In a dubious reincarnation of the 1890’s world of Yukon poet Robert Service, we deal here a century later with “strange things done in the midnight sun by the men who moil for gold.”1 We are asked to determine the validity of Environmental Protection Agency regulations under the Clean Water Act which govern placer mining and have particular impact on the gold-rich streambeds of Alaska. Because the regulations are complex and the issues raised are multitudinous, our opinion (written, alas, in arid prose) is outlined in some detail:
I. Background.1282
A. Placer Mining.1282
B. Statutory Framework.1282
C. Rulemaking History.1283
II. Discussion.1284
A. Standard of Review.1284
B. The EPA’s Authority Under the Clean Water Act.1285
C. Notice-and-Comment Procedures.1286
*12821. Additional Documents in the Record.1286
2. Best Management Practices.1286
3. Adoption of BAT for All Mines.1288
D. The Final Rule.1289
1. Merits of the Limitations.1289
a. Determination of BPT .1289
b. Determination of BAT.1290
(1) Analysis for Costs.1290
(2) Total Suspended Solids Limitations.1291
(3) Settleable Solids Limitations .1291
(4) Presence of Toxic Pollutants.1292
c. New-Source Criteria.1292
2. Selection, Presentation, and Analysis of Data.1293
a. Processing of Data.1294
b. Presentation of Analyses.1295
c. Falsified Evidence.1295
d. Deference to Other Agencies.1296
e. Consideration of Costs in Promulgating BMPs.1297
f. Non-Water Quality Environmental Impacts.1297
g. Hydraulic Mining Data. 1297
3. Mandating of Technology.1298
4. Availability of Variances.1298
5. Timeliness of Rule Promulgation.1299
6. Time of Compliance .1299
7. Constitutionality of the Final Rule.1300
III. Conclusion .1301
I
BACKGROUND
A. Placer Mining
Placer mining is one of the four basic methods of mining metal ores; it involves the mining of alluvial or glacial deposits of loose gravel, sand, soil, clay, or mud called “placers.” These placers often contain particles of gold and other heavy minerals. Placer miners excavate the gold-bearing material (paydirt) from the placer deposit after removing the surface vegetation and non-gold-bearing gravel (overburden). The gold is then separated from the other materials in the paydirt by a gravity-separation process known as “sluicing.”
In the sluicing process, a miner places the ore in an on-site washing plant (usually a sluice box) which has small submerged dams (riffles) attached to its bottom. He causes water to be run over the paydirt in the sluice box; when the heavier materials (including gold) fall, they are caught by the riffles. The lighter sand, dirt, and clay particles are left suspended in the waste-water released from the sluice box.
Placer mining typically is conducted directly in streambeds or on adjacent property. The water usually enters the sluice box through gravity, but may sometimes also enter through the use of pumping equipment. At some point after the process described above, the water in the sluice box is discharged. The discharges from placer mining can have aesthetic and water-quality impacts on waters both in the immediate vicinity and downstream. Toxic metals, including arsenic, cadmium, lead, zinc, and copper, have been found at a higher concentration in streams where mining occurs than in non-mining streams.
It is the treatment of the sluice-box discharge water before it reenters a natural water course that is at the heart of this case.
B. Statutory Framework
Congress enacted the Clean Water Act to “restore and maintain the chemical, physical, and biological integrity of the Nation’s waters.” 33 U.S.C. § 1251(a) (Supp. V 1987).2 Under the Act, the EPA must im*1283pose and enforce technology-based effluent limitations and standards through individual National Pollutant Discharge Elimination System (“NPDES”) permits. See 33 U.S.C. § 1342 (1982 & Supp. V 1987). These permits contain specific terms and conditions, as well as numerical discharge limits, which govern the activities of pollutant dischargers. Through the Clean Water Act, Congress has directed the EPA to incorporate into the permits increasingly stringent technology-based effluent limitations.
Congress specified a number of means for the EPA to impose and to enforce these limitations in NPDES permits. For instance, it requires the Agency to establish effluent limitations requiring dischargers to use the “best practicable control technology currently available” (“BPT”) within an industry. 33 U.S.C. §§ 1311(b)(1)(A), 1314(b)(1)(A) (1982). These limits are to represent “the average of the best” treatment technology performance in an industrial category. See EPA v. National Crushed Stone Ass’n, 449 U.S. 64, 76 n. 15, 101 S.Ct. 295, 303 n. 15, 66 L.Ed.2d 268 (1980). The EPA is further required to promulgate limitations both for discharge of toxic pollutants by mandating that an industry use the “best available technology economically achievable” (“BAT”) and for discharge of conventional pollutants by requiring the use of the “best conventional pollution control technology” (“BCT”); the congressionally imposed deadline for promulgation of these limitations was March 31, 1989. 33 U.S.C. §§ 1311(b)(2)(A), (C), (D), and (E); 1314(a)(4), (b)(2), and (b)(4) (1982 & Supp. V 1987). Whether a pollutant should be considered toxic has been left to the discretion of the EPA. See 33 U.S.C. § 1317(a)(1) (1982); see also 40 C.F.R. § 401.15 (1989) (designating, pursuant to 33 U.S.C. § 1317(a)(1), 65 toxic pollutants).
In addition, new pollution sources in an industry must meet a separate set of standards called new-source performance standards (“NSPS”). 33 U.S.C. § 1316 (1982). These standards limit the discharge of pollutants by new sources based on the “best available demonstrated control technology” (“BDT”). Id. Finally, the EPA is authorized to establish best management practices (“BMPs”) “to control plant site runoff, spillage or leaks, sludge or waste disposal, and drainage from raw material storage” in order to diminish the amount of toxic pollutants flowing into receiving waters. 33 U.S.C. § 1314(e) (1982).
C. Rulemaking History
On November 20,1985, proceeding under the Clean Water Act, the EPA proposed regulations for placer mining. See 50 Fed. Reg. 47,982 (1985). For most mines 3 processing fewer than 500 cubic yards of ore per day (“yd8/day”), the EPA proposed BPT effluent limitations of 0.2 millilitres per litre (“ml/1”) of discharge for settleable solids and 2,000 milligrams per litre (“mg/1”) for total suspended solids. For mines processing more than 500 yd3/day of ore, the EPA proposed more-stringent BCT and BAT limitations, as well as new-source performance standards (NSPS) prohibiting the discharge of processed waste-water. Twice during the rulemaking process, the Agency published notices of new information and requested public comment on additional financial and technical data. See 51 Fed.Reg. 5,563 (1986); 52 Fed.Reg. 9,414 (1987).
*1284As a result of its studies, the comments received during the review-and-comment periods, and new studies undertaken in response to the submitted comments, the EPA promulgated final effluent-limitation guidelines and standards on May 24, 1988. See 53 Fed.Reg. 18,764 (1988). The EPA established a BPT limitation, based upon simple-settling technology, for settleable solids of 0.2 ml/1 for virtually all mines.4 The final rule also established BAT limitations and NSPS, based on recirculation technology, restricting the flow of processed wastewater that could be discharged.5 In addition, the EPA promulgated five BMPs to control discharges due to mine drainage and infiltration. These regulations were to become effective on July 7, 1988.
During the rulemaking process, commen-ters had expressed concerns about the impact that the proposed regulations might have on small placer mines. In promulgating the final rule, the EPA therefore solicited, for a sixty-day period, further public comment on the economic impact of the rule on small mines. See id. at 18,779. The EPA stated that it would modify the rule if “significant additional data [were] presented to [it] on small placer mines during this comment period demonstrating that different effluent guidelines limitations and standards are warranted on a national basis.... ” Id. Following the close of the special comment period, the EPA published a notice stating that it had determined not to modify the rule and making available the record construing the data and analyses that the Agency had generated in response to the comments. See 54 Fed.Reg. 25, 28 (1989).
The Alaska Miners Association (“AMA”) and Stanley and Rosalie Rybachek timely petitioned this court for review of the EPA’s regulations.6 We ordered the petitions consolidated.
II
DISCUSSION
A. Standard of Review
Our review of the EPA’s regulations is governed by the Administrative Procedure Act (“APA”), the Clean Water Act, and the Constitution. Under the APA, we may set aside the EPA’s actions here if we find them to be “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law”; or “in excess of statutory jurisdiction, authority, or limitations, or short of statutory right”; or “without observance of procedure required by law.” 5 U.S.C. § 706(2)(A), (C), (D) (1988). Our function is to determine whether the Agency “has considered the relevant factors and articulated a rational connection between the facts found and the choice made.” Baltimore Gas & Elec. Co. v. Natural Resources Defense Council, Inc., 462 U.S. 87, 105, 103 S.Ct. 2246, 2256, 76 L.Ed.2d 437 (1983). We must base this determination on a “review [of] the whole record or those parts of it cited by a party, and due account shall be taken of the rule of prejudicial error.” 5 U.S.C. § 706(2) (1988).
Because the EPA has been charged with administering the Clean Water Act, we must show great deference to the Agency’s interpretation of the Act. See National Crushed Stone Ass’n, 449 U.S. at 83, 101 S.Ct. at 306 (court must show great deference to the interpretation given to a statute by the officers or agency charged with its administration). We especially defer where the Agency’s decision on the meaning or reach of the Clean Water Act involves reconciling conflicting policies committed to the Agency’s care and expertise under the Act. See Chevron U.S.A. Inc. v. Natural Resources Defense Coun*1285cil, Inc., 467 U.S. 837, 844, 104 S.Ct. 2778, 2782, 81 L.Ed.2d 694 (1984).7
B. The EPA’s Authority Under the Clean Water Act
The parties dispute whether placer mining is even subject to regulation under the Clean Water Act.
The Act charges the EPA with developing comprehensive programs aimed at “preventing, reducing, or eliminating the pollution of the navigable waters and ground waters and improving the sanitary condition of surface and underground waters.” 33 U.S.C. § 1252(a) (1982); see also E.I. du Pont de Nemours & Co. v. Train, 430 U.S. 112, 116, 97 S.Ct. 965, 969, 51 L.Ed.2d 204 (1977) (noting that Congress’s goal in enacting the Clean Water Act was to eliminate by 1985 all pollutant discharges into the nation’s waters). To assist the EPA in this program development, Congress made unlawful “the discharge of any pollutant by any person” except as in compliance with the Clean Water Act. 33 U.S.C. § 1311(a) (1982). It defined “discharge of a pollutant,” in part, as “any addition of any pollutant to navigable waters from any point source_” 33 U.S.C. § 1362(12) (1982).
The AMA seizes upon this statutory scheme to argue that placer mining is not subject to regulation under the Clean Water Act for at least two reasons: (1) placer mines do not discharge into “navigable waters”; and (2) placer mining does not “add” pollutants to water within the meaning of the Act.8 We reject both of these arguments and find that the EPA did not exceed the scope of its authority under the Clean Water Act in promulgating the challenged regulations.
First, the Clean Water Act covers the waters at issue here. Congress views broadly the words “navigable waters” in the Clean Water Act: it has defined them simply as “the waters of the United States, including the territorial seas.” 33 U.S.C. § 1362(7) (1982). Here, the parties agree that placer mines discharge into nearby streams and rivers. These are clearly among “the waters of the United States.”
Second, we will not strike down the EPA’s finding that placer mining discharges pollutants within the meaning of the Act. Placer miners excavate the dirt and gravel in and around waterways, extract any gold, and discharge the dirt and other non-gold material into the water.
On the one hand, if the material discharged is not from the streambed itself, but from the bank alongside, this is clearly the discharge into navigable waters of a pollutant under the Act. Congress defined “pollutant” as meaning, among other things, “dredged spoil ..., rock, sand, [and] cellar dirt....” 33 U.S.C. § 1362(6) (1982). The term “pollutant” thus encompasses the materials segregated from gold in placer mining. Congress defined “discharge” as any “addition [] to navigable waters from any point source.” 33 U.S.C. § 1362(12) (1982). Because, under this scenario, the material discharged is coming not from the streambed itself, but from outside it, this clearly constitutes an “addition.”
And on the other hand, even if the material discharged originally comes from the streambed itself, such resuspension may be interpreted to be an addition of a pollutant under the Act. See Avoyelles Sportsmen’s League, Inc. v. Marsh, 715 F.2d 897, 923 (5th Cir.1983) (stating that “[t]he word ‘addition,’ as used in the definition of the term ‘discharge,’ may reasonably be understood to include ‘redeposit’ ”), later proceeding, 786 F.2d 631 (5th Cir.1986) (concerning attorneys’ fees); United States v. M.C.C. of Florida, Inc., 772 F.2d 1501, 1506 (11th Cir.1985) (action of digging up sediment *1286and redepositing it on sea bottom by boat propellers constitutes an addition of pollutants), vacated and remanded on other grounds, 481 U.S. 1034, 107 S.Ct. 1968, 95 L.Ed.2d 809 (1987), readopted in part and remanded on other grounds, 848 F.2d 1133 (11th Cir.1988) (interpreting Supreme Court’s action as affecting only a different part of the original opinion), reh’g granted in other part, 863 F.2d 802 (11th Cir.1989). We will follow the lead of the Fifth and Eleventh Circuits and defer to the EPA’s interpretation of the word “addition” in the Clean Water Act. See Chevron U.S.A. Inc., 467 U.S. at 844, 104 S.Ct. at 2782; see also National Crushed Stone Ass’n, 449 U.S. at 83, 101 S.Ct. at 307 (stating that “this Court shows great deference to the interpretation given the statute by the officers or agency charged with its administration”) (quotation omitted).
In short, the EPA’s regulation of placer mining here was within its mandate from Congress and therefore not “in excess of statutory jurisdiction.” 5 U.S.C. § 706(2)(C) (1988).
C. Notice-and-Comment Procedures
Petitioners claim that the EPA’s notice- and-comment procedures violated their due process rights. The violations, according to petitioners, arose in several different forms. We consider each in turn.
1. Additional Documents in the Record
The Rybacheks allege that the EPA’s addition of over 6,000 pages to the administrative record, after the public review-and-comment period had ended, violated their right to comment on the record.
We disagree. The EPA has not violated the Rybacheks’ right to meaningful public participation. The additional material was the EPA’s response to comments made during a public-comment period.9 Nothing prohibits the Agency from adding supporting documentation for a final rule in response to public comments. In fact, adherence to the Rybacheks' view might result in the EPA’s never being able to issue a final rule capable of standing up to review: every time the Agency responded to public comments, such as those in this rulemak-ing, it would trigger a new comment period. Thus, either the comment period would continue in a never-ending circle, or, if the EPA chose not to respond to the last set of public comments, any final rule could be struck down for lack of support in the record. Cf. BASF Wyandotte Corp. v. Costle, 598 F.2d 637, 644-45 (1st Cir.1979) (noting that it is “perfectly predictable” that an administrative agency will collect new data during the comment period “in a continuing effort to give the regulations a more accurate foundation” and stating that “[t]he agency should be encouraged to use such information in its final calculations without thereby risking the requirement of a new comment period”), cert. denied, 444 U.S. 1096, 100 S.Ct. 1063, 62 L.Ed.2d 784 (1980), later proceeding, 614 F.2d 21 (1st Cir.1980) (dismissing petitions for review). The Rybacheks’ unviolated right was to comment on the proposed regulations, not to comment in a never-ending way on the EPA’s responses to their comments. Cf. Chemical Mfrs. Ass’n v. EPA, 870 F.2d 177, 200-02 (rejecting argument that agency was required to reopen notice-and-comment period before relying on economic data updated and expanded after the close of a comment period), clarified, 885 F.2d 253, later proceeding, 885 F.2d 1276 (5th Cir.1989) (concerning attorneys’ fees).
2. Best Management Practices
Petitioners also challenge the EPA’s promulgation of five best management practices (BMPs) to be included in NPDES permits. According to petitioners, the EPA failed to provide adequate notice of the BMPs contained in the final rule.10 They contend (1) that no actual wording for *1287the BMP part of the regulations was ever proposed, and (2) that only fourteen sentences in one Federal Register comprise the EPA’s total notice that it was considering promulgating BMPs, and these sentences implied that the EPA was considering requiring BMPs only on a case-by-case basis. We reject this challenge, but address each contention in turn.
The EPA’s failure to propose in advance the actual wording of the BMPs does not make the BMPs invalid. The EPA need not “publish in advance every precise proposal which it may ultimately adopt as a rule.” California Citizens Band Ass’n v. United States, 375 F.2d 43, 48 (9th Cir.), cert. denied, 389 U.S. 844, 88 S.Ct. 96, 19 L.Ed.2d 112 (1967). Instead, the EPA is only required to publish in this context the “terms or substance of the proposed rule or a description of the subjects and issues involved.” 5 U.S.C. § 553(b)(3) (1988).
The EPA’s discussions of the BMPs in both the original proposal and the second notice clearly describe “the subjects and issues” the BMPs involve. See 50 Fed. Reg. 47,982, 48,000 (1985) (original proposal); 52 Fed.Reg. 9,414, 9,416 (1987) (second notice of new information and request for comments). For instance, the EPA discussed in the preamble to the proposed rule and the draft Development Document its authority to impose BMPs under 33 U.S.C. § 1314(e) (1982). It also specifically discussed the possible BMPs that a permit-issuing authority could adopt in NPDES permits to minimize the amount of excess water entering the mine site, to divert water around the mine, and to prevent the release of settling pond sludge into receiving waters. See 50 Fed.Reg. 47,982, 48,000 (1985); EPA, Development Document for Proposed Effluent Limitations Guidelines and New Source Performance Standards for the Ore Mining and Dressing Point Source Category, Gold Placer Mine Subcategory, at VIII-34 to VIII-37 (1985), reprinted in R.A. 531, 532-35.11 Given these descriptions, the failure of the EPA to propose in advance the actual wording of the BMPs does not render them invalid.
We turn next to petitioners’ other two allegations: that the fourteen sentences in the original proposal which stated that the EPA was considering the BMPs are insufficient notice to the public, and that, unlike the final rule, the original proposal implied that the BMPs would be required on a case-by-case basis.12
First, the fourteen sentences about the BMPs comprise an entire section in the EPA’s original proposal. They were not difficult to find. And so long as enumeration is desired here, we note that these fourteen sentences contained more than 340 words. We will not strike down the EPA’s promulgation of BMPs merely on the grounds that this notice was short.
Instead, the crucial inquiry is whether this section gave adequate notice of the BMPs. Petitioners are correct that whereas the final rule mandated BMPs, the discussion in the original proposal and the second notice mentioned only the possibility that the EPA would “develop[] case-by-case BMP requirements for NPDES permits.” 50 Fed.Reg. 47,982, 48,000 (1985). Yet the fact that a final rule varies from a proposal, even substantially, does not automatically void the regulations. Rather, we must determine whether the inclusion of *1288the BMPs in the final rule was in character with the original proposal and a logical outgrowth of the notice and comments received. See, e.g., American Paper Inst. v. EPA, 660 F.2d 954, 959 n. 13 (4th Cir.1981) (noting that an agency may make substantial changes if they “are in character with the original proposal and are a logical outgrowth of the notice and comments already given”); Chemical Mfrs. Ass’n, 870 F.2d at 203 (upholding limitations not originally specifically proposed by the EPA because they were a logical outgrowth of notice and comments). Here, the EPA decided to include the BMPs in the final rule only after public comments strongly recommended adoption of mandatory BMPs for gold placer mines. See, e.g., Proposal Comment G-6, in 1 EPA, Summary Comments 1 Through 45 and Responses, at 76, reprinted in R.A. 217, 318. These comments emphasized the importance of BMPs to ensure cost-effective, environmentally sound mining operations. See, e.g., id. The EPA apparently found the comments persuasive, and it therefore included the BMPs in the final rule. Given all this, we conclude that the inclusion of BMPs in the final rule was both very much in character with the original proposal and a logical outgrowth of the notice and comments.
Informed changes and distinctions are the very raison d’etre of the notice-and-comment period. We conclude that the EPA provided sufficient notice with regard to the BMPs.
3. Adoption of BAT for All Mines
The AMA claims that the EPA’s determination in the final rule that recycling technology is the best available technology economically achievable (BAT) for all mines was sprung on the public without notice. The AMA argues that all of the public notices preceding the final rule indicated that recycling was not economically possible for small mines. The EPA’s sudden turnaround, as the AMA views it, deprived the public of the opportunity to comment on the feasibility of recycling for small mines.
We disagree with the AMA’s contention that the public was not on notice to comment as to whether recycling could be BAT for all mines. The EPA published a second notice of new information and request for comments on March 24, 1987. See 52 Fed. Reg. 9,414 (1987). There the EPA stated that its economic analyses indicated that “recycling of process wastewater is economically achievable for small open-cut mines processing between 1,500 and 70,000 yd3/year.” Id. at 9,423. The EPA then stated that “BAT limitations for these mines, therefore, would be zero discharge of process wastewater.” Id. at 9,423-24.13 The Agency also noted that it was “considering new limitations based on best practicable technology (BPT), best conventional technology (BCT), best available technology (BAT), and new source performance standards (NSPS),” and that it was soliciting comments on these possible changes. Id. at 9,414.
These statements conflict with the AMA's assertion that the EPA never indicated that it was contemplating adoption of recycling as BAT for small mines. Furthermore, the record demonstrates that during the spring and summer of 1987 the EPA received numerous comments on the changes proposed in this second notice of information; several of these addressed the question of whether recycling should be BAT for the placer mining industry. See, e.g., Summary Comment E-14, in EPA, Summary Comments 46 Through 130 and Responses, at 86, reprinted in R.A. 341, 465 (“Commenters supported the designation of complete recycling of wastewater as the ‘Best Available Technology’ for the placer mining industry in Alaska.”). Com-menters clearly recognized that the EPA was considering designating recycling as BAT for small mines.
In short, we conclude that the EPA provided sufficient notice with regard to the proposed adoption of recycling as BAT for small mines.
*1289D. The Final Rule
Petitioners make a host of arguments about the content of the final rule. For instance, they attack the EPA’s setting of BPT and BAT limitations. They also allege various errors by the EPA in its promulgation of BMPs and its enunciation of new-source criteria. We group most of these arguments into two broad categories: challenges to the merits of the limitations, and arguments concerning the EPA’s selection and presentation of data and its methods of analysis. We address each group in turn and then discuss the remaining challenges to the final rule.
1. Merits of the Limitations
Petitioners challenge the merits of the EPA’s regulations on a number of grounds; indeed, virtually every aspect of the regulations is attacked. To the extent the regulations may be divided into component parts (e.g., the BPT limitations, the BAT limitations, and new-source criteria), we address petitioners’ arguments along those lines.
a. Determination of BPT
We turn first to petitioners’ argument that the EPA erred in its determination that settling ponds are the best practicable control technology currently available (BPT) within the placer mining industry. There is no dispute that settling ponds are currently available pollution control technology; in fact, the AMA concedes that they are now used by almost all miners. Rather, petitioners contend that the EPA failed to use a “cost-benefit analysis” in determining that settling ponds were BPT for placer mining. They also argue that the EPA failed to consider costs when it set forth BPT limitations governing settleable solids for small mines.
The Clean Water Act controls when and how the EPA should require BPT. Under 33 U.S.C. § 1311(b)(1)(A), the Act requires “effluent limitations for point sources ... which shall require the application of the best practicable control technology currently available [BPT].” Under this section, the EPA is to determine whether a technology is BPT; the factors it considers “shall include ... total cost of” the technology “in relation to effluent reduction benefits to be achieved” from it, the age of equipment, engineering aspects, “non-water quality environmental impact ... and such other factors as the Administrator deems appropriate.” 33 U.S.C. § 1314(b)(1)(B) (1982).
From this statutory language, it is “plain that, as a general rule, the EPA is required to consider the costs and benefits of a proposed technology in its inquiry to determine the BPT.” Association of Pacific Fisheries v. EPA, 615 F.2d 794, 805 (9th Cir.1980). The EPA has broad discretion in weighing these competing factors. Id. It may determine that a technology is not BPT on the basis of this cost-benefit analysis only when the costs are “wholly disproportionate” to the potential effluent-reduction benefits. Id.; Chemical Mfrs. Ass’n, 870 F.2d at 205.
We look first to whether the EPA properly considered the costs of BPT and second to whether it properly weighed these costs against the benefits.
First, the record shows that the EPA properly considered costs in conducting the analysis which led to the determination that settling ponds are BPT and to the establishment of BPT effluent limitations for settleable solids. The EPA used a model-mine analysis to estimate the costs to mines of installing settling ponds. The Agency developed several model mines to represent the typical operating and compliance costs that open-cut mines and dredges of various sizes would incur.14 Commen-ters attempted to ensure that the model-mine analysis reflected actual industry conditions, and the EPA accordingly modified *1290the analysis when it thought it appropriate during the rulemaking.15 The EPA then determined, for each of its model mines, the incremental costs that would be incurred to construct and operate settling ponds to retain wastewater long enough to achieve a certain settleable solids level.16 It proceeded to conduct a detailed and complex assessment of the effect of the compliance costs on the mining industry’s profits.
The EPA then properly weighed these costs against the benefits of settling ponds. Its data indicated that placer mine waste-water contained high levels of solids and metals that were reduced substantially by simple settling. The upshot of the EPA’s analysis was its estimation that installation of settling ponds by open-cut mines industry-wide would remove over four billion pounds of solids at a cost of approximately $2.2 million — a removal cost of less than $1 per pound of solids. See EPA, Cost-Effectiveness Analysis of Effluent Limitations for the Gold Placer Mining Industry, Table 9 (1988), reprinted in R.A. 717, 736. The Agency therefore concluded that “the pollutant reduction benefits associated with compliance justify the costs for this subcategory.” 53 Fed.Reg. 18,764,18,772 (1988).
We uphold the EPA’s determination of BPT.
b. Determination of BAT
(1) Analysis of Costs
We next confront the AMA’s challenge to the EPA’s determination that recirculation of process wastewater is the best available technology economically achievable (BAT) in the placer mining industry. See 33 U.S.C. § 1311(b)(2)(A) (1982). By definition, BAT limitations must be both technologically available and economically achievable. 33 U.S.C. § 1314(b)(2)(B) (1982); see also Natural Resources Defense Council, Inc. v. EPA, 863 F.2d 1420, 1426 (9th Cir.1988). We conclude that the EPA’s BAT limitations were both and therefore uphold them.
The technological availability of recirculation of process wastewater is not in dispute; in fact, placer miners commonly practice it. It is recirculation’s economic achievability that petitioners challenge.
In determining the economic achievability of a technology, the EPA must consider the “cost” of meeting BAT limitations, but need not compare such cost with *1291the benefits of effluent reduction. See National Crushed Stone Ass’n, 449 U.S. at 71-72, 101 S.Ct. at 300-01; Association of Pacific Fisheries, 615 F.2d at 818. The Agency measures costs on a “reasonableness standard”; it has considerable discretion in weighing the technology’s costs, which are less-important factors than in setting BPT limitations. Natural Resources Defense Council, Inc., 863 F.2d at 1426; American Iron & Steel Inst. v. EPA, 526 F.2d 1027, 1052 n. 51 (1975) (“it is clear that for ‘BATEA’ [i.e., BAT] standards, cost was to be less important than for ‘BPCTCA’ [i.e., BPT] standards”), modified in other part, 560 F.2d 589 (3d Cir.1977), cert. denied, 435 U.S. 914, 98 S.Ct. 1467, 55 L.Ed.2d 505 (1978).
The record demonstrates that the EPA weighed the costs that recirculation would impose on gold placer mining. These include the costs for pumps to recirculate water from the settling ponds, as well as for piping, fuel, installation, and maintenance of equipment.17 Using various projections of gold prices, the Agency then calculated the number of mines that would be forced to close.18 Based on these figures, the EPA concluded that recirculation of process wastewater was economically achievable and therefore BAT.19 This conclusion was based on its data and can be supported by the data. In short, we find that the EPA “considered the relevant factors and articulated a rational connection between the facts found and the choice made.” Baltimore Gas & Elec. Co., 462 U.S. at 105, 103 S.Ct. at 2256.
(2) Total Suspended Solids Limitations
We come to petitioners’ claim that the EPA has impermissibly established BAT standards to regulate the discharge of total suspended solids. Petitioners argue that total suspended solids are conventional pollutants and therefore subject to BCT (best conventional pollution control technology), rather than BAT, standards. EPA’s adoption of recirculation as BAT to control total suspended solids was arbitrary, petitioners contend, because recirculation could not pass the cost-reasonableness test required in determining BCT.
The EPA declined to establish BCT for total suspended solids because test results indicated that settling technology could not consistently control the level of total suspended solids. Moreover, recirculation failed the BCT cost-reasonableness test.
Petitioners are incorrect in contending that the EPA instead adopted BAT to regulate the level of total suspended solids. The EPA’s discussion of BAT in the final rule makes no reference to controlling total suspended solids. Instead, the EPA set BAT standards to control the discharge of toxic pollutants — a category which, the parties agree, does not encompass total suspended solids. See 33 U.S.C. § 1314(a)(4) (1982) (designating suspended solids as a conventional pollutant). We therefore reject the contention that the EPA was arbitrary in establishing BAT standards.
(3) Settleable Solids Limitations
Petitioners also claim that settleable solids are a component of total suspended solids and that the EPA should have classified settleable solids as a conventional rather than a nonconventional pollutant. Petitioners contend that BAT-based effluent limitations are therefore inappropriate for settleable solids. We disagree.
In the Clean Water Act, Congress classified suspended solids as a conventional pollutant. 33 U.S.C. § 1314(a)(4) (1982). Congress did not classify settleable solids. We must determine, therefore, whether the EPA’s classification of settleable solids as a nonconventional pollutant “is based on a permissible construction of the [Clean Water Act].” Chevron U.S.A. Inc., 467 U.S. at 843, 104 S.Ct. at 2782. This court may *1292not substitute its own construction of the Act if the EPA’s interpretation is reasonable. Id. at 844, 104 S.Ct. at 2782.
The EPA argues that because settleable solids were not designated by Congress as either a conventional or a toxic pollutant, they should be considered a nonconventional pollutant under 83 U.S.C. § 1311(b)(2)(F) (1982 & Supp. V 1987).20 This argument is buttressed by the fact that the EPA has subjected settleable solids to BAT-level controls in other regulatory areas. See, e.g., 40 C.F.R. § 434.53(a) (1989) (governing post-mining drainage in coal mining). And even if settleable solids should more properly be considered a conventional pollutant, we note that the EPA has determined that settleable solids in placer mining effluent are a toxic pollutant indicator and thus may be subject to BAT-level limitations. See American Petroleum Inst. v. EPA, 858 F.2d 261, 262-63 n. 2 (1988), modified in other part, 864 F.2d 1156 (5th Cir.1989). We find, therefore, that the EPA’s decision to treat settleable solids as a nonconven-tional pollutant and thus subject to BAT standards was both reasonable and permissible. See Chevron U.S.A. Inc., 467 U.S. at 843, 104 S.Ct. at 2781.
(4) Presence of Toxic Pollutants
Finally, we turn to petitioners’ contention that any pollutants which are present in placer mine discharges may not be subjected to BAT limitations because the pollutants are in solid, natural form, not in their toxic form. We reject this challenge to the BAT limitations.
The record demonstrates that placer mine discharges are indeed toxic and that naturally occurring metals in treated wastewater cause environmental damage.21 Here, the EPA determined that the removal of solids in mine discharges is associated with substantial reduction of the concentration of all of the toxic metals in treated wastewater. Because this determination is substantially supported by the record, we uphold it.
c. New-Source Criteria
The AMA challenges the EPA’s criteria for determining what constitutes a “new source” for placer mines. According to the AMA, the EPA’s criteria are overly broad and contradictory to Congress’s intent. The AMA therefore urges this court to overturn the criteria for new-source determinations as arbitrary and capricious. Disagreeing with this characterization of the EPA’s criteria, we decline to do so.
The general definition of “new source” comes from the Clean Water Act. The Act defines “new source” as any source whose construction begins after the EPA publishes proposed regulations prescribing a performance standard applicable to it. 33 U.S.C. § 1316(a)(2) (1982). The Act broadly defines “construction” to mean “any placement, assembly, or installation of facilities or equipment (including contractual obligations to purchase such facilities or equipment) at the premises where such equipment will be used, including preparation work at such premises.” 33 U.S.C. § 1316(a)(5) (1982).
It devolves upon the EPA to apply these definitions to sources and to promulgate regulations establishing performance standards for new sources. See 33 U.S.C. § 1316(b)(1)(A), (B) (1982). The EPA thus *1293promulgated the following regulations for new-source placer mines:
(e) Notwithstanding any other provision of this chapter, the Regional Administrator or Director of a State agency with authority to administer the NPDES program shall in designating new source gold placer mines, take into account and base the decision on whether one or more of the following factors has occurred after May 24, 1988.
(1) The mine will operate outside of the permit area which is covered by a currently valid NPDES Permit.
(2) The mine significantly alters the nature or quantity of pollutants discharged.
(3) The mine discharges into a stream into which it has not discharged under its currently valid NPDES permit.
(4) The mine will operate in a permit area that has not been mined during the term of the currently valid NPDES permit.
(5) Such other factors as the Regional Administrator or state Director deems relevant.
40 C.F.R. § 440.144(c) (1989). The AMA contends that these criteria are inconsistent with the above-quoted congressional definitions.
The EPA has not acted arbitrarily or capriciously in setting forth these criteria for new-source placer mines. According to the AMA, these criteria make it likely that certain placer mines, which in the past have been mined periodically, will be considered new sources and therefore will require full review of their permits. That may be so, but the EPA might have chosen more-restrictive criteria that would have insured that all placer mines would be considered new sources every time the miners moved during the normal course of operations or reinstalled the mines each spring. The EPA instead chose a middle path and developed factors which the reviewing authority will “take into account” in determining whether a particular mine is a new source. This approach, which allows for an ad hoc, case-by-case application of the criteria, cannot fairly be said to be overbroad.
As a “particularly burdensome’ instance of arbitrariness or caprice, the AMA cites the EPA’s “last minute decision” to include in the new-source criteria the words “currently valid NPDES permit.” Petitioner AMA’s Opening Brief at 43. The AMA contends that these words, combined with the fact that placer mining permits run only one or two years, means that mines may be considered “new sources” if they go unworked for a season or if an application for a new permit for these mines is not made within six months of the expiration of an existing permit. Furthermore, according to the AMA, “[t]he new definition ... immediately creates a new source permit whenever the miner fails to get a permit for the previous season.” Id.
We disagree. The AMA loses sight of the fact that these are merely criteria to be considered in determining new sources. No one factor alone controls, no one factor automatically or immediately makes a mine a new source. See 53 Fed.Reg. 18,764, 18,776 (1988) (stating that meeting one criterion is “not intended to be conclusive in making a new source determination [] [but] [r]ather, the criteria are to be considered and taken into account ... in assessing all of the circumstances of a particular mine”).
We find that the criteria set forth on what is a new-source placer mine are within the ambit of authority which Congress gave the EPA in the Clean Water Act and cannot be considered arbitrary or capricious.
2. Selection, Presentation, and Analysis of Data
The AMA claims a host of factual inconsistencies and arbitrary assessments by the EPA. The arguments may be summarized thus: (1) certain of the EPA’s analyses and calculations were not scientifically sound or were unreasonable, inarticulate, or internally inconsistent, and (2) the analyses were presented to the public in such a way that the public had no meaningful opportunity to participate in the rulemaking process. The common base of the various *1294complaints is their concern with procedural fairness.
a. Processing of Data
The AMA claims that the EPA described the costs of recycling and settling ponds in an inconsistent and illogical manner, stated inconsistently the costs of running a bulldozer, distorted raw data, failed to consider storm events, failed to account for energy use, arbitrarily changed the description of a small mine, arbitrarily reduced costs of recycling, and wrongly estimated ore grades.
We first consider the AMA’s claim that the EPA inconsistently, arbitrarily, and illogically described the costs of recycling and settling ponds. As an example of inconsistency, the AMA notes that the cost calculations for ponds that appear at one point in the record do not match the construction costs in the Final Development Document. The costs in the Final Development Document do, however, correctly summarize the costs contained in the Final 1987 Costing Study. Other cost data cited by the AMA encompass only the costs of building ponds and not the additional costs of influent and effluent control materials, which the final cost estimates include. There is no inconsistency in the EPA’s description of costs.
As an instance of arbitrariness, the AMA asserts that the EPA used different pond dimensions and construction costs for the same model mine at two different times in the rulemaking process. The mines cited, however, are not the same. Because the model mine used in the EPA’s final cost analysis is twice the size of the mine used in the proposal, the EPA used pond sizes proportionate to the respective mine sizes.22 The EPA’s estimates and descriptions of the manner of building and the allocation of costs are both consistent and explicable. The AMA’s claim of illogicalness and arbitrariness thus fails.
The AMA also alleges that the EPA arbitrarily assumed different costs for the same piece of equipment (a particular bulldozer) in its compliance cost estimates and its baseline cost model analysis. Again, the reason for the difference in costs is easily explained. The compliance cost estimate consists of the hourly costs of leasing a new bulldozer, which include the cost of the equipment itself, as well as labor, maintenance, fuel, and insurance. See EPA, Final Development Document, at 163, reprinted in R.A. 29, 112. The baseline cost estimate is made up of only the direct costs of operating a used bulldozer (i.e., labor, energy, and supplies), and not the cost of the machine itself. See EPA, Final Economic Impact Analysis, app. A, at A-7, reprinted in R.A. 558, 692; see also id. at IV-9, reprinted in R.A. 558, 612 (including cost of bulldozer itself as an indirect spending cost). Such explicable and justifiable differences do not constitute arbitrariness on the part of the EPA.
One more instance of alleged arbitrariness or inconsistency may suffice. The AMA charges that the EPA’s cost calculations for ponds failed to take into account the costs of treating stormwater. The EPA, however, points to cost estimates for increased pond size volume for “freeboard” and says that this accounts for storm events.23 The Agency’s assertion that free-board volume accounts for precipitation is corroborated by its discussions elsewhere of the volume needed to retain stormwater. See, e.g., EPA, Final Development Document, at 141-42 (1988), reprinted in R.A. 29, 107-08; Kohlmann Ruggiero Engineers, Placer Mining Industry, 1987 Costing Study, Basic Costing Data (Pond Construction Costs for Small Open-cut Mine), reprinted in R.A. 773, 774 (third column of table, second line of calculations below ta*1295ble, concerning volume required for free-board).
We acknowledge that the EPA’s data may not be consistent to every decimal point. It is difficult to envision how any data that are accumulated, processed, commented upon, and revised, over a period stretching more than three-and-one-half years, could be entirely consistent. We are obligated, however, to take “due account ... of the rule of prejudicial error.” 5 U.S.C. § 706(2) (1988). We have made a great effort to immerse ourselves in the record in this case. Having done so, we find that any slight misstatements in some of the EPA’s calculations seem fairly understandable and that the misstatements are so minor that neither the AMA nor the Rybacheks could have been harmed by them. The prejudicial-error rule therefore applies. See Aero Mayflower Transit Co. v. ICC, 711 F.2d 224, 232 (D.C.Cir.1983) (prejudicial-error rule, mandated by the APA, means that court may not strike down agency action on basis of error unless petitioners have been harmed by error).
b. Presentation of Analyses
The AMA also alleges that the manner in which the EPA presented the data described above and the other data created an incomprehensible morass. As a result, the AMA alleges, the public did not have a meaningful opportunity to participate in the rulemaking process.
In support of this argument that the regulations must be struck down, the AMA cites Weyerhaeuser Co. v. Costle, 590 F.2d 1011 (D.C.Cir.1978). In Weyerhaeuser, however, the EPA admitted that it had used erroneous computations in its cost determinations. Id. at 1030. The EPA there could justify its rule only by referring to data which were outside the administrative record and which had been unavailable for public review and comment. Id. at 1030-31. Weyerhaeuser is inappo-site to this case because here the EPA can easily justify its rule by reference to the record.
It is certainly true that the data in the record here were often complex. But this is true of countless other administrative agency rulemakings. See, e.g., Chemical Mfrs. Ass’n v. EPA, 870 F.2d 177, clarified, 885 F.2d 253, later proceeding, 885 F.2d 1276 (5th Cir.1989) (considering challenges to EPA’s regulation of manufacturing plants in the organic chemicals, plastics, and synthetic fibers industries — a rule-making which involved an administrative record of 600,000 pages); Reynolds Metals Co. v. EPA, 760 F.2d 549 (4th Cir.1985) (rejecting challenge to effluent limitations promulgated by the EPA for the canmak-ing industry). We do not think that the complexity here has resulted in incomprehensibility. We find that the EPA’s presentation and analysis of the data were justifiable and coherent, and provided the public a meaningful opportunity to participate. We therefore reject this challenge to the EPA’s actions.
c. Falsified Evidence
We come to the AMA’s claim that the EPA intentionally and fraudulently distorted the processes and therefore the results of the Agency’s testing. According to the AMA, the EPA both falsified evidence showing that toxic metals are present in placer mine discharges and “cooked the data” by ignoring tests which would have shown its analysis to be faulty. We are unpersuaded and conclude that the EPA has neither acted arbitrarily or capriciously nor abused its discretion.
The AMA points out that when the EPA conducted tests to determine the amount of toxic pollutants that treatment could remove from placer mine discharges, the Agency assumed that tested water, in which toxic pollutants had been detected before but not after treatment, contained one-half the detectable level of toxic pollutants. In the AMA’s view, this assumption was a falsification of data and led to an inflated estimate of the amount of pollutants removed from the water.
The EPA’s assumption, however, was chosen as a middle course. When water which had contained detected pollutants be*1296fore testing contained no pollutants at the detectable level after testing, the EPA had to estimate how much pollutant remained below the detectable level. An estimate that the treated water contained no pollutants might have overstated the benefits of treatment; on the other hand, an estimate that the treated water contained a level of pollutants just under the detectable limit might have understated treatment benefits. We therefore find that the EPA’s estimate that toxic pollutants remained in the treated water at one-half the detectable level was reasonable and within the EPA’s “broad discretion in its selection of data and in the method of calculation.” Id. at 565.
Similarly, the AMA supports its “cooked data” allegation by pointing to facts which it claims show that the EPA arbitrarily disallowed a certain test result. According to the AMA, the disallowed test result showed that a particular mine’s wastewater contained less pollutant than had been in the mine’s intake-water supply. This inaccurately describes what the EPA disallowed in this specific instance. In actuality, the EPA did not include data from one mine when pollutant levels in the mine’s wastewater were higher after settling than they had been before treatment; measurement of intake water was not involved.24 When the EPA could find no explanation for the anomaly, it properly exercised its discretion and excluded that result.
Because the EPA discounted data which it reasonably believed skewed the test results, its action was not arbitrary. We therefore reject this challenge to the EPA’s testing procedure.
d. Deference to Other Agencies
The AMA charges that the EPA abused its discretion by failing to defer to federal agencies with special expertise in the realms of ore grades and profit margins for small mines. Specifically, the AMA argues that the expertise of the Bureau of Mines and of the Small Business Administration should be “controlling.”
The AMA has cited no authority for this extraordinary argument, and this court will not on its own compel one view of efficient agency administration. The prospect of various agencies contending in court, each claiming that another must defer to it in some particular area, is not a happy one. Resolution of such contentions, without some constitutional or statutory mandate, is not for this court.25
*1297 e. Consideration of Costs in Promulgating BMPs
We confront the AMA’s bald assertion that the “EPA never considered the cost of any one of the BMPs.... ” Petitioner AMA’s Opening Brief at 46. The assertion is erroneous; the record reflects that the EPA has met its duty to consider the BMPs’ costs. See, e.g., EPA Memorandum, Cost of Required Best Management Practices at Gold Placer Mines, at 1-S, reprinted in R.A. 795-97 (discussing the five BMPs and the costs of each “as a possible cost which would apply to operating costs of gold placer mines as a result of implementing the regulation for gold placer mines”). We therefore reject the allegation.
f Non-Water Quality Environmental Impacts
We turn to the petitioners’ contention that the EPA did not “take into account” the “non-water quality environmental impact (including energy requirements),” 33 U.S.C. § 1314(b)(1)(B), (b)(2)(B) (1982), in setting the effluent limitations based on settling and recycling. According to the petitioners, the EPA failed to comply with its statutory mandate by not truthfully considering environmental impacts other than water quality or by considering them only superficially. We find that the EPA has given adequate consideration to these impacts.
At the outset, we note that the EPA was not required to consider the impacts of some aspects of its regulations as petitioners contend. For instance, petitioners argue that the EPA failed to consider the environmental impact of plastic liners used to insulate settling ponds. But the EPA merely mentioned the use of these plastic liners as one option available to miners seeking to ensure the impermeability of their ponds; the EPA has not required the use of the liners. It therefore did not have to take into account their environmental impact in establishing the BPT and BAT limitations.
Furthermore, the record refutes petitioners’ other arguments that the EPA failed to consider or considered inadequately environmental impacts other than water quality. For instance, the AMA contends that the EPA inadequately considered additional energy requirements that would be needed for recirculation of process wastewater. Yet the record demonstrates that the EPA determined that compliance by the placer mining industry with BPT and BAT requirements would require a certain specified number of gallons of fuel per year; it further demonstrates how the EPA reached that determination and considered it in issuing its regulations. See 53 Fed. Reg. 18,764, 18,780 (1988); EPA, Final Development Document, at 171-72 (1988), reprinted in R.A. 29, 120-21. The EPA has thus satisfied its mandate to consider energy requirements necessitated by its rule. Cf. BASF Wyandotte, 598 F.2d at 662 (even a cursory discussion can satisfy the statutory mandate to take into account non-water quality environmental impacts).
The Clean Water Act leaves the EPA with broad discretion in deciding how non-water quality environmental impacts will be taken into account. Weyerhaeuser, 590 F.2d at 1049-53. We therefore conclude that the EPA did not abuse its discretion in its consideration of these impacts.
g. Hydraulic Mining Data
The Rybacheks claim that hydraulic mining should not be subject to regulation under the EPA’s new rule.26 Specifically, they contend that the EPA failed to study adequately the discharges associated with overburden removal through hydraulic processes and, therefore, failed to determine the appropriate effluent limitations for hydraulic mining. We reject the Ryba-cheks’ claim.
The EPA found that hydraulic mining operations are less efficient and more ex*1298pensive in comparison to mechanical mining operations (the use of mechanized earth-moving equipment). Further, the EPA found that because mechanical mining can be used effectively in all mining situations, miners considering the use of hydraulic mining would have to weigh the benefits of hydraulic mining against the more-economical mechanical mining approach. The availability of mechanical mining for all mines led the EPA to subject hydraulic mining to regulation under the new rule.
Furthermore, the EPA explained during the comment period that the final rule applies only to discharges resulting from the processing of ore and not the removal of overburden. See Response to Notice Comment F-4, in EPA, Summary Comments 46 Through 130 and Responses, at 98-99, reprinted in R.A. 341, 477-478. Discharges resulting from overburden removal would instead be regulated on a case-by-case basis through permit limitations. The EPA was not required, therefore, to study issues or consider data related to hydraulic overburden removal, which is different from hydraulic mining.
Accordingly, we find that the EPA was not arbitrary or capricious in including hydraulic mining within the regulations.
3. Mandating of Technology
The AMA next claims that by forbidding the discharge of any process waste-water, the EPA is mandating that placer miners use recirculation technology. According to the AMA, the EPA’s action violates Congress’s intent to avoid dictating technologies and to encourage innovation. While admitting that the wastewater flow standards are currently achievable only through certain technology, the EPA responds that the regulations only prescribe limitations reflecting actually achieved wastewater flow reductions. We agree with the EPA.
In the Clean Water Act, Congress expressed “the national goal that the discharge of pollutants into the navigable waters be eliminated by 1985.” 33 U.S.C. § 1251(a)(1) (1982); accord Natural Resources Defense Council, Inc v. EPA, 822 F.2d 104, 123 (D.C.Cir.1987). To accomplish this ambitious goal, Congress created a statutory scheme involving a “series of progressively more demanding technology-based standards” designed “not only to stimulate but to press development of new, more efficient and effective technologies.” Id. at 124. The AMA correctly notes that Congress sought to avoid requiring specific technologies and instead to encourage experimentation. See id. at 123.
The EPA has not mandated use of a particular technology. The Agency first determined that recirculation is BAT for the control of discharges by placer mines of toxic metals and settleable solids. Based on this determination, the EPA established that zero discharge of process wastewater is achievable and should be the BAT limitation and new-source performance standard. That the standards and limitations are stringent and currently may be achievable only through certain technology is true. However, nothing in the EPA’s regulations specifies the use of any particular technology to meet the BAT limitations and new-source performance standards achievable through recirculation. Cf. id. (rejecting the argument that particular EPA regulations “dictated] specific treatment technologies” and finding instead that they required “technology-based standards”).27 In fact, the EPA has encouraged miners to employ innovative technologies and to seek compliance extensions and alternative BAT limitations under section 301(k) of the Act. See 53 Fed.Reg. 18,764, 18,770 (1988); 33 U.S.C. § 1311(k) (Supp. V 1987). We find that the EPA’s setting of zero-discharge limitations based on recirculation results was within its mandate under the Clean Water Act.
4. Availability of Variances
Petitioners claim that the EPA has contravened Congress’s intent by failing to *1299allow miners to obtain variances for site-specific conditions. We first note that this assertion is flatly contradicted by the final rule’s express language allowing miners to apply for fundamentally different factor (“FDF”) variances for both the BPT and BAT limitations. See 40 C.F.R. § 125.30-.32 (1989). Petitioners’ claim of the unavailability of variances also ignores repeated statements by the EPA throughout the rulemaking. See, e.g., 50 Fed.Reg. 47,982, 48,001 (1985) (original proposal); Response to Proposal Comment F-6, in 1 EPA, Summary Comments 1 Through 45 and Responses, at 72, re-printed, in R.A. 217, 314 (“Finally, to the extent a mine does not have adequate space, the operator can apply for a ‘fundamentally different factors’ variance under the Act, which provides for variances subject to various substantive and procedural conditions.”); 53 Fed.Reg. 18,764, 18,786, 18,789 (1988) (final rule). Nevertheless, because we must determine for ourselves whether the EPA has properly handled the issue of variances, we address petitioners’ two main concerns on this issue.
Petitioners argue that the EPA’s classification of settleable solids as a toxic pollutant indicator will prevent miners from obtaining a variance. Normally, BAT limitations for nonconventional pollutants (here, settleable solids) are subject to modification under sections 301(c) and (g) of the Clean Water Act. See 33 U.S.C. § 1311(c), (g) (1982 & Supp. V 1987). In this instance, modifications for settleable solids under these provisions are unavailable because settleable solids are considered an indicator of toxic pollutants. See 33 U.S.C. § 1311(1) (Supp. V 1987). This does not mean, however, that no variance in the BAT limitations for settleable solids is available; miners may still apply for an FDF variance under section 301(n) of the Act. See 33 U.S.C. § 1311(n) (Supp. V 1987).
Petitioners further argue that the EPA’s criteria for “new sources” will result in all existing mines being reclassified as “new sources.” The mines will thus be ineligible to obtain variances for the BAT limitations, the argument proceeds, because new sources which are subject to NSPS are ineligible for variances. This argument assumes too much.
It is true that new mines may not obtain FDF variances for the new regulations. See 33 U.S.C. § 1311(n) (Supp. Y 1987) (prescribing factors for granting of FDF variances to existing mines). As discussed earlier, however, reclassification of existing mines as new sources is not automatic and will be done on a case-by-case basis by the permit-issuing authority only after consideration of the pertinent factors. Variances are still available under certain conditions for those mines or miners who do not merit new-source status. We therefore conclude that the EPA has adequately provided for variances under the regulations.
5. Timeliness of Rule Promulgation
The Rybacheks claim that even if the EPA can properly classify settleable solids as a nonconventional pollutant subject to BAT, it failed to promulgate the rules within 270 days as required under section 307(a)(2) of the Clean Water Act. See 33 U.S.C. § 1317(a)(2) (1982). Further, the Rybacheks argue that the EPA should have granted their request for a hearing under section 307. See id.
We reject both arguments for the same reason. The requirements of section 307 of the Clean Water Act apply only when the EPA promulgates health-based limitations for specific toxic pollutants pursuant to that section. See Environmental Defense Fund v. EPA, 598 F.2d 62, 72-73 (D.C.Cir.1978). Here, the EPA has promulgated technology-based effluent limitations pursuant to sections 304 and 306 of the Clean Water Act. See 33 U.S.C. §§ 1314, 1316 (1982 & Supp. V 1987). Therefore, the Rybacheks’ claims under section 307 of the Act must fail; the section is inapposite.
6. Time for Compliance
The AMA contends that Alaskan miners were not given sufficient time to comply with the EPA’s regulations. The AMA points out that the EPA promulgated the final rule on May 24, 1988, after most Alaskan miners were already in the field *1300for the season and therefore unable to learn of the rule and its details. According to the AMA, the miners had been advised that variances would be available and, moreover, the EPA’s decision not to amend the rule was not published until January 3, 1989. The AMA thus contends that enforcement of the regulations for the 1989 mining season was unreasonable.
Congress mandated that the BCT and BAT limitations must be achieved no later than March 31, 1989. 33 U.S.C. § 1311(b)(2)(C), (E) (1982 & Supp. V 1987). The EPA does not have the discretion to extend this deadline. See Chemical Mfrs. Ass’n, 870 F.2d at 242. This does not mean, however, that the EPA must be inflexible in enforcing the regulations. See id.
“Section 309 of the Act provides that, if a discharger fails to comply with a ‘final deadline,’ the [EPA] shall schedule a ‘reasonable’ time for compliance ‘taking into account the seriousness of the violation and any good[]faith efforts to comply with applicable requirements.’ ” Id. at 242 (quoting 33 U.S.C. § 1319(a)(5)(A) (1982)) (our emphasis and our emendation to conform to original statute). It was “Congress’ intention that good faith noneompliance [in appropriate instances] would be accommodated by the EPA’s post-deadline enforcement policy.” Id. Thus, while we acknowledge the miners’ concerns, we do not think that as a general matter enforcement of the new regulations for the 1989 mining season was unreasonable.
7. Constitutionality of the Final Rule
Finally, we turn to the Rybacheks’ argument that the regulations promulgated by the EPA constitute a taking under the fifth amendment. They seek just compensation for this alleged taking. There are at least two flaws in the Rybacheks’ argument.
First, no takings claim here is ripe for judicial resolution. A taking occurs in this context only when the EPA’s regulations are applied to particular property. See United States v. Vogler, 859 F.2d 638, 642 (9th Cir.1988) (“it is only after a permit has been denied and the denial’s effect prevents the ‘economically viable’ use of land that it can be said that a ‘taking’ occurred”) (citing United States v. Riverside Bayview Homes, Inc., 474 U.S. 121, 127, 106 S.Ct. 455, 459, 88 L.Ed.2d 419 (1985)), cert. denied, — U.S. —, 109 S.Ct. 787, 102 L.Ed.2d 779 (1989). Because the Ryba-cheks do not claim that a particular item of their property has been taken by the EPA’s final rule, a takings claim is unripe. See Hodel v. Virginia Surface Mining & Reclamation Ass’n, 452 U.S. 264, 293-97, 101 S.Ct. 2352, 2369-71, 69 L.Ed.2d 1 (1981) (claim of taking regarding implementation of surface-mining statutes and regulations is premature until statutes and regulations are actually applied to particular property of which a taking is claimed); see also Trustees for Alaska v. EPA, 749 F.2d 549, 560 (9th Cir.1984) (a justiciable takings claim presumes a “concrete controversy concerning the application of the permit conditions to a[] particular operation[] or to a[] particular estimate[] of economic impact and ultimate valuation”).
Second, ours is an inappropriate forum for such a takings claim. Our jurisdiction over the challenge to EPA regulations comes from a particular statutory grant. See 33 U.S.C. § 1369(b)(1) (Supp. V 1987) (“Review of the Administrator’s action [in promulgating regulations or in taking certain actions under the Clean Water Act] may be had by any interested person in the Circuit Court of Appeals of the United States for the Federal judicial district in which such person resides or transacts business which is directly affected by such action upon application by such person.”). Congress has specified other fora for resolution of claims against the United States that seek just compensation for a taking. See 28 U.S.C. § 1491(a)(1) (1982) (granting jurisdiction of certain takings claims to the Claims Court); 28 U.S.C. § 1346(a)(2) (1982) (“[t]he district courts shall have original jurisdiction, concurrent with the United States Claims ■ Court, of” certain claims against the United States, including those for certain takings). Thus, our authority for hearing the petitioners’ challenges to *1301the EPA’s regulations does not suffice to hear a takings claim.28
Ill
CONCLUSION
The EPA regulations contested here are not the most rousing words ever written about gold mining. Consider, for instance, the excitement missing from the technical discussion above:
Gold! We leapt from our benches! Gold! We sprang from our stools.
Gold! We wheeled in the furrow, fired with the faith of fools.
Fearless, unfound, unfitted, far from the night and the cold,
Heard we the clarion summons, followed the master-lure — Gold!
Men from the sands of the Sunland; men from the woods of the West;
Men from the farms and the cities, into the Northland we pressed.
Graybeards and striplings and women, good men and bad men and bold,
Leaving our homes and our loved ones, crying exultantly — “Gold!”
R. Service, The Trail of Ninety-Eight, in Collected Poems of Robert Service 144 (1940).
Gold! The trail of the 1990’s may be burdened by twentieth century laws and regulations, but the promise Service sings still awaits those who persevere.
PETITIONS FOR REVIEW DENIED.
. R. Service, The Cremation of Sam McGee, in Collected Poems of Robert Service 33 (1940). Regardless of what Service may have indicated, women moil for gold as well; one of the petitioners, Rosalie Rybachek, has articulately presented her own cause before this court.
. We shall refer to the entire statutory scheme found in 33 U.S.C. §§ 1251-1387 as the "Clean Water Act.”
. For purposes of the EPA’s regulations affecting placer mining, the term "mine” means "a place where work or other activity related to the extraction or recovery of ore is performed.” 40 C.F.R. § 440.141(a)(8) (1989). Compare id. with 40 C.F.R. § 440.132(g) (1989) (defining "mine” generally for regulations affecting other types of ore mining). As a colloquial matter, placer mining frequently occurs by the simple use of crude equipment on the bank of a stream. Because of the unique nature of placer mining, the activity ordinarily does not involve the large holes in the ground with which we usually associate the term "mine.” See, e.g., Webster's Third New International Dictionary 1437 (1986) (giving first definition of "mine" (n.) as "a pit or excavation in the earth from which mineral substances (as ores, precious stones, or coal) are taken by digging or by some other method of extraction”).
. Simple settling involves the retention of waste-water in ponds to allow solid materials to settle before the water is discharged.
. Recirculation, or recycling, employs pumps to route water from the settling ponds back to the sluice for reuse in the gold recovery process.
.We have jurisdiction under 33 U.S.C. § 1369(b)(1) (Supp. V 1987).
. The instant controversy does not compel us to become involved in the debate concerning "whether [Chevron ] applies with full force ... when the controversy involves a pure question of statutory construction.” Scalia, Judicial Deference to Administrative Interpretations of Law, 1989 Duke LJ. 511, 512 (quotation omitted).
. The AMA does not contend that placer mining does not involve a “point source.” See 33 U.S.C. § 1362(14) (Supp. V 1987) ("The term ‘point source’ means any discernible, confined and discrete conveyance ... from which pollutants are or may be discharged.”).
. The comment period followed the EPA’s publication on March 24, 1987 of a second notice of new information and request for comments. See 52 Fed.Reg. 9,414 (1987).
. Petitioners also argue that the BMPs set forth by the EPA are flawed because the EPA failed to consider the costs of the BMPs. Because this argument is not a challenge to the sufficiency of notice and comment, we consider it below. See infra p. 1297.
. As we noted earlier, in conducting our review of the EPA’s action we are under a statutory duty to “review the whole record or those parts of it cited by a party.” 5 U.S.C. § 706(2) (1988). Not surprisingly, many of the disagreements among the parties concern what the record does and does not contain or demonstrate. We therefore do not content ourselves with mere assertions of what the record contains or demonstrates; where appropriate, we will cite to the record. When we cite to certain documents that are not easy to find, such as those not published in the Federal Register, we will, where possible, identify its location in the Respondent’s Appendix ("R.A.”). Citation to the Respondent’s Appendix is especially appropriate both because of the parties’ easy access to this Appendix and because the Appendix is a public record.
. The EPA published its original proposal on November 20, 1985. See 50 Fed.Reg. 47,982 (1985). This contained the fourteen sentences to which petitioners refer. In its second notice of new information and request for comment, the EPA again included an entire section on BMPs. See 52 Fed.Reg. 9,414, 9,416 (1987).
. These considerations were also highlighted in the EPA’s tables summarizing the possible revisions to the proposed regulations. See 52 Fed. Reg. 9,414, 9,421 (1987).
. See EPA, Economic Impact Analysis of Final Effluent Limitations and Guidelines for the Gold Placer Mining Industry, at VI-1 to VI-6 [hereinafter "Final Economic Impact Analysis”], reprinted in _R.A. 558, 635-640; EPA, Development Documeht for Final Effluent Limitations Guidelines and New Source Performance Standards for the Ore Mining and Dressing Point Source Category, Gold Placer Mine Subcategory, at 168-69 (1988) [hereinafter "Final Development Document”], reprinted in R.A. 29, 117-18.
. We cannot agree with the AMA’s claim that the EPA "did not study actual expenses at any real mines." Petitioner AMA’s Opening Brief at 18. The EPA based the model-mine analysis on data collected directly from miners and on costs quotations from equipment manufacturers, transportation companies, and other sources. See EPA, Final Economic Impact Analysis, at IV-2 to IV-9, reprinted in R.A. 558, 605-612; Response to Proposal Comment A-l, in 1 EPA, Summary Comments 1 Through 45 and Responses, reprinted in R.A. 217, 243-45.
. The EPA determined that the settleable solids level achievable with settling ponds was 0.2 ml/1. The Rybacheks assert that the EPA has inadequate support for this determination and was required to set instead a higher level. We disagree.
Between 1983 and 1986, the EPA took 73 samples of full-scale settling ponds at 39 different gold placer mines; 50 of these samples indicated levels of settleable solids at or below 0.2 ml/1 after treatment. The EPA determined that many of the ponds not achieving this level had discernible design or operating deficiencies. See 53 Fed.Reg. 18,764, 18,768-69 (1988); Kohl-mann Ruggiero Engineers, Summary of Analytical Data from the 1983, 1984, 1985 and 1986 Alaskan Placer Mining Studies (Preliminary Draft), Table 4 (1987), reprinted in R.A. 546, 549-51; [EPA] Memorandum, Summary of In-place Treatment at Alaska Gold Placer Mines, at 1-3 (May 9, 1988), reprinted in R.A. 554, 554-556 (summarizing and explaining data from Table 4).
Depending on the position they seek to support, the parties here variously point to miners during the 1984 season who were or were not achieving the 0.2 ml/1 level of settleable solids. Based on our review of the record, we conclude that the EPA reasonably determined that the level of 0.2 ml/1 represented the "average of the best" mines. See American Frozen Food Inst. v. Train, 539 F.2d 107, 117 (D.C.Cir.1976) (BPT limitations should represent the "average of the best” existing performances of plants in the industrial category); American Meat Inst. v. EPA, 526 F.2d 442, 453 (7th Cir.1975) ("The ‘best practicable technology’ will normally be defined based on the average performance of the best existing plants.’’). Both because, as discussed in the text, the proper factors were considered, and because, as discussed in this note, the EPA’s determination of the 0.2 ml/1 level is supportable by the record, we uphold the EPA’s determination that settling ponds were BPT.
.See EPA, Final Development Document, at 162-70, reprinted in R.A. 29, 111-19; Kohlmann Ruggiero Engineers, 1987 Costing Study on Systems to Treat Wastewater Discharges in the Placer Mining Industry (1987), reprinted in R.A. 768, 771-72.
. See EPA, Final Economic Impact Analysis, at VII-14, VII-15, VII-19, VII-20, reprinted in R.A. 558, 671, 672, 676, 677.
. See id. at VII-20, reprinted in R.A. 558, 677.
. Section 1311(b)(2)(F) provides, in substance, that all pollutants not classified as toxic or conventional pollutants are subject to BAT limitations.
. See, e.g., Versar, Inc., Environmental Assessment of Discharges of Wastewater from the Placer Mining Subcategory of the Ore Mining and Dressing Industry (Revised Draft), at i-ii, 13, 18 (1987), reprinted in R.A. 145-148; Response to Notice Comment in K-8, in EPA, Summary Comments 46 Through 130 and Responses, at 135, reprinted in R.A. 341, 514.
Petitioners’ reliance on a study determining that the sludge in settling ponds would not be classified as hazardous solid waste under the Resource Conservation and Recovery Act, 42 U.S.C.A. §§ 6901-6992 (West 1983 & Supp. 1989), is misplaced. That study drew no conclusions regarding the toxicity of mine discharges in receiving waters. See EPA Memorandum, Toxicity Characteristic Contaminants and Regulatory Levels Analysis of Sludge, reprinted in R.A. 793-94; 53 Fed.Reg. 18,764, 18,-779-80 (1988).
. Compare, e.g., EPA, Final Development Document, at 169, reprinted in R.A. 29, 118 (describing model mine processing 340,000 cubic yards of paydirt annually) with, e.g., EPA, Development Document for Proposed Effluent Limitations Guidelines and New Source Performance Standards for the Ore Mining and Dressing Point Source Category, Gold Placer Mine Subcategory, at IX-14 (1985), reprinted in R.A. 531, 538 (proposal example describing model mine processing 153,000 cubic yards of paydirt annually).
. Freeboard is the portion of pond volume that exceeds the pond’s design capacity.
. See EPA Memorandum, Statistical Analysis of Metals Concentrations in Gold Placer Mining Effluent, at 2, reprinted in R.A. 699, 700.
. On January 30, 1989, the AMA filed in this court a motion to take judicial notice of seven documents, five of which had been submitted to the EPA during the special comment period. The documents contain reports, analyses, and comments by other federal agencies and the AMA criticizing the placer mining regulations. On February 3, 1989, this court ordered the motion construed as a motion to supplement the record and referred it to this panel for disposition.
We deny the AMA's motion. Judicial review of agency actions should generally be confined to the original record upon which the actions were based. Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402, 420, 91 S.Ct. 814, 825, 28 L.Ed.2d 136 (1971); Love v. Thomas, 858 F.2d 1347, 1356 (9th Cir.1988), cert. denied, - U.S. -, 109 S.Ct. 1932, 104 L.Ed.2d 403 (1989). The reviewing court may consider information supplemental to the record only exceptionally: for instance, if the information is necessary as background "to explain the basis of the agency’s action and the factors the agency considered." Id. In short, it is not "appropriate ... for either party to use post-decision information as a new rationalization either for sustaining or attacking the Agency’s decision.” Association of Pacific Fisheries, 615 F.2d at 811-12.
The original record here adequately explains the basis of the EPA’s placer mining regulations and demonstrates that the EPA considered the relevant factors. The EPA did not rely on the information contained in the AMA’s proffered documents, nor does the information address issues not already present in the record. See Friends of the Earth v. Hintz, 800 F.2d 822, 829 (9th Cir.1986) (discussing "exceptions ... to the rule that review of agency action is limited to the administrative record"). The AMA therefore cannot use the information, as it appears to be doing, to attack the regulations.
On the same principles, we grant both the EPA’s motion to strike portions of the AMA’s and Rybacheks’ briefs, and the AMA’s motion to strike portions of intervenor Trustees for Alaska’s brief, all of which rely on information outside the administrative record. Further, we grant the EPA’s motion to supplement the *1297record because the documents offered are part of the original administrative record and were relied upon by the Agency.
. Hydraulic mining involves the use of highly pressurized water to break up the ore and transport it to the sluice box.
. The EPA’s authority to impose zero-discharge limitations based on recirculation technology has been upheld by other courts. See, e.g., Kennecott Copper Corp. v. EPA, 612 F.2d 1232, 1244 (10th Cir.1979) (upholding zero-discharge limitation achievable by recycling technology for gold and silver cyanidation mills).
. We have discussed at length and rejected with particularity virtually all of both the AMA’s and the Rybacheks’ multitudinous challenges to the EPA's regulation of placer mining. Finding such level of detail inappropriate for the remaining allegations by petitioners, we also reject them, variously for failure to raise them below, for lack of merit, and for want of prejudicial error.
For instance, the Rybacheks argue that the EPA may not regulate their mining operations because to do so would conflict with their water rights that are protected by federal law. See 30 U.S.C. § 51 (1982) ("Whenever, by priority of possession, rights to the use of water for mining, agricultural, manufacturing, or other purposes have vested and accrued, and the same are recognized and acknowledged by the local customs, laws, and the decisions of courts, the possessors and owners of such vested rights shall be maintained and protected in the same_”). The Rybacheks do not dispute Congress’s power to affect rights protected under 30 U.S.C. § 51, but contend that, if it had intended to do so, Congress would have spoken more clearly than it has in its statutes controlling the EPA.
We do not pass on the question of whether the Rybacheks have rights protected under 30 U.S.C. § 51. For if they do not, then the claim that Congress did not intend, in passing the Clean Water Act, to affect section 51 is irrelevant. And if they do, we nonetheless think it clear that Congress has spoken with easily sufficient clarity to affect these rights. The Clean Water Act may be varia variis, but none can contend either that, in passing it, Congress did not intend to affect significantly the status quo or that Congress did not make this intention clear. Cf. National Crushed Stone Ass’n, 449 U.S. at 79, 101 S.Ct. at 304 (Court noted that legislative history "amply supported" argument that "Congress foresaw and accepted the economic hardship, including the closing of some plants, that effluent limitations would cause”).