ORDER
Plaintiffs, Oklahoma citizens, filed this class action in state court asserting nuisance, trespass, strict liability and unjust enrichment claims, based on the defendants’ alleged contamination of their property through operation of the Blackwell Zinc Smelter in Blackwell, Oklahoma. The defendants are Blackwell Zinc Company, Inc. (“BZC”), the company that operated the smelter, its parent companies and their successor corporations — Freeport-McMoRan Copper & Gold, Inc., Phelps Dodge Corp., Cyprus Amax Minerals Co., and Amax, Inc., — the Blackwell Industrial Authority (“BIA”), and BNSF Railway Co. The defendants removed the case, asserting jurisdiction under the Class Action Fairness Act of 2005 (“CAFA”), the Comprehensive Environmental Response, Compensation, and Liability Act (“CERCLA”),1 and 28 U.S.C. § 1442(a)(1), federal officer removal.2 The plaintiffs have filed a motion to remand, contending the court lacks subject matter jurisdiction.
BACKGROUND3
BZC owned and operated a smelter, used to refine zinc and cadmium ore concentrates, in Blackwell, Oklahoma from 1916 until 1974. In 1974, it dismantled the facility and donated the land to the Blackwell Industrial Authority (“BIA”).4 The BIA developed the property as an industrial park. Currently known as the Blackwell Industrial Park, the site5 consists of approximately 160 acres located one-half mile west of downtown Blackwell.
In 1992 the Environmental Protection Agency (“EPA”) indicated Blackwell might be designated a Superfimd site and placed on the National Priorities List if investigation and remedial efforts did not begin. That year BZC, the BIA and the City of Blackwell entered into a Consent Agreement and Final Order with the Oklahoma State Department of Health “to characterize and remediate the environmental contamination at the old Smelter site and on *1261city property.” Petition, ¶ 12. Two years later, in 1994, the Oklahoma Department of Environmental Quality (“ODEQ”)6 and the EPA entered into a Memorandum of Understanding (“MOU”)7 regarding remedial action to be taken in response to the release of hazardous substances at the Blackwell Zinc Site (“Site”).8 The purpose of the MOU was to “ensure prompt CERCLA-Quality Cleanup of the Site,” but conserve EPA resources by allowing ODEQ to develop and implement a remediation plan for the Site.
The MOU required ODEQ to submit drafts of its proposed Plan, Decision Document,9 and Remedial Design to the EPA for its review and comment, present to the EPA the procedures selected to complete the Remedial Design/Remedial Action (“RD/RA”), submit written quarterly and yearly progress reports, and arrange for EPA representatives to have access to the Site. The MOU provided that “[i]f, at any time, EPA determines that ODEQ is overseeing or conducting remedial action, at the Site, which is inconsistent with CERCLA, or the NCP, or which is not a CERCLA-Quality Cleanup, this MOU shall terminate.” MOU, p. 10.10 See Notice of Removal, Exhibit C, Record of Decision Document (“ROD”), p. 7 (“EPA agreed to not make a final determination to list the Site on the National Priorities List (NPL) as long as the pilot project proceeds in a timely manner and achieves CERCLA quality results.”).
The Site was divided into three operable units — the Soil Remediation Unit (“SRU”), the Ecological Remediation Unit (“ERU”) and the Ground Water Remediation Unit (“GRU”) — to allow soil remediation to proceed in advance of the other units. The SRU dealt “with soil contamination in residential, recreational, and commercial/industrial areas on the site.” ROD, p. 1. The ERU included “areas subject to ecological risk-based remediation goals,” such as grasslands, riparian areas, and streams. The GRU pertained to contaminated groundwater.11
In April, 1996, ODEQ issued its Record of Decision, which specified the remediation plan for the SRU. Remediation efforts began, principally consisting of soil removal and the capping in place of impacted soil. In July, 2001, the Final Remedial Action Completion Report (“Final Report”) for the SRU was issued, which documented that “the construction of the *1262remedy as prescribed within the Final Remedial Design Report, Blackwell Zinc Site Soil Remediation Unit, Blackwell, Oklahoma, PTI, May 1999 was substantially complete in 1999 with final completion in 2000.” Notice of Removal, Exhibit E, Second Five-Year Review Report (“Second Report”), p. iii. The Second Report, dated April, 2008, states that BZC had undertaken a second residential soil sampling program and that “[t]hrough May 2008, access agreements for over 8,400 properties had been obtained and sampling is in progress. Based on results, some properties will require cleanup. Cleanup will begin in May 2008.” Id. The Report also states that:
Construction is complete for those properties within the SRU that were identified and addressed prior to ODEQ’s approval of the FRACR.12 Additional properties are currently being evaluated under a supplemental soil sampling program. With respect to those additional properties, the remediation status is “under construction.”
Id. at p. iv.
In April, 2008, the plaintiffs filed this class action in state court, asserting state common law claims and seeking both monetary and equitable relief.13 The defendants removed the case, asserting three bases for federal jurisdiction: CAFA, CERCLA and federal officer.
CAFA
CAFA “extends the subject matter jurisdiction of the federal courts to encompass putative class actions in which at least one plaintiff class member is diverse from one defendant and where the amount in controversy exceeds $5 million.” Weber v. Mobil Oil Corp., 506 F.3d 1311, 1313 (10th Cir.2007). “In enacting CAFA, Congress was responding to what it perceived as abusive practices by plaintiffs and their attorneys in litigating major interstate class actions in state courts, which had ‘harmed class members with legitimate claims and defendants that ha[d] acted responsibly,’ ‘adversely affected interstate commerce,’ and ‘undermined public respect for our judicial system.’ ” Amoche v. Guarantee Trust Life Ins. Co., 556 F.3d 41, 47 (1st Cir.2009) (quoting CAFA, Pub. L. No. 109-2, § 2(a), 119 Stat. 4, 4 (2005)). Although “CAFA’s language favors federal jurisdiction over class actions,” Evans v. Waiter Indus., Inc., 449 F.3d 1159, 1163 (11th Cir.2006), Congress “did not give federal courts jurisdiction over ail class actions, specifically excluding those consisting of ‘primarily local matters.’ ” Johnson v. Advance America, 549 F.3d 932, 938 (4th Cir.2008), citing Sen. Rep. No. 109-14, at 6 (2005), U.S.Code Cong. & Admin.News 2005, at 3, 7. The plaintiffs argue that their lawsuit constitutes a “truly local dispute,” which falls within the “local controversy” exception to CAFA. That exception provides in relevant part:
“A district court shall decline to exercise jurisdiction ...—
(A)(i) over a class action in which—
(I) greater than two-thirds of the members of all proposed plaintiff classes in the aggregate are citizens of the State in which the action was originally filed;
(II) at least 1 defendant is a defendant—
*1263(aa) from whom significant relief is sought by members of the plaintiff class;
(bb) whose alleged conduct forms a significant basis for the claims asserted by the proposed plaintiff class; and
(cc) who is a citizen of the State in which the action was originally filed; and
(III) principal injuries resulting from the alleged conduct or any related conduct of each defendant were incurred in the State in which the action was originally filed; and
(ii) during the 3-year period preceding the filing of that class action, no other class action has been filed asserting the same or similar factual allegations against any of the defendants on behalf of the same or other persons.”14
28 U.S.C. § 1332(d)(4)(A). While the defendants must establish under CAFA that the amount in controversy and minimal diversity requirements are met, Amoche, 556 F.3d at 48, the plaintiffs bear the burden of proving that one of the exceptions to federal jurisdiction applies. Evans, 449 F.3d at 1164-65.
There is no dispute here that the defendants have met their initial burden of showing the existence of diversity jurisdiction under CAFA. See 28 U.S.C. § 1332(d)(2). The question is whether the plaintiffs have met their burden to establish an applicable exception. In particular, have the plaintiffs satisfied the “significant defendant” prong of the local controversy exception? See generally Evans, 449 F.3d at 1163 (“CAFA’s legislative history suggests that Congress intended the local controversy exception to be a narrow one, with all doubts resolved ‘in favor of exercising jurisdiction over the case.’ ”) (quoting S.Rep. No. 109-14 at 42, U.S.Code Cong. & Admin. News 3, 40). For the exception to apply, the proposed class must seek significant relief from an instate defendant whose conduct forms a significant basis for their claims. The plaintiffs argue that both Blackwell Zinc Company (“BZC”) and the Blackwell Industrial Authority (“BIA”) are “significant defendants.”
BZC
The initial question is whether BZC is an in-state/Oklahoma corporation. As BZC is incorporated in New York, its principal place of business must be in Oklahoma for it to be considered an Oklahoma citizen. See Gadlin v. Sybron Int’l Corp., 222 F.3d 797, 799 (10th Cir.2000). The defendants claim BZC has not conducted sufficient business to have a principal place of business in Blackwell “since at least the time that Phelps Dodge Corporation (now Freeport-McMoRan Corporation) acquired BZC’s parent corporation Cyprus Amax, in 1999.” Defendants’ response, p. 29. BZC’s corporate secretary has attested that, since 1999, the company “has conducted no business, other than following the formalities to keep the corporation in existence.” Defendants’ Notice of Removal, Exhibit G, Currault affidavit. The defendants also have submitted testimony that BZC has no significant assets *1264and is not engaged in any revenue-generating operations. However, in 2006, BZC filed a permit application with ODEQ for the Blackwell Zinc Site Groundwater Treatment Plant, whose stated purpose was to treat groundwater contaminated “by past operation — at the former Blackwell Zinc Smelter.” 15 Plaintiffs’ motion, Exhibit 4, water permit. BZC also has engaged in continued remediation efforts in conjunction with the ROD, including “an extensive community outreach program.” 16 Second Report, pp. 14-17 (“BZC also performed several response actions in 2007 in connection with a supplemental residential soil sampling program (SSP) initiated by BZC that year.”).
While, as the defendants assert, an inactive corporation that is merely responding to claims or lawsuits arising from its prior operations may not have a “principal place of business,” as the term is used in 28 U.S.C. § 1332(c)(1), the court is not persuaded that the “various environmental response activities that [BZC] has conducted or overseen beginning in the 1990’s,” defendants’ response, p. 29, do not constitute “transacting business” for jurisdictional purposes. The plaintiffs argue that “[t]he only purpose for the continued existence of BZC is to clean up pollution in Blackwell left behind by the smelter.” Plaintiffs’ motion, p. 11. Whether or not that is the sole reason for the company’s existence, it appears to be a substantial activity in which it currently is engaged. The court finds that activity suffices to establish Oklahoma as BZC’s principal place of business. See Gadlin, 222 F.3d at 799 (“When determining a corporation’s principal place of business, a court should look to the ‘total activity of the company’ or the ‘totality of the circumstances,’ considering ‘the character of the corporation, its purposes, the kind of business in which it is engaged, and the situs of its operations.’ ”) (quoting Amoco Rocmount Co. v. Anschutz Corp., 7 F.3d 909, 915 & n. 2 (10th Cir.1993)).
Resolving the question of BZC’s citizenship does not end the inquiry though. To establish the local controversy exception, the plaintiffs also must show that they seek significant relief from the former smelter operator and that BZC’s conduct forms a significant basis for their claims. CAFA offers little guidance in defining or supplying “standards for determining whether the relief sought is ‘significant,’ or for determining which bases for the plaintiffs’ claims are ‘significant.’ ” Ava Acupuncture P.C. v. State Farm, Mut. Auto. Ins. Co., 592 F.Supp.2d 522, 528 (S.D.N.Y.2008). Many courts have concluded that the terms “significant relief’ and “significant basis” are ambiguous, allowing resort to legislative history, a Senate Committee *1265Report, for their interpretation.17 E.g., Kearns v. Ford Motor Co., 2005 WL 3967998, at *9-11 (C.D.Cal.2005). Contra Caruso v. Allstate Ins. Co., 469 F.Supp.2d 364, 370 (E.D.La.2007) (“When, as here, there is no ambiguity in the statutory language that would warrant looking beyond the plain language of the statute for additional understanding of Congress’s intent, resort to the legislative history is unnecessary.”) (internal quotation omitted).
Courts generally have required that the local defendant’s conduct be significant when compared to the alleged conduct of the other defendants and that “the relief sought against that defendant is a significant portion of the entire relief sought by the class.” Evans, 449 F.3d at 1167. Several courts have stated that the significant relief inquiry “ ‘includes not only an assessment of how many members of the class were harmed by the defendant’s actions, but also a comparison of the relief sought between all defendants and each defendant’s ability to pay a potential judgment.’ ” Id. (quoting Robinson v. Cheetah Tramp., 2006 WL 468820, at *3 (W.D.La.2006) (opinion of Magistrate Judge), aff'd, 2006 WL 1453036 (W.D.La.2006) (district court opinion)).
The defendants do not dispute that BZC’s alleged conduct satisfies the “significant basis” requirement. As the sole owner and operator of the smelter, BZC clearly played a principal role in the alleged contamination. The defendants also do not dispute that the class seeks to recover extensive damages from the instate defendant. Instead they focus on BZC’s alleged lack of assets, arguing that a class cannot seek significant relief from a party that is unable to satisfy any judgment entered against it. Essentially, the defendants claim that a class can only seek “significant relief’ under CAFA from a defendant with “appreciable assets.” Defendants’ response, p. 13.
The construction of the phrase “from whom significant relief is sought,” urged by the defendants originated in Robinson, 2006 WL 468820, at *3. That class action was filed in state court in Louisiana on behalf of all person who resided or worked in Caldwell Parish on a certain date and were affected by the closure of the Columbia bridge, which occurred after it was struck by a tractor-trailer driven by John Gaston. The putative class representative sued the truck driver, his employer, the employer’s insurer, and the owner of the truck’s cargo. All the defendants, with the exception of Gaston, were citizens of states other than Louisiana. They removed the case to federal court pursuant to CAFA and the plaintiff sought to have it remanded under the statute’s local controversy exception.
The court held that the putative class did not seek significant relief from Gaston, the sole in-state defendant, relying on an example the Senate used to illustrate the local controversy exception. The example described a products liability class action brought in Florida on behalf of Floridians against an out-of-state automobile manufacturer and a few in-state dealers alleging a defective transmission. The Report stated that the
automobile dealers are not defendants whose alleged conduct forms a significant basis of the claims or from whom significant relief is sought by the class. Even if the plaintiffs are truly seeking relief from the dealers, that relief is just small change compared to what they are seeking from the manufacturer. Moreover, the main allegation is that the *1266vehicles were defective. In product liability cases, the conduct of a retailer such as an automobile dealer does not form a significant basis for the claims of the class members.
S.Rep.No. 109-14, at 41 (2005), U.S.Code Cong. & Admin.News 2005, at 39.
The Robinson court focused on the Senate’s “small change” comment, concluding that, “[a]s this statement makes clear, whether a putative class seeks significant relief from an in-state defendant includes not only an assessment of how many members of the class were harmed by the defendant’s actions, but also a comparison of the relief sought between all defendants and each defendant’s ability to pay a potential judgment.” Robinson, 2006 WL 468820, at *3. The court found that Gaston, the only individual named in the suit whom the plaintiffs had not ever served with process, was “just small change,” in comparison to the remaining defendants, all out-of-state national companies. While his “alleged negligence may have substantially contributed to the class members’ damages,” the judge held that the plaintiff would seek most of the relief “from those who are capable of paying it: the corporate defendants.” Id., at *4.
Robinson has been quoted in subsequent cases. In some of them, while a defendant’s financial status was listed as a consideration, it was not actually discussed. Evans, 449 F.3d at 1167; Cox v. Allstate Ins. Co., 2008 WL 2167027, at *3 (W.D.Okla.2008); Dunham v. Coffeyville Res., LLC, 2007 WL 3283774, at *3-4 (D.Kan.2007); see Cooper v. R.J. Reynolds Tobacco Co., 586 F.Supp.2d 1312 (M.D.Fla.2008) (noting that, for purposes of § 1332(d)(4)(B), courts used a variety of tests to define the term “primary defendants,” including whether a defendant was “most able to satisfy a potential judgment,” but finding it unnecessary to adopt any test); see generally Phillips v. Severn Trent Envtl. Servs., Inc., 2007 WL 2757131, at *3 n. 4 (E.D.La.2007) (“The Court acknowledges that Plaquemines Parish contends in its cross-claim that its contract with Severn Trent for the maintenance of the Parish’s water treatment contains an indemnity provision, but that has no effect on the determination that the Parish is a significant defendant.”). In others, in which the courts factored in a defendant’s ability to satisfy a judgment, the facts are distinguishable. For example, in Escoe v. State Farm Fire and Cas. Co., 2007 WL 1207231 (E.D.La.2007), the court concluded that an insurance agent, whom it presumed sold insurance policies to only a fraction of the class members and whose ability to pay an potential judgment obviously was much smaller than State Farm’s,18 was not a significant defendant.19 See Cox, 2008 WL 2167027, at *3-4. In none of these cases did the court conclude that a defendant in a position akin to that of BZC was not a significant defendant because of an asserted lack of assets. Also, here, in contrast to most of the decisions in which a defendant’s ability to satisfy a judgment has been considered, all putative class members allege claims against the in-state defendant and seek the identical — injunctive and monetary — relief *1267from all defendants. Petition, ¶¶ 37, 111. See generally Notice of Removal, ¶¶ 30-36.
Defendants argue that unless the phrase “from whom significant relief is sought” is construed as “from whom significant relief may actually be obtained,” the two separate elements of the local controversy exception — significant relief and significant basis — merge. The argument is unpersuasive. While the requirements may, to some extent, overlap, the court concludes that the separate significant relief element is satisfied here, where the petition claims that every potential plaintiff is entitled to recover from the BZC and the proposed class seeks to collect damages from all the defendants jointly and severally. See Kearns, 2005 WL 3967998, at *10 (“[Significant] relief must be measured with respect to that sought by the entire class.”).20
The plaintiffs’ inclusion of other members of the corporate structure as defendants does not mean the class is not seeking to recover its full damages from BZC. Although the plaintiffs have “sued a number of legal entities, but one defendant clearly stands before all the others in importance: [BZC] which own[ed] and operate[d] the [smelter] in question.” Dunham, 2007 WL 3283774, at *3. See generally Laws v. Priority Trustee Servs. of North Carolina, L.L.C., 2008 WL 3539512, at *5 (W.D.N.C.2008) (when plaintiffs were seeking relief from fees law firm received in connection with foreclosures and codefendant trust service, allegedly owned by law firm, neither at the time of the lawsuit nor during the time frame relevant to the plaintiffs’ claims had any assets, trust service was not a defendant from whom significant relief was sought).
Although the question is not free from doubt, the court concludes plaintiffs view of the “significant defendant” element is most consistent with the statutory language actually employed by Congress. The CAFA exception refers to a defendant from whom significant relief is “sought,” rather than a defendant from whom the relief “may be obtained” or “can be collected” or words of similar import. Further, in other portions of the statute, Congress used a different term — that of “principal defendants” — to identify defendants from whom the bulk of any financial recovery would expect to be collected. The Senate Report states:
For purposes of class actions that are subject to subsections 1332(d)(3) and (d)(5)(A), the Committee intends that “primary defendants” be interpreted to reach those defendants who are the real “targets” of the lawsuit — i.e., the defendants that would be expected to incur most of the loss if liability is found.
S.Rep. No. 109-14, at 43, U.S.Code Cong. & Admin.News 2005, at 41. Although it is possible that Congress intended the terms “principal defendant” and “from whom significant relief is sought” to mean the same thing, it did not say so. It expressed the principles or elements in different language and the court assumes that was intentional.
Based on the statutory language and the specific facts before it,21 the court con-*1268eludes that BZC’s status as a significant defendant does not turn on its ability to satisfy any judgment the class may obtain against it.22 Recognizing that the local controversy exception “is a narrow exception that was carefully drafted to ensure that it does not become a jurisdiction loophole,” the court nonetheless concludes that BZC is a significant in-state defendant for purposes of the local controversy exception23 and that this case is “a truly local controversy — a controversy that uniquely affects a particular locality to the exclusion of all others.” Evans, 449 F.3d at 1164 (quoting S.Rep. 109-14, at 39, U.S.Code Cong. & Admin.News 2005, at 38).24 The court does not have subject matter jurisdiction under CAFA.
CERCLA25
“Congress enacted CERCLA to provide a mechanism for the prompt and efficient cleanup of hazardous waste sites.” Cannon v. Gates, 538 F.3d 1328, 1332 (10th Cir.2008), cert. denied, — U.S. -, 129 S.Ct. 1669, 173 L.Ed.2d 1035 (2009) (quoting United States v. City and County of Denver, 100 F.3d 1509, 1511 (10th Cir.1996)). Federal district courts have “exclusive original jurisdiction” of CERCLA actions. 42 U.S.C. § 9613(b). CERCLA does not, however, completely preempt state law claims. See General Elec., 467 F.3d at 1244 (“Given these saving clauses, as well as the spirit of cooperative federalism running throughout CERCLA and its regulations, we may safely say Congress did not intend CERCLA to completely preempt state laws related to hazardous waste contamination.”). “CERCLA as enacted provides no private right of action for personal or economic injury caused by the release of hazardous substances.” Id. at 1246 n. 34. Individuals may “proceed with their state law claims concurrently with a CERCLA remediation in place,” *1269 New Jersey Dep’t of Environmental Protection v. Minnesota Mining & Man. Co., 2007 WL 2027916, at *6 (D.N.J.2007), so long as they are not seeking relief that is inconsistent with ongoing CERCLA-mandated remediation. See Cannon, 538 F.3d at 1332.
The defendants assert that the court has jurisdiction under CERCLA because the plaintiffs, in their petition, challenge an “ongoing CERCLA-quality cleanup.” Defendants’ response, p. 40. They claim the plaintiffs seek “sweeping injunctive relief to dramatically alter the ongoing cleanup that is progressing under the mandate of CERCLA and with EPA oversight....” Notice of Removal, ¶49. The plaintiffs respond that they have simply pleaded state, common law causes of action, which do not provide a basis for removal jurisdiction.
“Congress has provided for removal of cases from state court to federal court when the plaintiffs complaint alleges a claim arising under federal law.” Rivet, 522 U.S. at 472, 118 S.Ct. 921. Generally the existence of federal-question jurisdiction is governed “by the “well-pleaded complaint’ rule, under which a suit arises under federal law only when the plaintiffs statement of his own cause of action shows that it is based on federal law.’ ” Turgeau v. Admin. Review Bd., 446 F.3d 1052, 1060 (10th Cir.2006) (quoting Schmeling v. NORDAM, 97 F.3d 1336, 1339 (10th Cir.1996)). A case does not arise under federal law by either “the plaintiffs anticipation of a federal defense or the defendant’s assertion of a federal defense.” Id. “The plaintiff is the ‘master of the claim’ and may prevent removal [to federal court] by choosing not to plead a federal claim even if one is available.” Id. (internal citation omitted).
The Supreme Court has developed two narrow exceptions to the well-pleaded complaint rule: the complete preemption doctrine, Nicodemus v. Union Pacific Corp., 440 F.3d 1227, 1232 n. 4 (10th Cir.2006) and the substantial federal question doctrine. Grable & Sons Metal Prods., Inc. v. Dame Eng’r & Mfg., 545 U.S. 308, 312, 125 S.Ct. 2363, 162 L.Ed.2d 257 (2005). The complete preemption doctrine allows a state law cause of action to be removed to federal court on the theory that “federal preemption makes the state law claim necessarily federal in character.” Turgeau, 446 F.3d at 1061 (internal quotation omitted). Because CERCLA does not completely occupy the field of environmental regulation, the complete preemption doctrine does not apply here. Under the substantial-federal-question doctrine, a state law claim can give rise to federal question jurisdiction, even though Congress has not created a private right of action, if the right to relief depends on the validity, construction, or effect of federal law. Grable, 545 U.S. at 312, 125 S.Ct. 2363.
The plaintiffs contend the defendants have not identified an actual dispute about federal law and, at best, have a preemption defense to a potential remedy. They assert that their “request for equitable relief, [which] only potentially, if at all, implicates federal law,” does not confer jurisdiction over their claims. Plaintiffs’ motion, p. 26. Relying principally on Ninth Circuit cases, including Fort Ord Toxics Project, Inc. v. California Envtl. Protection Agency, 189 F.3d 828 (9th Cir.1999) and ARCO Envtl. Remediation, L.L.C. v. Dep’t of Health & Envtl. Quality, 213 F.3d 1108 (9th Cir.2000), the defendants respond that jurisdiction exists because federal courts have jurisdiction under CERCLA § 113(b) over any challenges to a CERCLA cleanup and the plaintiffs, through their request for affirmative injunctive relief, have challenged the cleanup.
Few cases have addressed the issue of whether “CERCLA’s involvement with ... *1270state law claims [is] sufficient to warrant the district court’s independent exercise of federal jurisdiction,” General Elec., 467 F.3d at 1242 n. 29 (citing Grable, the Tenth Circuit noted, but did not address the issue, as the district court properly exercised supplemental jurisdiction over the plaintiffs state law claims). Those that have are not in agreement. Compare ARCO, 213 F.3d at 1115-1626 with New Jersey Dep’t of Envtl. Protection v. Occidental Chemical Corp., 2006 WL 2806231 (D.N.J.2006)(limiting federal jurisdiction to completely preempted state law claims)27 In Grable the Supreme Court announced a new test for determining when a state-law claim gives rise to federal jurisdiction, occasioned by a split among the Circuit Courts “on whether Merrell Dow Pharmaceuticals Inc. v. Thompson, 478 U.S. 804, 106 S.Ct. 3229, 92 L.Ed.2d 650 (1986), always requires a federal cause of action as a condition for exercising federal-question jurisdiction.” Id. at 311-12, 125 S.Ct. 2363. The Court concluded that the lack of a federal cause of action did not preclude a lawsuit’s removal to federal court. The test is whether “a state-law claim necessarily raise[s] a stated federal issue, actually disputed and substantial, which a federal forum may entertain without disturbing any congressionally approved balanee of federal and state judicial responsibilities.” Grable, 545 U.S. at 314, 125 S.Ct. 2363.
The plaintiff in Grable brought a quiet title action against the purchaser of property it had previously owned, but which the Internal Revenue Service (“IRS”) had seized and sold to the defendant at a tax sale. The plaintiff claimed the IRS had given a deficient notice of seizure under a federal statute. The defendant removed the ease to federal district court, contending that the plaintiffs claim of title depended on the interpretation of a federal tax law. The Supreme Court concluded that the meaning of the federal statute was actually in dispute because an essential element of the plaintiffs claim was whether it had been given notice as required by the tax law. The Court also found that, because state quiet title actions rarely involve contested issues of federal law, exercising federal jurisdiction in the case “would not materially affect, or threaten to affect, the normal currents of litigation.” Id. at 319, 125 S.Ct. 2363.
As this case is factually distinct from Grable, it offers little guidance in the court’s task of determining whether the plaintiffs’ claims present a substantial federal question justifying removal or if “this is a case of conflict preemption.”28 Gener *1271 al Elec., 467 F.3d at 1244. Support for the conclusion that the plaintiffs’ claims raises disputed and substantial federal issues regarding the scope and impact of CERCLA is found in ARCO, 213 F.3d at 1115-16. In that case the Ninth Circuit had to determine whether the district court had removal jurisdiction over a suit ARCO brought in state court to obtain documents from the State of Montana pertaining to a CERCLA cleanup and to enjoin certain closed-door meetings. ARCO’s claims were based entirely on state law. The answer depended, the Ninth Circuit found, on “whether CERCLA’s jurisdictional provisions deprive[d] the Montana state court of jurisdiction over [the] case, thereby making ARCO’s claims ‘necessarily federal in character.’” Id. at 1115. The court noted that it had expansively read § 113(b) of CERCLA, finding that it conferred on the federal district courts exclusive original jurisdiction over both claims created by/controversies arising under CERCLA and “ ‘any challenge to a CERCLA cleanup.’ ” Id. (quoting Fort Ord, 189 F.3d at 832). ARCO’s claims were “necessarily federal claims under section 113(b),” the Ninth Circuit stated, “only if they constitute] a ‘challenge to a CERCLA cleanup.’” Id.
The court defined such a challenge as an action that is “related to the goals of the cleanup,” id. (internal quotation omitted), giving as examples cases in which the plaintiff sought to dictate specific remedial actions, to postpone the cleanup and to terminate the Remedial Investigation/Feasibility Study and to alter the method and order of cleanup. Id. The court emphasized that “an action d[id] not become a challenge to a CERCLA cleanup simply because the action has an incidental effect on the progress of a CERCLA cleanup.” Id. Because the relief sought by ARCO did not “alter cleanup requirements or environmental standards,” or “terminate or delay the ... cleanup,” it was not “related to the goals of the ... cleanup” and did not arise under CERCLA § 113(b). Id.; see Fort Ord, 189 F.3d at 832 (court rejected plaintiffs’ “cramped interpretation” of § 113(b) of CERCLA, instead reading it to “cover any ‘challenge’ to a CERCLA cleanup”).
Although a close question, the court agrees with the Ninth Circuit’s conclusion that a district court has jurisdiction over state law claims that challenge a CERCLA cleanup because they implicate significant federal issues.29 See North Penn Water Auth. v. BAE Systems, 2005 WL 1279091, at *10-11 (E.D.Pa.2005) (state law claims involving a direct challenge to a CERCLA cleanup completely preempted). Both the government and parties participating in a CERCLA cleanup have a strong interest in ensuring that CERCLA’s objectives are accomplished without “time-consuming litigation which might interfere with CERCLA’s overall goal of effecting the prompt cleanup of hazardous waste sites.” Denver, 100 F.3d at 1514. The court finds that “providing a federal forum for the resolution of this issue will not disrupt the sound division of labor between state and federal courts.” Nicodemus, 440 F.3d at 1237 (internal quotation omitted).
Having concluded that a federal court would have jurisdiction over a plaintiffs state law claims which challenge a CERCLA cleanup, the question then becomes *1272whether the cleanup at issue in this case is “a CERCLA cleanup” for federal question purposes and whether plaintiffs are challenging that cleanup.
The latter question is easy enough. The plaintiffs’ petition specifically attacks, in multiple ways, the “purported” remedial efforts to date:
Indeed, none of Defendants’ purported “remediation efforts” were effective or comprehensive. In fact, Defendants’ investigation and attempted remediation has been and continues to be nothing more than a sham designed to deceive the citizens of Blackwell. The testing protocols and threshold clean up levels employed by Defendants are designed to minimize findings of contamination, wholly fail to investigate contamination inside residents’ homes, and turn a blind eye to well-established and widely accepted benchmarks, such as those adopted and endorsed by the EPA.
Petition, ¶ 14. The plaintiffs further allege the manner in which the defendants operated, closed and remediated the site was “woefully inadequate and further jeopardized the safety and welfare of all individuals residing in and around Blackwell.” Id., ¶ 17. It is plain that the plaintiffs are more than a little critical of the cleanup efforts to date.
Simply being critical is not enough, though, to constitute a “challenge” to a cleanup in this context. The question is whether the plaintiffs’ lawsuit challenges the remediation action by calling into question the remediation plan. Cannon, 538 F.3d at 1335. As explained by the Tenth Circuit, “a suit challenges a removal action if it ‘interferes with the implementation of a CERCLA remedy’ because ‘the relief requested will impact the [removal] action selected.’ ” Id. (quoting Broward Gardens Tenants Ass’n v. EPA, 311 F.3d 1066, 1072 (11th Cir.2002)). However, given the pervasive nature of the plaintiffs’ criticisms of the cleanup to date and the scope of relief sought relative to the existing plan,30 the court readily concludes that plaintiffs have “challenged” the cleanup plan for present purposes. See General Elec., 467 F.3d at 1250 (“The State’s argument that remediation in the South Valley is not working as the EPA and NMED claim constitutes a dispute over environmental cleanup methods and standards.”).
The question of whether the plaintiffs are challenging a “CERCLA cleanup,” as that phrase is used in the context of determining federal question jurisdiction, is considerably closer. The plaintiffs essentially argue that the remediation efforts to date have been an ODEQ-mandated cleanup, rather than a “CERCLA cleanup.” They argue that the EPA has made no formal determination of a hazardous release within the meaning of CERCLA, that it did not place the Site on the National Priorities List, and that it had no direct role in the cleanup efforts. They therefore argue that the EPA was not acting under CERCLA to “arrange for the removal of ... [a] hazardous substance.” 42 U.S.C. § 9604(a)(1). They assert that the MOU with ODEQ was an express deferral of its authority under CERCLA, rather than an exercise of it.
Defendants respond that they were required to remediate the Site to federal standards set out in, or promulgated pursuant to, CERCLA, and that this dispute therefore constitutes a controversy arising *1273under CERCLA. They note the extensive involvement of the EPA in reviewing the plan at various times prior to its adoption and to its extensive monitoring role during plan implementation.
The parties have not supplied, and the court has not otherwise located, statutory or other authority specifically addressing the circumstances here — where the EPA plainly wanted a cleanup to be done, insisted that the cleanup be done according to CERCLA standards and subject to extensive EPA oversight and monitoring, but did not actually do the cleanup itself or order it to be done. The evidence suggests that the approach adopted by the EPA as to the Blackwell site was part of a pilot program designed to rely on state enforcement resources, recognizing that the EPA did not have the resources to directly address every potential site.
While there is considerable force to the defendants’ argument that the cleanup efforts in Blackwell were, in substance, a CERCLA cleanup, the court nonetheless concludes that, in these circumstances, the formal steps taken or not taken by the parties control the designation of the cleanup. The most pertinent document for purposes of characterizing the cleanup efforts at the Blackwell site is the MOU entered into between the EPA and ODEQ, which specifically addressed the basis upon which the parties would go forward. That document recites EPA has received information that there “may” have been a release of hazardous substances but that, as of that date, it had “made no determination as to whether there may be a release or a threatened release.” MOU, ¶ I-B. It also reflected EPA’s recognition that, under CERCLA, a determination by it that a release had occurred would trigger its authority to initiate, or arrange for the initiation of, appropriate removal or remedial action. However, the MOU also explicitly reflects the EPA’s decision to defer making that release determination, in deference to the ongoing state enforcement efforts and in recognition of its own limited resources. Although the EPA retained the right under the MOU to declare it terminated if dissatisfied with the nature or progress of cleanup efforts, there is no indication in the submissions of the parties to suggest that it has ever done so. The references to the site in EPA status records continue to identify the site as one “deferred for further action under State authority” with no suggestion of any formal determination or assertion of authority having been made by the EPA.31
In these circumstances, the court concludes there has been neither a determination of release sufficient to trigger a formal EPA response, 42 U.S.C. § 9604, nor the “selection” of a remedy within the meaning of 42 U.S.C. § 9621 and 40 C.F.R. § 300.430. A cleanup initiated solely under State authority, even if agreed to be conducted consistent with “CERCLA standards” and with the threat of EPA enforcement lurking in the background, does not constitute a “CERCLA cleanup” such as would, upon a challenge to it being initiated, implicate federal jurisdiction. Any other conclusion would permit non-federal parties to turn any cleanup into a “CERCLA cleanup” merely by agreeing to do the cleanup to CERCLA standards. The cases addressing federal question jurisdiction in this context plainly contemplate something more.
The court concludes the plaintiffs’ claims, though highly critical of the Blackwell site cleanup, do not constitute a challenge to a “CERCLA cleanup.” As a result, they do not create a basis for subject matter jurisdiction in this court. And even if a challenge to a CERCLA cleanup was involved, the existence of jurisdiction would depend on a further determination *1274of whether the cleanup is ongoing. Where a CERCLA cleanup is ongoing,32 a federal court lacks jurisdiction to “review any challenges to removal or remedial action.” 42 U.S.C. § 9613(h).
FEDERAL OFFICER REMOVAL JURISDICTION
As the final basis for their removal of this action, the defendants rely on 28 U.S.C. § 1442(a),33 which provides:
(a) A civil action or criminal prosecution commenced in a State court against any of the following may be removed by them to the district court of the United States for the district and division embracing the place wherein it is pending:
(1) The United States or any agency thereof or any officer (or any person acting under that officer) of the United States or of any agency thereof, sued in an official or individual capacity for any act under color of such office or on account of any right, title or authority claimed under any Act of Congress for the apprehension or punishment of criminals or the collection of the revenue.
The defendants’ argument that jurisdiction exists under § 1442(a) is based on the role the smelter played in conjunction with the United States’ efforts to prepare for and enter World War II. In 1942, the Metals Reserve Company (“MRC”), a wartime organization whose mission was to obtain strategic metals for the United States, entered into an agency agreement with American Metals Company, Ltd. (“AMC”), the predecessor entity to Cyprus Amax Minerals and parent company of BZC. Under the terms of the agreement,34 AMC was to act as MRC’s “Agent,” Notice of Removal, Exhibit D, ¶ 24, and was to buy, “receive, smelt and/or refine the ores and concentrates.... ” Id. at ¶¶ 24-25. AMC also agreed to stockpile MRC-owned zinc. Id. at ¶¶ 26, 31. AMC and its subsidiaries, including BZC, operated as a consolidated organization in complying with the agreement.
“A private corporation35 may remove a case under § 1442(a)(1), if it can show: (1) that it acted under the direction of a federal officer; (2) that there is a causal nexus between the plaintiffs claims and the acts the private corporation performed under the federal officer’s direction; and (3) that there is a colorable federal defense to the plaintiffs claims.”36 *1275 Greene v. Citigroup, Inc., 2000 WL 647190, at *2 (10th Cir. May 19, 2000) (unpublished opinion). Most courts, when considering the extent to which a defendant was acting under federal direction “at the time they were engaged in conduct now being sued upon ... require ‘direct and detailed control’ by the federal officer over the defendant.” Bahrs v. Hughes Aircraft Co., 795 F.Supp. 965, 969 (D.Ariz.1992) (quoting Ryan v. Dow Chem. Co., 781 F.Supp. 934, 950 (E.D.N.Y.1992)). Such control was found to exist in Winters v. Diamond Shamrock Chem. Co., 149 F.3d 387, 398—401 (5th Cir.1998). The Fifth Circuit concluded in that case that a private company acted under the direction of a federal officer in conjunction with a contract to supply Agent Orange where the government maintained strict control over the product’s development and production, required the production to be done in accordance with specifications set forth in the contract and documents referenced in the agreement, and performed inspections. Winters, 149 F.3d at 398-401. When a plaintiffs claims are based on pollution or contamination resulting from a facility’s waste disposal, courts have required the removing party to show “the federal government exercised control over the disposal of hazardous substances.” Ward v. Lockheed Martin Corp., 2006 WL 889729, at *5 (M.D.Fla.2006).
The plaintiffs contend that, while the defendants have demonstrated that BZC contracted with the federal government to produce and store zinc, they cannot show that the contamination was done under the direction of a federal officer.37 The defendants respond that the plaintiffs ignore both the fact the AMC/BZC acted as an express agent of the government and the extent to which the Smelter “was dedicated to meeting the wartime needs of the Federal Government.” Defendants’ response, p. 63. The defendants claim that they have shown that AMC and BZC were “acting under” a federal officer, as that term has been recently defined by the Supreme Court in Watson v. Philip Morris Cos., Inc., 551 U.S. 142, 127 S.Ct. 2301, 168 L.Ed.2d 42 (2007). They assert that the Court explained that “the private person’s ‘acting under’ must involve an effort to assist, or to help carry out the duties or tasks of the federal superior,” id. at 152, 127 S.Ct. 2301, and identified “assisting” relationships, two of which are present here. Because AMC and BZC “provided] the Government with a product that it used to help conduct a war,” id. at 154, 127 S.Ct. 2301, and had a “special [agency] relationship” with the government, the “acting under” requirement of § 1442(a) has, the defendants contend, been met.
The defendants focus on only part of the Supreme Court’s opinion, however, ignoring its statement that the word “under” as used in the statute
must refer to what has been described as a relationship that involves “acting in a certain capacity, considered in relation to one holding a superior position or office.” 18 Oxford English Dictionary *1276948 (2d ed. 1989). That relationship typically involves “subjection, guidance, or control.” Webster’s New International Dictionary 2765 (2d ed. 1953). See also Funk & Wagnalls New Standard Dictionary of the English Language 2604 (1942) (defining “under” as meaning “[s]ubordinate or subservient to,” “[s]ubject to guidance, tutorship, or direction of’); 18 Oxford English Dictionary, supra, at 949 (“[sjubject to the instruction, direction, or guidance of’).
Id. at 151-52, 127 S.Ct. 2301. The requirement that the private person be an “assisting relationship” is in addition to the requirement that the person be in a relationship involving “subjection, guidance, or control.” Id. In Watson, the court held that “[t]he upshot is that a highly regulated firm cannot find a statutory basis for removal in the fact of federal regulation alone.” Id. at 153, 127 S.Ct. 2301. Here, a company that was not closely regulated or supervised cannot find a statutory basis for removal solely in the fact that it “help[ed] the Government to produce an item that it need[ed].” Id.38
The defendants have demonstrated that the government clearly had a “significant wartime interest in the operation of the Smelter,” defendants’ response, p. 61. However, while operation of the smelter may have been “of utmost importance to the country’s war effort” and the smelter may have been “in large part dedicated to performing tasks for the Federal Government during this critical time period, id. p. 64, the defendants have not demonstrated an unusually close [relationship] involving detailed regulation, monitoring, or supervision” id. at 153, 127 S.Ct. 2301 required to bring AMC/BZC “within the scope of the statutory phrase ‘acting under* a federal ‘officer.’” Id. at 153, 127 S.Ct. 2301; Bahrs, 795 F.Supp. at 969-70; Brown v. Delta Air Lines, Inc., 2004 WL 5041257 (W.D.Okla.2004) (to demonstrate they were acting under the direction of a federal officer or agency defendants had to “show that their actions were performed pursuant to an officer’s direct orders or comprehensive and detailed regulations .... The official must have [had] direct and detailed control over the defendant[s].”) (internal quotation omitted); see Ward, 2006 WL 889729 at *4-5. Removal was not, therefore, authorized under § 1442(a)(1).39
While this case presents a number of close and difficult questions, the court concludes defendants have not established a basis for federal jurisdiction under any of the theories they have identified. Accordingly, plaintiffs’ motion to remand [Doc. #78] is GRANTED. This case is REMANDED to the District Court of Kay County, State of Oklahoma. As the court does not find the issues the plaintiffs seek to address in their supplemental brief to be dispositive, their motion for leave to file *1277a supplemental brief [Doc. # 179] is DENIED.
IT IS SO ORDERED.