Full opinion text
MEMORANDUM AND ORDER MILTON L. SCHWARTZ, District Judge. These consolidated cost recovery actions arise under the Comprehensive Environmental Response, Compensation, and Liability Act of 1980 (“CERCLA”), as amended by the Superfund Amendments and Reauthorization Act of 1986 (“SARA”), 42 U.S.C. § 9601 et seq. The actions concern Iron Mountain Mine, a parcel of land northwest of Redding, California, that for 100 years was mined for iron, zinc, copper, silver, gold, and pyrite ores. According to plaintiffs, intensive mining has fractured the land, causing severe acid mine drainage (“AMD”) that poses a threat to the environment. I. Procedural Background. On August 11, 1992, the court heard plaintiff United States of America’s motion for partial summary judgment on defenses, plaintiff State of California’s motion to strike defenses of defendant Rhóne-Pou-lenc Basic Chemicals Co. (“RP”), and plaintiff State of California’s motion to strike defenses of defendants Iron Mountain Mines, Inc., and T.W. Arman (“IMMI/Ar-man”). David B. Glazer appeared on behalf of plaintiff United States, Lisa Trank-ley Sato appeared on behalf of plaintiff State of California, Paul B. Galvani appeared on behalf of defendant RP, and Edward S. Atkinson appeared on behalf of defendants IMMI/Arman. The defenses subject to plaintiffs’ motions are enumerated below as they are enumerated in defendants’ answers and each defense is fully described in the text of this memorandum. For convenience, the defenses are cited herein as follows: for example, “RP1-US” denotes the first defense of defendant Rhóne-Poulenc Basic Chemicals Co. in the United States case; “IMMI5-CA” denotes the fifth defense of defendants Iron Mountain Mines, Inc., and T.W. Arman in the California case. Rulings on the motions as to particular defenses employ the same citation form. One procedural nicety must be mentioned. Although California moves to strike particular defenses (see Fed.R.Civ.P. 12(f)), the United States seeks summary adjudication of many of the same defenses (see Fed.R.Civ.P. 56). The United States has also joined in California’s motion to strike. Both motions raise similar or identical legal issues and, with some exceptions, are directed to the same defenses. Some defenses may be appropriately stricken as to both plaintiffs. But, because California has not sought summary adjudication, summary judgment will not be awarded in California’s favor on any defense. II. Substantive Background. Plaintiffs’ motions arise from CERCLA’s scheme of quasi-strict liability. See, e.g., United States v. Aceto Agric. Chems. Corp., 872 F.2d 1373, 1377 (8th Cir.1989) (“Most courts have held CERCLA imposes strict liability”); United States v. Monsanto Co., 858 F.2d 160, 167 (4th Cir.1988) (“We agree with the overwhelming body of precedent that has interpreted section 107(a) as establishing a strict liability scheme”), cert. denied, 490 U.S. 1106, 109 S.Ct. 3156, 104 L.Ed.2d 1019 (1989); United States v. Northeastern Pharmaceutical & Chem. Co., 810 F.2d 726, 732 n. 3 (8th Cir.1986) (“NEPACCO ”) (“Most cases have imposed strict liability”), cert. denied, 484 U.S. 848, 108 S.Ct. 146, 98 L.Ed.2d 102 (1987). CERCLA section 107(a)(4)(A) holds certain persons liable for “all costs of removal or remedial action incurred by the United States government or a State” so long as those costs are “not inconsistent with the national contingency plan [“NCP”].” 42 U.S.C. § 9607(a)(4)(A). Persons liable include “the owner and operator of a vessel or a facility” and “any person who at the time of disposal of any hazardous substance owned or operated any facility at which such hazardous substances were disposed of.” 42 U.S.C. § 9607(a)(1) and (2). The first sentence of section 107 states that CERCLA liability is “subject only to the defenses set forth in subsection (b) of this section.” Id. Only three defenses are listed in section 107(b): (1) “act of God”; (2) “act of war”; and (3) “act or omission of a third party other than an employee or agent of the defendant....” 42 U.S.C. § 9607(b). A fourth provision states that “any combination of the foregoing” defenses will also establish no liability- III. Standards of Review. A. Summary Judgment under Rule 56. Federal Rule of Civil Procedure 56(c) provides that summary judgment is appropriate when the moving party shows that “there is no genuine issue as to any material fact and ... the moving party is entitled to judgment as a matter of law.” Federal Rule of Civil Procedure 56(c); see also Adickes v. S.H. Kress & Co., 398 U.S. 144, 157, 90 S.Ct. 1598, 1608, 26 L.Ed.2d 142 (1970); Jung v. F.M.C. Corp., 755 F.2d 708, 710 (9th Cir.1985). In summary judgment practice, the moving party always bears the initial responsibility of informing the district court of the basis for its motion, and identifying those portions of “the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any,” which it believes demonstrate the absence of a genuine issue of material fact. Celotex Corp. v. Catrett, 477 U.S. 317, 323, 106 S.Ct. 2548, 2553, 91 L.Ed.2d 265 (1986). “[Wjhere the nonmoving party will bear the burden of proof at trial on a dispositive issue, a summary judgment motion may properly be made in reliance solely on the ‘pleadings, depositions, answers to interrogatories, and admissions on file.’ ” Id. at 324, 106 S.Ct. at 2553. Indeed, summary judgment should be entered, “after adequate time for discovery and upon motion, against a party who fails to make a showing sufficient to establish the existence of an element essential to that party’s case, and on which that party will bear the burden of proof at trial.” Id. at 322, 106 S.Ct. at 2552. “[A] complete failure of proof concerning an essential element of the non-moving party’s case necessarily renders all other facts immaterial.” Id. at 323, 106 S.Ct. at 2552. In such a circumstance, summary judgment should be granted, “so long as whatever is before the district court demonstrates that the standard for entry of summary judgment, as set forth in Rule 56(c), is satisfied.” Id. at 323, 106 S.Ct. at 2553. If the moving party meets its initial responsibility, the burden shifts to the opposing party to establish that a genuine issue of material fact actually exists. Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 585-86, 106 S.Ct. 1348, 1355, 89 L.Ed.2d 538 (1986). In attempting to establish that a factual dispute exists, the opposing party may not rely upon its pleading denials, but must tender evidence of specific facts in the form of affidavits or admissible discovery material, or both, in support of its contention that a dispute exists. Rule 56(e); Matsushita, supra, 475 U.S. at 586-87, 106 S.Ct. at 1356. The opposing party must demonstrate that the fact in contention is material, i.e., a fact that might affect the outcome of the suit under the governing law and that the dispute is genuine, i.e., the evidence is such that a reasonable jury could return a verdict for the nonmoving party. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 247-48, 106 S.Ct. 2505, 2509-10, 91 L.Ed.2d 202 (1986). B. Motion to Strike under Rule 12(f). Federal Rule of Civil Procedure 12(f) provides that, on motion of either party or of the court, the court may order stricken from any pleading any insufficient defense. While a party’s motion to strike must be made before the party files a responsive pleading, the court acting on its own initiative may order material stricken at any time. Thus, where the motion to strike is untimely, the court may yet strike an insufficient defense. 5A C. Wright and A. Miller, Federal Practice and Procedure § 1880 (2d ed. 1990). “All well-pleaded facts are taken as admitted on a motion to strike.” Kelly v. Kosuga, 358 U.S. 516, 79 S.Ct. 429, 3 L.Ed.2d 475 (1959). The motion to strike helps avoid a waste of resources by eliminating spurious issues before trial. Sidney v. A.H. Robins Co., 697 F.2d 880, 885 (9th Cir.1983). However, the motion is not favored by courts because it is thought to be dilatory. United States v. 416.81 Acres of Land, 514 F.2d 627, 631 (7th Cir.1975). As to insufficient defenses, the prevailing view is that the district court has broad power and can strike any insufficient defense, including constitutional ones. Id. at 630 n. 3. However, a motion to strike an affirmative defense should only be granted if the affirmative defense is insufficient as a matter of law. See Memorex Corp. v. International Business Mach. Corp., 555 F.2d 1379 (9th Cir.1977). IV. Analysis. Because of the breadth of plaintiffs’ motions, the following analysis is largely organized by topic. A. Has Plaintiff Stated a Claim under CERCLA? Plaintiff United States first moves for summary judgment on RP1-US and IMMI1-US, in which defendants assert that plaintiff has failed to state a claim upon which relief can be granted. Plaintiff California moves to strike IMMI1-CA, which alleges that California has also failed to state a claim for relief. Each plaintiffs complaint endeavors to state a single claim for relief seeking recovery of past and future response costs under CERCLA section 107, 42 U.S.C. § 9607. Plaintiffs premise defendant RP’s liability on its alleged status as successor in interest to liabilities of Mountain Copper, Ltd., who is said to have mined the site for several decades. Defendants IMMI/Ar-man’s liability is premised on their alleged status as present owner of the site. The Ninth Circuit Court of Appeals has written that To state a claim under 42 U.S.C. § 9607(a) [i.e., a CERCLA cost recovery claim], a plaintiff must allege that (1) the waste disposal site is a “facility” within the meaning of 42 U.S.C. § 9601(9); (2) a “release” or “threatened release” of any “hazardous substance” from the facility has occurred, id. § 9607(a)(4); and (3) such “release” or “threatened release” has caused the plaintiff to incur response costs that are “consistent with the national contingency plan,” id. §§ 9607(a)(4) & (a)(4)(B). [Citations.] In addition, the defendant must fall within one of four classes of persons subject to CERCLA’s liability provisions: (1) the owner and operator of a vessel or a facility, (2) any person who at the time of disposal of any hazardous substance owned or operated any facility at which such hazardous substances were disposed of.... 42 U.S.C. § 9607(a)(l)-(4). Ascon Properties, Inc. v. Mobil Oil Co., 866 F.2d 1149, 1152-1153 (9th Cir.1989). The familiar defense of failure to state a claim “test[s] the formal sufficiency of the statement of the claim for relief; it is not a procedure for resolving a contest about the facts or the merits of the case.” 5A Federal Practice and Procedure, supra § 1356. The defense must be read in conjunction with Federal Rule of Civil Procedure 8(a), which requires “a short and plain statement of the claim showing that the pleader is entitled to relief.” Id. There is no real question that plaintiffs have stated formally sufficient claims for relief under the applicable law. Plaintiff United States’ complaint alleges (1) that the “site is a ‘facility’ within the meaning of Section 101(9) of CERCLA” (U.S. Compl. at H 25); (2) that defendants IMMI/Arman are current owners or operators of the mine (id. at ¶ 26); (3) that defendant RP owned or operated the mine “during the time in which hazardous substances were disposed of” (id. at ¶ 27); (4) that the substances disposed of “are ‘hazardous substances’ within the meaning of Section 101(14)" (id. at if 28); (5) that " 'releases’ or ‘threatened releases’ of hazardous substances at or from the site” have occurred (id. at fl 29); (6) that “[t]he releases or threatened releases ... have caused the United States to incur ... ‘response’ costs within the meaning of Section 101(25) of CERCLA” (id. at ¶ 30); (7) that each defendant is a person liable under section 107 (id. at ¶ 31); (8) that such persons are liable for all removal and remedial costs not inconsistent with the NCP (id. at II 32); and (9) that costs incurred were not inconsistent with the NCP (id. at ¶ 33). This pleading exceeds the requirements imposed by the Ninth Circuit in Ascon Properties, supra, 866 F.2d at 1152-1153. Plaintiff California’s pleading is similarly complete. Accordingly, plaintiffs’ motions regarding RP1-US, IMMI1-US, and IMMI1-CA will be granted. Defendant RP also insists that relief cannot be granted on plaintiffs’ claims for four separate reasons listed in RP2-US and RP2-CA. Each reason is considered immediately below. Defendants IMMI/Arman join in that section of RP’s opposition assailing the availability of relief. The United States has moved for summary judgment on all four of the issues raised in RP2-US and California has moved to strike three of the four allegations made in RP2-CA. 1. Releases “at” Rather than “from” a Facility. Plaintiffs’ complaints allege that hazardous substances have been released “at or from” Iron Mountain Mine. According to RP, the usage “at or from” expands CERCLA’s reach, which is limited to cases where wastes have been released or threaten to be released “from” a facility. Thus, in RP1-US, RP2-US, and RP2-CA, RP alleges that the complaint fails to state a claim insofar as plaintiff seeks to recover costs for responding to a release “at” property now owned by IMMI/Arman near Redding, California. Cases such as Ascon Properties, supra, repeat the statutory language regarding facilities “from which there is a release” or threatened release, suggesting that, to be liable, a person must own or have owned or operated a facility from which a hazardous substance has been or threatens to be released. 42 U.S.C. § 9607(a) (emphasis added). The question to be answered here is whether a release “at” a facility — meaning a release that never leaves the facility — is a covered release. Plaintiffs argue that CERCLA section 104(a), 42 U.S.C. § 9604(a)(1), which gives the President authority to remedy any release of wastes “into the environment,” should be read in conjunction with section 101(8), 42 U.S.C. § 9601(8), which defines “environment” as (in part) “surface water, ground water, ... land surface or subsurface strata ... within the United States.... ” Such a reading, say plaintiffs, establishes liability for releases at a facility so long as the facility itself fits the statutory definition of environment. RP complains that such a result would impose liability even if acidic water never left the mountain because the mountain itself constitutes land or subsurface strata. RP Opp. (U.S.) at 11:12-15. RP relies on Fertilizer Inst. v. United States E.P.A., 935 F.2d 1303 (D.C.Cir.1991), which concerns an EPA reporting regulation promulgated under CERCLA section 102(a). The regulation defined a release as “placement of a hazardous substance into any unenclosed containment ... wherein the” substance “is exposed to the environment.” Id. at 1307. In rejecting the regulation the court held that the regulation “cannot be reconciled with CERCLA’s express terms [that define] a release as the movement of a substance from a facility into the envi-ronment_” Id. Rather, the EPA regulation reached substances that were merely “exposed to the environment.” Id. Fertilizer Institute, while it draws a nice and probably correct distinction, does not respond to the argument that “environment” as defined by CERCLA may include the facility itself. Fertilizer Institute was concerned only with a regulation that had potential to reach harmful substances that, although stored in open containers, had never reached and may never reach the environment as defined at section 101(8). Moreover, the CERCLA provision construed in Fertilizer Institute did not cover threatened releases, while section 107(a) does. It has been more plainly held that release of hazardous wastes at a facility are sufficient to trigger section 107 liability for costs. See United States v. Mottolo, 695 F.Supp. 615, 623 (D.N.H.1988) (hazardous chemicals dumped onto site surface constitute releases or threatened releases within the meaning of CERCLA). Mottolo also notes that there is no requirement under CERCLA that off-site pollution occur. Id. More important, the liability provision plaintiffs seek to apply to RP indicates that RP may be held liable for a release at rather than from the facility. RP’s liability is premised on its status as successor to the liabilities of Mountain Copper, Ltd., who is said to have mined Iron Mountain. Assuming RP may be held to answer for Mountain Copper, its liability rests on section 107(a)(2), which states that “any person who at the time of disposal of any hazardous substance owned or operated any facility at which such hazardous substances were disposed of” shall be liable. Accordingly, summary judgment should be granted in plaintiff United States’ favor on RP1-US and RP2-US to the extent that those defenses pose a question whether a release or threatened release “at” a facility is within the scope of section 107(a). For the same reasons, the same assertion in RP2-CA should be stricken as to plaintiff California. 2. Exclusion of Mining Wastes from “Hazardous Wastes” under CERC-LA. In the second part of RP2-US and RP2-CA, RP contends that plaintiff has stated no claim warranting relief because mining wastes are excluded from CERCLA’s definition of hazardous substances. Plaintiff United States seeks summary judgment on RP2-US and plaintiff California seeks to strike identical language in RP2-CA. CERCLA section 101(14) defines hazardous substances at length. Subdivision (C) of section 101(14) purports to exclude from CERCLA solid wastes that are not now federally regulated. Plaintiffs insist that the subdivision (C) mining waste exclusion does not preclude CERCLA coverage of a waste if the constituents of that waste are described in any other part of section 101(14). The background of the mining exclusion is as follows. In 1976, Congress enacted the Resource Conservation and Recovery Act (“RCRA”), which amended the Solid Waste Disposal Act (“SWDA”). 42 U.S.C. § 6901 et seq. Subtitle C of RCRA covers hazardous wastes and subtitle D of RCRA covers non-hazardous wastes. Section 3001(b)(1) of RCRA requires EPA to identify wastes that, because they are hazardous, are subject to regulation under subtitle C. 42 U.S.C. 6921(b)(1). In 1978, EPA proposed that “special wastes,” which it defined as high-volume, low-toxicity wastes, be subject to fewer regulations. 43 Fed.Reg. 58,946 (1978). Among wastes that EPA classified as “special” were mining wastes, i.e., wastes from “extraction, beneficiation, and processing of ores and minerals.” 43 Fed.Reg. 59,016 (1978). But when in 1980 EPA issued final regulations identifying hazardous wastes under RCRA, it failed to separate “special wastes” for lesser regulation. 45 Fed.Reg. 33,066 (1980). Before EPA’s final regulations took effect, Congress enacted the Solid Waste Disposal Act Amendments of 1980, Pub.L. No. 96-482, 94 Stat. 2334, known as the Bevill Amendment. The Bevill Amendment suspended RCRA regulation of “solid wastes from the extraction, beneficiation, and processing of ores and minerals” until at least six months after EPA completed a comprehensive study of the adverse effects of these low-hazard, high-volume wastes. 42 U.S.C. § 6921(b)(3)(A)(ii). In 1980, Congress also enacted CERCLA, incorporating the Bevill Amendment mining waste exclusion in CERCLA’s definition of hazardous substances. See 42 U.S.C. § 9601(14)(C). Between 1980 and 1986, EPA proposed several interpretations of the scope of the Bevill Amendment. In a 1986 regulation, EPA exempted extraction and beneficiation wastes from RCRA. 51 Fed.Reg. 24, 496 (1986); also see Environmental Defense Fund v. U.S. Environmental Protection Agency [sic], 852 F.2d 1309, 1312 (D.C.Cir.1988), cert. denied, 489 U.S. 1011, 109 S.Ct. 1120, 103 L.Ed.2d 183 (1989). Such wastes, more precisely defined, are “[sjolid waste from the extraction of ores and minerals consisting of very large volumes of overburden and waste rock excavated during mining [and] [s]olid waste from the benefi-ciation of ores and minerals including large volumes of crushed rock tailings.” Environmental Defense Fund v. Environmental Protection Agency, 852 F.2d 1316, 1327 (D.C.Cir.1988), cert. denied sub nom., American Mining Cong. v. Environmental Defense Fund, 489 U.S. 1011, 109 S.Ct. 1120, 103 L.Ed.2d 183 (1989). EPA did not immediately decide which particular wastes would be excluded under the Bevill Amendment. In 1988, the District of Columbia Circuit Court of Appeals held that Congress intended the Bevill Amendment exclusion “to encapsulate the ‘special waste’ concept articulated by EPA in 1978,” meaning that all wastes of high volume and low toxicity would be excluded from regulation under subtitle C of RCRA. Environmental Defense Fund, supra, 852 F.2d at 1329. The court also imposed a deadline on EPA to determine which particular processing wastes would fall within the Bevill Amendment exclusion. Id. at 1331. By September 1989, EPA produced final exclusion criteria and began to evalúate particular processing wastes to determine which qualified for exclusion. The final list of excluded mining wastes is found at 40 C.F.R. 261.4(b)(7) (1990). Several courts have considered the question whether mining wastes excluded from RCRA regulation by the Bevill Amendment may be regulated under CERCLA notwithstanding the exclusionary reference in section 101(14)(C). The leading case is Eagle-Picher Indus. v. United States E.P.A., 759 F.2d 922 (D.C.Cir.1985), which held that section 101(14) limits the exclusion to wastes defined solely by subdivision (C), i.e., by SWDA. Id. at 928. Thus, under Eagle-Picher, a substance may be regulated under CERCLA if it falls within another subdivision of section 101(14) although it may be a Bevill Amendment waste excluded by subdivision (C). Id. at 930. The court reasoned that Congress, had it intended the exclusion to apply throughout section 101(14), would have placed the exclusion at the end of the section as was done with the exclusion for petroleum and natural gas. Id. at 927. Several district courts have also held that the mining waste exclusion applies only to those wastes defined solely by subdivision (C) and not by other subdivisions of section 101(14). See United States v. Metate Asbestos, 584 F.Supp. 1143, 1147 (D.Ariz.1984); Idaho v. Bunker Hill Co., 635 F.Supp. 665, 673 (D.Idaho 1986); Idaho v. Hanna Mining Co., 699 F.Supp. 827, 833 (D.Idaho 1987). RP contends that Eagle-Picher1 & reasoning is flawed. Eagle-Picher’& reliance on placement of the exclusion within subdivision (C) rather than at the end of subsection 14 conflicts with a sounder explanation for that placement. Each subdivision of 101(14) defines hazardous substances by reference to other environmental statutes. See n. 2, supra. Subdivision (C) incorporates those wastes defined by SWDA section 3001. Because SWDA section 3001 itself contains the Bevill Amendment exclusion, subdivision (C) of CERCLA section 101(14) is the rational place to include that exclusion. The legislative history of CERCLA is instructive given the parties’ conflicting interpretations of the meaning and scope of the mining waste exclusion. RP relies on a Senate Report (also cited in Eagle-Picher) that states: It should be noted that any substance or material for which regulation is specifically suspended by Act of Congress under the Solid Waste Disposal Act is excluded from designation as a hazardous substance for the purpose of S. 1480, notwithstanding the presence in such substance of any hazardous or toxic chemical. S.Rep. 848, 96th Cong., 2d Sess. 28 (1980). Although Eagle-Picher acknowledged that this language purports to establish that the mining waste exclusion applies without regard to the other wastes defined in section 101(14), the court nevertheless viewed congressional intent as unclear because the exclusion fell within subdivision (C) rather than at the end of section 101(14). 759 F.2d at 928-29. Because the language of the Senate Report conflicted with the plain meaning of the statute itself, said the court, the statute would control. Id. at 928-929. In this court’s view, Eagle-Picher misreads the statute. First, to give effect to the exclusion in section 101(14)(C), the statute must be read to exclude all Bevill Amendment wastes from CERCLA regulation. It is “ ‘an elementary canon of construction that a statute should be interpreted so as not to render one part inoperative.’ ” Mountain States Tel. & Tel. v. Pueblo of Santa Ana, 472 U.S. 237, 249, 105 S.Ct. 2587, 2594, 86 L.Ed.2d 168 (1985), quoting Colautti v. Franklin, 439 U.S. 379, 392, 99 S.Ct. 675, 684, 58 L.Ed.2d 596 (1979). Second, placement of the exclusion within subdivision (C) is sensible in that all SWDA wastes are defined in that subdivision. Third, to effectuate the exclusion regardless of its placement in the statute comports with clear legislative history. Plaintiff United States argues that by enacting SARA in 1986 and thereby reauthorizing CERCLA, Congress made clear that it approved Eagle-Picher and that it agreed that CERCLA applied to cleaning of mining waste sites in the same manner as it applied to other hazardous waste sites. Plaintiff notes that one goal of CERCLA reauthorization was to clean Iron Mountain Mine. 132 Cong.Rec. H9628 (daily ed. Oct. 8, 1986) (letter from L. Thomas, Admin, of EPA, to Sen. Stafford and Cong. Dingell). Plaintiff also notes that Congress did not adopt an amendment limiting EPA regulation of mining wastes. Plaintiffs arguments do not consider the limited nature of the Bevill Amendment exclusion. Nothing in plaintiffs arguments shows that Congress believed all mining wastes to be regulated by CERCLA or whether it merely recognized that non-Bevill Amendment wastes may be regulated. The plain meaning of section 101(14) suggests that only wastes not excluded by the Bevill Amendment or by some other exclusion may be regulated by CERCLA. In this court’s view, even if mining wastes are covered by CERCLA, certain wastes are excluded from coverage by reference to the Bevill Amendment in section 101(14)(C). Having decided that Bevill Amendment wastes are excluded from coverage even though they may fall within other parts of section 101(14), the next question is whether the particular waste at issue here—AMD—is among those excluded by the amendment. The parties have not raised or briefed this question. Rather, plaintiffs ask either to strike, or for summary judgment on, the defense that they have not stated claims because mining wastes are specifically excluded from the definition of hazardous substances. In fact, the Bevill Amendment excludes only “special” mining wastes defined at 40 C.F.R. 261.4(b)(7), see n. 4, infra. Not all mining wastes are excluded by the regulation. As a matter of law, plaintiffs’ motions must be denied. While it appears that plaintiff United States is entitled to a ruling in its favor on the question whether mining wastes generally are excluded under CERCLA, it also appears that some mining wastes are excluded, namely those found at 40 C.F.R. 261.4(b)(7). It cannot be decided on the present record that the part of RP2-US and RP2-CA pertaining to exclusion of mining wastes is insufficient. 3. Liability for Releases Permitted by Federal Law. RP asserts in RP2-US that plaintiff United States’ complaint also fails to state a claim to the extent that plaintiff seeks to recover response costs resulting from federally permitted releases; a similar allegation is made in RP12-US, which asserts that, because releases that may have occurred when RP’s predecessors owned the mine were federally permitted, plaintiff cannot recover under section 107 of CERC-LA. Plaintiff United States moves for summary judgment as to both defenses, asserting that they are legally insufficient. Section 107(j) provides that “[rjecovery ... for response costs ... resulting from a federally permitted release shall be pursuant to existing law in lieu of this section.” 42 U.S.C. § 9607(j). One such permitted release may arise under section 402 of the Clean Water Act, which authorizes issuance of a permit to discharge pollutants under the National Pollutant Discharge Elimination System (“NPDES”). 33 U.S.C. § 1342. Plaintiff acknowledges that from 1978 to 1983 an NPDES permit governed discharge of metals from two copper ce-mentation plants at Iron Mountain. Under CERCLA, costs of responding to a federally permitted release may not be recovered unless it is shown that non-federally permitted releases contributed to the natural injury. See In re Acushnet River & New Bedford Harbor, 722 F.Supp. 893, 897 (D.Mass.1989). While Acushnet suggests that plaintiffs have the burden to prove that non-permitted releases contributed to the harm, id., it places on defendants the burden to prove that the injury is divisible, id. at 897 n. 9, so that the award of response costs may be reduced to reflect the unrecoverable portion attributable to a permitted release. Even where releases may have been permitted, response costs may be recovered for any releases that (1) were not expressly permitted, (2) exceeded the limitations of the permit, or (3) occurred at a time when there was no permit. Idaho v. Bunker Hill, 635 F.Supp. 665, 673-74 (D.Idaho 1986). Plaintiff insists that flows covered by the 1978-1983 permit represent only a portion of AMD flows at the site. Plaintiff submits evidence that “a large percentage of metals released into the environment at the site come from waste rock piles, tail-ings piles, seeps and sediments. These metals-bearing flows are not treated in the copper cementation plants operated at the sites.” Deck of R. Sugarek at 1112. Plaintiff also submits evidence that all flows since 1983 have been unpermitted, decl. of J. Pedri at HU 6 and 8; that permits were limited to copper loading; and that full compliance with the permits was never achieved, id. at UK 6-7. RP does not address the question whether nonpermitted releases contributed to the alleged injury; instead, RP argues that plaintiff has not met its burden to produce evidence that the harm is divisible. Plaintiff replies that, even if the burden of production is properly plaintiffs, that burden is satisfied, by its evidence that only some releases were permitted. Plaintiff is correct: evidence that some wastes were not treated in the permit-covered cementation plant and that post-1983 releases were not permitted is sufficient to suggest that non-permitted releases contributed to the harm. More important, that a release may have been federally permitted does not alone prevent recovery of response costs. Recovery is prevented only where defendant proves that the injury is divisible. Acushnet places the burden to prove divisibility on defendant RP, who has offered no evidence to establish that the injury alleged by plaintiff may be attributed in measurable proportion to permitted releases. RP thus fails to sufficiently establish an essential element of its defense: divisibility of the injury. See Celotex v. Catrett, supra, 477 U.S. at 322, 106 S.Ct. at 2552. Accordingly, summary judgment is appropriately awarded in plaintiff United States’ favor on that part of RP2-US alleging that the United States has stated no claim to the extent that it seeks to recover for permitted releases. The same result holds as to RP12-US, which alleges that releases that may have occurred during ownership by RP’s predecessors were federally permitted. 4. Whether Mining Constitutes “Disposal” under CERCLA. In the fourth part of RP2-US and RP2-CA, RP alleges that plaintiffs fail to state a claim for relief because mining, including the'creation of tailings piles, does not constitute disposal within the meaning of CERCLA. The United States moves for summary judgment on this defense and California moves to strike it, both plaintiffs asserting that discharge of AMD and abandonment of tailings constitute disposal under CERCLA. •CERCLA defines disposal by reference to SWDA section 1004, codified at 42 U.S.C. § 6903. 42 U.S.C. § 9601(29). Under SWDA, “disposal” means the discharge, deposit, injection, dumping, spilling, leaking, or placing of any solid waste or hazardous waste into or on any land or water so that such solid waste or hazardous waste or any constituent thereof may enter the environment or be emitted into the air or discharged into any waters, including groundwaters. 42 U.S.C. § 6903(3). Plaintiff United States offers evidence that ore extraction at the mine has fractured mineralized zones, allowing formation of AMD that is discharged through the watersheds into Spring Creek Dam and the Sacramento River. Deck of R. Sugarek at ITU 6-8, 11. Plaintiff also submits evidence that heavy metals leach from tailings piles that were created during mining. Id. at ¶ 12. Plaintiff contends that discharge of AMD and abandonment of tailings constitute disposal under CERCLA. Opposing summary judgment, RP first insists that only disposal of solid or hazardous waste, rather than disposal of hazardous substances, can subject it to CERCLA liability, citing 3550 Stevens Creek Ass’n v. Barclays Bank, 915 F.2d 1355, 1361 (9th Cir.1990), cert. denied, — U.S. -, 111 5.Ct. 2014, 114 L.Ed.2d 101 (1991). Stevens Creek, which refused to allow cost recovery for removal of asbestos-laden building materials on grounds that the materials were not solid or hazardous wastes, draws a fine distinction in the statutory scheme. First, it notes that liability under CERCLA depends on occurrence of “a ‘release’ or ‘threatened release’ of a hazardous substance_” Id. at 1359; 42 U.S.C. § 9607(a)(4). Second, it notes that a CERC-LA “defendant must be a person ‘who at the time of disposal of any hazardous substance owned or operated any facility at which such hazardous substances were disposed of.’ ” Id.; 42 U.S.C. § 9607(a)(2). Third, it notes that the action constituting “disposal of any hazardous substance” must involve “any solid waste or hazardous waste_” Id. at 1361; 42 U.S.C. § 9601(29) and 42 U.S.C. § 6903. Thus, “[o]n its face, ‘disposal’ pertains to ‘solid waste or hazardous waste,’ not to building materials which are neither.” Id. Having established that a covered disposal must involve waste, RP narrows its argument, insisting that because CERCLA incorporates the RCRA definition of hazardous waste but not its definition of solid waste, see 42 U.S.C. § 9601(14), a disposal of solid mining wastes is not a disposal within the meaning of CERCLA. In fact, the Stevens Creek court read CERCLA to reach disposal of either solid or hazardous waste, not hazardous waste alone. 915 F.2d at 1361. So does the statute: the SWDA definition of disposal, incorporated in CERCLA at section 101(29) (42 U.S.C. § 9601(29)), defines a disposal of “solid waste or hazardous waste.” RP’s conclusion that disposal under CERCLA pertains only to hazardous wastes is a leap not supported by the statute. RP does not otherwise dispute that “discharge” of AMD from the mine constitutes a “disposal” under CERCLA. For this reason alone, summary judgment must go to plaintiff on the assertion in RP2-US that mining does not constitute disposal. For the same reason, the same allegation in RP2-CA is insufficient and plaintiff California’s motion to strike it will be granted. B. Failure to Mitigate Damages and Recovery of “Unnecessary” Costs. Plaintiff United States moves for summary judgment on RP4-US, which alleges that plaintiff failed to mitigate its damages. The United States also moves for summary adjudication of RP3-US, RP13-US, and RP18-US; in each of these, defendant denies that it is liable for “unnecessary” response costs. Plaintiff California moves to strike the same defenses, which are numbered RP3-CA, RP4-CA, RP13-CA, and RP18-CA. Under section 107, a person who is liable under CERCLA (e.g., because a hazardous waste was released, or threatens to be released, from a facility the person owns or operates) is liable for “all costs of removal or remedial action incurred by the United States Government or a State ,.... not inconsistent with the national contingency plan.” 42 U.S.C. § 9607(a)(l)-(4)(A). Where “any other person” than the United States, a State, or an Indian tribe incurs costs, a CERCLA defendant is liable for “any other necessary costs_” 42 U.S.C. § 9607(a)(l)-(4)(B). Because the statute provides that all costs may be recovered, courts have rejected the defense of failure to mitigate damages. “CERCLA does not impose a duty upon the Government to mitigate response costs.” United States v. Kramer, 757 F.Supp. 397, 420 (D.N.J.1991); Kelley v. Thomas Solvent Co., 714 F.Supp. 1439, 1451 (W.D.Mich.1989) (section 107 “imposes no duty upon the United States to mitigate costs,” so defense “is insufficient as a matter of law and will be stricken”); United States v. Marisol, Inc., 725 F.Supp. 833 (M.D.Pa.1989) (striking failure to mitigate damages defense). Opposing such a rule, RP cites United States v. Hardage, 116 F.R.D. 460, 466 (W.D.Okla.1987), which refused to strike the failure-to-mitigate defense in light of potential delay and misfeasance by the government. In the instant case, RP neither suggests nor offers evidence of misfeasance or delay. In other words, defendant gives no persuasive reason to maintain an otherwise insufficient defense. Accordingly, plaintiffs’ motions relative to the failure to mitigate defenses (RP4-US and RP4-CA) will be granted. As to recovery of unnecessary costs, governments have generally been allowed to recover all response costs, as is permitted by the statute. See United States v. R.W. Meyer, Inc., 889 F.2d 1497, 1504 (6th Cir.1989) (government entitled to recover “indirect” or overhead costs, although costs were higher than other entities might have incurred), cert. denied, 494 U.S. 1057, 110 S.Ct. 1527, 108 L.Ed.2d 767 (1990); United States v. Bell Petroleum Servs., Inc., 734 F.Supp. 771, 780-781 (W.D.Tex.1990) (costs need not be shown to be “reasonable and necessary”); United States v. Hardage, 733 F.Supp. 1424, 1432 (W.D.Okl.1989) (“[c]ourts have emphasized that liability extends to ALL response costs”). For the above reasons, plaintiff United States’ motion for summary judgment should be granted on RP3-US, RP13-US, and RP18-US to the extent that those defenses deny liability for unnecessary response costs. Similar defenses in the California case — RP3-CA, RP13-CA, and RP18-CA — should be stricken to the same extent. C. Constitutional Defenses. Numerous constitutional issues are raised by RP5-US and RP5-CA, which allege that imposition of CERCLA liability violates the due process, equal protection, taking, and ex post facto clauses of the United States Constitution. RP13-US and RP13-CA allege that the remedy selection process (i.e., the process of determining responses necessary to releases or potential releases of hazardous substances) violates the due process and equal protection clauses. These defenses are subjects of the United States’ motion for summary judgment and of California’s motion to strike. 1. Deprivation of Procedural Due Process in Remedy Selection. RP13-US and RP13-CA allege in part that RP was denied due process in the selection of remedies to be used at Iron Mountain Mine. In opposition to plaintiffs’ motions, RP argues that it was denied an opportunity to examine witnesses and that it was also denied an opportunity to have remedy selection issues resolved by an impartial decision-maker. The same contentions were considered at the June 12, 1992, hearing on the United States’ motion to limit review of remedy selection to the administrative record. In an order dated June 19, 1992, the court stated that it “is satisfied that the informal remedy selection hearing process in which the administrative record is made (see 42 U.S.C. § 9613(k)) comports with due process, as does section 113(j) [42 U.S.C. § 9613(j) ]. See United States v. Rohm & Haas Co., Inc., 669 F.Supp. 672, 680 (D.N.J.1987).” The court also rejected RP’s argument regarding partiality of the remedy-selection decision maker: Nor does the court believe, as defendants argue, that EPA must erect an impartial tribunal to consider challenges to its remedy selections. First, the court recognizes that administrative agencies “ ‘ “should be free to fashion their own rules of procedure and to pursue methods of inquiry capable of permitting them to discharge their multitudinous duties.” ’ ” Vermont Yankee Nuclear Power Corp. v. National Resources Defense Council, Inc., 435 U.S. 519, 524 [98 S.Ct. 1197, 1202, 55 L.Ed.2d 460] (1978). Second, nothing in EPA’s remedy selection process suggests the kind of person-" al interest or bias condemned in the due process cases. The June 19 order requires that RP’s due process challenge to the remedy selection process be rejected. Accordingly, summary judgment should be granted in the United States’ favor on RP13-US to the extent that the defense alleges a due process violation; in the California case, RP13-CA should be stricken to the same extent. 2. Deprivation of Substantive Due Process. RP insists in RP5-US and RP5-CA that CERCLA deprives it of substantive due process in that the statute may lead to imposition of substantial retroactive monetary liability “on a corporate entity that neither was ‘responsible’ for, ‘created,’ nor ‘profited from’ releases from the Iron Mountain superfund site.” RP Opp. (U.S.) at 38:20-39:1. The burden to overcome the presumption that federal legislative acts are constitutional is RP’s: “ ‘legislative acts adjusting the burdens and benefits of economic life come to the Court with a presumption of constitutionality,’ ” so that “ ‘the burden is on one complaining of a due process violation to establish that the legislature has acted in an arbitrary and irrational way.’ ” Pension Benefit Guar. Corp. v. R.A. Gray & Co., 467 U.S. 717, 729, 104 S.Ct. 2709, 2717, 81 L.Ed.2d 601 (1984), quoting Usery v. Turner Elkhorn Mining Co., 428 U.S. 1, 15, 96 S.Ct. 2882, 2892, 49 L.Ed.2d 752 (1976). Where legislation operates retroactively, the retroactive aspects must themselves meet the test of due process. Id. 467 U.S. at 730, 104 S.Ct. at 2718. “But that burden is met simply by showing that the retroactive application of the legislation is itself justified by a rational legislative purpose.” Id. The retroactive reach of CERCLA has been explained as follows: Although CERCLA does not expressly provide for retroactivity, it is manifestly clear that Congress intended CERCLA to have retroactive effect. The language used in the key liability provision, ... 42 U.S.C. § 9607, refers to actions in the past tense: “any person who at the time of disposal of any hazardous substances owned or operated,”.... NEPACCO, supra, 810 F.2d at 732-733. Courts “have held uniformly that retroactive operation [of CERCLA] survives the Supreme Court’s tests for due process validity.” United States v. Monsanto Co., 858 F.2d 160, 174 (4th Cir.1988); NEPACCO, supra, 810 F.2d at 733 (“retroactive application of CERCLA does not violate due process”). RP argues that its due process rights are violated in that its alleged immediate predecessor, Stauffer Chemical Company, “had no connection with the mining activities at Iron Mountain and in no way profited from those activities_” RP Opp. (U.S.) at 39 n. 42. While the argument is attractive, “ ‘the burden is on one complaining of a due process violation to establish that the legislature has acted in an arbitrary and irrational way.’ ” Pension Benefit Guar. Corp., supra, 467 U.S. at 729, 104 S.Ct. at 2717. RP makes no such showing. But, more important, RP’s liability is not predicated on RP’s or Stauffer’s status as miner of the land; rather, liability is premised on RP’s status as successor to the liabilities of the alleged miner, Mountain Copper, Ltd. Because RP fails to overcome the presumption of constitutionality, its substantive due process challenge must be rejected and summary judgment granted in favor of plaintiff United States on that part of RP5-US alleging a substantive due process violation; RP5-CA will be stricken to the same extent as to plaintiff State of California. 3.Equal Protection as to Liability and Remedy Selection. RP has also pled equal protection challenges that, like the due process challenges, have two components. In RP13-US and RP13-CA, RP asserts that the remedy selection process violated its right to equal protection of the laws. In RP5-US and RP5-CA, RP asserts that imposition of CERCLA liability also violates the equal protection clause. Because RP does not assert membership in a suspect class, equal protection review is done under the rational basis test. See Massachusetts Bd. of Retirement v. Murgia, 427 U.S. 307, 313-314, 96 S.Ct. 2562, 2566-67, 49 L.Ed.2d 520 (1976). Regarding RP5-US and RP5-CA, RP asserts that statutory joint and several liability under CERCLA encourages selective prosecution of those who are best able to reimburse governments for response costs, i.e., those who are wealthiest. But RP does not establish that such selectivity, if it occurs, is irrational. It has been held that CERCLA’s “imposition of joint and several liability ... does not raise a constitutional equal protection issue and, even if it did, would not violate the constitutional standard requiring equal protection of the laws.” United States v. Conservation Chem. Co., 619 F.Supp. 162, 214 (D.Mo.1985); also see United States v. Kramer, supra, 757 F.Supp. at 432. Because RP offers no persuasive reason to reject the quoted conclusion, plaintiffs’ motions concerning RP5US and RP5-CA will be granted to the extent that the defenses allege that CERCLA liability violates equal protection. In its opposing memoranda, RP makes no effort to establish that its right to equal protection of the laws was violated in the remedy selection process. Accordingly, plaintiffs’ motions regarding the equal protection issue raised in RP13-US and RP13-CA will also be granted. 4. Ex Post Facto Law. RP5-US and RP5-CA also allege that CERCLA violates the prohibition against ex post facto laws found in Article I, Section 9, of the United States Constitution. The defense has often been rejected. See Monsanto, supra, 858 F.2d at 174-175 (CERCLA, which “does not exact punishment,” is not an ex post facto law); United States v. Kramer, supra, 757 F.Supp. at 431 (D.N.J.1991) (“[a]ll courts to have considered the issue have rejected” argument that CERCLA is penal and ex post facto). RP offers no persuasive reason to vary that conclusion. Rather, RP notes that the ex post facto clause has been applied analogically in civil cases. See Ralis v. RFE/ RL, Inc., 770 F.2d 1121 (D.C.Cir.1985); Louis Vuitton S.A. v. Spencer Handbags Corp., 765 F.2d 966 (2d Cir.1985). But RP gives no reason for analogical application of the ex post facto clause to this case. The Louis Vuitton case notes that the ex post facto prohibition “generally applies to criminal statutes [but] may also be applied where the civil disabilities disguise criminal penalties.” Id. at 972. Here, it can hardly be said that CERCLA’s liability scheme, which limits liability to response costs incurred by the government, provides for imposition of criminal penalties. Finally, in its opposition to the United States’ motion, RP attempts ex post facto to recast its ex post facto challenge as a challenge under the excessive fines clause of the Eighth Amendment. Because a violation of the Eighth Amendment is not among the defenses pled by RP, the arguments need not be reached. For the above reasons, RP’s ex post fac-to challenge must also be rejected and summary judgment awarded to the United States on that part of RP5-US assailing CERCLA as an ex post facto law. The same part of RP5-CA must be stricken. 5. Taking for Public Use without Just Compensation. The last constitutional principle invoked in RP5-US and RP5-CA is the taking clause of the Fifth Amendment, which states that private property shall not be taken for public use without payment of just compensation. RP asserts that CERC-LA violates the clause. Again, the defense has been rejected. NEPACCO, supra, 810 F.2d at 734 (government’s cleanup does not deprive property owner of any property interest); Conservation Chemical, supra, 619 F.Supp. at 215-217 (“[w]hat defendants have referred to as a ‘taking’ is, in reality, nothing more than an attempt to transform a substantive due process challenge of an economic regulation into a confiscation of defendant’s property rights”); United States v. Cannons Engineering Corp., 720 F.Supp. 1027 (D.Mass.1989), aff'd, 899 F.2d 79 (1st Cir.1990). RP cites no cases and makes no arguments to support a contrary conclusion. Accordingly, plaintiffs’ motions concerning the allegations in RP5-US and RP5-CA that CERCLA will cause an unconstitutional taking will be granted. D. Equitable Defenses. Plaintiff United States moves for summary judgment on RP’s equitable defenses: RP6-US, which alleges that plaintiff is es-topped to recover, and RP8-US, which alleges that plaintiff’s hands are unclean. Each of these defenses rests on the allegation that incurring of response costs was caused by actions of the United States Bureau of Reclamation (“USBR”). California moves to strike similar defenses, RP6-CA and RP8-CA, in the California case. Plaintiffs also seek rulings on the merits of defendants IMMI/Arman’s equitable defenses, IMMI11-US and IMMI11-CA, which allege that each plaintiff waived its right to recover. Plaintiffs argue that equitable defenses are precluded by the limitation in section 107(a) to defenses listed in section 107(b). 42 U.S.C. § 9607. Section 107 specifies that CERCLA liability is “subject only to the defenses set forth in subsection (b) of this section.” Id. Only three defenses are listed in section 107(b): (1) “act of God”; (2) “act of war”; and (3) “act or omission of a third party other than an employee or agent of the defendant_” Id. Plaintiffs argue that, because equitable defenses such as waiver, estoppel, and unclean hands are not section 107(b) defenses, they are not available to defeat CERCLA claims. Courts are divided on the question whether equitable defenses are available in section 107 cases. A number of cases have allowed equitable defenses, reasoning that because the government in a cost recovery case seeks the equitable remedy of restitution defendants should not be barred from raising equitable defenses. Conservation Chemical, supra, 619 F.Supp. at 206; Violet v. Picillo, 648 F.Supp. 1283, 1294 (D.R.I.1986); Mardan Corp. v. C.G.C. Music, Ltd., 600 F.Supp. 1049, 1056 n. 9 (D.Ariz.1984), aff'd 804 F.2d 1454 (9th Cir.1986). Other courts have reasoned that, because equitable concerns may be raised elsewhere under CERCLA (e.g., under section 113, which provides that a CERCLA defendant may seek contribution from other responsible parties during or after a section 107(a) action, 42 U.S.C. § 9613(f)(1)), the limitation of defenses expressed by section 107 should be observed. United States v. Kramer, supra, 757 F.Supp. at 427. The Supreme Court has noted that “Congress may ... guide or control the exercise of” equitable discretion on “clear and valid legislative command.” Weinberger v. Romero-Barcelo, 456 U.S. 305, 313, 102 S.Ct. 1798, 1803-04, 72 L.Ed.2d 91 (1982). Accordingly, courts have held that section 107(a), which states that CERCLA liability is “subject only to the defenses set forth” at section 107(b), is a valid restriction under Romero-Barcelo. See United States v. Stringfellow, 661 F.Supp. 1053, 1062 (C.D.Cal.1987); Kramer, supra, 757 F.Supp. at 425-428 (striking defenses of unclean hands, estoppel, waver, laches, and in pari delicto). Cases precluding equitable defenses are more persuasive than those allowing the defenses for the simple reason that the former comport with the statute while the latter do not. Additionally, the particular defenses at issue here — waiver, estoppel, and unclean hands — may not be asserted against sovereigns who act to protect the public welfare. Stringfellow, supra, 661 F.Supp. at 1062, citing Chesa peake & Delaware Canal Co. v. United States, 250 U.S. 123, 125, 39 S.Ct. 407, 407-08, 63 L.Ed. 889 (1919). For these reasons, RP’s and IMMI/Arman’s equitable defenses (RP6-US, RP8-US, IMMI11-US, RP6-CA, RP8-CA, RP11-CA) are insufficient and plaintiffs’ motions will be granted. E. Other USBR Defenses. In several additional defenses, defendants seek to place responsibility for the alleged environmental harm on USBR. In RP7-US and RP7-CA, RP claims that to hold it liable for USBR’s actions violates due process and equal protection. In RP9-US and RP9-CA, RP claims that plaintiffs, through government dam-building, caused the injury and assumed the risk of the injury for which they now seek recovery. In IMMI17-US and IMMI17-CA, IMMI/Arman also claim that USBR caused plaintiffs’ injuries. Regarding RP7-US, plaintiff United States notes that CERCLA liability is joint and several so that one defendant may be held liable for actions of another. Plaintiff acknowledges that it, too, may be held liable under section 120(a), which states that “[e]ach department, agency, and instrumentality of the United States” is “subject to” CERCLA “in the same manner, and to the same extent, both procedurally and substantively, as any nongovernmental entity, including liability under section 9607....” 42 U.S.C. § 9620(a)(1). But plaintiff sees no irrationality in holding RP jointly and severally liable for damage that may have been done by USBR. Nor does RP establish that joint and several liability is irrational and thus outside the parameters of due process and equal protection. Also, section 113 permits RP to seek contribution from other responsible parties during or after the section 107(a) action. 42 U.S.C. § 9613(f)(1). Thus, as a matter of law, no constitutional harm is done by holding RP liable for damage that may have been done by USBR where RP’s own liability for injury is established. If RP is found to be liable under section 107, then it is liable for all the damage although it may not have caused the damage. Accordingly, the defenses numbered RP7-US and RP7-CA are insufficient and plaintiffs’ motions will be granted. As to RP9-US, RP9-CA, IMMI17-US, and IMMI17-CA, which allege that USBR caused injury, both such defenses are subsumed in part by the third-party defense authorized by section 107(b)(3). See 42 U.S.C. § 9607(b)(3). While RP admits that its ninth defense simply buttresses its section 107(b)(3) defense, no buttressing is needed. Defendants IMMI/Arman state that they do not contest plaintiffs’ motions relative to IMMI17-US and IMMI17-CA. Section 107(b)(3) provides that “[tjhere shall be no liability ... for a person otherwise liable who can establish ... that the release or threat of release of a hazardous substance and the damages resulting therefrom were caused solely by ... (3) an act or omission of a third party_” All defendants have pled the third-party defense: RP in RP10-US and RP10-CA, IMMI/Ar-man in IMMI5-US and IMMI5-CA. None of these defenses are subjects of plaintiffs’ motions. To the extent that a release, if caused by USBR, was not caused solely by USBR, defendants are protected by section 113(f), which permits equitable contribution. 42 U.S.C. § 9613(f). As to defenses RP9-US, RP9-CA, IMMI17-US, and IMMI17-CA, the United States’ motion for summary judgment will be denied on grounds that judgment on these defenses is premature until the question of USBR’s liability is addressed. The question is raised both by counterclaims filed by all defendants and by the third-party defense that all defendants have pled; the court wishes to avoid any potential argument that litigation of USBR’s responsibility is precluded by the grant of summary judgment in the United States’ favor on the USBR defenses. On the other hand, RP9-US and RP9-CA, which allege assumption of risk, may be stricken as insufficient. Assumption of risk is not a valid defense to a CERCLA cost recovery action. See United States v. Marisol, Inc., 725 F.Supp. 833, 839-840 (M.D.Pa.1989). Also, defendants IMMI/Arman have stated non-opposition to the striking of IMMI12-US, IMMI12-CA, IMMI14-US, IMMI14-CA, IMMI17-US, IMMI17-CA, IMMI18-US, and IMMI18-CA, which they believe to be related to the third-party defenses (IMMI5-US and IMMI5-CA) that are not subjects of plaintiffs’ motions. Accordingly, these defenses will be stricken as to both plaintiffs. F. Release of a Naturally-Occurring Substance. Plaintiff United States moves for summary judgment on RP11-US, which alleges that RP is not liable for releases of naturally occurring substances because EPA lacks authority to respond to such a release. The defense is premised on section 104(a)(3)(A), which says: “[t]he President shall not provide for a removal or remedial action !.. in response to a release or threat of release ... of a naturally occurring substance in its unaltered form, or altered solely through naturally occurring processes or phenomena, from a location where it is naturally found_” 42 U.S.C. § 9604(a)(3). Plaintiff admits that RP is not liable for the costs of responding to releases of naturally occurring substances but believes that the proposition does not apply in this case. Plaintiff submits evidence that AMD flowing from the mine, although it consists of naturally occurring substances, is not itself naturally occurring but rather was created by mining. Deck of R. Sugarek at ¶¶ 6-10. In opposition, RP argues that plaintiff misconstrues the statute, which “is not phrased in terms of the process by which the hazardous substances are released.” RP Opp. (U.S.) at 60:13-16. The argument is confusing. If RP means to say that section 104 prohibits response to releases of natural substances, it is mistaken: the statute permits response to release of any natural substance released in altered form, or to release of a substance not altered by natural processes. RP also argues that mining, if it could cause harm, could merely expose more ore surface to natural weathering processes that combine with ore to create AMD. But it is not disputed that mining constitutes an artificial alteration rather than a naturally occurring process or phenomenon. To the extent that mining may have exposed more ore to weather, it may have non-naturally increased the amount of AMD released into the environment. In Mid-Valley Bank v. North Valley Bank, 764 F.Supp. 1377 (E.D.Cal.1991), relied on by both parties, defendant offered evidence that metals found at the site were within natural or background levels; defendant argued that the metals were thus naturally occurring. Plaintiff, on the other hand, offered evidence that the metals exceeded background levels. Recognizing a dispute, the court denied summary judgment. That dispute concerned the material question whether a hazardous substance triggering liability had been released. Here, defendant RP offers no evidence to dispute that AMD released from the mine exceeds natural levels. In fact, RP points to no evidence to show that any release to which EPA has responded is naturally occurring. Rather, RP’s evidence merely establishes that discharge of AMD occurred before mining did. Decl. of L. Erickson at ¶ 11. Plaintiff, on the other hand, offers evidence that present AMD releases from the mine are extremely elevated and that they exceed California water quality standards. Supp. deck of R. Sugarek at ¶¶ 9-10. RP also argues that, even if EPA has authority to respond to releases at the mine, a disputed issue of fact exists regarding divisibility of harm. RP again relies on In re Acushnet River & New Bedford Harbor, supra, 722 F.Supp. at 897, which assigns to RP the burden to prove that the injury is divisible. Id. at 897 n. 9.