base can hardly be said to rest on inaccurate or incomplete assumptions concerning water flow and geologic conditions. It is well settled that EPA may use the “best plant” standard in formulating the data base underlying the effluent limitations. American Petroleum Institute v. EPA, 540 F.2d 1023, 1034 (10th Cir.), cert. denied, 430 *1244U.S. 922, 97 S.Ct. 1340, 51 L.Ed.2d 601. See also S.Rep. No. 92-414 at 50 (1971): “The [EPA] Administrator should establish the range of best practicable levels based upon the average of the best existing performance by plants of various sizes, ages, and unit processes within each industrial category.”
Petitioner also argues that the BPT is neither practicable nor available and would have us find that its mine is “fundamentally different.” Whether this is true is not for us to decide in the first instance; this question is most appropriately addressed in the context of a permit proceeding. Natural Resources Defense Council, Inc. v. EPA, 537 F.2d 642, 647 (2d Cir.). There, the EPA will be free to exercise the flexibility built into the statute through its permit proceedings by acting on a record that focuses on petitioner’s individual circumstances. The Supreme Court’s holding in E. I. duPont de Nemours & Co. v. Train, 430 U.S. 112, 128, 97 S.Ct. 965, 51 L.Ed.2d 204, requires EPA to grant variances where appropriate. And “[a]s a result of G & W’s submissions and EPA’s own review, the agency recognizes that G & W may well be entitled to a variance.” (Respondent’s Brief at 62).
The regulations are upheld.
NOS. 76-1242, 76-1287, 78-1686
HOMESTAKE MINING COMPANY
Homestake attacks in general the sufficiency of the data base used by EPA and alleges that EPA’s determinations are unsupported by the record. The major challenge by Homestake concerns EPA’s determination that zero discharge of waste water from gold and silver cyanidation mills is Best Practicable Technology Currently Available (BPT). Finally, Home-stake attacks EPA’s subcategorization of the gold industry and its consideration of nonwater quality impacts and costs.
The Zero Discharge Limitations:
The challenge by Homestake on the zero discharge limitations is based on the refusal of EPA to modify the no discharge limitation for gold mills that use the cyani-dation process after the cyanide limitations for froth-flotation process mills had been withdrawn by EPA. Homestake contends that if the total cyanide limitation for froth-flotation mills was withdrawn because inadequately supported by the record, then the no discharge limitation for cyani-dation mills cannot stand since both are based upon the same record and technology. Homestake thus urges that the retention by EPA of the no discharge limitation is arbitrary and capricious.
EPA on the other hand urges that the no discharge limitation may be met regardless of regulations on cyanide as the two limitations are mutually exclusive.
EPA argues that the only effluent limitations that must be met by Homestake are mine drainage restrictions in 43 Fed.Reg. 440.22(a)(1) (1978). That section does not contain any cyanide limitation. The basic requirement or method to meet the no discharge requirement is to recycle waste water within the mining and milling system. EPA notes that five active gold and silver cyanidation mills are in compliance with the no discharge requirement (Mills 4101, 4119, 4122 and 4131). In this connection, as we have elsewhere pointed out, EPA may use the average of the best technology in setting BPT. As this recycle method has been demonstrated to be feasible, it appears that zero discharge of process waste water is not unreasonable.
The Data Base:
Homestake urges that it was arbitrary and capricious for the EPA not to consider data on Canadian mines. However, this point only goes to the general sufficiency of the data base and cannot be considered an independent ground to challenge the regulations. See American Frozen Food Institute v. Train, 176 U.S.App.D.C. 105, 539 F.2d 107 (D.C.Cir.).
As to the general challenge to the data base it is apparent that the EPA used three gold cyanidation mills out of some fourteen available. The record shows some but little further sampling after the Inter*1245im Final Regulations were promulgated. However, we cannot say on the record before us that the data base was inadequate to support the regulations under the standards of review. Many of Homestake’s comments on the data base relate to conditions existing and the extent of data available at a time too remote to be persuasive. As indicated above, we must hold that the data base was not exhaustive but was adequate to meet Homestake’s challenge.
The Ferroeyanide Issue:
EPA asserts that Homestake did not raise during rulemaking the issue of ferroeyanide buildup in recycling waste water, and consequently is precluded from urging such issue on appeal as a basis for invalidating the guidelines. Homestake argues that the problem arose from the recycling system that only began operating in December of 1977, and that as soon as the problem became apparent, EPA was notified. The Interim Final Regulations were issued on November 6, 1975, and suspended on May 24, 1976. The Final Regulations were published July 11,1978, about six months after the recycling operation had begun. Only free cyanide matters were raised by Homestake during rulemaking.
Homestake states that EPA was notified of the problem and has been supplied with an extensive analysis of the matter. However, there are no record citations to this material, and the record itself does not appear to contain any reference by Homestake of the problem. As elsewhere held, it is well settled that industry must first utilize the opportunity for comment before it may raise issues on appeal. American Iron & Steel Institute v. EPA, 526 F.2d 1027, 1050 (3d Cir.). Absent proof that this particular problem was raised before publication of the Final Regulations, this court will not now consider the ferroeyanide issue. At any rate, if the problem is unique to Homestake’s recycling as to make the operation “fundamentally different,” then application for a variance would be appropriate.
Subcategorization :
EPA placed in the same subeátegory all mills using the cyanidation process alone. The challenge to EPA’s subcategorization of gold mines and mills has four parts: EPA’s consideration of mineralogy; consideration of climate, rainfall and location; failure to establish a subcategory for mines and mills using sand backfilling; EPA’s separate classification of the Knob Hill mill and Homestake mill.
It is true, as Homestake argues, that all of these factors were discussed in a cursory manner, and it was determined that the most effective means of categorizing was to base it on the type of beneficiation process used, Dev.Doc. 161. It does not appear from the record that the mineralogy was seriously considered in the context of treatment problems. EPA’s rationale for this appears to be that:
“The specific beneficiation process adopted is based upon the mineralogical characteristics of the one; therefore, the waste characteristics of the mine or mill reflect both the ores mined and the extraction process used. For these reasons, ore mineralogy was determined to be a primary factor affecting subcategorization in all categories.”
This channeling of ore mineralogy considerations into an overall characterization based on beneficiation process is a rough sort of basis for categorization, and has some relationship to the chemistry of the waste water. It cannot be said to be arbitrary and capricious.
It seems clear that climate, rainfall, and location were adequately considered and provided for by 43 Fed.Reg. 440.22(a)(4), which allows discharge from gold cyanidation mills in net precipitation areas, and by 43 Fed.Reg. 440.81(c) which allows discharges under conditions causing storm runoff.
With regard to sand backfilling, Home-stake does not demonstrate why mines-mills using that device should be placed in a different subcategory. Homestake bases its argument for the most part on a statement in the Development Document that sand backfilling causes cyanide discharge *1246problems. Dev.Doc. 256. However, cyanide discharge issues are not before us.
The final argument by Homestake is the placing of Knob Hill Mill and Home-stake Mill in different subcategories is not proper. The EPA classifies Knob Hill (Mill 4104) as a froth flotation mill, apparently based on the recovery from that process, and thus under 43 Fed.Reg. 440.22(a)(2). It appears that Knob Hill processes about five percent of its ore by froth flotation and does recover eighty-five percent of its product therefrom; however, it processes the remaining ninety-five percent of its ore by the cyanidation process. The classification based on the percentage of product alone is arbitrary when it ignores the volumes of ore processed. On the record before us, it is unreasonable and arbitrary to place the Knob Hill mill and the Homestake mill in different subcategories.
Consideration of Non-Water Quality Impact:
Non-water quality impact must be taken into account as a factor in determining BPT. 33 U.S.C. § 1314(b)(1)(B). Home-stake’s attack on EPA’s consideration of non-water quality impact is directed specifically to the purported lack of attention given to the effect of water seepage containing cyanide on groundwater. The Development Document (p. 718) devotes one paragraph to groundwater contamination in its review of non-water quality impact, and one page discusses waste water impoundment systems such as tailing ponds (pp. 494-5). Treatment technology for tailing ponds is discussed in some detail in the Development Document at pages 425-7. Problems of seepage are specifically treated. EPA’s discussion of non-water quality impacts in general constituted only four pages of generalized discussion (Dev.Doc. pp. 718 — 21). It is evident, however, that treatment of non-water quality issues was dealt with adequately.
Courts have been liberal in reviewing EPA’s treatment of non-water quality impact. In Weyerhaeuser Co. v. Costle, 191 U.S.App.D.C. 309, 348, 590 F.2d 1011, 1050 (D.C.Cir.), the court stated:
“. . . As we have discussed, we believe Congress entrusted the manner of deliberation about all of the ‘consideration factors’ to EPA’s discretion, and we are prepared to uphold EPA on that basis alone.”
Also, the court there said (191 U.S.App.D.C. at 351, 590 F.2d at 1053):
“Thus, since Congress intended EPA’s internal structure to protect the non-water environment, the judicial function is completed when we have assured ourselves that EPA expressly considered the probable environmental impacts of its regulations.”
In American Paper Institute v. Train, 177 U.S.App.D.C. 181, 543 F.2d 328 (D.C.Cir.), the same court found EPA’s consideration of non-water quality environmental impact to be sufficient. There only a few pages were devoted to discussion of air pollution, noise potential, solid waste disposal, and by-product recovery. This was the express consideration required. The court there also declined to find invalid EPA’s treatment when certain treatment processes were not discussed.
A case with facts not unlike the case at bar is California & Hawaiian Sugar Co. v. EPA, 553 F.2d 280 (2d Cir.). In this case the industry complained that sludge treatment was not sufficiently considered, and the court found that EPA had recognized the problem and indicated a solution of sludge drying. The court also held that the problems were adequately considered when the agency “took into consideration the likelihood that design modification and careful placement could abate the noise and fogging problems.” In our case EPA did indicate solutions to seepage in tailing ponds, and it did “take into consideration” the fact that proper design of impoundment facilities could avoid the problem. Thus it appears that this recognition of the problem is sufficient to meet the standard for review.
Consideration of Costs:
The standard of review and case law trends concerning EPA’s consideration *1247of costs is set forth elsewhere herein. The record here shows that costs were considered not only in the Development Document (616-37, considering costs for individual mines and mills), but also in an Economic Analysis of Effluent Guidelines, Ore Mining and Dressing Industry, May 1977, and the Calspan Technical Report: Cost-Data Development and Economic Analysis Supplement B-2, April 18, 1975. Capital investment, annual cost, and operation and maintenance are analyzed in detail by EPA in these reports. This is sufficient.
The challenged regulations are upheld with the exception of the placing of the Knob’ Hill Mill and Homestake Mill in the same subcategory, as described in pages 32 and 33 above.