From Casetext: Smarter Legal Research

United States v. Doe Run Res. Corp.

UNITED STATES DISTRICT COURT FOR THE NORTHERN DISTRICT OF OKLAHOMA
Sep 26, 2017
Case No. 15-CV-0663-CVE-JFJ (N.D. Okla. Sep. 26, 2017)

Opinion

Case No. 15-CV-0663-CVE-JFJ

09-26-2017

UNITED STATES OF AMERICA and STATE OF OKLAHOMA, Plaintiffs, v. THE DOE RUN RESOURCES CORPORATION and NL INDUSTRIES, INC., Defendants. ASARCO LLC, Plaintiff-Intervenor, v. UNITED STATES OF AMERICA, STATE OF OKLAHOMA, THE DOE RUN RESOURCES CORPORATION, AND NL INDUSTRIES, INC. Defendants.


OPINION AND ORDER

Now before the Court is plaintiffs' Motion to Enter Consent Decree and Memorandum in Support (Dkt. # 35). Plaintiffs United States of America (United States) and State of Oklahoma (the State) request that the Court approve a proposed consent decree (Dkt. # 3-1) resolving plaintiffs' claims under the Comprehensive Environmental Response, Compensation, and Liability Act, 42 U.S.C. § 9601 et seq. (CERCLA) against defendants The Doe Run Resources Corporation (Doe Run) and NL Industries, Inc. (NL). Plaintiff-Intervenor ASARCO LLC (ASARCO) objects to the proposed consent decree, and it asks the Court to deny plaintiffs' motion to approve consent decree. Dkt. # 42.

I.

Early in the twentieth century, significant deposits of lead and zinc were discovered in southeastern Kansas and northeastern Oklahoma, and mining for these minerals followed shortly after this discovery. This area became known as the Tri-State Mining District (TSMD). Originally, mining activities were relatively small operations, but in the 1920s larger mining companies operated in the TSMD. The peak years of mining production occurred during the 1920s, but mining continued until the 1970s. Mining companies removed ore from the ground, stripped the ore of any valuable metals, and deposited waste material on the surface. Millions of tons of waste were left in the area in the form of chat piles and tailings ponds. Chat was created when mining companies stripped material of its useful ore and the remaining byproduct, chat, was stored on the surface in the form of chat piles. Tailings, another byproduct of mining activity, was left in unlined flotation ponds on the surface. Many of the chat piles and or former chat piles, known as chat bases, still remain; an inventory conducted in 2005 showed that there were still 83 chat piles and 243 chat bases. Dkt. # 35-12, at 8.

In 1980, Congress passed CERCLA due to the "serious environmental and health risks posed by industrial pollution." United States v. Bestfoods, 524 U.S. 51 (1998). As part of the effort to clean up hazardous waste sites, Congress created the National Priorities List (NPL), which is a "list of sites where releases have occurred that have the highest national priority for remediation efforts." Morrison Enterprises v. McShares, Inc., 302 F.3d 1127, 1132 (10th Cir. 2002). Once a site is designated as a Superfund site, the Environmental Protection Agency (EPA) must conduct a remedial investigation and feasibility study (RI/FS). Federal regulations define an RI as:

This case was filed by the United States on behalf of the EPA, and at times the Court will refer to the United States and the EPA separately. References to the United States will mean the named plaintiff in this case, because it is necessary to distinguish between the plaintiff and other federal agencies having a role in this case. To maintain consistency with the relevant statutes and the evidentiary record, the Court will refer to the EPA as the entity taking action to remedy environmental contamination. In addition, the Court will refer to the United States Department of the Interior (DOI) separately from the United States, because DOI is a settling party and it has an adverse interest to the EPA in this case.

a process undertaken by the lead agency to determine the nature and extent of the problem presented by the release. The RI emphasizes data collection and site characterization, and is generally performed concurrently and in an interactive fashion with the [FS]. The RI includes sampling and monitoring, as necessary, and includes the gathering of sufficient information to determine the necessity for remedial action and to support the evaluation of remedial alternatives.
40 C.F.R. § 300.5. An FS more closely concerns the evaluation of possible remedies rather than the broad investigation of a possible environmental hazard. Regulations define an FS as a:
study undertaken by the lead agency to develop and evaluate options for remedial action. The FS emphasizes data analysis and is generally performed concurrently and in an interactive fashion with the [RI], using data gathered during the RI. The RI data are used to define the objectives of the response action, to develop remedial action alternatives, and to undertake an initial screening and detailed analysis of the alternatives. The term also refers to a report that describes the results of the study.
Id. Following the completion of an FS, the EPA must follow a two-step process to select a final remedy. First, the EPA selects the preferred remedy and presents the proposed plan to the public for "review and comment." 40 C.F.R. § 300.430(f)(ii). Second, the EPA must review the public comments and consult with the state (or Indian tribe) to determine if the selected remedy is appropriate. Id. The final remedy is documented in a record of decision (ROD) showing "all facts, analyses of facts, and site-specific policy determinations considered in the course of carrying out activities." Id. at § 300.430(f)(5)(i). The EPA may work toward a comprehensive resolution of the environmental hazards at a complex site by dividing its remedial action into operable units. Id.

In 1983, the EPA placed approximately 40 square miles of northeastern Oklahoma on the NPL and designated the area as the Tar Creek Superfund Site (Tar Creek). Dkt. # 35-12, at 8. The EPA estimates that there is a total of 4,481 "waste acres" within Tar Creek. Dkt. # 35-2, at 4. The EPA's initial investigation identified two risks associated with contamination of groundwater and surface water caused by pollutants in mine shafts. Dkt. # 35-12, at 9. The EPA was concerned about the discharge of acid mine water into Tar Creek and the threat of contamination to the Roubidoux Aquifer, a source of drinking water. Id. The EPA initially determined that chat piles and tailings ponds did not pose a health risk to humans or the environment.

For the purpose of plaintiffs' motion, "waste acres" is defined as "tailings covered land" that includes a "buffer" or "transition" zone. Dkt. # 35-2, at 3. The "buffer" or "transition" zone extends out a set distance from the tailings and is not necessarily covered with tailings, but the soil is likely contaminated with high levels of cadmium, zinc, and lead. Id.

The cleanup of groundwater and surface water was designated as Operable Unit 1 (OU1) and a ROD was issued in selecting a remedy for OU1. Id. The remedy included surface water diversion, construction of dikes, and plugging wells to prevent acid mine water migration. Id. In 1994, the EPA received a report from Indian Health Services that Indian children living in the former TSMD had elevated levels of lead in their blood, and the EPA began to conduct soil sampling at areas where children would congregate, such as schools, parks, and day care centers. Id. In 1995, the EPA began to excavate soil at high access areas using its removal action authority, and the EPA also opened an RI/FS to study soil contamination within residential areas. Id. The soil removal action was officially designated as Operable Unit 2 (OU2) and was converted from a removal action into a remedial action. Id. In 1997, the EPA issued a ROD for OU2 choosing soil removal and backfilling of residential yards as a permanent remedy and, to date, the EPA has cleaned up 2,940 residential yards as part of OU2. Dkt. # 35-4, at 3. Of these 2,940 residential yards, 443 of the residential yards were owned by members of the Quapaw Tribe of Oklahoma (the Tribe) and almost of this land was held in restricted status. Id. at 3. The EPA represents that the cleanup work for OU2 is nearly complete. Id.

A removal action "costs less, takes less time, and is geared to address an immediate release or threat of release." Colorado v. Sunoco, Inc., 337 F.3d 1233, 1240 (10th Cir. 2003). In contrast, a remedial action "seeks to effect a permanent remedy to the release of hazardous substances when there is no immediate threat to the public health." Id.

On March 23, 1999, the Tribe notified the EPA that it had discovered drums of mining chemicals in an abandoned building. See Quapaw Tribe of Oklahoma v. Blue Tee Corp., 2008 WL 2704482, *3 (N.D. Okla. July 7, 2008). The EPA designated the cleanup of these chemicals as Operative Unit 3 (OU3), and the chemicals were quickly removed. Dkt. # 35-12, at 9.

In 2008, the EPA issued its ROD for Operative Unit 4 (OU4), which targeted the cleanup of mill and mining waste left within Tar Creek by historical mining activities. Dkt. # 35-6, at 2-3; Dkt. # 35-12, at 3-6. The selected remedy will be completed in two phases and the EPA originally estimated that the work would cost $167,288,000. Dkt. # 35-12, at 4-5. Phase 1 of the cleanup will "address voluntary relocation of residents, chat sales, and address source materials in a manner that will reduce the overall footprint of contamination and reduce the need for land use restrictions . . . ." Id. at 4. Chat piles and bases will be cleared and the soil underneath the chat piles and bases will be excavated. Id. Marketable chat will be sold for certain uses and non-marketable chat and mining waste will be disposed of in an on-site repository or mine workings. Id. Interim measures will be taken to prevent further contamination of waterways, and waterways will be excavated or a flexible membrane liner will be installed. Id. Phase 2 will address any chat bases, chat piles, and unmarketable chat that remain after the completion of Phase 1, and the selected remedy will be reviewed at least every five years. Id. at 5. As of December 31, 2015, the EPA had incurred almost $170 million in response costs for OU4. Dkt .# 35-13, at 4.

The State and Indian tribes had expressed concern about sediment and surface water contamination downstream of the central mining area in Tar Creek, and the EPA established Operable Unit 5 (OU5) to address these concerns. Dkt. # 35-3, at 3. OU5 is still in the RI/FS stage and no remedy has been selected, but the EPA will examine the nature and extent of contamination in the following bodies of water:

• Four Mile Creek (an upstream background or reference location unaffected by historical mining activites)

• Elm Creek

• Tar Creek (including Lytle Creek)

• Neosho River

• Beaver Creek

• Lost Creek

• Lower Spring River (portion of Spring River downstream of Empire Lake in Kansas, and ending at the headwaters of Grand Lake O' the Cherokees)
Dkt. # 42-13, at 10. As of December 31, 2015, the EPA had incurred $3,106,553 in response costs for OU5. Dkt. # 35-3, at 3.

The mining companies that formerly engaged in mining activities were no longer in business or had been merged into other companies, and the EPA sought to hold the predecessors-in-interest to the former mining companies liable under CERCLA. The EPA requested information from ASARCO, Blue Tee Corporation (Blue Tee), Childress Royalty Company, Incorporated (Childress), Gold Fields Mining Corporation (Gold Fields), NL, and Doe Run about the historical mining operations of the predecessors of these companies. The companies retained an expert, Gary Uphoff, to prepare a Joint Response to September 29, 1994 Information Request, and the parties refer to this document as the Joint Response. Dkt. # 29-1. ASARCO's predecessor was the Federal Mining and Smelting Company (Federal), and it operated on 11 properties within the TSMD from 1918 to 1952. Id. at 19-20. Federal mined a total of 8,172,207 tons of crude ore during its operations, which was about 4.5 percent of the total crude ore mined in the TSMD. Id. at 79-80. Doe Run's predecessor, Kansas Exploration Company (Kansas Exploration) operated on three properties and mined a total of 1,623,863 tons of ore between 1927 and 1949. Id. at 56, 84. This amount represented .9 percent of the total ore mined in the TSMD. Id. at 84. NL was the successor-in-interest to St. Louis Smelting and Refining Company (St. Louis), and the Joint Response suggests that Uphoff had incomplete information about the scope of St. Louis' mining activities. Id. at 64. Based on the available information, Uphoff determined that St. Louis operated at five properties and produced a total of 2,381,694 tons of ore. Id. at 64, 87.

The United States has settled with several potentially responsible parties (PRPs). Eagle-Picher Industries, Inc. (Eagle-Picher) settled with the United States in bankruptcy proceedings, and the United States did not seek to impose site-wide liability on Eagle-Picher. Instead, the settlement amount as to response costs was calculated based on Eagle-Picher's proportionate share of liability based on the cost per acre to remediate surface mining waste. In re Eagle-Picher Indus., Inc., 197 B.R. 260, 267 (Bankr. S.D. Ohio 1996). In October 2007, ASARCO settled with the United States as part of bankruptcy proceedings, and ASARCO agreed to pay approximately $32.7 million to resolve the United States' CERCLA claim for response costs at Tar Creek. Dkt. # 29-4, at 7. ASARCO waived its right to bring any claim against the United States, and ASARCO received protection from the contribution claims of other PRPs. Id. at 13-14. Childress settled its CERCLA liability to the United States for $810,918. United States of America v. Childress Royalty Co., Inc., 14-CV-633-CVE-FHM, Dkt. # 6 (N.D. Okla. Dec. 15, 2014).

Plaintiffs have submitted a proposed consent decree (Dkt. # 3-1) that would resolve plaintiffs' claims against Doe Run and NL under the CERCLA as to OU2, OU4, and OU5 of the Tar Creek Superfund Site. In addition, DOI seeks to resolve its contribution liability for certain parcels of land that were held in restricted status by the Bureau of Indian Affairs (BIA) on behalf of Native Americans. Doe Run has agreed to pay $3,433,137 to the EPA and $62,000 to the State, and NL has agreed to pay $,6,603,590 to the EPA and $225,000 to the State. Dkt. # 3-1, at 11-13. In exchange for these payments, the United States and the State agree not to sue or take administrative action against Doe Run and NL in relation to OU2, OU4, and OU5. Id. at 18-19. However, as to OU4 and OU5, the Unites States and the State reserve the right institute new proceedings if "conditions previously unknown to the EPA and/or the State are discovered" or "information, previously unknown to the EPA and/or the State, is received." Id. at 20, 22. The United States or the State would also be required to show that this newly discovered information indicates that the remedial actions selected by the EPA for the affected properties "is not protective of human health or the environment." Id. The United States and the State also reserve the right to bring a claim against Doe Run and NL for "injury to, destruction of, or loss of natural resources, and for the costs of any natural resource damage assessments." Id. at 24. DOI has agreed to pay $5,000,000 to the EPA, and DOI is subject to the same conditions as Doe Run and NL in terms of reopening the proceedings based on new information or a claim for natural resources damages (NRD). Id. at 14, 20-24. The consent decree would protect Doe Run and NL from contribution claims by any potentially responsible party (PRP). Id. at 27.

The proposed consent decree was published in the Federal Register and the United States has reviewed the 28 public comments that were submitted. Dkt. # 35-24; Dkt. # 35-25. The United States has found no reason to reject or modify the proposed consent decree based on the public comments. Dkt. # 35-10. Plaintiffs filed this case seeking judicial approval of the proposed consent decree. Dkt. # 2. ASARCO filed a motion to intervene in this case on the ground that it has an interest in protecting its right to contribution against Doe Run and NL. Dkt. # 8. The Court permitted ASARCO to intervene, but it denied ASARCO's request to conduct discovery. Dkt. # 28. The United States filed a motion (Dkt. # 35) for approval of the proposed consent decree, and ASARCO has filed a response (Dkt. # 42) in objection to the motion. NL, Doe Run, and the United States have filed replies in support (Dkt. ## 50, 52, 53) of the motion to approve consent decree, and the motion to approve consent decree is fully briefed. The Court has reviewed the parties' briefing and finds that an evidentiary hearing is unnecessary.

II.

The United States asks the Court to approve the proposed consent decree (Dkt. # 3-1), because the payments made by defendants would constitute a substantial payment in light of the risks associated with litigation and the defenses available to Doe Run and NL. ASARCO argues that the proposed settlement is inadequate when the overall scope of contamination at the entire Tar Creek is considered. ASARCO also claims that it was not permitted to participate in the settlement negotiations and that the settlement should be rejected due to procedural unfairness.

Although there are two plaintiffs, the briefing concerning the motion to approve the consent decree is focused exclusively on the settlement of the claims asserted by the United States. ASARCO's complaint in intervention (Dkt. # 29) does not assert any challenge to Doe Run's and NL's settlement with the State. The Court will refer only to the United States, not plaintiffs, when discussing ASARCO's arguments in opposition to the proposed consent decree.

A consent decree "is primarily a means by which parties settle their disputes without having to bear the financial and other costs of litigating." Local No. 93, Int'l Ass'n of Firefighters, AFL-CIO C.L.C. v. City of Cleveland, 478 U.S. 501, 528-29 (1986). A district court may approve or deny the issuance of a consent decree, but a district court may not generally modify the terms of the consent decree. United States v. Colorado, 937 F.2d 505, 509 (10th Cir. 1991). If the district court finds any problems with the consent decree, the district court should advise the parties of any issue that prevents approval of the consent decree and allow the parties to revise it. Id. Public policy favors the early settlement of CERCLA cases in which a government agency charged with the protection of the public interest has negotiated a settlement. United States v. ASARCO, Inc., 814 F. Supp. 951, 954 (D. Colo. 1993).

In the context of CERCLA litigation, a district court should approve a consent decree if it is "reasonable, consistent with CERCLA's goals, and substantively and procedurally fair." United States v. George A. Whiting Paper Co., 644 F.3d 368, 372 (7th Cir. 2011). In evaluating procedural fairness, the district court should consider whether the settlement was negotiated at arms length, and the court should "look to the negotiation process and attempt to gauge its candor, openness and bargaining balance." In re Tutu Water Wells CERCLA Litigation, 326 F.3d 201, 207 (3d Cir. 2003). "Substantive fairness introduces into the equation concepts of corrective justice and accountability: a party should bear the costs of the harm for which it is legally responsible." United States v. Cannons Engineering Corp., 899 F.2d 79, 87 (1st Cir. 1990). The Court has granted ASARCO leave to intervene, but it does not have a right to veto the settlement. United States v. Albert Inv. Co., Inc., 585 F.3d 1386, 1398 (10th Cir. 2009). The Court can approve the proposed consent decree if it finds that the parties' settlement is "reasonable, fair and consistent with CERCLA," even if this cuts off ASARCO's right to contribution from Doe Run and NL. Id.

A.

ASARCO objects to the proposed consent decree on the ground of procedural unfairness, and it claims that the United States has wholly failed to present evidence that the settlement was the result of an arms-length negotiation. Dkt.# 42, at 25-26. ASARCO also objects to the proposed consent decree because it was not permitted to participate in the settlement negotiations. Id. at 27. The United States responds that the settlement was negotiated by experienced counsel on both sides, and it was not necessary for the parties to include ASARCO in the negotiations. Dkt. # 35, at 39.

When evaluating the procedural fairness of a consent decree, the primary consideration for a district court is whether the settlement was negotiated at arms length. Cannons Engineering Corp., 899 F.2d at 86. This requires a reviewing court to examine the "candor, openness, and bargaining balance" of the parties and the sophistication of the parties and their representatives. United States v. Davis, 261 F.3d 1, 23 (1st Cir. 2001). Factors that a court may consider include whether the negotiations were adversarial, the skill and experience of counsel, and the adequacy of discovery or other information-sharing procedures to provide the parties with evidence to support their positions. 55 Motor Ave. Co. v. Liberty Indus. Finishing Corp., 332 F. Supp. 2d 525, 530 (E.D.N.Y. 2004). CERCLA does not require the EPA to permit all PRPs to participate in settlement negotiations, and the EPA is free negotiate and settle with whomever it chooses as long as the EPA acts in good faith. United States v. Grand Rapids, Michigan, 166 F. Supp. 2d 1213, 1221 (W.D. Mich. 2000). The EPA is "under no obligation to telegraph its settlement offers, divulge its negotiating strategy in advance, or surrender prerogatives of strategic flexibility which any negotiator cherishes." City of Bangor v. Citizens Communications Co., 532 F.3d 70, 96 (1st Cir. 2008).

ASARCO argues that the United States has the burden to come forward with evidence concerning the procedural fairness of the proposed consent decree, and it claims that the United States has not offered any evidence tending to show that the parties reached a settlement as the result of an arms-length negotiation. Dkt. # 42, at 26-27. According to ASARCO, the United States has provided only conclusory allegations that the parties conducted arms-length negotiations, and the Court, not the United States, should decide whether the proposed consent decree was the result of a procedurally fair and arms-length negotiation. Id. at 27. ASARCO claims that it did not forfeit a seat at the negotiating table by entering a settlement with the United States, and it argues that it should have been permitted to participate in settlement negotiations due to its right to contribution against Doe Run and NL. Id. at 28.

The Court will initially consider ASARCO's argument that the decision not to include ASARCO in the settlement negotiations is evidence of a lack of procedural fairness. ASARCO settled its CERCLA liability with the United States, and it had a right to seek contribution from other PRPs. There is no dispute that ASARCO had the right to file a separate lawsuit seeking contribution from Doe Run and NL, and approval of the consent decree will cut off ASARCO's right to contribution against Doe Run and NL. See 42 U.S.C. § 9613(f). However, ASARCO has cited no authority that it had a right to be included in settlement negotiations between the United States and other PRPs. CERCLA clearly states that the right of contribution of a settling PRP against non-settling PRPs is subordinate to the interests of the United States, and CERCLA permits the United States to cut off a settling PRP's right to seek contribution by settling with other PRPs. 42 U.S.C. § 9613(f)(2)(C); United States v. Bay Area Battery, 895 F. Supp. 1524, 1535 (N.D. Fl. 1995). ASARCO has been permitted to intervene in this case and file a lengthy brief in opposition to the motion to approve consent decree, and its objections to the proposed consent decree will be fully considered. This is sufficient involvement to ensure that ASARCO's objections to the proposed settlement and its right to contribution will be considered by the Court, and ASARCO's argument that it was not permitted to participate in settlement negotiations does not show a lack of procedural fairness in the proposed consent decree.

ASARCO argues that the United States has not met its burden to produce evidence establishing that the settlement is the result of an arms-length negotiation. The United States responds that formal efforts to gather information from the mining companies, including Doe Run and NL, began in 1994 after the United States received the Joint Response from the mining companies. Dkt. # 53, at 9. The United States reviewed the evidence submitted by the mining companies and gathered its own evidence to determine what areas were affected by the mining activities of each company. Dkt. # 35-2, at 3-4. The United States, Doe Run, and NL were parties in numerous lawsuits arising out of mining activity in Tar Creek, including a lawsuit brought by the the Tribe against ASARCO, Doe Run, NL, and other mining companies. The Quapaw Tribe of Oklahoma et al. v. ASARCO, Inc. et al., 03-CV-846-CVE-PJC (N.D. Okla.). The prior lawsuits are relevant to the negotiations in this case for at least two reasons. First, the parties conducted extensive discovery in Quapaw and other cases, and the parties, including the United States, are in possession of a substantial amount of information about the location and sources of contamination at Tar Creek. This supports a finding that the United States, Doe Run, and NL were in possession of sufficient information to conduct meaningful settlement negotiations and that each party was aware of the risks presented by litigation. Second, the undersigned handled Quapaw and other cases involving Tar Creek, and the cases were vigorously litigated by experienced counsel. The mining companies offered factually and legally complex defenses, particularly as to causation, and the federal defendants in Quapaw successfully defended against the the Tribe's CERCLA claims. See Quapaw Tribe of Oklahoma, 2008 WL 2704482 at *9-22. ASARCO has offered no evidence suggesting that there was any collusion among the United States, Doe Run, and NL, and the parties' conduct in prior litigation gives the Court no reason to believe that the settlement was not the result of an arms-length negotiation.

The Court finds no basis to reject the proposed consent decree due to a lack of procedural fairness. ASARCO's primary arguments are that it was not permitted to participate in settlement negotiations and that the United States has not come forward with evidence describing the settlement negotiations. Neither of these reasons is a basis to reject the proposed consent decree due to a lack of procedural fairness. The record shows that settlement negotiations lasted several years and the parties had a substantial body of evidence to measure the strengths and weaknesses of their respective cases. ASARCO has not shown that a settling PRP with a statutory right to contribution is entitled to participate in the United States' negotiations with other PRPs, and the mere fact that its right to contribution may be cut off by entry of a consent decree does not show that the consent decree lacks procedural fairness. The Court permitted ASARCO to intervene in this case and file its objections to the proposed consent decree, and ASARCO's objections will be thoroughly examined in the evaluation of the substantive fairness of the proposed consent decree. This adequately protects ASARCO's right to voice its objections to the proposed consent decree, and ASARCO has not shown that approval of the proposed consent decree should be denied to a lack of procedural fairness.

B.

ASARCO argues that the United States has not met its burden to show that the proposed consent decree is substantively fair, because the United States has failed to produce sufficient evidence to support the cost estimates of cleanup work and the mining companies' proportional share of liability. Dkt. # 42, at 32-43. ASARCO also argues that the $5 million settlement on behalf of the DOI is not supported by sufficient evidence. Id. at 44-46.

The substantive fairness of a proposed consent decree is primarily based on principles of comparative fault and apportionment of liability. "The logic behind these concepts dictates that settlement terms must be based upon, and roughly correlated with, some acceptable measure of comparative fault, apportioning liability among the settling parties according to rational (if necessarily imprecise) estimates of how much harm each PRP has done." Cannons Engineering Corp., 899 F.2d at 88. The EPA's calculation of comparative fault "should be upheld unless it is arbitrary, capricious, and devoid of a rational basis." George A. Whiting Paper Co., 644 F.3d at 372; see also In re Tutu Water Wells CERCLA Litigation, 326 F.3d 201, 207 (3d Cir. 2003) ("A consent decree only need be 'based on a rational determination of comparative fault, . . . whether or not [a district court] would have employed the same method of apportionment"). The level of precision with which a court evaluates the method of allocution of fault may be reduced if the historical record is incomplete or reliable information is unavailable. United States v. Charles George Trucking, Inc., 34 F.3d 1081, 1088 (1st Cir. 1994). When the cost of remedial measures is uncertain, a consent decree should be approved if the EPA has provided figures that "derive in a sensible way from a plausible interpretation of the record." ASARCO, Inc., 814 F. Supp. at 955. A district court reviewing the substantive fairness of a proposed consent decree should also consider that a principal purpose of CERCLA was to achieve prompt settlement of claims and the expeditious recovery of funds for the cleanup of environmental contamination. Grand Rapids, Michigan, 166 F. Supp. 2d at 1224.

The Court will begin the substantive fairness analysis by reviewing the parties' settlement as to each of the operable units for which Doe Run's and NL's CERCLA liability is being resolved by the consent decree. The EPA initiated OU2 to remedy high levels of lead contamination in the soil at Tar Creek, and this work primarily involved the excavation and removal of contaminated soil. Doe Run has agreed to pay a total of $3,433,137 to the EPA for the cleanup of OU2, OU4, and OU5. This settlement includes $292,344 for the cleanup of OU2, even though the EPA admits that it had little evidence linking Doe Run to the contamination of residential yards. Dkt. # 35-2, at 5. The EPA assigned liability to Doe Run for the cleanup of six residential properties located within 200 feet of tailings linked to the historical mining operations of Kansas Explorations, for a total of $267,552, and Doe Run agreed to pay an additional $97,878 for its percentage of site-wide costs of OU2 remediation. Id. at 17. Doe Run received a settlement discount of 20 percent, resulting in a total payment of $292,344 for its liability as to OU2. Id. The EPA also had little evidence linking NL to the contamination of residential areas, but for the purpose of settlement the parties agreed that there were two residential properties within 200 feet of tailings associated with NL. Id. at 5. NL agreed to pay $89,184 for remediation of the two properties, plus an addition $32,626.15 for site wide cleanup costs, resulting in a total $121,810.15. Id. at 11. The final amount of NL's settlement for OU2 was $97,448.12 after a 20 percent settlement discount was applied. Id.

ASARCO objects to the proposed settlement as to OU2, because the EPA did not rely on an updated cost estimate to determine defendants' liability for OU2. Dkt. # 42, at 33. ASARCO's expert, Uphoff, states that the 200 foot transition zone used by the EPA to identify properties potentially affected by tailings from defendants' mining waste is not a useful measure in this instance, because chat was removed from defendants' properties and used throughout the local communities as building material. Dkt. # 42-5, at 9. Uphoff asserts that the settlement "greatly underestimates" defendants' potential liability for cleanup costs as to OU2. Id. at 8. The United States responds that it was reasonable to rely on the amount of costs incurred as of 2012, instead of a cost estimate from 2015, because the properties at issue had already been remediated and it was more accurate to rely on the actual costs to determine defendants' settlement liability. As to Uphoff's criticisms of the 200 foot transition zone, the United States responds that this was a reasonable method to calculate defendants' liability for settlement purposes, because this figure is based on data collected by the United States Fish and Wildlife Service showing that very few properties outside of the 200 foot transition zone had elevated levels of metals in the soil. Dkt. # 35-18, at 18-19. The Court finds that it was reasonable for the United States to use a 200 foot transition zone to calculate the amount of defendants' settlement liability, because this accounts for the risks of proving causation if the case were litigated and it is supported by data. ASARCO argues that the United States should have sought to recover more from defendants, because mining waste was moved from defendants' property throughout the local community and defendants may have greater site wide liability for OU2. However, ASARCO has offered no meaningful way to measure how much chat or mining waste from each defendant was used as building material, and ASARCO's argument actually highlights how difficult it would be for the United States to establish the source of the contamination of residential areas if the case were litigated. The Court finds that the settling parties used a reasonable method to apportion defendants' liability for OU2, and the amount of the settlement is reasonable as to OU2.

ASARCO makes this argument concerning the redistribution of chat or mining waste in contexts other than in opposition to the proposed settlement as to OU2. The argument suffers from the same flaw in every other context, because ASARCO has provided no useful means to determine how much chat from each PRP's property has been redistributed. The EPA has used the number of waste acres as a means to calculate the amount of contamination caused by a settling party, and the Court separately will consider ASARCO's objections to this method of calculating defendants' settlement liability. However, when the available evidence is imprecise and subject to dispute, the Court finds that it should defer to the EPA's methodology in crafting a settlement with PRPs, because interpretation of this type of evidence is within the scope of the EPA's expertise and there is a strong judicial policy in favor of resolving disputes by means of settlement. See WildEarth Guardians v. Jackson, 2011 WL 4485964 (D. Colo. Sep. 27, 2011). ASARCO's argument concerning the redistribution of chat falls squarely within the EPA's expertise in evaluating historical evidence as to causes of contamination and the amount of a settling defendant's liability, and the Court does not find that ASARCO's argument tends to show that the proposed consent decree is substantively unfair or unreasonable.

OU4 is intended to remedy the high volume of surface mining waste in the form of chat piles, chat bases, and mill ponds, because the mining waste contains hazardous substances such as lead, cadmium, and zinc. The EPA developed maps showing the location of historical mining activity in Tar Creek, and determined that 46.5 waste acres are attributable to the activities of Doe Run's predecessor, Kansas Exploration, and 107.9 waste acres are attributable to NL's predecessor, St. Louis. Dkt. # 35-2, at 8. To determine the amount of waste acres, the EPA used a 50 foot transition zone beyond the location of any tailings or chat on property associated with defendants' historical mining activity, because this land likely contains contaminated soil even if no tailings or chat still remain. Dkt. # 35-2, at 3. There is a total of 4,481 waste acres in Tar Creek, and defendants' combined percentage of waste acres is 3.4 percent. Id. The EPA sought a settlement for three separate types of costs as to OU4: "(1) a proportionate share of OU4 past response costs; (2) a proportionate share of OU4 site wide costs; and (3) EPA's estimate of future response costs that will be incurred to clean up the 154.4 waste acres attributable to both Doe Run and NL operations." Dkt. # 35, at 30-31. Doe Run has agreed to pay $360,855 for past response costs, $947,240 for its proportional share of site wide response costs, and $2,106,086 for future estimated response costs. Dkt. # 35-2, at 4. This results in a total of $3,706,525, but the parties agreed to a reduced amount of $3,414,181 after further settlement negotiations. Id. NL has agreed to pay $837,804 for past response costs, $2,667,804 for its proportional share of site wide costs, and $2,767,097 for future estimated response costs, resulting in total settlement of $6,272,705 for OU4.

The EPA established OU5 to address "contaminated sediments and surface waters within and downstream of the central mining area" at Tar Creek. Dkt. # 35-3, at 3. OU5 will address sediment contamination in Elm Creek, Tar Creek, Lytle Creek, and the Neosho River, but the entire Spring River is not within the scope of OU5 and the parties have not reached a settlement as to any potential liability for the Spring River. Dkt. # 3-1, at 8; Dkt. # 35-3, at 3. The EPA is currently conducting a RI/FS for OU5 and it has not yet selected a remedy. Dkt. # 35-3, at 3. As of December 31, 2015, the EPA had incurred $3,106,553 in response costs for OU5, and the estimated future cleanup costs for OU5 is approximately $25 million. Dkt. # 35-3, at 4-5. This estimate is based in part on the actual costs incurred in OU4 for sediment removal and the installation of a flexible membrane liner within streambeds. Id. at 3-4. The EPA used a 200 foot transition zone for OU5 to determine if defendants' tailings or chat were located close enough to the relevant streams or waterways to have potentially caused contamination of surface water. Dkt. # 35-2, at 6. The EPA has determined that Doe Run does not currently have any liability as to OU5, because the mining waste left behind by Doe Run's predecessor is not located near the streams being remediated in OU5. Id. at 6. However, NL has agreed to pay $300,000 to resolve its liability for future response costs as to OU5. This amount includes a payment of $175,094 for its proportional share of liability for future response costs, plus and additional payment of $124,906 due to the uncertainty associated with the remedy for OU5. Dkt. # 35-2, at 6.

ASARCO argues that the cost estimates for OU4 and OU5 are speculative and imprecise, and it further claims that the cost estimates are not supported by sufficient evidence. Dkt. # 42, at 33. The United States has submitted evidence showing that the initial cost estimate for OU4 was $167,288,000, but the costs incurred to date have exceeded that amount by nearly $3 million. Dkt. # 35-12, at 4; Dkt. # 35-13, at 4. ASARCO argues that it is "impossible" for the Court to evaluate the proposed settlement as to future response costs for OU4 without solid evidence establishing the cost of the selected remedy for OU4. Dkt. # 42, at 33. The Court finds that the settlement accounts for uncertainty as to future response costs as to OU4, because the settlement amounts for future response costs include a multiplier for uncertainty and a cost overrun premium and the settling parties were clearly aware that the actual cost of completing the work could be greater than the EPA's initial estimate. Dkt. # 35-12, at 13-15, 19-20. NL's settlement as to OU5 includes almost $125,000, in addition to a payment of $175,000 for its proportional share of future response costs, to account for the uncertainty as to the remedy. The Court also notes that the proposed consent decree contains a provision allowing the United States to institute a new civil case or administrative proceedings against Doe Run or NL if the EPA receives new information unknown at the time of settlement that requires the EPA to incur unforseen response costs for the completion of OU4 and OU5. Dkt. # 3-1, at 20, 22. The Court finds no basis to reject the proposed consent decree as to OU4 or OU5 due to uncertainty as to future response costs.

ASARCO makes a related argument that the consent decree should be rejected to due the United States' failure to distinguish between near stream response costs for OU4 and the sediment and water cleanup contemplated in OU5. Dkt. # 42, at 52. The documents submitted by the United States are clear that OU5 primarily concerns areas downstream of the chat piles and bases being remediated as part of OU4, and ASARCO's argument is not a basis to reject the proposed consent decree. --------

ASARCO claims that the use of waste acres to determine the amount of a settling party's liability has been "debunked" as a viable means to measure the amount of contamination caused by a PRP. Dkt. # 42, at 36. The settlements for OU2 and OU5 use a 200 foot transition zone and the settlement for OU4 uses a 50 foot transition zone to calculate the total number of waste acres for which each defendant is responsible for response costs, and the transition zones extend out from tailings located on property associated with the predecessors of Doe Run and NL. Dkt. # 35-2, at 3-6. ASARCO argues that there is evidence showing that there is a high concentration of metals in the soil and water beyond these transition zones. Dkt. # 42, at 37-39. According to ASARCO, it would be more accurate to use the amount of ore mined by each PRP, because the small transition zones placed around existing mining waste on defendants' properties significantly underrepresents defendants' CERCLA liability. Id. at 39. ASARCO is correct that there is data showing that areas beyond the transition zones can have concentrations of metals in excess of the EPA's action levels. See Dkt. # 35-18, at 17. However, this does not by itself show that the proposed consent decree is substantively unfair. The evidence shows that the highest concentration of metals in the soil is located 50 feet from chat piles, even though there is metal in the soil at greater distances, and the mean transition zone for the entire site is 167 feet. Id. This shows that a transition zone of 200 feet for OU2 and OU5 has evidentiary support, and it also shows that defendants have an evidentiary basis to contest a transition zone greater than 200 feet. The Court finds that it is appropriate to defer to the expertise of the EPA in crafting a transition zone for settlement purposes, and the Court also finds that the transition zones used to calculate defendants' CERCLA liability take into account the substantial litigation risks of proving that defendants are responsible for metals outside of the selected transition zones. See United States v. Kerr-McGee Corp., 2008 WL 863975 (D. Colo. Mar. 2008) (district court should take into account the potential risks of litigation and the parties' interest in reaching a compromise when reviewing a consent decree).

ASARCO argues that the proposed consent decree is inadequate as to NL, because the EPA underestimated the scope of St. Louis' mining activities and failed to impose liability for all of the locations where NL's predecessor operated. Dkt. # 42, at 39-43. The EPA states that it relied on "[h]istoric information, EPA documents and the Joint Reponse" to determine the scope of historic mining activity by the predecessors of Doe Run and NL. Dkt. # 35-2, at 2-3. However, ASARCO faults the United States for attaching only the Joint Response to its motion to approve the consent decree. The Joint Response was prepared by Uphoff on behalf of several mining companies, including NL, and Uphoff now claims that his prior assessment of NL's leaseholds and mining activity in Tar Creek was based on an incomplete record. Dkt. # 42-5, at 5. Uphoff states that he has reviewed new information and he believes that NL's predecessor may have reprocessed tailings or received royalties for mining production on leased that property. Id. at 6. ASARCO has submitted newspaper articles from 1917 and 1918 suggesting that NL may have had broader mining operations than is represented in the Joint Response. Dkt. # 42-14; Dkt. # 42-15. The Court does not find that ASARCO's arguments concerning Uphoff's reassessment of NL's historic mining activities provide any basis to reject the proposed consent decree. The EPA represents that it considered evidence other than the Joint Response in determining NL's CERCLA liability, and the EPA sought to hold NL liable for lands where its predecessor operated and lands that were leased to other mining companies. Dkt. # 35-2, at 4. Uphoff now claims that there may be more mining production for which NL could be held liable under the theory advanced by ASARCO, but the Court has found that it was reasonable for the EPA to base the settlement on waste acres and not the volume of ore mined or milled. ASARCO has not identified any specific parcel of land or additional acreage that could be attributed to mining activities by NL's predecessor, and the Court finds that ASARCO's speculation that such properties could exist does not show that the EPA conducted an inadequate investigation into NL's potential CERCLA liability.

The Court has considered the evidence submitted by the United States and ASARCO's arguments in opposition to the proposed consent decree, and the Court finds no basis to reject the proposed consent decree as substantively unfair. The settlement amounts are supported by sufficient evidence and the amounts are reasonable in light of the risks and expenses associated with litigation.

C.

ASARCO argues that the proposed consent decree is unreasonable as to DOI, and the $5 million settlement significantly underestimates the potential liability of DOI for its role at Tar Creek. Dkt. # 42, at 45-47. ASARCO claims that DOI should be treated as any other PRP, and ASARCO asserts that DOI has substantial liability for its role as an owner or operator of lands where contaminants were deposited. Id. The United States responds that the settlement is reasonable as to DOI, because the settlement only includes specific parcels of land that were held in restricted status by the BIA and there is a significant risk that DOI would prevail if the case were litigated. Dkt. # 35, at 46; Dkt. # 53, at 29.

DOI has agreed to pay $5 million to resolve its liability as to OU2, OU4, and OU5. Dkt. # 3-1, at 14. The EPA and the State agree that they will not take any administrative action against DOI pursuant to CERCLA or the Resource Conservation and Recovery Act, 42 U.S.C. § 6901 et seq. (RCRA) for cleanup costs for OU2, OU4, and OU5 upon payment of the $5 million. Id. at 19. The settlement is not for the entire Tar Creek site but is limited to DOI's liability for land within OU2, OU4, and OU5 that was or is held in restricted status for the benefit of Native Americans. Id. at 21, 23. The EPA and the State have reserved the right to seek response costs from the DOI if "conditions at these Restricted Parcels, previously unknown to EPA and/or the State, are discovered" or the EPA discovers new evidence that would show that the selected remedial action at OU4 or OU5 is inadequate. Id. The United States represents that DOI took the position in settlement negotiations that it had no liability at Tar Creek, because DOI did not own the land in question and it did not engage in mining activity. Dkt. # 35, at 33. In addition, DOI claimed that the fiduciary liability provisions of CERCLA limited DOI's liability. Id. The parties characterize the liability of DOI not as direct liability under CERCLA but, instead, contribution liability to mining companies that actually operated on the restricted parcels. Dkt. # 35, at 33-34; Dkt. # 53, at 26-27. In determining the amount of the settlement, DOI agreed to pay only a 30 percent share of the response costs for the restricted parcels based on its assertion that the mining companies were primarily responsible for any contamination and that DOI did not own the land or engage in mining activity on the restricted parcels. Id. at 34.

ASARCO argues that it was unreasonable for the settling parties to limit DOI's liability to contribution claims, because DOI was an owner or operator under CERCLA and DOI can be held jointly and severally liable for the entire amount of the response costs for OU2, OU4, and OU5. Dkt. # 42, at 44-46. ASARCO's expert, Uphoff, opines that "DOI made day-to-day operation decisions, defined where facilities could be located, where wastes should be deposited and collected and dispersed money to the Native American landowners," and he claims that DOI is subject to liability as an owner or operator under CERCLA. Dkt. # 42-5, at 12. However, ASARCO has not actually cited a case or statute suggesting that CERCLA imposes joint and several liability on a government agency for supervising mineral leases on restricted lands. ASARCO has advanced a theory under which DOI might be held liable as an owner or operator, but this would be a novel application of the law and the factual record supporting ASARCO's argument is not well developed. The Court finds that there is substantial uncertainty that DOI could be held liable as an owner or operator under CERCLA, and it is reasonable for the settlement with DOI to account for this uncertainty by calculating DOI's liability in terms of contribution.

ASARCO claims that the proposed settlement does not impose liability on DOI for all of the restricted parcels within OU2, OU4, and OU5, and it argues that the United States has not provided sufficient information for the Court to fully evaluate the proposed settlement. Dkt. # 42, at 47. The settlement amount has been calculated based on DOI's liability for contribution to Gold Fields, Blue Tee, NL, and Doe Run. The United States has submitted charts showing the scope of each mining company's activities, and the charts also show which specific parcels of land were held in restricted status. Dkt. # 35-2, at 9-10; Dkt. # 53-2, at 5-6. ASARCO argues that other mining companies operated within Tar Creek and there are restricted parcels on those lands that have not been included in the settlement, and ASARCO asserts that DOI should be required to settle its liability as to these other restricted parcels. Dkt. # 42, at 47. However, DOI's settlement is based on its liability for contribution, and the other mining companies have already settled with the United States. Eagle-Picher, ASARCO, and Childress have waived the right to bring any claim against the United States, and this includes the right to seek contribution from DOI. The United States is entitled to keep the benefit of its bargain with prior settling PRPs. The Court does not find that the failure to include restricted parcels within the scope of prior settlements with PRPs that waived the right to seek contribution from the United states renders the proposed consent decree in this case unreasonable.

The Court finds that the $5 million settlement of DOI's liability for response costs for OU2, OU4, and OU5 is reasonable in light of the substantial litigation risks faced by the United States in prosecuting a CERCLA claim against DOI. There is a significant risk that DOI would be found to have no liability for historical mining activities on restricted parcels, and the settlement represents a meaningful compromise by the DOI and the United States to account for the uncertainty of litigation. The United States has presented sufficient evidence for the Court to identify the restricted parcels at issue, and it was reasonable for the parties to assess DOI's liability in terms of contribution.

D.

ASARCO contends that the proposed consent decree is unreasonable and inconsistent the purpose of CERCLA, because the settlement fails to address sediment and surface water contamination for all rivers and streams within the scope of OU5. Dkt. # 42, at 49-52. ASARCO also argues that the settlement unreasonably rewards late settling parties and is insufficient to compensate the public for the harm caused by the acts of defendants' predecessors. Id. at 53.

A consent decree should be approved if the settlement is reasonable and not inconsistent with the purposes of CERCLA. CERCLA was enacted with the purposes of providing the federal government a tool to promptly and efficiently clean up environmental contamination and to hold those who caused the contamination responsible. Cannons Engineering Corp., 899 F.2d at 90. When reviewing the reasonableness of a settlement, a district court should consider "the strength of a plaintiff's case, the good faith efforts of negotiators, the possible risks of and transaction costs involved in litigation under CERCLA, and the effect of settlement on non-settlers." Atlantic Richfield Co. v American Airlines, 836 F. Supp. 763, 772 (N.D. Okla. 1993). A district court assessing the reasonableness of a proposed consent decree should take a "broad view" of the settlement and "leav[e] highly technical issues and relatively petty inequities to the discourse between the parties." ASARCO, Inc., 814 F. Supp. at 955.

ASARCO argues that the proposed consent decree is unreasonable as to NL's settlement of its liability as to OU5, because the settlement does not include all waters within the RI/FS for OU5 and NL is not making any payment for the historical mining operations of its predecessor along Beaver Creek, Spring River, or Lost Creek. Id. at 49. The EPA has not selected a remedy for OU5 and ASARCO argues that any settlement of NL's liability as to OU5 is premature until the RI/FS is completed. Id. at 51. The Court does not find that any of these arguments is a basis to reject the proposed consent decree. The consent decree purposefully defines OU5 for settlement purposes more narrowly than the scope of the EPA's RI/FS for OU5. Dkt. # 3-1, at 8. The parties are aware that OU5 as defined in the consent decree does not include the entire Spring River and other streams, and the EPA states that it will separately address the settling parties' potential liability for the Spring River. Dkt. # 35, at 24. The EPA acknowledges that it has not selected a remedy for OU5 and it does not have a complete evidentiary record, but it states that uncertainty as to future cleanup costs was considered and a premium was added to NL's settlement as to OU5 to account for this uncertainty. Dkt. # 35-2, at 6; Dkt. # 53, at 23-24. The Court finds that this adequately accounts for any gaps in the evidence at this stage of the investigation of OU5.

ASARCO argues that the amounts being paid by the settling defendants are inadequate to compensate the public in light of the cost of site-wide remedial measures, and the settlement does not reflect the strength of the United States' CERCLA claims against the settling defendants. ASARCO's argument concerning the public interest in the settlement is simply a reassertion of its argument that Doe Run and NL should be held liable on a site-wide or volume based theory, rather than a waste acres calculation, and the Court has already determined that it was reasonable for the United States to apportion defendants' settlement liability based on waste acres. As to the strength of the United States' case against Doe Run and NL, ASARCO argues that Doe Run and NL have significant monetary resources and the United States has a strong case against these defendants. The Court has already discussed the potential defenses that could be asserted by Doe Run and NL, and it is clear that litigation of CERCLA claims against defendants would result in extended litigation with an uncertain outcome. ASARCO's brief argument that the United States has a "strong" case against defendants, without any further explanation, ignores the significant risks of litigation and the expenses associated with that litigation.

The Court finds that the settlement amounts in the proposed consent decree are reasonable in light of the amount of contamination that can be tied to the historical mining activities of defendants' predecessor and the risks associated with litigation. The Court must also take into account the strong policy in favor of settlement of CERCLA liability. ASARCO argues that it paid more to resolve its CERCLA liability, and that defendants' settlement should be deemed unreasonable when compared to the amount paid by ASARCO to settle its CERCLA liability. However, the fact that ASARCO may have voluntarily agreed to pay more does not have any bearing on the reasonableness of the proposed consent decree in this case. Judicial approval of the proposal will cut off ASARCO's right to contribution from Doe Run and NL and the Court has taken this factor into account in assessing the reasonableness of the proposed consent decree. The Court does not find that this factor renders the settlement unreasonable, because there is a strong interest in resolving CERCLA claims by means of settlement and ASARCO's right to contribution does not outweigh the United States' interest in obtaining funds to clean up environmental contamination.

IT IS THEREFORE ORDERED that plaintiffs' Motion to Enter Consent Decree and Memorandum in Support (Dkt. # 35) is granted.

IT IS FURTHER ORDERED that the plaintiffs are directed to submit a copy of the consent decree to the CM/ECF intake box in Wordperfect format by September 28, 2017.

DATED this 26th day of September, 2017.

/s/_________

CLAIRE V. EAGAN

UNITED STATES DISTRICT JUDGE


Summaries of

United States v. Doe Run Res. Corp.

UNITED STATES DISTRICT COURT FOR THE NORTHERN DISTRICT OF OKLAHOMA
Sep 26, 2017
Case No. 15-CV-0663-CVE-JFJ (N.D. Okla. Sep. 26, 2017)
Case details for

United States v. Doe Run Res. Corp.

Case Details

Full title:UNITED STATES OF AMERICA and STATE OF OKLAHOMA, Plaintiffs, v. THE DOE RUN…

Court:UNITED STATES DISTRICT COURT FOR THE NORTHERN DISTRICT OF OKLAHOMA

Date published: Sep 26, 2017

Citations

Case No. 15-CV-0663-CVE-JFJ (N.D. Okla. Sep. 26, 2017)