City of Wichita, Kansas v. Trustees of APCO Oil Corp. Liquidating Trust

306 F. Supp. 2d 1040, 58 ERC (BNA) 1364, 2003 U.S. Dist. LEXIS 24812, 2003 WL 23315705

• Court: District Court, D. Kansas
• Decided: December 31, 2003
• Docket: CIV.A.98-1360-MLB
• Status: Published

Opinion

MEMORANDUM DECISION

BELOT, District Judge.

TABLE OF CONTENTS

I. INTRODUCTION.1045

II. SUMMARY OF THE DISPUTE . 1046

III. CERCLA LIABILITY. 1048

A. Covered Person Under CERCLA.-..1052

B. Release of any Hazardous Substance.'..1056

1. APCO.1057

2. Reid Supply Company. 1069

3. Land Tool Company. 1075

4. Tri-Supply..'.1076

C. The Releases Caused Plaintiff to Incur Costs..1076

D. Response Action Compliance with the NCP .'.1076

1. The NCP. 1077

2. General Description of City’s Response Action.:..1078

3. Standard of Review for KDHE Determinations...1081

4. Presumption of NCP Compliance. 1082

5. Defendants’ Contentions.'.1084

6. Conclusion.:.1090

E. Necessity of Response Costs.1090

1. CDM Costs.1092

2. KDHE Costs.'...,.1093

3. Miscellaneous Vendor Costs.'.'.1094

4. Stinson, Mag & Fizzell Costs.......1095

5. City Payroll Costs . 1095

6. Motor Pool Expenses. 1096

7. Summary of Recoverable Costs.•... '._1096

F. Arguments Raised in Defense...1096

IV. ALLOCATION OF RESPONSE COSTS.1097

A. Equitable Factors..'.’.1098

1. Gore Factors.‘.'.1098

2. Other Equitable Factors.-..1101

3. Conclusions Regarding Equitable Factors. 1105

B. Methods for Allocating Response Costs.1106

1. Motion to Strike Testimony of Michael Smith...'... 1107

2. ALT’s Modeling and Allocation ..:.1110

3. Reid’s Allocation. 1113

4. Tri-Supply’s Allocation.1114

5. Land Tool Company’s Allocation.1115

C. Prejudgment Interest....1115

D. Declaratory Judgment. 1117

V. REID’S MOTION TO RECONSIDER. 1118

VI.CONCLUSION....1120

APPENDICES . 1120

I. INTRODUCTION

Plaintiff City of Wichita (City) brings this private party action against (1) the Trustees of the APCO Oil Corporation Liquidating Trust (ALT); (2) Reid Supply Company, Inc., Charles P. Trombold, David G. Trombold and Walter S. Trom-bold; (3) Land Tool Company and E'H. Land; and (4) Tri-Supply Company, 'Inc., and Gordon Kratz, pursuant to the Comprehensive Environmental Response, Corn-pensation and Liability Act (CERCLA), 42 U.S.C. § 9601 et seq., as amended by the Superfund Amendments and Reauthoriza tion Act of 1986 (SARA), Pub.L. No. 99-499, 100 Stat. 1613 (1986). The City seeks to recover past response costs from defendants, in the form of contribution pursuant to CERCLA § 113(f). 42 U.S.C. § 9613(f)- The City also seeks declaratory relief that two of the defendants, ALT and Reid Supply, are responsible for future response costs, pursuant to CERCLA § 113(g)(2). 42 U.S.C. § 9613(g)(2).

At the outset, it is important to comment on the size and complexity of this case and on the level of professionalism of counsel and their support staff. This Memorandum Decision represents the culmination of countless hours and resources expended by each of the parties involved in preparing for and litigating the case, and by the court in attempting to decide the many factual and legal issues. The initial complaint was filed on October 7,1998, and was the first of over 1,440 documents filed in the case. Numerous pretrial rulings were made, either by the magistrate judge or by this court. 1 The trial itself included 8 weeks of testimony concerning CERC-LA’s various complexities and the technical issues related to the parties’ potential liability, and an on-site visit during which the court and counsel viewed first-hand the geographic area at issue. The parties provided the court with an array of computer monitors in the courtroom which allowed the court, counsel, and all 38 witnesses immediate access to the hundreds of exhibits presented during the trial. Technicians were present throughout the trial to operate the computer equipment and were very efficient in doing so. The trial ultimately resulted in a transcript exceeding 5,000 pages. All counsel conveyed a level of preparation and skill in their representation that surpassed the court’s expectations in light of the difficult nature of the subject matter. The court sincerely appreciates the efforts made by all those involved in resolving this case.

In reaching its decision, the court has thoroughly considered the record in its entirety, including the voluminous trial transcript, the post-trial briefing and the court’s own assessments of witness credibility. For the reasons stated, the court rules that ALT, Reid Supply Company, Land Tool Company, Walter Trombold, and E.H. Land are liable for past response costs, but only for amounts proportional to the groundwater contamination caused by the respective business entities. The court furthermore declares that ALT and Reid Supply are liable for future groundwater remediation costs in proportion to their individual contributions to the groundwater contamination at the site. Finally, the court declares that ALT and Reid will be liable for any future source control measures at the former APCO and Reid Supply facilities, respectively.

II. SUMMARY OF THE DISPUTE

In 1986, the Kansas Department of Health and Environment (KDHE), an administrative agency of the State of Kansas, collected groundwater samples at multiple locations within the Gilbert and Mosley Site (the Site) and discovered high levels of chlorinated volatile organic compounds, also known as chlorinated solvents. Chlorinated solvents do not occur naturally either in soil or groundwater. They include the commercially-available products Tetrachloroethene or Perchloroethylene (PCE) and Trichloroethene (TCE), and their degraded products, Dichloroethene (DCE) and Vinyl Chloride (VC), neither of which is commercially available. The Site, located near the center of Wichita, Kansas, is named after the intersection of Gilbert Street and Mosley Street, which lies near the middle of the contaminated area. The boundaries of the Site have been delineated through a series of investigations conducted by the KDHE and others since 1986. The Site encompasses an area of roughly four and a half miles from north to south and one to two miles from east to west. It covers approximately 3,850 acres of both residential and commercial property. The Site includes about 8,000 separate parcels of land, and the City itself owns hundreds of parcels within the Site. The groundwater at the Site generally flows in a southerly direction, although it can shift to the south/southwest or the south/southeast in certain locations. The chlorinated solvents moving within the flow of groundwater have created “plumes” of contamination, designated plumes A-F. The plumes lie beneath approximately 1,800 acres of land and contain over 2.75 billion gallons of ground water. See App. C-l. The principal contamination plumes involved in this case are A, B, and E.

Following the discovery, the KDHE entered into a Multi-Site Cooperative Agreement with the EPA (EPA/KDHE multi-site agreement), which required investigations pursuant to CERCLA and the National Oil and Hazardous Substances Pollution Contingency Plan (NCP), to determine whether the Site, along with other areas, qualified for the EPA’s National Priorities List (NPL). 2 The City became aware of the groundwater contamination upon completion by the KDHE of a Listing Site Investigation in August 1990 and it subsequently entered into an agreement with the KDHE in March 1991 to address the resulting threat to public health and the environment and to avert further economic problems. The agreement was entitled “Settlement Agreement for [RI/FS], and for Certain Remedial Actions to be Determined Following Public Involvement” (City/KDHE agreement). Around that time, the City retained the environmental consulting firm of Camp, Dresser & McKee (CDM) to assist with the project. Following completion of the RI/FS, the KDHE issued its final Corrective Action Decision (CAD) for Interim Groundwater Remediation, in which the KDHE identified the pump and treat system as the preferred remedial alternative for groundwater remediation at the Site. 3 The pump and treat system was being installed at the time of trial.

A. Description of Defendants

APCO Oil Corporation (APCO) was a public corporation organized and existing under the laws of the state of Delaware. APCO owned a facility located at 1001 East Lincoln in Wichita until 1971. The East Lincoln location is within the Site. APCO was completely dissolved in 1978 and its assets and liabilities were transferred to ALT, a court created and supervised trust organized and existing under the laws of Delaware, pursuant to an order of the Delaware Court of Chancery. ALT claimed prior to trial that it could not be sued, pursuant to the Chancery Court’s order, that it had no capacity to be sued, pursuant to Rule 17(b), and that it consequently could not be held liable for any judgement against APCO. See Fed. R.Civ.P. 17(b) (“The capacity of a corporation to sue or be sued shall be determined by the law under which it was organized.”). The court ruled in its December 22, 2000, Memorandum and Order that the City’s case against ALT is not barred by the Chancery Court’s order establishing the trust and transferring to it “all funds, assets, properties and claims ... to provide for any other legitimate claims asserted against such liquidating trust,” that CERCLA preempts Fed.R.Civ.P. 17(b), and that in any event, ALT has the capacity to be sued under Rule 17(b).

ALT filed a motion to reconsider, which the court denied. ALT again raises Rule 17(b) in defense. ALT has not presented any new persuasive arguments or authority and the court again adheres to its initial ruling. ALT has the capacity to be sued under Rule 17(b) and is therefore liable for any and all judgments against APCO stemming from the present dispute.

Reid Supply was a Kansas corporation engaged in the business of selling and distributing chemicals. Reid Supply owned a facility located at 911 E. Indianapolis from 1975 to 1988. Walter Trom-bold was the sole stockholder and president of Reid Supply during most of the years that it was in business. Two of his sons, David and Charles Trombold, were also involved in the business. Charles began working in the business in 1978 as a corporate officer and director. He became more active in Reid’s operation in the early 1980s and eventually took over for his father as president in 1986. David came to the business in 1981 and held various offices. Reid was dissolved in 1988.

Land Tool Company operated a helmet manufacturing business at 650 E. Gilbert Street from 1975 to 1988, at which it handled, stored and used chlorinated solvents, including TCE. E.H. Land owned and operated the company during that time period.

Tri-Supply operated a dry cleaning supply facility at 330 S. Commerce, from approximately June 1, 1993, to November 80, 1996, at which it received PCE in bulk and delivered it to Wichita-area customers. Gordon Kratz resided in Illinois and was the president and majority stockholder of Tri-Supply Inc.

III. CERCLA LIABILITY

Congress enacted CERCLA in the wake of the Love Canal hazardous waste disaster, to establish “ ‘a comprehensive response and financing mechanism to abate and control the vast problems associated with abandoned and inactive hazardous waste disposal sites.’ ” Morrison Enter, v. McShares, Inc., 302 F.3d 1127, 1132 (10th Cir.2002) (quoting Pub. Serv. Co. of Colo. v. Gates Rubber Co., 175 F.3d 1177, 1181 (10th Cir.1999)). As amended by SARA in 1986, CERCLA authorizes two types of actions through which a party may recoup costs incurred in hazardous waste cleanup: cost recovery actions under § 107(a) and contribution actions under § 113(f)- See 42 U.S.C. §§ 9607(a), 9613(f); United States v. Colo. & E. R.R., 50 F.3d 1530, 1535 (10th Cir.1995).

Section 113(f) incorporates the liability provisions of § 107(a). Therefore, an understanding of the interrelationship between the two sections is helpful. Section 107 imposes liability on four classes of potentially responsible parties (PRPs), including current and former owners of facilities involved in hazardous waste disposal. Tosco Corp. v. Koch Indus., 216 F.3d 886, 891 (10th Cir.2000). Liability under § 107 is strict, joint, and several. Thus, under § 107, once liability is demonstrated, a PRP may be held liable for the entire cost of cleanup, even if multiple PRPs are in volved. See Colo. & E. R.R., 50 F.3d at 1535 (“CERCLA, as originally enacted, left a PRP faced with the prospect of being singled out as the defendant in a cost recovery action without any apparent means of fairly apportioning CERCLA costs awarded against it to other PRPs.”). Section 113 supplements § 107 by providing a contribution mechanism for a party saddled with more than its proportionate share of response costs. Thus, an individual PRP which has been left with the entire cleanup cost under § 107, or which voluntarily undertakes it, may seek contribution from and attempt to apportion liability to other PRPs under § 113(f). See Morrison, 302 F.3d at 1133. Unlike § 107, however, § 113 is governed by equitable apportionment principles rather than the confines of joint and several liability.

A prima facie showing under § 113 is simply a determination that the defendant PRP would otherwise be held jointly and severally liable under § 107(a). See County Line Inv. Co. v. Tinney, 933 F.2d 1508, 1516 (10th Cir.1991); Farmland Indus. v. Colo. & E. R.R., 922 F.Supp. 437, 440 (D.Colo.1996). In order to make the required showing, a § 113 plaintiff must establish (1) that defendant is a “covered person” under CERCLA; (2) that a release of any hazardous substance at the site in question has occurred; (3) that the release caused plaintiff to incur costs; (4) that plaintiffs response action or cleanup was consistent with the NCP; and (5) that plaintiffs costs are “necessary” costs of response. See 42 U.S.C. §§ 9607(a), 9613(f); Morrison, 302 F.3d at 1135-36. CERCLA liability may be inferred from the totality of the circumstances and need not be proved by direct evidence. Tosco, 216 F.3d at 892.

Because § 107 imposes strict liability upon PRPs regardless of fault, causation is not part of the liability inquiry under § 113. See id. at 891 (“To establish liability under § 9613(f), it is sufficient for the plaintiff to establish a connection between a particular defendant and the incurred response costs vis-a-vis the defendant’s identification as a responsible person as defined in § 9607(a).”); Control Data Corp. v. S.C.S.C. Corp., 53 F.3d 930, 935 & n. 8 (8th Cir.1995); Gen. Elec. Co. v. Litton Indus. Automation Sys., 920 F.2d 1415, 1417-18 (8th Cir.1990) (stating that courts look only to see if there has been a release or threatened release for which a defendant is responsible); City of Tulsa v. Tyson Foods, 258 F.Supp.2d 1263, 1280 (N.D.Okla.2003) (“CERCLA does not impose a causation requirement as a predicate to liability when a defendant falls into one of the classes of liable parties”); Farmland Indus., 922 F.Supp. at 440 (“Neither § 9613(f) nor the Tenth Circuit’s opinion in U.S. v. Colorado & Eastern Railroad Co., 50 F.3d 1530 (10th Cir.1995), require that such causation be proven to establish liability under CERCLA’s contribution provision.”). In other words, a plaintiff in a § 113 case involving multiple PRPs need not prove a specific causal link between the costs incurred and an individual PRP’s waste. See Tosco, 216 F.3d at 891 (citing United States v. Alcan Aluminum Corp., 964 F.2d 252, 264-66 (3d Cir.1992)). Rather, § 113(f) only requires a plaintiff to prove liability or potential liability under § 107(a) and the plaintiffs incurrence of response costs. See Farmland Indus., 922 F.Supp. at 440; La.-Pac. Corp. v. Beazer Materials & Servs., 811 F.Supp. 1421, 1426-30 (E.D.Cal.1993) (statute does not impose a causation element where defendant falls within one of four statutorily defined classes and there has been an actual release). An assessment of liability in this case therefore requires the City to prove each of the five prima facie elements of a § 113 action for each defen dant upon which the City seeks to place liability.

The court has previously concluded that the City’s action will proceed solely as a contribution action under § 113(f) and adheres to that conclusion here. 4 See City of Wichita v. Aero Holdings, Inc., 177 F.Supp.2d 1153, 1169-73 (D.Kan.2000) (“The court concludes the City is a PRP pursuant to § 107(a)(1) based on its present ownership of a portion of the Coleman B property and the Bus Barn property and that the City is not entitled to maintain its § 107(a) action despite the ‘unique’ nature of this case.”). The City claims that, despite the court’s prior ruling, it is now entitled to maintain a cost recovery action under CERCLA § 107 due to intervening changes in controlling law.

Subsequent to commencement of the present suit, Congress amended CERC-LA’s innocent landowner defense. See Small Business Relief and Brownfields Revitalization Act, Pub.L. 107-118, § 223, 115 Stat. 2356 (2002). In. order to qualify as an innocent landowner, and thereby avoid liability under CERCLA § 107(b), the City must show that when it purchased the properties, it “did not know and had no reason to know” that chlorinated solvents had been disposed of on the land. 42 U.S.C. § 9601(35)(A)(i) (emphasis added); see also id. § 9607(b)(3). The 2002 amendments changed the definition of “had no reason to know” to include the requirement that the innocent purchaser take reasonable steps to stop and prevent further releases at the facility. Id. § 9601(35)(B)(ii)(II). 5

The City contends that this change requires the phrase “did not know and had no reason to know” be read disjunctively, even though the amendments did not expressly change that provision. In other words, the City claims that the relevant language should be read as “did not know [or] had no reason to know.” The effect of such a change would permit a person who buys land with actual knowledge of contamination to successfully assert an innocent landowner defense, and thereby avoid CERCLA liability, so long as that person otherwise complies with the “reason to know” requirements of 42 U.S.C. § 9601(35)(B). The City urges its interpretation because a conjunctive reading of section (35)(A) requires both that the City lack actual knowledge and had no reason to know of the contamination at the facilities. The City observes that a person without actual knowledge could not possibly take the “reasonable steps” to stop and prevent releases required by section (35)(B)(i)(II). Thus, it argues, a conjunctive reading of section (35)(A) will prevent anyone from being an innocent landowner because one cannot take steps to stop and prevent releases without actual knowledge of the contamination, and actual knowledge automatically disqualifies one as an innocent landowner.

Before analyzing this issue, the court notes that the only other court to have considered the effect of the 2002 amendments concluded that Congress did not intend to make them retroactive. See United States v. Domenic Lombardi Realty, Inc., 290 F.Supp.2d 198, 210 (D.R.I.2003). 6 That issue need not be decided here because, regardless of whether the amendment applies to this case, the City does not qualify as an innocent landowner. Under the City’s reading of the 2002 amendments, section (35)(A) and (35)(B) do appear to be at odds. Nonetheless, a better reading of (35)(B) requires that an innocent landowner make all appropriate inquiries prior to purchasing a piece of property and lack actual knowledge of the pollution at the time of purchase; then, whenever the landowner subsequently discovers the contamination, he must take reasonable steps to control the problem as prescribed in (35)(B)(i)(II). Under this reading, (35)(A) can still be read in a conjunctive fashion, requiring both a lack of actual knowledge and no reason to know of contamination on the property at the time of purchase. This interpretation adheres to the plain meaning of 42 U.S.C. § 9601(35)(A) as it is written. It is further supported by the legislative history of the 2002 amendments, wherein the changes to section (35)(B) were said to be for the purpose of clarifying existing law, not changing it. S.Rep. No. 107-2, at 12 (2001).

The City appears to confuse changes to the innocent landowner defense with the portion of the 2002 amendments that created a bona fide prospective purchaser defense. See Small Business Relief and Brownfields Revitalization Act, Pub.L. 107-118, § 222, 115 Stat. 2356 (2002); 42 U.S.C. §§ 9601(40), 9607(r). The bona fide prospective purchaser defense does permit a person with actual knowledge to purchase a contaminated facility without becoming liable under CERCLA. See 42 U.S.C. §§ 9601(40), 9607(r); see also S.Rep. No. 107-2 at 11-12. Unfortunately, the City cannot qualify as a bona fide prospective purchaser because the amendment made clear that this new defense would only apply for purchases that occurred after the changes were enacted. 42 U.S.C. § 9601(40). In contrast to this new bona fide purchaser defense, neither the 2002 amendments nor the legislative history suggests that Congress intended to change the innocent landowner defense such that those who purchased with actual knowledge of contamination might avoid CERCLA liability. Indeed, although the City argues that its interpretation avoids creating some minor surplusage under section 35(B), the irony is that the City’s interpretation would render the entire bona fide purchaser defense as surplusage because all of the protections of that defense would be duplicated in the innocent landowner defense. The 2002 amendments clearly distinguish between the innocent landowner and the bona fide prospective purchaser, and this court must do the same. Compare Small Business Relief and Brownfields Revitalization Act § 222 with id. § 223. Since the City stipulated that it had actual knowledge of TCE contamination at the Bus Barn prior to purchasing the property, the City cannot maintain an innocent landowner defense. Accordingly, the City may not maintain a cost recovery action under CERCLA § 107(a); instead, the City may only bring a contribution action under CERCLA § 113(f). 7

A. Covered Person Under CERCLA

The City must initially establish, as part of its prima facie case of CERCLA liability pursuant to § 113, that each of the named defendants was a covered person under CERCLA. See Morrison, 302 F.3d at 1135. “Covered persons” include “any person who at the time of disposal of any hazardous substance owned or operated any facility at which such hazardous substances were disposed of.” 42 U.S.C. § 9607(a)(2). The term “facility” includes any building or structure, or “any site or area where a hazardous substance has been deposited, stored, disposed of, or placed, or otherwise come to be located.” 8 42 U.S.C. § 9601(9). CERCLA defines “disposal” indirectly in 42 U.S.C. § 9601(29) by referencing the Resource Conservation and Recovery Act of 1976 (RCRA), 42 U.S.C. § 6901 et seq. That Act defines “disposal” to mean “the discharge, deposit, injection, dumping, spilling, leaking, or placing of any solid waste or hazardous waste into or on any land or water so that such solid waste or hazardous waste or any constituent thereof may enter the environment or be emitted into the air or discharged into any waters, including ground waters.” 42 U.S.C. § 6903(3).

The court concludes that the City has met its burden on the “covered person” element of CERCLA liability with respect to APCO, Reid Supply, Tri-Supply, and Land Tool. APCO owned a facility, located at 1001 E. Lincoln, at which it received, stored, sold and distributed PCE and TCE from approximately 1967 to 1971. Reid Supply owned a warehouse, located at 911 E. Indianapolis, at which it received, stored and distributed PCE and TCE from approximately 1975 until at least 1978. Tri-Supply operated a dry cleaning supply facility at 330 S. Commerce, from approximately June 1, 1993, to November 30, 1996, at which it received at least eight bulk tank truck deliveries of PCE between June 1994, and June, 1995, totaling more than 8,100 gallons, filled and stored 55-gallon drums of PCE, and delivered PCE to Wichita-area customers. Finally, Land Tool operated a helmet manufacturing business at 650 E. Gilbert Street at which it handled, stored and used chlorinated solvents, including TCE.

The court also concludes that hazardous substances were disposed of, as defined in RCRA, at defendants’ respective facilities during each defendant’s tenure as owner or operator, 9 as more fully explained in section III.B. 10 See 42 U.S.C. § 6903(3).

Besides the business entities from which the City seeks contribution, several individuals are also named as defendants. These include Gordon Kratz, E.H. Land, Walter Trombold, Charles Trombold, and David Trombold. The City seeks to impose liability on these men as operators of their facilities pursuant to CERCLA § 107(a)(2). 42 U.S.C. § 9607(a)(2). This subsection imposes liability on both owners and operators of facilities when hazardous substances were disposed of during their tenure. Courts have lamented CERCLA’s unfortunate lack of clarity where it circularly defines an “operator” as one who “operates” a facility. 42 U.S.C. § 9601(20)(A)(ii); see United States v. Bestfoods, 524 U.S. 51, 66, 118 S.Ct. 1876, 1887, 141 L.Ed.2d 43 (1998); United States v. Township of Brighton, 153 F.3d 307, 313-14 (6th Cir.1998); East Bay Mun. Utility Dist. v. U.S. Dep’t of Commerce, 142 F.3d 479 (D.C.Cir.1998); Edward Hines Lumber Co. v. Vulcan Materials Co., 861 F.2d 155, 156 (7th Cir.1988). Observing the effects of that lack of clarity, the Tenth Circuit noted that other circuits are. split on how to define “operator.” FMC Corp. v. Aero Indus., 998 F.2d 842, 846 (10th Cir.1993). Some courts have imposed operator status on persons or entities with the “authority to control” operations at a facility, while others have required “actual control.” See id. Regrettably, FMC failed to adopt either test, finding instead that the defendant would be considered an operator regardless of which standard was used. Id.

In Bestfoods, the Supreme Court appeared to clarify the definition of “operator” under CERCLA, finding that “an operator must manage, direct, or conduct operations specifically related to pollution, that is, operations having to do with the leakage or disposal of hazardous waste, or decisions about compliance with environmental regulations.” Bestfoods, 524 U.S. at 67, 118 S.Ct. at 1887. This definition clearly requires actual participation, not merely the potential to do so. See Brighton, 153 F.3d at 314 (interpreting Best-foods as adopting the “actual control” test and rejecting the “authority to control” test). The court notes that in Bestfoods, the alleged operator was a parent corporation whose subsidiary was the owner of the facility in question. Bestfoods, 524 U.S. at 55, 118 S.Ct. at 1881. Similarly, in Brighton, the alleged operator was a municipality that exerted considerable influence over a privately owned dump site. Brighton, 153 F.3d at 310-12. Nonetheless, the rationale in those cases applies more broadly and, as here, makes just as much sense when applied to individuals.

1. The Trombolds

David and Charles Trombold began working at Reid Supply in 1981. 11 Although Charles was on the Reid board of directors and had held a position as a corporate officer since 1978, he was not actively involved in the day-to-day operations and decision-making until he began working at Reid in 1981. Even though the City was able to elicit a statement from Charles that he regularly participated in management meetings from the time he became an officer through the company’s dissolution in 1988, the court finds that Charles’ answer likely focused on the post-1981 period. Charles was simultaneously pursuing two degrees at the University of Kansas until 1980. Immediately thereafter, he studied chemistry at a university in England. Accordingly, he had little, if any, time to participate in the “actual control” of Reid Supply’s affairs to the point that he would be subject to operator liability under CERCLA. As for David Trom-bold, the City makes no allegations that he did anything to subject himself to operator liability before 1981. Thus, prior to 1981, neither David nor Charles held duties at Reid that would qualify them as operators. See 42 U.S.C. § 9601(20)(A)(ii); Bestfoods, 524 U.S. at 67, 118 S.Ct. at 1887.

In order to be considered covered persons under the relevant CERCLA provisions, David and Charles would have to be considered operators during a time when chlorinated solvents were disposed of at 911 E. Indianapolis. As discussed in much more detail in Section III.B.2, the only releases that the City has proven at the Reid facility occurred primarily from 1975-78 and ceased not later than 1981. None occurred during the time David Trombold worked at Reid. Since any disposals at 911 E. Indianapolis occurred prior to the time that David and Charles Trombold could be considered operators, neither David nor Charles Trombold is a covered person under CERCLA § 107(a)(2). 42 U.S.C. § 9607(a)(2).

Unlike his sons, Walter Trombold was involved with operations at the Reid facility during the period when bulk deliveries were made. Hence, his involvement in Reid’s affairs occurred during a time when PCE was disposed of at 911 E. Indianapolis. Thus, the real question for Walter is whether his activities made him an operator under CERCLA.

The City makes much of the fact that Walter was president of Reid Supply and owned most (and at times, all) of Reid’s outstanding stock. However, Walter’s titles and stock ownership suggest only some authority to control, not actual control. Bestfoods rejected authority to control as a basis for operator liability. Bestfoods, 524 U.S. at 67, 118 S.Ct. at 1887. Instead, an operator must be actively involved in decisions regarding disposal of hazardous substances or environmental compliance. See id. Moreover, courts applying the actual control test have consistently required more than casual or occasional involvement in such decisions. Instead, an operator under CERCLA must make the relevant decisions on a frequent, typically day-to-day, basis. See East Bay Mun. Util. Dist., 142 F.3d at 485 (citing United States v. Cordova Chem. Co. of Mich., 113 F.3d 572, 579-81 (6th Cir.1997); Redwing Carriers, Inc. v. Saraland Apartments, 94 F.3d 1489, 1504-05 (11th Cir.1996); Schiavone v. Pearce, 79 F.3d 248, 253-54 (2d Cir.1996); United States v. Gurley, 43 F.3d 1188, 1193 (8th Cir.1994); John S. Boyd Co., Inc. v. Boston Gas Co., 992 F.2d 401, 408 (1st Cir.1993); Lansford-Coaldale Joint Water Auth. v. Tonolli Corp., 4 F.3d 1209, 1220-22 (3rd Cir.1993); Joslyn Manuf. Co. v. T.L. James & Co., Inc., 893 F.2d 80, 83 (5th Cir.1990); Hines Lumber Co., 861 F.2d at 157-59).

As president of Reid Supply, Walter Trombold was two layers removed from the day-to-day supervision of operations at 911 E. Indianapolis. The evidence showed that Reid’s warehouse foreman was the person directly responsible for day-to-day operations at the 911 facility. For most of the period in which disposals occurred, that person was Jerry Letterman. The warehouse foreman reported to Reid’s general manager. The general manager was in charge of operations at all of the Reid facilities. During the relevant time frames, Gene Stamm was Reid’s general manager. Stamm reported to Walter Trombold. The evidence does not indicate that Walter Trombold ignored this organizational structure, nor that he otherwise engaged in activities that would rise to the level of control over PCE handling that would give him operator status. In other words, Walter Trombold’s connection with day-to-day operational decisions over PCE handling was too remote to establish operator liability. To rule otherwise would subject corporate officers to operator liability based on their mere authority to control, a standard rejected in Bestfoods.

On the other hand, Bestfoods also imposes operator liability on those making decisions about compliance with environmental regulations. Bestfoods, 524 U.S. at 67, 118 S.Ct. at 1887. Although Walter Trombold was not out on the warehouse floor telling employees how to fill barrels or how to clean up spills, the evidence shows that environmental compliance issues were addressed at weekly management meetings. Furthermore, Gene Stamm presented es sentially uncontroverted testimony that Walter Trombold was present at those nieetings, and that no decisions were made at those meetings without Walter’s approval. Based on the frequency of those meetings, and the fact that Walter Trombold was actively involved in deciding matters of environmental compliance, the court finds that he was an operator. Accordingly, Walter Trombold is a covered person under CERCLA § 107(a)(2).

2. Gordon Kratz

Gordon Kratz was the president and majority stockholder of Tri-Supply during its tenancy at 330 S. Commerce. The City seeks to hold Kratz liable as an operator based on his activities associated with the 330 S. Commerce facility. Kratz lived in Illinois, and only visited the Wichita facility about once every eight or nine months, staying only a day for each visit. He entrusted the management and environmental compliance of the Wichita facility to his local office manager, Bob Raveill. The City claims Kratz is an operator because he established PCE handling procedures and enforced them at the TriSupply facility. On the contrary, Kratz’s involvement at the Wichita location was nowhere near the day-to-day management and control required of an operator under Bestfoods. Likewise, his so-called. procedures amounted to nothing more than guidance to limit the number of times drums were re-used for PCE storage. Beyond that, he merely admonished Raveill to follow the general safe-handling procedures supplied by the manufacturer, and to avoid any spills. Indeed, Kratz noted the peculiarity of the City’s position at trial when he said “You keep calling it my procedures; but I — I don’t know why.” Kratz’s general admonitions for safe handling hardly rise to the level of control that would give rise to operator liability, nor do they amount to the type of decisions that would impose liability for directing environmental compliance matters. Accordingly, Kratz is not an operator, and not a covered person under CERCLA § 107(a)(2).

3. E. H. Land

E. H. Land stipulated that he both owned and operatéd the Land Tool facility during the relevant time periods. Furthermore, the evidence clearly established that chlorinated solvents were disposed of at the Land Tool facility. See infra, section III.B.3. Accordingly, E. H. Land is a covered person under CERCLA § 107(a)(2).

B. Release of any Hazardous Substance

The City must next establish, as part of its prima facie case of CERCLA liability pursuant to § 113, the occurrence of a release at the defendants’ respective facilities. See Morrison, 302 F.3d at 1133. The term “release” means “any spilling, leaking, pumping, pouring, emitting, emptying, discharging, injecting, escaping, leaching, dumping, or disposing into the environment (including the abandonment or discarding of barrels, containers, and other closed receptacles containing any hazardous substance or pollutant or contaminant).” 42 U.S.C. § 9601(22)! The term “environment” includes any “surface water, ground water, drinking water supply, land surface or subsurface strata, or ambient air within the United States or under the jurisdiction of the United States.” 42 U.S.C. § 9601(8). The City contends that a release of the chlorinated solvents PCE and/or TCE occurred at defendants’ respective facilities. To understand the discussion of evidence regarding releases, some background on the activity of chlorinated solvents in the environment is helpful.

Chlorinated solvents are hazardous substances. See Bob’s Beverage, Inc. v. Acme, Inc., 264 F.3d 692, 694 (6th Cir.2001). A spill of PCE or TCE which stays on a surface such as concrete tends to evaporate quickly. Once it reaches the soil, however, chlorinated solvents can stay in the soil for years, particularly if the soil is saturated. The contamination gradually moves or “leaches” into the groundwater and then moves along preferential flow paths to the bottom of the aquifer because PCE and TCE are heavier than water. Dechlorination occurs during that slow process. Chlorinated solvents dechlori-nate as the result of a chemical reaction which occurs when the chlorinated solvent loses a chlorine atom, which is then replaced with a hydrogen atom. The chlorinated solvents involved in this case lose one chlorine atom at each step of the dechlorination process: PCE (four chlorine atoms) dechlorinates to TCE (three chlorine atoms); TCE dechlorinates to DCE (two chlorine atoms); DCE dechlori-nates to VC (one chlorine atom) and VC ultimately dechlorinates to ethane, a gaseous hydrocarbon. The process is sequential and the steps are not skipped.

There was disagreement regarding whether dechlorination can occur in soil. Olsen, the City’s witness, testified that de-chlorination occurs only in groundwater, when carbon is available, and that it does not occur in soil. Eversman, Reid’s witness, testified on cross-examination by the City’s counsel that in the presence of a carbon source, PCE in the soil can degrade to daughter products, including DCE, by the time it enters the groundwater. The court accepts Eversman’s testimony on this point. There are places at the Site where PCE, TCE and DCE have been found in the soil at a single location, e.g., RSI-7. Coupled with evidence of a PCE spill at the location, the only reasonable explanation for the presence of TCE and DCE is dechlorination because it is undisputed that DCE exists only as the result of the dechlorination process.

While the dechlorination process is sequential, it is not inexorable. The presence of oxygen, nitrates, and ferric iron can affect dechlorination. For purposes of this case, however, carbon products and sulfates are the compounds of most concern. Carbon, which generally is found in the form of petroleum, can speed up the dechlorination process. The presence of sulfate, an inorganic compound, has the opposite effect; it can slow the process and even stop dechlorination from DCE to VC. Olsen explained that in the presence of a carbon source, PCE and TCE can degrade, but when the concentration of sulfate is “high enough,” dechlorination stops at DCE until the microorganisms which facilitate dechlorination can get rid of the sulfate. Once that happens, dechlo-rination can proceed to VC and beyond. Eversman agreed that sulfates can inhibit or retard the degradation process. An exhaustive (and exhausting) source of information regarding dechlorination at the Site can be found in exhibits P-2219A and P-2333. With this background in mind, the court will discuss the evidence of chlorinated solvent releases at each of defendants’ facilities. The trial testimony regarding releases consumed many days of testimony, was quite complex, and is difficult to summarize. In retrospect, it is tempting to simply find that releases occurred at each facility without discussing the details of the evidence. Because chlorinated solvents are not found naturally in soil or groundwater, and because they were found in both at each facility, the inevitable conclusion is that releases occurred. The court has resisted this temptation out of respect for the efforts of the parties and the inevitable likelihood of review.

1. APCO

From approximately 1967 to 1971, APCO received PCE and TCE by bulk transport truck. It stored, sold and distributed these chlorinated solvents at its 1001 E. Lincoln Property during that five-year period. APCO housed its “Chemicals and Solvents” and “Fuels” divisions in an office building located at the corner of Lincoln and Washington Streets. The chemicals and solvents group utilized a warehouse and loading dock, located just to the southeast of the office building, to store, ship, and receive product, including chlorinated solvents. See App. C-2.

APCO also utilized a fenced-in area, located directly to the east of the warehouse, to store empty 55-gallon drums. APCO stored anywhere from 50 to 60 drums in this area, known as the “pen,” including used drums that previously contained chlorinated solvents. APCO employees stacked used drums on their sides in the pen until they could be shipped to a drum-reconditioner for reworking. Some of the used drum heads were in “pretty bad shape” while others “weren’t quite so bad.” The pen also contained a 1,000 .gallon above-ground storage tank, filled • with PCE. Other than the “pen,” which was gravel, almost the entire surface area of the property at 1001 E. Lincoln was concrete.

When a chlorinated solvent was delivered to the facility, it was either “drummed” on the loading dock or pumped directly into the above ground storage tank in the pen. If the solvent was to be drummed, the transport driver “would head the tractor directly up toward the loading dock from the north,” and would connect a hose from the transport truck to a nozzle, which could be pointed directly into the drums on the dock. The drums were then filled, weighed, and moved to the back of the dock. Except for the type of nozzle used, the procedure for filling the above ground tank in the “pen” was basically the same. The transport driver would drive up from Lincoln Street into the same area, head the tractor to the east toward the alley, and connect a hose onto the fill pipe of the above ground tank.

At least two hose connections occurred every time a chlorinated solvent was delivered. First, the transport driver ran a hose from the discharge pipe on the bottom of the truck’s tank up to a pump located either on the side of the truck or on the tank itself. The driver then ran another hose from the pump to the fill pipe of the above ground tank or onto the dock so that solvent could be drummed.

APCO followed the same basic procedure when transferring chlorinated solvent out of its above-ground tank for delivery to outside customers. First, a hose was connected to the discharge pipe on the above-ground tank and then hooked to a portable pump. Another hose was then run from the portable pump into either drums or a tanker truck for filling.

Larry Richwine worked as a driver and warehouseman at APCO from 1966 to 1972 and was the only fact witness to have actually observed operations at the facility. 12 The court finds Richwine’s testimony as to APCO’s general operations and tanker-truck delivery procedures generally credible. On the other hand, his testimony as to spill location, frequency, clean-up procedure, and size was highly influenced by one of ALT’S counsel and largely lacking in credibility.

According to Richwine, spills of chlorinated solvents occurred as transport hoses were being disconnected at least 50% of the time chlorinated solvents were delivered, for a total of approximately 30 spills. The hoses were always disconnected-and the spills always oecurred-on concrete-covered property immediately north of the above-ground storage tank and immediately north of the loading dock. The spills varied from smaller to larger amounts, Larger spills were either wiped up with rags or absorbed with an absorbent material, while spills of a half-gallon or less were left to evaporate.

Shortly before trial, one of ALT’s counsel showed Richwine a photo purporting to depict the surface area that is covered when a half-gallon of liquid is spilled onto concrete. 13 Although Richwine testified about witnessing spills in excess of two to three gallons in his deposition, 14 he changed his trial testimony, based on the photo, to reflect that no more than a gallon or two of chlorinated solvent was ever spilled. 15

Although Richwine indicated prior to his deposition that solvent sometimes “spilled onto the gravel ground when transferring solvent from the above ground tank to drums for delivery,” Richwine insisted in his deposition and at trial that all spills occurred on concrete. The evidence of contamination in the pen area does not support his testimony.

The frequent changes and inconsistencies in Richwine’s testimony, and his general demeanor while testifying, made it apparent that he was prepared at trial to say anything he could to assist in ALT’s defense. Richwine lost what little credibility he had remaining when, at the beginning of his redirect examination, he looked directly at defense counsel, immediately before answering the question posed to him, as if seeking some form of direction.

Given his ever-evolving testimony and defense counsel’s obvious influence, it is reasonable to infer that Richwine personally observed spills larger and more frequent than those he testified about. The court has no doubt that frequent, sometimes significant spills occurred. The sheer volume of chlorinated solvent that was being stored, shipped, and received at APCO on a daily basis,, combined with APCO’s apparent failure to take spill precautions during solvent transfers, makes frequent spills quite likely.

The court concludes' that at least 30 spills of chlorinated solvents-and likely more-occurred at APCO between 1967 and 1971. Most spills occurred on the concrete-covered property between the office building and the loading dock/warehouse as transfer hoses were being disconnected. Due to the nature of APCO’s operations, spills and leaks of chlorinated solvents also occurred on the gravel surface area of the pen and the loading dock itself. Whether these spills occurred as a result of residual solvents leaking from the used drums in the pen, from PCE transfers involving the above ground tank, or from general operations on and around the loading dock, all had the potential to allow chlorinated solvents to seep directly through to the soil and, ultimately, to the groundwater.

ALT sought to deflect the damaging effects of Richwine’s testimony in two ways: first, it presented testimony by Steven L. McCabe to the effect that chlorinated solvents cannot pass through concrete into soil. Second, it vigorously challenged the validity of test data taken at the APCO site, which showed the presence of chlorinated solvents in soil and groundwater. McCabe’s opinions apply primarily to APCO and can be dealt with rather summarily. The arguments regarding test data have application throughout the Site, and require more extended discussion.

After a series of field observations, concrete core-sample collections, spill tests, crack evaluations, and mathematic computations, McCabe concluded that “the penetration of the CVOC product through the concrete is unlikely given the quality of the materials and the basic mechanisms that control the penetration of liquids through concrete.” The original concrete, however, no longer exists in the areas where Ri-ehwine testified that spills occurred. McCabe’s opinion therefore hinges upon the very large assumption that the concrete existing in the spill areas between 1967 and 1971 would have been similar, in all significant respects, to concrete of the same mid-1960’s vintage still existing on the property immediately north of the loading dock.

The court finds McCabe’s testimony highly speculative and largely contrived. As will be explained in greater detail, chlorinated solvents were detected in the soil, only two feet below the ground surface, immediately below where the original concrete existed, in an area where Richwine testified that spills occurred. McCabe could not offer-and the court cannot imagine-any other reasonable explanation for the presence of chlorinated solvents so near the ground surface in that area, except that they passed through the overlying concrete.

Indeed, it is the presence of chlorinated solvents in the soil that most clearly confirms, in the court’s view, that significant releases actually occurred at 1001 E. Lincoln. While plausible arguments have been made regarding the extent to which chlorinated solvents in the groundwater underneath 1001 are attributable to upgra-dient sources and about the legitimacy of on-site versus off-site data, 16 there is simply no getting around the presence of chlorinated solvents, at shallow depths, in the soil. Although the parties hotly contest the validity of the test data which showed the presence of chlorinated solvents in soil, the disputes center around the extent of the contamination, not the fact of contamination. It is agreed that chlorinated solvents do not occur naturally in soil and groundwater, and there is no credible evidence that they migrate through the soil from upgradient sources. Not even McCabe suggested that groundwater contaminated with chlorinated solvents from upgradient sources conveniently “wicked up” through 15 feet or more of soil to places where these contaminants had been spilled. 17 The presence of soil contamination is illustrated by soil samples taken at various locations on the APCO property designated APC-5, APC-9, and APC-9A.

APC-5 is located directly underneath the concrete covered area between APCO’s office building and its loading dock, the location where, according to Richwine, spills occurred. 18 On-site testing at APC-5 indicated high levels of PCE in the soil at concentrations of 1403 parts per billion (ppb) only two feet below ground. Although off-site testing of the same samples yielded dramatically lower results, the off-site lab result nonetheless confirmed the presence of PCE in the soil at concentrations of 58 ppb, only two feet below the surface.

Data from APC-9 and APC-9A also revealed the presence of PCE in the soil at shallow depths. APC-9 and APC-9A are located east of APCO’s warehouse, in the gravel area known as the “pen.” On-site PCE readings at APC-9 indicated the presence of PCE in the soil at concentrations of 13.4 ppb and 76 . ppb at depths of two and four feet, respectively. The off-site lab confirmed the presence of PCE, at 23 ppb, four feet below ground. Similar data from APC-9A indicated the presence of PCE at concentrations of 291 ppb (on-site) and 16 ppb (off-site) two feet below the surface.

Considering Richwine’s testimony together with the data collected from APC-5, APC-9 and APC-9A, the court finds overwhelming evidence that chlorinated solvent spills occurred at 1001' E. Lincoln and that at least some of those spills reached the soil. The question then becomes whether, and to what extent, 1001 E. Lincoln is a source of contamination to the groundwater. Resolution of this question requires a somewhat extended review of the evidence pertaining' to the taking and analysis of soil and groundwater samples at APCO and other locations at the Site.

In February 1998, CDM subcontracted with Environmental Priority Service (EPS) to perform soil, soil gas, and groundwater sampling at 100,1 E. Lincoln. Pat Martin is the owner of EPS. Martin had a lot of training and experience in the collection and analysis of samples. His job was to take soil and groundwater samples at locations selected by CDM. In March 1998, Martin and his crew collected soil and groundwater samples at 1001 E. Lincoln. Martin utilized an on-site laboratory consisting of a van fitted with a geoprobe and various other equipment for analyzing samples, such as a gas chromatograph. A geoprobe is a hollow tube which is driven through layers of soil and into groundwater. Geoprobing is a relatively quick and inexpensive way to preliminarily locate or screen for the presence of contaminants in soil and/or groundwater at various depths. The soil and/or groundwater samples are analyzed in either on- or off-site laborato ries, or both. The laboratory analysis generates test data which reveals concentrations of contaminants such as chlorinated solvents, petroleum and sulfates. The types and concentrations of contaminants, as well as them locations in the soil and groundwater, yield information which, in this case, provided grist for complex and vastly differing opinions regarding sources, concentrations, degradation, direction of movement, areal dispersion and other aspects of chlorinated solvent contamination in the soils and groundwater at the Site.

Martin employed a modified version of the EPA SW-846 Method 3810 to analyze soil samples on-site. Method 3810 is described as a “heated headspace method” because it requires the heating of a sample to release volatile compounds for analysis. It is a simple method that allows large numbers of samples to be screened in a relatively short period of time. Due to the variability of the method, however, it is recommended for use only as a screening procedure for other, more accurate and determinative methods.

All sampling conducted at 1001 E. Lincoln, including Martin’s use of the modified Method 3810, was to be performed according to a number of quality assurance and quality control (QA/QC) protocols referred to in the Plume B Workplan, the City/ KDHE Agreement, 19 and the 1995 Final Quality Assurance Project Plan for Prede-sign Data Acquisition and Bioremediation Pilot Demonstration (QAPP). Twenty seven borings were taken on and immediately adjacent to the APCO property and over 100 samples were collected. Of the locations sampled, the most notable contamination was detected at APC-3, APC-5, APC-7, APC-9, and groundwater sampling location 910.

The court was impressed with Martin’s forthright demeanor and finds that Martin knew what he was doing and did it properly. The court was not impressed with cross-examination whose apparent purpose was to demonstrate that Martin was un-knowledgeable and sloppy. Despite ALT’s criticisms of Martin’s work and the absence of “chain of custody” logs, the court finds that he sufficiently documented his analysis of samples at 1001 E. Lincoln. His field notations include the time of day for placing the sample in the laboratory instrument, the chromatogram readings, the sample identification number, the sample depth, the sample type, and the reported concentrations. Martin’s typed “field sheets and analytical reports” for APC series data points located on or near 1001 E. Lincoln were included in the City’s Data Quality Reports approved by the KDHE. The court finds Martin’s on-site data substantially complies with CERC-LA’s requirements and is reliable to assist in determining ALT’s (and Reid’s) liability.

In May 1998, the City retained David Hendron and his company, GeoSyntec Consultants, to assist in determining whether specific properties located within the Site were sources of groundwater contamination. 20 According to Hendron, groundwater flows across the Site generally in a north-to-south direction, with a downward gradient of about five feet per mile. In keeping with the known groundwater flow direction, Hendron generally compared shallow groundwater data from wells upgradient from the subject property with data from on-site wells and those immediately downgradient from the property. If lower concentrations of chlorinated solvents were discovered upgradient than were discovered on-site and immediately downgradient to a property that was known or suspected of having handled chlorinated solvents, Hendron generally deemed that property a source. Hendron focused upon shallow groundwater results because shallow groundwater is generally the first to be impacted. Once chlorinated solvents hit the shallow groundwater at a given location, they tend to sink, because of their density, and to diffuse out into plumes of contamination. Because the contamination generally moves slightly downward within a given plume as it migrates off site, a shallow depth of contamination indicates a likelihood that a nearby property is a source.

Hendron compared the data collected from upgradient wells 828-1A and 828-4A to downgradient samples collected from APC-5, APC-7, 828-8B, WM-1, WM-2, APC-9 and APC-9A. Data collected from the upgradient wells showed little or no chlorinated solvents, while data collected on site and immediately downgradient of the APCO property showed significantly higher chlorinated solvent concentrations in both the soil and groundwater. At APC-5, for example, PCE concentrations of 1,403 ppb were discovered in the soil at 2 feet below ground surface (bgs) near the southeast corner of the facility. Groundwater readings taken at monitoring well WM-1, just south of APCO’s property line, revealed concentrations of PCE in the shallow groundwater (18 feet bgs) at 1200 ppb, TCE at 7300 ppb, and eis-l-2-DCE at 17,000 ppb, respectively. 21

Additional geoprobe sampling was performed on the former APCO property in March 2000. 22 The samples were taken at location 910B, directly beneath the location of APCO’s loading dock. The geoprobe readings indicated the following contaminant levels in the shallow groundwater: 1800 ppb PCE, 1500 ppb TCE, 1600 ppb cis-l,2-DCE, and <500 ppb VC. After analyzing the geoprobe results, CDM installed a permanent monitoring well at 910B to monitor the remedial system. Like the geoprobe samples, the monitoring well indicated significant levels of chlorinated solvents in the groundwater at 910B. Shallow groundwater readings taken in May 2002 indicated the presence of not only PCE (1400 ppb), but also of TCE (1600 ppb), DCE (3800 ppb), and VC (130 ppb). Each of these readings dramatically exceeds the Alternate Cleanup Levels (ACLs) established by the KDHE. 23 Hendron asserted- and this court now agrees-that these sampling results, together with the results from APC-7, 828-8B, WM-2, APC-9 and APC-9A, indicate APCO is a source of chlorinated solvent contamination to the groundwater.

ALT mounted a vigorous defense to the City’s evidence regarding chlorinated solvent contamination at APCO. According to ALT, the on-site soil data taken at 1001 E. Lincoln is unreliable for both qualitative and quantitative purposes because Martin utilized the modified 3810 method without EPA or KDHE approval and did not perform the required QA/QC protocols and data validation tests required to ensure the reliability of the onsite soil data. 24 ALT points out that Method 3810 requires “[s]tandard quality assurance practices” to be used, including data validation tests on any samples taken such as matrix spikes, matrix spike duplicates, and surrogate spikes. ALT also asserts that Martin and members of his crew did not follow established QA/QC, chain of custody, and data validation procedures as required by Method 3810, the Settlement Agreement, and the QAPP. The court finds that any failure by Martin to perform analyses such as matrix spikes and matrix spike duplicates did not impair either Martin’s credibility or the weight of his testimony.

Even ALT’s own expert, Dr. Edward McBean, admitted that APCO is a source. ALT asserts nonetheless that its contribution of contamination to the groundwater is so insignificant that it does not exceed the ACLs. According to McBean, therefore, no significant clean-up is required.

McBean considered several factors in reaching his opinions, including solvent handling operations at APCO, the. presence of concrete at the facility, the soil and soil data, adjacent land uses, and the contribution of contamination from upgradient sources. McBean utilized ALOHA and EMSOFT modeling to demonstrate that only minute amounts- of contaminant could reach the soil through concrete. He concluded that a half-gallon spill onto concrete would likely evaporate in 12 minutes in the summer and 81 minutes in the winter. According to McBean, only 1% of the spill would seep into the expansion joints or cracks in the concrete, 94 to 99% of which would ultimately evaporate. 25

Because on-site data largely contradicted McBean’s findings, he rationalized that the quality of the on-site data is so poor that it cannot be relied upon except, for screening purposes, to show the presence or absence of a particular contaminant. To illustrate his opinions, McBean prepared several exhibits graphically depicting the dramatic differences between the on-site and off-site soil laboratory samples from 1001 E. Lincoln. ALT-1536 is a comparison of on-site and off-site soil samples taken from APC-5, APC-7, and APC-9A-three locations where testing indicated the soil was most contaminated. On-site PCE sampling at APC-5 showed PCE in the soil at concentrations of 1,403 ppb (two feet), 559 ppb (four feet), 1,340 ppb (six feet), 571 ppb (eight feet), and 605 ppb (ten feet), respectively, whereas off-site testing indicated PCE concentrations of only 58 ppb (two feet), 12 ppb (four feet), 20 ppb (eight feet) and non-detect (twelve feet). On-site sampling at APC-9A indicated concentrations of PCE at 291 ppb, two feet below the ground, while off-site testing showed concentrations of only 16 ppb at the same depth. The highest reading at APC-7 was 449 ppb PCE in the soil at 12 feet, whereas off-site readings are less than 5 ppb (a non-detect) at all depths. The on-site sampling results were never duplicated by the City in subsequent testing at the same location and were never, except on an intra-laboratory basis, compared to the off-site results for consistency.

Based on these and other factors, McBe-an opined that the off-site soil data is more appropriately utilized because it has been subjected to better QA/QC procedures and is more reliable than what he considered to be the City’s inflated, on-site results. McBean sought to reinforce this opinion by pointing out that the City’s choice to use the on-site data for modeling purposes at APCO directly conflicts with the data selection approach taken by the City at its own Bus Barn property. 26 There, the City used off-site laboratory data and referred specifically to the on-site data as being “for screening level purposes only.” On-site results were consistently higher than the off-site results at the Bus Barn, just as they were at APCO.

The City’s response to McBean’s criticisms has been that chlorinated solvents collected from the soil at APCO volatilized as they were transported from 1001 E. Lincoln to the off-site lab in capped, sealed plastic sleeves, resulting in the dramatically lower concentrations reported by the off-site lab. 27 The City claims the samples did not volatilize to the same extent at the Bus Barn because the soils there contain high levels of organic carbon, or “cinders,” which adsorb TCE, thereby minimizing chlorinated solvent volatilization. The City asserts that the soil at 1001 E. Lincoln does not contain cinders and, as a result, the soil samples at 1001 E. Lincoln volatilized more rapidly during transport to the off-site lab.

To justify its theory that the soil at the Bus Barn contains cinders, the City performed a total organic carbon (TOC) test on the Bus Barn soil. McBean considered the City’s TOC test to be unreliable because it does not distinguish between the carbon content in the soil/cinders and the carbon content in the chlorinated solvents themselves. It is his belief that the relatively high TOC concentrations in the soil at the Bus Barn were probably indicative of high chlorinated solvent contamination rather than high organic carbon content in the soil.

Not only were off-site results used for modeling purposes at the Bus Barn, but lower cleanup standards were applied there than were applied at APCO. Under the KDHE’s Interim Remedial Guidelines (IRG), which the City used at the Bus Barn, the PCE soil cleanup standard was 50,000 ppb. Under the KDHE’s March 24, 1999 RSK guidance, which the City claims is applicable to 1001 E. Lincoln, the PCE soil cleanup standard is only 180 ppb for PCE.

Several exhibits graphically depict comparisons between chlorinated solvent concentrations and the cleanup standards applied at the Bus Barn as opposed to 1001 E. Lincoln. ALT-1554 compares the most stringent cleanup standards for the two properties. ALT-1557, which compares the off-site laboratory soil chlorinated solvent concentrations at the Bus Barn prior to remediation and the off-site concentrations at 1001 E. Lincoln, illustrates that the off-site soil concentrations of chlorinated solvents at 1001 E. Lincoln are so small that they can barely be plotted. From this evidence, McBean concluded that, even after remediation of the Bus Barn soil, the chlorinated solvent levels from both the on-site and off-site laboratories are dramatically higher there than the chlorinated solvent levels at 1001 E. Lincoln. 28

According to ALT, this evidence illustrates a great hypocrisy: The City left most of the contamination in the soil at the Bus Barn, while at 1001 E. Lincoln, where no significant contamination was detected by the off-site lab, the City is demanding cleanup. While the City’s decision to treat the Bus Barn property to an opposite approach is troubling, the court finds that legitimate concerns regarding the volatili-zation of chlorinated solvents at APCO, during transport to the off-site lab, justify reliance on the on-site data at that location.

ALT additionally argues that the City’s approach at 1001 E. Lincoln is not only inconsistent with the approach taken at the Bus Barn but is also inconsistent with CDM’s conclusion, in its draft source control measures report, that soil contamination at APCO is not contaminating the groundwater at concentrations above the ACL’s. In 1998, CDM prepared an APCO draft source control measures report, which utilized VLEACH modeling to estimate likely PCE contributions from soil to groundwater at the APCO site. CDM’s modeling concluded that “[t]he predicted PCE concentrations in groundwater are well below the ACL of 14 ugL for both the no action and SVE scenarios.”

McBean performed the same VLEACH modeling, using even more conservative estimates, and again concluded that the concentrations of chlorinated solvents in the soil at APCO would not leach to the groundwater at concentrations above the ACLs. Instead, it is McBean’s opinion that the high concentrations of groundwater contamination underlying 1001 E. Lincoln are the result of upgradient sources from Pride, from the Bus Barn, from Wichita Eagle, from the so-called South Wash source, from an upgradient source in the vicinity of well TW-1 (located on the Tennyson Sheet Metal property), and possibly from Walter Morris, the property immediately south of 1001 E. Lincoln. In reaching this opinion, McBean considered the data at the site, the groundwater flow directions and velocities surrounding 1001 E. Lincoln, and the degradation rate characteristics of the chemicals of concern. He utilized DYNFLOW, DYNTRACK and CHAIN computer modeling. McBean deliberately chose to use DYNFLOW and DYNTRACK because they were the City’s experts’ models, which he claimed enabled him to make an “apples to apples comparison.” He also used an “octanol/water effect” or “sponge” theory to explain why, in his view, the contamination found at TW-1 indicates an upgradient source caused the contamination discovered at APC-910B and APC-7, despite the fact that the concentrations of chlorinated solvents found at TW-1 are less than the concentrations of chlorinated solvents detected at those downgradient locations on APCO’s property. He employed a “fingerprinting” analysis to show that TW-1, as opposed to APC-5, was the source of chlorinated solvent contamination at 910B.

McBean’s conclusion that APCO is not contributing to groundwater contamination above the ACLs, however, is necessarily dependent upon his ability to explain away the contaminant concentrations reported at virtually every sampling location on and around the APCO property. According to McBean, nothing at APC-3 indicated that APCO contributed contamination above the ACLs because the on-site data was not reliable, the off-site data was below RSK cleanup levels, and the groundwater was not sampled. He reached the same conclusion as to APC-5 and APC-9A because the on-site data was not reliable, the off-site data was below the RSKs, and there was allegedly an improper chain of custody on the groundwater samples. At APC-7, he contended there is no shallow groundwater contamination and the deep soil contamination is likely from the wicking up of contaminants in the groundwater below. At APC-9, he alleged that both on- and off-site soil concentrations were below the RSKs, and samples were not analyzed for groundwater contamination. At 910B, he opined the groundwater detections are from upgradient sources in the vicinity of TW-1.

In addition to McBean’s charge that the City’s data was unreliable, he also questioned the City’s use of certain data, including the City’s attribution of certain contamination to the APCO property. The most contaminated groundwater sample collected in the vicinity of 1001 E. Lincoln was collected at WM-1. WM-1 is located about six feet south of APCO’s southern property line. Groundwater concentrations sampled in March 1996 indicated the presence of PCE at 1200 ppb, TCE at 7300 ppb, cis-1,2 DCE at 17,000, trans-1-2DCE at ND 250 and vinyl chloride at ND 450. Those readings are approximately five times as high as the next highest groundwater sample readings at location 910B, which is located a few feet and slightly northwest of WM-1.

In McBean’s opinion, the contamination at WM-1 did not come from APCO, but came from contaminated soil overlying the Walter Morris property. McBean came to this conclusion after considering the data from well 828-8B, located about four feet upgradient from WM-1 on the Walter Morris property. Although 828-8B lies directly upgradient of WM-1, between WM-1 and the APCO property, contaminant concentrations there were approximately 50 times less, or roughly 2%, of the levels reported at WM-1. Because the groundwater flows generally north-to-south, it is highly unlikely, in McBean’s opinion, that chlorinated solvents migrated from the shallow groundwater at APCO past 828-8B and into the groundwater at WM-1. Nor is it likely, according to McBe-an, that chlorinated solvents could have migrated in the dry soil above the groundwater past 828-8B and into the groundwater at WM-1. The only reasonable explanation, in McBean’s view, is that the groundwater concentrations at WM-1 were caused either from contaminated soil overlying WM-1 leaching to the groundwater or from contaminated soil getting pushed down to the groundwater through geoprobe sampling at WM-1.

To emphasize his conclusion, McBean calculated how many drops of PCE in the soil at the surface it would take to reach the groundwater contaminant concentrations noted in the shallow groundwater at WM-1. Using a two-inch Geoprobe, McBe-an opined only 1.7 drops of PCE from an eye-dropper would yield the chlorinated solvent concentrations reported at WM-1. Using a one-inch geoprobe, it would take only approximately .8 drops to cause the levels of contamination reported there. The City never took soil samples at WM-1. Therefore, McBean contends that the City has faded to properly investigate whether the groundwater contamination at WM-1 might be due to a direct release onto the soil at the Walter Morris property. Instead, the City simply attributes that contamination to APCO. McBean makes a persuasive argument. It is quite peculiar that just a few feet north of WM-1 (and still on the Walter Morris property), well 828-8B showed DCE concentrations far less than both WM-1 and some of the sample points underlying APCO. This strongly suggests that wherever the WM-1 contamination originated, it did not migrate south from APCO. It also raises the unanswered question of why the City did not consider Walter Morris as a PRP.

In the end, the dispositive fact is that chlorinated solvents were found in the soil and groundwater at several locations on the APCO property. The debate about whether groundwater contamination may have migrated under the APCO property from other locations can never be conclusively resolved one way or the other. However, the only credible explanation for finding chlorinated solvents in the APCO soil is that they were spilled on the ground at the APCO facility. Moreover, the contamination was measured over 20 years after any spills at APCO would have occurred. The court reasonably infers that 20 years of leaching and diluting has reduced the chlorinated solvent concentrations, particularly in the soil, and that readings taken closer to the time of the spills would have yielded even higher concentrations than we see today. The court further concludes that chlorinated solvent concentrations in the shallow groundwater underlying APCO, combined with the evidence of significantly lower chlorinated solvent concentrations in the shallow groundwater upgradient of the facility, show that the chlorinated solvents released to the soil at APCO have leached to the underlying groundwater in concentrations sufficient to exceed the ACLs for some or all of those contaminants and their daughter products.

2. Reid Supply Company

Reid Supply Company sold supplies to dry cleaners, mainly in Wichita. Among the supplies were chlorinated solvents, primarily PCE, although Reid also handled some TCE. In 1975, Reid purchased a warehouse located at 911 E. Indianapolis (warehouse or 911). See App. C-3. 29 It occupied the warehouse until 1988. PCE, and to a lesser extent TCE, was stored at 911, primarily in 55 gallon steel drums. The drums containing the chlorinated solvents were delivered by truck to Reid’s customers, which used the solvents in their dry cleaning operations.

The exact date of original construction of the warehouse at 911 was not established, but it was at least 25 years old when Reid purchased it. The floor of the warehouse is concrete of unknown thickness. It is elevated approximately 3-4 feet above ground level, presumably to accommodate deliveries by truck or rail to a covered loading dock which runs part way along the west side of the warehouse. It was not established at trial whether there is open space between the warehouse floor and the ground or whether the space is filled with dirt.

Starting in 1975 and continuing through 1978 (and perhaps as late as 1981), PCE was delivered to 911 by Vulcan Chemical using a bulk tank truck which parked at the loading dock near what was later to become a geoprobe site called RSI-7 and a monitoring well known as 946B. The amount and frequency of the PCE deliveries was not precisely established at trial and is not all that important. It certainly amounted to between several hundred and 3,000 gallons per month.

The evidence regarding what happened during the transfers was not uniform, due in part to the passage of time and the differing responsibilities of the Reid employees who testified. Jerry Letterman supervised the warehouse from 1975 to 1978. He recalled that PCE was stored and sold in 55 gallon drums and perhaps 5 gallon and 1 gallon cans. When bulk deliveries were made, Vulcan’s tank truck pulled up to the loading dock. A hose from the truck was run through one of the doors on the warehouse to a line of empty drums. The drums were then filled with PCE and stored in the north end of the warehouse until they were delivered to Reid’s customers. The 5 and 1 gallon containers of PCE were manually filled from “spigots” on the 55 gallon drums. Mike Letterman, Jerry’s son, who worked at the warehouse from about 1975-78, testified similarly.

The evidence established that from time to time, “spills” of PCE (and perhaps TCE) occurred inside the warehouse. The amount ran the gamut from 4 or 5 spills of 5 to 25 gallons of PCE to drips or small overflows on top of the drums which occasionally got onto the concrete floor. Usually, the drips were wiped up with a cloth. Gene Stamm, who acted as general manager of the warehouse, could not recall any spills but he also testified that he would not have been present if spills had occurred when drums were filled from the tanker. Charles and David Trombold testified that they were not aware of any spills, but Charles did not work at the warehouse until the early 1980s and David not until 1981, after bulk deliveries had ceased. Charles Millsap saw no spills, but he was not hired by Reid until 1986. David Shaw also saw no spills but he did not begin work until 1979. The testimony of these late-comers regarding the absence of spills was not persuasive. Mike Evers-man, one of Reid’s experts, testified that there were spills totaling 5-7 gallons, in addition to drips, based upon his review of deposition testimony of Reid’s employees.

The court finds from the credible evidence that from 1975 until 1978, and perhaps as late as 1981, there were occasional spills of chlorinated solvents, primarily PCE, onto the concrete floor. The court accepts Eversman’s testimony regarding the amount spilled, although the amount is not material for the reasons hereafter stated. It is inconceivable that no spills occurred over this period given the frequen cy and amounts of chlorinated solvents stored and transferred at 911. Moreover, there is no evidence that spills, particularly small ones, were events of environmental concern to Reid at the time. Indeed, based upon the totality of the evidence at trial, relatively small spills or releases of chlorinated solvents were not especially worrisome to anyone in the 1970s, except perhaps environmentalists.

PCE can permeate concrete, depending on factors such as the concrete’s age, thickness and the existence of cracks or voids. The thickness of the concrete floor was not established. Walter Trombold testified that there were small cracks in the floor, which he filled. David Trombold saw “hairline” cracks. There were cracks in the floor at the time of trial but the age and depth of the cracks were not established. Eversman “scanned” the cracks and construction joints with an organic vapor analyzer to screen for the presence of volatile organic vapors or compounds. The results were negative but also inconclusive as to the presence of chlorinated solvents during the period of Reid’s occupancy of 911. The most persuasive evidence that PCE or TCE spills permeated the concrete floor would have been from soil samples taken directly beneath the building. No such samples were taken.

Based on the totality of the evidence, the court finds that the City has not met its burden to demonstrate that spills or drips of chlorinated solvents inside the warehouse permeated the concrete floor and reached the soil and groundwater underneath the building. This finding does not end the inquiry, however, because of the evidence of chlorinated solvents in the soil and groundwater at RSI-7 and in the groundwater at 946B. Some additional background information is needed to place this evidence in perspective.

RSI-7 is a geoprobe location about 13 feet west of the loading dock on the west side of the warehouse. It is where the Vulcan tank truck parked when PCE bulk deliveries were made. When Martin geo-probed RSI-7 in April 1998, he followed the mandated procedures. -Another of Reid’s experts, David Gallis, criticized Martin’s method of analysis and described supposedly more accurate methods such as off-site testing. As previously noted with respect to Martin’s testing at APCO, the court is not persuaded that Martin’s methods were unorthodox or so prone to error that Martin’s on-site test results should be disregarded. Martin understood that if screening revealed the presence of chlorinated solvents, additional samplés would be taken and sent to an off-site certified laboratory. Nevertheless, for reasons never fully explained, it does not appear that any off-site testing was done at RSI-7. That was not Martin’s responsibility or fault.

Martin’s on-site chemical analysis of the soil at RSI-7 revealed the presence of PCE, TCE and DCE at 1, 4, 8, 12 and 14.5 feet bgs. PCE, TCE and DCE, as well as TPH (gasoline) were found in groundwater at depths ranging from 19-26 feet. The presence of PCE in the soil at RSI-7 is consistent with spills having occurred there. Later, in March and May 2002, PCE, TCE, DCE and VC were found in groundwater samples taken at permanent monitoring well 946B, which is adjacent to and slightly south of RSI-7.

At the KDHE’s request, CDM did extensive testing of carbon and sulfate concentrations at the Site, including at 911. As explained elsewhere in this decision, the presence of carbon, such as petroleum or other organic materials, potentially accelerates the degradation or dechlorination process. Sulfates, on the other hand, potentially retard the process. CDM determined that sulfate concentrations at the Site are “naturally very high” except, curi ously, at APCO and 911, where “extra carbon” is said to be present.

The presence of “extra carbon” at 911 may (or may not) be explained by historical uses of the property. During the 1930s and 40s, prior to the construction of 911, the area occupied by the northern part of the building was a coal yard. In the middle was a bulk oil station. Just south of 911 were two underground fuel storage tanks, which were found to be leaking when removed in the 1980s. In addition, Reid had an underground gasoline storage tank near the northeast corner of 911. All of these locations were potential sources of petroleum hydrocarbons and high levels of petroleum were found in the groundwater near where the underground tanks were removed. In addition, prior to 1950, the area immediately to the northwest of 911 was a junkyard. Farther to the north was a railroad maintenance yard. Because of the generally north-to-south direction of groundwater flow, both of these areas were theoretical sources of petroleum hydrocarbons, as well as chlorinated solvents, although no soil or groundwater samples were taken in either location.

The City’s expert witness Olsen offered his opinion that 911 was a source and is a continuing source of chlorinated solvent contamination at the Site. Olsen focused on RSI-7 as a source of contamination based on evidence of PCE and TCE in the soil. Because gasoline was also present, he believed that significant dechlorination has occurred which also explains the concentrations of VC in the shallow groundwater. Olsen testified that petroleum products such as gasoline are depleted during the dechlorination process and that as the petroleum products are depleted, sulfate concentrations in shallow groundwater can increase from upgradient flow. It was Olsen’s opinion that the current high readings of sulfate at RSI-7 could be accounted for because the gasoline has been depleted and sulfates have flowed back into RSI-7 from upgradient sources. When the chlorinated solvents reach areas where the sulfate levels are high (and levels of carbon are correspondingly low), dechlorination of PCE and TCE slows down and dechlorination of DCE to VC may cease altogether, thus accounting for DCE’s presence in the groundwater. Olsen’s ultimate conclusion was that the DCE present in shallow groundwater at RSI-7 and 946B is the result of dechlori-nation at that location, not from an upgra-dient source.

Olsen’s views were both supported and refuted by the City’s witness Hendron, who offered opinions regarding sources of contamination. Hendron accepted the soil data at RSI-7, which showed PCE, TCE and DCE in the soil. Because DCE is a daughter product, a reasonable inference from this testimony is that Hendron also would accept that dechlorination has occurred in the soil at RSI-7, in contrast to Olsen’s statements that dechlorination occurs only in groundwater.

Olsen’s opinions regarding 911 were vigorously challenged on cross-examination and by the testimony of Reid’s witness, Eversman. No useful purpose will be served by summarizing the details of Olsen’s cross-examination testimony. Olsen did admit that no shallow groundwater reading at 911 exceeds either the ACL or Maximum Contaminant Level (MCL) for PCE or TCE. 30 However, shallow groundwater readings for DCE dramatically exceed the ACL at RSI-7 and 946B. Olsen steadfastly held to his opinion that migration was not a factor at RSI-7 because it is a source of contamination. Notwithstanding the opinion of their witness, the City has acknowledged the obvious: that high readings of DCE at 911 can be partially explained by groundwater migration of “very low levels of DCE” from upgradient sources.

Reid’s explanation for the chlorinated solvents at RSI-7 and 946B is multifaceted. First, Reid asserts that the KDHE has not determined that 911 is a source of contamination. Second, it contends that any contamination at and around 911 migrated there from upgradient sources, not from dechlorination of spills at 911. Third, it argues that any contamination around its property can be explained by sewer leaks.

The evidence regarding the KDHE’s determination of 911 as a source of chlorinated solvent contamination is based on the testimony of Christine Jump, KDHE project manager of the Site from 1994 to present, who stated on direct examination that:

Q Similarly, based on the soil and groundwater data as well as information about chlorinated solvent at the facility, has KDHE determined whether there has been a release of chlorinated solvents from the former Reid Supply facility at 911 E. Indianapolis?

A I believe that Reid Supply is a source of contamination to the groundwater that needs additional investigation as well.

On cross-examination, Jump testified:

Q Okay. And, in fact, you’ve had discussions as recently as I think the first quarter of this year with Roger Olsen that you needed more information about what was going on at Gilbert and Mosley before you would, you the KDHE, would be willing to sign off on whether a party was a source or a continuing source?

A Yes. We asked for additional monitoring wells.

Q And as I understand it, you are still not in a position today to determine whether Reid Supply is a continuing source at Gilbert and Mosley?

A I believe it is based on the data I’ve seen; but I want additional data to confirm that, yes.

Q Right. You don’t have enough information right now to confirm one way or the other. You need more?

A Yes, that’s accurate.

The court finds from this testimony that Jump believes 911 is a source of contamination but is reserving her opinion regarding whether Reid is a continuing source. Significantly, Jump never testified that the KDHE did not consider 911 to be a source. Rick Bean, the KDHE project manager who preceded Jump, did not testify one way or another whether he considered 911 to be a source. Based on all the evidence, the court finds that the KDHE believes 911 is a likely source of chlorinated solvent contamination at the Site.

Reid’s next contention is that any contamination found at and nearby 911 came from upgradient sources. It bases this contention on a combination of three factors: (1) groundwater contamination readings from geoprobe and monitoring well locations north of 911; (2) the presence of potential upgradient sources of contamination and (3) the generally north to south flow of groundwater throughout the Site.

There are four wells north of 911 which were sampled for groundwater contamination. They lie in an west-east line along the north side of Indianapolis Street. 911 is on the south side of Indianapolis Street. Well 253 is northwest of 911; wells RSI-5 and RSI-1 are immediately north and PTW-2 is northeast. The groundwater at 911 flows about one foot per day, generally in a north to south direction, although the direction of localized flow is a matter of dispute. Obviously, though, the presence of chlorinated solvents in groundwater north of 911 cannot be due to spills or releases at 911.

Eversman was Reid’s principal witness on the issues pertaining to chlorinated solvent contamination at 911. In a nutshell, Eversman’s opinion was that 911 is not a source of contamination at the Site. He testified that dechlorination, which he called “biodegradation,” was not occurring at 911 and that the DCE found in groundwater samples was not the -degraded result of a release of PCE at 911, but instead was the result of migration from upgradient sources. Eversman opined that remediation “may not be necessary” because chlorinated solvents are not leaking into groundwater. Eversman noted that the City was not remediating similar conditions at the Bus Barn and Kellogg.

To explain his opinion that dechlorination was not occurring at 911, Eversman discussed both the soil and groundwater test results. Although he disagreed with the quality of the soil data taken in 1998 at RSI-7,- he accepted it. He noted the absence of data regarding carbon in the soil at RSI-7, which presumably supported his view that dechlorination was not occurring in the soil. Comparing and contrasting the soil and groundwater data readings, Eversman opined that if dechlorination was still occurring in the soil at RSI-7, he would expect higher PCE and TCE readings in the groundwater, and more DCE in the soil as well as at the interface between the soil and groundwater.

Eversman next compared the groundwater readings taken in 1998 and 2002 at RSI-7 and 946B, respectively. He noted that the 2002 DCE readings decreased in the order of 65-70% from the 1998 readings, which he considered “indicative of a source perhaps not as much mass or having expired as a source area .... ” He pointed out that in 1998, the TPH (gasoline) reading at RSI-7 was 857 ppb. By 2002, it was a non-detect, which suggested the absence of carbon, a “food source” for dechlorination.

Eversman acknowledged that the sulfate reading of 249 ppm from CDM’s tests was sufficient to inhibit the dechlorination process. He tested for sulfides. A positive test is an indicator of reactions between carbon (TPH) and sulfates, which would point to ongoing dechlorination. The results were negative.

Having eliminated, in his view, dechlori-nation in the soil as an explanation for chlorinated solvents in the groundwater at Reid, Eversman amplified on his opinion that upgradient sources accounted for the contamination. Eversman noted that some of the test wells north and west of 911 had groundwater PCE readings equal to or higher than any at 911, which suggests dechlorination of PCE north of 911 as a possible explanation for the DCE readings at 911. On cross-examination, Eversman acknowledged that PCE also was detected throughout the soil at RSI-7, which is consistent with PCE spills at that location.

Finally, Reid presented rather minimal evidence that chlorinated solvent contamination could be from leaks in sewers. Ev-ersman identified two particular points: (1) northwest of 911 in a line running from west to east along Indianapolis and (2) approximately one-third of the way down the east side of 911 in a line running in an alley. Eversman relied upon soil gas studies done in 1989 showing elevated levels of chlorinated solvents at those locations. He admitted, however, that he had no evidence that chlorinated solvents were in the sewers or that they had even been discharged into the sewers.

Taking all the evidence into consideration, as well as the reasonable inferences therefrom, the court finds that the City has met its burden to show that Reid is responsible for some of the chlorinated solvent contamination at and around 911. It is important to keep in mind that at 911, as well as the other locations at the Site, chlorinated solvent spills occurred many years ago. The contamination first had to leach through the soil, a slow process. Finally, if and when the contamination reached the groundwater' — whether at a specific test location or upgradient of one — dechlorination was not instantaneous but rather, too, was slow. The observations and widely divergent opinions of the experts were based on relatively current events. Thus, the fact that readings taken in 1998 or 2000 do not exceed the ACL does not mean that they never did. On the contrary, when evidence of spills is combined with evidence of contamination at the same location, it is reasonable to infer that had readings been taken closer to the time of the spills, they would have been higher. Therefore, current readings below the ACLs do not automatically exonerate a defendant from responsibility for past contamination.

The evidence of chlorinated solvents in the soil and groundwater at RSI-7, as well as the groundwater at 946B, supports a reasonable inference that PCE was spilled there during bulk deliveries by Vulcan. Since chlorinated solvents do not occur naturally in soil or groundwater, what other explanation can there be? Although upgradient sources are a plausible explanation for groundwater contamination, they simply do not account for contamination in the soil at 911. Reid is responsible for the contamination resulting from these spills because of its contractual relationship with Vulcan. The evidence supports the conclusion that PCE from these spills has dechlorinated. It is reasonable to assume that PCE spilled in the area of RSI-7 and 946B eventually entered the groundwater and degraded to DCE and VC as it moved to the southeast. This tends to explain the presence of DCE and VC in the groundwater at RSI-10 and RSI-4 and RSI-11, all located southeast of RSI-7 and 946B.

The court also finds from the generally more persuasive evidence presented by Reid that some of the contamination at 911 migrated there from upgradient sources. The court rejects, however, leaks from sewers as a contributing factor. There is no evidence that chlorinated solvents were discharged upstream into the sewers and/or that chlorinated solvents, in fact, were in the sewers. Therefore, the only way chlorinated solvents could get into the sewers would be from infiltration from surrounding soil. The chlorinated solvents then would have to exfiltrate, or leak out, at other locations. This is simply too speculative to credit.

3. Land Tool Company

During the period from approximately 1975 to 1988, Land Tool Company and E.H. Land operated a motorcycle and athletic helmet manufacturing business at 650 E. Gilbert Street, where chlorinated solvents, including TCE, were handled, stored, and used. See App. C-4. 31 Land Tool used a TCE vapor bath to polish its helmets. TCE was poured into a heated container which created a vapor mist. Helmets were placed into this vapor mist to create a polished finish.

Land Tool kept a 55-gallon drum of TCE in the same room as the vapor bath so that additional TCE could be poured in as it was needed, approximately twice each day. TCE was hand-pumped from the drum into a one-gallon bucket, which was then used to pour the TCE into the vapor bath container. During this process, TCE periodically sloshed or dripped onto the floor. Land Tool employees occasionally mopped the floor to keep it clean. The mop water was then dumped outside the building. About once per week, after they had pumped as much TCE out of a drum as possible, Land Tool employees set the drum outside the building and brought in another drum. Some TCE was always left in each barrel when it was taken outside.

Shallow groundwater samples taken immediately upgradient of 650 E. Gilbert contained non-detectable or very low concentrations of PCE and TCE while shallow groundwater samples taken immediately downgradient of 650 E. Gilbert contained substantially higher concentrations of TCE, which were above the ACL. Sample location BOG-10, which contained ■ the highest concentration of TCE in the shallow groundwater, was taken inside the footings of the building that had been located at 650 E. Gilbert. Considering the totality of the circumstances, chlorinated solvents were released at the Land Tool facility, causing groundwater contamination above the ACLs.

4. Tri-Supply Company

From approximately June 1, 1993 through November 30, 1996, Tri-Supply operated 330 S. Commerce as a dry cleaning supply facility at which Tri-Supply filled and stored 55-gallon drums of PCE, and from which it delivered PCE to Wichita area customers. See App. C-5. 32 Between June 1994 and June 1995, Tri-Sup-ply received at least 8 bulk tank truck deliveries of PCE totaling more than 8,100 gallons.

Shallow groundwater samples taken immediately upgradient of 330 S. Commerce contained maximum PCE concentrations of 52 ppb, while immediately downgradient samples contained 2,600 ppb of PCE. Soil samples taken near the 330 S. Commerce loading dock contained 937 ppb of PCE at 12 feet, 351 ppb of PCE at 14 feet, and 626 ppb of PCE at 15 feet. In addition to the analytic data evidencing groundwater and soil contamination, there is documented evidence of a spill of chlorinated solvent occurring at this facility. On June 21, 1994, a Tri-Supply employee spilled 2 gallons of PCE at 330 S. Commerce while filling a 55-gallon drum. Based on the evidence of soil and shallow groundwater contamination near the facility, the court concludes that a release of chlorinated solvents occurred at 330 S. Commerce that has contaminated the underlying groundwater above the ACLs.

C. The Releases Caused Plaintiff to Incur Costs

The court must now turn to the City’s response to the presence of chlorinated solvents at the Site. A prima facie case of CERCLA liability pursuant to § 113 requires the City to show that the respective releases of chlorinated solvents at the Site caused the City to incur response costs. See Morrison, 302 F.3d at 1135. To this end, the City presented evidence that releases at the Site have caused it to incur $13,329,954.50 in response costs through June 15, 2002. The court is satisfied, based in part on the prior discussion regarding the release of chlorinated solvents and on the other evidence adduced at trial, that the release of chlorinated solvents from each of defendants’ facilities to the underlying groundwater was sufficient to require the City to incur response costs.

D. Response Action Compliance with the NCP

The City must furthermore establish, as part of its prima facie case of CERCLA liability pursuant to § 113, that its response action or cleanup was consistent with the NCP. See Morrison, 302 F.3d at 1135-36; see also 42 U.S.C. § 9607(a)(4)(B) (liability includes “any other necessary costs of response ... consistent with the national contingency -plan”).

1. The NCP

“The NCP is EPA’s regulatory template for a ‘CERCLA quality cleanup.’ ” Pub. Serv. Co. of Colo. v. Gates Rubber Co., 175 F.3d 1177, 1181 (10th Cir.1999) (citation omitted). It sets forth a very detailed and comprehensive list of requirements governing various aspects of the City’s response action. See 40 C.F.R. § 300.700(c)(5)-(6) (listing NCP requirements and the corresponding regulations explaining each); Morrison, 302 F.3d at 1136 (“The NCP is a long and detailed list of procedures that must be carried out by federal and state governments when they are responding to hazardous waste releases.”); OHM Remediation Servs. v. Evans Cooperage Co., Inc., 116 F.3d 1574, 1579 (5th Cir.1997) (noting that the NCP “sets performance standards, identifies methods for investigating the environmental impact of a release or threatened release, and establishes criteria for determining the appropriate extent of response activities”).

Despite the rather detailed manner in which these requirements are set forth, strict, to-the-letter compliance is not necessary. A “substantial compliance” standard was added to the NCP by a revision on March 8, 1990. The revision became effective on April 9, 1990. See National Oil and Hazardous Substances Pollution Contingency Plan, 55 Fed.Reg. 8666, 8666 (March 8, 1990). Because the City first incurred response costs in 1991, the court will apply the 1990 revisions to the City’s response action. See Morrison Enters. v. McShares, Inc., 13 F.Supp.2d 1095, 1112 (D.Kan.1998) (reversed on other grounds by Morrison, 302 F.3d at 1130). “Substantial compliance” is therefore the relevant touchstone. See 40 C.F.R. § 300.700(c)(3)(i); see also 55 Fed.Reg. at 8793 (noting that “strict compliance with [the] list [provided in § 300.700(c)(5)-(7) ] of NCP provisions is not required in order to be ‘consistent with the NCP’; the list is provided ... as guidance to private parties on those requirements that may be pertinent to a particular site”); Morrison, 302 F.3d at 1136 (“Private parties need only show ‘substantial compliance’ with the NCP in order to meet the requirements of § 9607(a), § 300.700(c)(3)(i) ....”).

The specific requirements of the NCP depend also on “whether the response action is characterized as a removal or a remedial action.” Pub. Serv. Co., 175 F.3d at 1182. A remedial action includes “those actions consistent with [a] permanent remedy.” 42 U.S.C. § 9601(24). A removal action is less permanent and “is generally an emergency, interim response to particular site conditions.” County Line, 933 F.2d at 1512 n. 6. It typically shall not exceed $2,000,000 in costs and shall not take longer than 12 months to complete. See 42 U.S.C. § 9604(c)(1); see also Exxon Corp. v. Hunt, 475 U.S. 355, 360, 106 S.Ct. 1103, 1108, 89 L.Ed.2d 364 (1986) (“Governmental response consists of ‘removal,’ or short-term cleanup, § 9601(23), and ‘remedial action,’ or measures to achieve a ‘permanent remedy’ to a particular hazardous waste problem, § 9601(24).”).

The court has heard no evidence that the contamination at the Site posed a threat to human health or the environment which required an immediate response. See Carson Harbor Village, Ltd. v. Unocal Corp., 287 F.Supp.2d 1118, 1157-58 (C.D.Cal.2003) (finding that the response action was a remedial action in part because there was “no evidence in the record that the materials posed the type of threat to human health and welfare that required immediate action”). And as far as the court can tell, defendants do not claim otherwise. In light of the various remedial goals expressed by the City at trial, and because the City’s response action exceeds the bounds specified in § 9604 in terms of both cost and the length of time required for completion, the court deems the City’s action remedial in nature and characterizes it as such when judging the City’s compliance with the NCP.

The requirements governing the City’s remedial action relate to “(1) worker health and safety; (2) documentation of cost recovery; (3) permit requirements; (4) identification of applicable or relevant and appropriate requirements (ARARs); (5) remedial site evaluation; (6) remedial investigation/feasibility study and selection of remedy (RI/FS), and (7) ... an opportunity for public comment.” Pub. Serv. Co., 175 F.3d at 1182 (citation omitted). Some of these requirements were not the subject of testimony. There was substantial disputed testimony regarding the City’s selection of a remedy and, in the process of doing so, its rejection of what has been referred to as “natural attenuation,” or “monitored natural attenuation,” which essentially means allowing the groundwater to clean itself over time. The court thus thinks it necessary to set forth in more detail the requirements governing the City’s remedy selection.

To select a remedy in a manner consistent with the NCP, the City must first develop an RI to “collect data necessary to adequately characterize the site for the purpose of developing and evaluating effective remedial alternatives.” 40 C.F.R. § 300.430(d)(1). Using that data, the City must then “conduct a site-specific baseline risk assessment to characterize the current and potential threats to human health and the environment.” Id. § 300.430(d)(4). The NCP next requires the City to conduct an FS “to ensure that appropriate remedial alternatives are developed and evaluated.” Id. § 300.430(e)(1). Pursuant to the FS, the City must outline various alternatives, including a “no action” alternative, that are protective of human health and the environment. Id. at § 300.430(e)(2),(6).

The NCP requires the City to screen the alternatives that it identified in the FS in light of their effectiveness, implementa-bility, and cost. Id. at § 300.430(e)(7)(i)-(iii). The City must then conduct a more detailed analysis of the alternatives still considered viable after the initial screening. Pursuant to the detailed screening, (1) each alternative must satisfy two threshold criteria: overall protection of human health and the environment and compliance with ARARs, and (2) each must be considered in light of five primary balancing criteria: long-term effectiveness and permanence, reduction of toxicity, mobility, or volume through treatment, short-term effectiveness, implementability, and cost; and two modifying criteria: state and community acceptance. Id. at § 300.430(e)(9)(iii),(f)(l)(i)(A)-(C). The City must additionally present the selected remedial alternative to the public in a proposed plan. Id. § 300.430(f)(2).

2. General Description of City’s Response Action

During investigations conducted pursuant to the EPA/KDHE multisite agreement, the KDHE detected chlorinated solvent levels in the groundwater at the Site that exceeded federal drinking water standards. It expected the contamination to pass beneath the Arkansas River, thus threatening public water supply wells to the south of the river.

As a result of its findings, the KDHE determined that a remedial investigation and feasibility study (RI/FS) was necessary. In 1992-1993, pursuant to the City/ KDHE agreement, the City collected and analyzed data, which it compiled into its draft RI/FS and submitted to the KDHE. 33 Following a public review and comment process, the KDHE approved the final RI on January 26, 1994, and the final FS on April 8, 1994. These reports described the geology and hydrogeology of the Site and surrounding region and the extent of groundwater contamination and other environmental conditions at the Site, evaluated contaminant fate and transport and biodegradation processes, identified potential contaminant source areas, and included a baseline risk assessment evaluating the threats to public health and environment posed by the chlorinated solvent contamination.

The KDHE prepared a draft Corrective Action Decision for Interim Groundwater Remediation (CAD), which summarized the RI/FS and, based on the results it contained, initially approved a “pump and treat system” as the preferred remedial alternative for groundwater remediation. Following a public review and comment process that included an opportunity for the public to comment on the selected remedial alternative from June 6, 1994, to July 6, and a public meeting held by the KDHE and the City on June 21, 1994, the KDHE issued a final CAD for Interim Groundwater Remediation on September 28, 1994. In it, the KDHE ultimately rejected various alternatives for remediation and verified the pump and treat system as the preferred remedial alternative.

The pump and treat system essentially involves the following: (1) hydraulic containment involving groundwater extraction and treatment to acceptable levels; 34 (2) institutional controls, including the prohibition of newly constructed private water wells for private or public drinking' water purposes; (3) compliance monitoring; (4) long-term monitoring; (5)' individual source control activities; and (6) microbiological studies to determine the efficiency of microbiological'enhancement.

The pump and treat system is designed to pump groundwater, contaminated above the ACLs, out through extraction wells, treat it to reduce chlorinated solvent levels to the MCLs, and then ultimately release the water into the Arkansas River. The system required the City to identify areas where the contamination exceeded ACLs, develop the number, location and size of extraction wells and create a system for groundwater transport, treatment and disposal.

After the KDHE issued the 1994 CAD, the City continued to collect data and submit drafts of work plans and reports to the KDHE for its review. In 1995, the City collected .groundwater data using geo-probes and permanent monitoring wells and prepared a QAPP to ensure the quality of the data collected. The City and CDM compiled the data and presented it to the KDHE in two reports: a Pre-De-sign Data Acquisition Draft Field- Activities Summary Technical Memorandum, dated September 1995, and a Preliminary Design Report, dated February 1996. Using the 1995 data, CDM updated a mathematical model of groundwater flow and contaminant transport that it initially created to assess the movement of groundwater and contaminants within the Site, to evaluate the effectiveness of remedial alternatives and to determine the extent of the contaminant plumes from the various sources within the Site.

The KDHE reviewed and commented upon the refined groundwater model and its comments were incorporated into a Preliminary Design Report. As required by the CAD, the City also conducted a bio-pilot demonstration study to determine whether innovative remedial alternatives involving bioremediation might be feasible for the Site. In 1997 and 1998, the City collected additional data to investigate each ACL plume pursuant to the KDHE-reviewed and approved work plans to determine where to locate remedial systems within plumes and to conduct a more detailed investigation at source areas.

The City compiled and assessed the quality of the data in two reports: the Data Report and Assessment of Data Quality dated June 1998, and the Data Report and Assessment of Data Quality Addendum dated October 1998 (Data Quality Reports). The KDHE reviewed and commented on the Data Quality Reports and approved them on June 9, 2000. It determined that the data was suitable for the KDHE’s regulatory purposes, including finding sources of contamination, designing the remedy, and determining the nature and extent of contamination at the Site.

The City summarized and evaluated the data from plume investigations and evaluated remedial alternatives for design in its March 1, 1999, “Remedial Investigation and Feasibility Study Addendum.” The City further refined the groundwater model with the additional data and used it to re-evaluate remedial alternatives. The KDHE rejected monitored natural attenuation as a remedial alternative because it found that natural attenuation would take much longer than the pump and treat remedy and would not degrade chlorinated solvent contaminants before they reached the Arkansas River. The KDHE directed the City to proceed with the design of Alternative 2b in the RTFS Addendum, providing for the pump and treat containment and remediation of the commingled plumes A, B and E. The KDHE determined that the pump and treat system is necessary because the contamination plumes were becoming larger, were migrating towards the Arkansas River and into previously uncontaminated areas and were degrading natural resources, and that contamination must be contained and remediated, pursuant to the NCP, to restore aquifers to their most beneficial use.

Based on the data from plume investigations, the City completed the design of the pump and treat remedial system and submitted the corresponding design reports to the KDHE. After the City responded to the KDHE’s comments on the design reports, the KDHE reviewed and approved the Final Design Report and directed the City to prepare the Remedial Action Work Plan for construction of the remedy.

On February 23, 2001, the City submitted to the KDHE the Final Remedial Action Work Plan for the downgradient groundwater remediation of plumes A, B, and E and subsequently began to construct the pump and treat remedial system. According to the work plan, the system was to include 13 extraction wells, placed at various locations within the Site based on the groundwater and transport model, and approximately 5.6 miles of piping for conveying the extracted groundwater to a treatment facility located at Herman Hill Park. The system also was to include six venturi air strippers that strip contaminants out of the groundwater.

In March 2001, the KDHE issued an “Explanation of Significant Differences to the Final Corrective Action Decision, Gilbert and Mosley Interim Groundwater Remediation” (ESD) to describe to the public certain changes to the remedy selected in the KDHE’s 1994 CAD. The ESD informed the public that the remediated groundwater from the Site was to be discharged to the Arkansas River. The ESD also notified the public that a minimum of seven air strippers were to be utilized in the remediation of the groundwater and estimated the time necessary to achieve the remedial goals of the CAD for plumes A, B and E to be 60 years. The KDHE approved all modifications described in the ESD and concluded that the revised remedy complies with CERCLA, is necessary and protective of human health and the environment and is cost effective.

3. Standard of Review for KDHE Determinations

All parties are in apparent agreement, and correctly so, that a de novo standard applies to the court’s review of any legal determinations made by the KDHE in this matter. The parties dispute, however, the correct legal standard applicable to the KDHE’s factual findings and draw opposite conclusions from the Tenth Circuit’s statements in Amisub (PSL), Inc. v. State of Colo. Dep’t of Soc. Servs., 879 F.2d 789 (10th Cir.1989). 35

The court recognizes that CERCLA authorizes the President to “provide for remedial action relating to such hazardous substance ... which the President deems necessary to protect the public health or welfare or the environment.” 42 U.S.C. § 9604(a)(1). As the President’s delegated agent, the EPA can also initiate a remedial action. See Key Tronic Corp. v. United States, 511 U.S. 809, 814, 114 S.Ct. 1960, 1964, 128 L.Ed.2d 797 (1994). CERCLA specifies that “the court shall uphold the President’s [or EPA’s] decision in selecting the response action unless the objecting party can demonstrate, on the administrative record, that the decision was arbitrary and capricious.” 42 U.S.C. § 9613(j)(2). But based on the evidence presented at trial, the court finds that the EPA played only a minimal role in the remedy selection process beyond its initial entry into the EPA/KDHE multi-site agreement.

The City nevertheless states that “the President delegated his authority to enter into cooperative agreements to EPA,” see Exec. Order No. 12,580 § 2(i), 52 Fed.Reg. 2,923 (Jan. 23, 1987), and that “[a]fter the President issued this order, KDHE and EPA entered into the cooperative agreement.” The City thus concludes that “[h]aving acted within the authority given to it under federal law, KDHE’s findings and determinations are entitled to deference unless they are found to be arbitrary and capricious.” But this contractual relationship between the EPA and the KDHE and finally the City seems too attenuated, by itself, to justify review according to the arbitrary and capricious standard of § 9613(j)(2).

The court is reluctant to defer to the factual determinations made by the KDHE, a state agency, regarding the City’s compliance with the NCP, a regulation enacted pursuant to federal law, without clear, on-point authority to do so. Therefore, the court will evaluate the KDHE’s factual determinations regarding NCP compliance de novo. See United States v. Akzo Coatings of Am., Inc., 949 F.2d 1409, 1425 n. 12 (6th Cir.1991) (“If EPA plays only a limited role in formulating a plan, then the President cannot be deemed to have taken or ordered the remedy, and a reviewing court would not be bound by the administrative record and the arbitrary and capricious standard”).

4. Presumption of NCP Compliance

The City contends that it is entitled to a rebuttable presumption that its response actions are consistent .with the NCP because the Site is included in the same pilot deferral program under the EPA-KDHE Cooperative Agreement as is the Scoular site at issue in Morrison, 302 F.3d at 1136-38. In Morrison, the Tenth Circuit, noting that the KDHE had approved the documents prepared by plaintiff in connection with the clean-up of the Scoular site, declared that

in the context of implementing this particular deferral program, Region VII of the EPA strongly indicated that compliance with the state orders, and acceptance of state oversight, would establish compliance with the NCP. Under 'the Cooperative Agreement between the EPA and the KDHE, the EPA stated that the KDHE’s system of oversight and implementation of cleanups of hazardous waste sites in Kansas is consistent with the NCP and achieves CERC-LA-protective cleanups.

Id. at 1138. The Tenth Circuit further noted without disagreement that this court concluded that the pilot deferral program under which the KDHE acted at the Scoular site, which is the same as at the Gilbert and Mosley Site, is the “ ‘functional equivalent of going through the federal Superfund process with EPA.’ ” Id. The Tenth Circuit concluded that “[gjiven the specifies of the EPA pilot program in this case, ... Morrison was entitled to a rebuttable presumption of compliance with the NCP based on the fact that its actions were undertaken pursuant to a consent order with the KDHE.” Id. at 1138.

As in Morrison, the City/KDHE Agreement and its Technical Scopes of Work for RI/FS and RD/RA require the KDHE’s oversight and approval of all of the City’s work at the Site. With respect to these Site activities, the court finds that the KDHE has directed, reviewed, commented upon, and approved all work performed by the City pursuant to the City/KDHE Agreement. Defendants do not directly dispute the KDHE’s actions in this regard, nor could they given the extensive interaction between the City and the KDHE. Instead, defendants argue that the City is not entitled to a Morrison rebuttable presumption because (1) the Morrison-KDHE agreement is a consent order whereas the City-KDHE agreement is a settlement agreement; (2) the Morrison presumption applies only to the unique facts of that case; (3) there are significant factual distinctions between this case and Morrison and (4) the City did not claim entitlement to a presumption in the pretrial order.

Defendants first try to distinguish a consent order from a settlement agreement. Defendants cite United States v. Acton Corp. ex. rel. Vikoa, 131 F.R.D. 431, 436 (D.N.J.1990) for the proposition that “[i]f there had been a consent order in this case, interested parties ... could have intervened and sought relief.” Defendants equate themselves with the “interested parties” in Acton.

Acton was a CERCLA case brought by the United States. Several PRPs sought leave to intervene pursuant to 42 U.S.C. § 9613(i) to oppose a proposed consent decree filed in accordance with 42 U.S.C. § 9622(d). A consent decree is an agreement that the parties desire and expect will be reflected in, and be enforceable as, a judicial decree. Rufo v. Inmates of Suffolk County Jail, 502 U.S. 367, 378, 112 S.Ct. 748, 757, 116 L.Ed.2d 867 (1992). In contrast, the Morrison consent order was not entered into pursuant to section 9622(d) and did not provide for, or even contemplate, submission to a court as a judicial decree. Indeed, the settlement agreement in this case contains no provision for enforcement as a judicial decree. Thus, Acton (the only decision cited by defendants) has no application to this case.

Morrison does not discuss the distinction between a consent decree and a consent order, nor a distinction, if one exists, between a consent order and a settlement agreement. Both the Morrison consent order and the settlement agreement at issue in this case were entered into with the KDHE pursuant to K.S.A. 65-3453 which uses the words “orders” and “agreements” without differentiating between the two. No evidence was offered through the KDHE witnesses regarding the difference to the KDHE, if any, between a consent order and a settlement agreement. The court concludes, for purposes of the Morrison rebuttable presumption, that there is no difference.

Defendants’ second argument is grounded on the following language from the Morrison opinion:

We hasten to add that we are not holding that a blanket presumption applies to all private party cleanups of hazardous waste sites undertaken as part of a state deferral program; our conclusion is specific to this program and this case. Cf. Pub. Serv. Co., 175 F.3d at 1179 [(10th Cir.1999)] (stating that in considering whether compliance with state orders should result in a presumption of compliance with the NCP, “the proposition may be sustainable under certain circumstances, [but] the facts here permit no such conclusion”). Pilot deferral programs may vary in their specifics across the country, and the EPA may also decide that any national program that it implements may be very different in its specifics, particularly in how it handles the issue of consistency with the NCP. We are wary of making any sweeping conclusions on the issue in light of the EPA’s cautionary language in 1990. See NOHSPCP, 55 Fed.Reg. at 8796-97.

Morrison, 302 F.3d at 1138-39. Defendants’ position is not clearly articulated, but they apparently believe that the Tenth Circuit was signaling that the Morrison presumption is unique to that case.

This court disagrees with defendants’ narrow interpretation. The Tenth Circuit panel noted that “[g]iven the specifics of the EPA pilot [deferral] program in this case, we conclude Morrison was entitled to a rebuttable presumption of compliance .... ” The “specifics” of the pilot deferral program require that all activities be consistent with the NCP and include a finding by the EPA that the KDHE’s site investigation and cleanup process are consistent with the requirements of CERCLA and the NCP. The City is part of the same pilot deferral program. Both in Morrison and in this case, the KDHE’s involvement has been comprehensive, including the KDHE’s approval of the City’s many proposals regarding clean up of the Site. Although the EPA’s involvement has been “minimal,” it has been no less so than in Morrison and there is no evidence that the EPA has rejected or criticized any action by the KDHE or the City as not being consistent with the NCP. On the contrary, the evidence is that the EPA was “on board” with the KDHE throughout the process. The panel’s cautionary admonition regarding application of a “blanket presumption” appears to be directed at pilot programs in other states, or at least at different pilot programs. Significantly, other than to cite the panel’s words, defendants make no argument as to why the Tenth Circuit would view this case as so different from Morrison that it would refuse to apply the presumption.

Defendants’ third argument — advanced without elaboration — is that the facts in Morrison and in this case are different. To be sure, but the factual differences are irrelevant, at least insofar as application of the presumption is concerned.

Defendant’s final argument is that the presumption is not mentioned in the pretrial order. Be that as it may, the City argues for the presumption in its trial brief. Defendants do not claim to be surprised by the argument, nor can they. Defendants are not really prejudiced by the presumption because all of their arguments regarding NCP non-compliance will be considered.

Application of the Morrison rebuttable presumption is therefore factually and legally appropriate. The presumption is not, however, irrebuttable. Its effect is to “shift the burden of proving inconsistency with the NCP.” Wash. State Dep’t of Transp. v. Wash. Natural Gas Co., 59 F.3d 793, 796 (9th Cir.1995). Defendants can thus rebut the presumption by demonstrating that the City’s remedy is in fact inconsistent with the NCP. See Morrison, 302 F.3d at 1139 (noting that defendant could perhaps rebut the presumption- by showing that “the KDHE-approved work-plan. was itself inconsistent with the NCP”).

5. Defendants’ Contentions

Defendants dispute the City’s presumptive NCP compliance on several fronts. Though the court will expressly address defendants’ primary contentions-those set forth mainly by ALT in Doc. 1432-in this Memorandum Decision, defendants asserted other arguments throughout the trial and in their various pleadings that relate to the consistency of the City’s response action with the NCP and which the court does not expressly address here. The court has considered the positions of defendants in their entirety, however, and determines that defendants have failed to rebut the presumption of NCP compliance due the City.

Defendants first contend that the City did not propose monitored natural attenuation as a remedial measure and apparently believe that the City had a duty to do so. They assert that because the City never officially proposed the no action or limited action alternatives discussed in the City’s FS, the KDHE consequently gave the options no credence or consideration. Defendants are correct in part. Jump testified that though “the information [regarding monitored natural attenuation] was presented to. KDHE ... there was never a request for KDHE to consider that [it] change the Corrective Action Decision to monitored natural attenuation.” Contrary to defendants’ assertions, however, the NCP simply requires that “[t]he no-action alternative ... shall be developed,” 40 C.F.R. § 300.430(e)(6), and does not otherwise place upon the City a burden to advocate for the use of monitored natural attenuation to the exclusion of other remedial alternatives.

Pursuant to § 300.430(e), the City submitted to the KDHE an FS that identified several remedial alternatives, including a “no action” alternative consisting of “no further remedial action.” The FS included a section titled “Detailed Analysis of Interim Remedial Action Alternatives” that further developed the “no action” alternative and a section comparing and summarizing each of the various alternatives presented by the City. The City later presented to the KDHE in the 1999 RI/FS Addendum an evaluation of the “bioremediation mechanisms within Plume B,” pursuant to its bio-pilot demonstration study, and concluded that the presence of excessive sulfate in the groundwater inhibits degradation at most locations encompassed by Plume B— a conclusion that undermines the selection of the “no action” alternative as the preferred remedial alternative.

The KDHE ultimately rejected the “no action” alternative in the Final CAD and affirmed its decision in the 2001 ESD. Because the NCP does not require the selection of natural attenuation in all eases but rather requires that a “no action” alternative be developed, and because the City did in fact present such an alternative in the FS, the court finds that the City’s ultimate development of a “no action” remedy was consistent with the NCP.

Defendants next contend that the KDHE’s selection of the pump and treat system over monitored natural attenuation was not consistent with the NCP in that operation of a monitored natural attenuation system 36 would cost far less and would require less time than will the pump and treat system. Cost-effectiveness is surely an important criterion to be considered in the selection of a remedial alternative. “Cost” itself is a factor to be used in both the initial screening and more detailed screening of remedial alternatives 37 and the NCP commands that “[ejach remedial action selected shall be cost-effective,” provided that it first is protective of human health and the environment and satisfies ARARs. 40 C.F.R. § 300.430(f)(l)(ii)(D). 38

ALT points to the City’s own groundwater modeling to conclude that monitored natural attenuation will achieve target cleanup levels in 50 years as opposed to the 60 to 70 years required by the pump and treat remedy. The City concluded in its 1999 RI/FS addendum, however, that monitored natural attenuation will require 130 years to treat the groundwater to desired levels due to various conditions at the Site which inhibit complete degradation of chlorinated solvents, 39 and that monitored natural attenuation will do so at a greater cost than will be required by the pump and treat system.

Though the parties argue, at length about the extent to which one alternative is more effective and less costly than another, the court is largely left with deter minations of credibility and generally views the conclusions reached by the City in this matter to be most credible. It indeed defies logic that a monitored natural attenuation alternative, whereby chlorinated solvents are left to degrade naturally, will actually require less time to restore groundwater to target levels than will a pump and treat system that, in addition to affirmatively removing chlorinated solvents via downgradient pumping, will also take advantage of any natural degradation tendencies. Apart from the cost/time dispute, however, the EPA’s expectations with respect to the selection of a remedy undermine any use of monitored natural attenuation as a remedy at the Site.

The EPA “expects to prevent further migration of the plume, prevent exposure to the contaminated ground water, and evaluate further risk reduction.” 40 C.F.R. § 300.430(a)(l)(iii)(F). To that end, defendants claim that the Arkansas River prevents further migration of contamination because the “plumes harmlessly enter the Arkansas River.” The river may in fact provide a natural barrier to the migration of Plume B beyond its borders and thus may prevent the endangerment of various public water supply wells south of the river. But the City expressed doubts about the “no action” and “limited action” alternatives because, as. stated in the FS, neither alternative truly “limit[s] migration of or remove[s] contaminants.” The City further stated more plainly in its 1999 RI/FS addendum that “during the long period before ACLs are achieved, the groundwater flow modeling indicates that Plumes A and B will continue to expand and Plumes A and E will discharge contaminants to surface water, the Arkansas River.”

The KDHE ultimately rejected the “no action” and “limited action” alternatives on this basis. It stated in the Final CAD that the “no action” and “limited action” alternatives “[do] not stop migration of contaminants.” Jump, who had ultimate authority over KDHE approval of the remedy, testified that she does not believe the KDHE would have accepted monitored natural attenuation if it had been proposed as an addendum to the CAD. Responding to an inquiry at trial by the City’s counsel about the reasons for her belief, Jump stated,

Because the data showed that there was continued migration and because the CAD, the intent of the CAD was to contain the plumes, not allow continued migration. I had told CDM that we could change the alternatives but we would not change the intent of the CAD. And also because in a meeting with the EPA and Coleman and the City of Wichita the, EPA said get on containment, things are continuing to move, proceed with containment. If you want to consider other alternatives, you can consider ‘em after you get containment on the plumes.

Because the NCP includes the protection of the environment among its threshold criteria for selecting a remedy, and furthermore because the EPA expects containment, at least in cases where returning usable groundwater to its beneficial use is not practicable, the court does not think it consistent with the NCP to select a remedy that allows contaminant flow into the river and away from the Site if another remedy will truly contain the contamination and compares favorably based on the NCP’s selection criteria.

ALT next claims that an actual threat to human health or the environment must exist in order for the City’s response action to be consistent with the NCP. See G.J. Leasing Co., Inc. v. Union Elec. Co., 854 F.Supp. 539, 561-62 (S.D.Ill.1994) (“In order to show that any response costs were necessary under CERCLA, plaintiffs must demonstrate they responded to a threat to public health or the environment.”)- ALT specifically contends that no remediation is necessary because exposure pathways to humans exist only hypothetically. ALT further notes that the City’s drinking water sources are upgra-dient to the Site, that the City itself has proclaimed that its water is safe, and that the RI specifically states that “[c]urrent exposures to residents are expected to be small to nonexistent, because current groundwater exposure pathways are incomplete. No significant ongoing exposures exist for current Wichita residents.”

ALT cites Bethlehem Iron Works, Inc. v. Lewis Indus., Inc., 1996 WL 557592 (E.D.Pa. Oct.1, 1996), and Southfund Partners III v. Sears, Roebuck and Co., 57 F.Supp.2d 1369 (N.D.Ga.1999), in support of its contention that the pump and treat system was not developed in response to a threat to human health or the environment. But the court in Bethlehem found it “most important ]” that the plaintiff failed to conduct a risk assessment study, Bethlehem, 1996 WL 557592, at *53, and the Southfund court stated that “Southfund has produced absolutely no evidence to suggest it posed any threat to the environment or public health.” Southfund, 57 F.Supp.2d at 1379. Here, however, far from an absolute lack of evidence, the City conducted an extensive Baseline Risk Assessment, pursuant to 40 C.F.R. § 300.430(d)(4), to “assess both existing environmental and human health risks by performing a no action alternative or baseline analysis, as well as potential risks associated with proposed remedial alternatives.”

Taking into account an assumption that most human exposure pathways are incomplete and might remain so, the City in its Risk Assessment combined chemical intake rates for various exposure scenarios with an assessment of the toxicity of the contaminants at the Site to conclude that cancer risks at the Site exceed 1x10 to 1x10 -6, the risk range deemed acceptable by the EPA. 40 See 40 C.F.R. § 300.430(e)(2)(i)(A)(2) (“For known or suspected carcinogens, acceptable exposure levels are generally concentration levels that represent an excess upper bound lifetime cancer risk to an individual of between 10 -4 and 10 -6 using information on the relationship between the dose and response.”). The City also identified over 400 water supply wells within a one-mile radius of the Site, including over 70 registered lawn and garden wells within the Site itself. The KDHE consequently determined that the contaminant level in the groundwater at the Site posed a human health risk.

Based on its independent review of the evidence, the court agrees. The court finds the Baseline Risk Assessment to be an accurate reflection of the risks in existence at the Site and determines that there is at least a potential risk to human health and the environment due to the contaminant level within the Site’s groundwater. See United States v. Burlington N. R.R., 200 F.3d 679, 684 (10th Cir.1999) (“Thus, the Risk Assessment has the specific purpose of providing a baseline snapshot of the potential risks, assuming that no remedial action has been taken.”) (emphasis added). The court thus rules that the City’s identification of potential risks and selection of the pump and treat system to effectively eliminate or reduce those risks satisfies the substantial compliance standard applicable to the City’s selection of a remedial alternative. See 40 C.F.R. § 300.430(a)(1) (“The purpose of the remedy selection pro cess is to implement remedies that eliminate, reduce, or control risks to human health and the environment.”).

ALT additionally alleges that the City’s extraction of the Site’s groundwater will be detrimental to human health and the environment. The potential for further detriment to human health is certainly a factor to be considered when choosing among vai’ious remedial alternatives. See also 42 U.S.C. § 9621(b)(1)(G) (requiring the President, when assessing alternative remedial actions, to take into account “the potential threat to human health and the environment associated with excavation, transportation, and redisposal, or containment”). But ALT cites no evidence that its contention is anything more than a self-serving assumption. The Ninth Circuit case that ALT references is of no help as the referenced language therein relates to the “covered person” analysis of 42 U.S.C. § 9607(a). See Kaiser Aluminum & Chem. Corp. v. Catellus Dev. Corp., 976 F.2d 1338, 1342-43 (9th Cir.1992). ALT’s position on this issue lacks merit.

ALT next argues that the groundwater at the Site should be classified as class III groundwater under the EPA’s preamble to the NCP. When selecting an appropriate remediation strategy for contaminated ground water at CERCLA sites, the EPA’s Superfund program utilizes as guidance a Ground-Water Protection Strategy that “establishes different degrees of protection for ground waters based on their vulnerability, use, and value.” The groundwater is ultimately classified as Class I, II or III and is treated accordingly. See 55 Fed.Reg. 8666, 8732.

For Class III groundwater in particular, drinking water standards are not ARAR and will not be used to determine preliminary remediation goals. Remediation timeframes will be developed based on the specific site conditions. The beneficial use of the ground water (e.g., agricultural or industrial use), if any, is determined; and the remediation approach will be tailored for returning the ground water to that designated use.

Id. ALT thus claims that a correct Class III classification of the groundwater at the Site precludes the City’s selection of the pump and treat system as the appropriate remedy.

Class III groundwater is defined as “groundwater that is unsuitable for human consumption-due to high salinity or widespread contamination that is not related to a specific contamination source-and that does not have the potential to affect drinkable or environmentally significant ground water.” 55 Fed.Reg. 8666, 8732. Class III groundwater specifically includes groundwater “(1) with a total dissolved-solids (TDS) concentration over 10,000 mg/1, or (2) that are so contaminated by naturally occurring conditions, or by the effects of broad-scale human activity (i.e., unrelated to a specific activity), that they cannot be cleaned up using' treatment methods reasonably employed in public water-supply systems.”

ALT apparently contends that groundwater should be categorized “Class III” if it has high salinity, a determination ALT claims should be made without reference to TDS. ALT furthermore emphasizes the extent to which levels of TDS exceed secondary MCL for chloride levels at the Site. Though the preamble to the NCP generally references “high salinity,” it directs one to “EPA Guidelines for Ground-Water Classification” for guidance in determining the correct classification of groundwater, and that document specifically states that Class III groundwater includes groundwater with a TDS concentration of over 10,-000 mg/1. The court thus deems that particular concentration-10,000 mg/l-to be the benchmark for determining high salinity.

The City collected over 1,000 samples from within the Site and determined that none exceeded 10,000 mg/1 for TDS. Few even exceeded 1,000 mg/1 and the average TDS level was less than 1,000 mg/1. At trial, Clegg, ALT’s witness, relied on a study produced by Lane and Miller that reviewed sampling of groundwater in the Wichita area, including sampling at the Site, and testified that TDS concentrations were found within a range of 450 to 11,150 mg/1. Clegg later admitted, however, that the samples collected pursuant to the Lane and Miller study included only one sample in excess of 10,000 mg/1 for TDS, and that this particular sample was collected not necessarily within the Site but somewhere within Wichita. The court finds city-wide TDS levels and testimony referencing those levels highly irrelevant to a determination of whether TDS levels at the Site itself exceed 10,000 mg/1. The court thus disregards much of Clegg’s testimony on this issue and, based on the evidence presented by the City, finds that TDS levels at the Site do not exceed the 10,000 mg/1 benchmark. 41

ALT further contends that the Site’s groundwater should be characterized as Class III groundwater due to the effects of broad scale human activity. To that end, ALT has identified hundreds of potential sources of groundwater contaminants, including numerous businesses that received liability releases from the city, operations located within the Northern Industrial Corridor, and the City’s own sewer system. The RI itself includes a statement that “[g]roundwater contamination in the form of chlorinated solvents and petroleum hydrocarbons has affected a large portion of the Gilbert-Mosley Site” and also outlines multiple types of contaminants throughout Wichita.

But given the existence of these contaminant sources and that various contaminants have reached the groundwater at the Site, ALT still fails to offer any evidence that this contamination cannot be “cleaned up using treatment methods reasonably employed in public water-supply systems,” as specified in “EPA Guidelines for Ground-Water Classification.” 42 In fact, the EPA’s “Guidelines for Ground-Water Classification Under the EPA Ground-Water Protection Strategy” identifies “Air Stripping,” the central means of treating groundwater in the pump and treat system, as a method “in common use in public water-treatment systems.” Accordingly, ALT’s argument and evidence regarding the classification of the groundwater at the Site is not persuasive and is not sufficient to rebut the presumption of consistency afforded the City.

ALT finally argues that the City did not attain community acceptance for its response actions. “The public comment/community relations requirements for remedial actions include interviewing interested parties, including local officials and community residents, developing a community relations plan, publishing a brief analysis of the remediation plan in a major local newspaper, and offering a public comment period following publication of the planned remediation that includes the opportunity for a public meeting.” Carson Harbor, 287 F.Supp.2d at 1160. While ALT generally claims that the City ignored certain public comment, never de fined the risks associated with air strippers, and issued an ESD rather than a revised CAD to circumvent the public comment requirements pertaining to the latter, it has failed to rebut the presumption that the City’s community relations efforts were consistent with the NCP.

The City did in fact produce and obtain the KDHE’s approval of a Community Relations Plan. The KDHE furthermore held a 30-day public comment period and public hearing regarding the draft CAD, whereby it sufficiently informed the public of the proposed remedy and included comments received during the period and the KDHE’s responses in the final CAD. 43 ALT can point to no specific departure by the City from the NCP’s public relations requirements, and neither can the court. See 40 C.F.R. § 300.430(c), (f)(3). The court thus rules that the City’s Community Relations Plan and its community relations efforts were not inconsistent with the NCP.

6. Conclusion

When evaluating whether the City’s response action is consistent with the NCP, the court must view the action “as a whole” and must examine it in light of the NCP’s “substantial compliance” standard. See 40 C.F.R. § 300.700(c)(3)(i) (“A private party response action will be considered ‘consistent with the NCP’ if the action, when evaluated as a whole, is in substantial compliance with the applicable [NCP] requirements ... and results in a CERC-LA-quality cleanup ... ”). And at the threshold of the court’s analysis, the City enjoys a presumption that its response action was in fact consistent with the NCP. Though defendants raised numerous issues, the evidence presented falls far short of that needed to rebut the presumption of NCP consistency due the City. To the contrary, when viewed as a whole, the evidence is more than sufficient to show that the City has substantially complied with the NCP, and as such is entitled to all costs necessary to its response. 44

E. Necessity of Response Costs

Finally, a prima facie case of CERCLA liability pursuant to § 113 requires the City to show that its costs are necessary costs of response. See Morrison, 302 F.3d at 1135. This element is derived from CERCLA § 107(a)(4)(B), which provides for recovery of “any other necessary costs of response incurred by any [private party] consistent with the national contingency plan.” 45 42 U.S.C. § 9607(a)(4)(B). Inter preting the phrase “necessary costs of response,” however, is problematic. See Daigle v. Shell Oil Co., 972 F.2d 1527, 1533 (10th Cir.1992) (“In keeping with its notorious lack of clarity, CERCLA leads us down a convoluted path to the definition of ‘any other necessary costs of response.’ ”).

While a strictly grammatical reading of the statute suggests that costs must be evaluated for consistency with the NCP, the Tenth Circuit has generally considered whether the response action was consistent with the NCP, and then whether the costs were necessary to complete the response action. See Bancamerica Commercial Corp. v. Mosher Steel of Kan., 100 F.3d 792, 796 (10th Cir.1996) (“[T]o obtain contribution ... [plaintiff’s] response actions must have been ‘consistent with the national contingency plan.’ ”) (emphasis added); United States v. Hardage, 982 F.2d 1436, 1443 (10th Cir.1992) (“Costs, by themselves, cannot be inconsistent with the NCP. Only response actions — i.e., removal or remedial actions — can be inconsistent with the NCP”); 46 see also 40 C.F.R. § 300.700(c)(3)-(4) (consistently discussing whether a “response action” is consistent with the NCP, and not whether particular costs are consistent with the NCP). Having already concluded that the City’s response action has thus far been consistent with the NCP, the court will now proceed to determine whether the City’s costs were necessary to fulfill the response action.

Necessary costs are those costs that are “necessary to the containment and cleanup of hazardous releases.” Hardage, 982 F.2d at 1448. Necessary costs include not only the cost of actual cleanup, but also include costs for investigation, planning, and remedial design. See Bancamerica Commercial Corp. v. Trinity Indus., 900 F.Supp. 1427, 1460 (D.Kan.1995), aff'd in part and rev’d in part on other grounds, 100 F.3d 792 (10th Cir.1996). “[C]ourts will deny recovery where the costs incurred were duplicative of other costs, wasteful, or otherwise unnecessary to address the hazardous substances at issue.” 47 Waste Mgmt. of Alameda County, Inc. v. East Bay Reg’l Park Dist., 135 F.Supp.2d 1071, 1099 (N.D.Cal.2001) (citing Lansford-Coaldale Joint Water Auth. v. Tonolli Corp., 4 F.3d 1209, 1219 (3rd. Cir.1993); United States v. Iron Mountain Mines, Inc., 987 F.Supp. 1263, 1272 (E.D.Cal.1997); Northwestern Mut. Life Ins. Co. v. Atl. Research Corp., 847 F.Supp. 389, 401 (E.D.Va.1994); Cent. Me. Power Co. v. F.J. O’Connor Co., 838 F.Supp. 641, 648-49 (D.Me.1993)). Nonetheless, mere redundancy of limited activities will not automatically render a cost unrecoverable. See Trinity Indus., 900 F.Supp. at 1460.

1. CDM Costs

The lion’s share of costs incurred by the City went to the environmental consulting firm of CDM. This is because the City essentially hired CDM to perform a turnkey job for the Site cleanup. CDM provided its own environmental consulting and engineering expertise. CDM also hired subcontractors and purchased equipment on behalf of the City. While some subcontractors were hired and paid directly by the City, most of the work was handled through CDM. Overall, CDM charged the City $10,254,210.36 as of June 15, 2002.

As evidence of these costs, the City points to CDM’s periodic invoices. ALT attacks these charges as being insufficiently documented. On the contrary, each invoice includes simple descriptions of the task being completed, as well as some indication of whether the charges are for labor or other direct expense. City expert Dennis Faulkner testified that all the money paid to CDM was properly supported by invoices and related documents. 48 Moreover, City Environmental Director Jack Brown testified that all the invoices were directed to him, and that he assigned his subordinates to perform a detailed review of each invoice to confirm that the City was being billed properly. One of Brown’s subordinates, Brian Fisher, also testified at trial. His testimony revealed that he worked extensively with CDM personnel doing field work within the Site. The evidence showed that Fisher was also called upon to review a large portion of the CDM bills. Defendants failed to ask Fisher anything substantive regarding how he reviewed the CDM invoices. Accordingly, Brown’s testimony that a review was performed is accepted as true.

The EPA has provided the following guidance on documenting response actions in order to support cost recovery:

During all phases of response, the lead agency shall complete and maintain documentation to support all actions taken under the NCP and to form the basis for cost recovery. In general, documentation shall be sufficient to provide the source and circumstances of the release, the identity of responsible parties, the response action taken, accurate account ing of federal, state, or private party-costs incurred for response actions, and impacts and potential impacts to the public health and welfare and the environment.

40 C.F.R. § 300.160(a)(1). While the language of this section indicates that it applies to government cost recovery actions, 40 C.F.R. § 300.700(c)(5)(ii) states that this provision may also be applicable to private party actions, as is the case here.

In United States v. W.R. Grace & Co.-Conn., 280 F.Supp.2d 1149 (D.Mont.2003), the district court conducted an extremely thorough examination of CERCLA’s documentation requirements in support of a cost recovery action. See id. at 1179-83. In evaluating 40 C.F.R. § 300.160(a)(1), Grace stated “[tjhis provision does not establish prescriptive standards for the content of cost documents,” and that the NCP “does not contain any specific standards concerning the documentation of costs.” Grace, 280 F.Supp.2d at 1179-80 (quotations omitted). Grace concluded that courts had consistently required that costs be proven by a preponderance of the evidence, and that detailed cost summaries, vendor invoices, payment vouchers, and contractor bills provided sufficient proof. The court finds the analysis in Grace to be persuasive, and need not repeat it here. CDM’s bills are documented on numerous multi-page invoices that sufficiently describe the work being billed. Jack Brown’s testimony shows that the bills were reviewed by City personnel for accuracy and consistency with the underlying contract. Defendants had ample opportunity in discovery and at trial to show that CDM’s bills were inappropriate, unnecessary, or outright fraudulent; however, no such proof was offered. The court finds that CDM’s costs were sufficiently documented and are necessary costs of response at the Site.

In reviewing the invoices that support CDM’s charges, the court finds that $218,591.79 was for site work at the City’s Bus Barn over the years 1997-1999. The Bus Barn project was essentially source control work to remove the contaminated soils found there. Upon further review, the court notes that the City has already deducted $93,794 of these charges, classifying them as unrecoverable remediation for petroleum contaminated soils. That leaves $124,797.79 of the $218,591.79 still unaccounted for. Thus, of the $10,254,210.36 in CDM charges that the court finds were necessary costs, $124,797.79 was for non-petroleum related source control work at the City’s Bus Barn, and the remaining $10,129,412.57 was for other response work at the Site.

2. KDHE Costs

The City also paid the KDHE $277,648.65. These costs reflect charges for KDHE oversight and participation in the Gilbert and Mosley project, as well as various laboratory and analytical charges incurred by the agency. The KDHE’s charges are documented in numerous invoices dating back to 1991. Most of the invoices provide summaries of the costs incurred, including a breakdown of labor charges and analytical work. Although these invoices are not as detailed as the CDM bills, the court finds that the KDHE invoices are sufficient to support the City’s claimed costs. Faulkner’s review indicates that the City’s payments to the KDHE are properly supported by invoices and records of payment. Defendants provided no evidence that any of the charges were false or unsupportable.

Costs for government oversight of private-party remedial actions are recoverable under CERCLA § 113(f). See Atl. Richfield Co. v. Am. Airlines, Inc., 98 F.3d 564, 569-70 (10th Cir.1996). The record is replete with evidence that the KDHE’s role was to provide oversight for the City’s work at the Site. Accordingly, the KDHE charges are necessary costs of response at the Site, and are fully recoverable.

3. Miscellaneous Vendor Costs

Next, the City paid $350,362.11 to various vendors which billed the City directly. This total is comprised of $46,105.95 from Continental Analytical Services for laboratory work analyzing samples from the Site; $56,013.91 from Cornejo & Sons, Inc. for what appears to be construction-type work; $32,213.93 from Geoprobe Systems for geoprobing tools; $179,091.54 from Haz-Mat Response, Inc. for source control work at the City’s Bus Barn; and $36,936.78 from other vendors who each received less than $5,000. The court will address each of these costs in turn.

First, the City seeks recovery of $46,105.95 from Continental Analytical Services. These costs are adequately documented by invoices and log sheets showing the locations from which samples were taken. Faulkner’s testimony further supports that these costs are adequately documented. The court also finds that these costs are necessary costs of response in that they provided laboratory measurements of the amount of contamination in soils and groundwater at the Site. However, of these charges, $34,303.60 is for source control measures at the City’s Bus Barn. That leaves $11,802.35 of Continental’s charges that are related to other aspects of the Gilbert and Mosley project.

Next, the City claims $56,013.91 in costs from Cornejo & Sons, Inc. All these costs were paid in 2002, and are reflected in plaintiffs exhibit P-2492. After a mind-numbing review of the 1,384 pages of documents that comprise P-2492, the court fails to find a sufficiently documented connection between these costs and the Gilbert and Mosley project. Cornejo appears to have done a great deal of work for the City. Indeed, several of the documents in P-2492 show that Cornejo billed the City over $100,000 for its work in 2001 and 2002. The problem arises in determining how much of that work amounts to response costs. Cornejo performs, among other things, construction work. The types of services provided by Cornejo may be useful to any number of City projects. Unfortunately, the documents that appear to be Cornejo invoices indicate nothing more than that work was done for the health department. There is no mention of Gilbert and Mosley on any of the bills. Thus, the court has no means to determine that certain charges are for work at the Site, while others are not. Trial testimony was equally unenlightening. A review of the trial transcript shows that the word Cornejo was spoken only once at trial, and it had nothing to do with these charges. Apparently, Faulkner was able to make the connection between the costs alleged here and the Gilbert and Mosley project, although he did not explain how he did so. The court cannot. Plaintiff has failed to meet its burden of proving that these were response costs related to the Site; therefore, no recovery will be permitted for the Cornejo & Sons, Inc. charges.

The City also claims $32,213.93 in costs from Geoprobe Systems for the purchase of a geoprobe and supporting equipment. After reviewing the invoices and related paperwork, along with Faulkner’s testimony, the court finds that these expenses are adequately documented. The geoprobe was clearly instrumental in source identification and sampling within the Site, and is therefore a necessary cost of response. On the other hand, the court notes that tools like this are useful at other contamination sites. In fact, trial testimony showed that the City’s geoprobe was also used to investigate another site with contaminated groundwater, the Northern In dustrial Corridor. Likewise, such a tool may be useful at other locations in the city. Accordingly, the court finds that approximately one-third of the costs associated with the geoprobe should be assigned to the Site, and will therefore recognize $11,000 in payments to Geoprobe Systems as costs eligible for recovery in this case.

Next, the City seeks contribution for $179,091.54 in costs from Haz-Mat Response, Inc. This charge is supported by invoices and minutes from a Wichita City Council meeting. A review of the documentation supporting this cost shows that it is for removal of contaminated soils at the City’s Bus Barn — in other words, this work is for source control at the Bus Barn. Since the Bus Barn is a source of contamination at the Site, this cost is a necessary cost of response.

Finally, the City seeks recovery based on $36,936.78 in costs from other vendors, which each received less than $5,000. Due to the relatively small charges and large numbers of vendors in this category, the City did not present these charges on a vendor-by-vendor basis. Nonetheless, the City submitted 4,630 pages of documents supporting its costs, which were replete with invoices, receipts, and supporting documentation for these charges. Combining this evidence with Faulkner’s testimony, the court concludes that these charges are adequately documented, and relate to Site response actions. They are therefore necessary costs of response, and will be recoverable costs in this action.

In sum, the City claims $350,362.11 in response costs from various vendors. Of that, $213,395.14 is potentially recoverable as response costs incurred for source control measures at the City’s Bus Barn, and $59,739.13 is potentially recoverable as response costs for other Site activities.

4. Stinson, Mag & Fizzell Costs

The City paid $205,047.78 to the law firm of Stinson, Mag & Fizzell for work related to the Gilbert and Mosley project. The invoices for those charges indicate that all the work billed was for PRP identification. Although litigation costs related to the City’s lawsuit would not normally be recoverable under CERCLA, attorney’s fees related to PRP identification are recoverable. Key Tronic, 511 U.S. at 819-20, 114 S.Ct. at 1967. Faulkner testified that many other charges from Stinson were excluded from the response costs the City seeks here because they are litigation costs. The court finds that these costs are sufficiently documented and are necessary response costs.

5. City Payroll Costs

Next, the City seeks $738,611.56 as part of its payroll burden for City employees who worked on the Site. ALT attacks these costs on the grounds that most of these people were already employed by the City, and their salaries would have been necessary without regard to the groundwater pollution problem. Despite this fact, any salary and benefits paid to these employees for work they did on the Gilbert and Mosley project was money spent in accomplishing the response action. ALT’s argument would allow the City to recover payroll burden for any new-hires, as well as for any costs of outsourcing the work, but would bar recovery if the City used existing workers. ALT has offered no legal authority to support its argument.

The real problem here is that no one kept any sort of contemporaneous records regarding how much time each employee devoted to Gilbert and Mosley response actions. Jack Brown, Brian Fisher, and Dennis Faulkner all testified that this payroll allocation was first developed in 2000 by reviewing project time lines and similar documentation and estimating from those records approximately how much time each employee devoted to the project each year. Thus, the time spent by each employee in 1991 was not estimated until 2000, over nine years later! As a former practicing lawyer who primarily billed by the hour, the court is aware how difficult it can be to make accurate estimates of time spent on particular tasks a day, or even a week, after the fact. To attempt to do so with any accuracy almost a decade later is utterly ludicrous. To make matters worse, even after Faulkner joined the project and made the City aware of its time-keeping problems, Brown and Fisher continued to use this same year-end estimation approach to allocating the 2000, 2001 and 2002 payroll burden for City employees who worked on the project.

In order to recover response costs under CERCLA, a plaintiff must maintain an “accurate accounting” of its costs. 40 C.F.R. § 300.160(a)(1). Courts that have considered recovery of payroll burdens or similar labor costs have frequently relied on the contemporaneous nature of time records to support such costs. See, e.g., United States v. Chapman, 146 F.3d 1166, 1171 (9th Cir.1998); California ex rel. Calif. Dept. of Toxic Servs. v. Neville Chem. Co., 213 F.Supp.2d 1134, 1138-40 (C.D.Cal.2002). Here, the City has failed to meet its burden of keeping an accurate accounting of these payroll costs. Estimating the number of hours worked by employees on the project months, years, and almost a decade after the fact is simply too speculative to be considered accurate. Accordingly, none of the City’s past payroll costs are eligible for recovery.

6. Motor Pool Expenses

Last of all, the City seeks to recover a portion of the $22,089.38 it spent for vehicle use related to the Gilbert and Mosley project. These costs are documented by quarterly inter-departmental transfer forms whereby the City’s motor pool bills other departments and projects for the use of the City’s vehicles. Faulkner testified that these costs were for use of the City’s geoprobe van on the project. The court finds that these cost are sufficiently documented and that they are necessary costs of response.

7. Summary of Recoverable Costs

Overall, the City claims that it actually spent $11,847,969.84 in response costs for the Gilbert and Mosley project. The court finds that $11,032,130.44 is properly supported, necessary costs of response. Of that figure, $338,192.93 was incurred for source control measures at the City’s Bus Barn, and $10,693,937.51 was incurred for response actions at other areas within the Site.

F. Arguments Raised in Defense

ALT asserts the equitable doctrines of laches, unclean hands and failure to mitigate, and claims that the City is consequently barred from prevailing in this matter. 49 But “[s]ection 9607(b) provides very limited defenses to liability,” and the court is unwilling, at least in this case, to expand the number of potential defenses to liability to include those asserted by ALT. Morrison, 302 F.3d at 1132; see also County Line, 933 F.2d at 1518 n. 15 (noting that CERCLA liability is “subject only to a few, very limited defenses”). The court has considered ALT’S conten tions and finds them to be without merit, at least insofar as ALT claims they completely bar liability. The court does, however, consider the relevant facts underlying these equitable doctrines, along with other equitable factors, when allocating response costs proportionally among defendants. 50

IV. ALLOCATION OF RESPONSE COSTS

Having found that the City has proven liability, “the court may allocate response costs among liable parties using such equitable factors as the court determines are appropriate.” 42 U.S.C. § 9613(f)(1). The City asks that essentially all of its unrecov-ered past response costs be allocated to the defendants, with ALT and Reid also bearing costs related to so-called orphan shares. 51 Furthermore, the City seeks to hold defendants liable for a sizable portion of future response costs. By contrast, defendants generally claim that they should have no liability; but, if they are found liable, they should bear little or none of the orphan shares.

Section 113 provides little guidance on how costs should be shared, stating only that “the court may allocate response costs among liable parties using such equitable factors as the court determines are appropriate.” 42 U.S.C. § 9613(f)(1). This broad language grants the court considerable discretion in allocating response costs between PRP’s. Bancamerica, 100 F.3d at 802. In making an equitable apportionment of response costs under CERCLA § 113(f)(1), “a court may consider several factors, a few factors, or only one determining factor.” Id. (quoting United States v. Colo. & E. R.R., 50 F.3d 1530, 1536 (10th Cir.1995)). Ultimately, “[t]he court must balance the equities in light of the totality of the circumstances.” FMC Corp., 998 F.2d at 847.

The court begins its analysis from the fundamental equitable position that individuals ought to take responsibility for their own actions. From an environmental perspective and when framed in a most simplistic manner, this suggests people should clean up after themselves. In the CERCLA § 113(f)(1) context, this often means that a responsible party should bear a share of the response costs that is in some way proportional to the extent of pollution that party has caused. See, e.g., Boeing Co. v. Cascade Corp., 920 F.Supp. 1121, 1136 (D.Or.1996) (finding that starting point for allocation was proportional to each party’s mass contribution of pollutants); Trinity Indus., 900 F.Supp. at 1473-74 (finding that starting point for allocation was proportional to the volume of lead pollution contributed by each party). This proportionality analysis is usually intended to identify a starting point for allocation based on some physical measurement of pollution at the site, such as mass contributed, Boeing Co., 920 F.Supp. at 1136, or volume contributed, Trinity Indus., 900 F.Supp. at 1473-74. The responsible parties are then assigned an initial percentage of response costs based on their relative contributions to the selected measurement standard.

While most courts begin the allocation process at proportionality, they often consider other mitigating and aggravating fac tors to determine whether an adjustment should be made to the proportional allocation. In this case, an upward adjustment would cause defendants to bear a portion of the orphan shares, while a downward adjustment would essentially cause the City to retain liability for some of the pollution caused by defendants.

A. Equitable Factors 52

1. Gore Factors

The congressionally-rejeeted “Gore Factors” often provide a convenient starting place for this second level of allocation analysis. 53 See Colo. & E. R.R., 50 F.3d at 1536 n. 5. The six Gore Factors are: (1) the ability of the parties to demonstrate that their contribution to a discharge, release or disposal of a hazardous waste can be distinguished; (2) the amount of the hazardous waste involved; (3) the degree of toxicity of the hazardous waste involved; (4) the degree of involvement by the parties in the generation, transportation, treatment, storage, or disposal of the hazardous waste; (5) the degree of care exercised by the parties with respect to the hazardous waste concerned, taking into account the characteristics of such hazardous waste; and (6) the degree of cooperation by the parties with the Federal, State or local officials to prevent any harm to the public health or the environment. Id.

The first three Gore Factors need not be analyzed here. The first and second factors are directly related to the amount of pollution caused by defendants, and are already accounted for by the court’s decision to make its initial allocations proportional to defendants’ contribution of pollution to the Site. The third Gore factor is inapplicable because the parties have not raised toxicity as a basis for apportioning costs.

The fourth factor, the parties’ degree of involvement in causing the pollution, is generally used to reduce a defendant’s liability because that defendant was only remotely involved in causing the release. See, e.g., Pneumo Abex Corp. v. Bessemer and Lake Erie R. Co., 936 F.Supp. 1250, 1269 (E.D.Va.1996); Am. Color & Chem. Corp. v. Tenneco Polymers, Inc., 918 F.Supp. 945, 959 (D.S.C.1995); Cent. Me. Power, 838 F.Supp. at 646 (concluding that, where both parties were “fully involved” in causing the pollution, degree of involvement would not shift response costs from one party to the other). Similarly, the fourth factor might reduce a defendant’s liability because the plaintiff was also involved in the activities that resulted in the release. Here, all defendants were owners or operators (or in ALT’s case, APCO’s successor-in-interest) of the facilities where releases occurred. The evidence shows that the most likely causes of releases on defendants’ properties were defendants themselves, their employees, or others with whom defendants conducted business. Conversely, no one has suggested that the City was responsible for releases on defendants’ land. 54 Since defendants must naturally bear a large share of responsibility for business activities conducted on their property, the court finds that the fourth Gore Factor does not favor a reduction in defendants’ share of the response costs.

The fifth Gore Factor suggests that the court consider “the degree of care exercised by the parties with respect to the hazardous waste concerned, taking into account the characteristics of such hazardous waste.” Colo. & E. R.R., 50 F.3d at 1536 n. 5. The City, ALT, and Reid presented evidence at trial on this factor. 55 The City maintains that Reid’s operating practices demonstrated a lack of care in handling and storing chlorinated solvents. Former Reid employees testified that relatively small amounts of chlorinated solvents were spilled when drums were filled and handled at the facility. The City also concludes that the mere fact that spills occurred at the APCO and Tri-Supply sites is sufficient to demonstrate a lack of due care. ALT attempts to rebut the City’s claims by showing that it had a policy and practice of cleaning up chlorinated solvent spills. Unfortunately for ALT, their star witness on this point, Riehwine, was not credible.

In assessing this factor, the court is mindful that most of the activity at issue in this case occurred in the 1960’s and 1970’s — a time when it was neither unlawful nor uncommon to dump chlorinated solvents on the ground. See Boeing Co., 920 F.Supp. at 1137; see also Comprehensive Environmental Response, Compensation, and Liability Act of 1980, Pub.L. No. 96-510 (showing CERCLA not enacted until 1980); Resource Conservation and Recovery Act of 1976, Pub.L. No. 94-580 (showing RCRA not enacted until 1976). In all likelihood, the employees at Reid and APCO probably were spilling relatively small amounts of chlorinated solvents without giving it a thought. But that was a different era, and while the court is prepared to hold defendants responsible for their own actions, there is no reason to consider this conduct as an aggravating factor with respect to ALT and Reid. 56

Turning next to Tri-Supply, the City maintains that a single two-gallon spill of PCE at that site demonstrates lack of due care. Unlike Reid and APCO, Tri-Supply operated during the 1990’s, a time when environmental awareness was expected from businesses. That observation notwithstanding, evidence of a single spill does not support a conclusion that TriSupply failed to use due care in its operations. Accordingly, this factor does not provide a basis to increase Tri-Supply’s allocation.

The last of the Gore Factors to be considered is “the degree of cooperation by the parties with the Federal, State or local officials to prevent any harm to the public health or the environment.” Colo. & E. R.R., 50 F.3d at 1536 n. 5. This factor only appears relevant to ALT, since the City failed to raise it regarding the other defendants. The City contends that ALT failed to cooperate with the KDHE and, in fact, provided false information to the KDHE regarding APCO’s use of chlorinated solvents at the Site.

This dispute largely focuses on ALT’s responses to a KDHE inquiry regarding use of chlorinated solvents at APCO’s Wichita facility. The KDHE asked questions regarding historical operations at the site, use of chlorinated solvents, details of any chlorinated solvent releases, and contact information for other persons who might have such knowledge. The responses were provided by a former APCO accountant who served as a records custodian for the Trust. Most of ALT’s answers simply stated a lack of knowledge or information sufficient to answer the questions. Based on information the City obtained through discovery, the City asserts that ALT did have information available to answer the questions, but purposely failed to do so. This argument is further intertwined with a motion for sanctions filed by the City, which seeks to impose discovery sanctions on ALT, along with a request that the court consider ALT’s alleged conduct under the sixth Gore Factor.

After reviewing the allegations and explanations surrounding the alleged conduct, the court concludes these issues are fact-intensive, highly disputed, and were not fully developed at trial. Accordingly, they will not be decided here. Instead, the court will take up those matters separately should it be necessary to decide the City’s motion for sanctions before entry of final judgment. 57 Without deciding the matter, the court notes that at the time of the KDHE inquiry, APCO 'had ceased to exist as a going concern for over 20 years. Indeed, APCO’s Wichita facility had been closed for over 25 years. Furthermore, the record indicates that a substantial portion of APCO’s operating records had been legitimately destroyed in the 1980s.. Thus, it is conceivable that ALT’s response to the KDHE inquiry was made in good faith. Therefore, the court will not consider ALT’s disputed conduct as an equitable factor in this decision.

2. Other Equitable Factors

While the Gore Factors provide only limited value in deciding this case, other circumstances exist that are more hélpful in determining the allocation of response costs. These factors include 1) economic benefits received by the City as a result of its cleanup actions; 2) ALT’s efforts to conduct source control at its facility; 3) windfalls from settlements the City made with other PRPs; and, 4) the City’s decision to release some PRP’s from Lability. Each merits discussion.

i. The City’s Economic Benefits Due to the Cleanup

As a result of the pollution underlying the Site, the City was faced with a substantial erosion of its property tax base. By the City’s own estimates, approximately 18% of this tax base was located within the Site. Due to the uncertain nature of liability associated with the Site, banks were unwilling to provide financing for real estate transactions on affected property. Thus, real estate in parts of the downtown area was becoming unmarketable. Accordingly, some property owners began to seek devaluations in their real estate tax appraisals. These circumstances helped convince the City that a proactive approach was needed in order to avert the impending fiscal calamity.

In furtherance of that goal, the City entered into the previously described agreement with the KDHE whereby the City agreed to assume financial responsibility for investigating and remediating the Site, subject to oversight by the KDHE. That agreement was instrumental in persuading financial institutions to resume financing real estate transactions within the Site, thereby restoring marketability of affected properties and shoring up a substantial part of 'the City’s tax base. Accordingly, the City realized an economic benefit through 'its assumption of responsibility for the cleanup. Although none of the parties attempted to quantify the dollar value of the affected tax base, the court notes that for a city the size of Wichita, 18% of the property tax base considered over many years must be substantial.

In determining allocation of response costs under CERCLA § 113(f)(1), courts have often considered the economic benefits realized by a party as a - result of remediation efforts. See, e.g., Dent v. Beazer Materials and Servs., 993 F.Supp. 923, 951, n. 30 (D.S.C.1995), aff'd, 156 F.3d 523 (4th Cir.1998); Farmland Indus. v. Colo. & E. R.R., 944 F.Supp. 1492, 1500-01 (D.Colo.1996); Hatco Corp. v. W.R. Grace & Co.-Conn., 836 F.Supp. 1049, 1091 (D.N.J.1993). In Farmland Industries, the district court allocated 85% of the response costs to Colorado & Eastern, based in large part on the observation that Colorado & Eastern’s property value had increased by over $600,000 as a result of the cleanup. See Farmland Indus., 944 F.Supp. at 1500-01. Thus, parties that benefitted financially from remediation are often allocated a larger portion of the response costs.

Like Colorado & Eastern, the City has realized a sizable economic benefit through the restoration of the tax base within the Site. Unlike Colorado & Eastern, however, the City also suffered the previous decline in its tax base as a result of pollution at the Site. Considering these facts collectively, the benefit received by the City is simply a restoration of the status quo. Whatever benefit the City has received (and will receive) is merely an offset to the damage it suffered as a result of the Gilbert and Mosley pollution problem. Despite this fact, the court is mindful that these defendants were not the sole cause of the problem. Had they never spilled a drop of chlorinated solvent, the City would still have faced the same economic problems due to the actions of other polluters who are not parties to this suit. Thus, the court finds that in balancing the equities on this factor, defendants should still be held accountable for their proportionate contribution to the problem, but should not be allocated an additional portion of the orphan shares.

ii. ALT’s Efforts to Conduct Source Control at Its Facility

On April 24, 2001, the KDHE sent letters to ALT and Reid requesting that they enter into consent orders to conduct source control measures at their respective sites. Within two months, ALT responded with an initial proposal for conducting source control at the APCO facility. The trustees initially offered to remove the contaminated soil and perform post-removal testing to ensure that chlorinated solvent levels in the soil had been reduced to a level acceptable to the KDHE. The trustees also desired to perform their own groundwater remediation efforts. In exchange for their work, the trustees sought some form of contribution protection from the KDHE. The KDHE rejected that proposal, but negotiations continued for almost a year. The KDHE concluded that it would not be appropriate to allow ALT to perform groundwater remediation, since ALT’s plan might interfere with the City’s efforts to clean up all the affected groundwater at the Site. By contrast, the KDHE did not appear to rule out source control, since soil removal at individual sites would not have a significant impact on the groundwater cleanup.

Ultimately, the KDHE identified three deficiencies in the ALT proposal that had to be addressed. In a letter dated February 19, 2002, the trustees appear to have capitulated on all three issues, and invited the KDHE to continue working toward a solution that would permit ALT to conduct source control measures at its facility. The KDHE never responded to that letter. Indeed, according to KDHE attorney Erika Bessey, she barely read past the first page of the letter, after which she “got real annoyed and threw it in a pile,” never bothering to look at it again. Moreover, Bessey noted that even if she had read the letter, she would not have been willing to reach an agreement with ALT — an agreement which she had initially invited, even though ALT was agreeable to everything the KDHE had previously requested. The court finds Bessey’s conduct was counterproductive to the KDHE’s obligations to oversee environmental protection in Kansas.

ALT contends that the KDHE refused to work toward a source control agreement because the City pressured the agency to avoid any agreements that might jeopardize the City’s ability to recover from ALT. Indeed, the KDHE’s letter to ALT requesting that the trustees undertake source control at the APCO facility shows that the KDHE felt source control was certainly necessary at the APCO site. Moreover, CDM employee Olsen testified that source control was recommended for the APCO site, and even outlined some of the options that he was considering for the job. Thus, although all sides appeared to agree that source control was needed at APCO, the KDHE was unwilling to pursue a prompt means of accomplishing the cleanup. Hence, according to the City’s expert, APCO remains a continuing source of chlorinated solvents entering groundwater within the Site.

Unfortunately for ALT, it has failed to provide sufficient evidence to show that there was some sort of collusion between the City and the KDHE to prevent ALT from performing source control. The City did object to ALT conducting its own groundwater remediation; however, that was justified on the grounds that it might interfere with the overall groundwater remediation efforts that the City was planning. Similarly, the City also objected to the KDHE’s entering any agreement that would provide contribution protection to defendants in this lawsuit; but, that does not amount to collusion or coercion. The City simply wanted to avoid any deals that would prevent it from recovering legitimate costs. Presumably, if ALT performed source control, the City would have no future costs related to source control at the APCO site. Hence, there would be no costs to recover. Thus, the City’s requests were reasonable to the extent it did not want the KDHE to exonerate defendants for costs that the City would ultimately have to pay.

Although there is insufficient evidence to show that the City actively interfered with ALT’s efforts to make source control arrangements with the KDHE, there is evidence that shows the City’s own dilatory conduct with regard to source control may have exacerbated the groundwater pollution problem. As early as 1996, the KDHE was harping on the City to pursue source control; yet, to this day the only source control efforts undertaken at the Site were conducted at one of the Coleman facilities and the City’s Bus Barn. Indeed, the KDHE identified the City’s deficiencies with regard to source control in a letter dated August 2, 1996, wherein KDHE project manager Christine Jump said

KDHE and the City have related concerns regarding source control; however, the primary objective of the City is cost recovery from PRPs, while KDHE’s primary objective is the investigation and implementation of a remedial action. ... Since [other activities] will delay the original schedule for implementation of a complete remedial system for the Gilbert and Mosley Site, KDHE feels it is even more critical now to implement source control activities in a timely manner.

(Emphasis added). Both of the KDHE’s project managers at the Site, Bean and Jump, testified about the City’s slow progress regarding source control. Furthermore, the City has characterized APCO and Reid as two of the “largest sources of CVOC groundwater contamination at the Site,” yet no effort to control the sources at those sites has been performed. Thus, as a result of the City’s delays at source control, chlorinated solvents have had an additional seven years to enter the aquifer and aggravate the groundwater contamination problem. ALT offered to perform source control, and was turned down. The City was asked to do it, but has failed to comply.

Overall, the court finds that this combination of circumstances should be considered in allocating response costs to defendants, particularly ALT. Moreover, there is something to be said for allowing the party who pays the bill to be involved in selecting the solution, especially where the KDHE has already invited it to do so. Here, the City appears to want to dictate source control measures at the APCO site, while passing the costs on to the trustees. ALT offered to perform its own source control measures while still remaining accountable to the KDHE for doing a satisfactory job. In doing so, ALT would have been able to better control its own costs while still being obligated to perform the cleanup according to KDHE standards. Although ALT could be precluded from performing a separate groundwater cleanup, as that might significantly interfere with the City’s plan, the court is very troubled by the KDHE’s irresponsible rejection of ALT’s source control proposals with no apparent effort to reach a solution. The court is further concerned that the City’s own delays in performing similar source control activities has allowed even more chlorinated solvents to leach into the groundwater. Accordingly, this factor weighs in favor of limiting all the defendants’ liability, particularly ALT.

iii. Windfalls from Settlements the City Made with Other PRPs

The City claims to have currently spent over $11.8 million for investigation and remediation of contamination at the Site. In seeking to recover these response costs, the City has negotiated settlements with twelve PRPs. All totaled, the City has recovered almost $9 million through these agreements. Thus, the City has managed to recover over 75% of its costs-to-date. Looking at it from another perspective, the City allocated a total of $5.5 million in past response costs to the settlors. Through its settlements, the City has recovered almost twice that amount. The court notes, however, that the settlements are intended to offset not only past costs, but also future costs of the project. Current estimates suggest that the total cost to remediate the Site may be upwards of $32 million.

The settlement figures make clear that the City has, on occasion, done very well for itself. For example, while the City’s own estimates show that the Wichita Eagle was only responsible for $196,002 in past costs, the City reaped $1.2 million from its settlement with the Eagle. Using the City’s figures of approximately $11 million spent out of a total of $32 million for the project, that leaves about two-thirds of the costs still remaining. Applying that ratio to the Wichita Eagle, if its past costs are $196,000, then it would be roughly responsible for $392,000 in future costs, for a total of $588,000 out of the entire $32 million in project costs. Thus, the Eagle paid $1.2 million when the City’s analysis says the Eagle would really only be responsible for $588,000, giving the City a windfall of $612,000 to cover orphan shares.

Similarly, the City recovered $225,000 from Vulcan Chemicals, another PRP at the Site. Unlike most of the other PRPs, however, Vulcan did not own a facility within the Site. Instead, Vulcan was a transporter who delivered chlorinated solvents to facilities, including Reid Supply. While there is certainly nothing wrong with the City’s recovery from Vulcan, the peculiar thing is that the City did not account for Vulcan’s contribution when it allocated response costs to defendants in this lawsuit. This has the effect of making Vulcan’s settlement appear as another windfall, thereby providing the City with an opportunity to recover from these defendants, and Reid in particular, an amount that the City has already recovered from Vulcan.

Finally, the City also received $200,000 in exchange for indemnifying a former owner of the City’s Bus Barn property against environmental liability. Like the Vulcan settlement, the City failed to include this figure in its accounting for past settlements, thereby providing yet another opportunity for double recovery through these defendants. Although the Bus Barn agreement was an indemnification for environmental liability, rather than a formal settlement agreement with a PRP, the court sees little substantive difference between the two under these particular facts.

In considering these facts, the court also notes that the City suffered shortfalls in some of its settlements. Nonetheless, the City was in control of its own destiny when it agreed to those numbers. Taking all these facts into consideration, the Court finds that this factor also weighs in favor of limiting defendants’ liability for any orphan shares.

iv. The City’s Decision to Release Some PRP’s from Liability

As part of its efforts to stabilize property values within the Site, the City initiated a Certificate and Release (CAR) program, whereby certain property owners were released from liability for the Site cleanup. As a result of the CAR program, the City has released over 3,500 property owners. In defense of that program, the City contends that it has performed over 5,000 samples in search of pollution within the Site, and none of those investigations suggest that any released person was a source of groundwater pollution.

By contrast, ALT contends that the City engaged in scant efforts to investigate many of the CAR applicants prior to issuing a release. While a review of the examples provided by ALT indicates that some of them appeared adequately investigated, other applicants indicated that solvents and similar pollutants were actively used on their property. ALT’s expert testified that he could find no evidence that the City had performed any sampling on many of these properties, and the City made no effort to rebut that claim.

These facts raise at least the possibility, and perhaps the probability, that somewhere among the 3,500 releases issued, the City has missed some PRPs who may have, or may be, contributing to the problems at the Site. The City initially took on the responsibility to remediate the Site when it entered into the cleanup agreement with the KDHE. The City compounded its financial risk when it decided to start issuing formal releases to potential polluters. Issuing those releases with only cursory investigations raises the risk even more. Accordingly, the court finds that this factor weighs in favor of limiting defendants’ liability for orphan shares.

3. Conclusions Regarding Equitable Factors

Considering the equitable factors discussed here, as well as the record as a whole, the court concludes that defendants should be allocated response costs in proportion to their individual contributions to the chlorinated solvent pollution within the Site. The court has considered and rejected the City’s arguments that, since it did not spill chlorinated solvents at the Site, it should not be allocated any response costs. Although the City cites cases where so-called “innocent” PRPs are allocated no response costs in a contribution action, those cases involved releases on one parcel, or perhaps a few small parcels of land. Indeed, if this were a case involving a few parcels of land where the City was completely innocent and all responsible parties had been identified, the court would be much more inclined to allocate little or no cost to the City. By contrast, this Site involves over 8,000 parcels of land, and the court is not remotely convinced that the City has identified all sources of chlorinated solvent releases. 58 The City voluntarily undertook this cleanup, and should have been aware of the Herculean effort required to identify PRPs. With so much lingering doubt regarding whether yet unidentified parties may have substantially contributed to the Gilbert and Mosley pollution problem, the court finds that the more equitable course is to hold defendants responsible for the pollution they caused. Accordingly, defendants will not be held responsible for all the orphan shares, or any other shortfalls which the City has suffered or may suffer in its efforts to recover response costs.

B. Methods for Allocating Response Costs

Having determined that defendants should be responsible for cleaning up the pollution they caused, the next obvious question is how can the court take that principle and turn it into a measure of actual dollars. Courts have considered various measures to allocate response costs. Some courts have considered the mass of pollutants contributed by individual parties. Boeing Co., 920 F.Supp. at 1137. Others have looked to the relative toxicity of contaminants. See generally United States v. Cantrell, 92 F.Supp.2d 718, 732 (S.D.Ohio 2000). Here, the City argues that the relative size of the contaminant plume originating from each source provides a fair basis on which to allocate investigation and preliminary design costs. ALT appears to dispute that point (as it does most things) but ultimately relies heavily on contaminant plume size to support a much smaller allocation to the trust. On the other hand, ALT rejects this method for allocating groundwater remediation costs, arguing instead that the trust should only pay a fraction of the capital and operating costs relating to the single extraction well that draws groundwater from APCO’s contaminant plume.

The court rejects ALT’s distinction. The relative size of a party’s contaminant plume is a fair and equitable measure of that party’s response costs under CERCLA § 113(f) for all response costs except facility-specific source control measures. The driving force behind groundwater remediation costs is the total volume of contaminated water that must be extracted from the aquifer. It does not matter whether any particular gallon of water is highly contaminated or only marginally polluted; if the water exceeds the ACL for any of the chlorinated solvents of interest, it must be dealt with. Since the aquifer underlying the Site has a relatively constant thickness, the areal extent of contaminant plumes provides a reasonable approximation of the relative contribution of chlorinated solvents from each source.

In order to estimate the size of each source’s plume, both ALT and the City engaged in computer modeling of the Site groundwater system. If properly used, computer models appear to be an invaluable tool in approximating the complexities of underground fluid flow. Without these models, the scientists and engineers would be limited to guessing at sources and fluid flow characteristics based on the limited number of wells that penetrate the aquifer. Through modeling, reservoir flow and contaminant transport can be calculated using complex mathematical operations that simulate the aquifer characteristics. From that effort, the model can simulate the progression of contaminant plumes from each source, thereby providing an estimate of the size of each plume at any given time. Unfortunately, there are no true crystal balls — the models are only as good as the data placed into them. In this ease, the data inputs and methods for configuring the models provided fertile ground for disagreement. Nonetheless, the court concludes that computer modeling of plume size is an appropriate basis for allocating costs. The real question is simply whose model to use.

1. Motion to Strike Testimony of Michael Smith

Prior to trial, ALT and Reid moved in limine to bar the testimony of Michael J. Smith, the City’s expert witness regarding modeling.. By Minute Orders of September 10, 2002, the court denied the motions without prejudice to reassertion in post-trial submission. In their post-trial submissions, ALT and Reid have renewed and supplemented their motions with references to Smith’s trial testimony. The City has responded. The court has read and reread Smith’s trial testimony, which spans almost 350 pages of transcript, and has concluded that it will not be in the best interest of a comprehensive and complete resolution of the issues to strike Smith’s testimony in its entirety. Indeed, despite ALT’s and Reid’s vigorous attacks on Smith’s testimony and opinions, ALT’s experts adopted a portion of Smith’s work in their own opinions. In particular, McBean performed some of his modeling based on Smith’s source loadings, and Frank Rovers recommended using McBean’s model with Smith’s source numbers for allocation of response costs. However, for the following reasons, the court declines to credit the majority of Smith’s testimony because it fails to meet the requirements of the Federal Rules of Evidence and applicable case law.

Smith was the City’s expert witness on groundwater modeling, and contaminant fate and transport modeling issues. In close consultation with Olsen, Smith used CDM’s proprietary computer codes DYN-FLOW and DYNTRACK to develop models to simulate the areal location, movement and size of plumes of contaminants emanating from sources at the Site, including, but not limited to, sources representing the businesses operated by defendants. DYNFLOW is a flow model and DYN-TRACK is a contaminant transport model. Smith also used models such as CHAIN, a government model used to analyze contaminant transport when there is degradation of the contaminants. Smith’s models were submitted to, commented upon and ultimately approved by the KDHE. The KDHE used Smith’s final model to determine and approve the number and location of groundwater extraction wells, an integral part of the pump and treat remedy selected for the Site. Olsen later employed Smith’s conclusions to assign areas of responsibility, expressed in percentages, for Site-wide costs and individual plume-specific costs to sources of contamination at the Site. Nevertheless, the KDHE never examined or approved Smith’s allocation modeling, which utilized different parameters. Therefore, the court does not consider the KDHE’s review and approval of Smith’s models as any sort of official endorsement of the reliability and. relevance of Smith’s and Olsen’s allocation testimony and opinions.

ALT’s and Reid’s attacks on Smith’s testimony and opinions are premised upon Fed.R.Evid. 702, Daubert v. Merrell Dow Pharm. Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), Kumho Tire Co. v. Carmichael, 526 U.S. 137, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999) and cases discussing those rules and decisions. The court and the parties are familiar with the applicable law and no useful purpose will be served' by an extended general discussion of the case law pertaining to the admission of expert testimony.

Rule 702 first requires an expert witness to be qualified. Neither ALT nor Reid directly challenges Smith’s experience, training or education, which appear to be extensive. Instead, they attack the facts and data supporting Smith’s opinions, argue that his testimony is not the product of reliable principles and methods and that he has not reliably applied the principles and methods to the facts. The City, as the proponent of Smith’s opinions, asserts just the opposite.

Once an expert meets Rule 702’s qualification requirements, his or her opinions must be based on sufficient facts or data. This is the rule’s requirement for foundation. In the main, the facts and data used by Smith were supplied by Olsen. Smith did not independently confirm the accuracy of any of the data. This created problems for him during his testimony. For example, in connection with the preparation of his first allocation model, Olsen told Smith that Reid was a source of TCE contamination at a concentration of 1000 ppb in groundwater near Reid. In fact, no such concentration level was ever recorded.

iii. Rule 702(2) and (3)

Rule 702(2) and (3) require that the expert’s testimony be the product of reliable principles and methods and that the expert apply the principles and methods reliably to the facts. These requirements are discussed extensively in the Daubert line of cases. To be reliable, the expert’s testimony must be based on the “methods and procedures of science” and reflect more than the witness’ “subjective belief or unsupported speculation.” The court must assess whether the “reasoning or methodology” underlying the testimony is scientifically valid and whether the reasoning or methodology properly can be applied to the facts in issue. A determination of reliability requires consideration of the flexible, non-exclusive and now-famous Daubert factors: (1) whether the theory or technique can be or has been tested; (2) whether the theory or technique has been subjected to peer review or publication; (3) whether there is a potential or known rate of error for the theory or technique; and (4) whether the theory or technique is accepted in the scientific community. 509 U.S. at 589-94, 113 S.Ct. at 2796-97. It is here that the battle is truly joined regarding Smith’s testimony and opinions.

Computer modeling is an accepted and, in appropriate circumstances, reliable method for use in determining groundwater flow and contaminant transport in an aquifer, and to evaluate the effectiveness of remedial alternatives. This is especially true when there is extensive and overlapping contamination from multiple sources, where contamination cannot be traced entirely to a specific source and when the extent of contamination is difficult to determine by “field methods” such as geoprobing and drilling test wells. Nevertheless, even in the best of circumstances, a model is only an estimate and the accuracy of the estimate depends to a considerable extent on the data selected for use in the computer model, the quality and reliability of that data and, of course, the skill of the modeler. Smith acknowledged that modeling results are not a perfect match to actual conditions but rather an approximation of what happens in reality. He conceded that his modeling techniques and methodology are not based on any specific guidelines or standards, but rather on his “professional judgment.”

Models are tested by calibration, which Smith generally described as:

Well, model calibration is the process whereby the parameters that control flow and flow characteristics, flow directions, are adjusted to match to a reasonable degree the data that we have from the field. Because we’re dealing with the subsurface, you can’t measure characteristics at every location at the site and so the calibration process is how you adjust for that.

The actual process itself consists of repeatedly running the model, making changes to the most sensitive parame ters, comparing the results of those simulations to the field data and then adjusting in an iterative type fashion.

* * * * * *

Once you meet this criteria of a reasonable fit to the field data, then you consider the model calibrated.

Examples of field data include source locations and degradation rates of the contaminant, in this case chlorinated solvents, which degrade from PCE.to TCE to DCE to VC. If the area of contamination or “plume” simulated by the model does not match the area defined by field data, the modeled plume can be “truncated” to, in effect, eliminate any areas where the model says there is contamination but the field data says there is not. However, apparently using his- “professional judgment,” Smith did not always “truncate” areas where the model showed contamination but the field data did not. He acknowledged, for example, that his model at APCO was not consistent with reality because it modeled contamination north or upgradient of APCO which was not confirmed by field data. Yet he did not truncate that area. The effect of this use of “professional judgment” was to allocate contamination to APCO which APCO could not have caused.

Smith prepared two reports regarding allocation models. All of the information he used to prepare his second report was available when he prepared' his first report. Nevertheless, Smith’s modeling methodology or technique changed. A predominant feature of the model in his second report was a “plug,” a subsurface condition of reduced hydraulic conductivity or lower permeability — a measure of how well fluid flows through the subsurface medium. The existence or non-existence of the plug could have been confirmed by drilling a test well, but Smith did not request that be done. The size of the Reid plume was dramatically increased by the addition of the unconfirmed “plug.”

In his first report, Smith modeled both PCE and TCE, the so-called “parent products” which he assumed had been spilled or otherwise introduced into the environment at numerous source areas at the Site. The significance of using PCE or TCE is important because they are produced and used commercially. In his second report, at the Reid and APCO locations only, and in consultation with Olsen, Smith modeled DCE as a parent product. DCE is a “daughter product” which exists only as a product of dechlorination. Even' though he was modeling DCE, Smith varied from accepted or customary modeling techniques by modeling the ACL for VC at APCO and Reid locations only, instead of following the usual practice of using the same ACL as the modeled product, e.g., PCE/PCE. Smith did not disclose this divergence from usual practice in his second report; it was discovered at his deposition. Smith claimed that after talking with Olsen, they decided to model DCE to simulate the degradation of DCE to VC at the Reid plume. The court cannot accept this all-too-convenient explanation. Smith acknowledged that never before in his professional career had he modeled a parent product using the ACL for a daughter or surrogate product. He had never seen any literature which approved the use of a surrogate! Smith admitted that if he had modeled DCE using the corresponding ACL line for DCE, the resulting simulated plume would have been “considerably smaller.” In contrast to what appears to be a ■ shortcut by Smith, when ALT’s McBean-wanted to know the plume sizes for parent and daughter products, he modeled them both. Had Smith done something similar, his work might have retained more credibility. Finally, Smith and Olsen knew that allocation modeling was for use in this case. In other-words, they knew the rules of the game. Smith’s failure to follow the rules by not disclosing his change from usual modeling methodology cannot be seen as an innocent mistake. This damaged not only Smith’s, but Olsen’s, credibility, too.

The remaining Dauberb factors can be quickly disposed of. While the evidence was conflicting about whether DYNFLOW and DYNTRACK had been peer-reviewed, it is clear beyond question that Smith’s particular use of DYNFLOW and DYN-TRACK in connection with his allocation modeling for the Site had never been peer-reviewed. There was no evidence that his allocation modeling technique was accepted by the scientific community. Similarly, since Smith’s allocation modeling technique was unique to this case, there could be no known rate of error. These are merely examples of problems associated with Smith’s modeling techniques. No useful purpose will be served by describing others.

Rule 702’s third requirement is that the witness must apply the principles and methods reliably to the facts of the case. For the reasons already noted, the Court finds that Smith did not do that. As the Supreme Court observed in Gen. Elec. Co. v. Joiner, 522 U.S. 136, 118 S.Ct. 512, 519, 139 L.Ed.2d 508 (1997): “nothing in either Daubert or the Federal Rules of Evidence requires a district court to admit opinion evidence that is connected to existing data only by the ipse dixit of the expert. A court may conclude that there is simply too great an analytical gap between the data and the opinion offered.” Joiner involved exclusion of expert testimony in connection with a summary judgment ruling. The Court’s words apply with even more force to consideration of opinions following a full court trial. There are simply too many gaps in Smith’s testimony.

In countless opinions, courts have observed that an overriding purpose of expert testimony is helpfulness to the trier of fact. The vast majority of these opinions analyze a judge’s decision to admit or exclude expert testimony offered in a jury trial. In this court trial, the written transcript does not adequately reflect the difficulties with the presentation of Smith’s unnecessarily long, complex and confusing testimony. Both Smith and his presenting counsel were admonished that the court was having great difficulty understanding Smith’s testimony. The admonitions went unheeded. Eventually, the problem became so bad that the court threatened Smith with contempt and restricted him to “yes or no” answers on cross-examination, a step which was not necessary with any of the other witnesses, many of whom were experts. In the end, Smith’s testimony did little to help the court reach a decision in this difficult case. Accordingly, Smith’s testimony and his modeling work are accorded little weight.

2. ALT’S Modeling and Allocation

ALT retained Dr. Edward McBean as their modeling expert. Although McBean thought other modeling tools would have been better suited for the Site, for sake of consistency he used the same DYNFLOW, DYNTRACK, and CHAIN models employed by the City’s modeler, Smith. Although McBean used the same modeling tools as Smith, McBean applied the models differently, independently determining the proper data and parameter inputs to best simulate the Site aquifer.

Even if the value of Smith’s modeling had not already been discounted, the court would conclude (and does conclude) that McBean’s modeling is more reliable and provides a more accurate basis on which to allocate costs. First of all, McBean modeled both the parent and the daughter products emanating from each source. Smith modeled only the parent products for most sources. Smith based this decision on his conclusion that significant deg radation of PCE and TCE was not occurring for the upgradient sources; hence, there were no significant concentrations of DCE or VC from those sources, which were not defendants in this lawsuit. However, the evidence belies this conclusion. As previously discussed, dechlorination occurs under particular reservoir conditions, including an anaerobic environment and the presence of electron donors. The court finds it highly unlikely that such conditions exist only in places where APCO and Reid are the only sources. Even more concrete, however, is the fact that the evidence clearly shows DCE concentrations in excess of the ACLs in the deep portion of the aquifer upgradient of Reid. The evidence also shows that chlorinated solvents in the deep portion of the aquifer most likely came from relatively distant sources. Thus, the only possible conclusion is that parent products from sources in the northern part of Plume B were degrading to DCE in excess of the ACLs. Smith did not model that. McBean did.

Next, Smith’s model contained a peculiar anomaly that substantially benefitted the City. Although the predominant direction of groundwater flow in Plume B is north-to-south, and although CDM’s own documents specifically show groundwater flow in the vicinity of the City’s Bus Barn as being north-to-south, Smith’s model projected the contamination plume from the Bus Barn toward the southwest. This had the effect of quickly driving the City’s plume outside the known boundaries of Plume B. Since Smith’s protocol was to truncate or ignore any modeled plume that was outside the known boundaries of the actual plumes, a substantial portion of the City’s pollution plume was truncated and thus ignored. The irregularity of this modeling result is emphasized by a quick review of the other sources Smith modeled, virtually all of which show the plumes expanding almost due south, even in the area of the City’s Bus Barn. Conversely, McBean’s models provided a result for all sources that appeared more consistent with the actual field data. For these reasons, the court concludes that McBean’s modeling is more credible. Accordingly, McBean’s modeling will form the basis for ALT’s allocation.

The next choice becomes which of McBean’s models should be adopted for allocation purposes. The primary difference between the various McBean models is the source loading. Source loading is essentially the amount of a particular chlorinated solvent that the model assumes is coming from a specific source. McBean modeled the final source loadings used by Smith, as well as various source loadings that he calculated himself. Although the court might ordinarily have to decide which of the McBean models to adopt, ALT simplified the process when the trustees’ allocation expert, and McBean coworker, Frank Rovers, recommended adoption of the McBean model using Smith’s source loadings. 59 The court ac cepts that invitation, noting that Rovers was a very credible witness whose testimony proved quite helpful in deciding this matter. 60

The results of McBean’s modeling are summarized in ALT-1445. 61 The model estimated the total size of the APCO contaminant plume, including daughter products, at 897,854 square feet. This plume encompassed 1.72% of the total contamination in Plumes A, B, and E. Rovers applied this percentage directly to investigation and design costs submitted by the City’s allocation expert, Olsen. However, Rovers suggested an alternative method for allocating groundwater remediation costs. 62 Rather than apply this percentage based on individual plume size, Rovers proposed that ALT’s share of downgradient remediation costs should be based only on the number of extraction wells actually removing APCO’s contamination. Since McBe-an’s modeling showed that only one of the thirteen extraction wells actually draws from APCO’s plume, Rovers suggested that ALT’s share of downgradient remediation costs should begin at l/13th of the City’s overall downgradient remediation costs. Rovers also opined that ALT’s share should be further reduced in proportion to the number of other individual contaminant plumes that are pumped through the same well. Since McBean’s model showed that the well which draws from the APCO plume also takes in contamination from five other sources, Rovers felt that ALT’s l/13th share should be reduced to l/6th of l/13th. The net result of this mathematical exercise would require ALT to pay 1.28% of downgradient remediation costs, rather than the 1.72% calculated using APCO’s plume size.

While Rover’s analysis for allocating downgradient costs is certainly not baseless, the court perceives this as an effort to parse too finely the costs at issue. Although the computer model separated the contamination plumes out by individual sources, the reality is that chlorinated solvents, gasoline, oil, and numerous other pollutants have commingled into the proverbial witch’s brew beneath the Site. Anyone would be hard pressed to distinguish one gallon of polluted water from another as it is withdrawn from the aquifer. APCO was part of the overall problem; thus, it should be part of the overall solution. It is merely fortuitous that APCO’s plume is drawn into a single well rather than two or more extraction wells. If the latter had been the case, Rover’s method may well have produced a higher percentage allocation for ALT. Accordingly, the court concludes that the better method is to allocate groundwater remediation costs to ALT based on the size of APCO’s contaminant plume. Since the pump and treat system is employed across plumes A, B, and E, ALT’S allocation is based on the size of its plume relative to the combined areas of plumes A, B, and E — 1.72%.

In sum, ALT will bear 1.72% of all past costs, other than those attributable to source control at the City’s Bus Barn. Applying this percentage to the $10,693,937.51 the court previously found recoverable for response actions other than at the Bus Barn, ALT is liable for $183,935.73 in past costs, exclusive of any prejudgment interest.

3. Reid’s Allocation

Unlike ALT, Reid did not present its own modeling and allocation experts at trial. Instead, Reid chose to base its main defense on the theory that Reid was not a source of groundwater contamination at the Site. Although somewhat equivocal at first, Reid ultimately adopted the recommendations of McBean and Rovers on the alternative basis that, if the court finds Reid liable, McBean’s modeling and Rovers’ allocation figures are more accurate than the City’s. Reid reiterated this position in its post-trial briefs. Having already ruled that Reid is a source of groundwater contamination, and generally finding that the work of McBean and Rovers is more credible than that of the City’s experts, the court will allocate costs to Reid based on Rovers’ recommendation; 63 however, like ALT, Reid’s future costs will be based on plume size, and not by dividing costs related to a particular extraction well. 64

Pursuant to McBean’s modeling and Rovers’ allocations, Reid will bear 0.15% of the City’s past response costs, exclusive of any source control actions at the City’s Bus Barn. Applying that percentage to the $10,693,937.51 in costs that the court previously found recoverable, Reid will be liable for $16,040.91, exclusive of prejudgment interest.

4. Tri-Supply’s Allocation

Unlike the situation at the APCO and Reid facilities, 330 S. Commerce has been the home to at least two businesses that handled PCE. Thus, even though the evidence shows soil and groundwater contamination at 330 S. Commerce, the court finds it much more difficult to determine who caused the pollution. The only evidence of a release during Tri-Supply’s tenancy shows two gallons of PCE were spilled during a delivery to the facility. The spill occurred on cement during the summer of 1994. Furthermore, there was a great deal of evidence at trial which showed that only a small fraction of spills on cement could actually penetrate to the underlying soil. Moreover, evidence also showed that spills evaporate more quickly during periods of warmer weather. Thus, only a tiny fraction of this two-gallon spill, if any at all, actually made it to the soil beneath the cement floor. In addition, any PCE that made it to the soil would have a difficult time migrating down approximately 18 feet to the underlying aquifer since that soil was overlain by the cement floor. The evidence showed that one of the principle mechanisms to wash the chlorinated solvents down to the aquifer was runoff from rainwater. Soil immediately beneath cement would be sheltered from most, if not all, of the runoff; thus, the PCE in the soil would be less likely to leach down to the aquifer. Finally, even after reaching the soil, some of the PCE coats the soil grains and does not continue leaching to deeper strata. 65 Accordingly, the court finds that no appreciable amount of this two-gallon spill contaminated the groundwater.

In addition to the tenuous link between this two gallon spill and the levels of contamination detected at and under 330 S. Commerce, the court notes that Tri-Sup-ply was not the only entity at that location to handle PCE. Tri-State Laundry operated there from 1987-1990. Tri-State maintained a 4,500 gallon PCE tank on the premises. The City’s expert on sources, Hendron, submitted an expert report that showed Tri-State Laundry received 596,-000 pounds of PCE during its tenure at 330 S. Commerce, almost 6 times as much as Tri-Supply handled. Tri-State Laundry apparently did a substantial amount of business repackaging PCE into smaller containers for re-sale. Finally, although Hendron opined at trial that Tri-Supply was the source of PCE groundwater contamination from 330 S. Commerce, in his expert report he stated that both Tri-State and Tri-Supply had releases that contaminated the groundwater.

In contrast to the City’s version of facts, Kratz credibly testified that he was very concerned about PCE handling at 330 S. Commerce. He cautioned his employees to follow appropriate safe-handling procedures and prevent any spills. Furthermore, it appears that the only time PCE was openly handled at Tri-Supply was when a tanker truck delivered to the facility. During delivery, PCE was transferred from the truck to 55-gallon drums. Thereafter, Tri-Supply simply delivered those drums to its customers. It was during one of the bulk deliveries to the TriSupply facility that the 2-gallon spill was reported on a United States Department of Transportation Incident Report form. The fact that this relatively small spill was officially reported during a delivery suggests that either the delivery company, Tri-Supply, or both took PCE spills seriously. The absence of reported spills during other deliveries suggests that no spills occurred during those deliveries. This observation, when combined with the fact that Tri-Supply merely delivered the filled drums to its customers, suggests that PCE spills were not particularly likely in TriSupply’s operations. The City has provided no evidence to the contrary. Finally, unlike APCO, which operated in the 1960’s and 1970’s, an era when operators were not particularly sensitive to the dangers of chlorinated solvents, Tri-Supply operated in the mid-1990’s, a time when environmental awareness was substantially higher and, more noteworthy, a time when the problems at the Site were already known to the public.

Taking all these facts and circumstances into account, the court concludes that the City has proved only a 2-gallon spill of PCE during Tri-Supply’s tenure at 330 S. Commerce. Unlike the other defendants, the City has not been able to establish a pattern of behavior at Tri-Supply that would lead the court to conclude Tri-Sup-ply was routinely spilling PCE. The City did little to identify the operating practices of a prior tenant, Tri-State Laundry. The fact that Tri-State Laundry had a 4500-gallon PCE tank suggests that PCE was being transferred between the tank and other containers on a more routine basis, thereby giving many more chances for spills. Having already found that no appreciable amount of Tri-Supply’s two-gallon spill could have contaminated the underlying aquifer, the court concludes that Tri-Supply should not be liable for any of the City’s past or future response costs. 66

5. Land Tool Company’s Allocation

Land Tool Company and E.H. Land presented evidence only on the issue of liability. They presented no modeling of their own, nor did they adopt any modeling or allocation schemes presented by other parties. Since the court has concluded that Land Tool is liable for contribution under CERCLA § 113(f), some method of allocation must be applied. Lacking any better means of reducing Land Tool’s allocation to an actual dollar figure, the court will use the City’s proposed allocation. Accordingly, Land Tool Company and E.H. Land are responsible for 0.04% of the City’s past costs. Applying that percentage to the $10,693,937.51 in costs that the court previously found recoverable, Land Tool Company and E.H. Land will be liable for $4,277.58 in past costs, exclusive of prejudgment interest. The City has not asked for a declaratory judgment against these defendants as to future costs; thus, Land Tool Company and E.H. Land are not liable for any future costs.

C. Prejudgment Interest

The final issue directly related to allocation of response costs is whether the City is entitled to prejudgment interest. Pursuant to CERCLA § 107(a), a party can recover prejudgment interest that accrues “from the later of (i) the date payment of a specified amount is demanded in writing, or (ii) the date of the expenditure concerned.” 42 U.S.C. § 9607(a). This provision also applies to a contribution ac tion under CERCLA § 113(f). See Bancamerica, 100 F.3d at 801.

The City claims it is due prejudgment interest from the date its complaint was served on defendants. The complaint states that the City has spent over $3.8 million in response costs as of the date of filing, as well as an additional $250,000 in recoverable attorneys’ fees. The complaint then goes on to name twenty-nine different defendants against whom the City’s claims are made. This lack of specificity as to how much was being demanded from each defendant is the crux of the problem in determining whether the City is entitled to prejudgment interest from the date of its complaint.

In Bancamerica, the Tenth Circuit concluded that a complaint stating that the plaintiff had spent “in excess of $1 million” was sufficient to satisfy the written demand requirement of CERCLA § 107(a). Bancamerica, 100 F.3d at 801. In that case, however, there were only two corporate defendants, one of which was a subsidiary of the other. See Trinity Indus., 900 F.Supp. at 1436. In support of its decision, Bancamerica cited Matter of Bell Petroleum Servs., 3 F.3d 889 (5th Cir.1993). Bell held that a complaint could satisfy § 107(a)’s written demand requirement, even if it did not specify an amount of money sought. Bell, 3 F.3d at 908. Moreover, Bell involved three separate defendants. Id. at 892. Nonetheless, Banc- america’s reliance on Bell does not mean that the Tenth Circuit has adopted Bell wholesale.

In United States v. Consolidation Coal Co., 345 F.3d 409 (6th Cir.2003), the Sixth Circuit concluded that a complaint was not sufficient to satisfy § 107(a)’s written demand requirement when the complaint named 59 different defendants. Id. at 416. The Sixth Circuit noted that both Bell and Bancamerica held a complaint could provide the required written notice, but reasoned that when the complaint named a large number of defendants, with no indication as to how much was sought from each defendant, the complaint could not be construed to meet the “specified amount” requirement of § 107(a). Id. The court finds the reasoning in Consolidation Coal to be applicable here. While a complaint has the potential to meet the written demand requirements of § 107(a), Bancamerica, 100 F.3d at 801, the mere fact that a complaint states a dollar amount does not necessarily meet the specificity requirement — the inquiry must be decided on the facts of each case. In Bancamerica, and similar cases with one defendant or a few financially related defendants, even a vague estimate of the dollar amount sought puts the defendant on notice as to the magnitude of the costs at issue, as well as the fact that the defendant is being asked to pay the entire amount sought. By contrast, where the complaint names over two dozen defendants without even hinting at how much money is being sought from each defendant, the complaint cannot satisfy § 107(a)’s mandate that a “specified amount” be demanded in writing. 67 Defendants had no way of discern ing what portion of the overall amount is claimed to be due from each individual defendant (or group of financially related defendants). Accordingly, the City is not entitled to prejudgment interest from the date its complaint was served.

Even so, ALT concedes that a satisfactory demand was made on the trustees during mediation on April 11, 2002. Thus, the City is entitled to prejudgment interest from ALT accruing from April 11, 2002. Conversely, the City points to no other evidence showing appropriate written demand on Reid or Land Tool. 68 Hence, the City is not entitled to any prejudgment interest from those defendants.

D. Declaratory Judgment

In addition to past costs, the City also seeks a declaratory judgment against ALT, Reid Supply, and Walter Trombold for future response costs at the Site. 69 CERCLA permits entry of declaratory judgment as to liability for future costs. See 42 U.S.C. § 9613(g)(2). However, such a declaration does not settle the matter as to recoverability of any particular costs. See Hardage, 982 F.2d at 1445. ALT argues vigorously that the City may not obtain a declaratory judgment for source control measures at the APCO facility because the City has not established that any particular source control remedy would be consistent with the NCP. However, lack of compliance with the NCP “is a defense to the recoverability of particular response costs, not a defense to liability for those costs.” Hardage, 982 F.2d at 1445. Accordingly, if the other prima facie elements of liability have been met, the court may enter a declaratory judgment as to liability for future costs, without depriving defendants of an opportunity to challenge the recoverability of those costs at a future date.

Here, the court has spent almost 100 pages determining that the prima facie elements of CERCLA liability have been met with respect to groundwater remediation. Accordingly, the City is entitled to a declaratory judgment as to liability for future costs of groundwater remediation. Future costs for such remediation should be allocated on the same basis as past costs: the relative size of each defendant’s contaminant plume. Thus, ALT and Reid are liable for 1.72% and 0.15% respectively of the City’s future groundwater remediation costs for the Site.

With respect to source control measures, the prima facie elements of covered person, release or threatened release, and incurrence of response costs are the same as for groundwater remediation, and have already been proved. Since NCP consistency and necessity of costs are defenses to recoverability, and not liability, in the declaratory judgment context, the City is entitled to a declaratory judgment for future source control costs. Interestingly, the 'City asks that ALT and Reid be declared liable for a flat percentage of all source control measures within the Site. That is not appropriate. The mobility of contaminants in groundwater makes it difficult to precisely identify the source; thus, requiring all PRPs to bear a percentage of the overall costs makes sense. Unlike groundwater contamination, chlorinated solvents in the soil do not travel horizontally for any great distances. Thus, the soil contamination at APCO came from APCO, and the soil contamination at Reid came from Reid. Accordingly, ALT will be liable for 100% of the source control measures at 1001 E. Lincoln. Similarly, Reid and Walter Trombold are liable for 100% of the source control costs at 911 E. Indianapolis. Conversely, defendants are not liable for any source control costs at other locations. ALT and Reid will be free to challenge the recoverability of any particular source control costs once those costs are incurred , and the City makes demand for reimbursement.

V. REID’S MOTION TO RECONSIDER 70

Reid asks the court to reconsider the equitable allocation assigned to Reid pursuant to CERCLA § 113(f). Reid argues that all of the contamination at 911 was caused by Vulcan Chemical, not by Reid. Thus, Reid claims that its allocation should be shifted to Vulcan. In addition, Reid asks the court to clarify the terms under which it may have to pay the City’s response costs for performing soil remediation at 911.

A. Reid’s Equitable Allocation

The court found Reid liable for response costs at 911, and allocated Reid a mere 0.15% of past and future response costs related to groundwater remediation, as well as 100% of any recoverable response costs related to future soil remediation at the Reid facility. Reid now challenges that allocation. Based on the court’s conclusion that the only releases proved by the City for 911 occurred outside Reid’s warehouse and were likely related to bulk deliveries by Vulcan, Reid argues that its allocation should be further reduced (or altogether eliminated) by shifting Reid’s share to Vulcan.

Quite frankly, the court is surprised that Reid raises this issue. The City, relying on its own modeling, sought over $400,000 in past costs from Reid, plus 3.1% of all future groundwater remediation costs. Moreover, the City also requested that a substantial portion of the orphan share be shifted to Reid, -thereby increasing Reid’s allocation to over $1 million for past costs, plus 8.9% of all future groundwater remediation costs. Conversely, Reid provided no modeling data of its own, steadfastly maintaining that it was not responsible for any spills at its facility. But as a fallback plan, Reid expressly adopted, without reservation, the costs estimated by ALT’S allocation expert, Frank Rovers.

In reliance on the express adoption that Reid now tries to repudiate, the court used ALT’S modeling to allocate costs to Reid. As a result, Reid avoided hundreds of thousands, and perhaps millions, of dollars in response costs, receiving instead an allocation of only $16,040.91 in past .costs, and a similarly meager allocation of future groundwater remediation costs. Not once did Reid suggest that it merely wanted to adopt the plume modeling performed by ALT’s expert, McBean, with some intention to divide responsibility for that plume between itself and Vulcan. Instead, Reid specifically stated that its allocation for past costs “should be limited to the $23,888 as determined by ... Rovers. 71 ” Those numbers represented 100% of the past costs associated with remediating the groundwater contaminated by spills at 911. Reid never suggested that it wanted to be responsible for only a portion of those costs. Nor did Reid suggest that those figures ought to be applied to Vulcan. On the contrary, Reid stated that 100% of the past costs for groundwater remediation at 911, as estimated by Rovers, should be “assessed against Reid Supply.”

Having found that Reid was liable for spills at 911, the court was faced with the dilemma of trying to decide how to allocate costs to Reid. While the City provided extensive modeling to estimate Reid’s share of the contamination, Reid did not put forth any modeling or any other evidence on which to base its allocation. Were it not for Reid’s adoption of Rovers’ figures, the court could quite easily have imposed the City’s numbers on Reid, since Reid failed to put forth any evidence to the contrary. However, having found that ALT’s modeling was generally more credible than the City’s, the court elected to use ALT’s modeling for Reid as well. Inseparably bound up in that decision was the court’s awareness that Vulcan might have been a participant in any spills at 911. Accordingly, the court concluded that any potential unfairness toward Reid by not allocating some of the 911 pollution to Vulcan was mitigated by giving Reid the benefit of ALT’s modeling, rather than using the City’s figures. In any event, the evidence showed, and Reid conceded, that Reid’s employees were intimately involved every time Vulcan delivered to 911. 72 Accordingly, there was no reason for the court to allocate a substantial part of the 911 costs to Vulcan. Thus, by allowing Reid to use ALT’s figures, the court has been more than fair with Reid in this case. Reid’s motion to reconsider this decision is DENIED.

B. Recoverability of Future Soil Remediation Costs

Reid also asks the court to clarify its ruling regarding recoverability of any future soil remediation costs. Reid seeks assurances that it will be able to defend a source control cost recovery action on the ground that 911 is not a continuing source of chlorinated solvents to the underlying groundwater in excess of the ACLs. Such a ruling is not appropriate at this time.

CERCLA permits recovery of necessary response costs that are consistent with the NCP. 42 U.S.C. § 9607(a)(4)(B). It also directs that a declaratory judgment be entered as to liability for future response costs whenever a contribution action is decided. Id. § 9613(g)(2); see also Hardage, 982 F.2d at 1445. However, a party may not challenge the recoverability of any particular costs until recovery is sought by the plaintiff. See Hardage, 982 F.2d at 1445.

In order to recover any costs for soil remediation at 911, the City will have to show that its actions were consistent with the NCP, and that its response costs were necessary to accomplish the cleanup. Reid will have the benefit of whatever defenses the law will permit it to raise at that time, including that the City’s actions are not NCP compliant. While groundwater contamination has been the main focus of cleanup efforts so far, the court finds it premature to say that the City may only remediate the soil if it proves that chlorinated solvents are still leaching to the groundwater in excess of the ACLs. If the City finds and proves that the soil contamination at 911 poses other environmental hazards for which the NCP authorizes cleanup, then the City should be able to recover those costs. 73

VI. CONCLUSION

IT IS THEREFORE ORDERED that defendants are liable for the City of Wichita’s response costs as follows:

1. The Trustees of the APCO Oil Corporation Liquidating Trust are liable for $183,935.73 in past costs, plus prejudgment interest accruing from April 11, 2002; the trustees are also liable for 1.72% of all future costs of groundwater remediation for Plumes A, B, and E; and finally, the trustees will be liable for 100% of future source control measures taken at the former APCO facility.

2. Reid Supply Company, Inc. and Walter Trombold are liable for $16,040.91 in past costs plus 0.15% of all future costs of groundwater remediation for Plumes A, B, and E; Reid Supply Company, Inc. and Walter Trombold are also liable for 100% of future source control measures at 911 E. Indianapolis.

3. Land Tool Company and E.H. Land are liable for $4,277.58 in past costs, with no liability for any future costs or for prejudgment interest.

4. Tri-Supply Company and Gordon Kratz are not liable for any costs, past or future.

APPENDICES

APPENDIX A

Significant Pretrial Rulings

1. July 28, 2000 Memorandum and Order (Doc. 732), 177 F.Supp.2d 1153. The Court determined that the City was limited to an action for contribution under CERCLA § 113(f), and granted defendants’ joint motion for summary judgment with respect to the City’s cost recovery claims under CERCLA § 107(a), because the City was a PRP based on its current ownership .of contaminated property consisting of a portion of the Coleman B and the Bus Barn properties. Id. at 1165. The Court also held that the City’s claims were not time-barred by the limitations periods provided in CERCLA § 113(g), and denied defendants’ joint motion for summary judgment with respect to the statute of limitations.

2. December 15, 2000 Memorandum and Order (Doc. 950). The Court held that the City’s CERCLA § 113(f) contribution action against the ALT Trustees was not barred even though no CERCLA § 107 civil action or § 106 administrative action had been brought against the City. The Court denied the ALT Trustees’ motion for summary judgment with respect to this issue.

3. December 22, 2000 Memorandum and Order (Doc. 960). The Court held: (1) that the City’s case against APCO and the APCO Liquidating Trust is not barred by the terms of the Chancery Court’s order establishing the Trust; (2) that APCO and the APCO Liquidating Trust have the capacity to be sued for APCO’s CERCLA liability; and (3) that the City has made out a prima facie case that its response costs are reasonable and necessary and that genuine issues of material fact exist with respect to the City’s response costs. The Court denied the ALT Trustees’ motion for summary judgment with respect to each of these three issues.

4. March 13, 2001 Memorandum and Order (Doc. 1036). The Court denied the ALT Trustees’ motion for reconsideration of the ruling in the Court’s December 22, 2000 Memorandum and Order (Doc. 960), that the City’s case against APCO and the APCO Liquidating Trust is not barred by the terms of the Chancery Court’s order establishing the Trust. The Court also denied the ALT Trustees’ motion for certification of interlocutory appeal of both the Court’s December 15, 2000 Memorandum and Order (Doc. 950) and the Court’s December 22, 2000 Memorandum and Order (Doc. 960).

5. September 25, 2001 Memorandum and Order (Doc. 1146). The Court denied Gordon Kratz’s motion for summary judgment with respect to CERCLA “operator liability.” With respect to the City’s claim to hold Mr. Kratz liable by piercing TriSupply’s corporate veil, the Court determined that the City had presented sufficient evidence upon which a jury could conclude that the first prong of the Greater Kansas City Roofing veil-piercing test had been satisfied. However, the Court directed the parties to file supplemental briefs addressing in more detail the evidence the City relies upon to support Mr. Kratz’ personal liability under the second prong of the Greater Kansas City Roofing test.

6. January 3, 2002 Memorandum and Order (Doc. 1192). The Court determined that the City has presented sufficient evidence upon which a jury could conclude that the second prong of the Greater Kansas City Roofing veil-piercing test had been satisfied. The Court denied Mr. Kratz’ motion for summary judgment as to the City’s veil-piercing claim.

7. September 9, 2003 Memorandum and Order (Doe. 1315). The Court determined that the City’s current claims against the Trombolds for CERCLA “operator liability” are not “substantially related” to any of the matters involved in attorney Larry Meeker’s former representation of Reid Supply and/or the Trom-bolds in mid-1980s. The Court denied the motion of Reid Supply and the Trombolds for disqualification with respect to the City’s operator liability. In the Memorandum and Order, the Court noted that two other issues raised in the motion for disqualification were resolved in a telephone conference held on July 26, 2003. During that conference, it was determined that disqualification of Stinson Morrison was not necessary insofar as the City’s CERC-LA claims against Reid Supply are concerned. It was also determined that a decision regarding disqualification could be deferred insofar as the City’s claims to pierce Reid Supply’s corporate veil to reach the Trombolds individually are concerned. The Court severed the Reid Supply veil-piercing claims from the trial of the CERCLA liability claims.

APPENDIX B

Summary of Key Witnesses

(Doc. 1424 at 5-12)

1. Abedini, Mike. (Tr. Vol. 16, 10/25/02). Former employee of Land Tool Company. He testified handling and disposal of chlorinated solvents at Land Tool.

2. Bean, Rick. (Tr. Vols. 1-2, 10/1/02 — 10/2/02). The KDHE’s project manager for the Site until 1994. He testified regarding the interactions between the City and the KDHE during the investigation of the Site.

3. Bessey, Erika. (Tr. Vol. 9, 10/15/02). Ms. Bessey is an attorney with the KDHE. She testified regarding negotiations between the KDHE and ALT regarding source control measures at the former APCO facility

4. Brown, Jack. (Tr. Vols. 6 & 8, 10/8/02, 10/10/02). Former Director of Environmental Health for the City of Wichita.

5. Clegg, Bruce. (Tr. Vols. 23 & 24, 11/7/02, 11/12/02). APCO’s expert witness on data quality and NCP consistency issues.

6. Egan, Joseph. (Tr. Vol. 19, 10/30/03). The City’s expert witness regarding the necessity and consistency of the City’s response costs with the NCP.

7. Eversman, Michael. (Tr. Vols. 29-30, 11/20/02 — 11/21/02). Reid Supply’s source identification expert.

8. Faulkner, Dennis. (Tr. Vols. 18-19, 10/29/03, 10/30/03). The City’s expert regarding response costs.

9. Fisher, Brian. (Tr. Vols. 7-8, 10/9/02 — 10/10/02). Employed in the Wichita Department of Environmental Health. Fisher worked closely with the Gilbert and Mosley project, and testified regarding his work, including geoprobe operations.

10. Gallis, David. (Tr. Vol. 29, 11/20/02). Reid Supply’s expert witness regarding on-site laboratory data quality.

11. Hendron, David. (Tr. Vols. 15-17,-10/24/02, 10/25/02, 10/28/02). The City’s source identification expert.

12. Henning, Shawn. (Tr. Vol. 18, 10/29/03). Henning is the City’s Controller. He testified regarding response cost documentation; and City financial/aceount-ing/audit/budget procedures.

13. Jump, Christine. (Tr. Vols. 9-10, 10/15/02 — 10/16/02). KDHE project manager for the Site from 1994 to present. She testified about the Gilbert and Mosley project work, including the various phases of investigation and remediation planning, as required by the NCP..

14. Kratz, Gordon. (Tr. Vol. 18, 10/29/03).Mr. Kratz is a defendant. He is the president and 100% shareholder of Tri-Supply.

15. Lamb, Melvin. (Tr. Vol. 16, 10/25/02). Former Land Tool Company employee. He testified regarding handling ánd disposal of chlorinated solvents at Land Tool.

16. Land, Elvert. (Tr. Vol. 16, 10/25/02). Mr. Land is a defendant. Mr. Land owned the property located at 650 E. Gilbert. He also owned 90% of the stock of Land Tool Company.

17. Letterman, Jerry. (Tr. Vol. 11, 10/17/02). Former Reid Supply employee. He testified regarding handling and disposal of chlorinated solvents at the Reid facility, including spills in which he was involved.

18. Letterman, Mike. (Tr. Vol. 13, 10/22/02). Former Reid Supply employee. He testified about handling and disposal of chlorinated solvents.

19. Martin, Pat. (Tr. Vol. 13, 10/22/02). Geoprobe operator for EPS. He testified regarding his geoprobe sampling at the Site. .

20. McBean, Edward. (Tr. Vols. 24-26, 11/12/02 — 11/14/02). ALT’s modeling expert.

21. McCabe, Steven. (11/14/02). ALT’s expert regarding concrete.

22. Olsen, Roger. (Tr. Vols. 2, 3, 5-7, 21-22, 10/2/02 — 10/3/02, 10/7/02-10/8/02,-11/5/02 — 11/6/02). City’s expert regarding NCP compliance and cost allocation. Olsen was a vice-president with CDM, and the project leader for the City’s remediation work.

23. Richwine, Larry. (Tr. Vol. 12, 10/18/02). Former APCO employee who worked at the Wichita facility. He testified regarding handling and disposal of chlorinated solvents at the APCO facility.

24. Rovers, Frank. (Tr. Vol. 27, 11/15/02). ALT’s allocation expert.

25. Smith, Michael. (Tr. Vols. 19-21, 10/30/03, 10/31/03, 11/5/03). The City’s modeling expert.

26. Stamm, Gene. (Tr. Vol. 11, 10/17/02). General Manager for Reid Supply until July 1986. He testified regarding the operations and management of Reid Supply.

27. Trombold, Charles. (Tr. Vol. 13-14, 10/22/02 — 10/23/02). Defendant Charles Trombold was the son of Defendant Walter Trombold. He was an officer, director and shareholder of Reid Supply.

28. Trombold, David. (Tr. Vols. 10-11, 10/16/02 — 10/17/02). Defendant Charles Trombold was the son of Defendant Walter Trombold. He was an officer, director and shareholder of Reid Supply.

29. Trombold, Walter. (Testimony by deposition transcript.) Mr. Trombold is a defendant. Until his retirement in May 1986, he was the chairman of the board, president, treasurer and principal shareholder of Reid Supply.

30. Voorhees, Michael. (Tr. Vols. 28-29, 11/19/02 — 11/20/02). ALT’s expert regarding groundwater flow.

APPENDIX C

MAP OF THE GILBERT AND MOSLEY SITE

[[Image here]]

SKETCH OF THE APCO FACILITY

SKETCH OF THE REID SUPPLY FACILITY

SKETCH OF THE LAND TOOL FACILITY

SKETCH OF THE TRI-SUPPLY FACILITY

APPENDIX D

Glossary of Abbreviations and Technical Terms

[[Image here]]

[[Image here]]

[[Image here]]

[[Image here]]

[[Image here]]

[[Image here]]

[[Image here]]

[[Image here]]

[[Image here]]

[[Image here]]

1 . See App. A for a summary of significant pretrial rulings.

2 . The NPL is a listing of the nation's most contaminated sites. The Site was not placed on the NPL but was later included in a State Deferral Pilot Program, a subset of the EPA/ KDHE multi-site agreement that provided for the clean-up of certain sites not yet placed on the NPL.

3 . The pump and treat system essentially involves the use of water wells to pump the contaminated water out of the aquifer. The ''treat” part of the system uses air strippers to atomize the contaminated water. Since chlorinated solvents are highly volatile, they flash to vapor from the atomized water droplets and are released to the atmosphere. The residual water is then collected and discharged to the Arkansas River. For a more in-depth discussion of the pump and treat system, see section III.D.2 in this Memorandum Decision.

4 . The court also adheres to its previous ruling that the City's claims were not time barred by the applicable statute of limitations provision of 42 U.S.C. § 9613(g). Defendants revisit the issue in the Pretrial Order and in post-trial submissions. The court has reviewed these submissions with the particular purpose of identifying any material facts developed at trial which would persuade the court to reverse its prior decision. The court has also reviewed Colorado v. Sunoco, Inc., 337 F.3d 1233 (10th Cir.2003), the only intervening Tenth Circuit decision regarding CERCLA's statute of limitations. The court sees no reason to depart from its initial ruling, however, and again determines that the City's action is timely.

5 . The relevant parts of the statute now read as follows:

(35)(A) The term "contractual relationship”, for the purpose of section 9607(b)(3) of this title, includes, but is not limited to, land contracts, deeds, easements, leases, or other instruments transferring title or possession, unless the real property on which the facility concerned is located was acquired by the defendant after the disposal or placement of the hazardous substance on, in, or at the facility, and one or more of the circumstances described in. clause (i), (ii), or (iii) is also established by the defendant by a preponderance of the evidence:

(i) At the time the defendant acquired the facility the defendant did not know and had no reason to know that any hazardous substance which is the subject of the release or threatened release was disposed of on, in, or at the facility.

(ii) The defendant is a government entity which acquired the facility by escheat, or through any other involuntary transfer or acquisition, or through the exercise of eminent domain authority by purchase or condemnation.

(iii) The defendant acquired the facility by inheritance or bequest.

(B) Reason to know

(i) All appropriate inquiries

To establish that the defendant had no reason to know of the matter described in subparagraph (A)(i), the defendant must demonstrate to a court that—

(I) on or before the date on which the defendant acquired the facility, the defendant carried out all appropriate inquiries, as provided in clauses (ii) and (iv), into the previous ownership and uses of the facility in accordance with generally accepted good commercial and customary standards and practices; and

(II) the defendant took reasonable steps to—

(aa) stop any continuing release;

(bb) prevent any threatened future release; and

(cc) prevent or limit any human, environmental, or natural resource exposure to any previously released hazardous substance.

42 U.S.C. § 9601(35) (emphasis added).

6 . The City states that the 2002 amendments to CERCLA § 101(35) apply retroactively, citing the Historical and Statutory Notes to 42 U.S.C. § 9607. The Notes contain no such express statement. The section entitled Effect on Concluded Actions speaks to amendments under Title I of Pub.L. 107-118 and states that the amendments thereunder shall not affect settlements entered into before January 11, 2002. However, a review of Pub.L. 107-118 shows that Title I did not amend 42 U.S.C. § 101(35). Instead, sections (35)(A) and (35)(B) were amended under Title II of Pub.L. 107-118. Accordingly, the statement on which the City appears to rely for its claim of retroactivity does not apply to the changes in 42 U.S.C. § 101(35). See Pub.L. 107-118 § 103, 115 Stat 2356 (2002).

7 . ALT and Reid maintain that the City is barred from bringing a contribution action under CERCLA § 113(f) because the City has not been sued in an action under CERCLA §§ 106 or 107. The court rejected this argument in ruling on a previous motion for summary judgment, and adheres to that conclusion here.

8 . Defendants contend that the City must prove that the Site is a "facility.” The City concludes that the Site is a facility, and defendants do not dispute that conclusion. However, a review of CERCLA § 107(a) shows that the court need only determine whether defendants owned or operated a facility at which hazardous substances were disposed. See 42 U.S.C. § 9607(a). Having found that the individual properties that defendants owned, operated, or to which they were otherwise connected, were facilities, the court need not address whether the entire Site is a facility.

Similarly, defendants contend that the court must determine whether Gilbert and Mosley is one site or multiple sites. A review of CERCLA §§ 101, 107, and 113 reveals that the only relevant use of the word "site” in those statutes relates to the definition of "facility”. 42 U.S.C. § 9601(9); see 42 U.S.C. §§ 9601, 9607, 9613. Having decided that the court need not determine whether Gilbert and Mosley is a facility, there is likewise no reason to determine if it is a site. Although Gilbert and Mosley is referred to as "the Site” throughout this opinion, that represents convenient semantics, not a legal conclusion. To the extent defendants contend that each contamination plume within Gilbert and Mosley is a separate site (and that any response costs allocated to them should be based solely on the size of their respective plumes) that argument is rejected for reasons discussed later in this Memorandum Decision.

9 . CERCLA § 107(a)(2) speaks in terms of whether a hazardous substance was disposed of, and relies on the definition of disposal found in RCRA. 42 U.S.C. § 9607(a); see id. § 9601(29). Interestingly, RCRA defines "disposal'' in terms of hazardous waste. See 42 U.S.C. § 6903(3). Since CERCLA § 107(a)(2) specifically addresses disposal of hazardous substances, this language can only mean that CERCLA intends to substitute "hazardous substance” for "hazardous waste” in RCRA's definition of "disposal”. In any event, the argument is academic since PCE and TCE are both hazardous substances and hazardous wastes. See 40 C.F.R. § 261.31 (2002); Bob's Beverage, Inc. v. Acme, Inc., 264 F.3d 692, 694 (6th Cir.2001).

10 . Section III.B analyzes whether a release occurred at the facilities in question. Although the definitions of disposal and release are not identical, they overlap considerably. Compare 42 U.S.C. § 6903(3) with id. § 9601(22). Under the facts of this case, where the chlorinated solvents reached the ground and groundwater through spilling and leaking, any release would also be a disposal.

11 . Although the City stipulated to these dates in the pretrial order, the City attempts to dispute these same dates in its post-trial brief. The court has reviewed the highly equivocal testimony that the City cites in support of its new conclusions, and finds the testimony unconvincing. Accordingly, the stipulations in the pretrial order will control.

12 . Although Richwine claims to have worked at 1001 E. Lincoln until 1972, virtually all the other references to APCO’s operations at that location indicate that APCO closed it down in 1971. In any event, the discrepancy is immaterial.

13 . The liquid was orange juice.

14 . Richwine's spill estimates were even higher in a written statement taken before his deposition. The statement was written by one of plaintiff's counsel, who interviewed Ri-chwine on August 17, 2000. Richwine initialed every page of the statement at the time it was written, and certified that it was true. The statement reads: "For an average-size spill (about two gallons) we did not attempt to clean up the spill but let it evaporate and dry. For larger spills (for example 4-5 gallons) we would clean the spilled solvent by using rags and/or Floor Dry.”

15 . Apparently Richwine thought the photo depicted a one-gallon spill, as opposed to a half-gallon spill, and changed his trial testimony accordingly. The change in Richwine's testimony is best illustrated by the following colloquy between Richwine and Bob Driscoll, plaintiff’s lead counsel.

Q: If I understand you, 2 to 3 gallons was our original estimate of an average spill. Correct?

A: Yes.

Q: Of chlorinated solvents at the APCO facility. Correct?

A: I believe that's right, yes. I don't recall that for sure.

Q: And your estimate today here in court of the amount of an average spill of chlorinated solvents at the APCO facility is what?

A: I don’t believe there was ever maybe more than a gallon or two.

Q: And, now, would you explain why you have changed from your original estimate to the estimate you just gave?

A: Because I seen how much a gallon of spillage makes and I don’t think there was that big of an area on the concrete whenever they disconnected those [hoses].

Q: And how did you come to make that observation?

A: I seen a demonstration of the amount that was spilled.

Q: Where did you see that demonstration and when?

A: Kasey showed it to me?

Q: Kasey who?

A: Rogg.

Q: Kasey Rogg showed it to you. And when was that?

A: I don't recall.

Q: Was that between the time you gave your written statement and the time that you gave your deposition in this case?

A: No. It was after the deposition.

Q: Okay. Is it recent? Your deposition was in August 2000.

A: Yeah.

Q: And can you place it in time between then and now?

A: I don’t remember exactly when it was.

Q: Was it this year?

A: Yeah.

Q: The last three months?

A: I don’t remember. Yeah, I think it was within the last three months. I don't remember exactly when it was.

Q: How did this demonstration get displayed to you?

A: He just showed it to me.

Q: Would you just describe, please, what happened at this demonstration? Were you looking at a video or were you looking at a gallon bucket dumped on the floor? What were you looking at?

A: I was looking at a picture.

Q: A picture?

A: Uh-huh.

Q: What kind of picture?

A: I don't know what kind it was.

Q: Photograph?

A: Yes.

Q: Still photograph?

A: Yes.

Q: Not a video?

A: No.

Q: What did Mr. Rogg tell you about that photograph? Or were there more than one?

A: One is all I seen.

Q: All right. What did he tell you about that photograph?

A: That the demonstration had been made and that’s how big a spot it made.

Q: Describe as best you can this photograph. Was it black and white?

A: Yes.

Q: How big was it?

A: Three by five.

Q: Three inches by five inches?

A: Yes.

Q: And it depicted a spill of water?

A: Water or something mixed with it. I’m not sure what it was.

Q: Some kind of fluid?

A: Yeah.

Q: Was it a straight down shot on the fluid?

A: I wasn't there, I don’t know.

Q: Well, as you saw it, was it a straight down shot, a side angle shot in this photograph?

A: It was kind of above the fluid.

Q: What was the surface on which the fluid had been spilled?

A: Concrete.

Q: Did Mr. Rogg tell you how he came to be in possession of this photograph?

A: No, he didn’t.

Q: Did you ask.

A: No, I didn't.

Q: Did you accept his word?

A: Yes, I did.

Q: So based on that, your original estimate of the amount spilled at 1001 E. Lincoln of chlorinated solvents has gone down in number. Right?

A: Yes.

Q: Gone down in volume?

A: Yes.

Q: Okay. And today your testimony is that an average spill of chlorinated solvent which occurred 50% of the' time upon the disconnecting process of chlorinated solvent transfer at 1Ó01 E. Lincoln is 1 to 2 gallons; is that right?

A: Yes, it is.

16 . Generally speaking, the on-site results at APCO are consistently several orders of magnitude higher than the off-site results. Regardless of that discrepancy, the presence of PCE in the soil is confirmed at multiple locations on the APCO property by both on-site and off-site data. Even if the court were to find that PCE is not present in the concentrations reported by the on-site lab, it simply does not view the data so inadequate as to yield false-positive results throughout the property. Thus, while plausible arguments can be made about the degree of contamination in the soil, its presence cannot be disputed.

17 . ALT suggested that the soil contamination at the APCO facility may have been caused by excavation of underground storage tanks pri- or to sampling. However, it has not pressed that argument with any fervor. The theory gleans only a passing mention in post-trial briefs. Moreover, ALT put on only meager evidence at trial to support this theory. Indeed, the court questioned ALT's expert, McBean, regarding this issue. His responses were equivocal and quite unconvincing.

18 . Appendix C-2 is a diagram of the APCO property showing the general location of various test locations which will be referred to in this decision, e.g. APC-5. Similar diagrams are appended for the Reid, Land Tool, and Tri-Supply properties. See App. C-3 to C-5. The diagrams are intended to help any reader who lacks access to the many other exhibits depicting more detailed information but which cannot be readily reproduced in readable form.

19 . Pursuant to the 1991 City/KDHE Agreement, the City was required to comply with a number of QA/QC requirements including:

45. All samples analyzed pursuant this Agreement shall be analyzed by a laboratory which participates in a quality assurance/quality control program equivalent to that specified in the document entitled “USEPA Contract Laboratory Program Statement of Work for Organic Analysis” and USEPA Contract Laboratory Program Statement of Work for Metals Analysis” ("Contract Lab Statement of Work”) and is approved by the State of Kansas.

46. All sample collection and analysis shall be performed in compliance with EPA-approved methods, including timing of analyses, documentation of sample collection, handling and analysis.

47. Laboratory deliverables consistent with and equivalent to those specified in the Contract Lab Statement of Work shall be submitted to KDHE for any analytical work that is identified in the Work Plan approved by KDHE.

48. The City of Wichita shall use the quality assurance, quality control, and chain of custody procedures specified in the Quality Assurance Project Plan, for all sample collection and analysis performed pursuant to this Order.

20 . In assessing potential sources at the Site, Hendron relied upon the property's current and historical uses, the current and historical industrial processes conducted at the facility, and the analytical results collected on or around the property. Analytical data for each location was compared with the site's use and industrial process information for consistency. Source areas were identified as those areas that generally showed elevated shallow groundwater contamination or soil contamination on a property consistent with the site's operational history.

21 . In addition to storing and selling chlorinated solvents, APCO stored fuel products of various types. There were significant indications of BTEX (petroleum) compounds in the groundwater samples tested at 91 OB, and strong fuel odors were noted in the groundwater at certain depths. According to Hen-dron, the presence of petroleum compounds in the groundwater at 91 OB is consistent with APCO's fuel operations, and is a likely contributor to PCE dechlorination at the site, resulting in the high presence of PCE daughter products, i.e., TCE, DCE and VC, at that location.

22 .The court granted ALT's continuing objection to 91 OB samples, which were taken after the close of discovery. The objection now is overruled. The 2002 test results merely confirm prior readings. No prejudice to ALT results from their consideration.

23 . For more information on ACLs, see infra note 34.

24 . ALT previously filed a motion in limine to strike the on-site data due to lack of reliability. The court denied that motion without prejudice in a minute order. For the reasons stated in this decision, the court finds the data sufficiently reliable.

25 .If McBean is correct in this regard, then the fact that chlorinated solvents were found in the soil and groundwater at numerous locations at APCO, more than 20 years after APCO ceased doing business there, gives rise to a reasonable inference that a lot of chlorinated solvents were spilled during APCO's occupancy of the property.

26 . The City is the current owner of property located at 777 E. Waterman, otherwise referred to as the "Bus Barn.” Although the City knew prior to purchase that the property was contaminated with, among other things, TCE, the City purchased the property and indemnified the seller against environmental liability. The City performed source control work at the Bus Bam in 1998 under the less stringent standards set forth in the Interim Remedial Guidelines. Subsequently, the KDHE enacted significantly more stringent soil remediation standards, which will presumably apply to any future source control work within the Site.

27 . The City has adhered to this conclusion despite the fact that the on-site soil data at the Bus Barn and 1001 E. Lincoln were both collected and transported using Geoprobe liners or "sleeves.”

28 . McBean testified that approximately 4500 and 3500 ppb of PCE and TCE respectively remain in the soil at the Bus Bam. He contended that even after the Bus Barn was remediated, it still has chlorinated solvent concentrations that are orders of magnitude higher than the highest levels measured by the onsite laboratory at 1001 E. Lincoln.

29 . The sketch of the Reid Supply facility in Appendix C is merely provided to show the general layout of the property, as well as the approximate location of the various sampling points addressed in this section.

30 . The MCL is a drinking water standard.

31 . The sketch of the Land Tool facility in Appendix C is merely provided to show the general layout of the property.

32 . The sketch of the Tri-Supply facility in Appendix C is merely provided to show the general layout of the property.

33 . After requesting bids from numerous environmental consultants in 1990-1991, as required by the City/KDHE agreement, the City retained CDM, an engineering consulting firm, to complete the RI/FS. CDM did much of the actual investigation the court attributes to the City.

34 . As part of its plan for hydraulic containment, the KDHE established ACLs based on risk formulae provided by the EPA. The ACLs were less stringent .than the applicable MCLs, but were still within ranges deemed acceptable in CERCLA. Specifically, the KDHE established the following ACLs for the contaminants at issue: 14 parts per billion for PCE, 21 parts per billion for TCE, 70 parts per billion for DCE and 2 parts per billion for VC.

35 . The court in Amisub noted that a "state agency’s determination of procedural and substantive compliance with federal law is not entitled to the deference afforded a federal agency,” and held that "the district court improperly limited its review as to whether the [state agency's] findings and assurances were arbitrary and capricious to determine whether the state plan violated federal law.” 879 F.2d at 795-96. But the court later applied an arbitrary and capricious standard to the state agency’s findings that the particular state program at issue in that case complied with federal law. Id. at 801.

36 . The parties apparently agree that monitored natural attenuation essentially includes the "no action” remedy with additional monitoring. The court understands the various reasons expressed throughout this case by the City and the KDHE for rejecting the "no action” remedy to apply equally to monitored natural attenuation.

37 . With respect to cost, the NCP provides the following:

Costs that are grossly excessive compared to the overall effectiveness of alternative may be considered as one of several factors used to eliminate alternative. Alternatives providing effectiveness and implementability similar to that of another alternative by employing a similar method of treatment or engineering control, but at greater cost, may be eliminated.

40 C.F.R. §§ 300.430(e)(7)(iii).

38 . The NCP further provides that

[cjost-effectiveness is determined by evaluating the following three ... criteria: long-term effectiveness and permanence, reduction of toxicity, mobility, or volume through treatment, and short-term effectiveness. Overall effectiveness is then compared to cost to ensure that the remedy is cost-effective. A remedy shall be cost-effective if its costs are proportional to its overall effectiveness.

40 C.F.R. § 300.430(f)(l)(ii)(D).

39 . Olsen testified that within plumes A, B, and E, PCE and VC were not degrading at all and concluded as a result that "monitored natural attenuation was not an appropriate alternative.”

40 . As a reference point, 1x10 -6 refers to 1 excess cancer in one million exposed individuals over a lifetime.

41 . For similar reasons, the court rejects ALT’s assertion that the groundwater at the Site is highly saline due to natural conditions, such as the influx of saltwater from the Wellington aquifer into fresh water sources.

42 . Indeed, the City and the KDHE originally planned to send the treated water back into the public drinking water system. However, there was strong public opposition to that idea, resulting in the decision to discharge the treated water to the Arkansas River, instead. Thus, the treated water appears to have been technically capable of being restored to drinkable quality. It was simply unpalatable to do so in light of public opinion.

43 . The comments and KDHE responses included those pertaining to (1) the use of ACLs rather than MCLs, (2) beneficial reuse of remediated groundwater, (3) air emissions, and (4) bioremediation.

44 . ALT furthermore argues that the City's failure to collect reliable sampling data, the criteria for which are set out in the City’s QAPP, renders its conclusions regarding an appropriate response unreliable. The court finds ALT’s argument insufficient to demonstrate any inconsistency with the NCP for reasons explained elsewhere in this decision.

ALT additionally claims that the City has failed to follow fundamental NCP requirements with regard to source control measures at 1001 E. Lincoln and nevertheless seeks a declaratory judgment for future response costs associated with source control measures at the APCO facility. The court will address arguments related to declaratory judgment later in this decision.

45 .This language creates different standards for recovery by a private party than those created by § 107(a)(4)(A) for the government or an Indian tribe. Under § 107(a)(4)(A), the government can recover “all costs of removal or remediation ... not inconsistent ” with the NCP. 42 U.S.C. § 9607(a)(4)(A) (emphasis added). By contrast, § 107(a)(4)(B) only permits private parties to recover “necessary costs of response ... consistent with ” the NCP. Id. § 9607(a)(4)(B) (emphasis added). Thus, private parties generally have the additional burden of proving consistency with the NCP and necessity of costs.

46 . Hardage further stated, "If the government's response actions are in harmony with the NCP, the costs incurred pursuant to those actions are recoverable from liable parties.” Hardage, 982 F.2d at 1441. Although Hard-age involved the United States seeking response costs, and thus applied the CERCLA § 107(a)(4)(A) standards for NCP consistency and cost recovery, Hardage's insistence that only the response action be consistent with the NCP is just as valid.

47 . ALT argues that the City failed to comply with its own competitive bidding ordinances when it entered into agreements with CDM, and that this violation prevents the City from proving that certain costs were necessary. According to ALT, the City was required to select the least expensive means to complete the response action; hence, by failing to bid all the work, the City can't prove that it got the cheapest price for the job. Defendants failed to raise the competitive bidding issue in the pretrial order. Accordingly, that issue was not preserved for trial. See Fed.R.Civ.P. 16.

Alternatively, even if the issue was preserved under the umbrella of proving necessity of response costs, it is still unavailing. Purely as a practical matter, competitive bidding would have lengthened the already prolonged project. The contract between the City and CDM had been amended 10 times at the time of trial. If each of these amendments had been competitively bid, it is possible that different contractors would have been involved in the project, with the attendant problems. Finally, defendants have offered no evidence that competitive bidding, in fact, would have reduced the costs of the project. ALT cites no legal authority, nor has the court found any, that says "necessary costs” means the least expensive means available. As stated above, costs are denied when they are duplicative, wasteful, or otherwise unnecessary. East Bay, 135 F.Supp.2d. at 1099. Failure to comply with bidding ordinances does not render an otherwise recoverable cost unnecessary. Similarly, ALT’s argument that failure to comply with the bidding ordinances renders the resulting contracts void is also irrelevant. CERCLA does not require that costs be incurred pursuant to a contract. Instead, it merely requires that costs be ¡necessary to accomplish the response action. Thus, with or without a contract, the City has incurred costs for the Gilbert and Mosley,response action. Those costs will be considered on the merits, regardless of whether the City followed its prescribed bidding practices.

48 . ALT has argued that Faulkner’s work was essentially worthless, in that he failed to investigate the work behind the invoices to ensure it was done properly, met contract specifications, etc. Despite ALT's objections, the court accepts Faulkner’s work for what it is— a verification that sufficient documentation exists to show that the money paid by the City was properly billed and the paperwork indicates that the work is related to the Site. Indeed, as a result of Faulkner’s work, some costs originally included as Gilbert and Mosley response costs were removed because they were for work outside the Site, or for costs that are not recoverable under CERCLA.

49 . ALT originally contended that City’s claims were also limited by the fact that the City had taken in tax money to pay some of the response costs. See 42 U.S.C. § 9614(b). That issue was not pressed at trial, nor was it addressed in post-trial briefs. Accordingly, the court finds that theory has been abandoned.

50 . The City also asserted an apparent defense to any contribution claims against it based on a portion of the City/KDHE Agreement. However, defendants dropped all counterclaims against the City that might give rise to a contribution claim, and the City did not raise the issue at trial, nor in its post-trial briefs. Hence, the issue is moot.

51 . The term “orphan shares” refers to the response costs that are attributable to a source for which there is no identifiable PRP, or for which all PRPs are insolvent. Sun Co., Inc. (R & M) v. Browning-Ferris, Inc., 124 F.3d 1187, 1193 n. 4 (10th Cir.1997).

52 . As previously noted, ALT suggests that the court should consider the equitable doctrines of laches, unclean hands, and failure to mitigate as equitable factors in allocating response costs. Rather than speak in terms of doctrines, the court will analyze ALT's theories based on the relevant underlying facts that arise in support of these equitable doctrines. In reviewing ALT’s arguments, the court finds that the most relevant conduct undergirding claims of unclean hands and failure to mitigate has to do with the City’s dilatory conduct in performing source control within the Site. That issue is addressed in section IV.A.2. With respect to the claim that the City applied less stringent soil cleanup standards at its Bus Barn property, the court finds that the City complied with the standards in place at the time of cleanup. Any prejudice toward ALT as a result of intervening enactment of more stringent standards is addressed in section IV.A.2, regarding the City’s delay in conducting source control at the APCO facility.

ALT also claims that the City failed to mitigate damages by not allowing the taxpayers to pay for the cleanup through CERCLA’s Superfund. With this argument, ALT seeks to turn CERCLA on its head and undermine the very purpose of the act. CERCLA was enacted "to shift the cost of environmental response from the taxpayers to the parties who benefitted from the wastes that caused the harm.” Pub. Serv. Co., 175 F.3d at 1181 (quotation omitted). By contrast, ALT’s theory is that CERCLA should be used to shift response costs from the responsible parties to the taxpayers. The absurdity of that statement speaks for itself, and need not be addressed further.

Finally, ALT asserts that the City's delay in bringing the suit gives rise to the doctrine of laches. However, given the massive scope of the Gilbert and Mosley problem, any delay was explainable and excusable. Moreover, the delay has not prejudiced ALT other than to allow more pollution to leach from the soil into the groundwater. As already mentioned, that particular problem is addressed in section IV.A.2.

53 . The "Gore Factors” are so named because they were part of an amendment to CERCLA sponsored by Representative Albert Gore. See 126 Cong. Rec. 26781 (1980). He proposed the factors as a moderate approach to joint and several liability, to apportion contribution claims under § 113(f). See Colo. & E. R.R., 50 F.3d at 1536 n. 5. The House of Representatives adopted Gore’s amendment, 126 Cong. Rec. at 26788; however, the language was rejected by the Senate and ultimately removed in the compromise bill. See Boeing Co., 207 F.3d at 1187. Nonetheless, courts have used the Gore Factors as a non-exhaustive list of factors to consider in allocating response costs. See Colo. & E. R.R., 50 F.3d at 1536 n. 5.

54 . Although ALT and Reid have suggested that City-owned sewers may be a source of chlorinated solvents at defendants’ facilities, no credible evidence was produced to support that contention beyond the mere hypothetical possibility that it might have occurred.

55 . The City does not allege that Land Tool or E.H. Land failed to use due care. Since none of the parties address this factor, nor does the evidence suggest that an unusually high or low degree of care was exercised at Land Tool, the court will not consider that factor with respect to E.H. Land or Land Tool.

56 . Based on the court's observation that spilling chlorinated solvents was not unlawful during much of the time period when these defendants operated their respective facilities, they may understandably wonder why they are still being held liable for cleanup costs. The answer, once again, is grounded in equity. With respect to releases at APCO and Reid, the City is an innocent third party. As between an innocent party and a responsible party, equity demands that the responsible party bear the burden of its past actions. Accordingly, if anyone must bear the costs of cleaning up pollution from the APCO and Reid sites, it should be ALT and Reid.

57 . Likewise, ALT filed a motion for discovery sanctions against the City for failure to produce a report regarding source control measures at the APCO facility. ALT's motion was denied without prejudice so that the court and the parties could focus on the upcoming trial. ALT maintains that the report, prepared by CDM, shows that source control is not necessary at the APCO facility. The City counters that the report is only a draft that was never finalized, and that ALT misinterprets the report's conclusions. Like the City's motion for sanctions, the court finds that the conduct here is also highly disputed, and was not fully developed at trial. For those rea sons, and as a matter of equity, the court will not consider these allegations as an equitable factor in allocation.

58 . For example, the evidence showed a strong possibility that there may have been chlorinated solvent releases on the Walter Morris property, immediately south of the APCO facility. See supra section III.B.l.

59 . McBean made a good case for using smaller source loadings at APCO. In particular, he pointed out that Smith’s source loadings were based on an average of several DCE measurements in the shallow groundwater at different locations on and near the APCO facility. One of those measurements was from a geoprobe measurement taken at WM-1, located six feet south of the APCO facility on what is referred to as the Walter Morris property. WM-1 produced a DCE measurement of 17,000 ppb, the highest reading anywhere near APCO. Interestingly, another DCE measurement was taken in the shallow groundwater two feet south of APCO and four feet north of WM-1. This measurement location was designated 828-8B, and yielded a DCE concentration of 340 ppb. McBean concluded that the fact that this well, which lies between WM-1 and APCO, produced a DCE measurement 50 times léss than WM-1 strongly suggests that the contamination mea sured at WM-1 did not migrate from APCO. Given ALT's adoption of McBean’s model using Smith's source loadings, the court need not decide this difficult point. Nevertheless, it is illustrative of the many complex, technical issues running throughout this case for which there are no ready answers.

60 .This conclusion also renders moot ALT's arguments concerning application of the Uniform Comparative Fault Act (UCFA). Under ALT’s interpretation of UCFA, the court would need to determine the equitable share of each solvent PRP (including a portion of the overall orphan share) and then compare that to the total amount paid by each settling defendant. ALT further contends that any amounts underpaid by the settling defendants would be the City's responsibility. However, the court has taken a different approach to allocation — rather than determining each PRP's share and pushing all the shortfalls down to the non-settling defendants, the court is simply allocating to the non-settlors a share of the cleanup costs based on how much pollution they contributed to the Site. This method ensures defendants do not suffer from any previous settlement shortfalls. Furthermore, ALT’s allocation modeling assigned an equal portion of the orphan share to all solvent parties, including ALT, Reid Supply, and the City. Accordingly, to the extent ALT's interpretation of the applicability and effect of the UCFA is correct, its concerns are satisfied.

61 . The City contends that Rovers produced higher allocation percentages for ALT and Reid Supply. While it’s true that Rovers arrived at higher numbers when he apportioned costs based on the City’s allocation model, Rovers ran several allocation scenarios during his investigation. Ultimately, he rejected the City’s modeling and recommended adoption of McBean's modeling with Smith’s source loadings. Those are the numbers that the court uses for allocating response costs to defendants.

62 . Rovers also suggested distinguishing between Site-wide costs and costs that could be isolated to Plume B. While Rovers' suggestion is understandable, the court finds it most appropriate to treat the groundwater contamination across Plumes A, B, and E collectively. The evidence suggests that the three plumes are contiguous and may have even undergone some commingling. Since much of the investigation and design work between the plumes overlapped, the most straightforward approach is to treat them as a single unit for purposes of this suit.

63 . Even though McBean and Rovers were not Reid witnesses, their opinions appeared far more objective towards Reid than those of the City. Indeed, McBean’s client, ALT, would have benefitted from a larger plume of contamination associated with Reid because if Reid's plume overlapped with APCO’s, ALT would have received a further reduction in allocated response costs. Nonetheless, McBe-an’s modeling showed that Reid had a much smaller pollution plume than the City’s model showed. Since McBean’s simulation of Reid's plume worked a detriment to ALT, the court finds that the only logical explanation therefor is that McBean was simply doing his best to provide an objective assessment of the problem at the Reid facility.

64 . The City maintains that Reid’s share should be increased based on the fact that Reid was delivering and spilling PCE at various locations within the Site. Although the City cites testimony of Jerry Letterman in support of that conclusion, Letterman’s testimony was not credible. Moreover, the court notes a glaring inconsistency between the City's proposed treatment of Reid’s delivery activities and those of Vulcan Chemical. The City admits that Vulcan was spilling chlorinated solvents at delivery points within the Site; yet, the City maintains that all responsibility for those spills should be assigned to the facility owner or operator, and not to Vulcan. By contrast, the City wants to indirectly assign responsibility to Reid for alleged spills when delivering to other facilities by allocating a larger portion of the orphan share to Reid. The City cannot have it both ways. Reid's share will be based on the contamina tion at its facility, not on alleged spills at other facilities within the Site.

The City made similar claims against ALT in the pretrial order, but those were apparently abandoned at trial, and are notably absent from the City’s post-trial brief.

65 . This analysis also applies to spills inside the Reid Supply facility.

66 . This conclusion also eliminates the need to consider piercing the corporate veil as to Kratz.

67 . A review of the case law from the various districts within the Tenth Circuit shows that all cases wherein a complaint was held to satisfy § 107(a)'s written demand requirement involved a single defendant or a handful of related defendants. See, e.g., Dow Chem. Co. v. Sinclair Oil Corp., 3 F.Supp.2d 1252 (D.Wyo.1998); Horsehead Indus. v. St. Joe Minerals Corp., 1996 WL 33415778, at *1 (N.D.Okla. Apr.2, 1996); State of Colo. v. Idarado Min. Co., 735 F.Supp. 368, 371 (D.Colo.1990) (holding that complaint satisfied written demand requirement against three defendants who were all related as parent corporations or subsidiary corporations (this relationship is described in State of Colo. v. Idarado Min. Co., 916 F.2d 1486, 1489 (10th Cir.1990))). In that respect, these cases are consistent with the court's view of Bancamerica — that a complaint seeking overall response costs satisfies § 107(a)’s written notice requirement for a specified amount only when the complaint names one defendant or a few financially related defendants.

68 . The question of pre-judgment interest is moot as to Tri-Supply, since no costs are allocated to it.

69 . The City also seeks a declaratory judgment against Charles and David Trombold. Since they are not covered persons under CERCLA, they cannot be found liable for future costs. Accordingly, they are not discussed here.

70 . This decision was originally filed on December 31, 2003. Subsequently, Reid filed a motion asking the court to reconsider certain aspects of its opinion. The court denied Reid's motion by separate order, but the body of that order is included here for sake of completeness.

71 . The difference between the $23,888 requested and the $16,040.91 actually allocated was based on various costs that the court concluded were not recoverable.

72 . Indeed, Reid admitted that two of its employees “handled 90% of the offloading of PCE at 911," and that one of its employees was “constantly watching employees who were offloading the PCE or TCE.” Accordingly, Reid was not an uninvolved, innocent bystander during PCE deliveries. Instead, Reid employees were “constantly" involved in accomplishing and supervising the unloading process.

73 . To the extent Reid feels that this decision gives the City unfettered discretion to conduct source control at 911, Reid's fears are misplaced. The City bears the risk that it alone will be responsible for the costs of any unnecessary soil remediation at 911.