Ekotek Site PRP Committee v. Self

MEMORANDUM AND ORDER LUNGSTRUM, District Judge. Plaintiff The Ekotek Site PRP Committee (the Committee) brought this contribution action against a number of defendants pursuant to the federal Comprehensive Environmental Response, Compensation and Liability Act (CERCLA), 42 U.S.C. §§ 9601-9675, to recover a portion of its past and future costs incurred in cleaning up a site contaminated by hazardous substances. After a number of settlements and dismissals, only plaintiffs claim against defendant Morrison Knudsen Corporation (MK) remained for trial. Phase I of plaintiffs litigation against MK ended -on January 24, 1997, when MK consented to judgment with respect to its liability for response costs as a generator and an owner-operator. Phase II (damages) and phase III (allocation) issues with respect to MK were tried to the court in Salt Lake City, Utah, over 10 days from January 12 to January 23, 1998. This Memorandum and Order constitutes the court’s findings of fact and conclusions of law pursuant to Fed. R.Civ.P. 52(a). After considering all of the evidence, as well as the arguments of counsel, the court concludes that the equitable allocation of response costs to MK should reflect MK’s relatively minimal contribution to contamination at the Ekotek Site, as more fully set forth below. Accordingly, the court apportions one (1) percent of all recoverable past and future response costs incurred by plaintiff to MK. Judgment is ordered in favor of plaintiff against MK in the amount of $179,638.84 for past costs and interest on those costs. In addition, a declaratory judgment is ordered in favor of plaintiff against MK, by which MK is deemed liable for one (1) percent of all future recoverable response costs incurred by plaintiff with respect to the Ekotek Site. FINDINGS OF FACT Site Background This case involves the environmental cleanup of a contaminated site, known as the Ekotek Site (the Site), located at 1628 North Chicago Street in Salt Lake City, Utah. The Site is approximately 6.6 acres in area and is divided by railroad tracks into western and eastern parts that are roughly equal in area. MK acquired the eastern half of the Site in pieces between 1929 and 1948. In 1975, MK conveyed all interest in the eastern half to Bonus International Corporation (Bonus) and vacated the premises. From 1953 to 1968, O.C. Allen Oil Company (O.C.Allen) owned the western half of the Site and operated a used oil refinery there. In 1968, Flinco, Inc. purchased the western *1286 half. Flinco ran the refinery until 1978, during which time the company changed its name to Bonus. In 1975, Bonus consolidated the two sides of the Site. In 1978, Ekotek, Inc., a Delaware corporation (Ekotek of Delaware), owned the Site and operated the facility. In 1981, Steven Self and Steven Miller purchased the property and incorporated under the name Ekotek, Inc., a Utah corporation (Ekotek of Utah). This company operated the re-refining facility until 1987, when the company declared bankruptcy. Pe-troChem Recycling Corporation (PetroChem) then leased the property and operated the facility until February of 1988, when the State of Utah issued notices of violation and operations ceased. Morrison Knudsen Operations MK is a large construction company with its principal place of business in Idaho. From the 1940s through the 1970s, MK engaged in more than 100 construction projects in the Utah area, including the Great Salt Lake Causeway project in the late 1950s, which took approximately three years to complete. MK operated a heavy equipment storage and maintenance facility (the yard) on the eastern portion of the Site from the late 1940s until 1975. Aerial photographs of the yard during that period generally show between 40 and 80 pieces of equipment present at the yard and an adjacent parcel at any one time, with more than 100 pieces present on two occasions. The photos also suggest that the yard was much less busy before 1958 than after. The yard contained a number of buildings along its periphery, including warehouses, offices, storage sheds, and a paint shop. The buildings were generally situated along the southern and eastern edges of the yard. Repair and maintenance work was generally performed in the “old shop”, which contained only one or possibly two equipment bays, until approximately 1960, when the “new shop” was constructed at the north end of the yard. The new shop contained four or five equipment bays. The yard was served by a railroad spur, maintained by Union Pacific Raih’oad, that acted as the western boundary of the yard. The yard was also served by two concrete loading docks on the western side of the yard. An underground storage tank containing gasoline was located in the southeastern corner of the yard (UST # 1); another UST, which was actually a buried railroad tanker, sat in the southwestern part of the yard along the raih’oad tracks and contained virgin diesel fuel (UST # 2). In the early 1970s, MK kept some old railroad ties at the yard for use in bracing equipment during transport. The ties, which mainly lay in stacks around the railroad spur, had been treated against rotting. Equipment usually arrived at the yard for repair on MK’s one low-boy truck; less frequently, equipment came by rail. The great majority of the equipment repaired at the yard was powered by diesel fuel. MK did not bring equipment to the yard merely for routine maintenance; minor repairs were made and oil and other fluids were changed in the field. There, waste oil would be dumped into a pit or fill area; it would then be covered or taken to a landfill. This practice was followed even at a project within one-half mile of the yard. Nor was equipment always returned to the yard between construction jobs in the area. MK brought equipment to the yard for significant repair work only, such as a major overhaul of a piece of equipment. Only a small number of mechanics worked at the yard at any one time. Herbert Guk-eisen, the shop foreman at the yard from 1970 to 1975, testified that the number of mechanics varied from five to as many as 25 or 30 on one occasion. Harold Walker, a mechanic at the yard from 1957 to 1969, testified that there were never more than five mechanics employed at the yard at any one time. Payroll records for the period from 1961 to 1975 indicate that the number of mechanics generally ranged from three to seven, averaging the equivalent of 4.6 full-time mechanics. The records show only one week with as many as 19 mechanics at the yard at a time. Equipment would generally be steam-cleaned upon arrival at the yard. The steam-cleaning area was located in the southeastern portion of the yard, outside the old shop. An employee would use a high-pres *1287 sure soap and water spray to dislodge dirt, rocks, and grease, which would be collected on a concrete pad and dumped off of the Site. Liquids would collect in a drain in the center of the concrete pad. No degreasing agent was used in this process. Liquid would occasionally spill off of the concrete pad, but Robert Harvey, who spent most of his employment with MK from 1952 to 1975 at the yard and who oversaw the cleaning and painting operations, testified that such spills did not occur very often. The soil located south of the steam-cleaning pad did become stained from the steam-cleaning process. Most of the repair and maintenance work was performed on equipment in one of the shops, although a small amount of work took place outside in the yard. The equipment bays in the new shop were sometimes full and sometimes empty or only partially full. A major overhaul of a large piece of equipment would normally take two mechanics three or four weeks or even longer to complete. Parts were often sent from the yard to Caterpillar, an equipment manufacturer, for repair. In repair or maintenance of a piece of equipment, crankcase oil, transmission oil, or hydraulic fluid was drained out and changed only when necessary to effectuate the repair. In addition, even when hydraulic fluid was removed, the greater amount of fluid in the equipment’s reservoir would not be touched. Draining oil would be caught in cut-off 55-gallon drums or in cans. Parts were cleaned in solvent. When work was performed outside, oil or other fluids would occasionally splatter or spill onto the ground. In addition, as much as 70 percent of the equipment coming into the yard leaked oil or fluids onto the ground. Minor spills would also occur when mechanics carried waste oil or fluids in the cut-off drums, or when they emptied the drums into waste barrels or the waste tank. There is no evidence of any major spills occurring at the yard, however. Mr. Walker testified that only a quart or two of oil at most was ever spilled onto the ground at the yard at any one time, and then only very rarely. Oil was not permitted to be drained or dumped directly onto the ground at the yard as it might be in the field because the yard was a permanent facility. Waste oil and fluids were collected in 55-gallon drums that generally sat on the east side of the new shop. Waste oil may also have been collected at one time on the west side of the new shop. Mr. Walker testified that there were never more than eight waste oil drums. In the 1970s, waste oil was deposited into a waste oil tank on the east side of the new shop, which held between 200 and 500 gallons. The waste oil would then be picked up by the recycler operating on the western half of the Site. The waste oil was not picked up on a regular basis; rather, employees at the yard called the recycler to pick up the waste oil whenever the drums or tank became full. Mr. Walker testified that, if the yard were very busy, it would take two or three months to fill the waste oil drums used in the 1960s. Mr. Harvey testified that it would take a few weeks or even a few months for the waste oil tank to become full. Waste call-in sheets maintained at the Site during the period from June of 1973 to March of 1975 indicate that the waste oil tank was emptied on a fairly regular basis every four to six weeks. Small amounts of oil were occasionally spilled while being poured into the waste oil tank, and oil spilled out of the top of the tank at least once. Some of the soil around the tank became stained from spilled oil. After a repair was completed, the equipment might be painted. First, loose paint and sand would be scraped away, and the equipment would be wiped with a solvent. Such preparation usually took place in front of the old shop and the paint shop. Some of the painting was also performed outside. Paint was thinned with a solvent, almost all of which was re-used. During the painting process, paint overspray would fall on the bare ground, and solvent would occasionally spill. Ted Orn, who worked at the yard for a year in the early 1960s, and Mr. Harvey testified that a Stoddard solvent was used at the yard. Mr. Gukeisen also testified that he believed the solvent used was a Stoddard *1288 solvent. A Stoddard solvent is not chlorinated. Some spilling also occurred at UST # 2, the diesel tank, staining some of the soil in that area. Although no one is aware of any leaks in the tank, it is likely that the tank has leaked at some point. Mr. Gukeisen left the tank completely dry when MK left the yard in 1975. The MK yard was generally very clean. Some staining did occur at various times throughout the yard, especially in the steam-cleaning area, in front of the new shop, by the waste oil tank, and around the loading docks. The aerial photographs show such staining beginning in 1955. The photos, which were taken approximately annually, show stains recurring only in the steam-cleaning area and by the waste oil tank. Waste oil from the Causeway project most likely was not taken to the recycler on the west side of the Site. Mr. Orn testified that, at the Causeway project, waste oil was deposited into a truck, although he did not know where the truck took the waste oil. Mr. Harvey testified that he was not aware of any used oil ever brought in from projects to the yard or the recycler. Lee Westwood, an MK engineer who spent most of his time in the field, testified that waste oil from a project was deposited at the project site. Operations of Other Owners (Ekotek) O.C. Allen accepted used oil to its re-refinery on the western part of the Site as early as 1953. A 1949 aerial photograph of the Site shows buildings on the west side that might have been the beginning of a refining facility, but there is no othér evidence that used oil was received there at that time. The Environmental Protection Agency (EPA) documents relating to the Site are consistent in stating that O.C. Allen began its refining operation in 1953. According to the EPA Record of Decision (ROD) for the Site, approximately 50 million gallons of used oil were sent to the facility for recycling from 1953 to 1988. An entry in the Federal Register concerning a de minim-is settlement offer estimated that 23,454,592 gallons of used oil went to the Site. Records dating back to the late 1970s that were found at the Site and information provided by members of plaintiff Committee account for 27,412,789 gallons of used oil sent to the Site. The ROD also states that 335,000 gallons of solvent were sent to the Site. The refining operation (Bonus at that time) took over the east side of the Site in 1975. At that time stained soil to a depth of approximately two inches was removed from in front of the new shop, and three to four inches of stained soil was dug up and hauled off from around UST #2. Fill was brought in to make a road running through the east side. At some point, the southern half of the east side was covered with asphalt. The owners of the Site after MK also used UST # 2, but put burner fuel or distillate, including solvents, into the tank with diesel fuel. Six large tanks were brought in after MK left in 1975; they were located in the central western portion of the east side of the Site, and they contained waste diesel and oil. Spills from these tanks did occur. These six tanks were later replaced by six other tanks, collectively called the south tank farm. A larger tank was also added on the east side of the south tank farm. The business of re-refining oil continued after MK left the Site. The majority of the material brought to the Site, in one or two truckloads per day, was used motor oil, used as feedstock in making the recycled oil. The oil was first heated to remove water and compounds, which were vented into the atmosphere. Until 1981 or 1982, an acid process then took place. Sulfuric acid was added to remove additives, forming an acid sludge. This sludge contained sulfated organic matter, lead and other metals, residues from anti-oxidants, polymers from lubricating oil, and detergent additives. At various times from 1967 to 1980, the acid sludge was taken to a landfill, where it was neutralized and disposed of. Beginning in 1980, the sludge was neutralized at the Site. The sludge was mixed with a sugar beet lime, and the mixture was stored on Site, and then periodically taken to a landfill. Documents relating to the clean-up of the Site refer to an acid neutralization pit and a sugar beet pile in the central portion of the east side of the Site. *1289 The oil was then polished with clay. The resulting distillate from that process was reused or burned as fuel. The base oil was placed into tanks and received additives. The finished oil was then stored in tanks on the eastern side until placed into drums and cans for sale. The spent filter cake clay from the polishing process, which contained carbons, impurities, and residual oil, was originally hauled from the Site. Money then became scarce, and the clay began to be stored on the east side of the Site. It was piled behind the new shop, all along the back and down the eastern side of the shop. When that space filled up, the clay was placed in a pile in front of the new shop. Cleanup documents prepared by plaintiffs contractors refer to the clay behind the shop as pile # 3 and the clay in front of the shop as pile # 1. Pile # 2 is used in those documents to refer to clay placed along the front of a warehouse along the eastern edge of the Site. Aerial photographs taken in 1984 and 1988 confirm the existence of these black piles. Portions of the piles were occasionally hauled away from the Site when the owners had enough money to do so. In 1981, a fire destroyed most of the refinery, but the facility was rebuilt. A new polishing and filter system was begun, but that process still produced spent clay. In 1982, Ekotek switched from the acid process to a distillation process. That process produced distillation bottoms that were temporarily stored on Site. Again, spent clay was stored in piles on the east side until it could be hauled away. As a result of the re-refining operations at the Site, substances were introduced into the soil and groundwater, including benzene, toluene, ethylbenzene, and xylenes (BTEX); polynuclear aromatic hydrocarbons (PAHs); PCBs; chlorinated solvents; and heavy metals. The used motor oil received at the Site contained a variety of substances found at the Site, including BTEX and PAHs. According to EPA and other documents relating to the cleanup of the Site, 60 above-ground storage tanks (ASTs), ranging in capacity from 2,900 to 87,000 gallons, were found at the Site, containing 500,000 gallons of liquid. Sources of contamination of the Site included the 60 ASTs, 1,130 drums, 1,500 smaller containers, and numerous piles and pits of waste material. These sources were poorly contained, leaking, and unlined. The documents report leaks, contaminant ponds, waste piles, and surface migration of waste, including waste migrating off-Site and across the street. Allegations were reported of improper handling and disposal of wastes. At the time the EPA commenced its removal action, 1,100 tons of filter cake sludges and sugar beet waste were found on the eastern side of the Site; 125 tons yet remain in a warehouse there. Aerial photos of the Site show extensive areas of stained soil. Committee Plaintiff Committee is a voluntary association of approximately 56 members (membership sometimes changes). The Committee was formed in November of 1988 to negotiate with the EPA to undertake response activities with respect to the Site. Most of the Committee members had been identified by the EPA as parties liable for the cleanup of the Site. The members all sent used oil to the Site for recycling. For the most part, they sent their oil to the Ekotek Site because recycling was preferred, the Site offered the only recycling facility in the area, and the State of Utah recommended that used oil be sent there. According to the Site records and the Committee members’ own information, the members sent a total of 8,777,041 gallons of material to the Site, 97 percent of which was used oil. That figure includes 500,000 gallons attributed to Axel Johnson, the parent company of Ekotek of Delaware, as a result of the settlement of plaintiffs dispute with Axel Johnson concerning that company’s liability and its corporate veil defense. Committee costs are allocated among the members pro rata on the basis of volumes of materials sent to the Site, again determined according to the Site documents dating to the late 1970s and the members’ own information. Union Pacific Railroad Company, to which a volume of 2.4 million gallons has been attributed, is the highest generator of used oil and other materials sent to the Site among the Committee members. *1290 Cleanup of Site After PetroChem was cited for certain violations, the State of Utah requested that the EPA initiate a removal action at the Site. In July and August of 1989, the EPA and the members of the Committee entered into an Administrative Order on Consent (AOC), which legally obligated the Committee to effect a removal action. During the removal, which lasted until early 1992, the Site was contained, sources of contamination were removed, and certain structures were dismantled. The EPA sent the Committee notice of completion of work under the emergency removal AOC in August of 1992. The Committee incurred response costs of $10,504,613.41 relating to the removal stage. The cleanup of the Site continued with the Remedial Investigation/Feasibility Study (RI/FS) stage. Again, an AOC was executed. During this phase, the Committee took samples, conducted tests, and otherwise studied the Site (RI), and then developed various alternatives for remediation of the Site (FS). Work in this stage has continued to the present, although the bulk of the work was completed in 1995. In late 1996, the EPA entered its Record of Decision (ROD), by which it chose Alternative 10 from the FS. That alternative involved the excavation of soil hotspots and other contaminated soil, with some of the soil being placed under a cap on-Site and the remainder hauled off-Site. Alternative 10 also called for cleanup of groundwater contamination by means of intrinsic bioremediation, with additional contingent remedies if needed. The EPA had originally favored a more costly alternative, but the Committee and its contractors convinced the EPA that Alternative 10 would provide sufficient remediation of contamination at the Site. The Committee thus reduced projected future cleanup costs by approximately $10 million. During the RI/FS stage, the Committee incurred $6,152,373.64 in response costs. Plaintiff then embarked on the Remedial Design/Remedial Action (RD/RA) phase, by which its contractors design and perform the actual cleanup of the Site. At the time of trial, the Committee members and the EPA were in the process of executing a Consent Decree for the RD/RA process. The RD/RA phase should take approximately three years, although the RD/RA also calls for long-term groundwater monitoring and periodic EPA review. Through June 30, 1997, plaintiff Committee incurred $234,812.92 in post-ROD response costs. The RD/RA phase is expected to cost an estimated $12 .2 million. The groundwater contingencies — specifically related to the natural attenuation of the organic contaminant plume and the arsenic levels at the Site — could add almost $10 million to future response costs. Plaintiff has also incurred $972,373.57 in costs to identify potentially responsible parties (PRPs) and to create and maintain a database of PRPs and the volumes they sent to the Site. The database was created from Site records and information from the PRPs, and was used in urging the EPA to send out information requests to PRPs. Plaintiff has paid over $2.5 million in fees and costs to its legal counsel. Denise Kennedy, one of plaintiffs attorneys, reviewed all fees paid and separated them into categories for litigation-related fees and expenses and those related solely to the cleanup. Ms. Kennedy calculated that plaintiff paid $1,873,877.44 in non-litigation attorney fees and expenses through June 30,1997. Other Committee Activity In 1988 and 1989, the EPA sent out official requests for information to various parties regarding them contribution to contamination at the Site. Many members of the Committee received such requests after they were identified as parties who had sent used oil to the Site. Such information requests, called section 104(e) requests, state that a response is mandatory and that the recipient must investigate in good faith in order to make that response. In 1991, the Committee sent out notices inviting all 104(e) recipients to a meeting to take place in Salt Lake City in April of 1991. Several settlements with plaintiff resulted from that meeting. In June of 1991, the Committee sent out notice of a de minimis settlement offer. In February of 1992, the Committee made a settlement offer by which parties contributing less than 20,000 gallons *1291 to the Site could completely settle their liability, parties contributing 20,000 to 100,000 gallons could settle with liability re-opened if response costs exceeded a certain figure, and parties contributing over 100,000 gallons could join the Committee. In October of 1992, plaintiff sent out notice of a settlement cut-off date and a possible lawsuit. Before suit was eventually filed, settlements were reached with over 200 parties for a total of approximately $5.5 million. Plaintiff filed suit against 1,100 defendants in March of 1994. The defendants were not served with process, however, and plaintiff did not engage in discovery until 1995. In total, plaintiff has settled with 638 parties for $9,517,251, with 5,160,346 gallons attributed to those parties. Those figures include de minimis settlements with 582 parties, with 2,401,416 gallons, for $8,313,-993. Financial hardship settlements were reached with the other 56 parties, with 2,758,930 gallons, for a total of $1,203,258. In reaching those hardship settlements, the Committee attempted to collect the most money that the settling parties could pay without being driven into bankruptcy. Plaintiff has not settled with 555 parties, to which 370,315 gallons have been attributed. Each of those parties contributed less than 2,000 gallons, and the Consent Decree reached with the EPA prohibits plaintiff from attempting to recover from them. Because no amounts were collected from these parties, plaintiff considers their shares of responsibility for cleanup costs to be orphan shares. Finally, plaintiff has not collected anything from 559 parties (3,755,822 gallons) because of incorrect addresses, 31 parties (1,909,467 gallons) because of bankruptcy, and 49 parties (7,600,412 gallons) because they are insolvent or defunct. Plaintiff first learned about MK as a former owner of a portion of the Site in 1989. MK was not sent a 104(e) request in 1988 or 1989, however. Nor did MK receive a 104(e) request sent in February of 1991, according to the testimony of Edwin Apel, an attorney and vice-president at MK. In early March of 1991, MK received correspondence from the EPA and notices from plaintiff concerning the Salt Lake City meeting. On March 18, 1991, MK received a 104(e) request from the EPA. In July of 1991, MK received the notice of the de minimis settlement offer from plaintiff. On September 9, 1991, MK received its second 104(e) request from the EPA. MK did not respond to either of the first two 104(e) requests it received. Mr. Apel testified that he and another employee were responsible for responding to those requests. The other employee spoke with someone at the EPA, and Mr. Apel believed that no responses were sent because he probably assumed in 1991 that the EPA had told the other employee not to worry about responding. Mr. Apel testified that the failure to respond was not intentional, but that MK did not have any information about the Ekotek Site to give to the EPA at that time at any rate. The EPA did not penalize MK for its failure to respond to the 104(e) requests. MK did not receive any other correspondence relating to the Ekotek Site until March of 1994, when it received the notice of the filing of the lawsuit. In July of 1994, MK waived service of process with respect to the suit. In August of 1996, the EPA sent MK another 104(e) request, and MK returned a response that same month. In October of 1996, plaintiff made its first settlement offer to MK. 1 CONCLUSIONS OF LAW I. Legal Standards Section 107(a) of CERCLA imposes liability on the following persons: 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, [and] any person who by contract, agreement or otherwise arranged for disposal or treatment ... of hazardous substances owned or possessed by such person, by any other party or entity, at any facility.... *1292 42 U.S.C. § 9607(a)(2), (3). Such persons shall be liable for “all costs of removal or remedial action incurred by the United States Government” and “any other necessary costs of response incurred by any other person consistent with the national contingency plan.” Id. § 9607(a). MK has consented to judgment on its liability under this section both as an owner-operator under section 107(a)(2), who owned and ran the yard on the eastern half of the Site, and as a generator under section 107(a)(3), who sent material containing hazardous substances to the western portion of the Site for recycling. The parties have stipulated that they are “persons” and that the Site is a “facility” for purposes of CERCLA liability, see id. § 9601(9), (21); that “a release or a threatened release” of hazardous substances has occurred at the Site, see id. §§ 9601(14), (22), 9607(a)(4); and that plaintiff has incurred response costs, which were in part necessary and consistent with the national contingency plan, see id. § 9607(a). In the present action, plaintiff Committee, consisting of potentially responsible parties (PRPs) under CERCLA, seeks to recover a portion of its response costs from another PRP; the action is therefore governed by the equitable contribution principles of CERCLA section 113, 42 U.S.C. § 9613. See Sun Co., Inc. (R & M) v. Browning-Ferris, Inc., 124 F.3d 1187, 1191 (10th Cir.1997); The Ekotek Site PRP Committee v. Self, 881 F.Supp. 1516 (D.Utah 1995) (present action is governed by section 113). Section 113(f) provides as follows: Any person may seek contribution from any other person who is liable or potentially liable under section 9607(a) of this title.... In resolving contribution claims, 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). Plaintiff also seeks a declaratory judgment under section 113(g), which provides: In any such action described in this subsection, the court shall enter a declaratory judgment on liability for response costs or damages that will be binding on any subsequent action or actions to recover further response costs or damages. Id. § 9613(g)(2). Section 113 does not create new liabilities, but rather incorporates the liability provisions of section 107(a). Sun Co., 124 F.3d at 1191. An action brought under section 113 is different from one under section 107 in important ways, however. For actions under section 107, strict liability is imposed on PRPs for response costs, and that liability is joint and several. Sun Co., 124 F.3d at 1190; United States v. Colorado & E.R.R. Co., 50 F.3d 1530, 1535 (10th Cir.1995). A defendant sued under section 107 can secure apportionment of response costs and escape joint and several liability only if it can demonstrate that the harm it caused is divisible, a nearly impossible burden. Sun Co., 124 F.3d at 1190; Colorado & E.R.R., 50 F.3d at 1535. Plaintiff attempts to invoke these standards in this case; such standards do not apply, however, under section 113. In response to the inequity of a defendant PRP’s burden under section 107, courts began to recognize an implicit right to contribution, and in 1986 Congress amended CERCLA, adding section 113, to codify that right. Sun Co., 124 F.3d at 1190; Colorado & E.R.R., 50 F.3d at 1535. An action under section 113 is governed by equitable principles, and liability is only several, not joint and several. Sun Co., 124 F.3d at 1191, 1193. The burden of proof rests on the “party seeking apportionment to establish that it should be granted.” Colorado & E.R.R., 50 F.3d at 1536. Causation of an increase in cleanup costs is not a required element, but is merely a factor to be considered in the apportionment. Id. at 1534. Section 113 states that a court may apportion response costs “using such equitable factors as the court determines are appropriate.” 42 U.S.C. § 9613(f). The Tenth Circuit has interpreted that language to give great latitude to the district court in allocating costs: A district court has considerable discretion in apportioning equitable shares of response costs. It can allocate the response costs among the liable parties using any *1293 equitable factors it deems appropriate. The court must balance the equities in light of the totality of the circumstances. FMC Corp. v. Aero Indus., Inc., 998 F.2d 842, 846 (10th Cir.1993) (citations omitted). “In any given case, a court may consider several factors, a few factors, or only one determining factor, depending on the totality of the circumstances presented to the court.” Colorado & E.R.R., 50 F.3d at 1536. The court need not weigh all possible factors, but only those equitable factors it deems appropriate. Bancamerica Commercial Corp. v. Mosher Steel of Kan., Inc., 100 F.3d 792, 802-03 (10th Cir.1996); see also Sun Co., 124 F.3d at 1193 (court may consider any factors it deems relevant). The Tenth Circuit has noted that many courts look to the “Gore Factors,” proposed by then-Senator Albert Gore as part of section 113, to help guide them in the exercise of them discretion. Colorado & E.R.R., 50 F.3d at 1536 n. 5. Those factors are as follows: (i) the ability of the parties to demonstrate that their contribution to a discharge, release or disposal of a hazardous waste can be distinguished; (ii) the amount of the hazardous waste involved; (iii) the degree of toxicity of the hazardous waste involved; (iv) the degree of involvement by the parties in the generation, transportation, treatment, storage, or disposal of the hazardous waste; (v) 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 (vi) 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 Tenth Circuit has stressed that the Gore Factors are neither an exhaustive nor exclusive list of possible factors to be considered. Id. In a section 113 action, responsibility for “orphan shares” — that is, “those shares of the waste responsibility which are attributable to PRPs who either are insolvent or cannot be located or identified” — should be equitably apportioned among all PRPs. Sun Co., 124 F.3d at 1193, 1193 n. 4. Finally, prejudgment interest is recoverable in a contribution action, running from the time a cost is incurred or from the time of a written demand for contribution to such cost. Bancamerica, 100 F.3d at 799; 42 U.S.C. § 9607(a). II. Recoverable Costs Under CERCLA sections 107(a) and 113, MK is responsible for its equitable share of “necessary costs of response” incurred by plaintiff that are “consistent with the national contingency plan. See 42 U.S.C. §§ 9607(a)(4)(B), 9613. Plaintiffs accountant, Melbourne Romney, testified that plaintiff incurred response costs through June 30, 1997 totaling $17,864,173.54. That amount did not include litigation expenses or recoverable interest. Nor did that amount include the $2 million estimated value of the Committee members’ own in-kind contributions and expenses; plaintiff concedes that such costs are not recoverable under CERCLA as response costs, but it urges the court to consider the fact of such contributions by its members as an equitable factor in allocation.” Mr. Romney’s figure included $2,562,197.05 in attorney fees. Ms. Kennedy testified that plaintiff incurred only $1,873,877.44 in non-litigation attorney fees and expenses through June 30, 1997. 2 Thus, Mr. Romney’s total for past costs must be modified to $17,175,-853.93. In addition, plaintiff urges the court to consider as orphan shares the equitable share of those parties with which it reached financial hardship settlements, on the basis that those parties did not pay their fair share of response costs. The court agrees that, if MK’s liability is determined at least in part in relation to its proportionate volumetric share, volumetric shares of the financial hardship parties, based on a total of 2,758,930 gallons, should be considered orphan shares. Accordingly, plaintiff agrees that its response *1294 costs should be reduced by $1,203,258 so that the solvent parties absorbing the orphan shares may receive credit for amounts actually collected from the financial hardship parties; otherwise, plaintiff would recover that amount twice—once from the hardship parties and once in the allocation of those costs to non-orphan PRPs. 3 Plaintiff thus seeks to recover MK’s share of response costs totaling $15,972,595.93. Before trial, MK stipulated that most of the response costs incurred by plaintiff were necessary and consistent with the national contingency plan, as required by section 107(a). MK did not stipulate that attorney fees originally included as response costs were recoverable. At trial, however, MK did not challenge the amount of attorney fees and expenses as calculated by Ms. Kennedy. The coui’t was persuaded by the testimony of Ms. Kennedy, and, in the absence of anything to the contrary, concludes that the amount of non-litigation fees and expenses calculated by Ms. Kennedy will be considered recoverable costs under CERCLA. MK does challenge certain response costs claimed by plaintiff with respect to its PRP database. Mr. Romney testified that plaintiff incurred a total of $972,373.57 for PRP identification and development of the PRP database. MK has refused to stipulate that certain line item costs within that category of costs, totaling $940,045.19, are recoverable. Specifically, MK argues that although portions of the disputed costs went toward PRP identification and are therefore recoverable, portions relating to management of the PRP database go beyond mere PRP identification and therefore are not recoverable under CERCLA. In Key Tronic Corp. v. United States, 511 U.S. 809, 114 S.Ct. 1960, 128 L.Ed.2d 797 (1994), the Supreme Court held that CERC-LA did not authorize an award of attorney fees associated with bringing a cost recovery action. Id. at 819. The Court stated, however, that not all payments made to lawyers are unrecoverable expenses under CERCLA. Id. The Court explained: On the contrary, some lawyers’ work that is closely tied to the actual cleanup may constitute a necessary cost of response in and of itself under the terms of § 107(a)(4)(B). The component of Key Tronic’s claim that covers the work performed in identifying other potentially responsible parties falls in this category.... [T]hese efforts might well be performed by engineers, chemists, private investigators or other professionals who are not lawyers. As the Tenth Circuit observed, the American rule [requiring statutory authority] ... does not govern such fees because they are not incurred in pursuing litigation. ... Tracking down other responsible solvent polluters increases the probability that a cleanup will be effective and get paid for. Key Tronic is therefore quite right to claim that such efforts significantly benefited the entire cleanup effort and served a statutory purpose apart from the reallocation of costs. These kinds of activities are recoverable costs of response clearly distinguishable from litigation expenses. Id. (citation and footnote omitted); see Atlantic Richfield Co. v. American Airlines, Inc., 98 F.3d 564, 571 (10th Cir.1996) (under Key Tronic, fees for identifying other PRPs are recoverable because such activity benefits the entire cleanup and serves a statutory purpose apart from reallocation). Finally, the Court held that its ruling permitting certain non-litigation attorney fees did not extend to legal services for negotiating with the EPA regarding a consent decree. Key Tronic, 511 U.S. at 820. The Court concluded that such expenses did not constitute necessary costs of response because the activity primarily served to protect the plaintiffs in *1295 terests regarding its own liability for the cleanup. Id. at 820-21. MK concedes in this case that plaintiffs costs in identifying PRPs would be recoverable under Key Tronic. MK argues, however, that costs related to plaintiffs PRP database go beyond mere identification of PRPs and so may not be recovered. The court agrees. Dennis Farley, a Union Pacific employee who sits on plaintiffs executive committee, testified that the database was completed from records found at the Site in only one year. Thus, the great majority of the work with the database came after the PRPs had been identified from the Site records. Maintaining the database for all these years since its creation no doubt helped plaintiff keep track of generator volumes and settlements. Such purposes are related to the reallocation of costs, however, and under Key Tronic those costs are not recoverable. Plaintiff has not segregated out those costs incurred solely in identifying PRPs; rather, the cost items making up the $972,373.57 total come under the general heading of “PRP Identification and Database Development and Management” in Mr. Romney’s accounting of past costs. Thus, plaintiff has not met its burden of affirmatively proving that any particular portion of the disputed costs items were necessary response costs. See United States v. Hardage, 982 F.2d 1436, 1442 (10th Cir.1992) (private parties seeking to recover response costs must prove that the costs are necessary and consistent with the national contingency plan). Accordingly, the court will not consider any portion of the disputed $940,045.19 as recoverable costs. MK also takes issue with $33,163.70 in costs incurred by plaintiff in having audits done on two separate occasions to review expenses that the EPA charged to the Committee. In arguing that the costs of the audits are not recoverable as response costs, MK compares such costs to attorney fees incurred in negotiating with the EPA, which costs are not recoverable under Key Tronic. Again, the court agrees that the costs are not recoverable. Key Tronic makes clear that costs associated with activities primarily related to the plaintiffs own liability for cleanup costs are not recoverable. Here, the audits were not necessary to the cleanup, but rather concerned how much plaintiff actually owed with respect to that cleanup. Subtracting the non-recoverable amounts relating to the database and the audits leaves a total of $14,999,387.04 in past recoverable response costs, of which plaintiff seeks to recover MK’s equitable share. Plaintiff also seeks contribution from MK for the estimated $12.2 million in future response costs that plaintiff will incur. III. Liability as a Generator The parties addressed MK’s responsibility for response costs for the entire Site by separately considering MK’s responsibility for contamination on the eastern half of the Site, which MK owned, and that on the western half, to which MK sent used oil. For MK’s west side responsibility, plaintiff has urged the court to consider MK’s volumetric share. The owners and operators of the recycling facility on that side are defunct or in bankruptcy; those parties have not contributed to the costs of the cleanup, and their equitable shares must be considered orphan shares. Because the remaining PRPs, including MK, all contributed to the contamination on the west side only by sending used oil and other materials to the refinery, plaintiff argues that the equitable shares of responsibility for contamination on the west side of the Site should be determined strictly on the basis of the volume of materials sent by the PRPs. Indeed, plaintiff has allocated costs among its members in that fashion. The court agrees that the responsibility for west side contamination is most equitably allocated among the non-orphan PRPs by considering the parties’ volumetric shares, at least as a starting point. Orphan shares may then be equitably apportioned among the other PRPs pro rata, on the same volumetric basis. The court stresses, however, that it will not treat the Site as two separate sites and determine MK’s liability with respect to each. The court finds it significant in that regard that the EPA also considered the Site as a single unit. Specifically, the court will not allocate every single cost item to one side *1296 or the other in an effort to assign to MK different percentages of each side’s response costs. Nevertheless, because the overwhelming majority of past and future response costs may clearly be traced to contamination on the west side of the Site, the court believes that it is useful to consider MK’S volumetric share. The court will therefore estimate the percentage of the total material sent to the Site by non-orphan PRPs that MK contributed. The EPA estimated in the ROD that 50 million gallons of material, almost all of which was used oil, was sent to the recycling facility over its lifetime. According to plaintiffs PRP database, which took its figures from documents found at the Site dating back to the late 1970s and from information provided by the PRPs themselves, a total of 27,412,789 gallons were actually sent. Of that amount, 12,338,493 gallons may be attributed to non-orphan PRPs, excluding MK. 4 Including the 500,000 gallons attributed to Axel Johnson as a result of that pai*ty’s settlement yields a final denominator of almost 13 million gallons to be used in calculating MK’s volumetric share. Plaintiff estimated at trial that MK sent 461,588 gallons of waste oil and other waste materials from the yard to the recycling facility. First, one of plaintiffs expert witnesses, Wayne Grip, noted the number and types of construction equipment appearing in aerial photographs of the yard. Another expert, Richard Scott, then used Mr. Grip’s equipment counts to estimate that a fleet of 104 pieces of construction equipment, of various kinds, was served by the yard. Mr. Scott then relied upon Caterpillar manuals and discussions with a Caterpillar employee to derive average crankcase oil, transmission oil, and hydraulic fluid capacities for the various types of equipment. Finally, Mr. Scott determined the recommended service intervals for the various equipment; for instance, Mr. Scott assumed five crankcase oil changes a year for every piece of equipment. Mr. Scott multiplied out these figures to reach a final opinion that MK generated over 450,000 gallons of used oil at the yard from 1949 to 1975. Mr. Scott believed that such a figure was corroborated in the sense that 15 to 20 employees at the yard working in five bays in the new shop could accomplish that many fluid changes. Mr. Scott also opined that 556,880 gallons of oil were sent from the Causeway project to the recycling facility. He calculated that figure in the same manner, except that he used documented figures for the number and types of equipment at the project. Mr. Scott testified that the waste oil from the project would likely have gone to the Site because it would not have been good practice to dump the waste oil at the project site, O.C. Allen provided the only recycling facility in the area, and MK had a preexisting relationship with the facility by virtue of sending waste oil from the yard. 5 *1297 The court concludes that Mr. Scott’s estimates of the amount of waste oil sent to the recycler by MK is grossly excessive. First and foremost, Mr. Scott’s estimates presume that each piece of equipment in MK’s fleet was brought in from the field to have its regular oil and fluid changes done at the yard. The evidence was to the contrary, however. MK employees working at the yard consistently testified that equipment was brought to the yard for repair, and not merely for routine maintenance; rather, oil changes were performed in the field. The employees who worked at the yard testified that fluid changes occurred only when incidental to a repair. Moreover, the evidence does not suggest that equipment came and went from the yard nearly as rapidly as Mr. Scott would have the court believe. A major overhaul might take two mechanics a month to complete. Witnesses also testified that the bays in the new shop might be full some of the time, but would be only partially full or even completely empty on other occasions. In addition, MK usually employed seven or fewer mechanics at the yard. The presence of only one relatively small waste oil tank at the yard and the limited means by which MK was able to transport equipment to the yard further undercut the suggestion that such a large amount of waste oil was generated. The fluid capacities used by Mr. Scott were also called into question at trial. The credibility of Mr. Scott’s figures was undermined upon cross-examination, when he acknowledged that certain Caterpillar manuals indicated capacities under 10 gallons where, for instance, he used a capacity of 80 gallons. Defendant’s expert, Paul Kuhlmeier, gave his opinion that Mr. Scott’s capacities were greatly excessive. The court believes that Mr. Kuhlmeier’s figures are more credible than Mr. Scott’s in part because Mr. Scott relied on representations of a Caterpillar employee for some of his figures, while Mr. Kuhlmeier relied strictly on documentation for the equipment. In addition, the only lay testimony on this issue, from Mr. Gukeisen, tended to confirm Mr. Kuhlmeier’s figures. Furthermore, it appears that Mr. Scott did not take into account that some of the fluid in a piece of equipment would be held in a reservoir and that that fluid would not necessarily be changed. Another question raised by plaintiffs approach here concerns its failure to attach more importance to the only known documentation for the amount of waste oil sent to the west side by MK. Waste-in sheets kept at the Site indicate calls from MK for pick-ups of oil by the recycling facility between November of 1972 and March of 1975. Plaintiff did not attempt to use those documents in determining MK’s volumetric share, even though it constructed volumetric shares for the other non-orphan PRPs, whose contributions are to be compared with that by MK, solely from Site documents and adjustments thereto based on PRP information. The persuasiveness of plaintiffs position is severely diminished by its decision instead to calculate MK’s volume by a method (namely, assuming that all equipment came to the yard for all oil and fluid changes) that has no apparent basis in fact. Extrapolating from the waste-in sheets, one of MK’s expert witnesses, Daniel Stephens, estimated that MK generated between 27,279 gallons and 68,196 gallons of waste oil, depending on whether the waste oil tank held 200 or 500 gallons. Mr. Kuhlmeier calculated a range of 24,000 to 57,000 gallons for MK based on the number of available bays, the average time needed for an equipment overhaul, and the fluid capacities for the equipment, using Mr. Scott’s description of the MK fleet. Mr. Kuhlmeier testified, however, that MK did not have enough mechanics at the yard to effect that many repairs, and, of course, the oil and fluids in a piece of equipment would not all have been actually changed during every repair. The court believes that the most reliable way to calculate the waste oil generated by MK is from the actual evidence of what happened at the yard. An analysis of the waste-in sheet entries for MK reveal that *1298 from June of 1973 to March of 1975 pick-ups of waste oil from the yard occurred at fairly regular intervals, approximately every four to six weeks. Mr. Orn, Mr. Gukeisen, and Mr. Walker, all of whom worked at the yard, testified that a tanker truck came to pick up MK’s waste oh whenever the waste oil tank or drums were full. MK was not on a regular periodic schedule for pick-ups; rather, MK employees would call next door when they needed their waste oil receptacles to be emptied. Mr. Gukeisen, who was in charge of making sure that the tank was emptied, was quite sure on this point. Plaintiff called William Hill as a witness on this issue. Mr. Hill oversaw the pick-ups of oil from MK after 1967. Mr. Hill testified that the pick-ups from the yard evidenced by the call-in sheets were not the only MK pickups, but rather represented only those calls from MK for emergency pick-ups. He testified that MK was on a regular route, with three to four weeks between pick-ups. Mr. Hill’s credibility on this point was impaired, however, by his statement in his deposition that, at least for the period from 1967 to 1969, regular pick-ups were not made from MK’s yard, but that MK’s oil would be picked up only when MK called; Mr. Hill didn’t know what the routine was after 1969. Moreover, the regularity of the pick-ups shown in the waste-in sheets suggest that those pickups were not merely additional emergency calls, but rather more likely constituted all of the trips made to the yard. The court rejects Mr. Hill’s testimony on this point as not credible, and finds that the waste-in sheets evidence the only waste oil sent to the recycling facility by MK after June of 1973. The court must also consider the amount of waste oil sent to the west side on each occasion. The waste oil tank was present at the yard only in the 1970s. Mr. Hill testified at first that the tank held between 500 and 800 gallons, but later in his testimony he seemed to adopt the 500-gallon figure. Moreover, in his deposition, Mr. Hill conceded that the capacity was more likely 500 gallons. Mr. Gukeisen testified that the waste oil tank held only 200 gallons. He was quite certain about the capacity because 200 gallons was the most that was ever pumped out of the tank, and he knew that because MK had to pay by the gallon. The court concludes that Mr. Gukeisen’s number is more likely accurate given his greater basis for knowing the capacity and Mr. Hill’s distinct lack of credibility. As stated above, the waste-in sheets suggest that the waste oil tank was emptied every four to six weeks. Mr. Harvey testified that it would take a few weeks or even a few months before the waste oil tank would be full. Even Mr. Hill stated that generally three to four weeks elapsed between pick-ups from the yard. When Mr. Walker worked at the yard from 1957 to 1969, waste oil was collected in empty oil drums. He testified that there were never more than eight waste oil drums and that, if the yard were very busy, it would take two or three months to fill those drums. That testimony also confirms Mr. Gukeisen’s estimate of the capacity of the waste oil tank — if it took two or three months to fill eight 55-gallon drums, it would take approximately one month to fill a 200-gallon tank. The court thus believes that MK likely sent approximately 2,000 gallons of waste oil to the west side each year, at least in the latter part of MK’s operation of the yard. The aerial photographs of the yard and the absence of the five-bay new shop indicate that activity at the yard was considerably greater in the 1960s and 1970s than in the 1950s. Therefore, the court would estimate that MK sent between 30,000 and 40,000 gallons to the west side between 1953 and 1975. 6 The court also concludes that plaintiff has failed to show that MK sent waste oil back to the Site from the Causeway project. Mr. Westwood convincingly testified that waste *1299 oil was routinely disposed of in the field.' Mr. Harvey also testified that he was not aware of used oil ever coming back to the yard or to the recycler from the field. The court will not attribute over 500,000 gallons to MK solely on the basis of Mr. Scott’s opinion that waste oil might have been sent to the Site from the Causeway. Accordingly, if plaintiffs figure for total non-orphan gallons is used, the court’s 30,-000- to 40,000-gallon estimate yields a volumetric share for MK of approximately one-quarter of one percent (0.0025). Although it is possible that MK’s gallonage might have been higher, it is just as likely that its overall percentage should be lower. The amounts of non-orphan gallons came from Site records dating back to the late 1970s and from PRP admissions. Additional amounts of waste oil may have been sent to the Site by some of the non-orphan PRPs from 1953 through the 1970s that are undocumented and therefore have not been attributed to those PRPs. With a larger total non-orphan volume, MK’s share would decrease. 7 In addition, the vast majority of the waste oil generated by MK came from diesel-powered equipment. The expert witnesses agreed that waste oil from diesel equipment was less toxic than waste oil from gasoline— powered vehicles. Therefore, the court sees some merit in MK’s position that its volumetric share should be adjusted downward to account for the low toxicity of its waste oil compared to the toxicity of the oil sent to the Site considered as a whole. Finally, the court does not adjust MK’s share upward because of Mr. Harvey’s testimony that he observed the recycler dump waste oil into a pit oh the west side. As plaintiff established in defense of its own members’s actions, the State of Utah recommended that used oil be sent there, and environmental regulations were practically non-existent. The court does not believe that such knowledge by one MK employee makes MK more culpable regarding contamination on the west side than other generators of used oil. IV. Liability as an Owner and Operator The court now turns to MK’s contribution to contamination on the east side of the Site, where it operated a construction equipment maintenance yard for 25 years. Plaintiff argues that MK contributed significantly to that contamination, and it suggests that 50 percent of the liability for cleanup associated with the east side should lie with MK. The court does not agree. After considering the evidence of the activities on the east side before and after 1975, the contamination found there, and the response costs and remediation necessitated, the court concludes that MK’s likely contribution to the environmental problems on the east side of the Site, in comparison to that by the subsequent owners, is minimal, and that MK’s equitably allocated share should reflect that relative lack of responsibility. Plaintiff presented abundant proof that MK did in fact contribute to the contamination on the east side, a point that MK and its experts do not seriously dispute. Employees who worked at the yard reported that various substances were introduced into the soil over the life of the yard. For instance, with respect to maintenance and repair activities, equipment leaked oil and fluids onto the ground while sitting in the yard, and oil and fluids were spilled or splattered when employees performed repairs outside, carried waste in unwieldy containers, and deposited that waste into the waste oil drums or tank. Solvents used to clean parts no doubt spilled onto the ground on occasion. The painting operation at the yard produced overspray of paint and thinner that fell to the bare ground. Loose paint and solvent fell onto the soil during the painting preparation process. During steam-cleaning of equipment, liquids and residue sometimes escaped off of the concrete pad. Such spills and leaks often resulted in stained soil, which employees reported in various places throughout the yard, particularly near the new shop and loading docks, in *1300 the steam-cleaning area, and around the waste oil tank. The aerial photographs' taken periodically of the yard confirm the existence of such dark staining of soils. In addition, it is likely that UST #2, in which MK stored diesel fuel, leaked at some point in the tank’s lifetime, as do most tanks of that age. Minor spills and attendant staining were also reported at UST # 2. These waste streams produced by MK contained various contaminants found at the Site. Used oils, fuels, and greases contain heavy metals, volatile organic compounds (VOCs) such as chlorinated solvents, polynu-clear aromatic hydrocarbons (PAHs), and other hydrocarbons. Used transmission and hydraulic fluids contain metals, VOCs, and sometimes PCBs. Paint waste and overspray would have contained heavy metals and hydrocarbons. Gasoline and diesel fuel leaked or spilled from their tanks and tank bottoms and sludges from the tanks would have contained metals, PAHs, and other hydrocarbons. Nevertheless, the evidence indicates that the resulting impact from MK’s introduction of these waste streams into the soil is minor. First, only a small amount of these materials was spilled or leaked onto the ground. There is no evidence of any major spills occurring, and one longtime employee at the yard testified that only a quart or two at most was ever spilled at any one time. Maintenance and repair work was generally done inside the shops, not outside in the yard. Staining did occur, but an analysis of the aerial photos reveals that stains generally did not recur in the same spot in the yard. For instance, only one percent of the entire area of the yard showed staining in more than three of the sixteen aerial photographs taken during MK’s operation of the yard. Staining was only persistent in the steam-cleaning area and around the waste oil tank. Moreover, employees at the yard testified that only soap and water were used in the steam-cleaning process, which testimony strongly suggests that the recurrent stains in that area were caused by water and not by other substances. In addition, plaintiff argues that the solvent used by MK was chlorinated, because such solvents were used in that time period and were more effective and less flammable. Chlorinated solvents are considered hazardous substances under CERCLA. Those who actually worked with the solvent at the yard, however, were consistent in their belief that they used only a Stoddard solvent, a widely used solvent that was not chlorinated. It is probable that, at the time response activities began at the Site in the late 1980s, MK’s contaminant “footprint” had almost entirely disappeared from the yard. Fust, MK left the yard very clean; the two aerial photos from early 1975 show very little soil staining. In addition, after MK departed, the next owner scraped out, to a depth of two inches, an area of stained soil in front of the new shop, and three or four inches of soil was taken in the vicinity of UST #2. A large amount of fill was also brought in to cover the soil and make a road running throughout the east side of the Site. Thus, by the end of 1975, the ground likely showed no visible sign of MK’s operation at the yard. Moreover, plaintiffs experts testified credibly that natural attenuation through biode-gradation would have significantly reduced the amount of contaminants in the soil in the 13 years from MK’s departure from the yard to the start of the cleanup. For instance, Dr. Robert James testified that only three percent of the original volume of PAHs typically remains after 13 years. Furthermore, substances introduced into the soil earlier in MK’s tenure will have had much longer than 13 years in which to biodegrade. Although a contaminated site may not be able to clean itself up entirely in this manner, the contamination would certainly become substantially reduced over time. In fact, the EPA expected groundwater contamination at the Site essentially to clean itself up when it selected natural attenuation as the groundwater remedy. All of which makes it extremely unlikely that MK could be responsible for much of the contamination noted on the east side of the Site. Most important in this regard is the fact that the great majority of the east side’s contaminant hits — that is, instances where contaminants were found at levels exceeding their background levels — that were noted in *1301 the soil by plaintiffs contractors cleaning up the Site occurred at the surface, in samples taken at a depth of zero to two centimeters. The court has difficulty fathoming how surface contamination could be attributed to MK’s operations at the yard, given the cleanliness of the yard when MK left, the additional soil removal by the next owner, the activities that have taken place at the yard in the many years since MK left, and the natural attenuation that would have occurred, especially at the surface where the oxygen needed for biodegradation is readily available. Gene Schmidt, an expert witness for plaintiff, testified that at a few particular locations at which he took soil samples in 1996, he found hydrocarbons that were over 24 years old, which would date them back to MK’s time at the yard. Mr. Schmidt’s method of dating relied on determining the ratio between nC17 and pristane, two chemicals in hydrocarbons that biodegrade at different rates. Mr. Schmidt determined the age of his samples by noting their places along an age-ratio curve developed from a 1993 scholarly paper and his own experiences with hydrocarbon spills of known ages. The euive yielded an age range of 16 to 21 years for several of the samples. Mr. Schmidt then increased those ages by 50 percent to account for the relative lack of precipitation in the area and the relatively non-permeable nature of the soil, which increases permitted the samples to date back to the time of MK’s ownership. 8 The court is not persuaded by Mr. Schmidt’s testimony, however. Dr. James Bruya, MK’s expert, who proved an extremely credible and knowledgeable witness, testified that the model used by Mr. Schmidt could not give a reliable result in this case. Dr. Bruya explained that one problem lay in not being able to know the starting point for the curve because one could not know a particular hydrocarbon’s nC17-pristane ratio at age zero. Dr. Bruya also noted that the study on which Mr. Schmidt relied found, no correlation between the ratio and soil conditions or nutrients; thus, there would be no reason to use an additional 50-percent factor because of soil conditions at the Site. According to Mr. Schmidt’s curve, only a ten percent addition was needed for a spill of known age in West Texas, where the amount of precipitation was similar to that in Utah. Mr. Schmidt attributed the larger factor used in this ease to the relative lack of soil permeability here, due in large part to a layer of clay in the soil. Mr. Schmidt nevertheless used the 50-percent factor for samples he took above that layer and at the surface, where oxygen availability would not be as much of a problem. This seems inappropriate to the court. Dr. Bruya also pointed out that the study required that soil samples from a depth of one meter be used, which fact further calls into question the reliability of Mr. Schmidt’s dating of some of his samples. Finally, as an example of the unreliability of Mr. Schmidt’s method, Dr. Bruya noted one sample location where Mr. Schmidt dated older oil at a depth in between instances of newer oil. For all of these reasons, the court generally discounts Mr. Schmidt’s testimony, which, at any rate, concerned only a few discrete locations in the yard. The court concludes that the contribution to the contamination on the east side by the subsequent owners of the Site (collectively referred to as Ekotek) absolutely dwarfs that by MK. While MK may have spilled small amounts of oil in performing repair and maintenance of equipment, Ekotek was in the business of handling oil, and it conducted a fair amount of that business on the east side after 1975. Ekotek kept numerous large tanks containing large amounts of oil and waste material on the east side. Leaks and improper handling of oil and wastes were reported, and the aerial photographs show great areas of dark staining around the tanks and elsewhere on the east side after 1975. Ekotek located an acid neutralization pit on the east side, and sugar beet waste was found there. Finally, millions of pounds of sludge and filter cake material, which contained oil, were dumped in large piles on the east side behind and in front of the new shop. The relationship between these activities and the soil contamination supports apportioning most of the responsibility for cleanup *1302 on the east side to Ekotek. Plaintiff argues that the contaminant hits on the east side correlate with types and- locations of MK’s activities at the yard, but that argument is largely unconvincing. One example of a possible correlation is with respect to MK’s painting operations contaminants associated with painting, such as chromium and other metals, were found near MK’s paint shop, and there is no evidence of a similar-operation in that area after MK left the Site. Contamination around UST # 2 almost certainly resulted from leaks and spills from the tank, and MK concedes that it did use UST # 2 for a number of years. One dioxin hit was found on the east side between the railroad tracks, where MK for a time stored used railroad ties that had likely been treated with substances that are sources for dioxin; no other credible explanation has been offered for the dioxin hit there. Otherwise, however, MK’s contributions of used oil and other materials to the soil were scattered throughout the yard and thus cannot be tied to any particular contamination. Moreover, contamination with oil as the likely source cannot be readily ascribed to MK, inasmuch as Ekotek dealt with that substance in far greater quantities. In fact, the contamination on the east side correlates much better with Ekotek’s activities. The RI report, prepared by plaintiffs contractoi’s under the supervision of plaintiffs expert, Sara Black, contains a summary of soil impacts on the east side. The largest area noted is in the vicinity of the loading docks, where 4,300 cubic yards of soil is affected. Sludge pile # 1 sat in this area, and the RI states that the impacts in this area appear to be related to the sludge storage by Ekotek. Similarly, the RI attributes 260 cubic yards of affected soil northeast of the new shop — where pile # 3 was located— to the sludge storage. In the area of UST # 2, which Ekotek continued to use after MK left, 2400 cubic yards of soil were affected by diesel fuel leaks and spills. (The RI report also states that groundwater impacts in the area of UST #2 appear limited.) Other impacted areas, totaling 1,300 cubic yards, include the area of Ekotek’s south tank farm on the east side, as well as other areas in the southern part of the east side. The particular contaminant hits also correlate well with Ekotek activities, with the highest contaminant levels routinely occurring where Ekotek’s sludge piles and tanks were located. For example, on the east side, levels for total extractable hydrocarbons (TEHs) are highest, both at the surface and at depth, in the areas of pile # 1, pile # 3, and the south tank farm. BTEX levels are highest at pile # 1, the acid neutralization pit, and around Ekotek’s tanks. Hits for chlorinated solvents are concentrated in the same areas. Lead was found in its highest concentrations near the acid neutralization pit and pile # 2; beryllium, in pile # 1 and pile # 3. PCBs are at their- highest level at pile # 1, and PCB hits follow a likely pattern of surface migration to the southeast part of the yard from that pile. Finally, Dr. Bruya testified very credibly that the gas chromatograms of the hydrocarbon samples taken by Mr. Schmidt are consistent with refined oil, and thus those hydrocarbons are likely attributable to Ekotek’s sludge piles, which retained oil from the refining process. Dr. Bruya showed that if the hydrocarbons were from used oil, such as that contributed by MK, the chromatograms would'have shown blow-by from use in an engine. The extent to which contamination on the east side presents a health risk and so drives the cleanup further supports allocating only a small share of responsibility for the east side to MK. The non-oil sources of contamination that may be attributed to MK most easily— namely, metals associated with MK’s painting operation — are not at such great levels that they present a health risk. In fact, the EPA has not even imposed soil performance standards, mandating cleanup to a certain level of concentration, for those substances. For contamination that may be traced to oil and other hydrocarbons, and that drives both risk and cleanup, Ekotek’s operations are the far more likely source. Again, Eko-tek’s contribution of such sources dwarfs that by MK, and attenuation has undoubtedly reduced MK’s contribution of hydrocarbons. In addition, because MK’s oil came mostly *1303 from diesel engines, MK’s hydrocarbons would have been less toxic on average than those from oil sent to Ekotek generally. The greatest risk at the Site is presented by the Light Non-Aqueous Phase Liquid (LNAPL) plume. The LNAPL is essentially a layer of oil floating on top of the groundwater at the Site. The LNAPL plume extends only over the western portion of the Site, however, and MK almost certainly did not contribute to that contamination by its operation on the east side of the Site. The relationship of contamination on the east side to the cleanup and the costs therefor also suggest minimizing MK’s equitable share. The removal stage of the cleanup was directly related to Ekotek’s operation of the facility. Action was taken to address problems with Ekotek’s many tanks and other containers and to remove the sludge. Samples were taken of the sludge itself, and soil sampling was directed to such Ekotek features as the piles, the acid neutralization pit, and the sugar beet waste pile. There are three total petroleum hydrocarbon (TPH) hotspots at the Site, from which soil must be excavated and hauled off-Site because of the risk imposed. One of those hotspots is on the west side, and the other two are where Ekotek’s piles # 1 and # 3 were located. Such response costs should not therefore be attributed to MK. with respect to groundwater cleanup, it is not clear that there is even any groundwater contamination on the east side to be remedied. One map prepared by plaintiffs contractors during the cleanup, Plate 4-22, shows the dissolved phase plume, an area of groundwater contamination, over most of the west side and extending over northern parts of the east side. Plate 4-22 also shows a small groundwater contaminant plume around UST # 2. The monitoring wells on the east side, however, have not shown any groundwater contamination; contaminant hits have come only inconsistently from geo-probes inserted into the ground to the groundwater. For instance, there were two groundwater hits for TEHs. At one location near UST # 2, there were no hits in later samples, and the RI report states that it is likely that there is no groundwater contamination there. The absence of any TEHs found in the soil at depth at that place also suggests that the one groundwater hit was not accurate. Also at the second location, TEHs were not found in other samples. Vinyl chloride groundwater hits have occurred only in samples taken by geoprobes. The EPA has expressed doubt about the reliability of such results, however, because they have not been duplicated in the more reliable well sampling. Moreover, documents prepared by plaintiffs contractors suggest that contamination indicated by the geoprobes may have resulted from the insertion of the geoprobes themselves. For these reasons, the FS report states that the groundwater does not appear to be affected at-UST #2. The other vinyl chloride hit occurred in a low-gradient area in the groundwater; any vinyl chloride there likely came from two buildings nearby on the west side that showed high levels of that contaminant. Finally, Michael DeDen, an employee of plaintiffs contractor working on the RD/ RA, whom plaintiff called as a witness, testified that the groundwater contaminant plume was now greatly reduced in size from the time of Plate 4-22, and that current monitoring detects groundwater contamination only on the west side. Even if there is any groundwater contamination left on the east side, the remedy chosen for that contamination is bioremediation. Therefore, the cleanup costs associated with groundwater contamination are relatively small. MK does not dispute that it should probably bear some responsibility regarding the cleanup of contamination associated with UST #2. It is likely that leaks and spills did occur, and MK did use the tank. Soil must be excavated there, but there is likely no groundwater contamination in the area of UST #2. According to the breakdown of estimated future remediation costs set out by the EPA and modified by plaintiff, cleanup of contamination from UST # 2 will involve approximately $30,000 in direct costs, with another $30,000 or so in indirect costs. Finally, although the court intends to treat the Site as a single entity requiring a single *1304 cleanup action, it is significant that response costs associated with the west side far outstrip those associated with the east side. Plaintiffs expert, Steven Newton, broke down the future estimated cleanup costs of $12.2 million for the Site as follows: $1 million, or 8.2 percent of the total costs, attributable to the east side of the Site; $10.4 million, or 85.2 percent, attributable to the west side of the Site; and $800,000, or 6.6 percent, attributable to the entire Site. Thus, even if MK were responsible for a significant portion of east side contamination, its relative responsibility with respect to cleanup of the entire Site would remain very minimal, in light of the fact that most of the future cleanup is intended to address contamination on the west side, to which MK contributed only as a generator. Gayle Koch, plaintiffs allocation expert, in allocating plaintiffs past response costs among the east side, the west side, and the entire Site, assigned more than one-half of those costs to the entire Site. Plaintiff would divide responsibility for such costs equally between the east and the west sides. The court disagrees with such an approach. Costs that benefitted both sides would not have been incurred but for the overall contamination of the Site, and the problems on the west side have been far greater than those on the east side. The court also rejects plaintiffs argument that the allocation of past costs should not necessarily mirror that of future costs for the actual remediation of the Site. Again, testing of the entire Site would not have been necessary but for the serious problems requiring remediation that predominate on the western half of the Site. It is worth noting that of the past costs that Ms. Koch did allocate to one side or the other, those for the west side exceed those for the east side by a 14 to 1 ratio. The result is that any MK responsibility for east side past costs is relatively minute when costs for cleanup of the entire Site are considered. Y. Allocation For purposes of allocation, the court will treat the Site as a single unit, as the EPA did. The court concludes that almost all of the environmental problems at the Site resulted from the operation of the recycling facility at the Site, and after 1975 Ekotek operated that facility on the entire Site, and not on two separate parcels that it considered distinct. Accordingly, the liability of MK and other PRPs as generators of used oil and other materials sent to the re-refinery must be imposed with respect to cleanup of the entire Site. The court will not attempt to divide the response costs by side in order to assign MK separate shares of costs as a generator and as an owner-operator. The court has examined MK’s contribution to and relative responsibility for contamination on the east side alone. It has done so, however, only for the purpose of comparing MK’s responsibility with that of other PRPs and determining the extent to which MK’s equitable share should reflect the fact that it also contributed to contamination of the Site as an owner-operator. 9 As explained more fully above, MK’s equitable share if MK is considered only as a generator would likely not exceed one-quarter of one percent. The court agrees that MK did contribute to contamination at the Site — specifically, to contamination on the east side — and its share should be increased above its generator share to reflect that fact. The court concludes that any such increase should be small, however, because that contribution was only minimal. 10 The court also agrees with plaintiff that MK’s share should be adjusted to account for its recalcitrancej in failing to respond to the EPA’s section 104(e) information requests in 1991, but that adjustment should be minor. The court does not condone or excuse MK’s failure in this regard, but plaintiff Committee *1305 and the remediation process suffered little prejudice as a result. The court is persuaded that MK could not have provided much information about contamination at the Site at that time that plaintiff or the EPA did not already have. That conclusion is strengthened by the fact that neither the EPA nor plaintiff pursued MK in search of more information despite early knowledge of MK’s ownership and Union Pacific’s own past relationship with MK. Plaintiff did not even make a settlement offer to MK until October of 1996, and MK certainly cannot be blamed for rejecting that offer given the terms of the proposed settlement. Plaintiff also contends that, in allocating a share to MK, the court should consider the in-kind costs and other expenses incurred by the Committee’s members, as well as the success enjoyed by the Committee in reducing the costs of cleanup. Certainly, plaintiff is to be commended for persuading the EPA to endorse a far less costly remediation alternative; but the members joined the Committee because the EPA said they were liable and they wanted to hold down the resulting costs by doing the cleanup themselves. Thus, only a slight adjustment in MK’s share is warranted on this basis. The court concludes under all of the circumstances of this case that MK should be allocated responsibility for one percent of plaintiffs past and future response costs that are recoverable under CERCLA. The court believes that such an apportionment is equitable here, in that it reflects MK’s contribution to contamination at the Site both as a generator and as an owner-operator. Considering that MK likely sent some one-quarter of one percent of the volume of oil generated by non-orphan PRPs, a one-percent share adequately accounts for MK’s minimal contribution to contamination of its yard and especially to the necessary costs of response, given the relatively small amount of costs resulting from east side contamination. A one-percent share also sufficiently covers the other adjustments discussed above, as well as the possibility that MK made other contributions to contamination at the Site that somehow have not been accounted for. Finally, the increase over the one-quarter of one percent figure is ample enough to encompass a substantial share of the costs specifically attributable to the cleanup of UST # 2. As set out above, MK’s share should be taken from a total of $14,999,387.04 in past recoverable response costs. Plaintiff is therefore entitled to contribution from MK constituting one percent of that amount, or $149,993.87. Plaintiff is also granted a declaratory judgment that MK is liable for one percent of future recoverable costs incurred by plaintiff in cleaning up the Ekotek Site. Plaintiff is also entitled to recover prejudgment interest on MK’s liability for past costs. Ms. Koch calculated that a total of $3,222,303.65 would apply to plaintiffs total past response costs, running from the later of the filing of the complaint on March 23, 1994, and the date a particular cost was actually incurred. Ms. Koch based that calculation on a past cost figure of $16,303,-804.54. Her total interest figure should be reduced to account for the various adjustments made to Mr. Romney’s past cost total (for costs for attorney fees, EPA audits, and the PRP database) and for the different financial hardship amount used by Ms. Koch, which adjustments resulted in the court’s total of $14,999,387.04 for past costs. If Ms. Koch’s total interest figure is reduced in the same proportion that the total past costs have been reduced, the result is $2,964,-497.00 for total interest. If MK is assessed one percent of that amount to mirror its liability for past response costs, MK’s liability for prejudgment interest comes to $29,-644.97. Thus, MK’s total liability is $179,-638.84, and plaintiff is awarded judgment against MK in that amount. IT IS THEREFORE ORDERED BY THE COURT THAT judgment is awarded in favor of plaintiff Committee against defendant Morrison Knudsen Corporation in the amount of $179,638.84 for past response costs and interest on those costs. IT IS FURTHER ORDERED BY THE COURT THAT a declaratory judgment is ordered in favor of plaintiff Committee against defendant Morrison Knudsen Corporation, by which Morrison Knudsen is deemed liable for one (1) percent of all future recoverable response costs incurred by plain *1306 tiff Committee with respect to the Ekotek Site. IT IS SO ORDERED. Footnotes [1] Additional factual determinations by the court have been incorporated into the court's Conclusions of Law. [2] Ms. Kennedy testified that she followed the dictates of Key Tronic Corp. v. United States, 511 U.S. 809, 114 S.Ct. 1960, 128 L.Ed.2d 797 (1994), in separating recoverable from non-recoverable attorney fees. She further testified that she calculated the amount of recoverable expenses by employing the ratio between the recoverable and non-recoverable attorney fees. [3] It appears from the testimony of Gayle Koch, plaintiff's allocation expert, that additional financial hardship settlements may have been reached after Mr. Romney put together his settlement summaries. Because the cost figures were derived from Mr. Romney's tables, the court has used Mr. Romney's settlement figures as well. The parties should treat future financial hardship settlements and other settlements not reflected in Mr. Romney's figures in the same manner in determining MK's liability in the future. Similarly, MK's share of recoverable costs incurred after June 30, 1997, even if technically past costs as of the time of trial, should be considered as a part of MK's future liability. [4] Again, gallons attributed to parties involved in financial hardship settlements are considered orphan shares. [5] At trial, the court took under advisement MK's motion under Daubert and Fed.R.Evid. 702 to exclude Mr. Scott’s testimony concerning waste oil from the Causeway project; at trial, the court denied MK’s Daubert !rule 702 motion as it pertained to other aspects of Mr. Scott’s testimony. The court also took under advisement MK’s motion at the end of trial for judgment as a matter of law under Fed.R.Civ.P. 52(c) on any claim by plaintiff relating to oil sent to the Site from the Causeway; the court denied a similar motion at the close of plaintiff's case-in-chief. The court now denies both remaining motions. Although the court has reservations about Mr. Scott’s methodology in this regard, the court does not deem it necessary in this instance, when acting as the trier of fact itself, to exercise its discretion as gatekeeper and exclude the testimony, especially given MK’s and the court's own ability to examine the witness and flesh out any weaknesses in that testimony. Here, Mr. Scott's approach was very straightforward and certainly not an example of novel scientific evidence. It simply rested on flawed assumptions. Had the issue been able to have been decided sufficiently far in advance of the trial to cause an appreciable savings of time or money if the evidence were excluded, an even harder look at admissibility might have been in order, but that was not the case here. The court therefore believes that it is appropriate to admit Mr. Scott’s testimony concerning the Causeway, and it has duly considered that testimony, although it ultimately rejects the opinions set forth therein. See Fierro v. Gomez, 865 F.Supp. 1387, 1396 n. 7 (N.D.Cal.1994) (better approach under Daubert in a bench trial is to permit the expert testimony and "allow ‘vigorous cross-examination, presentation of contrary evi *1297 dence' and careful weighing of the burden of proof to test 'shaky but admissible evidence' ”) (quoting Daubert v. Metrell Dow Pharmaceuticals, 509 U.S. 579, 596, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993)). [6] The court will not use 1949 as the starting date as plaintiff's experts did. The EPA documents and the cleanup documents created by plaintiff's contractors give 1953 as the starting date for O.C. Allen’s refining operation. Mr. Grip's opinion that some buildings were present in 1949 that might have been the beginning of a refining operation is not sufficient, in the court's estimation, to overcome the evidence in the documents. [7] Such undocumented volumes could account for some of the difference between the 50 million gallons estimated by the EPA and the 28 million gallons for which plaintiff has documentary support. [8] At trial, the court rejected MK’s challenge to this testimony under Daubert or rule 702. [9] At trial and in its trial briefs, plaintiff did not argue that MK should be allocated a super-share of liability on account of its mere ownership of the property. Rather, plaintiff has argued that MK's actual contribution to contamination while owning the yard was significant, although that argument has proved largely unsuccessful. [10] Any other responsibility for contamination on the east side falls upon orphan shares and would thus be apportioned among all non-orphan PRPs.